Developing Distributions of Annual Average Concentration with Dependency Among Daily Values. A. C. Cullen and C. S. Bretherton, University of Washington, Seattle, WA
One often overlooked complexity in the development of distributions to represent concentration of a contaminant in air or food for probabilistic assessment arises from the fact that the magnitude of one’s contact concentration on a given day is likely related to that on the previous day. This occurs in air (and subsequently inhalation exposure) because a weather pattern and its accompanying evaporative conditions tends to settle in an area for several days at a time. In food (and subsequently ingestion exposure) it occurs because an individual may consume several meals from a single produce item, such as a head of lettuce. We present a method for developing a distribution to represent annually averaged concentration based on a small set of concentration measurements, where the number of exposure/contact concentrations across the exposure duration is smaller than the number of days in that time period. Accounting for this complexity may result in a significant increase (up to an order of magnitude in the span of 5th to 95th percentiles in our case) in the overall uncertainty/variability in annually averaged concentration, compared to treating successive days as independent.
Sponsored by USEPA’s National Center for Research on Statistics and the Environment, UW.
Go to . . .