有害空氣污染物之來源分析與風險評估

Source Apportionment and Risk Assessment of Hazardous Air Pollutants

吳思穎(Szu-Ying Wu)

指導教授 ： 吳章甫

國立臺灣大學/公共衛生學院/環境衛生研究所/碩士(2006年)

https://doi.org/10.6342/NTU.2006.01773

摘要

近來，空氣品質隨著經濟的發展而日趨惡化，暴露於空氣污染對於人類健康所造成的不良效應也漸漸受到重視。空氣污染是由各種不同污染源累加而成的混合物，各個污染源對於健康風險的貢獻度也不同，因此，分析空氣污染中的主要來源及其健康風險將可以幫助有關當局建立更有效率的管制策略，以降低民眾因暴露於空氣污染所造成的危害。本研究分為兩部份，第一部份為分析主要的空氣污染來源，第二部份則利用分析結果進行各污染源的健康風險評估。
本研究利用受體模式 (receptor model) Multilinear Engine (ME)推估美國西雅圖地區主要的空氣污染源並量化其貢獻。有別於一般相關研究，懸浮微粒(PM2.5)和氣狀揮發性有機污染物(VOCs)的化學組成同時被放入模式中分析，並比較加入VOCs後是否可以幫助模式分辨更多的汙染源。結果顯示，利用PM2.5的化學組成資料即可辨識出大部分的汙染源，加入VOCs 資料僅幫助我們辨識出另一個VOCs 的背景來源。生質燃燒,二次生成硫酸鹽及硝酸鹽類是PM2.5的主要貢獻者,而VOCs的背景濃度、柴油、生質燃燒則是VOCs的重要來源。此外，本研究將黑碳 (black carbon)的測值與模擬之柴油及生質燃燒微粒進行回歸分析以評估黑碳是否為柴油微粒的代表性指標，結果發現，黑碳與柴油微粒和生質燃燒微粒於冬季期間 (10月至2月) 都呈現良好的相關性，因此於冬季期間，黑碳並非一個單純代表柴油微粒的指標。在風險評估部分，本研究利用來源分析的資料，以汙染源中各物質的濃度乘上單位風險 (unit risk) 估計各汙染源的致癌風險值。結果顯示，除了海洋氣膠外，其餘各汙染源的致癌風險值皆超過1×10-6，其中以VOCs的背景來源風險值最高。雖然加入VOCs 資料進行來源分析對於分辨相似汙染源的貢獻並不非常顯著,但是對於風險評估部分，VOCs確實可以幫助提供更完整的健康風險值。

關鍵字

受體模式 ； 來源分析 ； 風險評估

並列摘要

Recently, health impacts from exposures to air pollution have been widely studied. To develop effective control strategies to reduce population risks to complex air pollution, it is important to identify major sources and quantify their health risks. In this thesis, the sources of PM2.5 and VOCs were identified in the first part and the source specific health risks were calculated in the second part.
Speciated PM2.5 and VOCs collected at Beacon Hill, WA between 2000-2004 were analyzed with the multivariate receptor model, Multilinear Engine (ME). The model was performed with (1) only PM2.5 data, and (2) PM2.5 with gaseous air toxic data to compare the model’s performances. The results showed that ME model identified most sources regardless if the gaseous air toxic data were included or not. An additional source related to the background concentration of VOCs was identified when VOCs data were included. Black carbon (BC) measured from aethalometer was regressed against the apportioned diesel particle matter (DPM) and wood burning particle to evaluate if BC a good indicator of DPM. BC correlated well with both the DPM and vegetative burning particles during the heating season (Oct-Feb). Therefore, BC does not represent the actual DPM or wood burning particle mass concentrations without further detailed calibrations.
The source-specific risks were estimated by summing the products of individual air toxics concentrations in each source profile and its unit risk factor. The health risk estimates by sources showed that all sources except ‘marine’ gave a sum of cancer risks higher than 1×10-6, with the highest risk from the background VOCs sources. The improvement from having additional VOCs data in separating similar sources was not apparent. However, the combined data helped us to obtain a whole picture of the source-specific health risks.

並列關鍵字

receptor model ； source apportionment ； health risk assessment

參考文獻

USEPA. 2005b. Integrated Risk Information System Available from: .

USEPA. 2005c. Integrated Risk Information System Available from: .

Chapter 1

Begum BA, Biswas SK, Kim E, Hopke PK, Khaliquzzaman M. 2005. Investigation of sources of atmospheric aerosol at a hot spot area in Dhaka, Bangladesh. Journal Of The Air & Waste Management Association 55(2):227-240.

Buzcu B, Fraser MP. 2006. Source identification and apportionment of volatile organic compounds in Houston, TX. Atmospheric Environment 40(13):2385-2400.