Twenty-four communities in North America were monitored over one year for a variety of air pollutants as part of a cross-sectional epidemiological study on the respiratory health effects of exposure to fine particle acidity. The relationships between these single-year observations and the long-term community levels of ambient sulfate and acidity were examined. In the health study it was assumed that the single-year measurements were indicative of the lifetime or long-term exposures of the participants (eight-, nine-, and ten-year-olds). Therefore, a strong relationship between the long-term and single-year (24-community) particle acidity and sulfate concentrations was important. Ambient sulfate data from a variety of alternate sources were obtained from monitoring sites close to 20 of the 24 communities. Long-term averages, which were determined for the warm season (May to September), were derived from a minimum of four complete years of monitoring data at each site. Long-term acidity concentrations were derived from these sulfate data because multi-year measurements of acidity were not available. These concentrations were calculated by multiplying the sulfate concentrations by the mean warm season acid-to-sulfate ratios observed during the 24-community study. For each community, 25 random estimates (determined by allowing the observed mean ratio to vary randomly by +/- 0.2) of the mean warm season acidity were used to determine the community-to-community differences in the long-term acid concentrations. Overall, the long-term and 24-community warm season sulfate concentrations were correlated with an R(2), determined from linear regression, of 0.92 (slope = 0.90 +/- 0.13). With only two exceptions, regardless of which of these exposure estimates were used, the communities that were determined to experience high (> 8 mu g m-(3)), moderate (4-8 mu g m(-3)) and low (< 4 mu g m(-3)) sulfate exposures did not change. Similarly, few communities crossed exposure classes when the long-term and short-term acid concentrations were compared. However, due to the increased uncertainty arising from the lack of information on the long-term acid-to-sulfate ratio, the average correlation (R(2)) between the long-term and 24-community exposure estimates (the mean of the 25 separate random estimates for each community) was 0.85 (slope = 0.94).
