| 摘要: | 本論文包括兩大部分，第一部份旨在發展空氣中有機揮發性氣體 (VOCs)的監測與應用，其中又可分為兩單元，分別為1.交通源之成份與其對臭氧的影響，2. 台灣南部空氣汙染事件之成因分析。另一部分為建立高山測站溫室氣體監測方法，使溫室氣體成為我國高山背景空氣測站之固定監測項目。
論文的第一部份第一單元旨在探討交通車速對成份的影響與其對空氣品質的可能影響，以文獻中少見之長隧道作為取得純車輛排放之理想環境，在隧道中量測數十種非甲烷碳氫化合物 (NMHC)成份與一氧化碳 (CO)，並以NMHC/CO比值方式呈現不同行車速度下NMHC的組成變化與其對臭氧生成潛勢的影響，研究結果顯示車輛多且平均車速較慢時 (45 km hr-1)比值較高，以未燃燒完全之汽油─烷烴為主；隨著車速增加 (75-85 km hr-1)比值降低，此時污染物排放由烷烴轉為以烯烴、芳香烴為主，雖然總臭氧生成量降低，但因此兩類物種最大反應增量 (MIR)較高，佔臭氧生成貢獻比例較高。本研究之創新處在於1. 以長隧道取得純粹車輛排放之理想環境，相較於以往文獻研究多以動力計與短隧道實驗作為量測方式更具有代表性且不易受其他排放源之干擾；2. 採多點等距採樣，使隧道數據更具車輛尾氣之代表性；3. 以NMHC/CO比值作為比較相對排放之基礎，排除了車輛密度之變因。
論文的第一部份第二單元旨在探討南臺灣長年受高臭氧事件影響之成因，本研究為南台灣之單一個案分析，選擇都會區典型VOC成份作為汙染傳送之示蹤劑，並以光化學模式模擬空氣中指標VOC物質之濃度變化與擴散行為。結果顯示南部都會區排放之污染物在春季盛行東北季風時會被推送至西部海域，不會影響高雄、屏東地區；若東北季風轉弱，天氣型態轉變成海陸風型態時，都會區排放污染物則會被海風帶至內陸地區，並於傳輸路程中間接產生高臭氧而影響下風處。本研究首次成功利用已被驗證的光化學模式結合實地觀測，模擬一次污染物VOCs與二次污染物臭氧之濃度分佈與傳送行為，是文獻中少見的突破。
論文的第二部份旨在高山上建立背景測站溫室氣體監測方法，本研究首次於我國第一座高山背景空氣品質測站 (鹿林山測站)以光腔衰盪光譜儀 (CRDS)建立台灣背景大氣中二氧化碳 (CO2)與甲烷 (CH4)逐時濃度量測方法，以長期觀測此兩種重要溫室氣體的大氣濃度變化。研究結果顯示二氧化碳與甲烷測值與NOAA採樣結果呈現相當一致的結果，二氧化碳背景濃度呈現逐年上升趨勢亦與全球平均觀測值接近 (每年上升1.48 ppm)，兩者長期量測數據具有東亞地區之代表性。雖然兩者平均濃度與逐年趨勢與世界其他背景測站結果相似，但受本地與長程傳輸影響使得兩者濃度在背景值上出現較大的跳動，顯示其中蘊含了豐富的空氣流動特徵，將成為未來鹿林山長期觀測數據的獨特資產。
This thesis comprises two parts. The first part aims to developing monitoring and applications of ambient volatile organic compounds (VOCs), which contains two subjects, i.e., studying the composition of pure traffic emissions and interpreting the occurrence of air pollution events in southern Taiwan. The second part aims to develop monitoring capabilities of greenhouse gases (GHGs) at a high mountain station and to add these gases to the list of routine monitoring.
The first subject of the VOC work is to study the composition of vehicular emissions and the relationship between traffic speed and chemical composition. To obtain pure traffic emissions without interferences from other sources, a long-tunnel tunnel (12.9 km) was exploited in the study. A series of non-methane hydrocarbons (NMHCs) and carbon monoxide (CO) in multiple air samples collected from a moving vehicle inside the tunnel were analyzed. Instead of calculating emission factors as appeared in most of the related works in the literature, the compositional changes in emission at various speeds were discussed in a relative manner by normalizing individual NMHCs with CO concentrations to obtain relative emissions. Subsequently, the effects on ozone formation arisen from the compositional changes at different speeds were discussed. Results show that, at the slowest speed (45 km hr-1) due to traffic congestion in the tunnel, the group of alkanes dominates the NMHC composition. As the speed increases (75-85 km hr-1), alkenes and aromatics become the dominant constituents of the NMHC composition. Although, on the per-unit-mass basis, alkenes and aromatics have higher maximum increment reactivity (MIR) than alkanes, the much greater emissions at the slowest speed still induce the largest ozone forming potential as a whole. The novelties of this study lie in the following aspects: 1. Long tunnels are ideal for obtaining pure vehicular emissions, which is more advantageous than performing emission studies in shorter tunnels or with chassis dynamometers
2. Multiple sampling along the long tunnel can better represent pure traffic emissions with minimal interferences from other non-vehicular sources as compared to measurements in short tunnels
3. The use of NMHC/CO ratios can eliminate the factor of vehicle densities, enabling more straightforward comparison between different datasets.
The second subject of the VOC work is to study the cause of high-ozone episodes that frequently occurred in southern Taiwan. A case study in springtime was chosen for the study. The VOC monitoring was coupled with model simulations (PAMS-AQM) to help interpret dispersion of air pollutants, and toluene was employed as the surrogate pollutant in model simulations to display pollutants movements. Results show that when strong northwesterly winds prevail, pollutants are blown westwards to the sea not affecting the northern counties. In contrast, when the local circulation dominates weather conditions, pollutants are blown towards inland areas and, thus, the southern cities are more prone to pollution episodes. This work is one of the very few studies in the literature that coupled model simulations with VOC field measurements, which successfully demonstrates the transport of primary pollutants and the formation of secondary pollutants such as ozone during a pollution episode.
The second part of the thesis is to establish the measurement capabilities of GHGs at a high-mountain background station (Lulin Atmospheric background Station, LABS) using cavity ringdown spectrometry (CRDS) to routinely monitor carbon two major GHGs, namely dioxide (CO2) and methane (CH4). Our in-situ observations of CO2 and CH4 were found to be rather consistent with the NOAA flask results in terms of seasonal variations and average concentrations. The gradual increase in atmospheric CO2 concentrations at LABS also agreed well with global observations, suggesting the GHG observations are representative of this geophysical region. Despite the general agreement with other global stations, the variability was found to be much greater than that at other global stations, implying complex influences from local or long-range transport. To decipher the mechanisms embedded in the variability could make LABS a unique asset in terms of global observations of GHGs. |
