| As an important kind of atmospheric pollution,atmospheric particulate matter pollution has become one of the most serious environmental problems for many cities in China in recent years.Atmospheric particles have complex structure and large specific surface areas for absorbing many heavy metals in the atmosphere very easily.The presence of heavy metals absorbing in atmospheric particulate matter can not only pose a potential threat to the ecological environment but also may enter the human body through inhalation and have adverse physiological effects.Due to the differences of the geographical location,climatic conditions,industrial structure and management measures for different cities in China,heavy metals in atmospheric particulate matter always show various pollution characteristics.Thus,a comprehensive understanding of the pollution of atmospheric heavy metals such as their pollution characteristics,possible sources and environmental and health risk,is needed for us to gain insight into the air pollution status in China,which is also crucial for the government to make the environmental policies and carry out the prevention work for urban atmospheric pollution.In this study,PM2.5 samples were collected on clear and hazy days in four seasons from Xianlin and Gulou campus of Nanjing University in Nanjing,a typical city of Yangtze River Delta.The chemical composition and morphology of PM2.5 were analyzed to identify its main sources.The modified BCR(the European Community Bureau of Reference)sequential extraction procedure was applied for partitioning different chemical speciation of heavy metals(As,Cd,Cr,Cu,Fe,Mn,Ni,Pb and Zn)in PM2.5 samples.The seasonal variations and temporal distribution of heavy metals in PM2.5 were analyzed from a macro perspective.Meanwhile,the speciation distribution pattern of metals as well as the potential environmental and health risk caused by heavy metals in PM2.5 were analyzed and assessed from a micro perspective.Correlations of heavy metals in PM2.5 between meteorological factors(such as wind speed,wind direction,temperature,humidity and barometrie pressure)and other atmospheric pollutants(such as SO2,NO2,CO,O3,PM10 and PM2.5)were studied.Then the rapid prediction model for heavy metals in PM2.5 was established by using artificial neural networks and support vector machine.The main results of this study are as follows:(1)The highest concentrations of PM2.5 collected from the two sampling sites were found in winter,followed by spring,whereas the lowest concentrations were found in summer.The PM2.5 concentrations of most samples collected in winter and spring were more than the national ambient air quality standard.Traffic pollution is the important factor affecting PM2.5 pollution.Heavy metals accounted for 1.17%to 3.34%of the total mass of PM2.5.(2)The mean concentrations of metal elements in PM2 5 in different sites and seasons decreased as Ca>Na>K>Fe>Al>Mg>Zn>Pb>Mn>Ti>Cu>Cr>Ba>Ni>As>Sr>V>Cd>Mo>Co.Similar to the seasonal distribution of PM2.5,most metals exhibited the highest concentrations in winter and spring,and the lowest concentrations in summer or autumn.Heavy metals in PM2.5 in Nanjing showed a medium pollution level compared with other Chinese cities.However,the concentrationos of some metals exceeded the corresponding ambient air quality standards when haze occurred in the winter,which indicates the high pollution degree of atmospheric heavy metal pollution in Nanjing from another perspective.According to the source analysis,industrial emissions,transportation activities,residents’ daily life activities and road dust were the main sources of heavy metals in PM2.5.(3)Based on the analysis of chemical partitioning,most heavy metals that mainly come from anthropogenic sources had higher contents of exchangeable fraction,whereas the earth’s crust element Fe had higher contents of the oxidizable and residual fractions,which can be thus considered as relatively stable in the environment.The calculated bioavailability index(Bi)of different heavy metals decreased as Zn>As>Pb>Cd>Mn>Cu>Cr>Ni>Fe.The contents of the four fractions for most metals showed negative correlation with the environment temperature,and positive correlation with the atmospheric pressure and humidity.Wind had different effects on the speciation distribution of heavy metals.Compared with Xianlin,the traffic sources and human daily life activities in Gulou may increase the environmental activity for some metals.(4)The results of health risk assessment showed that the carcinogenic risk for adults and children caused by inhalation of heavy metals in PM2.5 were all above the acceptable level(10-6),whereas the hazard index values of non-carcinogenic risk for the studied metals were all lower than the safe level of 1.For the two subpopulations,ehildren had greater health risks than adults.The health risk calculated for the samples collected in Gulou was a little higher than Xianlin.(5)Cd,Mo,Pb,Zn,Cu and As should be paid special attention from the view of enrichment dgree;Special attention needs to be paid to the metals such as Zn,As,Pb and Cd judging from their environmental risk.Meanwhile,As,Mn and Cr should be noticed according to the health risk assessment results.(6)Heavy metals in PM2.5 showed significant positive correlations between the barometric pressure and atmospheric pollutants such as SO2,NO2,CO,PM10 and PM2.5,and significant negative correlations between wind speed,temperature and O3 concentrations.In order to forecast the concentrations of As,Cd,Cr,Cu,Fe,Mn,Ni,Pb and Zn in PM2.5,the rapid prediction model was then established which used tihie meteorological factors and atmospheric pollutants as input factors by principal component analysis-neural network and principal component analysis-support vector machine.The results showed such prediction model can gain ideal forecasting results. |
