AAAR 32nd Annual Conference
September 30 - October 4, 2013
Oregon Convention Center
Portland, Oregon, USA
Abstract View
Sensitivity of the Chemical Mass Balance Model to Different Molecular Marker Traffic Profiles
PALLAVI PANT, Jianxin Yin, Roy M. Harrison, University of Birmingham
Abstract Number: 68 Working Group: Source Apportionment
Abstract Road traffic is one of the key sources of particulate matter (PM) concentrations in urban areas, and a good understanding of the contribution of traffic emissions to PM mass is imperative for policy action. The Chemical Mass Balance (CMB) model, based on the principle of mass balance, is routinely used for quantitative estimation of source contributions. However, the model requires locally-relevant profiles for reliable source contribution estimates (SCEs). The traditional approach for creation of source profile involves source sampling although ambient data can also be used. The aim of this study was to assess the response of the CMB-MM model to molecular marker profiles for traffic derived using different sampling approaches, i.e. dynamometer (USA), twin-site (UK) and road tunnel (France) sampling. The traffic estimates thus obtained were cross-compared using other proposed methods.
Ambient data collected at different sites in UK was analysed using the CMB8.2 software for attribution of PM$_2.5 OC mass to different sources (vegetative detritus, wood combustion, coal combustion, natural gas, crustal dust and traffic).The marker species for the traffic profile across the profile runs were monitored and cross-compared to assess the influential species in the different runs. PM$_2.5 estimates were obtained using previously published OC to PM$_2.5 source conversion ratios.
Based on comparisons with estimates from the EC tracer method and mass closures, the ambient profile was found to estimate traffic source better than other profiles although higher standard errors were reported for this profile. It was also observed that selection of species for inclusion in the profile can determine the overall modelling output, both for estimation of traffic source and the overall estimations. It can be concluded based on preliminary analysis that ambient profiles can generate reasonable estimates for traffic source. Further analysis is currently underway using a UK tunnel profile and European dynamometer profiles.