中南半島近污染源生質燃燒氣膠特性及其傳輸演化與東沙島氣膠特性

Near-source characterization of biomass burning aerosols and transport evolution from Indochina peninsula and aerosol properties at Dongsha island.

張佑嘉(Yu-Chia Chang)

指導教授 ： 李崇德

國立中央大學/工學院/環境工程研究所/碩士(2011年)

摘要

每年春季中南半島北部生質燃燒旺盛，生質燃燒煙團經盛行西風傳送，分布廣泛遍及中南半島及東亞，影響這個區域對太陽輻射的收支。本文在泰國清邁(海拔1,396 m)、台灣鹿林山(海拔2,862 m)、台灣東沙島進行氣膠觀測，發現傳送至東沙島的氣流屬於地面傳輸，傳送至泰國清邁及鹿林山則屬於高層大氣傳送，在這三個地點觀測氣膠能分別瞭解受到高層大氣及地面傳送不同氣流來源影響的氣膠特性差異。
清邁位於泰國與緬甸交界，是一個相當靠近中南半島生質燃燒源區的地方，本研究以手動方式採集PM10及PM2.5氣膠進行化學成分分析，發現清邁氣膠質量濃度以PM2.5氣膠為主。PM2.5水溶性離子中硫酸根離子及鉀離子為優勢物種，氣膠碳成分以OC3及EC1-OP為主，水可溶有機碳(WSOC)佔有機碳比例為62%，二元酸以Oxalic acid濃度最高，至於氣膠單醣無水化合物則明顯以左旋葡萄糖為主。從樣本的分析可清楚瞭解生質燃燒源區氣膠特性，透過適當氣膠成分比值的計算，可推測PM2.5氣膠來自森林曠野燃燒，燃燒樹種多以軟木為主。
鹿林山春季受到盛行西風影響，是一個適合觀察中南半島生質燃燒氣團長程傳輸的地點。本文比較鹿林山與清邁源氣膠特性，發現生質燃燒氣膠經過傳輸以後，氣膠鉀離子增加4.4倍、硝酸根離子增加2.3倍、OC3 增加2.7倍、EC1-OP增加4.7倍、左旋葡萄糖增加3.4倍。本文也利用一些氣膠氧化特性指標來探討生質燃燒氣膠傳輸老化的現象，其中以WSOC/WIOC (水不可溶有機碳)最能突顯老化特性，WSOC在傳輸前後會由易揮發的有機碳物種組成轉變至不易揮發的有機碳物種，nss-SO42-/nss-K+及SOR (sulfur oxidation ratio)則會受到氣團經過中國南方的影響，使氣團nss-SO42-平衡狀態改變，在評估老化現象需特別注意。
本研究共進行四次雲霧事件的前、中、後氣膠採樣觀測，有三次雲霧事件會挾帶氣膠至鹿林山；雲霧事件後可能受到濃度較高的氣流影響使PM1氣膠濃度增加。在雲霧活化氣膠成分效率方面，硝酸根離子最容易受到雲霧活化，其次為硫酸根離子，銨根離子的活化效率最低。
東沙島座落於南海東北端，當它受到亞洲大陸傳輸氣流影響會明顯帶來氣體污染物，即使是受到海洋來源影響也是有少量人為污染物。
在氣膠中和程度與化合物結合形態方面，三個採樣地點PM2.5氣膠都有氨氣不足的現象，東沙島PM10氣膠也有氨氣不足的現象。清邁PM2.5氣膠NH4+及nss-SO42-結合形態為(NH4)3(H)(SO4)2，鹿林山生質燃燒事件與非生質燃燒期間結合形態都是(NH4)2SO4，東沙島氣膠受到海鹽與酸鹼氣反應的影響有氯損失現象，利用“氯損失原理”可獲得東沙島PM10氣膠結合形態為NH4HSO4，PM2.5氣膠為(NH4)3(H)(SO4)2。

關鍵字

南海氣膠 ； 氣膠中和 ； 氣膠老化指標 ； 燃燒區氣膠特性 ； 東南亞氣膠 ； 生質燃燒

並列摘要

Biomass burning (BB) is active in the northern part of Indochina Peninsula every spring. The BB plume transported by the prevailing westerly wind affects solar radiation budget in the Indochina Peninsula and East Asia. This study observed atmospheric aerosols at Chiangmai (1,396 m a.s.l., Thailand), Mt. Lulin (2,862 m a.s.l.), and the Dongsha Island. Trajectory analysis showed that the airmasses were transported near the surface at the Dongsha Island in contrast to the upper atmospheric transport at the Chiangmai and Mt. Lulin sites. The distinction of transported aerosol properties between surface and upper atmosphere can be appreciated at these three sites.
The Chiangmai site locates in the borderline of Thailand and Myanmar and is very close to BB source area in the Indochina Peninsula. Filter-based PM10 and PM2.5 samples were manually collected and analyzed for their chemical compositions. The result showed that Chiangmai aerosol mass was dominated by PM2.5 and sulfate and potassium ions were the major species in the water-soluble ions. OC3 and EC1-OP were predominant in aerosol carbonaceous fractions and 62% of organic carbon (OC) was in water-soluble organic carbon (WSOC). Oxalic acid was dominated in diacides and levoglucosan is undoubtedly the paramount fraction of anhydrous monosaccharide. The analyzed samples clearly reveal aerosol characteristics in the BB source area. The collected PM2.5 could be attributed to forest open-burning and most burnt tree species was softwood through the calculations of appropriate aerosol composition ratios.
Mt. Lulin is an appropriate site for observing BB plume transported from Indochina peninsula as it is located downstream of the prevailing westerly wind. This study compared aerosol characteristics between Mt. Lulin and Chiangmai and found that aerosol potassium ion increased 440%, nitrate ion increased 230%, OC3 increased 270%, EC1-OP increased 470%, and levoglucosan increased 340% in the BB plume during transport.
A few aerosol oxidation indices were employed to assess aging effect of biomass burning plume after transport. Among the selected indices, WSOC/WIOC (the ratio of WSOC over water-insoluble organic carbon) is the most significant index in showing transported aging aerosol. A conversion of more volatile to less volatile organics is noticed for WSOC. Note that the change of the equilibrium state of nss-SO42- influenced by the passage of airmasses through southern China makes nss-SO42-/nss-K+ and SOR (sulfur oxidation ratio) less effective in studying aerosol aging effect.
There are four cloud events observed in the state of before, during, and after periods in this study. Aerosol was brought by the airflow to the Mt. Lulin site for three events. PM1 concentration was likely to increase by the introduction of higher aerosol mass airflow after cloud event. In the activation efficiency of aerosol component in cloud, nitrate ion is activated easiest in the cloud followed by sulfate ion and ammonium ion is the least activated one.
The Dongsha Island is located in the northern tip in South China Sea. Its location makes it either affected by gaseous pollutants brought by Asian continental airmasses or minor anthropogenic pollutants when the airmasses are with oceanic origin.
In studying aerosol neutralization and compound form, ammonia deficiency is found for all at the three sites. The compound form of NH4+ and nss-SO42- at the Chiangmai site is (NH4)3(H)(SO4)2, while (NH4)2SO4 is inferred at the Mt. Lulin site. By adopting the “Chlorine Loss” mechanism to account for the reaction of sea-spray aerosol between acidic and basic gases, the inferred compound forms of NH4+ and nss-SO42- at the Dongsha Island are NH4HSO4 for PM10 and (NH4)3(H)(SO4)2 for PM2.5, respectively.

並列關鍵字

Aerosol in South China Sea ； Aerosol neutralization ； aerosol aging indices ； Southeast Asian aerosol ； Biomass burning aerosol ； Source characteristics of aerosol

參考文獻

Aggarwal, S., Kawamura, K., 2008. Molecular distributions and stable carbon isotopic compositions of dicarboxylic acids and related compounds in aerosols from Sapporo, Japan: implications for photochemical aging during long-range atmospheric transport. Journal of Geophysical Research-Atmospheres 113, D14301, doi:10.1029/ 2007JD009365.

Aggarwal, S., Kawamura, K., 2009. Carbonaceous and inorganic composition in long-range transported aerosols over northern Japan: Implication for aging of water-soluble organic fraction. Atmospheric Environment 43, 2532-2540.

Arimoto, R., Zeng, T., Davis, D., Wang, Y., Khaing, H., Nesbit, C., Huey, G., 2008. Concentrations and sources of aerosol ions and trace elements during ANTCI- 2003. Atmospheric Environment 42, 2864-2876.

Beig, G., Brasseur, G. P., 2000. Model of tropospheric ion composition: A first attempt. Journal of Geophysical Research, 105, 22671–22684.

Benner Jr., B.A., Wise, S.A., Currie, L.A., Klouda, G.A., Klinedinst, D.B., Zweidinger, R.B., Stevens, R.K., Lewis, C.W., 1995. Distinguishing the contributions of residential wood combustion and mobile source emissions using relative concentrations of dimethylphenanthrene isomers. Environmental Science & Technology 29, 2382-2389.