A Comparison of Hazard Vulnerability Indexes for Washington State

Tim Sheehan; Esther Min; Jeremy Hess

doi:10.1515/jhsem-2021-0066

Artikel Open Access

A Comparison of Hazard Vulnerability Indexes for Washington State

Tim Sheehan , Esther Min und Jeremy Hess

Veröffentlicht/Copyright: 27. März 2023

Veröffentlicht von

Veröffentlichen auch Sie bei De Gruyter Brill

Manuskript einreichen Informationen für Autor*innen Erkunden Sie dieses Fachgebiet

Aus der Zeitschrift Journal of Homeland Security and Emergency Management Band 20 Heft 2

Abstract

Factors associated with structural racism, disenfranchisement, poverty, and other persistent sources of inequity are associated with vulnerability and exposure to environmental hazards. Social, demographic, and environmental factors associated with vulnerability to environmental hazards have been used by many researchers to produce indexes of hazard vulnerability. In preparation for a climate change health risk assessment for Washington state, we compared methods and results from six indexes designed to support environmental health risk assessment. Production of these indexes varies in the number of variables considered, calculation complexity, and exposure of local causal pathways. Results for these indexes are generally very similar, especially at the highest decile of vulnerability, the exception being the Environmental Health Disparities index, the only one to consider hazard exposure. Some indexes used methods that hide causal pathways. Those that exposed causal pathways limited model structure. Results indicate that simpler indexes may be more appropriate for use in decision support tools as they require less overhead for data updates and scenario analysis and that other methodologies may provide a more useful framework for index generation.

Keywords: vulnerability index; hazard; climate change; decision support; Washington State

1 Introduction

Social and demographic factors play important roles in exposure to and adverse impacts from environmental hazards (Cutter, Boruff, and Shirley 2003). How these factors interact with hazard exposure to affect risk distributions is important for targeting risk reduction efforts and evaluating the success of these efforts over time (Bouwer 2019; Mechler and Bouwer 2015).

As our understanding of factors conferring vulnerability to hazardous environmental exposures has developed, certain consistent patterns have emerged. Unsurprisingly, poverty is associated with increased exposure and susceptibility through multiple pathways (Adger 2006; Philip and Rayhan 2004; Thomalla et al. 2006; Tellman et al. 2020). Factors associated with structural racism, disenfranchisement, and other persistent sources of inequity drive these relationships. Vulnerability factors and built environment factors associated with increased hazard exposure frequently overlap. These relationships have evolved over time and have been reinforced by the dynamics associated with structural inequities (Borden et al. 2007; Fuchs et al. 2019; Zahran et al. 2008).

Research has produced a large volume of information on social, demographic, and environmental factors associated with vulnerability to environmental hazards. Indexes, composite measures developed through a process of weighting, aggregation, and robustness checking (Greco et al. 2019), are often used to measure the central influences among these factors. Index creation focuses on certain high priority data elements and relationships, providing conceptually tractable explanations of observed relationships. This allows consistent measurement over time and supporting generalization of relationships from one context to another.

Indexes of social vulnerability to environmental hazards vary in their relative weighting of various factors and their methodologies (Schmidtlein et al. 2008; Yoon 2012). This variation is driven, at least in part, by the processes and communications that the indexes are designed to support (de Oliveira Mendes 2009), as well as by differences in the relative importance of vulnerability factors for specific hazards (Flanagan et al. 2018). Some indexes are meant to support risk assessment for multiple hazards on a large geographic scale and tend to include least common risk denominators, while others are more specific to particular hazards or local drivers.

This variability can complicate risk assessment and decision support, particularly over time, as relationships between social vulnerability factors and hazards are dynamic, and vulnerability is somewhat hazard specific. This is of particular concern in risk assessment related to climate change, which drives a wide range of hazards. In locations where climate change mitigation is expected to reduce fossil fuel pollution, exposure to air pollution hazards is likely to decrease over time as precursor emissions drop, but exposure to climate-sensitive hazards such as heat (Honda et al. 2014), flooding (Kundzewicz et al. 2014), and wildland fires (Abatzoglou and Williams 2016), among other exposures (McDermott-Levy et al. 2021) is likely to increase. Information about the correlations between various vulnerability Indexes can provide insight into the relevance of various Indexes for risk assessment.

As part of a broader effort to develop an interactive climate health risk tool for decision support, we are creating a set of climate health risk models for Washington State. Priorities for an efficient, useable, and useful tool include transparency of location-specific causal pathways for risk, ease of index computation, and limiting the number of required variables. To explore the relationships between these priorities in existing indexes, we surveyed six social vulnerability indexes and their results for Washington State: the Environmental Health Disparity (EHD) index, the Washington Social Vulnerability to Hazards (SVH) index, the Socioeconomic data layer from the EHD (EHDSE), the Social Vulnerability Index for the United States (SoVI^®), the Centers for Disease Control Social Vulnerability Index (SVI), and the Social Deprivation Index (SDI).

2 Methods

2.1 Study Area

Our study area comprises Washington State subdivided into 2010 census tracts (Figure 1).

Figure 1:

Washington State census tracts with inset of Puget Sound area census tracts.

2.2 The Indexes

The Environmental Health Disparities (EHD) map (Min et al. 2019) was created specifically for Washington State (WA). It is based on the CalEnviroScreen tool (2018). It is incorporated into the WA Department of Health (DOH) WA Tracking Network (WTN) Information by Location (IBL) tool (https://fortress.wa.gov/doh/wtn/WTNIBL/). EHD takes into account socioeconomic factors (low educational attainment, housing burden, linguistic isolation, poverty, race, transportation expense, and unemployment), health issues making populations sensitive (cardiovascular disease and low birthweight), environmental effects (lead risk and exposure, proximity to hazardous waste generators and facilities, proximity to Superfund sites, proximity to facilities with highly toxic substances, and wastewater discharge), and environmental exposures (diesel emissions, ozone, particulate matter 2.5 μm and smaller (PM2.5), toxic releases from facilities, and traffic density). EHD implements a three-tiered hierarchical model using means, weighted means, and products to produce decile relative rankings. Data used in this study were directly downloaded from the WTN website, where data from each factor used in the model can be viewed and explored interactively. EHD computes the rankings for combined socioeconomic factors as a separate data layer. We consider this as its own index (EHDSE for Environmental Health Disparities Socioeconomic) to compare with other indexes in this study.

The Social Vulnerability to Hazards (SVH) index was created by WA DOH and is available through the WTN IBL tool. Factors considered in this index are grouped as socioeconomic (access to private vehicle, high school diploma, population living in poverty, and unemployment), housing (housing with 10 or more units, mobile home occupancy, and overcrowded housing), and household (limited English, persons 65 and older living alone, disability, and single parent households). Calculation of this index is through multitiered means and weighted means. Data used in this study were directly downloaded from the WTN website, where data from each factor used in the model can be viewed and explored interactively.

The Social Vulnerability Index for the United States (SoVI^®) (Cutter, Boruff, and Shirley 2003) is a comparative metric that “measures the social vulnerability of U.S. counties to environmental hazards (http://artsandsciences.sc.edu/geog/hvri/sovi/)”. It is a product of the Hazards & Vulnerability Research Institute (VRI; http://artsandsciences.sc.edu/geog/hvri/) at the University of South Carolina. SoVI uses a set of socioeconomic variables to compute the index, and the variables used have changed over time. The current list includes variables regarding racial makeup, age, children in two-parent families, income level, social security recipients, female population, female head of household, access to healthcare, per capita nursing home residents, English as a second language, high school education, housing structure and cost, labor types, female participation in labor force, and vehicle ownership (http://artsandsciences.sc.edu/geog/hvri/sovi%C2%AE-evolution). SoVI is computed using a factor analysis. It is recommended that it be calculated over the geography of concern. For this study we used SoVI calculated by the National Oceanic and Atmospheric Administration (NOAA) for WA using 2010 census data (https://coast.noaa.gov/digitalcoast/data/sovi.html).

The Centers for Disease Control (CDC)/Agency for Toxic Substances and Disease Registry (ATSDR) Social Vulnerability Index (SVI) (Flanagan et al. 2011) is intended to “help local officials identify communities that may need support before, during, or after disasters (https://www.atsdr.cdc.gov/placeandhealth/svi/index.html).” Its method calculates values for each of four themes using associated census variables (in parentheses): socioeconomic status (below poverty, unemployment, income, high school diploma), household composition and disability (age 65 and older, age 17 and younger, over age 5 with disability, single-parent households), minority status and language (minority, speaks English less than well), and housing type and transportation (multi-unit structures, mobile homes, crowding, no vehicle, group quarters). For each variable, a percentile ranking over a 0 to 1 range is calculated. For each theme, the percentile rankings are summed, and the sums percentile ranked. The theme scores are then summed and the result percentile ranked for the overall score. For this study we downloaded rankings calculated using 2018 data for WA (https://www.atsdr.cdc.gov/placeandhealth/svi/data_documentation_download.html). Ranks were calculated among census tracts within the state.

Developed by the Robert Graham Center (RGC), the Social Deprivation Index (SDI; https://www.graham-center.org/rgc/maps-data-tools/sdi/social-deprivation-index.html) is “a composite measure of area level deprivation based on seven demographic characteristics collected in the American Community Survey.” The seven characteristics are percent of population: living under the federal poverty level, 25 years or older with less than 12 years of education, non-employed, living in renter-occupied and crowded housing units, living in crowded housing units, single parent households, with no car. Characteristics used in the survey were culled from a larger number of characteristics by discarding those with a loading of less than 0.60 in a factor analysis. The index is then calculated using the factor loadings for the remaining characteristics as weights. For this project we extracted values for WA from the 2015 census tract level SDI data.

2.3 Comparisons

We compared published results for six different hazard and health vulnerability indexes. Because three of these indexes are decile rank based (EHD, EHDSE, SVH) we converted each index into a decile ranking over the study area to compare them. We rendered a map of each index for census tracts over the entire state. Between each pair of indexes, we computed Spearman’s rho, which calculates correlation between rank values, and generated heat maps to indicate which rankings correlate more and less strongly. We also generated decile rankings for population density for the state and used the same method to compare each index’s rankings to state population ranking.

3 Results

Correlations among the non-EHD indexes are greater (Spearman’s rho 0.639–0.881; Table 1) than between EHD and any other index (Spearman’s rho 0.454–0.627). Non-EHD decile ranks match more closely with each other (Figure 2, second through fourth columns) than they do with EHD (Figure 2, first column). Matching is greatest at the highest and lowest decile ranks among non-EHD indexes, and at the highest decile between non-EHD indexes and EHD (Table 2).

Table 1:

Indexes reviewed in this study.

Index	Created by	Data source	Number of variables used
EHD	WA DOH	https://fortress.wa.gov/doh/wtn/WTNIBL/	19
SVH	WA DOH	https://fortress.wa.gov/doh/wtn/WTNIBL/	11
SoVI	VRI	https://coast.noaa.gov/digitalcoast/data/sovi.html	30
SVI	CDC	https://www.atsdr.cdc.gov/placeandhealth/svi/data_documentation_download.html	15
SDI	Robert graham center	https://www.graham-center.org/content/dam/rgc/documents/maps-data-tools/sdi/ACS2015_CTallvars.xlsx	7
EHDSE	WA DOH	https://fortress.wa.gov/doh/wtn/WTNIBL/	7

Figure 2:

Heat maps comparing index rankings for census tracts. Each cell in the graph represents a combination of ranks from the two models being compared. The shade of the cell indicates how many census tracts have the combination of ranks. Cells falling along the diagonal line (matching value line) have identical rank values in both index rankings. Darker shades towards the diagonal line indicate greater agreement between the two index rankings.

Table 2:

Spearman correlation coefficients between ranks of each pair of indexes (p-value below 2e⁻⁷⁴ for all pairs).

Index	EHD	SVH	SoVI	SVI	SDI	EHDSE
EHD	1	0.557	0.454	0.565	0.613	0.627
SVH	0.557	1	0.725	0.835	0.881	0.791
SoVI	0.454	0.725	1	0.746	0.696	0.639
SVI	0.565	0.835	0.746	1	0.857	0.852
SDI	0.613	0.881	0.696	0.857	1	0.843
EHDSE	0.627	0.791	0.639	0.852	0.843	1

Correlation with population density is much higher for EHD (Spearman’s rho 0.470; Table 3, Figure 3) than for non-EHD indexes (Spearman’s rho 0.097–0.262). All indexes indicate strong matching at the highest decile. EHD also shows strong matching at the lowest two deciles, while SoVI, SVI, and EHDSE show very strong mismatching at the lowest index decile (Figure 3).

Table 3:

Spearman correlation coefficients between each index and population density rank (p-value below 2.5e⁻⁴ for all pairs).

EHD	SVH	SoVI	SVI	SDI	EHDSE
0.470	0.152	0.149	0.097	0.262	0.166

Figure 3:

Heat maps comparing index rankings to population density rankings for census tracts in Washington state. Each cell in the graph represents a combination of ranks from the two rankings being compared. The shade of the cell indicates how many census tracts have the combination of ranks. Cells falling along the diagonal line have identical values in both rankings. Darker shades towards the diagonal line indicate greater agreement between the two rankings.

Tracts with high EHD decile ranks are concentrated in densely populated areas (Figure 4G), notably the eastern and southern Puget sound area (Figure 4A) with only one rural census tract (southeast of Yakima in the southern central portion of the state) falling into the ninth decile and no rural census tracts in the 10th decile. Nearly all of the rural tracts in the northwest quarter of the state fall into the lowest three deciles.

Figure 4:

Index and population ranks for census tracts for Washington State and the Puget sound area. (A) EHD index rank; (B) SVH index rank; (C) SoVI index rank; (D) SVI index rank; (E) SDI index rank; (F) EHDSE index rank; and (G) population density rank.

Ranks from the five non-EHD indexes are more evenly distributed among urban and rural census tracts (Figure 4B–F). Tracts generally have the same or similar values among the indexes, for instance with tracts along the northern border. There are, however, some exceptions, for instance the large tract south southeast of the Puget sound area near the southern border, which ranks in the lowest decile in four non-EHD Indexes (Figure 4B–E) and in the eighth decile for EHDSE (Figure 4F).

4 Discussion

4.1 Similarity, Complexity, and Causal Pathway Transparency

The high level of correlation among the non-EHD indexes indicates that simpler models, with fewer variables and involving less complex computation, may provide similar utility to indexes with greater complexity. With results most similar at higher decile rankings, this would be particularly true if an index is used to identify the most vulnerable census tracts. Using a smaller set of variables with a simpler method (for instance mean of decile rankings with seven variables for EHDSE versus a principal components analysis with 30 variables for SoVI) provides an easier path for analysis as well as updating index values as new datasets become available or if indexes are used for scenario-based planning. This is especially important if an index is used in a planning tool where users may want to interactively specify and analyze alternative scenarios such as climate change, hazard occurrence, or mitigation and adaptation actions.

EHD, EHDSE, SVH, and SVI all use simple, multi-tiered computations, which expose the causal pathways of risk in a straightforward manner. A user can examine the variables in a lower tier to see those factors contributing most highly to the next tier up. EHD, EHDSE, and SHV are incorporated in the WTN IBL tool which exposes the tiered calculations and results with a map display and allows users to explore local index drivers interactively. All four of these indexes are, however, limited to three tiers, which limits the potential to expand them using more complex relational models.

SoVI and SDI both utilize factor analysis, which combines input variables in complex ways. This prevents users from readily seeing and understanding the direct contributions of input variables to the final results.

The difference between EHD and the other indexes is striking but unsurprising. Other indexes, including the socioeconomic vulnerability layer (EHDSE) used by EHD, were designed to measure populations’ vulnerability to hazards. EHD “was based on a model that integrates measures of environmental exposures, adverse environmental effects, sensitivities, and sociodemographic vulnerabilities together to create a single composite score” (Min et al. 2019). Many non-socioeconomic factors used to compute EHD, such as those related to air pollution (e.g. NOx-diesel emissions, ozone, PM2.5, heavy traffic roadways, and toxic facilities release) and other environmental exposures (e.g. proximity to hazardous waste facilities, Superfund sites, and wastewater discharges) are generally concentrated in more densely populated areas and account for EHD’s higher correlation with higher population density compared to the other indexes.

4.2 Hazard-specific Vulnerability Indexes and Factors

The non-EHD socioeconomic-based indexes, while potentially useful, fail to consider specific socioeconomic and non-socioeconomic factors that increase vulnerability to specific hazards. Hazard-specific risk factors and indexes have been and continue to be a subject of research.

Vulnerability indexes for flooding such as Coastal City Flood Vulnerability Index (CCFVI) (Balica et al. 2012), Flood Vulnerability Index (FloodVI) (Fernandez et al. 2016), Livelihood Vulnerability Index (LVI) (Can, Tu, and Hoanh 2013), District Flood Vulnerability Index (DFVI) (Nasiri et al. 2019) consider factors exclusive to or highly influenced by flooding including population dependence on agriculture, fishing, and tourism; joblessness during floods; indicators of preparation such as hazard maps, warning systems, and flood insurance; and ecosystem resilience to flooding.

A number of vulnerability indexes for heat have also been created, including the Heat Vulnerability Index (HVI) (Reid et al. 2009, 2012), Extreme Heat Vulnerability Index (EHVI) (Johnson et al. 2012), Heatwave Vulnerability Index (HVI) for London (Wolf and McGregor 2013; Wolf, McGregor, and Analitis 2014). In addition to socioeconomic variables, these indexes include variables for age, health, geographic environment (e.g. urban setting, vegetation cover), exposure (e.g. insulation, residence position), and access to air conditioning. Future work could compare these hazard-specific Indexes with the more general Indexes examined here.

Sabrin et al. (Sabrin et al. 2020) have developed a combined air pollution and heat island vulnerability index for Camden, New Jersey that considers socioeconomic factors as well as environmental factors such as vegetation coverage, building characteristics, impervious surfaces, and water presence. However, air pollution vulnerability indexes are not widely published, even though factors affecting vulnerabilities to non-wildfire air pollution have been long recognized and include age, health status, gender, socioeconomic factors (Makri and Stilianakis 2008). Wildfire smoke can be considered as a hazard separate from or in addition to other air pollution. Specific vulnerabilities to wildfire smoke include or may include age (young children and older adults), preexisting cardiovascular disease, and preexisting asthma (Reid et al. 2016; Liu et al. 2015). While indexes for wildfire smoke vulnerability are not widely published, efforts to identify differential risk to wildfire smoke include online tool development (Vaidyanathan et al. 2018).

The diversity of general and hazard-specific Indexes poses a challenge for integrated analyses (Flanagan et al. 2018). This is particularly the case for analyses that aim to incorporate population vulnerability to a range of environmental hazards that are likely to systematically change over time (Martens et al. 2009). For example, vulnerability mapping to support climate change mitigation typically focuses on exposure to pollution from fossil fuel pollution (Shonkoff et al. 2011), while vulnerability mapping for climate change adaptation tends to focus on climate-sensitive hazards like heat, extreme precipitation, and flooding (Wilson et al. 2010). While the two sets of hazards have a common precursor, vulnerability patterns for the various hazards are distinct.

4.3 Towards More Sophisticated and Informative Methods and Tools

To be useful to decision makers, frameworks for decision support should identify the sources of vulnerability so that appropriate interventions can be undertaken. Yu et al. (Yu et al. 2021) do this by identifying factors affecting sensitivity, exposure, and adaptive capacity for four different hazards: heat, flooding and sea level rise, wildfire smoke, and ground-level ozone. They combine these results into an overall vulnerability index. These results are presented in separate maps, potentially allowing decision makers to direct specific interventions to areas based on vulnerabilities to specific hazards, however their use of principal component analysis complicates determining the local causal pathways for risk.

Such research has the potential to inform sophisticated methods and tools that provide researchers and managers with clear visualizations of vulnerability and the underlying causes of vulnerability. Existing online tools for health vulnerability (e.g. EHD, https://fortress.wa.gov/doh/wtn/WTNIBL/; CalEnviroScreen, https://oehha.maps.arcgis.com/apps/MapSeries/index.html?appid=8dad35dcd2274285874e60871c404edc) already provide users with the ability to explore multilayered datasets to discover the causes and effects of vulnerability. Online tools relating environment change to climate change (e.g. The California Climate Console; http://www.climateconsole.org/) provide examples for exploring the causes of and uncertainties in projected changes.

Fuzzy logic modeling (Zadeh 1973) presents an alternative multi-tiered modeling method with the potential to produce indexes whose drivers are easily discovered. Fuzzy logic models are not limited to an arbitrary number of tiers and the operators for combining variables can capture greater nuance in the relationships among variables than weighted sums or means. Fuzzy logic also allows for the combining of separate models by adding a tier via a single operator. The Environmental Evaluation Modeling System (EEMS) (Sheehan and Gough 2016) and its online implementation (EEMS online; www.eemsonline.org) provide a framework and tool that allow for the creation, exploration, and development of such models for environmental decision support. This methodology could be directly applied in the area of health vulnerability and risk.

Powerful, cloud-based spatial data systems such as Google Earth Engine (https://earthengine.google.com/) and Microsoft Azure Maps (https://azure.microsoft.com/en-us/services/azure-maps/) provide resources for spatially modeling complex datasets at global scales. These methods and resources provide a solid base for creating high quality, hazard-specific indexes and powerful decision support tools using them. Updates to datasets on these platforms would be readily available to any risk analysis frameworks interfacing with them, making automatically-updating tools a possibility.

5 Conclusion

We compared hazard vulnerability indexes and found that the five of them that considered only socioeconomic factors yielded similar results over Washington State, especially for the most vulnerable decile of census tracts. EHD, the index that also considered hazard exposures, differed from the rest and had higher values in areas with higher population densities where the hazard exposures it considered are more common.

Vulnerability to specific hazards and vulnerability indexes developed for specific hazards involve factors beyond the socioeconomic. A framework for developing an index using multiple hazards not only considered different factors for each index, but further classified the factors in terms of sensitivity, exposure, and adaptive capacity. This methodology, coupled with methods and technologies for geospatial modeling of other types of data provide the building blocks for powerful decision support tools that could help inform hazard vulnerability decisions on a global scale.

Research and development of vulnerability and risk indexes is important in light of the increasing hazards posed by climate change, and researchers continue to develop indexes using different methodologies and for different hazards. Fuzzy logic’s successful use in environmental decision support suggest it has high potential as an additional means of not only creating vulnerability and risk indexes, but also in exposing the local causal pathways of vulnerability and risk. This method’s applicability should be explored both for its ability to generate indexes and to provide decision makers with actionable information. Follow-on research should also explore its incorporation with the rich spatial data increasingly available on cloud platforms.

Whatever direction future climate vulnerability and risk index research takes, comparisons with existing indexes will be an important aspect. The indexes surveyed in this paper provide benchmarks for comparison and the methods used provide a framework for making comparisons with other indexes and indexes applied in other geographies.

Corresponding author: Tim Sheehan, Department of Environmental and Occupational Health Sciences, University of Washington, Box 351618, Seattle, WA, 98195-1618, USA, E-mail: tjshee@uw.edu

References

Abatzoglou, J., and A. P. Williams. 2016. “Impact of Anthropogenic Climate Change on Wildfire Across Western US Forests.” Proceedings of the National Academy of Sciences 113 (42): 11770–5.10.1073/pnas.1607171113Suche in Google Scholar

Adger, W. N. 2006. “Vulnerability.” Global Environmental Change: Human and Policy Dimensions 16 (3): 268–81. https://doi.org/10.1016/j.gloenvcha.2006.02.006.Suche in Google Scholar

Balica, S. F., N. G. Wright, and F. van der Meulen. 2012. “A Flood Vulnerability Index for Coastal Cities and its Use in Assessing Climate Change Impacts.” Natural Hazards 64 (1): 73–105. https://doi.org/10.1007/s11069-012-0234-1.Suche in Google Scholar

Borden, K. A., M. C. C. T. W. W. SchmidtleinEmrichPiegorsch, and S. L. Cutter. 2007. “Vulnerability of U.S. Cities to Environmental Hazards.” Journal of Homeland Security and Emergency Management 4 (2). https://doi.org/10.2202/1547-7355.1279.Suche in Google Scholar

Bouwer, L. M. 2019. “Observed and Projected Impacts from Extreme Weather Events: Implications for Loss and Damage.” In Loss and Damage from Climate Change: Concepts, Methods and Policy Options, edited by Mechler, R., Bouwer, L. M., Schinko, T., Surminski, S. and Linnerooth-Bayer, J., pp. 63–82. Cham: Springer International Publishing.10.1007/978-3-319-72026-5_3Suche in Google Scholar

Can, N. D., H. VoTu, and T. ChuHoanh. 2013. “Application of Livelihood Vulnerability Index to Assess Risks from Flood Vulnerability and Climate Variability—A Case Study in the Mekong Delta of Vietnam.” Journal of Environmental Science and Engineering 2: 476–86.Suche in Google Scholar

Cutter, S. L., B. J. Boruff, and W. Lynn Shirley. 2003. “Social Vulnerability to Environmental Hazards.” Social Science Quarterly 84 (2): 242–61. https://doi.org/10.1111/1540-6237.8402002.Suche in Google Scholar

Fernandez, P., S. Mourato, M. Moreira, and L. Pereira. 2016. “A New Approach for Computing a Flood Vulnerability Index Using Cluster Analysis.” Physics and Chemistry of the Earth, Parts A/B/C 94: 47–55. https://doi.org/10.1016/j.pce.2016.04.003.Suche in Google Scholar

Flanagan, B., E. W. Gregory, E. Hallisey, J. L. Heitgerd, and B. Lewis. 2011. “A Social Vulnerability Index for Disaster Management.” Journal of Homeland Security and Emergency Management 8 (1). https://doi.org/10.2202/1547-7355.1792.Suche in Google Scholar

Flanagan, B. E., E. J. Hallisey, E. Adams, and A. Lavery. 2018. “Measuring Community Vulnerability to Natural and Anthropogenic Hazards: The Centers for Disease Control and Prevention’s Social Vulnerability Index.” Journal of Environmental Health 80 (10): 34–6.Suche in Google Scholar

Fuchs, S., M. Keiler, R. Ortlepp, R. Schinke, and M. Papathoma-Köhle. 2019. “Recent Advances in Vulnerability Assessment for the Built Environment Exposed to Torrential Hazards: Challenges and the Way Forward.” Journal of Hydrology 575: 587–95. https://doi.org/10.1016/j.jhydrol.2019.05.067.Suche in Google Scholar

Greco, S., A. Ishizaka, M. Tasiou, and G. Torrisi. 2019. “On the Methodological Framework of Composite Indexes: A Review of the Issues of Weighting, Aggregation, and Robustness.” Social Indicators Research 141 (1): 61–94. https://doi.org/10.1007/s11205-017-1832-9.Suche in Google Scholar

Honda, Y., M. Kondo, G. McGregor, H. Kim, Y. L. Guo, Y. Hijioka, M. Yoshikawa, K. Oka, S. Takano, S. Hales, and R. S. Kovats. 2014. “Heat-Related Mortality Risk Model for Climate Change Impact Projection.” Environmental Health and Preventive Medicine 19: 56–63, https://doi.org/10.1007/s12199-013-0354-6.Suche in Google Scholar

Johnson, D. P., S. Austin, V. Lulla, and L. George. 2012. “Developing an Applied Extreme Heat Vulnerability Index Utilizing Socioeconomic and Environmental Data.” Applied Geography 35 (1): 23–31. https://doi.org/10.1016/j.apgeog.2012.04.006.Suche in Google Scholar

Kundzewicz, Z. W., S. Kanae, S. I. Seneviratne, J. Handmer, N. Nicholls, P. Peduzzi, R. Mechler, L. M. Bouwer, N. Arnell, K. Mach, R. Muir-Wood, G. R. Brakenridge, W. Kron, G. Benito, Y. Honda, K. Takahashi, and B. Sherstyukov. 2014. “Flood Risk and Climate Change: Global and Regional Perspectives.” Hydrological Sciences Journal 59 (1): 1–28, https://doi.org/10.1080/02626667.2013.857411.Suche in Google Scholar

Liu, J. C., G. Pereira, S. A. UhI, M. A. Bravo, and M. L. Bell. 2015. “A Systematic Review of the Physical Health Impacts from Non-occupational Exposure to Wildfire Smoke.” Environmental Research 136: 120–32. https://doi.org/10.1016/j.envres.2014.10.015.Suche in Google Scholar

Makri, A., and N. I. Stilianakis. 2008. “Vulnerability to Air Pollution Health Effects.” International Journal of Hygiene and Environmental Health 211 (3–4): 326–36. https://doi.org/10.1016/j.ijheh.2007.06.005.Suche in Google Scholar

Martens, P., D. McEvoy, and C. Chang. 2009. “The Climate Change Challenge: Linking Vulnerability, Adaptation, and Mitigation.” Current Opinion in Environmental Sustainability 1 (1): 14–8. https://doi.org/10.1016/j.cosust.2009.07.010.Suche in Google Scholar

McDermott-Levy, R., M. Scolio, K. Shakya, and C. Moore. 2021. “Factors That Influence Climate Change-Related Mortality in the United States: An Integrative Review.” International Journal of Environmental Research and Public Health 18 (15): 8220.10.3390/ijerph18158220Suche in Google Scholar

Mechler, R., and L. M. Bouwer. 2015. “Understanding Trends and Projections of Disaster Losses and Climate Change: Is Vulnerability the Missing Link?” Climatic Change 133 (1): 23–35. https://doi.org/10.1007/s10584-014-1141-0.Suche in Google Scholar

Min, E., D. Gruen, D. Banerjee, T. Echeverria, L. Freelander, M. Schmeltz, and E. Saganić. 2019. “The Washington State Environmental Health Disparities Map: Development of a Community-Responsive Cumulative Impacts Assessment Tool.” International Journal of Environmental Research and Public Health 16 (22): 4470. https://doi.org/10.3390/ijerph16224470.Suche in Google Scholar

Mendes, O., and José Manuel de. 2009. “Social Vulnerability Indexes as Planning Tools: Beyond the Preparedness Paradigm.” Journal of Risk Research 12 (1): 43–58.10.1080/13669870802447962Suche in Google Scholar

Nasiri, H., M. J. M. Yusof, T. A. M. Ali, and M. K. B. Hussein. 2019. “District Flood Vulnerability Index: Urban Decision-Making Tool.” International journal of Environmental Science and Technology 16 (5): 2249–58. https://doi.org/10.1007/s13762-018-1797-5.Suche in Google Scholar

OEHHA. n.d. 2018. CalEnviroScreen 3.0:Update to the California Communities Environmental Health Screening Tool. Also available at https://oehha.ca.gov/calenviroscreen/report/calenviroscreen-30.Suche in Google Scholar

Philip, D., and Md I. Rayhan. 2004. “Vulnerability and Poverty: What Are the Causes and How Are They Related.” Bonn: ZEF Bonn, Center for Development Research, University of Bonn. Also available at https://www.zef.de/fileadmin/downloads/forum/docprog/Termpapers/2004_3a_Philip_Rayan.pdf.Suche in Google Scholar

Reid, C. E., M. Jerrett, J. R. Balmes, and C. T. Elliott. 2016. “Critical Review of Health Impacts of Wildfire Smoke Exposure.” Environmental Health Perspectives 124 (9): 1334–43. https://doi.org/10.1289/ehp.1409277.Suche in Google Scholar

Reid, C. E., J. K. Mann, A. Ruth, P. B. English, R. A. Lincoln, and H. G. Margolis. 2012. “Evaluation of a Heat Vulnerability Index on Abnormally Hot Days: An Environmental Public Health Tracking Study.” Environmental Health Perspectives 120 (5): 715–20. https://doi.org/10.1289/ehp.1103766.Suche in Google Scholar

Reid, C. E., M. S. O’Neill, C. J. Gronlund, S. J. Brines, and J. Schwartz. 2009. “Mapping Community Determinants of Heat Vulnerability.” Environmental Health Perspectives 117 (11): 1730–6. https://doi.org/10.1289/ehp.0900683.Suche in Google Scholar

Sabrin, S., M. Karimi, and R. Nazari. 2020. “Developing Vulnerability Index to Quantify Urban Heat Islands Effects Coupled with Air Pollution: A Case Study of Camden, NJ.” ISPRS International Journal of Geo-Information 9 (6): 349. https://doi.org/10.3390/ijgi9060349.Suche in Google Scholar

Schmidtlein, M. C., R. C. Deutsch, W. W. Piegorsch, and S. L. Cutter. 2008. “A Sensitivity Analysis of the Social Vulnerability Index.” Risk Analysis: An Official Publication of the Society for Risk Analysis 28 (4): 1099–114. https://doi.org/10.1111/j.1539-6924.2008.01072.x.Suche in Google Scholar

Sheehan, T., and M. Gough. 2016. “A Platform-independent Fuzzy Logic Modeling Framework for Environmental Decision Support.” Ecological Informatics 34: 92–101. https://doi.org/10.1016/j.ecoinf.2016.05.001.Suche in Google Scholar

Shonkoff, S. B., R. Morello-Frosch, M. Pastor, and S. James. 2011. “The Climate Gap: Environmental Health and Equity Implications of Climate Change and Mitigation Policies in California—a Review of the Literature.” Climatic Change 109 (S1): 485–503. https://doi.org/10.1007/s10584-011-0310-7.Suche in Google Scholar

Tellman, B., C. Schank, B. Schwarz, P. D. Howe, and Alex de Sherbinin. 2020. “Using Disaster Outcomes to Validate Components of Social Vulnerability to Floods: Flood Deaths and Property Damage across the USA.” Sustainability: Science, Practice and Policy 12 (15): 6006. https://doi.org/10.3390/su12156006.Suche in Google Scholar

Thomalla, F., D. Tom, E. Spanger-Siegfried, G. Han, and J. Rockstrom. 2006. “Reducing Hazard Vulnerability: Towards a Common Approach between Disaster Risk Reduction and Climate Adaptation.” Disasters 30 (1): 39–48. https://doi.org/10.1111/j.1467-9523.2006.00305.x.Suche in Google Scholar

Vaidyanathan, A., F. Yip, and P. Garbe. 2018. “Developing an Online Tool for Identifying At-Risk Populations to Wildfire Smoke Hazards.” The Science of the Total Environment 619–620: 376–83. https://doi.org/10.1016/j.scitotenv.2017.10.270.Suche in Google Scholar

Wilson, S. M., R. Richard, L. Joseph, and E. Williams. 2010. “Climate Change, Environmental Justice, and Vulnerability: An Exploratory Spatial Analysis.” Environmental Justice 3 (1): 13–9. https://doi.org/10.1089/env.2009.0035.Suche in Google Scholar

Wolf, T., and Glenn McGregor. 2013. “The Development of a Heat Wave Vulnerability Index for London, United Kingdom.” Weather and Climate Extremes 1: 59–68. https://doi.org/10.1016/j.wace.2013.07.004.Suche in Google Scholar

Wolf, T., Glenn McGregor, and A. Analitis. 2014. “Performance Assessment of a Heat Wave Vulnerability Index for Greater London, United Kingdom.” Weather, Climate, and Society 6 (1): 32–46. https://doi.org/10.1175/wcas-d-13-00014.1.Suche in Google Scholar

Yoon, D. K. 2012. “Assessment of Social Vulnerability to Natural Disasters: A Comparative Study.” Natural Hazards 63 (2): 823–43. https://doi.org/10.1007/s11069-012-0189-2.Suche in Google Scholar

Yu, J., K. Castellani, K. Forysinski, P. Gustafson, J. Lu, E. Peterson, M. Tran, A. Yao, J. Zhao, and M. Brauer. 2021. “Geospatial Indicators of Exposure, Sensitivity, and Adaptive Capacity to Assess Neighbourhood Variation in Vulnerability to Climate Change-Related Health Hazards.” Environmental Health: A Global Access Science Source 20 (1): 31. https://doi.org/10.1186/s12940-021-00708-z.Suche in Google Scholar

Zadeh, L. A. 1973. “Outline of a New Approach to the Analysis of Complex Systems and Decision Processes.” IEEE Transactions on Systems, Man, and Cybernetics SMC- 3 (1): 28–44. https://doi.org/10.1109/tsmc.1973.5408575.Suche in Google Scholar

Zahran, S., S. D. Brody, W. G. Peacock, V. Arnold, and H. Grover. 2008. “Social Vulnerability and the Natural and Built Environment: A Model of Flood Casualties in Texas.” Disasters 32 (4): 537–60. https://doi.org/10.1111/j.1467-7717.2008.01054.x.Suche in Google Scholar

Received: 2021-09-03

Accepted: 2023-01-31

Published Online: 2023-03-27

This work is licensed under the Creative Commons Attribution 4.0 International License.

Artikel in diesem Heft

https://doi.org/10.1515/jhsem-2021-0066

Schlagwörter für diesen Artikel

vulnerability index; hazard; climate change; decision support; Washington State

Creative Commons

BY 4.0