Souce Apportionment of Carbonaceous Aerosols Using Integrated Multi-Variant and Source Tracer Techniques annd a Unique Molecular Marker Data Set.

摘要

Four hundred fine particulate matter samples from two sites in the Los Angeles Basin were analyzed for molecular marker source tracers, and the results were used in three source apportionment models to obtain daily, monthly and the annual average source contributions to fine particle organic carbon (OC). Good agreement between the source contribution from mobile sources and biomass burning for the chemical mass balance (CMB) model and the positive matrix factorization (PMF) models were obtained and provide additional weight of evidence that these source apportionment techniques aresufficiently accurate for policy development. However, the CMB model did not quantify primary biogenic emissions, which were quantified by the PMF model, and were included in other sources with secondary organic carbon (SOC) in the CMB model. The PMF apportionment results demonstrate seasonal patterns in the split between SOC and primary organic carbon (POC), which emphasize the biases that can result from previous short term intensive studies used to represent the annual average source contributions as well as source contributions in other seasons than those examined. The PMF model also provided new insight into the differences in composition and impacts of forest fires and high wintertime wood burning events. PMF and a second multi-variant receptor model (UNMIX) were unable to separate source contributions from diesel and gasoline engines. However, a new multi-variant receptor model, Interactive Confirmatory Factor Analysis (ICFA) was able to separate mobile sources into diesel, gasoline and smoking engines.