Hypothermia Mortality in the United States: A Quantitative Assessment of Meteorological Thresholds

Excessive heat and its associated hazards has an extensive body of literature that includes many aspects of vulnerability and warning methodology. In contrast, there is a comparatively small body of research examining these same aspects for excessive cold weather. Hypothermia is a drop in core body temperature caused by cold exposure, with social factors contributing to higher levels of risk among some populations. This research investigates the spatio-temporal distribution of hypothermia mortality in the United States for the years 1979-2004. Hypothermia death data were aggregated by U.S. Census Combined Statistical Area (CSA) and the weather conditions were assessed in each CSA using daily weather data. Descriptive statistics and binomial probability distributions were utilized to determine the probability of a hypothermia death event occurring within a CSA for wind temperatures temperatures ranging from -60º F to 50º F. NWS wind chill criteria were obtained for Advisories, Watches, and Warnings for all Weather Forecast Office (hereafter, WFO) county warning areas. Each CSA was then matched up with the appropriate county warning area wind chill alert criteria. The frequency and probability of hypothermia fatalities was then compared to the criteria to determine if NWS wind chill alert criteria encompass days with high levels of hypothermia mortality. Preliminary results indicate that there are several regions with high level of hypothermia fatalities: these include the Rust Belt and Megalopolis cities, mid-size cities in the southern Piedmont region, along with several areas of the Desert Southwest. The preliminary results also indicate that deaths occur before NWS wind chill alert criteria are reached.