095

1,450 千元

2006/03/02

2007/02/15

陳律言

南投縣政府環境保護局

國立聯合大學

大氣空氣

臭氧, 事件日, 前趨物

ozone, event day, precusor

張艮輝, 華梅英

陳律言

037-381082

lychen@nuu.edu.tw

彙整近年南投縣重大臭氧污染資料得知：在臭氧濃度超過120 ppb的事件日中，出現頻率前五高天氣型態依序為微弱東北季風、高壓迴流型、太平洋高壓、標準東北季風、低壓帶，且其出現機率合計佔所有臭氧超標站日的74.73%，與常年天氣型態分布不盡相同。由監測結果得知：臭氧月平均濃度測值在春、秋二季較高。埔里站臭氧月平均濃度測值較南投及竹山二站高。氮氧化物及非甲烷碳氫化合物月平均濃度在冬季較高。在2000~2005年中，南投縣空品測站曾有90個臭氧超標站日，微弱東北季風、高壓迴流型、及太平洋高壓等三類亦為全時距內常出現之天氣型態。本計畫蒐集及確認南投縣臭氧前驅物排放量資料，應用光化網格模式模擬解析臭氧事件形成機制。在事件日一(10/27/2003 ~ 10/31/2003)的高壓天氣型態中，海域東北風及向內陸遞減之風速，結合下沉氣流而構成利於沿海地區污染物傳輸至內陸堆積反應。事件日二(6/1/2003 ~ 6/5/2003)颱風外圍環流與地形綜合作用，使內陸山脈後方風速較低，結合強勁下沉氣流而利於沿海污染物傳輸至內陸而反應。事件日二梅雨鋒面型態中，因風速較低不利污染物擴散而在內陸反應。本計畫臭氧模擬值與測值具良好相關性。事件日一中高臭氧濃度區多生成於南投縣西南部，繼而依地形沿東北西南向擴張，隨入夜縮小。在事件日二中，颱風型事件日高臭氧濃度區生成於南投縣西北部，逐步朝內陸擴張，入夜後逐漸縮小。以H2O2/HNO3, O3/HNO3, H2O2/NOy, O3/NOy, 及HCHO/NOy等五種光化指標解析得知在模擬事件日中絕大多數高臭氧事件為氮氧化物主控，顯見尋求氮氧化物減量應為臭氧管制策略優先思考方向。不同天氣型態臭氧事件主要貢獻鄉鎮不盡相同，東勢鎮、彰化市及南投市為本計畫模擬事件日中較常出現者。惟主要貢獻仍以南投縣外中部空品區鄉鎮為主。南投縣有某些高臭氧污染事件日之臭氧濃度超標量，高於中部空品區排放貢獻之和，在本研究模擬之事件日中，此類事件日以低壓帶天氣型態為主。減少臭氧超標站日之策略，建議朝向針對臭氧濃度僅小幅超標之事件小時，以及由超標情形較不顯著事件小時構成之站日著手。以2000到2005年的資料彙整結果顯示：南投縣三處空品測站中，在總共227個超標站日中，臭氧濃度只在一個測站僅有一小時超標5ppb以內的站日數有69天，佔總超標站日數的30%強。

臭氧, 事件日, 前趨物

20070516臭氧期末定稿.pdf

12MB

[期末報告]公開版

2007/8/1 上午 12:00:00

Among the historical data of ozone event days, the top five frequent weather types, which contribute 74.43% of ozone event station days, are mild northeast monsoon, high-pressure system with recirculation, Pacific high-pressure system, standard northeast monsoon, and low-pressure zone. Higher monthly average ozone concentrations were observed in spring and autumn. The monthly average ozone concentrations measured in Puli are higher than the ones measured in Nanto and Chushan. Higher monthly average nitrogen oxides and non-methan hydrocarbons concentrations were observed in winter. There were 90 ozone event station days in Nanto County during 2000~2005, weather types with highest frequency are mild northeast monsoon, high-pressure system with recirculation, and Pacific high-pressure system. In this project, the mechanisms of ozone event are simulated by photochemical grid model. In event day 1 (10/27/2003 ~ 10/31/2003), lower wind velocities were simulated inland. The pollutants were transported from the areas near sea to the inland area and react to form ozone. In event day 2 (6/1/2003 ~ 6/5/2003), the circulation around typhoon leads to lower wind velocity and pollutant accumulation behind the montains. The simumlted and measured ozone concentrations exhibit good correlationship. During the event day 1, areas with high ozone concentration originate in the southwest part of Nanto County, and develop along the northeast-southwest direction, then diminish in the night. In the event day 2, areas with high ozone concentration originate in the northwest part of Nanto County, and develop from west to east, then diminish in the night. The dominating precursor of ozone can be determined as nitrogen oxide by using the photochemical indicators H2O2/HNO3, O3/HNO3, H2O2/NOy, O3/NOy, and HCHO/NOy. The reduction of nitrogen oxides emission should have higher priority in developing the ozone control strategy. The major contributing areas of ozone are different for ozone event days with different weather types. Doshe, Chanhwa City, and Nanto City are the frequent contributors. However, the ozone concentration is contributed mainly by emission sources outside the central air quality zone. The strategy to reduce the ozone violence days is suggested to focus on the event hours with only small ozone violence. From the data of 2000~2005, there are 69 days with only one hour ozone concentration exceedence among the 227 ozone violence station days.

ozone, event day, precusor