Research On Prediction,Forewarning And Health Effects Of Atmospheric Pollution In Main Cities Of China

Booming economic growth,rapid industrial development and accelerated urbanization have brought heavy atmospheric pollution in China,which is a threat to human health.For this purpose,several atmospheric pollutants concentration and meteorological data of major cities in China were collected.The spatial and temporal distribution characteristics of atmospheric pollutants were analyzed,and three models were applied to predict the concentration of atmospheric pollutants in typical cities.Meta-analysis was applied to preliminary explored the health effects of atmospheric pollutants on human health in China.Several conclusions were listed below:?1?The temporal and spatial distribution of air pollutants were obviously in China.PM_2.5,PM₁₀,SO₂,and CO showed a downward trend yearly,with the annual average of PM_2.5.5 Exceeding China's air quality secondary standard,while NO₂ and O₃showed a first decline and then an upward trend.The annual change of PM_2.5,NO₂,SO₂,and CO showed a“U”curve,peaking from June to August;O₃ showed an inverted“U”curve,with peak ranging from May to June;PM₁₀0 showed a bimodal curve peaking in March and December respectively.O₃ were significantly higher in summer and autumn than that in winter and spring,and the others pollutants were the opposite.In terms of spatial distribution characteristics,all pollutants in North China showed higher levels,and coarse-grain pollution in the northwest region showed a relatively high level,while O₃ was more serious in the entire northern region.?2?Three models including BP neural network,support vector machine and wavelet support vector machine,were selected to predict six pollutants in seven representative cities in China.It showed that the wavelet support vector machine was more accurate than single support vector machine and BP neural network.And the prediction accuracy was better,which showed that the model has certain generalization ability and effectiveness.The comparative study about the same pollutant in different cities showed that the fluctuation of the pollutant concentration sequence would worsen the prediction effect.?3?51 articles were collected into Meta-analysis.The results showed that the increasing concentrations of PM_2.5,PM₁₀,SO₂,NO₂,CO,and O₃ leads to total mortality,cardiovascular and cerebrovascular diseases mortality,respiratory disease mortality and outpatient hospitalization.For example,with 10?g/m³ increase of PM₁₀,SO₂,NO₂,and O₃,the total mortality increased by 0.3%?95%CI:0.2%,5%?,1%?95%CI:0.7%,1.3%??1.3%?95%CI:0.9%,1.8%?,and 0.5%?95%CI:0.3%,0.6%?.Meta-regression analysis showed that the region may be the main factor for impact of atmospheric pollutants on human health.Subgroup analysis showed that the health effects of different pollutants in the southern and northern subgroups may be enhanced,and only the effect of SO₂ on cardiovascular and cerebrovascular hospitalization existed Begg publication bias.