近年来，在我国工业化和城市化进程日益加快的同时，大气环境质量恶化现象也愈发凸显，尤其是以灰霾天气为特征的细颗粒物（PM2.5）污染频繁出现，且表现出很强的区域性和复合型的特征，引起各级政府和人民群众的广泛关注。北京及周边五省市由于人口密集，燃煤、钢铁、水泥等重化工产业结构明显，导致高强度的污染物排放，加之所处地形和气候条件不利于污染扩散，已成为我国大气PM2.5污染最严重的区域之一。为改善区域环境质量，保护公众健康，识别和量化区域大气PM2.5的主要污染来源重点地区和高污染行业，科学的控制削减污染源排放显得尤为必要。 本论文研究构建了WRF-SMOKE-CAMx复合模型系统，以本课题组建立的高分辨率北京及周边五省市主要大气污染物排放清单为输入，模拟了北京及周边五省市2012年1、4、7、10四个季节的代表月的气象状况和污染状况，并利用观测值对模拟结果进行了统计学评估和验证。在验证结果良好的前提下，分析考察了研究区域内不同季节的气象条件、污染源排放特征和污染模拟的时空分布特征，并对大气PM2.5开展来源解析研究，定量各排放源区和行业对受体城市不同季节大气环境中PM2.5浓度的贡献，进而鉴别和解析主要污染源。然后以重污染过程频发的冬季中1月为例，设置不同行业的减排情景，分析不同行业减排导致的大气PM2.5改善程度和变化的敏感度。 结果表明，在研究区域内大气PM2.5在春季的传输方向为从南至北，夏季为从东南至西北，冬季为西南和东北，而秋季的传输趋势不明显。从行业贡献角度来看，PM2.5在冬季最主要的排放污染来源行业为居民生活源，在春、夏和秋季为其他工业源（除钢铁厂和水泥厂之外的工业排放源）。以冬季为例分析PM2.5内部化学组分的来源发现，元素碳和铵盐主要来源于本地排放，硫酸盐和硝酸盐主要源于外部区域传输，居民生活源、其它工业源和钢铁厂源是PM2.5中元素碳和硫酸盐组分的主要来源行业，电厂源、交通源和其它工业源是PM2.5中硝酸盐的主要来源行业，农业源和居民生活源则是铵盐的主要来源行业。 为了探究重污染期间的大气PM2.5来源变化，解析了北京2012年1、4、7和10月日均污染来源区域和行业。结果发现,秋季和冬季重污染主要受本地和上风向区域的传输共同影响，而春季和夏季的重污染主要是由上风向区域的传输增加导致。居民生活源、其它工业源和钢铁厂源的贡献增加是冬季重污染出现的原因，农业源、电厂源和其它工业源的贡献在春季和夏季的重污染期间明显增加，其它工业源的贡献增加是秋季重污染出现的重要原因。 探究不同行业减排与大气环境中PM2.5的关系发现，1月份削减居民生活源和其它工业源的排放可使大多数受体城市中大气PM2.5浓度下降最为显著，而在唐山则是削减钢铁厂源的排放使大气PM2.5下降幅度最大。从敏感性的角度来看，居民生活源、钢铁厂源、水泥厂源和其它工业源的敏感度较高，总的来说，冬季在研究区域内控制居民生活源、钢铁厂源和其它工业源的排放可获得良好的PM2.5浓度下降和空气质量改善的正向反馈。

In recent years, with the acceleration of China's industrialization and urbanization, atmospheric environmental pollution has become more and more serious, especially the frequent occurrence of fine particulate matter (PM2.5) pollution characterized by haze weather, which shows strong regional and compound characteristics and has aroused extensive attention from the government and people. Beijing and the surrounding five provinces (BSFP) have become one of the most serious PM2.5 pollution areas in China due to the large amount of pollutant emissions, as well as the terrain and climate conditions that are not conducive to the diffusion of pollution. In order to alleviate this environmental pollution situation, it is particularly necessary to identify and quantify the major sources regions and sectors of PM2.5. In this paper, by applying the localized comprehensive emission inventory of BSFP region in 2012 as input, the WRF-SMOKE-CAMx model system is constructed to simulate the meteorological conditions and pollution conditions of BSFP region in January, April, July and October, representing four seasons in 2012. And the simulation results are statistically evaluated and verified using observations. Under the premise of accurate verification, the meteorological conditions, pollutant emission characteristics and temporal and spatial distribution characteristics of pollutant simulation in different seasons in the study area are analyzed. Then, the source apportionment of PM2.5 is conducted to quantify the contributions of various sources regions and sectors to PM2.5 concentration in the receptors in different seasons, so as to determine the dominant pollution sources. In addition, taking January as an example, set the emission reduction scenarios of different sectors, and analyze the improvement degree and sensitivity of PM2.5 caused by emission reduction in different sectors. The results show that the transport direction of PM2.5 in the study area is from south to north in spring, from southeast to northwest in summer, southwest and northeast in winter, and is not obvious in autumn. From the perspective of the sector, the dominant source of PM2.5 is the residential source in winter, and the remaining industrial source in spring, summer and autumn. Taking winter as an example to analyze the source of PM2.5 internal components, it is found that elemental carbon and ammonium are mainly derived from local emissions, and sulfates and nitrates are mainly derived from external regional transport. The residential sector, remaining industrial sector and iron and steel plants are the main sources of elemental carbon and sulfate, power plants, transportation sector and remaining industrial sector are the main sources of nitrate, agricultural sector and residential sector are the main sources of ammonium. In order to explore the PM2.5 source changes during heavy pollution period, we analyzed the source regions and sectors of daily average PM2.5 concentration in Beijing in January, April, July and October. It is found that heavy pollution in autumn and winter is mainly affected by the local sources and the transmission from upwind regions, while heavy pollution in spring and summer is mainly caused by the increase of transmission from the upwind regions. The increase of the contribution of residential sector, remaining industrial sector and iron and steel plants is the cause of heavy pollution in winter. The contribution of agricultural sector, power plants and remaining industrial sector increased significantly during heavy pollution periods in spring and summer. And remaining industrial sector is important source of heavy pollution in autumn. Exploring the relationship between emission reductions in different sectors and PM2.5 concentrations in the receptors found that the reduction ratio of PM2.5 concentrations in mostly receptors is the largest when reduce the emissions of residential sector and remaining industrial sector in January, while the reduction ratio of PM2.5 concentrations in Tangshan is largest when reduce the emission of iron and steel plants. The sensitivity analysis shows that sensitivity of residential sector, iron and steel plants, cement plants and remaining industrial sector is relatively high. In general, good air quality improvement feedback can be received by controlling emissions from residential sector, iron and steel plants and the remaining industrial sector in the study area in winter.