模擬與分析裸露河床地對空氣中懸浮微粒濃度之影響

Simulating and Analyzing the Effects on Concentration of Airborne Particulates derived from Unvegetated Riverbanks

林昆明(Kun-Ming Lin)

指導教授 ： 張能復

國立臺灣大學/工學院/環境工程學研究所/博士(2014年)

https://doi.org/10.6342/NTU.2014.02410

摘要

台灣地區為容易遭受颱風和地震等天然災害的地區，這些天然災害亦會伴隨著土石流的災害，土石流裡會挾帶著大量的泥砂，會隨著河流的沉積到下游處。當大氣條件為乾燥並伴隨著較強烈的東北季風時，往往會造成裸露河床地區附近懸浮微粒(PM10)濃度的升高而危害人體之健康。2009年經過莫拉克颱風之後，一些鄰近河川下游地區的空氣品質測站測得比往年高的PM10濃度，故本研究為欲了解莫拉克颱風對這些河川下游地區空氣品質測站PM10的影響。本研究為建立一揚塵之網格模式，首先比對台東空氣品質測站2004、2005年10 ~ 12月模擬與觀測值之結果，了解模式模擬PM10之可信度，接著迴歸2001 ~ 2008年10 ~ 12月之揚塵潛勢與PM10之平均濃度，顯示兩者有很高的相關性(0.78)，而截距28.7可表示在沒有揚塵情況時，台東測站的PM10背景濃度。接著把2009年10 ~ 12月的揚塵潛勢代入先前迴歸的方程式中，求得的PM10濃度為37.98 μg/m3, 但實際上台東測站2009年10 ~ 12月的平均濃度為61.67 μg/m3，而兩者的差異亦可解釋為莫拉克颱風後，帶來較往年多的砂石，沉積到河川下游處，故在揚塵潛勢沒有太大變化下，2009年 10 ~ 12 月卻監測到較高的PM10濃度。
除了了解莫拉克風災後對河川下游地區的空氣品質測站PM10的影響，本模式亦測試其應用在複雜地形的能力，可以看出在花蓮溪的揚塵沿著花東縱谷內傳輸。在垂直方向上，因揚塵模式只計算揚塵的通量，並沒有考慮垂直速度，故在垂直方向只靠擴散效應影響，故影響高度約在混合層高度附近，多在800公尺以下。
本研究亦利用濁水溪附近高密度的空氣品質監測站，了解濁水溪附近揚塵主要影響區域為崙背、麥寮和褒忠地區等，原因為河川揚塵為大顆粒粒徑為主，故較易沉降致使影響範圍不大。而在高風速下，除了在濁水溪鄰近區域有高PM10之濃度之外，其他地區可能因為料堆場或是廢棄的農地等，致使PM10濃度升高。

關鍵字

土石流 ； PM10 ； 莫拉克颱風 ； 揚塵模式 ； 揚塵潛勢

並列摘要

Landslides frequently occur during large earthquakes and storms in Taiwan, supplying large volumes of sediment to downslope areas. When coupled with the intense northeast monsoon over Taiwan in the dry winter season, this can lead to high concentrations of airborne particulates that are hazardous to human health. Air quality monitoring stations near unvegetated riverbanks recorded high concentrations of particulate matter less than 10 microns (PM10) after Typhoon Morakot in 2009. The objective of this study was, therefore, to analyze the effects on air quality of sediment caused by the typhoon. A deflation model was simulated, and the resulting estimates were compared with observed data from the Taitung monitoring station for 2004 and 2005. The relationship of dust flux to average atmospheric dust concentration was analyzed for October to December 2001 ~ 2010. Analysis showed that the 2001 ~ 2008 data are highly correlated (0.78) with the average concentration. The intercept of 28.7 represented the background concentration with no dust emission, from October to December of 2001 to 2008. Based on the dust potential in 2009, the average PM10 concentration would be 37.98 μg/m3; however, the measured concentration was 61.67 μg/m3 from October to December. This suggests the strong influence of dust re-suspended from unvegetated riverbanks by Typhoon Morakot.
Also, this model was very ideal when applying the transportation mechanism in the complex terrain. The pollution air mass could move alone the East Rift Valley. In the deflation module, due to lack the vertical velocity of the dust emission, the influence height of dust pollution is only the affected by the mechanism of vertical diffusion in the mixing height. In general, influence height of dust pollution is about 200-800 meters.
From the contour maps of high density air quality stations near Jhuoshuei River, main influence area is from Lunbei station to Mailiao and Baozhong Station. Under high-speed winds, besides bare lands of the river, there were also other PM10 emission sources, such as pile storage, blocks yards and bare or abandoned farmlands. Even the monitoring stations are not closed the Jhuoshuei River, sometimes they read high PM10 concentration.

並列關鍵字

Landslides ； PM10 ； Typhoon Morakot ； Deflation model ； Dust potential

參考文獻

Barnes SL (1973) Mesoscale objective analysis using weighted time-series observations. NOAA Tech Memo. ERL NSSL-62, National Severe Storms Laboratory, Norman, OK 73069

Taiwan EPA (2014) Analysis of the causes of fugitive dust at the downstream of Jhuoshuei River. Department of Environmental Monitoring and Information Management Press Release

Bagnold RA (1941) The physics of blown sand and desert dunes. London

Barnum BH, Winstead NS, Wesely J et al (2004) Forecasting dust storms using the CARMA-dust model and MM5 weather data. Environ Modell Softw 19(2): 129–140

Bott A (1989) A positive definite advection scheme obtained by nonlinear renormalization of the advective fluxes. Mon Wea Rev 117:1006–1015