中文摘要：

为对比霾判别方法的差异,探讨霾观测标准的再完善性,文中从空间分布与单点时间序列两方面分别分析研究了4种霾判别方法的特征与适用性.选取中国长江三角洲地区1980~2009年38个地面观测站的气象资料,根据使用日均值的方法 1、2和使用14:00观测值的方法 3分别统计各站点的霾日,分析3种方法的异同.发现这些方法都能够反映出霾的长期变化趋势,但存在差异,这种差异随着年代际变化逐渐减小.由方法 1得到的霾日数最多,方法 3考虑了天气现象,比方法 1和2更合理.依据南京北郊2012年5月~2013年4月的逐时PM2.5浓度、相对湿度和能见度等资料,分析比较了方法 4(霾的观测和等级预报,QX/T 113-2010)与方法 1、2、3的不同.结果表明,由方法 3统计出的霾日少于其他方法,由方法 4统计出的霾日数介于方法 1与方法 3的结果;方法 3不能分辨出霾的严重程度,而其他方法能较好地分辨出霾的严重程度.

英文摘要：

It is crucial to compare the difference and applicability of different haze discrimination methods and its criteria in haze study. Due to the requirement of methods and limitation of data, 4 common methods in two views of regional and temporal in one site are analyzed and studied. Based on the meteorological data from 38 observatories from 1980-2009, haze-day and haze-hour number are both counted for each station, employing Method 1 and 2 ( with daily mean observation) and Method 3 ( with 14:00 observation) . The characteristics and applicability of these three methods are also compared and summarized. Statistical results via these methods are all capable to represent the long-term trend of haze, but haze-day numbers counted via these methods show differences, which become less remarkable with decadal changes. Haze days are the most by using Method 1. Method 3 considering the weather phenomenon is more reasonable than Method 1 and 2. According to the data of visibility, relative humidity and PM2. 5 concentration in northern Nanjing from May, 2012 to April, 2013, employing 4 haze discrimination methods including the additional one (QX/T 113-2010, i. e. Method 4), haze-day and haze-hour numbers are counted and compared. The result shows that:the haze-day number obtained employing Method 3 is less than others, while the haze-day number obtained employing Method 4 is between those with Method 1 and 3;using all Methods but Method 3 can distinguish the severity of haze significantly.