An investigation of PM _2.5 climatology at a rural Midwest site

An intensive field campaign was conducted September - December 2002 to characterize the physical and chemical properties of ambient fine particulate matter at a rural Midwest site. This study, funded by the Central Regional Air Planning Association (CENRAP) through USEPA Region VII, was designed to elucidate aerosol dynamics and climatology near the centroid of the CENRAP domain. Measurements were conducted at the Reserve, Kansas, IMPROVE Protocol site which commenced operation in the summer of 2002. Aerosol properties measured at high (hourly or finer) time resolution included PM _2.5 mass, black carbon, sulfate and nitrate; daily 24-hour integrated measurements for PM _2.5 chemical speciation and also precursor gases were conducted during two six-week intensives. The data set is publicly available through the Visibility Information Exchange System (VIEWS, http://vista.cira.colostate.edu/ views/). Diurnal profiles for PM _2.5 mass, black carbon, nitrate, and sulfate demonstrate the dominant role of atmospheric ventilation as augmented by species-specific physicochemical dynamics. Daily 24-hour integrated PM _2.5 mass reconstructions show the episodic nature of sulfate in the summer/fall and nitrate in the fall/winter; organic carbon, elemental carbon, and crustal material concentrations are relatively constant on a day-to-day basis (at least for this study period). Ion balances performed on the daily 24-hour integrated samples are consistent with the conceptual model for the conditions conducive to ammonium nitrate formation. A regional sulfate episode is placed in a regional context by comparing to estimates from the Naval Research Laboratory (NRL) Aerosol Analysis and Prediction System (NAAPS) model and by comparing observed hourly sulfate concentrations to contemporaneous semicontinuous sulfate measurements in St. Louis, MO. Finally, a cluster analysis on air mass back trajectories demonstrates the extent to which synoptic scale weather conditions for the sampling period are representative of other years.