Spokane, WA, is a nonattainment area for airborne particulate matter smaller than 10 mum (PM10), so that a detailed emission inventory for PM10 is needed to evaluate various control strategies. It is thought that emissions from paved and unpaved roads in Spokane contribute three-fourths of the anthropogenic PM10 (neglecting wind-blown dust from agricultural areas). A study was conducted in the summer and fall of 1992 and again in the spring and summer of 1994 to measure PM10 emission rates from paved and unpaved roads in Eastern Washington state using a novel tracer technique. A known amount of an inert tracer (SF6) was released and concentrations of PM10 and SF6 downwind of the road, along with meteorological parameters and traffic volume, were measured. The results of the tracer experiments showed that within experimental uncertainties the PM10 and the tracer gas disperse in the same manner, suggesting that the use of a tracer in a line source to simulate roadway PM10 emissions can provide a tool for improving the existing emission inventories from roads. The emission factors obtained from two unpaved road experiments (136 g per vehicle per kilometer traveled, or g VKT-1, and 336 g VKT-1) were similar in magnitude to those predicted using currently accepted empirical algorithms. The factors determined from six paved road experiments were approximately 80% higher than that predicted using current formulate (6.7 +/- 3.7 g VKT-1 compared to 3.7 g VKT-1) for two-lane roads with daily traffic less than 10,000 vehicles. For major highways (4 + lanes and traffic in excess of 10,000 vehicles per day) the emission factors obtained from the tracer experiments were, on average, 44% lower than those predicted using standard formulae (1.0 +/- 0.5 g VKT-1 compared to 1.8 g VKT-1). The calculated emission factors for paved roads exhibited a wide range of variability, suggesting that in order to quantify PM10 emission rates from paved roads, more investigation is warranted.
