Chang Shuenn-Chin, Lee Chung-Te
Graduate Institute of Environmental Engineering, National Central University, Jhongli, Taiwan, ROC.
J Environ Manage. 2008 Mar;86(4):627-35. doi: 10.1016/j.jenvman.2006.12.029. Epub 2007 Feb 12.
Data collected from the five air-quality monitoring stations established by the Taiwan Environmental Protection Administration in Taipei City from 1994 to 2003 are analyzed to assess the temporal variations of air quality. Principal component analysis (PCA) is adopted to convert the original measuring pollutants into fewer independent components through linear combinations while still retaining the majority of the variance of the original data set. Two principal components (PCs) are retained together explaining 82.73% of the total variance. PC1, which represents primary pollutants such as CO, NO(x), and SO(2), shows an obvious decrease over the last 10 years. PC2, which represents secondary pollutants such as ozone, displays a yearly increase over the time period when a reduction of primary pollutants is obvious. In order to track down the control measures put forth by the authorities, 47 days of high PM(10) concentrations caused by transboundary transport have been eliminated in analyzing the long-term trend of PM(10) in Taipei City. The temporal variations over the past 10 years show that the moderate peak in O(3) demonstrates a significant upward trend even when the local primary pollutants have been well under control. Monthly variations of PC scores demonstrate that primary pollution is significant from January to April, while ozone increases from April to August. The results of the yearly variations of PC scores show that PM(10) has gradually shifted from a strong correlation with PC1 during the early years to become more related to PC2 in recent years. This implies that after a reduction of primary pollutants, the proportion of secondary aerosols in PM(10) may increase. Thus, reducing the precursor concentrations of secondary aerosols will be an effective way to lower PM(10) concentrations.
分析了台湾环境保护局于1994年至2003年在台北市设立的五个空气质量监测站收集的数据，以评估空气质量的时间变化。采用主成分分析（PCA）通过线性组合将原始测量污染物转换为较少的独立成分，同时仍保留原始数据集的大部分方差。保留了两个主成分（PC），它们共同解释了总方差的82.73%。代表一氧化碳、氮氧化物和二氧化硫等一次污染物的PC1在过去10年中呈现出明显下降。代表臭氧等二次污染物的PC2在一次污染物明显减少的时间段内呈逐年上升趋势。为了追踪当局提出的控制措施，在分析台北市PM10的长期趋势时，已排除了由越境传输导致的47天高PM10浓度天数。过去10年的时间变化表明，即使当地一次污染物得到很好控制，臭氧的适度峰值仍呈现出显著上升趋势。PC得分的月度变化表明，1月至4月一次污染显著，而臭氧从4月至8月增加。PC得分的年度变化结果表明，PM10在早期与PC1的强相关性逐渐转变为近年来与PC2的相关性更强。这意味着在一次污染物减少后，PM10中二次气溶胶的比例可能会增加。因此，降低二次气溶胶的前体浓度将是降低PM10浓度的有效方法。
