中文摘要：

利用HYSPLIT模型和全球资料同化系统（GDAS）气象数据，用聚类方法对2012年12月~2013年11月期间抵达天津的逐日72h气流后向轨迹按不同的季节进行归类.并利用相应的 PM10和 NO2浓度日监测数据，分析了不同季节气流轨迹对天津污染物浓度的影响.运用潜在源贡献（PSCF）因子分析法和浓度权重轨迹（CWT）分析法分别模拟了不同季节PM10和NO2潜在PSCF和CWT.结果表明，不同方向气流轨迹对天津PM10和NO2潜在源区分布的影响存在显著差异.天津PM10和NO2日均浓度最高值对应的气流轨迹均集中在冬、春和秋季等来自内陆的西北气流；夏季影响天津的气流轨迹主要来自西北和东南方向，对天津 PM10和 NO2的日均浓度贡献较小.天津 PM10和 NO2的PSCF与CWT分布特征类似，最高值主要集中在天津本地以及邻近的河北省和山东省，是天津这两种污染物主要潜在源区.

英文摘要：

HYSPLIT model and global data assimilation system （GDAS） meteorological statistics were applied to analyze the 72-hour backward trajectories of daily airflow in Tianjin during 2012 December~2013 November. At the same time, cluster method was used to classify the airflow backward trajectory in different seasons. Moreover, the daily monitoring data of PM10 and NO2 from December 2012 to November 2013 were used to analyze the effects that airflow trajectory of different seasons have on the pollutants concentrations in Tianjin. The potential source contribution function （PSCF） analysis method and concentration-weighted trajectory （CWT） analysis method were applied to model the PSCF and CWT of PM10 and NO2 in different seasons. There was an obvious difference in the potential source area distribution of PM10 and NO2 owing to the airflow in different directions. Corresponding to the airflow, the highest values of PM10 and NO2 were concentrated in the inland airflow from northwest in winter, spring and autumn. However, the airflow trajectories in summer were mainly from northwest and southeast, which had less effect on the daily pollutants concentration value of PM10 and NO2. The PSCF of PM10 and NO2 in Tianjin has similar distribution properties to CWT, and the highest value of PM10 and NO2 was concentrated in Tianjin, Hebei and Shandong province, which were the main potential source area of these two main pollutants.