中文摘要：

利用2014年北京市12个空气质量监测站的逐小时PM2.5地面观测资料,以及Terra和Aqua卫星的MODIS气溶胶光学厚度（AOD）产品,在时间和空间数据匹配的基础上,研究了PM2.5的5 h（10：00—14：00）和24 h（0：00—23：00）两种时段平均浓度及两颗卫星平均AOD的时空分布特征,并建立了AOD与不同时段平均PM2.5浓度之间的回归模型.结果表明：PM2.5的5 h平均浓度和24 h平均浓度值均在城区高、郊区低,最低值位于定陵站;匹配后逐时PM2.5浓度的日变化呈＂双峰型＂,最低值出现在下午,但北京西北部郊区的定陵和昌平镇站因局地山谷风环流和外部排放源的影响,其＂双峰型＂波动趋势较城区站偏弱,最低值出现在上午;AOD的空间分布特征与PM2.5浓度分布一致,但在郊区由于污染水平分布不均,卫星采集的样本可能来自于周围的清洁大气,导致AOD的最小值在郊区站点明显低于城区站点;两颗卫星平均的AOD与5h PM2.5平均浓度的决定系数高于AOD与24 h PM2.5平均浓度的决定系数;AOD与PM2.5的相关系数在城区高于郊区,郊区排放源分布不均和强的局地系统性环流是造成其相关系数低的重要原因.

英文摘要：

In this paper, surface PM2.5 observation data and aerosol optical depth （AOD） in 2014 are investigated. Hourly PM2.5 data are sampled from 12 air quality monitoring stations in Beijing. The AOD data are derived from TmTa and Aqua-MODIS with a 10 km resulution at about 10：30 AM amt I ：30 PM, respectively. In order to match these two kinds of data, 5-h averaged （10：00--14：00） and 24-h averaged （00：00-23：00） PM2s com：cntralions are calculated, and the AOD data are interpolated to the 12 air quality monitoring stations. By using the matched PM2.5 and AOI） data, the spatial and temporal distribution characteristics are discussed, and the regressions of PM2.5 versus AOD are estimated. The results show that the 5-h averaged PM2.5 concentration and the 24-h averaged PM2.5 concentration are higher in urban areas than those in suburban areas, with the lowest PM2.5 concntration at Dingling station of suburban areas. Daily time series of the matched PM2.5 concentration shows a bimodal pattern with the lowest value in the afternoon, with the amplitudes of the bimodal pattern at Dingling and Changpingzhen stations being weaker than those at the stations in urban arcas. The lowest values of the bimodal pattern at these two suburban stations appear in the nmrning due to the effects of local wind circulations and external emission sourcc, The spatial distribution of AOD is similar to that of PM2.5 concentration, with the minimum of AOD in the suburban areas being significantly lower than that m the urban areas. The minimum concentration of PM2.5 in the suburban and urban areas are similar, and the possible reasons are that the samples of AOD come from the grid of surrounding clean atmosphere while the samples of PM2.5 are from the fixed monitoring stations. AOD shows higher determination correlation with 5-h averaged PM2.5 concentrations than 24-h averaged PM2.5 concentrations. The correlations in the urban areas are significantly higher than those in the suburban areas due to the uneven distribution