Continuous weeklong measurements of personal exposures and indoor concentrations of fine particles for 37 health-impaired North Carolina residents for up to four seasons

A study of personal exposures and indoor and outdoor concentrations of particles was carried out in 2000–2001 for 37 health-impaired residents of North Carolina. Earlier papers have dealt with the 24-h integrated gravimetric samples; this report adds the continuous data (1-min resolution) using optical scattering devices (personal data RAMs, or pDRs) for personal and indoor measurements. Subjects and their households were sampled for 7 consecutive days in each of four seasons, although not all subjects completed all four seasons. Each subject completed an activity diary with 15-min resolution on each day giving his or her presence in one of six locations and describing activities with the potential for increased exposure to particles. More than 800 person-days and 1.1×106 person-min of valid personal exposures and indoor air concentrations were collected.The pDRs compared well with the PM2.5 gravimetric devices, with an R2 of 87% compared to the indoor Harvard impactor (HI), and 70% compared to the personal exposure monitor (PEM). As found in previous studies, (and as expected due to the calibration of these devices with an aerosol of higher density than normal ambient and indoor aerosols), the pDRs overestimated the gravimetric concentrations by 20–50%. Because the correction factor varied by season and by type of sample (indoor vs. personal), no overall correction factor could be applied. The estimated mean increases in exposure during three common activities (cooking, cleaning, and personal care) were 56, 28, and 20 μg m−3, respectively. Several less common activities, such as burning food and using a fireplace were found to result occasionally in very high (>1500 μg m−3) short-term peaks of exposure. Although households with smoking were ineligible for participation, two households did have significant smoking, and had the highest average personal exposures and indoor concentrations of all the study homes. The diurnal variation of personal exposures and indoor concentrations was bimodal, with peaks occurring between 10 and 11 AM and 5 and 6 PM. The “damping effect” of the home due to air exchange caused the infiltrated outdoor particles to show only a unimodal variation, with a single broad and mild peak occurring between 6 and 10 AM.A method for identifying peak concentrations due to indoor sources and personal activities was developed. Peaks due to personal activities and indoor sources occurred about 23% and 14% of the time, respectively. On average, outdoor particles contributed about half of the total personal exposures and indoor concentrations, but there was wide variation across subjects. These findings using the MIE data were then compared to previously published estimates using measurements of sulfur from the gravimetric data; good agreement (R2 values from 50% to 65%) was found, providing a measure of validation to both methods. Personal exposures exceeded co-located indoor concentrations by 2–3 μg m−3, confirming findings of a small but statistically significant “personal cloud” for PM2.5 made in previous studies using gravimetric 24-h integrating monitors.