Supplementary Material to "wintertime Aerosol Chemistry and Haze Evolution in an Extremely Polluted City of North China Plain: Significant Contribution from Coal and Biomass Combustions"

Haiyan Li,Qi Zhang,Qiang Zhang,Chun-Rong Chen,Litao Wang,Zhe Wang,Shunhua Zhou,Caroline L. Parworth,Bo Zheng,Francesco Canonaco,Andrê S. H. Prévôt,Ping Chen,Hongliang Zhang,Kebin He
DOI: https://doi.org/10.5194/acp-2016-1058-supplement
2017-01-01
Abstract:Analysis to determine the best solution for ME-2Using the multi-linear engine (ME-2) implemented with the toolkit SoFi (Source Finder), organic source apportionment was performed with the so-called a value approach.The a value determines the extent to which the resolved factor profiles (fj,solution) or time series (gi,solution) are allowed to vary from the input ones (fj, gi) (Canonaco et al., 2013):To separate an individual HOA factor in addition to CCOA, BBOA and OOA, we only constrained the mass spectra of HOA in this study, using the reference profile from Ng et al. (2011).Some important criterions to select the optimal solution with a value varying from 0 to 1 are shown in Fig. S2-S6.Generally, increasing a value decreases Q and the ratio of Q to Qexp, due to additional degrees of freedom.However, the size of the change of Q/Qexp over different a values is not significant (Fig. S2a).In addition, the explained variation (EV) could indicate how much variation in each variable is explained by each factor.According to Paatero ( 2004), a variable should be considered only if the unexplained EV for that variable is less than 25% (Fig. S2c).Combining the analysis of factor profiles and time series of each factor across different a values, there is no significant variations for CCOA and OOA.By contrast, with different constraint levels, large changes were found for HOA and BBOA, which were mixed during PMF analysis.As shown in Fig. S3a, moving from a constrained run to a less constrained situation apportioned more signals to larger m/z for HOA profile.For a values from 0.2 to 1.0, fractions of m/z smaller than 50 for BBOA profile were significantly enhanced and the time series of BBOA were similar to those of OOA.These results indicated that solutions for a value of 0 and 0.1 were more reasonable.This was further evidenced by the diurnal profiles of each factor and their correlations with external tracers over different a values.To allow some degrees of freedom for model run, we chose the results for a value of 0.1 to be the best solution in this study.
What problem does this paper attempt to address?