Natural Hazards Research Research Articles

Risk assessment concerning urban ecosystem disservices: The example of street trees in Berlin, Germany

Urban green infrastructure, including street trees, is thought to play an important role in providing urban ecosystem services (UES) such as carbon sequestration and air quality improvement. The concept of urban ecosystem disservices (UEDS) is comparably new in landscape ecology. Research into UEDS assessment is low compared to that of UES. This paper demonstrates an approach to assess the distribution of UEDS risks caused by street trees within the city of Berlin. The approach adapts concepts from urban ecosystem service and natural hazard and risk research to demonstrate the importance of including context-specific vulnerability indicators as additional information in the assessment of UEDS risks. A cluster analysis is carried out to recognize patterns in UEDS risk distribution throughout the city. The paper demonstrates, that using context indicators in addition to biophysical indicators supports the assessment of UEDS as part of the relationship between humans and the environment.

How demographic developments determine the management of hydrometeorological hazard risks in rural communities: The linkages between demographic and natural hazards research

AbstractThe management of hydrometeorological hazard risks increasingly builds on the notion that the main constituents—hazard risk and vulnerability—have to be understood as dynamic processes. Natural hazards and their impacts are influenced by changes in land use and impacts of climate change, whereas the vulnerability of communities is, inter alia, determined by changes in their demographic composition. Structurally weak rural regions have to face these challenges, while their capacities to manage hydrometeorological risks are affected by major demographic developments, such as population decline and demographic aging (especially in Europe). Nonetheless, academic research regarding this issue is still in its infancy. Research on the relationship between hydrometeorological hazards and long‐term demographic change often focuses on one side of the coin: impacts of hazards on the demographic composition of communities. The impacts of demographic change onto communities' ability to prepare for, respond to and recover from natural hazards appear as disregarded link. This article gives an overview over linkages by reviewing literature at the intersection of demographic and hydrological hazard research. The review of more than 90 publications provides insights into how particular demographic factors (socioeconomic status; migration, residency, and mobility; gender and gender relations; education and knowledge; and religion and beliefs) determine the management of hydrometeorological hazard risks. Furthermore, it identifies publications which consider long‐term developments. The overview shows that we lack a thorough connection between demographic change and hydrometeorological hazard risk management, which would be crucial for in‐depth studies as well as a more dynamic understanding of respective linkages.This article is categorized under: Science of Water > Water Extremes Engineering Water > Planning Water

GIS-Based Integration of Social Vulnerability and Level 3 Probabilistic Risk Assessment to Advance Emergency Preparedness, Planning, and Response for Severe Nuclear Power Plant Accidents.

In the nuclear power industry, Level 3 probabilistic risk assessment (PRA) is used to estimate damage to public health and the environment if a severe accident leads to large radiological release. Current Level 3 PRA does not have an explicit inclusion of social factors and, therefore, it is not possible to perform importance ranking of social factors for risk-informing emergency preparedness, planning, and response (EPPR). This article offers a methodology for adapting the concept of social vulnerability, commonly used in natural hazard research, in the context of a severe nuclear power plant accident. The methodology has four steps: (1) calculating a hazard-independent social vulnerability index for the local population; (2) developing a location-specific representation of the maximum radiological hazard estimated from current Level 3 PRA, in a geographic information system (GIS) environment; (3) developing a GIS-based socio-technical risk map by combining the social vulnerability index and the location-specific radiological hazard; and (4) conducting a risk importance measure analysis to rank the criticality of social factors based on their contribution to the socio-technical risk. The methodology is applied using results from the 2012 Surry Power Station state-of-the-art reactor consequence analysis. A radiological hazard model is generated from MELCOR accident consequence code system, translated into a GIS environment, and combined with the Center for Disease Control social vulnerability index (SVI). This research creates an opportunity to explicitly consider and rank the criticality of location-specific SVI themes based on their influence on risk, providing input for EPPR.

Revisiting Saarinen’s “Perception of the drought hazard on the Great Plains”

This piece revisits Saarinen’s 1966 work on farmers’ perceptions of the drought hazard on the Great Plains. A synopsis of the work is provided as well as a discussion of the lasting influence that this work has had on natural hazards research.

The importance of supervisory and organisational awareness of the risks for an early career natural hazard researcher with personal past-disaster experience

Higher education students are advised that research must be objective, focusing on research outcomes and a strict set of criteria upon which the student will be examined. Yet, for those undertaking natural hazard research, this advice is tested, particularly if the student themselves were also impacted by the same natural hazard. Such was the experience of this author. This paper reflects on the author's personal natural hazard experiences, and the drivers that resulted in the pursuit of a higher education degree which involved the phenomenological investigation of the impact of these disasters on the lives of those residing in isolated exposed locations on the Far North Queensland coast in Australia. The aim of this paper is to highlight the importance of supervisors and organisations charged with ensuring the students well-being understand why a student wishes to pursue emotion-charged research. The article concludes with recommendations that supervisors and Ethics Committees focus not only on the risks to participants, but also to the students themselves, by ensuring they also consider the student's own past-disaster experiences.

Interpretations of Resilience and Change and The Catalytic Roles of Media: A Case of Canadian Daily Newspaper Discourse on Natural Disasters.

The varied interpretations of the concept of resilience in natural hazards research literature has attracted numerous criticisms. A common criticism centers around a poor understanding of the changes caused by natural disasters by the research stream. Considering resilience as a metaphor of change, and newspaper as a catalyst that often highlights post-disaster opportunities for "forward looking" (rather than bouncing back) changes, we examined some specific aspects of change in Canadian communities by analyzing coverage of natural disasters in daily newspapers. We posit that post-disaster newspaper discourse on resilience and change can not only assist enhancing academic inquiries on resilience but also contribute to improving practices for transformative changes in post-disaster contexts. We adopted a social constructivist approach to analyzing newspaper discourse, using the ProQuest database to find articles from the 1996-2017 period. The findings exhibited a trend of the increased use of narratives on resilience in Canadian newspapers since the 1990s that substantiates the hypothesis that transformative change in the personal and practical spheres requires alteration of peoples' attitude, behavior, and thinking toward environmental risks. The discourse emphasized incremental changes at the policy level: (i) to improve response and recovery, and (ii) to address the needs of vulnerable and disaster-affected population. Our findings overall underscore the importance of documentation and efforts towards streamlining learning; application of learning at multiple interconnected levels for progressive changes to enhance community resilience, and the need for building consensus among academicians, practitioners and policy makers regarding the meaning and use of the concept of resilience.

TerraSAR-X/TanDEM-X data for natural hazards research in mountainous regions of Uzbekistan

Accurate and up-to-date digital elevation models (DEMs) are important tools for studying mountain hazards. We considered natural hazards related to glacier retreat, debris flows, and snow avalanches in two study areas of the Western Tien-Shan mountains, Uzbekistan. High-resolution DEMs were generated using single TerraSAR-X/TanDEM-X datasets. The high quality and actuality of the DEMs were proved through a comparison with Shuttle Radar Topography Mission, Advanced Spaceborne Emission and Reflection Radiometer, and Topo DEMs, using Ice, Cloud, and Land Elevation Satellite data as the reference dataset. For the first study area, which had high levels of economic activity, we applied the generated TanDEM-X DEM to an avalanche dynamics simulation using RAMMS software. Verification of the output results showed good agreement with field observations. For the second study area, with a wide spatial distribution of glaciers, we applied the TanDEM-X DEM to an assessment of glacier surface elevation changes. The results can be used to calculate the local mass balance in glacier ablation zones in other areas. Models were applied to estimate the probability of moraine-dammed lake formation and the affected area of a possible debris flow resulting from glacial lake outburst. The natural hazard research methods considered here will minimize costly ground observations in poorly accessible mountains and mitigate the impacts of hazards on the environment of Uzbekistan.

Living in areas with different levels of earthquake damage and association with risk of cardiovascular disease: a cohort-linkage study

Cardiovascular disease rates are known to increase immediately after a severe earthquake. However, less is known about the magnitude of this increase over time in relation to the amount of housing damage. We assessed the effect of area housing damage from a major earthquake sequence in Christchurch, Canterbury province, New Zealand, on cardiovascular disease-related hospital admissions and deaths. For this cohort-linkage study, we used linked administrative datasets from the Statistics New Zealand Integrated Data Infrastructure to identify individuals aged 45 years or older living in Christchurch from the date of the first earthquake on Sept 4, 2010. Individuals were assigned the average damage level for their residential meshblock (small neighbourhood generally comprising 10-50 dwellings) using the insurance-assessed residential building damage costs obtained from the Earthquake Commission as a proportion of property value. We calculated the rates of cardiovascular disease-related hospital admissions (including myocardial infarction) and cardiovascular disease-related mortality and rate ratios (adjusted for age, sex, ethnicity, small-area deprivation index, and personal income) by level of housing damage in the first year and the 4 subsequent years after the earthquake. The rate ratio association between earthquake housing damage and cardiovascular event was examined by Poisson regression, and linear test of trends across damage categories was done by regression modeling. We identified 179 000 residents living in the earthquake-affected region of Christchurch, of whom 148 000 had complete data. For the first 3 months after the Feb 22, 2011 earthquake, the Poisson regression-adjusted rate ratio (RR) for cardiovascular disease-related hospital admissions for residents from areas that were most damaged (compared with residents from the least damaged areas) was 1·12 (95% CI 0·96-1·32; test for linear trend p=0·239). In the first year after the earthquake sequence, for residents from areas that were most damaged (vs the least damaged areas), Poisson regression-adjusted RRs were 1·10 (1·01-1·21; test for linear trend p=0·068) for cardiovascular disease-related hospital admissions, 1·22 (1·00-1·48; p=0·036) for myocardial infarction-related hospital admissions, and 1·25 (1·06-1·47; p=0·105) for cardiovascular disease-related mortality, corresponding to an excess of 66 (95% CI 7-125) cardiovascular disease-related hospital admissions, including 29 (0-53) additional myocardial infarction-related hospital admissions and 46 (13-73) additional deaths from cardiovascular disease. In the 4 subsequent years, we found no evidence of an association of these outcomes with earthquake damage. Rates of cardiovascular disease and myocardial infarction were increased in people living in areas with more severely damaged homes in the first year after a major earthquake. Policy responses to reduce the effect of earthquake damage on cardiovascular disease could include pre-earthquake measures to minimise building damage, early wellbeing interventions within the first year to address post-earthquake stress, and enhanced provision of cardiovascular disease prevention and treatment services. Healthier Lives National Science Challenge and Natural Hazards Research Platform, Ministry of Business, Innovation and Employment.

Scientist and stakeholder perspectives of transdisciplinary research: Early attitudes, expectations, and tensions

Transdisciplinary approaches are becoming increasingly adopted as a way to research complex socio-environmental problems. Conceptually, transdisciplinarity aims to foster meaningful knowledge co-production through integrative and participatory processes that bring together diverse actors, disciplines, and knowledge bases. In practice, transdisciplinarity is more ambiguous. While there is a growing body of literature on such approaches, there remains no widely-accepted definition, concrete framework, or empirical strategy for how to carry out a transdisciplinary project. We propose that this lack of explicit structure and entrenched meaning leaves space for transdisciplinary approaches to be shaped by the evolving network of participating scientists and stakeholders, according to their perspectives of the approach and what it embodies. Here, we examine the perspectives of a diverse group of actors (n=42) embarking on a 10-year transdisciplinary research project focused on building resilience to natural hazards and disasters in New Zealand. We present the findings of qualitative surveys and group interviews that investigate stakeholders’ and scientists’ early perspectives of transdisciplinary, or co-created, research. The study represents the first stage of longitudinal research that will continue over the course of the project. Results show that early actors in the project share an overall consistent understanding of co-created research. Participants described a process that integrated diverse people and knowledge; created benefits on both a social and personal level; fostered clear, two-way dialogue; and overcame pragmatic and intrinsic challenges. Collectively, participants agreed with adopting transdisciplinary approaches to natural hazard, risk, and resilience research, with stakeholders showing a stronger degree of agreement than scientists. While attitudes towards transdisciplinarity were overall positive, a number of underlying conflicts emerged in regards to carrying out new modes of knowledge production within traditional social and institutional structures. These conflicts result in a tension that is felt by actors involved in transdisciplinary projects early on, and in some cases, influences perception of their ability to fully participate in such an approach. Evaluating actor perspectives and expectations early in the transdisciplinary process can give insight into how attitudes, expectations, and conflicts might shape transdisciplinary efforts, and can provide relevant parameters for assessing change over time.

Message from the Winner

I am honored to receive the JDR Award and am very grateful to the editors and staff members of the Journal of Disaster Research. I am also deeply indebted to the authors who contributed their excellent papers to the two special issues that I had the privilege of guest editing. One issue focused on tsunami forces and effects on structures (Vol.4 No.6) and the other on uncertainties in tsunami effects (Vol.11 No.4). I organized these special issues in cooperation with professors Shuto1 and Sato2, without whom I could not have achieved such fruitful special issues. So this award must be shared with both professor Shuto and professor Sato. The Journal of Disaster Research was founded in 2006, immediately after two impactful disasters: the 2004 Great Indian Ocean Earthquake and Tsunami, and the 2005 Hurricane Katrina. The founding of the journal was very timely, and I was inspired when I was invited to serve as an international member of the editorial board. Recognizing that research in natural hazards must be trans-disciplinary and interwoven with expertise in geophysics, engineering, and social sciences, I felt it was difficult to locate a proper arena for reporting research findings. And because natural hazards do not recognize national borders, I felt it was crucial that research be conducted with close international collaboration. In the Journal’s early days, it published collections of papers originating mostly from Japanese research activities. But contributions from other countries have increased dramatically, and today the Journal is in the process of establishing an exceptional reputation, attracting truly outstanding research articles from around the world. The Journal of Disaster Research now serves as an excellent dissemination outlet for cutting-edge natural hazard and disaster research, and I intend to continually contribute to this exemplary international journal.

The Morphotectonic units of Ikaria island – Contribution in the natural hazards research

Present morphological and tectonic image of Ikaria island is the complex result of both the tem-poral evolution of the island, as part of the geodynamical evolution of the Hellenic Arc, especially that of the last stages (from the upper Miocene and after), and the exogenous factors. Also human impact functions as an additional parameter for the formation of Ikaria geoenvironment.According to field data and their analysis, computing and estimation which took part with the use of modern technologies (use of GIS, analysis of geographical and descriptive databases) the mor-photectonic units of Ikaria Island were defined. The definition of the morphotectonic units was based on their special features and more precisely on the lithological differentiation (granite, metamorphic rocks and post alpine formations) as well as the complex ductile and brittle structures (large scale ductile shear-zones and brittle detachment faults, normal faults, strike-slip transfer faults and join sets) and the morphological features (slope distribution, asymmetry of the drainage system, planation surfaces, morphological discontinuities etc).The synthesis of the special features of each morphotectonic unit in combination with the human impact (fires, overgrazing etc), defines the kind, the intensity and the geographical localization of the natural hazards which occur in various sites of the island, such as landslides, erosion, floods, seismic activity, relief changes, coastline changes etc.

Decision Analysis for Management of Natural Hazards

Losses from natural hazards, including geophysical and hydrometeorological hazards, have been increasing worldwide. This review focuses on the process by which scientific evidence about natural hazards is applied to support decision making. Decision analysis typically involves estimating the probability of extreme events; assessing the potential impacts of those events from a variety of perspectives; and evaluating options to plan for, mitigate, or react to events. We consider issues that affect decisions made across a range of natural hazards, summarize decision methodologies, and provide examples of applications of decision analysis to the management of natural hazards. We conclude that there is potential for further exchange of ideas and experience between natural hazard research communities on decision analysis approaches. Broader application of decision methodologies to natural hazard management and evaluation of existing decision approaches can potentially lead to more efficient allocation of scarce resources and more efficient risk management.

Advancing Climate Change Adaptation: Borrowing from the Natural Hazard Research (NHR) Tradition in Geography

Extreme events have widespread impact on human and natural systems, and the response to such extreme events is examined by several academic and practicing communities. For decades, the Natural Hazard Research (NHR) tradition in geography, led by White and expanded by scholars like Mitchell, have looked at possibilities of blending and applying technological fixes and societal adjustments to reduce the risk of losses from extreme events. While knowledge about potential impacts of climate change related to extreme events has grown substantially, the consolidation of knowledge to plan for and respond to extreme climate events in the short and long terms has only begun in the past few years. These scholars have emphasized the importance of context, understanding and including the broader range of options, and systematic study of past experiences to elicit knowledge pertaining to effective societal responses to extreme events. Such lessons from the NHR tradition and the works of Mitchell are key to finding the way forward with climate change adaptation.

Role of Boundary Organization after a Disaster: New Zealand’s Natural Hazards Research Platform and the 2010–2011 Canterbury Earthquake Sequence

AbstractThe boundary organization concept has been used to establish that collaborative arrangements and outputs across science and policy domain boundaries need to be credible, relevant, and legitimate in order to be to be effective. Although widely accepted in other issue-driven fields, this concept does not have equivalent currency in the natural hazard and disaster risk reduction context. This paper uses the development of the New Zealand Natural Hazards Research Platform during a recent earthquake disaster to assess the utility of the concept in this topic area. Lessons are also identified concerning the use of larger consortium organizations to increase policy and other end-user involvement in the management and coordination of research funding, and the impact of a major disaster on this research-funding initiative. Mapping the Platform’s collaborative arrangements in relation to boundary tensions over time makes it possible to distinguish disaster effects from preexisting and ongoing structural eff...

Research Engagement after Disasters: Research Coordination before, during, and after the 2011–2012 Canterbury Earthquake Sequence, New Zealand

This article argues that active coordination of research engagement after disasters has the potential to maximize research opportunities, improve research quality, increase end-user engagement, and manage escalating research activity to mitigate the ethical risks posed to impacted populations. We focus on the coordination of research activity after the 22 February 2011 Mw6.2 Christchurch earthquake by the then newly formed national research consortium, the Natural Hazards Research Platform, which included a social science research moratorium during the declared state of national emergency. Decisions defining this organization's functional and structural parameters are analyzed to identify lessons concerning the need for systematic approaches to the management of post-disaster research, in collaboration with the response effort. Other lessons include the importance of involving an existing, broadly based research consortium, ensuring that this consortium's coordination role is fully integrated into emergency management structures, and ensuring that all aspects of decision-making processes are transparent and easily accessed.

The co-production of risk from a natural hazards perspective: science and policy interaction for landslide risk management in Italy

Despite continuing technological advancement in hazard and vulnerability assessment, risk modelling and hazard mitigation techniques, losses to disasters associated with natural hazards continue and in some cases are increasing across Europe and worldwide. This paper focuses on the need to bridge the gap between technical solutions and the sociopolitical contexts in which these are produced, to better understand and create more effective risk management regimes. We do so with application of the science–policy co-production frame to landslide risk management in Italy. The methodology deployed included a desk study informed by semi-structured interviews carried out with selected key stakeholders at national, regional and municipal level. We propose a normative and analytical framework for transferring co-production into natural hazard research by presenting a matrix identifying four contexts within which co-production may unfold. The matrix is based on two axes, which distinguish between innovation and its absence in science and policy domains. We examine several examples of co-production, such as the water–soil integrated approach to risk management or the implementation of hazard/risk assessment. The results highlight that the insulation of science from the institutional context within which knowledge is produced and used is a very problematic issue. This often hinders the implementation of desirable policies and undermines the effectiveness of interventions. Moreover, innovation in science and policy does not automatically result in successful solutions for landslide risk management. Finally, results confirm the utility of co-production but also highlight methodological challenges associated with the introduction of this new conceptual paradigm into the well-structured communities of scientists and policy-makers.

The impact of flooding on people living with HIV: a case study from the Ohangwena Region, Namibia

BackgroundFloods are a disaster situation for all affected populations and especially for vulnerable groups within communities such as children, orphans, women, and people with chronic diseases such as HIV and AIDS. They need functioning health care, sanitation and hygiene, safe water, and healthy food supply, and are critically dependent on their social care and support networks. A study carried out in the Ohangwena region, Namibia, where HIV prevalence is high and extensive flooding frequently occurs, aims to provide a deeper understanding of the impact that flooding has on people living with HIV (PLWHIV) as well as on HIV service providers in the region.DesignThe qualitative research applying grounded theory included semi-structured interviews with PLWHIV, focus group discussions with HIV service providers, and a national feedback meeting. The findings were interpreted using the sustainable livelihoods framework, the natural hazard research approach, and health behaviour theories.ResultsThe study reveals that flooding poses major problems to PLWHIV in terms of their everyday lives, affecting livelihoods, work, income, and living conditions. The factors threatening them under normal conditions – poverty, malnutrition, unsafe water, sanitation and hygiene, limited access to health facilities, a weak health status, and stigma – are intensified by flood-related breakdown of infrastructure, insecurity, malnutrition, and diseases evolving over the course of a flood. A potential dual risk exists for their health: the increased risk both of infection and disease due to the inaccessibility of health services and antiretroviral treatment. A HIV and Flooding Framework was developed to display the results.ConclusionsThis study demonstrates that vulnerabilities and health risks of PLWHIV will increase in a disaster situation like flooding if access to HIV prevention, treatment, care and support are not addressed and ensured. The findings and the HIV and Flooding Framework are not specific to Ohangwena and can be transferred to any flood-affected region that has a high HIV prevalence and relies mainly on subsistence agriculture. They serve as a model case for analysing vulnerabilities related to health and health service provision under disaster conditions. The impact will vary according to the physical, geographical, climatological, social, and behavioural characteristics of the region and the people affected. In the Ohangwena region, a disaster risk management mechanism is already in place which addresses people with HIV during flooding. However, preparedness could be improved further by applying the HIV and Flooding Framework.

Preface "Hillslope hydrological modelling for landslides prediction"

Looking at natural hazards and risks, such as landslides, debris flow and floods, we see a broad range of science fields involved: from earth science to engineering science, to management, policy and social science. Although research on natural hazards and risks contains mono-disciplinary aspects, it is best characterized by its multi-disciplinary nature. Moreover, the link to society has always been a very important driver in natural hazards research, pushed by the need for reliable prediction models to be implemented into early warning systems as well as by the required complete understanding of the physical mechanisms for the design of mitigation works. If we focus on landslide and debris flows as in this special issue of HESS, we will see impressive progress in the geotechnical and slope stability modelling needed for hazard analysis. Similarly, the risk management aspects, such as hazard mapping and risk assessment, have received much attention. Hydrology is an important aspect of landslide and debris flow assessment. Precipitation and snowmelt water infiltration, leading to local pore water pressure increase and/or matric suction decrease, is amongst the most common triggers of landslides. Fundamental knowledge about underlying processes affecting this infiltration process, such as macropore and fissure flow, water repellency, soil structure, soil–plant–atmosphere interactions as well as the effects of land use practices (e.g. deforestation, terracing, grazing), has strongly improved in the last decade or so, with a clear focus on more detailed knowledge of hydrological process dynamics. The hydrological process understanding progressed rapidly under pressure of societal needs such as prediction of discharge generation and contaminant transport, to name a few. Although hydrology research is very strongly linked to natural hazards, such as landslides and debris flows, this improved hydrological knowledge has found its way into the landslide community rather modestly. In particular, the incorporation of hydrological processes into large-scale models is still incomplete and their application to landslide prediction limited. Landslide research tends to be more focused on novel methods to include spatial data and on the practical applicability of, e.g. landslide triggering modelling and statistical analyses for regional hazard and risk assessment. However, without stating that these fields have been fully exploited, we see that in our quantitative landslide and debris flow modelling the inclusion of increased process knowledge seems to lag behind. This in depth process understanding needs to be incorporated in our technical predictions in order to improve the reliability of early warning systems, mitigation works and landslide zonation. This special issue aims to present innovative hydrological research applied to landslide studies to improve the understanding of the spatio-temporal patterns of slope movement mechanisms induced by precipitation. The initiative struck a sympathetic note, as many colleagues were facing these challenges. The topic of hydrology and landslides is finding more and more space in hydrological research, as witnessed by the number of relevant papers published, in special issues, e.g. on hydrology of unstable clay shales (Bogaard et al., 2012) and through the organization of workshops and special conference sessions, such as six EGU sessions on hydrology and landslides, three editions of the Italian Workshop on Landslides mainly dedicated to landslide hydrology (www.iwl.unina2.it), and special sessions at IAEG2014. Therefore, this Special Issue discusses the representation of

Bayesian network learning for natural hazard analyses

Abstract. Modern natural hazards research requires dealing with several uncertainties that arise from limited process knowledge, measurement errors, censored and incomplete observations, and the intrinsic randomness of the governing processes. Nevertheless, deterministic analyses are still widely used in quantitative hazard assessments despite the pitfall of misestimating the hazard and any ensuing risks. In this paper we show that Bayesian networks offer a flexible framework for capturing and expressing a broad range of uncertainties encountered in natural hazard assessments. Although Bayesian networks are well studied in theory, their application to real-world data is far from straightforward, and requires specific tailoring and adaptation of existing algorithms. We offer suggestions as how to tackle frequently arising problems in this context and mainly concentrate on the handling of continuous variables, incomplete data sets, and the interaction of both. By way of three case studies from earthquake, flood, and landslide research, we demonstrate the method of data-driven Bayesian network learning, and showcase the flexibility, applicability, and benefits of this approach. Our results offer fresh and partly counterintuitive insights into well-studied multivariate problems of earthquake-induced ground motion prediction, accurate flood damage quantification, and spatially explicit landslide prediction at the regional scale. In particular, we highlight how Bayesian networks help to express information flow and independence assumptions between candidate predictors. Such knowledge is pivotal in providing scientists and decision makers with well-informed strategies for selecting adequate predictor variables for quantitative natural hazard assessments.

A multi-criteria evaluation model of earthquake vulnerability in Victoria, British Columbia

Researchers have recently examined the geographic variability of the vulner- ability of populations to earthquakes. These studies focus mainly on the complex mod- elling of geophysical processes or identification of socio-economically disadvantaged populations. However, no studies to date have integrated different components of vul- nerability with metrics of travel distance to hospitals and trauma centres (systemic vul- nerability). We argue that this previously unaccounted component is an important conceptual and practical aspect of earthquake vulnerability. Accordingly, this paper pre- sents a multi-criteria model for combining physical, social, and systemic components, moving towards a more comprehensive assessment of vulnerability. An analytic hierarchy process is used to produce a place-specific index of social vulnerability, which we combine with soil liquefaction and amplification index and a road network model for access to hospitals and trauma services. Using a geographic information system, we implemented this model for the Greater Victoria region (483 km 2 , 2011 population: 345,000) in British Columbia, Canada. Clustering of total vulnerability was found in outlying areas, high- lighting the importance of access to trauma centres. We conclude by identifying challenges in measuring earthquake vulnerability and advocate integration of systemic vulnerability components in natural hazards research.