請用此 Handle URI 來引用此文件: http://hdl.handle.net/11455/5443
標題: 台灣中部都會與沿海地區PM2.5及PM2.5-10氣膠化學組成及污染源貢獻量之研究
Chemical Compositions and Source Apportionment to PM2.5 and PM2.5-10 Aerosols at Urban and Coastal Areas in Central Taiwan
作者: 邱嘉斌
Chio, Chia-Pin
關鍵字: Principal Component Analysis
主成份因子分析法
Chemical Mss Blance
Source Contribution
Bayesian Model
Mass Reconstruction
Vehicle Emissions
Vegetative Burning
Secondary Aerosol
化學質量平衡法
污染源貢獻量
貝氏模式
質量再重組法
交通排放
農廢燃燒
二次氣膠
出版社: 環境工程學系
摘要: 台灣中部都會與沿海地區最近幾年發生高污染事件仍以PM10為主要污染物,因此本研究的目的在於探討此地區PM2.5及PM2.5-10氣膠化學組成及污染源的貢獻量,都會及沿海地區的測站分別選擇台中市崇倫國中與梧棲鎮梧棲國小作為代表測站,當地的氣象觀測站與空氣品質監測站均鄰近此兩測站,實驗方面利用雙粒徑分道採樣器採集PM2.5及PM2.5-10氣膠微粒,並利用離子層析儀、元素分析儀與感應耦合電漿光譜分析儀分析其化學組成,主要採樣時間為1998年8月至1999年3月,都會測站共採集110組大氣氣膠樣本,而沿海地區則有92組。數據分析方面則利用主成份因子分析法與化學質量平衡法(CMB7)作為PM2.5及PM2.5-10氣膠微粒的污染源定性與定量之工具,同時應用貝氏模式、質量重組及高斯傳遞係數軌跡模式模擬案例。 依據PCA與CMB7模式分析中部都會與沿海地區之PM2.5及PM2.5-10氣膠微粒之結果,兩地區PM2.5及PM2.5-10氣膠微粒之污染源趨勢相當類似,PM2.5氣膠微粒之污染源以交通排放與二次氣膠為主,都會地區來自交通排放約佔34.1~84.8 %,沿海地區則約42.7~69.3 %來自交通排放,都會地區二次氣膠約佔17.8~28.4 %,沿海地區則介於18.1~25.9 %之間。而PM2.5-10氣膠微粒之污染源則以地殼物質與交通排放所佔比例較高,特別是都會地區中來自地殼物質約佔51.9~76.7 %,沿海地區則約24.0~74.8 %來自地殼物質。 結合兩氣膠粒徑數據,其中交通排放為都會區PM10氣膠微粒中最重要的污染源,其次分別為地殼物質、二次氣膠、農廢燃燒、工業排放及海鹽飛沫,而沿海地區除了海鹽飛沫的貢獻被發現較都會地區顯著外,亦有相同之污染源趨勢,研究發現兩地區之交通排放雖為第一大污染源,但高污染事件日發生時,農廢燃燒與二次氣膠為造成PM10濃度增加的主要來源,這些高污染事件日都會地區通常發生在秋季與冬季,而在沿海地區則只有可能發生在春季。 利用CMB7模式分析之個別污染源貢獻量散佈於貝氏模式所推估的5~ 95 %區間內,但適當的污染源選擇與足夠的大氣樣本往往可影響其結果,比較CMB7模式與其他模式之分析結果,CMB7模式分析之個別污染源貢獻量與質量再重組法(MRC)比較接近,而利用CMB7與GTx模式分析個別污染源貢獻量之差距較為顯著,特別是硫酸銨與硝酸銨,然而,研究發現PM2.5氣膠微粒兩者之解析有良好的關係,並以秋冬季節的數據解析結果之相關性較佳,GTx模式之優點是可提供受體點反軌跡途徑之污染貢獻量,其軌跡有助於解釋其影響的污染源。
PM10 was still the major pollutant during episodic events occurred in Taichung urban and coastal areas in recent years. Therefore the purpose of this study was to investigate the chemical compositions and the source contributions for PM2.5 and PM2.5-10 in these areas. In this study, Chunglun junior high school and Wuchi elementary school were chosen as the representative sites for Taichung urban and coastal areas, respectively. Weather service office and air quality monitoring stations were nearby the selected sites. Experimentally, dichotomous samplers were used to collect PM2.5 and PM2.5-10 aerosols, and the chemical compositions were analyzed by using ion chromatograph (IC), elemental analyzer (EA) and inductively coupled plasma mass spectrometry (ICP-MS). Totally 110 urban samples and 92 coastal samples were collected from August 1998 to March 1999. Principal component analysis (PCA) with varimax rotation and a chemical mass balance model version 7 (CMB7) were used to qualify and quantify, respectively, the source contributions to PM2.5 and PM2.5-10. In addition, Bayesian model, Mass Reconstruction (MRC) and Gaussian Trajectory transfer-coefficient modeling system (GTx) were also applied to analyze the special cases in this study. The results obtained by using the PCA and the CMB modeling showed a similar pattern on the source contributions to PM2.5 and PM2.5-10 for both areas. Vehicle emissions and secondary aerosols were the major sources for PM2.5. Quantitatively, vehicle emissions accounted for 34.1 % to 84.8 % of the urban PM2.5 and 42.7 % to 69.3 % for the coastal PM2.5. There were 17.8 % to 28.4 % of PM2.5 contributed from secondary aerosols in urban area, while the secondary aerosols contributed 18.1 % to 25.9 % of the coastal PM2.5. However, crustal materials and vehicle emissions were the major sources for PM2.5-10. Especially the crustal materials contributed 51.9 % to 76.7 % of the urban PM2.5-10 and 24.0 % to 74.8 % of the coastal PM2.5-10. In conclusion, vehicle emissions, crustal materials, secondary aerosols, vegetative burning, industry emissions and marine spray were the sources for PM10 in sequence in the urban area. The similar pattern was also observed in the coastal area, except the marine spray was found more significantly than that in the urban area. The contribution emitted from vehicle emission was indeed the most significant source during the episodic periods. But based on the increments of the contributory percentage, the vegetative burning and secondary aerosols were the influential sources causing the elevated PM10 at both sites during episodic periods. These episodic events always occurred in fall and winter in the urban area, but they could only happen in spring in the coastal area. The contributions from each source calculated by using CMB7 model ranged from 5% to 95 % of those estimated by using Bayesian model. However, the validity of the source compositions and the sufficient aerosol samples were the limitations for using Bayesian model. Intercomparison of these results showed a better agreement between the estimations by using CMB7 and MRC. Although, there were differences of the individual source contributions in PM10 obtained by CMB7 and GTx models, especially for the contributions of ammonium sulfate and nitrate, however they were well correlated for the data obtained in fall and winter. Furthermore, the advantages of using GTx model could provide the backward trajectory from the receptor and the information of the major pollution sources.
URI: http://hdl.handle.net/11455/5443
顯示於類別:環境工程學系所

文件中的檔案:
沒有與此文件相關的檔案。


在 DSpace 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。