| 摘要: | 大氣細懸浮微粒(氣動粒徑小於2.5μg m-3, PM2.5)會對民眾健康造成影響並導致大氣環境的變遷，檢測各地PM2.5化學特性，有助於瞭解PM2.5環境效應及污染來源。本文彙整2011年至2015年在台灣北(新莊、板橋)、中(忠明)、南(前鎮、前金、小港) 都會區環保署空氣品質監測站手動採集的PM2.5質量及化學成分濃度數據，探討台灣北、中、南部PM2.5的變化趨勢。在PM2.5採集過程，微粒半揮發性成分的揮發以及石英濾紙的吸附有機氣體會影響成分量測的正確性，本文藉由連續三張濾紙裝置進行干擾修正並探討揮發性成分特性以及影響因子。對於PM2.5污染來源，本文以受體模式PMF (Positive Matrix Factorization)進行推估，並以CPF (Conditional Probability Function)輔助判別本地污染源影響，最後探討影響大氣能見度的PM2.5化學成分、氣象因子及污染來源。
研究結果顯示北部地區PM2.5質量濃度以春季最高，且在三月有高濃度發生，四個季節PM2.5化學成分都以SO42-為優勢物種，但夏季SO42-與修正OC濃度濃度差異不大，顯示夏季有機物影響相對重要；中部地區PM2.5質量濃度以秋季最高，SO42-在四個季節都是最主要成分濃度；南部地區PM2.5質量濃度以冬季最高，PM2.5化學成分濃度在四個季節也都以SO42-為主要成分。
PM2.5水溶性無機離子NO3-及Cl-為半揮發性成分，在北部、中部及南部測站NO3-揮發/未修正濃度比值除了冬季以外，其他三個季節都大於100%，且以夏季比值最大；Cl-揮發/ 未修正濃度比值則在四個季節都大於100%，如果不進行修正，會造成PM2.5揮發成分低估。比較PM2.5在低濃度(<35μg m-3)和高濃度(>35μg m-3)各成分占比，SO42-、OC、EC占比下降、NH4+維持不變、NO3-從6%增加到13%，Cl-從1%增加到2%，NO3-增加的幅度最為明顯，顯示降低NO3-和Cl-前驅產生源排放，有助於降低高濃度事件的發生。本文以濾紙採集揮發NH4+，發現揮發NH4+當量數通常大於揮發(NO3-+Cl-)當量數，推測可能有空氣中NH3(g)穿透denuder的現象。
PM2.5碳成分採集過程會發生有機碳揮發以及石英濾紙吸附有機氣體的現象，北、中、南部揮發性有機碳占有機碳的比例平均分別為9%、11、8%，北、中、南部石英濾紙吸附有機氣體的占有機碳比例平均為18、16、14%，代表如果不進行修正，會造成北、中、南有機碳成分平均高估9%、5%、6%。討論揮發離子濃度與環境最高溫度(Tmax)及修正後的PM2.5質量濃度的關係指出揮發NO3- 和NH4+與環境最高溫度及PM2.5濃度有關，揮發Cl-與環境最高溫相關性則不顯著。
利用PMF推估並以CPF輔助判別污染來源，北部、中部、南部測站都以二次硫酸鹽和二次硝酸鹽及氯鹽污染源類別貢獻最大，顯示台灣都市地區PM2.5空氣品質受到化石燃料燃燒排放的影響為最主要。大氣能見度的衰減以南部最為嚴重，整個採樣期間能見度平均為7.5±3.16 km，北部和中部能見度較南部好，採樣期間能見度平均分別為12.57±4.13 km、12.72±3.4 km，能見度以中部>北部>南部。將PM2.5化學成分、污染來源和氣象因子對大氣能見度進行多元迴歸分析，北、中、南能見度主要受水溶性無機離子和氣象因子的影響。
;Atmospheric fine particulate matter (aerodynamic diameter less than or equal to PM2.5) will pose human health risk and cause atmospheric environmental change. In order to understand PM2.5 variation trends in Northern, Middle, and Southern Taiwan, this study summarized the results of PM2.5 mass and chemical components manually collected at Xinzhuang, Banqiao, Chungming, Qianzhen, Qianjin, and Siaogang, respectively from 2011 to 2015. The correctness of aerosol components is affected by volatilization of semi-volatile species and the adsorption of volatile organic gases during PM2.5 collection. In this study, a three-filter series was adopted to correct the interference of volatile species and affecting factors. Source apportionment of PM2.5 was conducted by using Positive Matrix Factorization (PMF) with the assistance of Conditional Probability Function (CPF) for local source evaluations. In the final step, the effects of meteorological factors and polluted sources were regressed on ambient visibility.
The results show that PM2.5 mass level was highest in March and SO42- was a predominant species in all seasons in Northern Taiwan. However, the corrected organic carbon (OC) competed with SO42- as the predominant component in summer which indicated organic compounds were important in summer. PM2.5 mass level is highest in autumn and SO42- is a predominant species in all seasons in Middle Taiwan. In addition, PM2.5 mass level is highest in winter and SO42- is a predominant species in all seasons in Southern Taiwan.
The volatilization of NO3- and Cl- are significant as the volatile ratio is greater than one in most occasions. The fraction of NO3- to PM2.5 mass level increased from 6% to 13% when PM2.5 mass level was low (<35μg m-3) as compared to high (>35μg m-3). In implies that the reduction of NO3- and Cl- precursor sources will help reduce high pollution events. Meanwhile, NH3(g) was found to penetrate from denuder adsorption as the equivalence of volatile NH4+ was greater than the equivalence of volatile (NO3-+Cl-) most of the time.
The fractions of volatile OC to retained OC level in the first filter were 9%, 11, and 8% in Northern, Middle, and Southern Taiwan, respectively. Moreover, the fractions of adsorbed volatile organic gases to retained OC level were 18, 16%, and 14% in Northern, Middle, and Southern Taiwan, respectively. If no correction was made, the OC fractions in retained OC mass level would overestimate 9%, 5%, and 6%, respectively. The volatilized NO3- and NH4+ are related to maximum ambient temperature and corrected PM2.5 mass level; however, the volatilized Cl- is not significantly affected by maximum ambient temperature.
The source apportionment using PMF and aided by using CPF showed that sources of secondary sulfate, nitrate, and chloride contributed greatly in Northern, Middle, and Southern Taiwan. It indicates fossil fuel combustion is the most significant source in affecting urban PM2.5 air quality. The degradation of atmospheric visibility is most serious in Southern Taiwan with visibility averaged at 7.5±3.16 km. In contrast, the visibilities in Northern and Middle Taiwan are better with values at 12.57±4.13 km and 12.72±3.4 km, respectively. From multiple regression analysis on ambient visibility by using PM2.5 chemical components, pollution sources, and meteorological factors, water-soluble inorganic ions and meteorological factors are found significant in affecting the visibility in Northern, Middle, and Southern Taiwan. |
