Sensitivity of source apportionment results to mobile source profiles

Abstract

The sensitivity of a source apportionment model to mobile source profiles was examined to determine the impact of using non-local mobile source profiles in chemical mass balance (CMB) models. We examined the impact of USA and Chinese mobile source profiles on source apportionment results in St. Louis, Missouri, and Beijing. The results showed that the use of non-local mobile source profiles did not impact the model apportionment results for vegetative detritus and biomass burning, but other primary source contributions were influenced by the use of non-local source profiles. Secondary organic carbon (SOC) contributions estimated by the CMB models with local and non-local profiles were compared to estimate of SOC from the EC tracer method and were found to be consistent with little bias. The results also showed that it is feasible to use the USA mobile profiles in China while model results were biased by using Chinese mobile profiles in the USA. Monthly and annual average concentrations of molecular markers in the source apportionment model showed lower sensitivity to source profiles than daily measurements, which has implications to the design of source apportionment studies.