Abstract
The Aerodynamic Particle Sizer (APS 3321, TSI, Inc., Shoreview, MN) rapidly measures particle number concentration by size from 0.5 to 20 μm. This work used simple assumptions for particle shape factor and density to estimate ambient coarse mode particulate matter, PM10–2.5, from APS number concentration data. This estimate was compared with that measured with time-integrated, filter-based federal reference method (FRM) samplers in four U.S. field studies: two in Phoenix, AZ; one in Gary, IN; and one in Riverside, CA. Near one-to-one agreement and a strong linear relationship were observed between APS-estimated and FRM-measured PM10–2.5 in the first Phoe nix, AZ study (slope = 0.90, R 2 = 1.00); the second Phoenix, AZ study (slope = 0.99, R 2 = 0.99); and the Riverside, CA study (slope = 1.00, R 2 = 0.84). In the Gary, IN study, PM10–2.5 estimates made with data from the APS tended to be less than that measured with the FRM samplers (slope = 0.57), but the linear relationship between the two methods was still strong (R 2 = 0.90). Particle-bound water associated with wet atmospheric conditions may account for these differences. Additional testing is required to resolve this issue.