新竹地區PM2.5空氣品質與氣象因子關係之研究

Abstract

本研究以行政院環境保護署民國95年至民國104年共計10年的空氣品質監測數據，包含空氣污染物：PM2.5、PM10、CO、NO、NO2、NOX、O3、SO2，及氣象相關因子：溫度、雨量、風向、風速、濕度，以每小時資料為依據，選取新竹縣、市共三站：新竹站、湖口站、竹東站之資料，以多變數線性迴歸的方式來進行統計分析。同時針對2015年新竹縣/市PM2.5事件日，利用美國國家海洋暨大氣總署大氣資源實驗室所提供的逆軌跡計算模式，推估可能的污染源。 結果顯示，除臭氧外，其餘空氣污染物均大致呈現逐年遞減之情形，PM2.5超標日數均也逐漸降低，顯示新竹地區空氣品質在政府和人民的共同努力下，正在逐年改善。在月分佈上除了O3外，其餘空氣污染物大致呈現V字型，低谷均出現於夏季到秋季。原因應為冬季逆溫層較厚，空氣對流較差，易導致空氣污染物濃度較高。 相關係數分析結果顯示，在日/月平均分析中可得知新竹縣/市PM2.5與氣象因子關係為：PM2.5與氣溫、雨量、相對溼度、風向、風速均大致呈現負低度至中度相關。多變數迴歸分析結果顯示PM2.5與風向以外的氣象因子均呈現負低度相關，與相關係數分析結果相似。R2值方面，最小值為0.17，最大值為0.38，均呈現低度相關，可能是由於PM2.5相關因子尚須納入空氣污染源，如交通工具數量、種類，工廠數量、種類等，亦或是臨近縣市污染狀況，或是境外移入空氣物染物等。 由2015 新竹縣/市三站PM2.5 事件日前三名之逆軌跡推估分析結果，由於前三名均發生於冬季，故可能是冬季逆溫層較厚，大氣擴散條件較原本就較差，再加上特殊事件引起PM2.5再升高。所以PM2.5事件日之主要原因可分為兩方面：第一係為中國大陸吹來之沙塵引起，因冬季主要風向為東北風，與本逆軌跡分析之氣流方向相符，且全島均受到影響，幾乎所有測站PM2.5均為事件日。第二為鄰近縣/市污染物境內傳輸所致，氣流由雲嘉南空品區往中部空品區，再經苗栗往新竹縣/市方向吹，造成PM2.5事件日的發生。故在空氣污染物防制上，建議新竹縣/市可與鄰近縣市聯手，共同研討防制策略。

This study used the 2006 to 2015 data of air quality monitoring stations from Environmental Protection Administration of Executive Yuan. The air pollutants: PM2.5, PM10, CO, NO, NO2, NOX, O3, SO2, and meteorological data: temperature, rainfall, wind direction, wind speed, relative humidity. All data are presented by hourly. Three air quality monitoring stations are all selected in the region of Hsinchu County: Hsinchu station, Hukou station, Zhudong station. And use the method of multiple linear regression approach to do statistical analysis. At the same time, according to the PM2.5 event day of 2015 in Hsinchu County, the possible sources of pollutants were calculated by the back trajectory analysis model provided by the National Oceanic and Atmospheric Administration(NOAA). The results show that all of the pollutants, except O3, are gradually declining year by year, PM2.5 exceeded standard days are also gradually reduced. Indicating that the air quality in the Hsinchu County is improving year by year by the joint efforts of the Government and the people. In the monthly distribution, all of the air pollutants, except O3, are generally V-shaped, bottom appears in the summer to autumn. The reason is that the inversion layer is thicker in winter and the air convection is poorer, which leads to higher air pollutant concentration. PM2.5 and meteorological factors correlation coefficient analysis in Hsinchu County shows that PM2.5 has a negative low to medium correlation with air temperature, rainfall, relative humidity, wind direction and wind speed. According to the results of multiple linear regression analysis, the PM2.5 was negative low correlation with the meteorological factors except the wind direction, which was comparable with the correlation coefficient analysis. In R2 value, the minimum value is 0.17, the maximum value is 0.38, showed a low correlation. It may because of the PM2.5 related factors may need to include air pollution sources, such as the number of vehicles, types, number of factories, types, etc., or pollution condition in neighboring counties, or foreign pollutants. According to the results of back trajectory analysis result in Hsinchu County, the results show that the top three of 2015 PM2.5 event day were all occurred in the winter, probably because of the thicker inversion layer in winter and the worse atmospheric diffusion conditions. Furthermore, a special event triggered PM2.5 increased. Therefore, the main reasons for PM2.5 event day can be separate into two aspects: 1. Dust blowing from mainland China, mostly in the winter, the main wind direction is northeasterly, the same as back trajectory analysis of the airflow direction , and the whole island are affected, almost all of the air monitoring stations are PM2.5 event day. 2. The transmission of pollutants in the neighboring counties. The airflow is from southern Taiwan to central Taiwan, and then flows to the Hsinchu County through Miaoli. And then PM2.5 event day occurs. Therefore, in the air pollution control, the Hsinchu County should co-work with the neighboring counties to discuss prevention strategies.