5. Using Geographic Information Science to Estimate Vulnerable Urban Populations for Flood Hazard and Risk Assessment in New York City

The research presented in this chapter seeks to demonstrate a new method to more accurately estimate populations vulnerable to hazards, especially in densely developed mega-cities, and to characterize at-risk populations based on measures of social, physical, and health vulnerability. Emergency management and disaster preparation, planning, mitigation, and recovery requires accurate estimation of potentially at-risk populations and sub-populations. Census data alone, however, cannot provide sufficiently detailed knowledge of population location and distribution, particularly in large, hyper-heterogeneous urban areas like New York City. Additionally, specific sub-populations (i.e., racial/ethnic minorities) may be at higher risk, yet under-counted by existing methods of calculating potentially exposed or impacted populations. We discuss two new inter-related methods that employ Geographic Information Science (GISc) to assess and quantify risk and vulnerability: the Cadastral-based Expert Dasymetric System (CEDS) and the New York City Hazard Vulnerability Index (NYCHVI). CEDS uses an expert system and dasymetric mapping to disaggregate population and sub-population data to the property tax lot level. The analysis shows that compared to CEDS, conventional areal weighting of census data and centroid-containment selection methods under count at-risk population for floods by 37 and 72%, respectively. We found that minorities and other vulnerable sub-populations are disproportionately underestimated using traditional methods, which impairs preparedness and relief efforts. NYCHVI provides a straightforward way of assigning a vulnerability rating to populations in potentially impacted areas, and incorporates locally significant factors that are not captured using national models. Used in tandem, CEDS and NYCHVI are effective in characterizing the vulnerable populations and areas subject to flooding and other hazards, enabling significant improvements in estimating vulnerability over prevailing methods.