| 摘要: | 細懸浮微粒(氣動直徑小於或等於2.5 μm粒狀物質, PM2.5)對於空氣品質評估、污染源管制策略制定、健康風險評估、環境變遷有重大的影響。本文為探究台灣都會區PM2.5可能污染源以及污染事件成因，於2015年7月至2016年5月在台灣北(板橋)、中(忠明)、南(小港)部環保署空氣品質監測站，觀測都會區PM2.5質量濃度與化學成分變化，採集PM2.5過程配置三張濾紙，修正採樣過程微粒吸附及揮發作用。對於解析的PM2.5質量濃度與化學成分，本文分別探討其季節變化趨勢，並結合手動採集數據與環保署空氣品質監測數據，探討污染事件成因。此外，透過EC-tracer與質量重組推估原生與二次生成碳成分在PM2.5占比，使用正矩陣因子法(Positive Matrix Factorization, PMF)推估貢獻因子(污染來源)，最後使用條件機率函數(Conditional Probability Function, CPF)結合風向資料，推估高貢獻當地污染源來向。
研究結果顯示北、中、南三個觀測地點PM2.5質量濃度，都以夏季最低，由北至南三個測站最高季節濃度分別發生於春季、秋季、冬季，造成季節差異可歸因於污染事件成因差異，例如：北部春季的境外傳輸、中部秋季受到上風污染區域傳輸與環境擴散不佳、南部冬季環境擴散不佳。觀測地區PM2.5化學成分組成大多以硫酸鹽與修正後有機碳成分為主，然而，當PM2.5濃度大於35 μg m-3時，硝酸根離子明顯有增量趨勢，顯示發生污染事件時，硝酸根離子前驅排放源具有重大影響，因此要降低PM2.5事件發生必須管制硝酸根離子相關前驅排放源。
透過EC-tracer方法推估原生與二次生成有機碳占比，結果顯示三個測站都是冬季有最高二次生成有機碳占比，分別為49%、48%、60%，這項結果顯示冬季低溫使有機物揮發損失減少以及冬季混合層高度低，環境中前驅物濃度高有利於二次生成反應的進行。質量重組的結果則顯示台灣北、中、南都會區有機物與有機碳間比值分別可使用1.56、1.6、1.66，而且發現比值在高污染時期有明顯提升，顯示高污染時期環境有利於有機物含量提升。
從金屬元素成分變化發現各觀測地區元素富集因子數值隨污染物濃度升高而提升，顯示污染源主要來自人為活動。當季節改變，風向轉為東北季風時，作為燃煤指標元素的Se在三站都明顯提升，富集因子數值同時驟升，推測受到境外傳輸或是鄰近都會區域傳輸影響。低污染濃度時期，如夏季中部與南部測站燃油指標元素V有明顯占比，並會隨季節逐漸降低，顯示低污染時期本地工業的燃油排放影響十分明顯。
PMF受體模式推估結合CPF顯示在低污染時期污染貢獻源主要來自本地排放源，高污染時期受到環境擴散不佳或是境外傳輸影響，因此二次污染物占比較大。
;Atmospheric suspended fine particles (particulate matter with an aerodynamic diameter less than or equal to 2.5 μm, PM2.5) play an important role in air quality assessment, setting up control strategies for polluting sources, health risk assessment, and environmental change. This study aims to investigate source contributions and causes of polluting events by observing PM2.5 mass and chemical component concentrations at the stations of Taiwan Environmental Protection Administration (TEPA) in the northern (Banqiao), middle (Zhongming), and southern (Xiaogang) parts of Taiwan from July 2015 to May 2016. During the studying period, the deployment of three-filter in series was adopted for PM2.5 collections to correct for the interference of volatilization of semi-volatile species and adsorption of gases from the environment. For the resolved PM2.5 mass and chemical component concentrations, this study investigated their seasonal variation trends and causes of polluting events by combining manual collections with air-quality monitoring data conducted at TEPA stations. In addition, weight percentages of primary and secondary organic carbons in PM2.5 were estimated by EC-tracer and mass reconstruction methods. Contributing factors (i.e., polluting sources) were apportioned by Positive Matrix Factorization (PMF) method and validated by Conditional Probability Function (CPF) with wind direction for high contributing local sources.
The results show that PM2.5 mass levels were lowest at all sampling sites in summer, and highest at stations in the northern, middle, and southern parts of Taiwan in spring, autumn, and winter, respectively. These seasonal variations can be attributed to causes of polluting events, for examples, transboundary transports in the northern Taiwan in spring, upwind regional transport and bad environmental ventilation in the middle Taiwan in autumn, and bad environmental ventilation in the southern Taiwan in winter. Major PM2.5 chemical components are sulfates and organic carbons in most samples. However, the weight percentages of nitrate ion in PM2.5 were apparently enhanced when PM2.5 concentrations were greater than 35 μg m-3 implying precursor sources of nitrate ion played a significant role. The precursor sources of nitrate ion need to be controlled to reduce PM2.5 polluting events.
The results of EC-tracer method in estimating weight percentages of primary and secondary organic carbons in PM2.5 show that the weight percentages of secondary organic carbons are highest in winter with 49%, 48%, and 60%, respectively, across all three sites. It indicates low ambient temperature in winter reduces losses of semi-volatile species and lower winter mixing layer in enhancing precursor pollutant concentrations both favor the process of secondary production. The mass closure results show that the multiplying factors of organic matter (OM) to OC in the northern, middle, and southern metropolis are 1.56, 1.6 and 1.66, respectively. In addition, the factor was apparently increased during high polluting period which implied that the content of OM was enhanced.
The computed values of Enrichment Factor (EF) from the analyzed metal elements across all stations are increased with the rises of pollutant concentrations which reveal that polluting sources mainly contributed from anthropogenic activities. The concentrations of coal burning tracer Se were increased with simultaneous increases of EF values across all three stations when the prevailing wind changed to northeast monsoon during the change of season. The cause was considered to be affected by transboundary or regional transports of nearby metropolis. During low polluting period, the weight percentages of oil combustion element V were enhanced in the middle and southern stations in summer but lowered according to the change of seasons. It indicated oil combustion emissions from local industry were relatively important during low polluting period.
The source apportionment from PMF coupling with CPF reveals that the emissions from local sources are mainly responsible during low polluting period. In contrast, the weight percentages of secondary pollutants were comparatively higher due to bad environmental ventilation or transboundary transport during high polluting period. |
