Biomass burning (BB) is a significant source of atmospheric particles in many parts of the world. Whereas many studies have demonstrated the importance of BB emissions in central and northern Europe, especially in rural areas, its impact in urban air quality of southern European countries has been sparsely investigated. In this study, highly time resolved multi-wavelength absorption coefficients together with levoglucosan (BB tracer) mass concentrations were combined to apportion carbonaceous aerosol sources. TheAethalometer model takes advantage of the different spectral behaviour of BB and fossil fuel (FF) combustion aerosols. The model was found to be more sensitive to the assumed value of the aerosol Ångström exponent (AAE) for FF (AAEff) than to the AAE for BB (AAEbb). As result of various sensitivity tests the model was optimized with AAEff = 1.1 and AAEbb = 2. The Aethalometer model and levoglucosan tracer estimates were in good agreement. The Aethalometer model was further applied to data from three sites in Granada urban area to evaluate the spatial variation of CMff and CMbb (carbonaceous matter from FF or BB origin, respectively) concentrations within the city. The results showed that CMbb was lower in the city centre while it has an unexpected profound impact on the CM levels measured in the suburbs (about 40%). Analysis of BB tracers with respect to wind speed suggested that BB was dominated by sources outside the city, to the west in a rural area. Distinguishing whether it corresponds to agricultural waste burning or with biomass burning for domestic heating was not possible. This study also shows that although traffic restrictions measures contribute to reduce carbonaceous concentrations, the extent of the reduction is very local. Other sources such as BB, which can contribute to CM as much as traffic emissions, should be targeted to reduce air pollution.

No abstract

The performance and the uncertainty of receptor models (RMs) were assessed in intercomparison exercises\ud employing real-world and synthetic input datasets. To that end, the results obtained by different\ud practitioners using ten different RMs were compared with a reference. In order to explain the differences\ud in the performances and uncertainties of the different approaches, the apportioned mass, the number of\ud sources, the chemical profiles, the contribution-to-species and the time trends of the sources were all\ud evaluated using the methodology described in Belis et al. (2015).\ud In this study, 87% of the 344 source contribution estimates (SCEs) reported by participants in 47\ud different source apportionment model results met the 50% standard uncertainty quality objective\ud established for the performance test. In addition, 68% of the SCE uncertainties reported in the results\ud were coherent with the analytical uncertainties in the input data.\ud The most used models, EPA-PMF v.3, PMF2 and EPA-CMB 8.2, presented quite satisfactory performances\ud in the estimation of SCEs while unconstrained models, that do not account for the uncertainty in\ud the input data (e.g. APCS and FA-MLRA), showed below average performance. Sources with well-defined\ud chemical profiles and seasonal time trends, that make appreciable contributions (>10%), were those\ud better quantified by the models while those with contributions to the PM mass close to 1% represented a\ud challenge.\ud The results of the assessment indicate that RMs are capable of estimating the contribution of the major\ud pollution source categories over a given time window with a level of accuracy that is in line with the\ud needs of air quality management

No abstract

No abstract