Few studies have examined heat vulnerability on a sub-metropolitan area level. This paper presents an analysis of heat vulnerability across Ohio (USA) on a county level. Each county is classified as 'urban', 'suburban', or 'rural'. Four different criteria defining what is meteorologically 'oppressive' are evaluated individually. Each of these criteria is associated with an increase in mortality of several percent statewide. Absolute increases in mortality are greatest across urban counties, as expected. When these values are evaluated as a percentage increase in mortality, rural and suburban counties actually show a greater response. The differences among the 3 groups are not statistically significant. This research thus suggests that merely being an urban resident does not make one more vulnerable to heat.KEY WORDS: Heat vulnerability · Human mortality · Synoptic climatology · Atmospheric hazard · Suburban · Urban · Rural · Ohio (USA)
Resale or republication not permitted without written consent of the publisherClim Res 24: [255][256][257][258][259][260][261][262][263][264][265] 2003 waves, which fall outside the acclimatized range of the local population (e.g. Kalkstein & Davis 1989). In extreme cases, mortality can increase 100% above average levels (Ellis 1972).Much less research has examined human vulnerability on a sub-metropolitan area scale. Initial inquiries suggest that several factors, including age, income level, and level of social isolation, may be important (Smoyer 1998, McGeehin & Mirabelli 2001. Racial responses appear to vary within individual heat waves (Whitman et al. 1997) and perhaps on a regional level (Kalkstein & Davis 1989), though no universal pattern has been noted. Mortality increases among those 65 yr and older are much more significant than those of younger persons. Mortality increases of those under 65 yr of age are often not statistically significant (Greenberg et al. 1983, Kilbourne 1997.Nearly all studies that have focused upon heatrelated mortality and morbidity have concentrated upon urban areas. Much of the rationale lies in the availability of data: large cities or metropolitan areas often have a greater availability of meteorological data, both in terms of period of record and the number of meteorological variables. Further, these locations contain far greater sample sizes of human-mortality totals, or any other health response. Nevertheless, much of the existing literature also suggests that urban residents are more significantly affected than rural residents by oppressive heat (e.g. McGeehin & Mirabelli 2001). Some modeling studies of the impacts of future climate change ignore the possibility that rural residents may be vulnerable as well (Tol 2002). The urban heat island, building types, greater social isolation, and higher cost of living are all often cited as the culprits for this differential. Events such as the heat wave of July 1995 have supported this notion.This research presents an initial analysis into heatrelated vulnerability across various 'lev...