Evaluating the performance of PMF model for Atmospheric PM source apportionment in multi-site
HUANGFU Yan-qi1, TIAN Ying-ze1, DONG Shi-hao1, DAI Qi-li1, SHI Guo-liang1, ZHOU Xiao-yu1, WEI Zhen2, QIAN Yong3, FENG Yin-chang1
1. State Environmental Protection Key Laboratory of Urban Ambient Air Particulate Matter Pollution Prevention and Control, College of Environmental Science and Engineering, Nankai University, Tianjin 300350, China; 2. Anhui Provincial Environmental Monitoring Center Station, Hefei 230071, China; 3. Hefei Environmental Monitoring Center Station, Hefei 230031, China
Abstract：With the improvement of spatial resolution for environmental monitoring, spatial information which normally displayed as multi-site datasets is available and could be used for atmospheric particular matter source apportionment. PMF was performed on the combined simulated multi-site datasets that including eight scenarios (three major types). Meanwhile, ambient data of PM2.5 in Hefei was used to evaluate the performance of PMF model for source apportionment of multi-sites data. When the time series of source contributions were fully consistent with each other, it turned out that the PMF results of combined multi-site datasets were not better than these of an individual site. And the combined multi-site datasets could yield a reliable PMF result with different time series of source contributions. What's more, when the time series of source contributions were partly consistent, the PMF results generally became better, but some specific sources may have high uncertainties for some specific sources.
Chan C K, Yao X H. Air pollution in mega cities in China[J]. Atmospheric Environment, 2008,42:1-42.
[3]
Zhang Y H, Hu M, Zhong L J, etl. Regional Integrated Experiments on Air Quality over Pearl River Delta 2004(PRIDE-PRD2004):Overview[J]. Atmospheric Environment 2008,42(25):6157-6173.
[4]
Laupsa H, Denby B, Larssen S, et al. Source apportionment of particulate matter (PM2.5) in an urban area using dispersion, receptor and inverse modeling[J]. Atmospheric Environment 2009,43:4733-4744.
[5]
Chen D S, Cheng S Y, Liu L, et al. An integrated MM5-CMAQ modeling approach for assessing trans-boundary PM10 contribution to the host city of 2008 Olympic summer games-Beijing, China[J]. Atmospheric Environment, 2007,41:1237-1250.
Paatero P. Least squares formulation of robust non-negative factor analysis[J]. Chemometrics and Intelligent Laboratory Systems 1997,37:23-35.
[12]
Escrig A, Monfort E, Celades I, et al. Application of optimally scaled target factor analysis for assessing source contribution of ambient PM10[J]. J. Air & Waste Manage, 2009,59:1296-1307.
[13]
Mooibroek D, Schaap M, Weijers E P, et al. Source apportionment and spatial variability of PM2.5 using measurements at five sites in the Netherlands[J]. Atmospheric Environment, 2011,45(25):4180-4191.
[14]
Larsen B R, Gilardoni S, Stenström K, et al. Sources for PM air pollution in the Po Plain, Italy:Ⅱ. Probabilistic uncertainty characterization and sensitivity analysis of secondary and primary sources[J]. Atmospheric Environment, 2012,50:203-213.
[15]
Tian Y Z, Shi G L, Han B, et al. The accuracy of two-and three-way positive matrix factorization models:Applying simulated multisite data sets[J]. Journal of the Air & Waste Management Association, 2014,64(10):1122-1129.
[16]
Hopke PK. Recent developments in receptor modeling[J]. Journal of chemometrics, 2003,17:255-265.
Javitz H S, Watson J G, Robinson N. Performance of the chemical mass balance model with simulated local-scale aerosol[J]. Atmospheric Environment, 1988,22:2309-2322.
[19]
Zhang T, Cao J J, Tie X X, et al. Water-soluble ions in atmospheric aerosols measured in Xi'an, China:Seasonal variations and sources[J]. Atmospheric Research, 2011,102(1):110-119.
