利用遮雨載玻片在台灣採樣鑑定大陸沙塵暴微粒之研究

A Sheltered Glass Slide Method for Sampling and Identifying China Sand Storm Particles in Taiwan

孫岩章(En-Jang Sun)

《環境保護》 29卷1期 (2006/06) Pp. 1-10

https://doi.org/10.30017/EP.200606.0001

摘要

目前在臺灣採樣及鑑定中國大陸沙塵暴的方法主要是利用PM10監測儀，然而PM10資料呈現的是各種污染微粒的總量，而非專指沙塵暴之顆粒，故其代表性十分不足，尤其是在其他污染量大於沙塵暴顆粒時。因此作者乃嘗試研發一種可以利用「遮雨載玻片」採樣及鑑定沙塵暴污染之方法，希望未來可以在台灣實際監測從大陸飛來之沙塵污染情況。本研究已設計出一種四面圍著雙層百頁窗遮雨柵欄之採樣器，以每週更換載玻片三次之方式，即可進行沙塵暴微粒之採樣及鑑定。此些經過曝露後的載玻片在暗視野光學顯微鏡下觀察、鑑定、計數。利用此一方法發現沙塵暴微粒在200倍暗視野光學顯微鏡下，為亮黃、或亮褐色不整形粒子，與其他黑煙、汽車排煙、海鹽鹽沫等其他污染微粒皆屬不同。其粒徑從可1微米以上到200微米皆有。本研究亦採取大陸2001年北京發生沙塵暴之微粒樣品，並與在台灣台北所採者相比，發現在物理性狀上兩者實無法分辨，唯若選取粒徑介於2~10微米做為指標範圍，發現在環保單位證實之「沙塵暴事件日」期間，每個觀測視野（直徑2 mm）中的沙塵微粒數會增高，且多高於10個以上，而在「非沙塵暴事件日」期間，每個觀測視野微粒數皆低於10以下，經統計14次之「非沙塵暴污染日」之微粒數，得其平均為2.3粒，標準偏差亦為2.3粒。若取「平均值大於3倍標準偏差」之值為「污染事件日」之判定標準，得值為9.2。此說明若增高至10個以上，即代表其與「非污染事件日」有極顯著差異。故可利用此「遮雨載玻片」上每一視野（直徑2mm）的沙塵微粒數，判定實際之「沙塵暴污染日」數。已初步利用此一方法，在2001年2月至5月的監測中，分別監測到2月有2次、3月有3次、4月有1次、5月有1次之沙塵暴污染事件(Episode)。證實此－「遮雨載玻片」方法可提供一種直接的採樣、監測之技術，也建議環保單位可用此進行長期偵測跨國沙塵微粒污染之工作或計畫。

關鍵字

沙塵暴 ； 顯微鏡 ； 微粒 ； PM10 ； 臺灣

並列摘要

At present time China sand storm particles were principally sampled and identified by the PM10 instruments in Taiwan. However the PM10 data represent only the total amount of polluted particles and is not specific for China sand storm particles. It means that the representiveness is not good especially when the other polluted particle amount is close or higher than the sand particles. Therefore a sheltered glass slide method was developed to sample and identify the sand particles flown from Mainland China. An anti-rain shelter with two layers of plastic shutter at four directions was designed for sampling the target sand particles. Within the shelter a petri dish with glass slide was disposed and replaced with new one three times for a week. The exposed slides were observed under dark field metallurgical light microscope at 200X. The sand particles were identified based on their physical characteristics. Results showed that sand particles were differentiable from other anthropogenic particles such as black carbon, vehicle smoke, and salt spray. Typical sand particles were amorphous with shiny yellow or brown color at size ranging from detectable (ca. 1 um) to 200 um. However, the sand storm particles collected from Beijing in 2001 showed similar characteristics as those collected at Taipei, indicating their similarity in morphology. When the counting size range were selected at 2-10 um, the particle number per field (diameter 2 mm) at 200X is greater in the pronounced sand storm episode days, and usually greater than 10. While on non-episode days and in other seasons, the particle number per field is usually lower than 10. The mean particle number of 14 non-episode samples was 2.3, with standard deviation of 2.3. If we take three folds of this standard deviation as the significant different level, the lowest episode level is 2.3+3×2.3=9.2. The results suggest that this is a reliable sampling and identifying method for sand storm particles. Preliminary results of particle sampling from February to May, 2001, showed that there were 2, 3, 1 and 1 episodes detected at Taipei, respectively. This sheltered glass slide method proved to be an ideal, simple and direct method for detecting the occurrence of sand storm. The author therefore recommended that Taiwan EPA should use this method to do the continuous long-term monitoring work in the future.

並列關鍵字

Sand storm ； Glass slide ； PM10 ； Taiwan