Pollution of polycyclic aromatic hydrocarbons (PAHs) in the environment is becoming increasingly serious with the development of the economy. Source analysis is the key step in reducing PAHs pollution in the environment, and existing methods are usually based on receptor models. However, when these methods are applied to the distribution of PAHs in sediments, overestimation or underestimation often occurs. Therefore, this paper proposes an improved model based on principal component analysis and multiple linear regression. The model has been applied to study the pollution of PAHs in sediments of Taihu Lake in China over several decades. Compared with existing models, principal component analysis followed by multiple linear regression of source (PCA-MLRS) can identify specific emission sources and quantify the impact of each emission, and the source resolution accuracy of the strong toxic molecules BaP and DBA is significantly higher than that of other models. Biomass combustion source is the main material source of Phe, Ant, Flu and Pyr in Taihu Lake sediment, and their contribution rates are 85.2%, 44.8%, 58.9% and 62.2%, respectively. The coal combustion source is the main material source of the 5-ring molecule BaP and the 6-ring molecules InP and BP. The contribution rates of coal combustion to BaP, InP and BP are 70.1%, 72.9% and 72.1%, respectively, and this contribution has increased since 1985. The contribution of petroleum oil combustion source and coal combustion source to PAHs in sediments of Taihu Lake is consistent. They are the main sources of heavy molecular PAHs such as BaA, BbF, BkF and DBA, and their contributions are 56.5%, 36.8%, 43.9% and 67.3% respectively. The results are related to the different emission and management characteristics of PAHs from different sources and the different behavior characteristics of low and high molecular weight PAHs in the environment.
Keywords: APCA-MLR; PAHs; PCA-MLRS; PMF; Source apportionment.
Copyright © 2019 Elsevier B.V. All rights reserved.