Abstract of Meeting Paper

Society for Risk Analysis 1996 Annual Meeting

Air Pollutant Time-Series Analysis: Outdoor Sources of Carbon Monoxide and Polycyclic Aromatic Hydrocarbons. N. Klepeis, Lockheed Environmental Systems and Technology, 980 Kelly Johnson Dr., Las Vegas, NV 89119; W. Ott and P. Switzer, Department of Statistics, Stanford University, Stanford, CA 94305; M. P. Zelenka and A. H. Huber, Air Resources Laboratory (MD-56), National Oceanic and Atmospheric Administration, Research Triangle Park, NC 27711; and W. C. Nelson, EPA National Exposure Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

The time series of airborne polycyclic aromatic hydrocarbons (PAH) and carbon monoxide (CO) continuously collected in five-minute intervals at a Redwood City, CA residence (both indoors and outdoors) over a period of about 3 months (December 1994 through March 1995) are analyzed with respect to possible outdoor pollutant sources. The time period of the study includes a significant portion of the San Francisco Bay Area’s wood burning season. Since the ratio of CO to PAH concentrations provides a "fingerprint" for different sources, it is calculated at different times of day to give an indication of when sources are active. In addition, daily counts of wood smoke plumes, which are available for the neighborhood around the residence, and minute-by-minute human activity patterns from the 1989-90 California Activity Pattern (CAP) study of adults and the 1992-94 National Human Activity Pattern Survey (NHAPS) study are used to estimate the diurnal time frames corresponding to increased automobile travel and wood burning. The observed plumes, activity patterns, CO/PAH ratios, and time series peaks all suggest that the two major outdoor sources for the given time period are diurnal traffic -- between approximately 4 AM and 11 AM -- and diurnal wood smoke -- between approximately 5 PM and midnight. Fortunately, there appears to be little overlap between these two source time frames -- even on weekends when activity patterns are somewhat shifted in time. CO and PAH concentrations that are linked to wood smoke sources have overall means that are X% and Y% higher, respectively, than CO and PAH concentrations that are linked to traffic sources.