Limitations in the 1996 National-Scale Air Toxics Assessment-Page4

EPA conducted the national-scale assessment variability analysis to understand how ambient air concentration, exposure and risk vary geographically, and not between specific individuals. Still, many of the components of variation of risk discussed in the page on Components were included. The U.S. was divided into census tracts, and EPA calculated the ambient air concentrations for each census tract based on the emission sources and meteorological conditions affecting those specific tracts. The ambient (outdoor) air concentrations, however, were time-averaged. The calculations of ambient air concentration, therefore, do not reflect temporal variation (the first bullet in the page on Components, but they do reflect geographic variation (the second bullet in the page on Components).

Individuals were placed at the centroid (center of population mass) of the census tract in which they live. This reflects variation in geographic location of individuals between census tracts, but it does not reflect variation in geographic location of individuals within a census tract (accounting partially for the third bullet in the page on Components. Activity patterns were included for each of 10 receptor populations defined by age and gender (racial/ethnic groups also were considered, but the activity patterns were not significantly different and so these groups were averaged), and even within a receptor population some variability in activity patterns between individuals was considered. This reflects in part the fourth bullet in the page on Components. The initial draft of this assessment determined that the results were not sufficiently different among racial/ethnic groups to justify retaining separate results for these groups in this final version. The results were, therefore, averaged over these groups. Differences in susceptibility and sensitivity (the fifth and sixth bullets in the page on Components were not included in the national-scale assessment. There were two reasons for not including them:

• First, the study considers only geographic differences in air toxic concentration, exposure and risk. The goal is to understand how these three factors differ as one moves between geographic areas. This is reflected, as mentioned above, by keeping track of differences in air concentration in different census tracts. This produces differences in the typical air toxic concentration, exposure, and risk found in different tracts. Susceptibility and sensitivity, however, produce differences in risk between two individuals in the same census tract, and the national-scale assessment was not intended to report these differences.
  
  

• Second, there is very limited information on differences in susceptibility and sensitivity between individuals. Even if it was decided to calculate and report differences between individuals in a census tract, scientifically reliable information necessary to produce these calculations is not available for many of the air toxics. It is possible, given current information, to at least estimate variability in the rates at which people breathe air, but this is only a small component of the overall variation in susceptibility and sensitivity. It was decided, therefore, not to incorporate this source of variation between individuals.

Taking into consideration these comments, EPA decided to incorporate differences in emissions and meteorology (resulting in differences in ambient air concentration) between census tracts, as well as differences in location of typical individuals (resulting in differences in exposure) between census tracts. In addition, variation in activity patterns for different age groups is reflected in the study. Variability in susceptibility and sensitivity is not, however, included for the reasons given above. In addition, temporal variation is included in the sense that it was used to develop time-weighted averages of emissions characteristics and meteorological conditions, but no temporal variation of ambient air concentration was estimated (only the time-weighted annual average).

What are the components of variability?
How was the variability analysis conducted?
What are the results of the variability analysis
How can these results be interpreted?

Return to the Main Variability Page