Residential flood vulnerability along the developed North Carolina, USA coast: High resolution social and physical data for decision support

This article presents an ArcGIS geodatabase of socio-demographic and physical characteristics derived from recent high resolution data sources to construct measures of population vulnerability to inundation in the 28 counties of coastal North Carolina, U.S.A. as presented in Pricope et al., 2019. The region is simultaneously densely populated, low-lying and exposed to recurrent inundation related to storms and incremental sea level rise. The data presented here can be used as a decision support tool in coastal planning, emergency management preparedness, designing adaptation strategies and developing strategies for coastal resilience. The socio-demographic data (population and housing) was derived from 228 tables at the block-group level of geography from the 2010 U.S. Census Bureau. These data were statistically analyzed, using Principal Component Analysis, to identify key factors and then used to construct a Social Vulnerability Index (SOVI) at the block-group level of geography which highlighted regions where socio-demographic characteristics such as family structure, race, housing (primarily owner vs. renter-occupied), special needs populations (e.g. elderly and group living), and household/family size play an overwhelmingly important role in determining community vulnerability from a social perspective. An index of physical exposure was developed using the National Flood Hazards Maps (available from North Carolina's Flood Risk Information System and FEMA) along with a novel building inventory dataset available from the North Carolina Department of Public Safety that contains the Finished-Floor Elevation of every structure in the state. We took advantage of the unprecedented high spatial resolution nature of the building inventory dataset to calculate an index of physical vulnerability to inundation of every block group in the 28 coastal counties relative to Base Flood elevations and identified hotspots where this intersection predisposes people to an increased risk of flooding. Here, we present the final derived dataset containing the social, physical and an integrative measure of vulnerability to flooding that can be used at multiple scales of analysis, starting with the regional, county, local, and neighborhood to identify areas of priority intervention for risk-reduction in coastal planning and emergency management preparedness as well as forward-looking adaptation strategies.

Modeling residential coastal flood vulnerability using finished-floor elevations and socio-economic characteristics

Densely populated coastal regions are vulnerable to threats associated with climate change and variability, especially storms. In the United States, millions of people are repeatedly at risk of flooding and because this number will only continue to grow, the identification of the intersection of social vulnerability and physical risk to flood inundation is essential for both coastal planning and adaptation purposes. Although a key tool to identify vulnerable populations, most vulnerability models are built at the county or coarser scales, thereby hindering the effectiveness of mitigation and adaptation planning at community scales, which are more socially and physically diverse than what county-scale analyses can reveal. We present an integrated social and physical model of vulnerability at the block-group level of geography using census data to measure social variability based population and housing data and physical exposure based on the intersection of finished floor elevation of all buildings in coastal North Carolina, USA with flood hazards maps. We identify, in a spatially-explicit manner and at multiple levels of governance, areas of high social vulnerability and their intersection with areas of high physical exposure to inundation. We found that in the 28 coastal counties of North Carolina, 45.3% of the structures within the 100-year floodplain were structurally exposed to potential damage from inundation. Supporting our hypothesized patterns of vulnerability to inundation, a significant clustering of highly vulnerable block-groups were located in Albemarle and Eastern Carolina coastal regions, yet high vulnerability outliers were also located at significant distance away from the highly physically-exposed coastline. Our findings suggest that the high-resolution block-group level analysis identified multiple levels of vulnerability to inundation at the sub-county scale and provide essential information for effective hazard mitigation within scales ranging from the community to transboundary governing bodies.