Study On The Methods Of Source Apportionment Of Polycyclic Aromatic Hydrocarbons And The Application In Songhua River

To identify sources of the pollutants and determine their contribution by using effective models can provide technical support for evaluation of water environmental quality and reduce the pollutant.The paper selects sixteen kinds of PAHs which are controlled preferentially by EPA and builds source apportionment complex receptor models for source apportionment of water environment.(1)Factor Analysis with Non-negative Constraints based on Back propagation Artificial Neural Networks(FA-NNC/BP) and Positive definite matrix factorization model based on Naive Bayesian Model(PMF/NBC) were proposed..(2)The sub-classifiers of Back propagation Artificial Neural Network Models and Naive Bayesian Models are trained by eight kinds fingerprint chromatograms of PAHs. The source spectrum which is deduced through FA-NNC and PMF are input into the trained classifier as the test sample, and is used to identify the source of the pollution source.(3)The source apportionment analysis of PAHs in the water phase were done in two sections of Songhua River( one is the section of Ersong-jilin, the other is the section of Harbin-Jiamusi) using FA-NNC/BP.It indicates that the main pollution sources of Ersong-jilin are coal burning and transportation combustion in May, with the contribution rates of 84.93% and 13.33% respectively;The main pollution sources in July are gasoline engine, diesel engine, transportation combustion and coal burning, with the contribution rates of 85.51%、10.61%、1.76% and 1.29% respectively; In October, the main pollution sources are gasoline engine, power plant, coal burning, transportation combustion and diesel engine, with the contribution rates of 78.96%、17.92%、1.75%、0.73% and 0.53% respectively. Meanwhile, the section of Harbin- Jiamusi, the main pollution sources of PAHs are power plant in May, coke oven, coal burning and transportation combustion, with the contribution rates of 87.42%、8.77%、1.69% and 1.29% respectively; the main pollution sources in July are power plant, coke oven and transportation combustion, with the contribution rates of 8.28% 、 16.28% and1.98% respectively; the main pollution sources in October are coke oven, power plant, transportation combustion and gasoline engine, with the contribution rates of 68.10%、26.14%、2.98% and 1.19%.(4)The source apportionment analysis of PAHs in water phase of all section of Songhua River use PMF/NBC. It indicates that the main sources of PAHs in all section in May are coal burning(64.3%), coke oven(14.4%), gasoline engine(9.3%), transportation combustion(8.6%), diesel engine(2.8%) and residential emission(0.6%); the main sources in July are residential emission and biomass burning(72.8%), coal burning(12.8%), transportation emission and gasoline engine(5.4%), coke oven(4.9%) and power plant(4.2%); the main sources of PAHs in October are transportation emission(41.8%), coal burning(34.4%), biomass burning(9.3%), power plant(6.6%), gasoline engine(4.6%) and diesel engine(3.3%).(5)The main sources of PAHs of Songhua River are coal burning, utilizing and pyrolysis of petroleum and transportation emission. Therefore, the control of fuel and content of PAHs in the atmosphere and waste water should be strengthened. The cleaner production and the application of clean energy should be promoted. The emission standards of automobile exhaust gas should be regulated strictly and the biomass burning such as straw combustion should be prohibited strictly.