Search

Abstract: The development of geographical information system techniques has opened up a wide array of methods for air pollution exposure assessment. The extent to which these provide reliable estimates of air pollution concentrations is nevertheless not clearly established. Nor is it clear which methods or metrics should be preferred in epidemiological studies. This paper compares the performance of ten different methods and metrics in terms of their ability to predict mean annual PM10 concentrations across 52 monitoring sites in London, UK. Metrics analysed include indicators (distance to nearest road, traffic volume on nearest road, heavy duty vehicle (HDV) volume on nearest road, road density within 150 m, traffic volume within 150 m and HDV volume within 150 m) and four modelling approaches: based on the nearest monitoring site, kriging, dispersion modelling and land use regression (LUR). Measures were computed in a GIS, and resulting metrics calibrated and validated against monitoring data using a form of grouped jack-knife analysis. The results show that PM10 concentrations across London show little spatial variation. As a consequence, most methods can predict the average without serious bias. Few of the approaches, however, show good correlations with monitored PM10 concentrations, and most predict no better than a simple classification based on site type. Only land use regression reaches acceptable levels of correlation (R 2 = 0.47), though this can be improved by also including information on site type. This might therefore be taken as a recommended approach in many studies, though care is needed in developing meaningful land use regression models, and like any method they need to be validated against local data before their application as part of epidemiological studies. [Copyright &y& Elsevier]

Abstract: Numerous epidemiological studies have shown health effects related to short- and long-term exposure to elevated levels of ambient particulate matter (PM). It is not clear however which specific characteristics (e.g., size, components) or sources of PM are responsible for the observed effects. The aim of RAPTES (Risk of Airborne Particles: a Toxicological–Epidemiological hybrid Study) was to investigate which specific physical, chemical or oxidative characteristics of ambient PM are associated with adverse effects of PM on health. This was done by performing experimental exposure of human volunteers to air pollution at several real-world settings that had high contrast and low correlation between several PM characteristics. For this goal, eight sites in the Netherlands that differed in local PM emission sources were chosen for extensive air pollution characterization. Measurement sites included an underground train station, three different road traffic sites, an animal farm, a sea harbor, a site located in the vicinity of steelworks, and an urban background site. Five- to six-hours average concentration measurements at each site were made between June 2007 and October 2009. We measured PM10, PM2.5, particle number concentration (PNC), oxidative potential of PM, absorbance, endotoxin content, as well as elemental and chemical composition of PM, and gaseous pollutants concentrations. This paper presents a detailed characterization of particulate air pollution at the sampling sites. We found significant differences in all PM characteristics between the sites. The underground train station, compared to each outdoor location, had substantially higher concentrations of nearly all PM characteristics. The average PM10 and PM2.5 mass concentrations at the underground train station were 394 μg m−3 and 137 μg m−3, respectively, which was 14.1 and 7.6 times higher than the urban background. The sum of the concentrations of trace metals in fine and coarse PM was nearly 20 times above the outdoor levels. Elemental carbon (EC) was elevated at the underground site in the fine but also in the coarse mode, in contrast to the traffic sites where EC was predominantly found in fine PM. The highest concentrations and contrasts in PNC were at the traffic sites (between 45,000 and 80,000 particles cm−3), which was several times higher than measured at any other site. Correlations of PNC with metals, PM10, PM2.5 and absorbance were low to moderate, while correlations between PM10, PM2.5 and the metals Cu and Fe were high. After excluding the underground train station data, correlations between PM10, EC and metals decreased whereas the correlation between PNC and EC increased. We conclude that we were able to successfully identify and characterize real-world situations with very different particle characteristics. High contrast and low correlations between PM characteristics, as well as consistency of these differences across sampling campaigns, provide a good basis for identifying health relevant PM characteristics in the upcoming analysis. [Copyright &y& Elsevier]