While much of the existing literature focuses on socio-ecological, socio-economic, infrastructural, or institutional vulnerability, characteristics of the built environment and particularly residential housing stock is essential for understanding the degree to which hazards may affect people and social systems. Citing the tsunami that struck Asia in 2004 as an example, Adger (2006) posits that stochastic weather events illustrate underlying vulnerability in a place. While susceptibility to damage from hazards centers around social systems or infrastructural systems (see discussion above), characteristics of the built environment (such as building materials, vintage, condition, location, and typology) affect the extent of damage from extreme events and are both sources and manifestations of vulnerability.

Based on a review of the literature and recent and ongoing changes to building code regulations in Harris County, I have opted to focus on social and built vulnerability for the purpose of this project. Social vulnerability is an established concept and is a topic of institutional and academic concern for Houston and coastal areas more broadly. I elected to include built characteristics as a domain of interest due to fact that much of the damage experienced through Hurricane Harvey was suffered by households outside of FEMA’s designated flood zones, suggesting that the presence of flood insurance based on existing FIRM maps alone is an insufficient marker of a property’s preparedness and ability to cope with flood events. Recent flood events, the rate of urbanization in the area, and the current discussion revolving around revising the building code and flood insurance requirements in Houston signal that the time is ripe to consider vulnerability of structures in the larger discussion about vulnerability, resilience, and adaptation.

The goal of this project is to explore social and built vulnerability to flood events across Houston. By applying statistical techniques, my project will provide a synthesis of vulnerability factors across domains. Existing web tools either rely on the Center for Disease Control’s conception of social vulnerability as a measure of vulnerability or present the user with a range of vulnerability-related datasets for exploration. This project extends beyond producing overlays by combining these disparate input layers into a synthetic measure of vulnerability while simultaneously calling attention to the explanatory factors. This tool provides a baseline for multi-domain vulnerability for Houston that can inform policy discussions, guide adaptation initiatives and resource distribution (Smit and Wandel, 2006 p.285), and influence data collection and measurement in the future.

There is a body of resilience and vulnerability self-assessment tools that allow communities to develop measures of vulnerability based on surveys, focus groups, and interviews to develop a shared definition of vulnerability and the ways in which that community might develop adaptations to changing conditions (Smit and Wandel, 2006 p.284). This qualitative process is instrumental in developing realistic and context-specific adaptations in an ever-changing world and relies wholly on individual input to guide the broader process. The current project relies upon publicly-available data from a number of sources to develop a relative measure of the spatial distribution of vulnerability across the region. Several limitations present themselves in this method, notably the availability of data at a useful scale, as well as the relevance of existing data for measuring vulnerability at all. An inherent bias exists in the construction of quantitative vulnerability indicators through the assumption that vulnerability can be quantified at all, and that the data that can be used to form a proxy understanding of vulnerability is available and current enough to form a faithful representation of vulnerability for use in policy making.