Characterizing local traffic contributions to particulate air pollution in street canyons using mobile monitoring techniques

Traffic within urban street canyons can contribute significantly to ambient concentrations of particulate air pollution. In these settings, it is challenging to separate within-canyon source contributions from urban and regional background concentrations given the highly variable and complex emissions and dispersion characteristics. In this study, we used continuous mobile monitoring of traffic-related particulate air pollutants to assess the contribution to concentrations, above background, of traffic in the street canyons of midtown Manhattan. Concentrations of both ultrafine particles (UFP) and fine particles (PM2.5) were measured at street level using portable instruments. Statistical modeling techniques accounting for autocorrelation were used to investigate the presence of spatial heterogeneity of pollutant concentrations as well as to quantify the contribution of within-canyon traffic sources. Measurements were also made within Central Park, to examine the impact of offsets from major roadways in this urban environment. On average, an approximate 11% increase in concentrations of UFP and 8% increase in concentrations of PM2.5 over urban background was estimated during high-traffic periods in street canyons as opposed to low traffic periods. Estimates were 8% and 5%, respectively, after accounting for temporal autocorrelation. Within Central Park, concentrations were 40% higher than background (5% after accounting for temporal autocorrelation) within the first 100 m from the nearest roadway for UFP, with a smaller but statistically significant increase for PM2.5. Our findings demonstrate the viability of a mobile monitoring protocol coupled with spatiotemporal modeling techniques in characterizing local source contributions in a setting with street canyons.