2015—2021年京津冀地区空气污染物的时空变化特征

自改革开放以来,中国经济持续快速发展,与此同时环境也面临着挑战。在过去一段时间,全国城市普遍受到空气污染的严重影响,尤其是颗粒物(particulate matter,PM)污染和近期的臭氧(O₃)污染。京津冀地区作为中国发展的关键板块,其环境问题更是引起了广泛的关注。在本文中,我们系统地研究了京津冀地区六种空气污染物(CO、NO₂、O₃、PM₁₀、PM_2.5和SO₂)的时空变化特征。在2015—2021年,京津冀地区的CO、PM₁₀、PM_2.5和SO₂年浓度整体呈下降趋势,平均浓度分别降低0.11 mg/m³、7.7 mg/m³、5.4 mg/m³和4.2 mg/m³,而NO₂和O_3-8h年平均浓度呈现先增后减的趋势。其中京津冀中南部地区SO₂     

新冠疫情期间深圳市空气质量变化分析

研究2022年新冠疫情重点管控时期空气质量变化,分析人为污染源受到限制时,空气污染物变化特征。基于时间序列分析方法,根据疫情管控政策发布实施时间,将深圳市2022年3月1日—13日定义为疫情管控前期,3月14日—20日定义为疫情严控期,3月21日—31日定义为复工复产期,分析AQI以及SO₂、NO₂、PM_2.5、PM₁₀、CO、O₃等6项空气污染物浓度变化特征,比较深圳市2018—2021年同一时期空气质量,分析疫情防控措施、气象要素对该城区空气质量的影响。结果表明,2022年疫情严控期平均AQI为58.6,较往年有一定程度改善;O₃浓度为近5年最高,成为造成空气质量下降的重要指标;疫情严控期平均AQI相较疫情管控前期同比降低5.18%,复工复产期平均AQI较疫情严控期降低41.98%,空气环境质量在严格的措施管控下在短期内处于上升趋势。     

河南省春节禁燃政策对大气污染的影响分析

烟花爆竹的燃放是短期内空气质量恶化的重要原因,严重危害人体健康.利用河南省18个地市2016—2019年空气质量指数、污染物浓度(SO₂、NO₂、O₃、CO、PM_2.5和PM₁₀)和气象因子(气压、气温、相对湿度、风速、降水)数据,采用距离倒数权重插值、变异系数分析及相似性指数等方法,从多角度探究河南省春节禁燃政策的实施成效.结果表明,2016—2019年河南省春节期间的空气质量呈现逐渐改良趋势,污染出现的时间稍有提前,多出现在春节前期.禁燃对控制SO₂、PM_2.5和PM₁₀的浓度骤升(“削峰”)有很好的效果,对NO₂、O₃、CO的影响较小. 2016年烟花爆竹燃放对PM_2.5、PM₁₀和SO₂贡献量最大、贡献时间持续最长,贡献率分别为66.98%、56.32%和56.49%;到2019年,随着禁燃成效的...     

2022年夏季高温干旱对四川盆地污染物浓度变化的影响分析

2022年夏季高温干旱严重影响中国长江流域,臭氧(O₃)等污染物也出现明显异常,为研究高温干旱对污染物的影响,利用2015-2022年夏季逐小时地面空气质量和气象监测数据以及气象再分析资料,分析了夏季高温干旱特征以及对O₃和细颗粒物(PM_2.5)浓度的影响。结果表明：2022年夏季受高原暖高压和西太平洋副热带高压西伸北抬的影响,中国长江流域出现极端高温干旱天气事件,持续时间长,影响范围广,其中四川盆地和长三角地区地面温度明显偏高,相对湿度和降水量偏低,对近地面O₃和PM_2.5浓度造成了一定的影响。7-8月高温干旱对四川盆地产生的影响尤其严重。异常的高温干旱增强了大气光化学反应能力,对O₃和二次气溶胶生成有贡献,且降水对污染物的湿清除作用大大减弱,导致四川盆地O₃浓度和超标天数明显增加,PM_2.5浓度也有所升高,甚至造成持续十多天的高温热浪和O₃污染复合事件,其中对成都平原O₃     

福建省PM2.5-O3双高特征与天气形势影响分析

为理清福建省天气形势对PM_2.5与O₃演变的影响,识别二者的不同趋势与特点,揭示双高过程的气象场特征,利用2015-2021年PM_2.5与O₃连续观测资料,采用统计合成、天气学诊断等方法,探究PM_2.5与O₃变化趋势、污染状况及其与主导天气形势的关系,阐明气象因素对PM_2.5与O₃(简称“PM_2.5-O₃”)双高过程的协同作用。结果表明：2015-2021年福建省PM_2.5质量浓度年均值呈明显下降趋势,超标天数从5.6 d(2015年)下降到0.3 d(2021年)。O₃日最大8 h平均(简写为O₃-8h)质量浓度的年均值呈先上升后略下降的趋势,2018年O₃超标天数为2016年的8倍以上。天气形势对PM_2.5与O₃的影响存在一致性特征,也...     

疫情管控对关中城市群大气污染特征及颗粒物化学组成的影响

新冠肺炎疫情的暴发对生产生活模式产生了重大影响,进而改变了大气污染现状和规律,是一次极限减排的“大气实验”.本研究以中国典型大气污染控制区的关中城市群为研究对象,考察该地区疫情管控下大气污染物污染特征、来源和形成机制,解析了大气细颗粒物(PM_2.5)中化学组分的浓度变化特征.结果显示,除O₃外,其余大气污染物浓度的整体变化趋势均表现为管控前>管控后,与全国趋势一致、但程度不同.新冠肺炎疫情的社会隔离措施大幅度削减了各类排放源,西安市、咸阳市、铜川市、宝鸡市、渭南市等5个城市PM₁₀、PM_2.5、NO₂和CO浓度值明显降低,其中PM₁₀、SO₂、NO₂、CO在疫情期间的浓度达到近5年来历史最低.O₃浓度在封城期间却有显著上升的现象,表明大气氧化性可能在燃烧排放减少的背景下由于NO₂滴定效应减小而得到强化.对西安市PM_2.5中水溶性阴阳离子浓度对...     

渭河流域PM2.5时空演化及人口暴露风险

研究PM_2.5时空演化及人口暴露风险,对于环境风险评价及人居环境改善、政府环保部门制定针对性的空气污染防控政策具有重要意义。渭河流域是国家重要工业基地,也是国家级城市群和关中—天水国家级经济区的核心区域,降低该区域PM_2.5人口暴露风险是实现高质量发展的必然途径。基于渭河流域2000-2020年PM_2.5遥感反演数据和人口格网分布数据,测算人口暴露风险指数。采用Theil-Sen Median与Mann Kendall检验法,分别识别PM_2.5质量浓度值和人口暴露风险指数时间演化特征,并通过GIS空间探索工具,分析其空间变化特征。结果表明,(1)2000-2020年渭河流域PM_2.5年均质量浓度为47.2μg·m^-3,最高值为2013年的57.6μg·m^-3,最低值为2020年的31.8μg·m^-3,呈现先上升后下降的变化趋势。趋势显著性检验发现,渭河流域PM_2.5污染情况呈现好转趋势。(2...     

基于颗粒物时空分布的街道峡谷空间形态研究——以合肥市同安街道为例

在城市化加速建设的情况下,街道峡谷已成为城市建成环境重要的空间组成部分。街道峡谷中PM_2.5和PM₁₀质量浓度存在显著差异,建筑空间形态是其重要的影响因素。选取安徽省合肥市包河区同安街道为研究对象,监测PM_2.5和PM₁₀质量浓度变化并对其进行空气质量评价,从多维度视角出发研究街道峡谷空间形态,并对街道峡谷中下沉广场这一特殊建筑空间形态提出优化策略。结果表明,1)街道峡谷内PM_2.5、PM₁₀的日均质量浓度均表现出多峰变化的特点,PM_2.5和PM₁₀质量浓度最大值出现在8:00-9:00区间,最小值出现在13:00-14:00区间,因此建议同安街道的居民尽量避开工作日早高峰时期,在下午13:00-16:00期间出行,以减少颗粒物对健康的危害。2)运用AQI对同安街道各监测点空气质量进行评价,结果表明其空气质量状况以优良为主,等级多分布在1级和2级。其中选点E所在的口袋公园空气质量最优,选点B所在的商业下沉广场空气...     

2020年COVID-19疫情期间河北省空气污染特征及来源识别

2020年初新冠肺炎疫情暴发,1月24日至4月30日河北省启动重大突发公共卫生事件一级响应.针对响应期间河北省空气质量从污染演变、时空特征、PM_2.5组分、污染来源等多方面开展研究.结果表明,河北省空气质量整体较好,但全省臭氧普遍反弹（-1.3%）,二次细颗粒物污染突出（SNA占比60%以上）,重工业城市（唐山）空气质量最差,太行山沿线地区（石家庄、保定等）颗粒物污染严重,非通道城市（承德和张家口）明显反弹(PM_2.5、SO₂及CO反弹比例均50%以上).春节至元宵节期间污染过程一方面与烟花爆竹燃放相关,另一方面是由于供热、电力等基础保障类工业生产稳定,废气排放量较大行业仍处于运行状态,各项污染物排放量并位出现较大幅度降低.“十四五”及以后更长时期河北省将聚焦以二次PM_2.5和臭氧为主的二次污染治理,建议加强省控站的标准化建设和区县级以下面源管控力度,钢铁企业从均衡发展、绿色发展、产业转型、工艺结构调整等长远角度考虑走高质量发展道路,有效促进科研成果落地支撑环境管理需求.     

2015—2020年海南岛大气PM2.5和PM10的时空分布特征

基于2015—2020年海南省18个市县环境监测数据和气象观测数据,结合Cressman客观差值、相关分析和气候倾向率等统计方法对PM_2.5和PM₁₀质量浓度时空分布特征进行深入分析.结果表明,PM_2.5和PM₁₀质量浓度空间分布上均呈北半部高于南半部的特征,同时2015—2020年均表现为快速下降的变化趋势,趋势系数分别为-0.982(PM_2.5)和-0.935(PM₁₀),通过了99.9%的信度检验.PM_2.5和PM₁₀质量浓度有明显的季节变化特征,冬季质量浓度最高,秋季和春季次之,夏季最低.PM_2.5和PM₁₀质量浓度呈现U形的逐月变化,最低值出现在7月,最高值出现在12月. PM_2.5和PM₁₀质量浓度呈“双峰双谷”型的日变化,受人为活动影响较为显著. PM_2.5和PM₁₀与...     

Investigating the impact of air pollution on AMI and COPD hospital admissions in the coastal city of Qingdao,China

• The impact of air pollution on AMI/COPD hospital admissions were examined. • Significant connection was found between air pollutants and AMI/COPD in Qingdao. • Nonlinearity exists between air pollution and AMI/COPD hospital admissions. Air pollution has been widely associated with adverse effects on the respiratory and cardiovascular systems. We investigated the relationship between acute myocardial infarction (AMI), chronic obstructive pulmonary disease (COPD) and air pollution exposure in the coastal city of Qingdao, China. Air pollution in this region is characterized by inland and oceanic transportation sources in addition to local emission. We examined the influence of PM_2.5, PM₁₀, NO₂, SO₂, CO and O₃ concentrations on hospital admissions for AMI and COPD from October 1, 2014, to September 30, 2018, in Qingdao using a Poisson generalized additive model (GAM). We found that PM_2.5, PM₁₀, NO₂, SO₂ and CO exhibited a significant short-term (lag 1 day) association with AMI in the single-pollutant model among older adults (>65 years old) and females, especially during the cold season (October to March). In contrast, only NO₂ and SO₂ had clear cumulative lag associations with COPD admission for females and those over 65 years old at lag 01 and lag 03, respectively. In the two-pollutant model, the exposure-response relationship fitted by the two-pollutant model did not change significantly. Our findings indicated that there is an inflection point between the concentration of certain air pollutants and the hospital admissions of AMI and COPD even under the linear assumption, indicative of the benefits of reducing air pollution vary with pollution levels. This study has important implications for the development of policy for air pollution control in Qingdao and the public health benefits of reducing air pollution levels.     

Collaborative control of fine particles and ozone required in China for health benefit

● Increased DAAO offsets 3/4 of the decrease of DAAP in 2013–2020. ● DAAO increases are mainly due to O₃ concentration increase and population aging. ● Health benefit from PM_2.5 reduction after 2017 is larger than that before 2017. ● Reducing PM_2.5 concentration by 1% results in 0.6% reduction of DAAP. ● Reducing O₃ concentration by 1% results in 2% reduction of DAAO. PM_2.5 concentration declined significantly nationwide, while O₃ concentration increased in most regions in China in 2013–2020. Recent evidences proved that peak season O₃ is related to increased death risk from non-accidental and respiratory diseases. Based on these new evidences, we estimate excess deaths associated with long-term exposure to ambient PM_2.5 and O₃ in China following the counterfactual analytic framework from Global Burden Disease. Excess deaths from non-accidental diseases associated with long-term exposure to ambient O₃ in China reaches to 579 (95% confidential interval (CI): 93, 990) thousand in 2020, which has been significantly underestimated in previous studies. In addition, the increased excess deaths associated with long-term O₃ exposure (234 (95% CI: 177, 282) thousand) in 2013–2020 offset three quarters of the avoided excess deaths (302 (95% CI: 244, 366) thousand) mainly due to PM_2.5 exposure reduction. In key regions (the North China Plain, the Yangtze River Delta and the Fen-Wei Plain), the former is even larger than the latter, particularly in 2017–2020. Health benefit of PM_2.5 concentration reduction offsets the adverse effects of population growth and aging on excess deaths attributed to PM_2.5 exposure. Increase of excess deaths associated with O₃ exposure is mainly due to the strong increase of O₃ concentration, followed by population aging. Considering the faster population aging process in the future, collaborative control, and faster reduction of PM_2.5 and O₃ are needed to reduce the associated excess deaths.     

Mass concentrations and temporal profiles of PM10, PM2.5 and PM1 near major urban roads in Beijing

Mass concentrations of PM₁₀, PM_2.5 and PM₁ were measured near major roads in Beijing during six periods: summer and winter of 2001, winter of 2007, and periods before, during and after the 2008 Beijing Olympic Games. Since the control efforts for motor vehicles helped offset the increase of emissions from the rapid growth of vehicles, the averaged PM_2.5 concentrations at roadsides during the sampling period between 2001 and 2008 fluctuated over a relatively small range. With the implementation of temporary traffic control measures during the Olympics, a clear “V” shaped curve showing the concentrations of particulate matter and other gaseous air pollutants at roadsides over time was identified. The average concentrations of PM₁₀, PM_2.5, CO and NO decreased by 31.2%, 46.3%, 32.3% and 35.4%, respectively, from June to August; this was followed by a rebound of all air pollutants in December 2008. Daily PM₁₀ concentrations near major roads exceeded the National Ambient Air Quality Standard (Grade II) for 61.2% of the days in the non-Olympic periods, while only for 12.5% during the Olympics. The mean ratio of PM_2.5/PM₁₀ near major roads remained relatively stable at 0.55 (±0.108) on non-Olympic days. The ratio decreased to 0.48 (±0.099) during the Olympics due to a greater decline in fine particles than in coarse-mode PM. The ratios PM₁/PM_2.5 fluctuated over a wide range and were statistically different from each other during the sampling periods. The average ratios of PM₁/PM_2.5 on non-Olympic days were 0.71.     

Impact of anthropogenic heat emissions on meteorological parameters and air quality in Beijing using a high-resolution model simulation

• The Large scale Urban Consumption of energY model was updated and coupled with WRF. • Anthropogenic heat emissions altered the precipitation and its spatial distribution. • A reasonable AHE scheme could improve the performance of simulated PM_2.5. • AHE aggravated the O₃ pollution in urban areas. Anthropogenic heat emissions (AHE) play an important role in modulating the atmospheric thermodynamic and kinetic properties within the urban planetary boundary layer, particularly in densely populated megacities like Beijing. In this study, we estimate the AHE by using a Large-scale Urban Consumption of energY (LUCY) model and further couple LUCY with a high-resolution regional chemical transport model to evaluate the impact of AHE on atmospheric environment in Beijing. In areas with high AHE, the 2-m temperature (T₂) increased to varying degrees and showed distinct diurnal and seasonal variations with maxima in night and winter. The increase in 10-m wind speed (WS₁₀) and planetary boundary layer height (PBLH) exhibited slight diurnal variations but showed significant seasonal variations. Further, the systematic continuous precipitation increased by 2.1 mm due to the increase in PBLH and water vapor in upper air. In contrast, the precipitation in local thermal convective showers increased little because of the limited water vapor. Meanwhile, the PM_2.5 reduced in areas with high AHE because of the increase in WS₁₀ and PBLH and continued to reduce as the pollution levels increased. In contrast, in areas where prevailing wind direction was opposite to that of thermal circulation caused by AHE, the WS₁₀ reduced, leading to increased PM_2.5. The changes of PM_2.5 illustrated that a reasonable AHE scheme might be an effective means to improve the performance of PM_2.5 simulation. Besides, high AHE aggravated the O₃ pollution in urban areas due to the reduction in NO_x.     

New insights into the formation of ammonium nitrate from a physical and chemical level perspective

● Factor analysis of ammonium nitrate formation based on thermodynamic theory. ● Aerosol liquid water content has important role on the ammonium nitrate formation. ● Contribution of coal combustion and vehicle exhaust is significant in haze periods. High levels of fine particulate matter (PM_2.5) is linked to poor air quality and premature deaths, so haze pollution deserves the attention of the world. As abundant inorganic components in PM_2.5, ammonium nitrate (NH₄NO₃) formation includes two processes, the diffusion process (molecule of ammonia and nitric acid move from gas phase to liquid phase) and the ionization process (subsequent dissociation to form ions). In this study, we discuss the impact of meteorological factors, emission sources, and gaseous precursors on NH₄NO₃ formation based on thermodynamic theory, and identify the dominant factors during clean periods and haze periods. Results show that aerosol liquid water content has a more significant effect on ammonium nitrate formation regardless of the severity of pollution. The dust source is dominant emission source in clean periods; while a combination of coal combustion and vehicle exhaust sources is more important in haze periods. And the control of ammonia emission is more effective in reducing the formation of ammonium nitrate. The findings of this work inform the design of effective strategies to control particulate matter pollution.     

PM2.5 over North China based on MODIS AOD and effect of meteorological elements during 2003‒2015

• The Taihang Mountains was the boundary between high and low pollution areas. • There were one high value center for PM_2.5 pollution and two low value centers. • In 2004, 2009 and after 2013, PM_2.5 concentration was relatively low. Over the past 40 years, PM_2.5 pollution in North China has become increasingly serious and progressively exposes the densely populated areas to pollutants. However, due to limited ground data, it is challenging to estimate accurate PM_2.5 exposure levels, further making it unfavorable for the prediction and prevention of PM_2.5 pollutions. This paper therefore uses the mixed effect model to estimate daily PM_2.5 concentrations of North China between 2003 and 2015 with ground observation data and MODIS AOD satellite data. The tempo-spatial characteristics of PM_2.5 and the influence of meteorological elements on PM_2.5 is discussed with EOF and canonical correlation analysis respectively. Results show that overall R² is 0.36 and the root mean squared predicted error was 30.1 μg/m³ for the model prediction. Our time series analysis showed that, the Taihang Mountains acted as a boundary between the high and low pollution areas in North China; while the northern part of Henan Province, the southern part of Hebei Province and the western part of Shandong Province were the most polluted areas. Although, in 2004, 2009 and dates after 2013, PM_2.5 concentrations were relatively low. Meteorological/topography conditions, that include high surface humidity of area in the range of 34°‒40°N and 119°‒124°E, relatively low boundary layer heights, and southerly and easterly winds from the east and north area were common factors attributed to haze in the most polluted area. Overall, the spatial distribution of increasingly concentrated PM_2.5 pollution in North China are consistent with the local emission level, unfavorable meteorological conditions and topographic changes.     

嘉善冬季PM2.5化学组分特征及来源分析

为研究嘉兴地区嘉善冬季污染时段和清洁时段PM_2.5化学组分特征,结合气象数据对2019年1月嘉兴市嘉善县善西超级站在线自动监测PM_2.5及化学组分数据、气态污染物(NO₂和SO₂)进行了分析.结果表明,2019年1月嘉善善西超级站污染时段PM_2.5浓度(97.18μg·m^-3)为清洁时段(36.77μg·m^-3)的2.6倍.污染时段水溶性离子浓度(41.58μg·m^-3)较清洁时段(19.82μg·m^-3)高21.76μg·m^-3,但占比有所降低,含碳组分比例增加.OC;EC比值为3.93,可能受到燃煤及机动车排放的共同影响.低风速及高湿有利于NO₂和SO₂等气态污染物进行二次转化,污染时段硫转化率和氮转化率均比清洁时段高,分别增高7.93%和54.11%,说明NOx向硝酸盐二次转化较为明显,导致颗粒物浓度升高.聚类分析结果显示67.34%气流来自北方,且相应的气流轨迹上污染物浓度比周边高,说明污染物存在一定的长距离输送.结合风玫瑰图可以看出,污染主要为本地及其周边的输送,污染物的长距离输送在短时会使污染浓度突增.因此,在重点关注本地及周边污染的同时,偏北气流下的污染物区域输送不可忽视.     

How aerosol direct effects influence the source contributions to PM2.5 concentrations over Southern Hebei,China in severe winter haze episodes

The aerosol direct effects result in a 3%–9% increase in PM_2.5 concentrations over Southern Hebei. These impacts are substantially different under different PM_2.5 loadings. Industrial and domestic contributions will be underestimated if ignoring the feedbacks. Beijing-Tianjin-Hebei area is the most air polluted region in China and the three neighborhood southern Hebei cities, Shijiazhuang, Xingtai, and Handan, are listed in the top ten polluted cities with severe PM_2.5 pollution. The objective of this paper is to evaluate the impacts of aerosol direct effects on air quality over the southern Hebei cities, as well as the impacts when considering those effects on source apportionment using three dimensional air quality models. The WRF/Chem model was applied over the East Asia and northern China at 36 and 12 km horizontal grid resolutions, respectively, for the period of January 2013, with two sets of simulations with or without aerosol-meteorology feedbacks. The source contributions of power plants, industrial, domestic, transportation, and agriculture are evaluated using the Brute-Force Method (BFM) under the two simulation configurations. Our results indicate that, although the increases in PM_2.5 concentrations due to those effects over the three southern Hebei cities are only 3%–9% on montly average, they are much more significant under high PM_2.5 loadings (~50 μg·m⁻³ when PM_2.5 concentrations are higher than 400 μg m⁻³). When considering the aerosol feedbacks, the contributions of industrial and domestic sources assessed using the BFM will obviously increase (e.g., from 30%–34% to 32%–37% for industrial), especially under high PM_2.5 loadings (e.g., from 36%–44% to 43%–47% for domestic when PM_2.5>400 μg·m⁻³). Our results imply that the aerosol direct effects should not be ignored during severe pollution episodes, especially in short-term source apportionment using the BFM.     

Physical and chemical processes of wintertime secondary nitrate aerosol formation

The UCD/CIT model was modified to include a process analysis (PA) scheme for gas and particulate matter (PM) to study the formation of secondary nitrate aerosol during a stagnant wintertime air pollution episode during the California Regional PM_2.5/PM₁₀ Air Quality Study (CRPAQS) where detailed measurements of PM components are available at a few sites. Secondary nitrate is formed in the urban areas from near the ground to a few hundred meters above the surface during the day with a maximum modeled net increase rate of 4 μg·m^-3·d^-1 during the study episode. The secondary nitrate formation rate in rural areas is lower due to lower NO₂. In the afternoon hours, near-surface temperature can be high enough to evaporate the particulate nitrate. In the nighttime hours, both the gas phase N₂O₅ reactions with water vapor and the N₂O₅ heterogeneous reactions with particle-bound water are important for secondary nitrate formation. The N₂O₅ reactions are most import near the surface to a few hundred meters above surface with a maximum modeled net secondary nitrate increase rate of 1 μg·m^-3·d^-1 and are more significant in the rural areas where the O₃ concentrations are high at night. In general, vertical transport during the day moves the nitrate formed near the surface to higher elevations. During the stagnant days, process analysis indicates that the nitrate concentration in the upper air builds up and leads to a net downward flux of nitrate through vertical diffusion and a rapid increase of surface nitrate concentration.     

Seasonal variations of transport pathways and potential sources of PM2.5 in Chengdu,China (2012–2013)

PM_2.5 in Chengdu showed clear seasonal and diurnal variation. 5, 5, 5 and 3 mean clusters are generated in spring, summer, autumn, and winter. Short-distance air masses are important pathways in Chengdu. Emissions within the Sichuan Basin contribute significantly to PM_2.5 pollution. Long-range transport from Southern Xinjiang is a dust invasion path to Chengdu. Seasonal pattern of transport pathways and potential sources of PM_2.5 in Chengdu during 2012–2013 were investigated based on hourly PM_2.5 data, backward trajectories, clustering analysis, potential source contribution function (PSCF), and concentration-weighted trajectory (CWT) method. The annual hourly mean PM_2.5 concentration in Chengdu was 97.4 mg·m^–3. 5, 5, 5 and 3 mean clusters were generated in four seasons, respectively. Short-distance air masses, which travelled within the Sichuan Basin with no specific source direction and relatively high PM_2.5 loadings (>80 mg·m^–3) appeared as important pathways in all seasons. These short pathways indicated that emissions from both local and surrounding regions of Chengdu contributed significantly to PM_2.5 pollution. The cities in southern Chengdu were major potential sources with PSCF>0.6 and CWT>90 mg·m^–3. The northeastern pathway prevailed throughout the year with higher frequency in autumn and winter and lower frequency in spring and summer. In spring, long-range transport from southern Xinjiang was a representative dust invasion path to Chengdu, and the CWT values along the path were 30-60 mg·m^–3. Long-range transport was also observed in autumn from southeastern Xinjiang along a northwesterly pathway, and in winter from the Tibetan Plateau along a westerly pathway. In summer, the potential source regions of Chengdu were smaller than those in other seasons, and no long-range transport pathway was observed. Results of PSCF and CWT indicated that regions in Qinghai and Tibet contributed to PM_2.5 pollution in Chengdu as well, and their CWT values increased to above 30 mg·m^-3 in winter.