Lu Hsin-Chung
Department of Environmental Engineering, Hungkuang Institute of Technology, Hungkuang University, No. 34 Chung-Chi Rd., Sha-Lu 433, Taichung, Taiwan.
Chemosphere. 2004 Feb;54(7):805-14. doi: 10.1016/j.chemosphere.2003.10.012.
Three theoretical parent frequency distributions; lognormal, Weibull and gamma were used to fit the complete set of PM10 data in central Taiwan. The gamma distribution is the best one to represent the performance of high PM10 concentrations. However, the parent distribution sometimes diverges in predicting the high PM10 concentrations. Therefore, two predicting methods, Method I: two-parameter exponential distribution and Method II: asymptotic distribution of extreme value, were used to fit the high PM10 concentration distributions more correctly. The results fitted by the two-parameter exponential distribution are better matched with the actual high PM10 data than that by the parent distributions. Both of the predicting methods can successfully predict the return period and exceedances over a critical concentration in the future year. Moreover, the estimated emission source reductions of PM10 required to meet the air quality standard by Method I and Method II are very close. The estimated emission source reductions of PM10 range from 34% to 48% in central Taiwan.
使用三种理论母体频率分布；对数正态分布、威布尔分布和伽马分布来拟合台湾中部地区的完整PM10数据集。伽马分布是最能代表高PM10浓度表现的分布。然而，母体分布在预测高PM10浓度时有时会出现偏差。因此，采用了两种预测方法，方法一：双参数指数分布和方法二：极值渐近分布，以更准确地拟合高PM10浓度分布。双参数指数分布拟合的结果比母体分布与实际高PM10数据的匹配度更高。两种预测方法都能成功预测未来年份超过临界浓度的重现期和超标情况。此外，通过方法一和方法二估计的达到空气质量标准所需的PM10排放源削减量非常接近。台湾中部地区估计的PM10排放源削减量在34%至48%之间。
