Watson et al

Aerosol size distributions are presented for a winter intensive study at the Fresno Supersite. The size distributions were consistent with and predictive for continuous PM2.5 measured by beta attenuation. They varied temporally with respect to source type and intensity, with the smallest mean diameters associated with high NOx concentrations during weekday morning rush hours. Conversely, small and large particle and black carbon (BC) concentrations were higher during Sunday and weekday evenings in response to traffic and residential wood combustion emissions. Ambient PM2.5 light scattering (Bsp) was precisely but systematically underestimated during winter, probably because of heating in the sample shelter. INTRODUCTION The U.S. Environmental Protection Agency (EPA) Supersites program examines atmospheric measurement technologies, source contributions and control strategies, and the effects of atmospheric particles on human health.1 A major research goal at the Fresno Supersite is to evaluate nonroutine monitoring methods.2 Measurements include high-resolution meteorology, PM2.5 and PM10 mass, black carbon (BC), NO, NO2, NOx, NOy, O3, PAH, particle NO3 –, and dry PM2.5 and ambient TSP light scattering (Bsp).2 In addition, particle size distributions from 0.009 to 10 μm are measured continuously. These measurements are motivated in part by concerns that ultrafine (i.e., nanometer-sized) particles adversely affect human health.3 In this paper, aerosol size distributions at the Fresno Supersite and their relationships with various chemical and physical parameters are described. Light scattering estimated from the particle size distributions is compared with measured Bsp. It is expected that long-term, continuous measurements will be of sufficient quality to address air-quality and public-health issues. The detailed characterization of the aerosol size distribution should be particularly useful in the latter regard. METHODS The Fresno Supersite is located at 3425 First Street. A fourlane road with residential areas lies to the east, and commercial facilities, schools, and office buildings lie to the west, north, and south. The sampler is located in a second-story room with inlets ~10 m above the ground. This paper focuses on the period from December 15, 2000, through February 3, 2001, when an intensive series of continuous gas and particle and filter-based measurements was made. Particle-size distributions are measured through a PM10 inlet with a TSI SMPS 3936L10 (scanning particle mobility particle sizer: 0.009–0.392 μm diameter in 52 channels) and a PMS Lasair (LAS) 1003 optical particle counter (0.1–2 μm in 7 channels). The SMPS sizes particles based on their mobility in an electric field. While the SMPS data were adjusted for the decrease in smallparticle counting efficiency, no attempt was made to correct for particle losses in the sample stream. However, such losses are expected to be on the order of 10% for particles with diameters of 20 nm. The LAS counts and sizes individual particles according to the amount of light each scatters at 633 nm. The instrument was factory-calibrated for size with polystyrene latex beads (Duke Scientific) with a refractive index of 1.59, i0.0. Because the office is environmentally controlled, the measured sizes are assumed to represent “dry” particles during the winter intensive. Hourly-average PM10 and PM2.5 mass were measured with Met One 1020 beta attenuation monitors (BAM). Meteorological measurements [wind speed and direction, temperature, pressure, relative humidity (RH), and solar irradiance] were made on a 5-min time base. TEI 42, API 400, and McGee AE14U aethalometer monitors were used Particle Size Relationships at the Fresno Supersite John G. Watson, Judith C. Chow, and Douglas H. Lowenthal Desert Research Institute, Reno, Nevada Mark R. Stolzenburg, Nathan M. Kreisberg, and Susanne V. Hering Aerosol Dynamics, Inc., Berkeley, California IMPLICATIONS Long-term, continuous measurements of particle size distributions at the Fresno Supersite exhibit consistent and reasonable relationships with gas and particle pollutant concentrations, source emissions, and light scattering. The results suggest that such measurements will be useful for addressing air-quality as well as human-health issues.