| 摘要: | 本文於2016年秋季在台灣竹子山(海拔1,103公尺)、鹿林山(海拔2,862公尺)，以及2017年春季在台灣鹿林山進行PM2.5化學特性分析，目的是瞭解秋季竹子山和鹿林山氣膠受東北季風以及春季鹿林山受中南半島生質燃燒煙團長程傳輸的影響。
2016年秋季期間竹子山受到了台北都會區以及境外傳輸影響，PM2.5質量濃度平均為7.5 ± 8.2 μg m-3，於污染事件時甚至高達31.1 μg m-3；相對地，鹿林山PM2.5質量濃度平均為2.7 ± 1.2 μg m-3，各樣本濃度變化不明顯，有山谷風現象時濃度有些許提升，水溶性離子濃度提升較為顯著，NH4+與SO42-分別增加了377%與282%。
2017年春季期間PM2.5與PM10質量濃度分別為8.6 ± 4.8 以及12.7 ± 6.3 μg m-3，PM2.5大約占了PM10的70%。氣膠水溶性離子不論PM2.5或PM10都以SO42-、NO3-與NH4+為主要成分。PM2.5與PM10碳成分則是以OC3最為主要，但OC4成長增加最為明顯。與春季背景相比，PM2.5水溶性離子NH4+與SO42-分別增加了136%與89%，OC3與OC4分別增加了220%與250%。春季鹿林山氣膠特性主要受到藉由盛行西風長程傳輸來自中南半島生質燃燒煙團以及發生頻率較低但來自中國北方的氣團影響。
本文利用revised IMPROVE方法計算出大氣消光係數(bext)，並與鹿林山春季監測的大氣消光係數進行比較，廻歸判定係數非常好(R2>0.85)，在PM2.5質量濃度低時，空氣分子和有機物對bext有重要貢獻，隨著PM2.5濃度的上升，在高污染期間(NH4)2SO4與NH4NO3對bext貢獻顯著。PM2.5質量濃度較低時(≦5 μg m-3)相對濕度對的影響並不是十分顯著，隨著PM2.5質量濃度增加，相對濕度帶來的影響變得明顯。當PM2.5從10到20.9 μg m-3時，revised IMPROVE模式計算與儀器量測的能見度有不錯的線性關係 (R2=0.70)，但IMPROVE模式計算值高出許多。整體來說，本文展現了從背景到污染事件的大氣成分光學效應變化。
;This study collected PM2.5 for chemical characterization at Mt. Bamboo (25.18°N, 121.53°E, 1,103 m a.s.l.) and Mt. Lulin (23.47°N, 120.87°E, 2,862 m a.s.l.) in autumn 2016, and at Mt. Lulin in spring 2017 in Taiwan. The aims of this study are to investigate the influences of the northeast monsoon on atmospheric aerosols at Mt. Bamboo and Mt. Lulin in autumn and that of the transported biomass-burning (BB) smoke from Indochina peninsula at Mt. Luin in spring.
In autumn 2016, PM2.5 mass concentrations were with an average of 7.5 ± 8.2 μg m-3 and as high as 31.1 μg m-3 in a pollution event under the combined effects of Taipei metropolis and transboundary transport. In contrast, PM2.5 mass concentrations were with a mean value of 2.7 ± 1.2 μg m-3 at Mt. Lulin with relatively less variation except for the concentration rise when the mountain-valley wind was prominent. The rise of water-soluble inorganic ions (WSIIs) was significant, e.g., the increases of NH4+ and SO42- were 377% and 282%, respectively.
In spring 2017, the mean values of PM2.5 and PM10 mass concentrations were at 8.6 ± 4.8 μg m-3 and 12.7 ± 6.3 μg m-3, respectively; the proportion of PM2.5 in PM10 was around 70%. For WSIIs, SO42-, NO3-, and NH4+ were major components in either PM2.5 or PM10. For carbonaceous content, OC3 (evolved between 281℃ and 480℃) was the most abundant fraction in PM2.5 and PM10, but OC4 (evolved between 481℃ and 580℃) increased the most. In contrast to spring background air, PM2.5 SO42- and NH4+ increased 136% and 89%, respectively; while the increases of OC3 and OC4 were 220% and 250%, respectively. The aerosol properties at Mt. Lulin were mainly under the influences of BB smoke from Indochina peninsula via prevailing westerly as well as less frequent air masses from North China in spring.
For the assessment of atmospheric optics, this study adopted the revised IMPROVE algorithm to compute light-extinction coefficients (bext) of the atmospheric species. The computed bext values agreed well (R2>0.85) with the observed ones at Mt. Lulin in spring. Raleigh scattering and organic species influenced atmospheric optics significantly in low PM2.5 levels; however, (NH4)2SO4 and NH4NO3 gained its dominance in the high PM2.5 pollution events. It is worthwhile to note that the relative humidity effect was not significant for low PM2.5 levels (≦5 μg m-3), but became prominent in higher PM2.5 levels. For the range of PM2.5 levels from 10 to 20.9 μg m-3, the visibility computed from the revised IMPROVE algorithm had a linear fashion (R2=0.70) with the measured visibility, although the computed values tended to exceed the observed values significantly. Overall, this study demonstrated that the optical effects of atmospheric species varying from atmospheric background condition to pollution events. |
