Receptor modeling of ambient and personal exposure samples: 1998 Baltimore Particulate Matter Epidemiology-Exposure Study

Sources of particulate matter exposure for an elderly population in a city north of Baltimore, MD were evaluated using advanced factor analysis models. Data collected with versatile air pollutant samplers positioned at a community site, outside and inside of an elderly residential facility were analyzed with a three-way analysis to identify the source(s) that contributed to all sample types. These sources were secondary sulfate, secondary nitrate, motor vehicles, and a organic carbon (OC). The OC source contained 96% OC and most likely represents positive volatile organic carbon artifact and other unidentified sources. No soil source was found that contributed significantly to these samples. A second set of data was collected with personal samplers (PEM) from 10 elderly subjects, their apartments, a central indoor location, and outdoors. The PEM data were analyzed using a complex model with a target for soil that included factors that are common to all of the types of samples (external factors) and factors that only apply to the data from the individual and apartment samples (internal factors). From these results, the impact of outdoor sources and indoor sources on indoor concentrations were assessed. The identified external factors were sulfate, soil, and an unknown factor. Internal factors were identified as gypsum or wall board, personal care products, and a factor representing variability not explained by the other indoor sources. The latter factor had a composition similar to outdoor particulate matter and explained 36% of the personal exposure. External factors contributed 63% to personal exposure with the largest contribution from sulfate (48%). (C) 2003 Elsevier Science Ltd. All rights reserved.