Natural Hazards and Earth System Sciences最新文献

{"title":"Low-regret climate change adaptation in coastal megacities – evaluating large-scale flood protection and small-scale rainwater detention measures for Ho Chi Minh City, Vietnam","authors":"Leon Scheiber, Christoph Gabriel David, Mazen Hoballah Jalloul, J. Visscher, H. Nguyen, Roxana Leitold, J. Revilla Diez, T. Schlurmann","doi":"10.5194/nhess-23-2333-2023","DOIUrl":"https://doi.org/10.5194/nhess-23-2333-2023","url":null,"abstract":"Abstract. Urban flooding is a major challenge for many megacities\u0000in low-elevation coastal zones (LECZs), especially in Southeast Asia. In\u0000these regions, the effects of environmental stressors overlap with rapid\u0000urbanization, which significantly aggravates the hazard potential. Ho Chi\u0000Minh City (HCMC) in southern Vietnam is a prime example of this set of\u0000problems and therefore a suitable case study to apply the concept of\u0000low-regret disaster risk adaptation as defined by the Intergovernmental\u0000Panel on Climate Change (IPCC). In order to explore and evaluate potential\u0000options of hazard mitigation, a hydro-numerical model was employed to\u0000scrutinize the effectiveness of two adaptation strategies: (1) a classic\u0000flood protection scheme including a large-scale ring dike as currently\u0000constructed in HCMC and (2) the widespread installation of small-scale\u0000rainwater detention as envisioned in the framework of the Chinese Sponge\u0000City Program (SCP). A third adaptation scenario (3) assesses the combination of both approaches (1) and (2). From a hydrological point of view, the reduction in various flood intensity\u0000proxies that were computed within this study suggests that large-scale flood protection is comparable but slightly more effective than small-scale\u0000rainwater storage: for instance, the two adaptation options could reduce the normalized flood severity index (INFS), which is a measure combining flood depth and duration, by 17.9 % and 17.7 %, respectively. The\u0000number of flood-prone manufacturing firms that would be protected after\u0000adaptation, in turn, is nearly 2 times higher for the ring dike than for\u0000the Sponge City approach. However, the numerical results also reveal that\u0000both response options can be implemented in parallel, not only without\u0000reducing their individual effectiveness but also complementarily with\u0000considerable added value. Additionally, from a governance perspective,\u0000decentralized rainwater storage conforms ideally to the low-regret paradigm:\u0000while the existing large-scale ring dike depends on a binary commitment (to\u0000build or not to build), decentralized small- and micro-scale solutions can\u0000be implemented gradually (for example through targeted subsidies) and add\u0000technical redundancy to the overall system. In the end, both strategies are\u0000highly complementary in their spatial and temporal reduction in flood\u0000intensity. Local decision-makers may hence specifically seek combined\u0000strategies, adding to singular approaches, and design multi-faceted\u0000adaptation pathways in order to successfully prepare for a deeply uncertain\u0000future.\u0000","PeriodicalId":18922,"journal":{"name":"Natural Hazards and Earth System Sciences","volume":" ","pages":""},"PeriodicalIF":4.6,"publicationDate":"2023-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41919493","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The potential of open-access data for flood estimations: uncovering inundation hotspots in Ho Chi Minh City, Vietnam, through a normalized flood severity index","authors":"Leon Scheiber, Mazen Hoballah Jalloul, Christian Jordan, J. Visscher, H. Nguyen, T. Schlurmann","doi":"10.5194/nhess-23-2313-2023","DOIUrl":"https://doi.org/10.5194/nhess-23-2313-2023","url":null,"abstract":"Abstract. Hydro-numerical models are increasingly important to\u0000determine the adequacy and evaluate the effectiveness of potential flood\u0000protection measures. However, a significant obstacle in setting up\u0000hydro-numerical and associated flood damage models is the tedious and\u0000oftentimes prohibitively costly process of acquiring reliable input data,\u0000which particularly applies to coastal megacities in developing countries and\u0000emerging economies. To help alleviate this problem, this paper explores the\u0000usability and reliability of flood models built on open-access data in\u0000regions where highly resolved (geo)data are either unavailable or difficult\u0000to access yet where knowledge about elements at risk is crucial for\u0000mitigation planning. The example of Ho Chi Minh City, Vietnam, is taken to\u0000describe a comprehensive but generic methodology for obtaining, processing\u0000and applying the required open-access data. The overarching goal of this\u0000study is to produce preliminary flood hazard maps that provide first insights\u0000into potential flooding hotspots demanding closer attention in subsequent,\u0000more detailed risk analyses. As a key novelty, a normalized flood severity\u0000index (INFS), which combines flood depth and duration, is proposed to\u0000deliver key information in a preliminary flood hazard assessment. This index\u0000serves as an indicator that further narrows down the focus to areas where\u0000flood hazard is significant. Our approach is validated by a comparison with\u0000more than 300 flood samples locally observed during three heavy-rain events\u0000in 2010 and 2012 which correspond to INFS-based inundation hotspots in\u0000over 73 % of all cases. These findings corroborate the high potential of\u0000open-access data in hydro-numerical modeling and the robustness of the newly\u0000introduced flood severity index, which may significantly enhance the\u0000interpretation and trustworthiness of risk assessments in the future. The\u0000proposed approach and developed indicators are generic and may be replicated\u0000and adopted in other coastal megacities around the globe.\u0000","PeriodicalId":18922,"journal":{"name":"Natural Hazards and Earth System Sciences","volume":" ","pages":""},"PeriodicalIF":4.6,"publicationDate":"2023-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44042829","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing long-term tephra fallout hazard in southern Italy from Neapolitan volcanoes","authors":"S. Massaro, Manuel Stocchi, Beatriz Martínez Montesinos, L. Sandri, J. Selva, R. Sulpizio, B. Giaccio, M. Moscatelli, E. Peronace, M. Nocentini, R. Isaia, Manuel Titos Luzón, P. Dellino, G. Naso, Antonio Costa","doi":"10.5194/nhess-23-2289-2023","DOIUrl":"https://doi.org/10.5194/nhess-23-2289-2023","url":null,"abstract":"Abstract. Nowadays, modeling of tephra fallout hazard is coupled with probabilistic analysis that takes into account the natural variability of the volcanic phenomena in terms of eruption probability, eruption sizes, vent position, and meteorological conditions. In this framework, we present a prototypal methodology to carry out the long-term tephra fallout hazard assessment in southern Italy from the active Neapolitan volcanoes: Somma–Vesuvius, Campi Flegrei, and Ischia. The FALL3D model (v.8.0) has been used to run thousands of numerical simulations (1500 per eruption size class), considering the ECMWF ERA5 meteorological dataset over the last 30 years. The output in terms of tephra ground load has been processed within a new workflow for large-scale, high-resolution volcanic hazard assessment, relying on a Bayesian procedure, in order to provide the mean annual frequency with which the tephra load at the ground exceeds given critical thresholds at a target site within a 50-year exposure time. Our results are expressed in terms of absolute mean hazard maps considering different levels of aggregation, from the impact of each volcanic source and eruption size class to the quantification of the total hazard. This work provides, for the first time, a multi-volcano probabilistic hazard assessment posed by tephra fallout, comparable with those used for seismic phenomena and other natural disasters. This methodology can be applied to any other volcanic areas or over different exposure times, allowing researchers to account for the eruptive history of the target volcanoes that, when available, could include the occurrence of less frequent large eruptions, representing critical elements for risk evaluations.\u0000","PeriodicalId":18922,"journal":{"name":"Natural Hazards and Earth System Sciences","volume":" ","pages":""},"PeriodicalIF":4.6,"publicationDate":"2023-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47258602","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing the coastal hazard of Medicane Ianos through ensemble modelling","authors":"C. Ferrarin, Florian Pantillon, S. Davolio, M. Bajo, M. Miglietta, E. Avolio, D. Carrió, I. Pytharoulis, C. Sánchez, Platon Patlakas, J. J. González-Alemán, E. Flaounas","doi":"10.5194/nhess-23-2273-2023","DOIUrl":"https://doi.org/10.5194/nhess-23-2273-2023","url":null,"abstract":"Abstract. On 18 September 2020, Medicane Ianos hit the western coast of Greece,\u0000resulting in flooding and severe damage at several coastal locations.\u0000In this work, we aim at evaluating its impact on sea conditions and the\u0000associated uncertainty through the use of an ensemble of numerical\u0000simulations. We applied a coupled wave–current model to an unstructured\u0000mesh, representing the whole Mediterranean Sea, with a grid resolution\u0000increasing in the Ionian Sea along the cyclone path and the landfall\u0000area. To investigate the uncertainty in modelling sea levels and waves\u0000for such an intense event, we performed an ensemble of ocean\u0000simulations using several coarse (10 km) and high-resolution (2 km)\u0000meteorological forcings from different mesoscale models. The performance of the ocean and wave models was evaluated against observations retrieved from fixed monitoring stations and satellites. All model runs emphasized the occurrence of severe sea conditions along the cyclone path and at the coast. Due to the rugged and complex coastline, extreme sea levels are localized at specific coastal sites. However, numerical results show a large spread of the simulated sea conditions for both the sea level and waves, highlighting the large uncertainty in simulating this kind of extreme event. The multi-model and multi-physics approach allows us to assess how the uncertainty propagates from meteorological to ocean variables and the subsequent coastal impact. The ensemble mean and standard deviation were combined to prove the hazard scenarios of the potential impact of such an\u0000extreme event to be used in a flood risk management plan.\u0000","PeriodicalId":18922,"journal":{"name":"Natural Hazards and Earth System Sciences","volume":" ","pages":""},"PeriodicalIF":4.6,"publicationDate":"2023-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44286225","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Modeling compound flood risk and risk reduction using a globally applicable framework: a pilot in the Sofala province of Mozambique","authors":"D. Eilander, A. Couasnon, F. Sperna Weiland, W. Ligtvoet, A. Bouwman, H. Winsemius, P. Ward","doi":"10.5194/nhess-23-2251-2023","DOIUrl":"https://doi.org/10.5194/nhess-23-2251-2023","url":null,"abstract":"Abstract. In low-lying coastal areas floods occur from\u0000(combinations of) fluvial, pluvial, and coastal drivers. If these flood\u0000drivers are statistically dependent, their joint probability might be\u0000misrepresented if dependence is not accounted for. However, few studies have examined flood risk and risk reduction measures while accounting for\u0000so-called compound flooding. We present a globally applicable framework for\u0000compound flood risk assessments using combined hydrodynamic, impact, and\u0000statistical modeling and apply it to a case study in the Sofala province of\u0000Mozambique. The framework broadly consists of three steps. First, a large\u0000stochastic event set is derived from reanalysis data, taking into account\u0000co-occurrence of and dependence between all annual maximum flood drivers.\u0000Then, both flood hazard and impact are simulated for different combinations\u0000of drivers at non-flood and flood conditions. Finally, the impact of each\u0000stochastic event is interpolated from the simulated events to derive a\u0000complete flood risk profile. Our case study results show that from all\u0000drivers, coastal flooding causes the largest risk in the region despite a\u0000more widespread fluvial and pluvial flood hazard. Events with return periods longer than 25 years are more damaging when considering the observed\u0000statistical dependence compared to independence, e.g., 12 % for the\u0000100-year return period. However, the total compound flood risk in terms of\u0000expected annual damage is only 0.55 % larger. This is explained by the\u0000fact that for frequent events, which contribute most to the risk, limited\u0000physical interaction between flood drivers is simulated. We also assess the\u0000effectiveness of three measures in terms of risk reduction. For our case,\u0000zoning based on the 2-year return period flood plain is as effective as\u0000levees with a 10-year return period protection level, while dry proofing up\u0000to 1 m does not reach the same effectiveness. As the framework is based on\u0000global datasets and is largely automated, it can easily be repeated for\u0000other regions for first-order assessments of compound flood risk. While the\u0000quality of the assessment will depend on the accuracy of the global models\u0000and data, it can readily include higher-quality (local) datasets where\u0000available to further improve the assessment.\u0000","PeriodicalId":18922,"journal":{"name":"Natural Hazards and Earth System Sciences","volume":" ","pages":""},"PeriodicalIF":4.6,"publicationDate":"2023-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44640044","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comprehensive landslide susceptibility map of Central Asia","authors":"A. Rosi, W. Frodella, Nicola Nocentini, Francesco Caleca, H. Havenith, A. Strom, Mirzo S. Saidov, Gany Amirgalievich Bimurzaev, V. Tofani","doi":"10.5194/nhess-23-2229-2023","DOIUrl":"https://doi.org/10.5194/nhess-23-2229-2023","url":null,"abstract":"Abstract. Central Asia is an area characterized by complex tectonics and active deformation; the related seismic activity controls the earthquake hazard level that, due to the occurrence of secondary and tertiary effects, also has direct implications for the hazard related to mass movements such as\u0000landslides, which are responsible for an extensive number of casualties\u0000every year. Climatically, this region is characterized by strong rainfall\u0000gradient contrasts due to the diversity of climate and vegetation zones.\u0000The region is drained by large, partly snow- and glacier-fed rivers that\u0000cross or terminate in arid forelands; therefore, it is also affected by a\u0000significant river flood hazard, mainly in spring and summer seasons. The\u0000challenge posed by the combination of different hazards can only be tackled by\u0000considering a multi-hazard approach harmonized among the different\u0000countries, in agreement with the requirements of the Sendai Framework for\u0000Disaster Risk Reduction. This work was carried out within the framework of\u0000the Strengthening Financial Resilience and Accelerating Risk Reduction in Central Asia (SFRARR) project as part of a multi-hazard approach and is\u0000focused on the first landslide susceptibility analysis at a regional scale\u0000for Central Asia. To this aim the most detailed landslide inventories,\u0000covering both national and transboundary territories, were implemented in a\u0000random forest model, together with several independent variables. The\u0000proposed approach represents an innovation in terms of resolution (from 30\u0000to 70 m) and extension of the analyzed area with respect to previous\u0000regional landslide susceptibility and hazard zonation models applied in\u0000Central Asia. The final aim was to provide a useful tool for land\u0000use planning and risk reduction strategies for landslide scientists,\u0000practitioners, and administrators.\u0000","PeriodicalId":18922,"journal":{"name":"Natural Hazards and Earth System Sciences","volume":" ","pages":""},"PeriodicalIF":4.6,"publicationDate":"2023-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46861805","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Scenario-based multi-risk assessment from existing single-hazard vulnerability models. An application to consecutive earthquakes and tsunamis in Lima, Peru","authors":"Juan Camilo Gomez- Zapata, M. Pittore, Nils Brinckmann, J. Lizarazo-Marriaga, S. Medina, N. Tarque, F. Cotton","doi":"10.5194/nhess-23-2203-2023","DOIUrl":"https://doi.org/10.5194/nhess-23-2203-2023","url":null,"abstract":"Abstract. Multi-hazard risk assessments for building portfolios\u0000exposed to earthquake shaking followed by a tsunami are usually based on\u0000empirical vulnerability models calibrated on post-event surveys of damaged\u0000buildings. The applicability of these models cannot easily be extrapolated\u0000to other regions of larger/smaller events. Moreover, the quantitative\u0000evaluation of the damages related to each of the hazard types\u0000(disaggregation) is impossible. To investigate cumulative damage on extended building portfolios, this study proposes an alternative and modular method to probabilistically integrate sets of single-hazard vulnerability models\u0000that are constantly being developed and calibrated by experts from various\u0000research fields to be used within a multi-risk context. This method is based\u0000on the proposal of state-dependent fragility functions for the triggered\u0000hazard to account for the pre-existing damage and the harmonisation of\u0000building classes and damage states through their taxonomic characterisation, which is transversal to any hazard-dependent vulnerability. This modular assemblage also allows us to separate the economic losses expected for each scenario on building portfolios subjected to cascading hazards. We\u0000demonstrate its application by assessing the economic losses expected for\u0000the residential building stock of Lima, Peru, a megacity commonly exposed to\u0000consecutive earthquake and tsunami scenarios. We show the importance of\u0000accounting for damage accumulation on extended building portfolios while\u0000observing a dependency between the earthquake magnitude and the direct\u0000economic losses derived for each hazard scenario. For the commonly exposed\u0000residential building stock of Lima exposed to both perils, we find that\u0000classical tsunami empirical fragility functions lead to underestimations of predicted losses for lower magnitudes (Mw) and large overestimations for larger Mw events in comparison to our state-dependent models and cumulative-damage method.\u0000","PeriodicalId":18922,"journal":{"name":"Natural Hazards and Earth System Sciences","volume":" ","pages":""},"PeriodicalIF":4.6,"publicationDate":"2023-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44484071","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A predictive equation for wave setup using genetic programming","authors":"Charline Dalinghaus, G. Coco, P. Higuera","doi":"10.5194/nhess-23-2157-2023","DOIUrl":"https://doi.org/10.5194/nhess-23-2157-2023","url":null,"abstract":"Abstract. We applied machine learning to improve the accuracy of present predictors of wave setup. Namely, we used an evolutionary-based genetic programming model and a previously published dataset, which includes various beach and wave conditions. Here, we present two new wave setup predictors: a simple predictor, which is a function of wave height, wavelength, and foreshore beach slope, and a fitter, but more complex predictor, which is also a function of sediment diameter. The results show that the new predictors outperform existing formulas. We conclude that machine learning models are capable of improving predictive capability (when compared to existing predictors) and also of providing a physically sound description of wave setup.\u0000","PeriodicalId":18922,"journal":{"name":"Natural Hazards and Earth System Sciences","volume":" ","pages":""},"PeriodicalIF":4.6,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48492664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Review article: A European perspective on wind and storm damage – from the meteorological background to index-based approaches to assess impacts","authors":"D. Gliksman, Paul Averbeck, N. Becker, B. Gardiner, V. Goldberg, J. Grieger, D. Handorf, K. Haustein, Alexia Karwat, F. Knutzen, H. Lentink, Rike Lorenz, Deborah Niermann, J. Pinto, Ronald Queck, A. Ziemann, C. Franzke","doi":"10.5194/nhess-23-2171-2023","DOIUrl":"https://doi.org/10.5194/nhess-23-2171-2023","url":null,"abstract":"Abstract. Wind and windstorms cause severe damage to natural and\u0000human-made environments. Thus, wind-related risk assessment is vital for the preparation and mitigation of calamities. However, the cascade of events leading to damage depends on many factors that are environment-specific and the available methods to address wind-related damage often require sophisticated analysis and specialization. Fortunately, simple indices and thresholds are as effective as complex mechanistic models for many applications. Nonetheless, the multitude of indices and thresholds available requires a careful selection process according to the target sector. Here, we first provide a basic background on wind and storm formation and characteristics, followed by a comprehensive collection of both indices and thresholds that can be used to predict the occurrence and magnitude of wind and storm damage. We focused on five key sectors: forests, urban areas, transport, agriculture and wind-based energy production. For each sector we described indices and thresholds relating to physical properties such as topography and land cover but also to economic aspects (e.g. disruptions in transportation or energy production). In the face of increased climatic variability, the promotion of more effective analysis of wind and storm damage could reduce the impact on society and the environment.\u0000","PeriodicalId":18922,"journal":{"name":"Natural Hazards and Earth System Sciences","volume":" ","pages":""},"PeriodicalIF":4.6,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49254295","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using machine learning algorithms to identify predictors of social vulnerability in the event of a hazard: Istanbul case study","authors":"Oya Kalaycıoğlu, Serhat Emre Akhanli, E. Menteşe, M. Kalaycıoğlu, S. Kalaycioglu","doi":"10.5194/nhess-23-2133-2023","DOIUrl":"https://doi.org/10.5194/nhess-23-2133-2023","url":null,"abstract":"Abstract. To what extent an individual or group will be affected by the damage of a hazard depends not just on their exposure to the event but on their social vulnerability – that is, how well they are able to anticipate, cope with, resist, and recover from the impact of a hazard. Therefore, for mitigating disaster risk effectively and building a disaster-resilient society to natural hazards, it is essential that policy makers develop an understanding of social vulnerability. This study aims to propose an optimal predictive model that allows decision makers to identify households with high social vulnerability by using a number of easily accessible household variables. In order to develop such a model, we rely on a large dataset comprising a household survey (n = 41 093) that was conducted to generate a social vulnerability index (SoVI) in Istanbul, Türkiye. In this study, we assessed the predictive ability of socio-economic, socio-demographic, and housing conditions on the household-level social vulnerability through machine learning models. We used classification and regression tree (CART), random forest (RF), support vector machine (SVM), naïve Bayes (NB), artificial neural network (ANN), k-nearest neighbours (KNNs), and logistic regression to classify households with respect to their social vulnerability level, which was used as the outcome of these models. Due to the disparity of class size outcome variables, subsampling strategies were applied for dealing with imbalanced data. Among these models, ANN was found to have the optimal predictive performance for discriminating households with low and high social vulnerability when random-majority under sampling was applied (area under the curve (AUC): 0.813). The results from the ANN method indicated that lack of social security, living in a squatter house, and job insecurity were among the most important predictors of social vulnerability to hazards. Additionally, the level of education, the ratio of elderly persons in the household, owning a property, household size, ratio of income earners, and savings of the household were found to be associated with social vulnerability. An open-access R Shiny web application was developed to visually display the performance of machine learning (ML) methods, important variables for the classification of households with high and low social vulnerability, and the spatial distribution of the variables across Istanbul neighbourhoods. The machine learning methodology and the findings that we present in this paper can guide decision makers in identifying social vulnerability effectively and hence let them prioritise actions towards vulnerable groups in terms of needs prior to an event of a hazard.\u0000","PeriodicalId":18922,"journal":{"name":"Natural Hazards and Earth System Sciences","volume":" ","pages":""},"PeriodicalIF":4.6,"publicationDate":"2023-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45369857","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}