Exposure to fine particulate air pollution in Prince George, British Columbia, Canada

The relationship between ambient fine particulate pollution (PM_2.5) and children’s personal exposure was investigated during the winter of 2001 in the city of Prince George, Canada. Harvard Personal Environmental Monitors (HPEMS) were used to collect personal exposure samples from 15 children aged 10 to 12 years and ambient samples at five outdoor monitoring sites located on their elementary school roofs. PM_2.5 mass, sulphate (SO₄^2-) and absorption coefficient (ABS) were determined for all samples (ABS as a surrogate for elemental carbon (EC)). SO₄^2- and EC personal/ambient ratios were used as tracers of ambient PM_2.5 to estimate personal exposure to ambient and non-ambient sources. Overall, ambient PM_2.5 concentrations were very high for the study period with levels at the downtown school exceeding the Canada-Wide Standard for 2001. Inversion conditions were found to be responsible for all high ambient concentrations (>30µg/m³) and personal exposures were also impacted. Although spatial differences were found for all three measures, high correlations were found between the central PM_2.5 TEOM monitor and each outdoor study site suggesting the TEOM could be used to represent relative ambient concentrations at all sites. Personal exposure and ambient concentrations showed stronger personal-ambient correlations for SO₄^2- (r=0.96) and ABS/EC (r=0.73) compared to total PM_2.5 (r=0.52). A strong association was found between estimates of ambient generated exposure (derived from personal/ambient SO₄^2- and EC ratios) and ambient concentration at both the closest school monitor (r=0.92) and the central site (r=0.88), suggesting that the central site TEOM was also suitable for assessing average ambient generated personal exposure throughout the city during the study period. Individual longitudinal regressions showed that ambient generated personal exposure was 45% of ambient concentration. Overall ambient generated exposures were slightly higher than non-ambient exposures with each contributing almost equal proportions to total personal exposure. This study describes the relationship between ambient concentration and personal exposure to PM_2.5 and demonstrates that the sulphate and elemental carbon components are reliable tracers for exposure to particles of ambient origin. The strong association found between ambient generated exposure and ambient concentration suggests that ambient levels are indeed an appropriate surrogate of exposure to ambient PM_2.5 sources. The equal contributions made by ambient and non-ambient sources to total PM_2.5 personal exposure, even when subjects were indoors 95% of the time, demonstrates the importance of managing ambient air quality. These findings strongly support the use of ambient data for a longitudinal health study in this city. Use of the regression equations provided will enable future estimation of ambient generated exposures for healthy elementary school children during the winter in the city of Prince George.