The Importance of Species-Specific Particle Size Distributions in the Development of Refined Surrogates for Exposure to PM2.5 of Outdoor Origin

NATASHA HODAS (1), Barbara Turpin (1), Melissa Lunden (2)

(1) Rutgers University, New Brunswick, NJ, (2) Lawrence Berkeley National Laboratory, Berkeley, CA

     Abstract Number: 247
     Last modified: November 9, 2009

Abstract
Epidemiological studies have established a positive relationship between increased morbidity and mortality and exposure to ambient fine particulate matter (PM$_(2.5)). Most epidemiological studies have used central site monitor PM$_(2.5) mass concentration as a surrogate for exposure. Human activity studies, however, show that people spend the majority of time indoors. Outdoor-generated particle concentrations and characteristics in the indoor environment differ from those in the outdoor setting. A better accounting of ambient particle penetration into and persistence in the indoor environment is crucial for reducing exposure misclassification. This research is performed under a cooperative agreement with USEPA/NERL to refine and evaluate alternative exposure metrics in the investigation of air pollution health effects.

Air exchange rate (AER), particle size, and particle chemical composition are key factors in determining the penetration and persistence of ambient PM$_(2.5) in the indoor environment. In this work, AERs derived from the Lawrence Berkeley National Laboratory (LBNL) infiltration model are paired with particle size and composition data to estimate indoor concentrations of ambient PM$_(2.5) in several New Jersey (NJ) cities and provide refined exposure surrogates for two epidemiology studies: the NJ Triggering of Myocardial Infarctions Study and the NJ Adverse Birth Outcomes Study. The LBNL infiltration model inputs, which include housing characteristics and meteorological data, are available through census and climate data archives. Chemical composition- and size-resolved particle concentration data, however, are available only in select geographic regions where intensive PM$_(2.5) studies have been performed. Size- and chemical composition-resolved PM$_(2.5) concentrations from the Pittsburgh (PITT) and New York City (NYC) EPA Supersites were used to deduce particle characteristics in NJ. Daily size distributions were examined for the number of modes, mass median diameters for all modes, and the fraction of total mass in each mode for all available sampling days for sulfate, nitrate, and carbon in PM$_(2.5). Size-specific particle penetration coefficients and deposition loss rates were assigned based on the results for each species. These values, along with the LBNL derived AERs provide inputs for the indoor-outdoor particle mass balance equation, which allows for the calculation indoor concentrations of outdoor-generated PM$_(2.5).

NJ sulfate (PM$_(2.5)) can be described as unimodal with a mass median diameter near 0.47 µm. Organic carbon and elemental carbon are bimodal with peaks near 0.08 and 0.47 µm. Values were assigned based on frequency of occurrence. For example, sulfate mass median diameter was 0.47 µm for 68% of sampling days in NYC and 60% in PITT. Similarly, a total of 76% of NYC sampling days showed a peak in nitrate PM$_(2.5) mass at a diameter of 0.47 µm, 79% of which could be described by a single mode. When results from the two sites varied considerably, a greater weight was placed on NYC data due to a higher degree of similarity between NYC (located in Queens, NY) and urban NJ PM$_(2.5) sources. Similar analyses have been repeated with data from other EPA Supersites with size- and chemical composition resolved particle concentrations to characterize PM$_(2.5) in various U.S. regions and provide important inputs for PM$_(2.5) exposure studies.