Quantifying Particulate Matter Emissions from Wind Blown Dust Using Real-time Sand Flux Measurements

An extensive sand flux monitoring network was installed on the dry lake bed at Owens Lake, California to determine hourly PM-10 emissions. The network consisted of 135 co-located electronic and passive sand flux samplers covering 135 square kilometers of the lake bed. The network measured the hourly sand flux at each site for 30 months. Previous researchers found that PM-10 emissions due to wind erosion are proportional to the saltation (or sand) flux. Hourly PM-10 emissions from each square kilometer were estimated by the equation, PM-10 = Kf x q, where q is the sand flux measured at 15 cm above the surface, and Kf is the proportional relationship between the sand flux and the PM-10 emissions. Kf values were determined by comparing CALPUFF model predictions to observed concentrations at six PM-10 monitor sites. The results showed that Kf changed spatially and temporally at Owens Lake and that the changes corresponded to different soil textures on the lake bed and to seasonal surface changes that affected erodibility. The results also showed that some source areas were active all year, while others were seasonal and sometimes sporadic. As part of the Owens Lake Dust Identification (Dust ID) Program, the locations of the active areas identified by the sand flux network were confirmed by observers who visually mapped the dust source areas, by remote time-lapse video cameras, and by surface inspections using a Global Positioning System. This paper also compares the Dust ID method to other methodologies used to estimate particulate matter emissions from wind blown dust, such as those contained in the USEPA’s AP-42 guidance document. Using the Dust ID results, hourly PM-10 emission rates for each square kilometer were input to the CALPUFF model to predict air quality impacts around the Owens Lake shoreline and at monitored receptors.