Determinants of black carbon, particle mass and number concentrations in London transport microenvironments. (July 2017)

• Record Type: Journal Article
• Authors: Rivas, Ioar; Kumar, Prashant; Hagen-Zanker, Alex; Andrade, Maria de Fatima; Slovic, Anne Dorothee; Pritchard, John P.; Geurs, Karst T.
• Is Part Of: Atmospheric environment. Volume 161(2017)
• Page Start: 247
• Page End: 262
• Publication Date: 2017-07
• Subjects: Personal exposure assessment -- Transport mode -- Commuting -- Linear regression -- Extrapolation; ANCOVA analysis of covariance -- BC black carbon -- CERC Cambridge Environmental Research Consultants -- GLM Generalised Linear Models -- GPS Global Positioning System -- IDW Inverse Distance Weighting -- IMD Index of Multiple Deprivation -- LSOA Lower Layer Super Output Areas -- NOW non-openable window -- OW openable window -- OD Origin-Destination -- PM particulate matter -- PNC ultrafine particle number concentrations -- r Pearson correlation coefficient -- UB urban background; Air -- Pollution -- Periodicals; Air -- Pollution -- Meteorological aspects -- Periodicals; 551.51
• DOI: 10.1016/j.atmosenv.2017.05.004
• Languages: English
• ISSNs: 1352-2310

Abstract

Abstract: We investigated the determinants of personal exposure concentrations of commuters' to black carbon (BC), ultrafine particle number concentrations (PNC), and particulate matter (PM1, PM2.5 and PM10 ) in different travel modes. We quantified the contribution of key factors that explain the variation of the previous pollutants in four commuting routes in London, each covered by four transport modes (car, bus, walk and underground). Models were performed for each pollutant, separately to assess the effect of meteorology (wind speed) or ambient concentrations (with either high spatial or temporal resolution). Concentration variations were mainly explained by wind speed or ambient concentrations and to a lesser extent by route and period of the day. In multivariate models with wind speed, the wind speed was the common significant predictor for all the pollutants in the above-ground modes (i.e., car, bus, walk); and the only predictor variable for the PM fractions. Wind speed had the strongest effect on PM during the bus trips, with an increase in 1 m s −1 leading to a decrease in 2.25, 2.90 and 4.98 μg m −3 of PM1, PM2.5 and PM10, respectively. PM2.5 and PM10 concentrations in car trips were better explained by ambient concentrations with high temporal resolution although from a single monitoring station. On the other hand, ambient concentrations with high spatial coverage but lower temporal resolution predicted better the concentrations in bus trips, due to bus routes passing through streets with a high variability of traffic intensity. In the underground models, wind speed was not significant and line and type of windows on the train explained 42% of the variation of PNC and 90% of all PM fractions. Trains in the district line with openable windows had an increase in concentrations of 1 684 cm −3 for PNC and 40.69 μg m −3 for PM2.5 compared with trains that had non-openable windows. The results from this work can be used to target efforts to reduce personal exposures of London commuters. Graphical abstract: Highlights: Wind speed and ambient concentrations were main predictors in above-ground modes. Ambient PM concentrations explained more variability than wind speed. Wind speed had a strongest effect on bus concentrations compared to car and walking. Underground line and train's type of windows explained 90% of the variation in PM. PM in bus was better explained by ambient concentrations at high spatial resolution.