The impacts of weather and pollution on human mortality in Birmingham, Alabama and Philadelphia, Pennsylvania

Past studies have examined how both extreme weather and atmospheric polluta nts influence human mortality. However, the differential and/or synergistic impacts of weather and pollution on mortality are poorly understood. This relationship is particularly notable in summer, when both extreme weather a nd high pollution episodes are frequent. The goal of this study is to descr ibe the relationship between atmospheric conditions las characterized by we ather and air pollution) and mortality in the summer season for Birmingham, Alabama and Philadelphia, Pennsylvania. To assess the health impacts of bo th weather and pollution (ozone and total suspended particulates, or TSP), we used a synoptic, or air mass-based, approach to take into account the en tire weather situation, rather than individual weather elements. This metho d was used to identify 'offensive' air masses, which are associated with el evated mortality, and then to investigate which individual or combination o f atmospheric conditions poses the greatest threat to human health in terms of acute (daily) mortality. In both cities, offensive weather events have a greater impact on acute mor tality than high concentrations of TSP or ozone. The highest mortality leve ls occur when the hottest, but not the most polluted, air mass is present i n each city. In philadelphia, neither TSP nor ozone appear to contribute an 'add on' effect to weather-related mortality. Under non-offensive weather situations, pollution concentrations are associated with increased mortalit y in Philadelphia. Yet, regardless of pollution concentration, mortality le vels are much lower for these air masses than for offensive events. Polluti on appears to be more important for mortality in Birmingham. Although Birmi ngham's high-mortality (offensive) air mass is not the most polluted, offen sive air mass days with high pollution concentrations nonetheless exhibit h igher mean mortality than offensive air mass days with low pollution concen trations. Also different from Philadelphia is a lack of a relationship in B irmingham between pollution levels and mortality on non-offensive air mass days. The relationship between summer weather and mortality is strong in bo th cities, while the role of pollution is less clear. This research undersc ores the imperative need for the development of a weather/health watch-warn ing system to alert the public that an offensive synoptic situation is immi nent. Copyright (C) 2000 Royal Meteorological Society.