Spatial heterogeneity in social vulnerability to flood exposure

The authors investigate the spatial dynamics of the relationships between social vulnerability indicators and flood exposure using a literature review and comparison of regression methods. This research was undertaken as the assumption of spatial stationarity can obscure important geographical variations in the relationship between social vulnerability indicators and the exposure to hazard, such as flooding. First, the authors undertook a systematic literature review assesses whether spatial heterogeneity is evident in existing studies. Then the authors - using Texas as a case study - apply two types of regression that assume spatial heterogeneity and stationarity respectively to find the nature of the relationships between twenty vulnerability indicators and flood exposure.

Some of the main findings of the study are as follows:

• Inconsistent results across studies: the literature review found little consistency in how indicators like race, income, or housing status relate to flood exposure across the United States (US), suggesting these relationships are highly context-specific.
• Only six of twenty indicators exhibit stationary relationships to flood exposure, while the majority demonstrate spatial heterogeneity, with localized variations in strength, direction, and significance.
• Underused indicators may be important: some potentially important indicators, such as foreign-born population, were rarely used in studies despite their likely relevance in specific places.
• Single indicators may miss intersectional vulnerability: disaggregating indicators may fail to capture unique and compounded vulnerability from overlapping characteristics.