Atmospheric Environment 161 (2017) 235e246
Contents lists available at ScienceDirect
Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv
Spatial and temporal variation of particulate matter and gaseous pollutants in China during 2014e2016 Rui Li a, Lulu Cui a, Junlin Li a, An Zhao c, Hongbo Fu a, b, *, Yu Wu a, Liwu Zhang a, Lingdong Kong a, Jianmin Chen a, ** a Shanghai Key Laboratory of Atmospheric Particle Pollution and Prevention, Department of Environmental Science & Engineering, Fudan University, Shanghai 200433, China b Collaborative Innovation Center of Atmospheric Environment and Equipment Technology (CICAEET), Nanjing University of Information Science and Technology, Nanjing 210044, China c Key Laboratory of Poyang Lake Wetland and Watershed Research, Ministry of Education, Jiangxi Normal University, Nanchang 330022, China
h i g h l i g h t s All of the pollutants except O3 decreased slightly during 2014e2016. All of the pollutants except O3 displayed the highest levels in the winter. PM10 is a major pollutant affecting the air quality of China.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 5 December 2016 Received in revised form 4 May 2017 Accepted 6 May 2017 Available online 8 May 2017
China is experiencing severe air pollution due to rapid economic development and accelerated urbanization. High-resolution temporal and spatial air pollution data are imperative to understand the physical and chemical processes affecting air quality of China. The data of PM2.5, PM10, SO2, CO, NO2, and O3 in 187 Chinese cities during January 2014 and November 2016 were collected to uncover the spatial and temporal variation of the pollutants in China. The annual mean concentrations of PM2.5 exceeded the Grade I standard of Chinese Ambient Air Quality (CAAQS) for all of the cities except several cities in Hainan, and more than 100 cities exceeded the CAAQS Grade II standard. The concentrations of PM2.5, PM10, SO2, CO, and NO2 decreased from 2014 to 2016, whereas the O3 level increased dramatically during this period. The concentrations of PM2.5, PM10, SO2, CO and NO2 exhibited the highest levels in winter and the lowest in summer, and evidently decreased from 2014 to 2016, whereas the O3 concentration peaked in spring and summer, and dramatically increased from 2014 to 2016. The non-attainment ratios were highest in winters, while high pollution days were also frequently observed in the Southeast region in autumn and in the Northwest region in spring. Pearson correlation analysis indicated that all of the pollutants exhibited significant correlation one another. PM10 was a major pollutant affecting the air quality of China in all of the seasons. Both SO2 and NO2 exerted significantly adverse effects on the air quality in spring and autumn, but CO played an important role on the air quality in winter. O3 was found to be the dominant species among the pollutants affecting the air quality in summer, suggesting that photochemical O3 formation should be paid more attention to improve the air quality in summer. The results of geographical weight regression (GWR) showed that more significant correlations among the pollutants and the highest air quality index (AQI) appeared in the south of China. The impacts of PM10 and NO2 on the air quality increased from the east to the west of China, while SO2 and O3 exhibited the opposite variation. The data presented herein supplied an important support for the future source apportionment and intra- and inter-regional transport modeling of pollutants. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Gaseous pollutants Particulate matter Spatial and temporal variation Non-attainment China
* Corresponding author. Shanghai Key Laboratory of Atmospheric Particle Pollution and Prevention, Department of Environmental Science & Engineering, Fudan University, Shanghai 200433, China. ** Corresponding author. E-mail addresses:
[email protected] (H. Fu),
[email protected] (J. Chen). http://dx.doi.org/10.1016/j.atmosenv.2017.05.008 1352-2310/© 2017 Elsevier Ltd. All rights reserved.
236
R. Li et al. / Atmospheric Environment 161 (2017) 235e246
1. Introduction Worse air quality of urban sites has become an increasingly concerned issue due to its adverse effects on human health and climate change. The increasing health risk is associated with the elevation of particulate matter (PM) such as PM2.5 and PM10. As a typical inhalable particle, the increase of PM2.5 is closely associated with the occurrence of respiratory and cardio-vascular diseases (Araujo, 2011; Kim et al., 2015). Besides, PM plays significant roles on climate change (Paasonen et al., 2013). Some aerosol particles probably absorb or scatter the solar radiation, thereby affecting the global climate (Anenberg et al., 2012). On the other hand, they can act as cloud condensation nuclei (CCN) to govern heat transfer properties in the atmosphere, consequently altering cloud formation process and rainfall patterns (Carslaw et al., 2013; Kalkavouras et al., 2017). Moreover, regional transport of PM2.5 has been regarded as a major cause of severe haze in some cities (Li et al., 2015a,b). Apart from the effects of PM, gaseous pollutants also play significant roles on human health and environment. For example, the accumulation of many gaseous pollutants including CO, SO2, NO2, and O3 increased the susceptibility to respiratory diseases and reduced the lung fraction, thereby leading to hematological problems and cancer. In recent years, severe haze pollution has appeared in many regions of China, such as Jing-Jin-Ji, Yangtze River Delta (YRD), and Pearl River Delta (PRD), all of which could be associated with the accumulation of gaseous pollutants. SO2 and NO2 could be transformed to SO24 and NO3 via gas-toparticle process under the condition of high relative humidity (RH), which has been confirmed to cause severe haze (Huang et al., 2011a,b; Niu et al., 2016). Moreover, some gaseous pollutants such as NOx could be potential precursors to photo-oxidants such as ozone in ambient air, resulting in fog-haze pollutions (Zhou et al., 2014). Although series of control measures have been adopted by Chinese governments to alleviate atmospheric pollution, a lot of fog-haze episodes still come up frequently in the metropolitans of China, such as Beijing and Shanghai. Therefore, to investigate the spatial-temporal variations of both PM and gaseous pollutants is essential to understand current air pollution status in China, to assess the effects of control measures applied by local and central governments, and to provide scientific judgment on haze governance. A growing body of studies about temporal and spatial variations of the pollutants have been reported (Johansson et al., 2007; Karar et al., 2006; San Martini et al., 2015). Chen et al. (2016) analyzed the spatial and diurnal variation of PM2.5 in Nanjing using complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) method and elucidated the relationship between PM2.5 concentrations and meteorological factors. Wu et al. (2015) predicted the spatial-temporal variation of the concentrations of the atmospheric pollutants using land use regression method and concluded that PM2.5 showed more remarkably temporal variation than spatial variation. Spatial and temporal variation of multiple areas and the pollutants have also been reported recently. Wang et al. (2014a,b) studied the spatial and temporal variation of six criteria pollutants in 31 provincial capital cities of China and observed that the concentrations of PM2.5, PM10, CO and SO2 peaked at the cities located in the North region, followed by those in the West and the East regions. In addition, Liu et al. (2016a,b,c) investigated the spatial and temporal variation of the air pollutants, API, and mortality in120 cities of China for one year and observed significant clustering of API, the concentrations of the air pollutants and mortalities in the northwest of China. Chai et al. (2014) analyzed the annual variation of the gaseous pollutants in 26 cities of China and concluded that China was still faced an arduous challenge. On the basis of such conclusion, they thus
suggested that more efforts should be taken to control air pollution. Although some studies on the spatial distribution of the pollutants in a large scale range have been reported, the pollutant levels in dozens of the cities can not accurately reflect the spatial distribution of the pollutants in China. The methods of spatial statistics and geo-statistics (e.g., ordinary Kriging interpolation and inverse distance weight) should be applied to evaluate the spatial-temporal variation of the ambient pollutants. To the best of our knowledge, hardly any studies have concerned about ambient pollutants covered the data of multi observation stations and assessed the spatial variation of the pollutants in China through spatial interpolation methods. Herein, the officially released data of PM2.5 and PM10 and four gaseous pollutants (CO, SO2, NO2, and O3) at 187 cities of China during the period of January 2014eNovember 2016 were collected in order to quantify the pollutant levels accurately in China, as well as spatial and temporal variation trend. The aim of this study is (1) to understand the correlation of six pollutants and determine the dominant factor for the air quality index, (2) to decipher the seasonal variation of six pollutants and identify key factors, and (3) to investigate the inter-annual variability of the pollutants and assess the effects of control measures adopted by governments on the pollutant levels in recent years. The quantification of the spatial and temporal variation of six criteria pollutants will raise the awareness of authorities about the air quality in China, provide a more comprehensive understanding of current situation of air quality, and obtain appropriate strategies to promote environmental protection. 2. Methods Three-year long ambient monitoring data of PM2.5, PM10, CO, SO2, NO2, and 8-h O3 in 187 cities were analyzed to assess the air quality status in China. The real-time hourly concentrations of PM2.5, PM10, CO, SO2, NO2, and 8-h O3 at 187 cities were downloaded from the website of China air quality monitoring platform (http://www.aqistudy.cn/). Hourly air quality data of six pollutants at individual monitoring site for major cities have been published through the website from January 2013. To date, the monitoring sites have covered Mainland China. This data is imperative to supply more detailed information about the current air quality situations in China. The data of six criteria pollutants at 187 cities from January 1 st, 2014 to November 15 th, 2016 were collected because many monitoring sites had not been established before 2014. All of the data were supplied by the national air quality monitoring sites located in each city. The monitoring sites have been designed as a mixture of urban and background sites, including most of the sites in urban area, and a few of sites in suburban and rural areas as the background sites. The mean concentrations of the species were calculated on the basis of the data supplied by the monitoring sites in each city. The automated monitoring systems were installed to determine the concentrations of the gaseous pollutants including SO2, NO2, CO and O3 at each site. The continuous monitoring system of PM2.5 and PM10 is composed of the sample collection unit, the sample measurement unit, the data collection and transport unit, and other accessory equipment. SO2, NO2, and O3 were measured using the ultraviolet fluorescence method (TEI, Model 43i from Thermo Fisher Scientific Inc., USA), chemiluminescence method (TEI Model 42i from Thermo Fisher Scientific Inc., USA), and the UVspectrophotometry method (TEI model 49i from Thermo Fisher Scientific Inc., USA), respectively. CO was measured using the nondispersive infrared absorption method and the gas filter correlation infrared absorption method (TEI, Model 48i from Thermo Fisher Scientific Inc., USA).
R. Li et al. / Atmospheric Environment 161 (2017) 235e246
The quality assurance of the data were conducted based on HJ 630e2011 specifications (http://kjs.mep.gov.cn/hjbhbz/bzwb/ other/qt/201109/W020120130585014685198.pdf) before being released to the open website. The accuracy, consistency, and validity of the data have been checked based on some methods, as well as comparison with the previous data. In our study, the released data have been further processed to evaluate the pollution status in China. First, the data has no missing values, confirming the data completeness. Besides, the validity of the data has been demonstrated based on GB 3095-2012 specifications (http://hbj.new.cqcs. gov.cn/upfiles/2013-3/2013327153015207.pdf). Moreover, at least 30 daily mean concentrations except in February were applied to calculate monthly mean concentration of six criteria pollutants. The quality assurance procedure was supplied to all of the operators at 187 cities and all of the gas analyzers were calibrated with the standard gases once a week. To better confirm the quality of the data, reliability analysis was also performed. The results of reliability analysis indicated Cronbach's Alpha was 0.853 (>0.6), suggesting the reliability of the data. In addition, the independence and normality of data were also analyzed through run test and Kolmogorov-Smirnov test, respectively, which implicated the data showed favorable independence and normality. Furthermore, some previous studies have also utilized some statistics methods to demonstrate the quality of the data is reliable (Zhao et al., 2016; Xie et al., 2015). One-Way Analysis of Variance (ANOVA) (Fisher Test, p < 0.05) was utilized to identify the significant difference of air quality index (AQI), gaseous pollutants, and PM during the sampling periods. Pearson correlation analysis was used to decipher the relationships between AQI and the concentrations of six criteria pollutants. Grey correlation analysis (GCA) was used to determine the predominant factor affecting air quality in China. The original independents and dependents were represented as Xi(k) and X0(k), respectively. Grey data processing must be transformed to be dimensionless before they couldbe calculated. Each series was normalized by dividing the data in the original series by their average. The grey correlation coefficient ðxiðkÞ Þ between the independents and dependents could be calculated as following (Tian et al., 2014):
xiðkÞ ¼
min minjX0 ðkÞ Xi ðkÞj þ 0:5 max maxjX0 ðkÞ Xi ðkÞj i
i
k
k
jX0 ðkÞ Xi ðkÞj þ 0:5 max maxjX0 ðkÞ Xi ðkÞj i
k
(1) Where min minjX0 ðkÞ Xi ðkÞj is the minimum absolute distance i k among all of the independents, max maxjX0 ðkÞ Xi ðkÞj is the i k maximum absolute distance. The grey correlation grade (gi ðkÞ) is an average of the grey correlation coefficients and it was defined as (Tian et al., 2014):
gi ðkÞ ¼
Pn
k¼1 xi ðkÞ
n
(2)
However, Pearson correlation analysis and grey correlation analysis could not be utilized to investigate the spatial correlation of AQI and the pollutant levels. Thus, geographical weight regression (GWR) model was used to produce the coefficient of determination (R2) and local regression coefficients for each city of the study areas, which were then mapped to show the spatial variability. Regression coefficients could be calculated using weighted least squares with the following weighting function (Brunsdon et al., 1996):
bðui ; vi Þ ¼ X T Wðui ; vi ÞX
237
1
X T Wðui ; vi ÞY
(3)
where bðui ; vi Þ denotes the local regression coefficient at city i; X is the matrix of the independent variables; Y is the vector of the dependent variable; and W(ui,vi) is an n order matrix in which the diagonal elements are the spatial weighting of the observed samples. The spatial weight function was calculated using the exponential distance decay form:
. Wðui ; vi Þ ¼ exp d2 ðui ; vi Þ b2
(4)
where d(ui,vi) is the distance between location i and j, and b is the kernel bandwidth. Inverse distance weighted (IDW), ordinary Kriging (OK), and Voronoi neighborhood averaging (VNA) were widely used to investigate the spatial variation of the pollutants. The leave-oneout cross-validation (LOOCV) method was applied to evaluate the performances of these models. R2 and root mean squared error (RMSE) were used to demonstrate the performances of these models. In our study, OK for all of six pollutants showed the highest values of R2 and the lowest values of RMSE compared with IDW and VNA (Table S1). Thus, OK interpolation method was selected to generate spatially continuous values to explore the spatial distribution of the pollutants. All of the statistical analysis and figures shown hereinwere performed by the software package SPSS 16.0 and ArcGIS 9.3 for Windows. 3. Results and discussion 3.1. Annual variation of the ambient pollutants in China 3.1.1. The concentration levels of the ambient pollutants The annual average concentrations of PM2.5 ranged from 16.28 mg/m3 (Sanya) to 101.63 mg/m3 (Baoding). PM2.5 in all of the cities exceeded the Grade I standard (15 mg/m3) of CAAQS (GB30952012) and 84% of the cities exceeded the Grade II standard (35 mg/ m3) (Fig. 1). The annual mean concentrations of PM10 varied between 31.30 mg/m3 (Sanya) and 179.66 mg/m3 (Korla). PM10 in all of the cities except Haikou and Sanya exceeded the Grade I standard (40 mg/m3) of CAAQS and 74% of the cities exceeded the Grade II standard of CAAQS (70 mg/m3). The mean concentrations of SO2 ranged from 2.77 mg/m3 (Sanya) to 85.29 mg/m3 (Zibo). SO2 in 122 cities exceeded the Grade I standard (20 mg/m3), while only several cities in Hebei and Shandong Province exceeded the Grade II standard of CAAQS (60 mg/m3). The mean concentrations of NO2 ranged from 12.74 mg/m3 (Beihai) to 59.31 mg/m3 (Zibo). NO2 in 56 cities exceeded the Grade I standard (40 mg/m3), and 171 cities exceeded the Grade II standard (20 mg/m3). Averaged CO and O3 concentrations ranged from 0.48 mg/m3 (Rongcheng) to 2.78 mg/ m3 (Taiyuan) and from 58.16 mg/m3 (Nanchong) to 121.75 mg/m3 (Yixing), respectively. To date, the standard value of CO or O3 were not defined in CAAQS to evaluate their pollution levels. Lower concentrations of PM and the gaseous pollutants were generally observed in the coastal city such as Haikou, Sanya, and Beihai, all of which were usually affected by strong wind and turbulence. However, some gaseous pollutants such as SO2, CO and NO2 usually peaked in the industrial cities such as Zibo and Taiyuan. A growing body of studies have attributed to the higher levels in some industrial cities to the emissions from coal combustion or biomass burning (Chai et al., 2014; Pui et al., 2014). The updated inventory of NO2 and SO2 showed that power plants and industry sector were main sources of NO2 and SO2 (Zhang et al., 2012; Lu et al., 2010). Wang et al. (2005) showed that the contribution of industrial sector
238
R. Li et al. / Atmospheric Environment 161 (2017) 235e246
Fig. 1. Inter-annual variation of PM2.5 and PM10 during 2014e2016 in China.
for CO emission was markedly higher than that of other sectors. Yixing is a typical industrial city in the YRD region. Rapid growth of transportation, industries, and urbanization in some cities of the YRD region has become the major driving force for the O3 pollution (Hu et al., 2014). Huang et al. (2011a,b) found that many VOCs species such as ethylene, m, p-xylene, o-xylene, toluene, 1,2,4trimethylbenzene, 2,4-dimethylpentane, ethyl benzene, propylene, 1-pentene, and isoprene were the main species accounting for 77% of the total ozone formation potential. Korla possessed the highest PM10 concentration (179.66 mg/m3) among 187 cities, which could suffer from frequent dust storm, particularly in spring (Liu et al., 2004). The highest PM2.5 concentration peaked in Baoding because of the combined effects of SO2 and VOCs. Liu et al. (2016a,b,c) reported that SO2 could enhance the formation of secondary aerosol from anthropogenic VOCs and promote new particle formation. Compared with other regions in China, NCP displayed higher SO2 concentration, which may cause the elevation of PM2.5. 3.1.2. Inter-annual variation of six pollutants The mean concentrations of PM2.5 in China significantly decreased from 60.82 ± 20.08 mg/m3 in 2014 to 45.13 ± 13.97 mg/m3 in 2016 (Fig. 1), which reflected the efficient reduction of the PM2.5 levels during the past years. The central and local governments implemented various control measures, such as the increased use of nuclear and hydroelectric power, restricting the building of new cement plants, ceramics factories, and glassworks, phasing out of small coal-fired power generation units, establishment of strict emission standards for industrial boilers, and improvements in energy conservation and emissions reduction of vehicle fuel (Fu et al., 2014). The inter-annual variation of PM10 was in accordance with that of PM2.5. The averaged PM10 concentration markedly decreased from 101.25 ± 34.37 mg/m3 in 2014 to 81.40 ± 28.43 mg/ m3 in 2016. Although the PM concentrations displayed a sharp reduction, they showed remarkable spatial heterogeneity. For
example, the PM10 concentration significantly decreased as a whole in recent years, whereas the averaged PM2.5 in Xinjiang Province and some cities of North China Plain (NCP) such as Luoyang increased slightly. During the past several years, Xinjiang Province has been suffering from severe haze due to increasing vehicle exhausts, domestic heating, and cooking (Zheng and Liu, 2012). Although many control measures such as strictly regulating vehicle travel and closing heavy industrial factories have been conducted in Beijing and its surrounding areas, the aim of “Zero Increase” of PM2.5 was not achieved yet. The mean concentrations of SO2, CO, and NO2 decreased from 33.77 ± 18.47 mg/m3 to 21.86 ± 11.79 mg/m3, from 1.19 ± 0.40 mg/m3 to 1.01 ± 0.32 mg/m3, and from 36.17 ± 11.01 mg/m3 to 32.00 ± 9.51 mg/m3 during the period of 2014e2016, respectively (Fig. 2). The main source of SO2 was coal-fired power plants in China (Lu et al., 2010; Zhao et al., 2008). The total industrial SO2 emission, particularly in Northeast China and NCP decreased considerably after the installation of flue gas desulfurization (FGD) systems in thermal power units and the closure of small and lessefficient power plants (http://www.stats.gov.cn/tjsj/ndsj/). In addition, most of sinter processes adopted electrostatic precipitators (ESP) for the PM removal and fabric filters (FFs) were more widely employed in steel-making, ironemaking and cement production process, which sharply reduced the emissions of SO2 (Wang et al., 2016a,b; Hua et al., 2016). It was well documented that industrial activities and vehicle emissions were major contributors of NO2 in China (Streets et al., 2003). The number of private automobiles linearly increased from 59.38 million in 2010 to 135.39 million in 2015 with a striking growth rate of 17.5% each year (http://www.stats.gov.cn/tjsj/ndsj/). However, the NO2 emissions per automobile decreased slightly after upgrading of oil product quality standards (Li et al., 2016). Meanwhile, many factories were obliged to employ low-NO2 burner technologies and import denitrification facilities after the implementation of emission standards
R. Li et al. / Atmospheric Environment 161 (2017) 235e246
239
Fig. 2. Inter-annual variation of the gaseous pollutants during 2014e2016 in China.
for coal-fired power plants. Recently, the updated standard (GB13223-2011) with most stringent emission limits (100 mg/Nm3) has been proposed, demanding that all of the newly built plants and most of the in-use plants must install advanced SCR or SNCR devices to reduce NO2 emission (Tian et al., 2013). Thus, the contribution of vehicle emissions was probably counteracted by controlling the NO2 emissions, thereby leading to the slight decrease of NO2 concentration in the ambient air. The CO
concentration only exhibited slight decrease from 2014 to 2016, in contrast to the dramatically decreased SO2. The predominant anthropogenic CO originated from the combustion of biomass and fossil fuels. Energy consumption structure in some regions of China has changed significantly in recent years, which could decrease the emission of CO (Lindner et al., 2013). For example, straw recycling has been conducted instead of open burning or fuels in many areas € bbing et al., of Inner Mongolia, which lead to less CO released (Ko
240
R. Li et al. / Atmospheric Environment 161 (2017) 235e246
2014). The mean concentration of O3 displayed opposite inter-annual variation to other gaseous pollutants, which increased from 87.65 ± 16.74 mg/m3 in 2014 to 98.57 ± 14.86 mg/m3 in 2016 (Fig. 2). O3 is a secondary pollutant, which is generally formed in the atmosphere through photochemical pathways of NOx and volatile organic compounds (VOCs). Remarkable increase of O3 was observed in NCP, YRD, Inner Mongolian Plateau, and even southeastern of Tibetan Plateau, although NO2 exhibited slight decrease recently. This may be attributed to the fact that the PM concentrations in most of the regions displayed remarkable decrease in recent years, which promote the formation of O3 because of stronger solar radiation. Besides, the emission of VOC increased considerably in some industrialized areas (Yuan et al., 2013), leading to the elevation of O3 through the reaction with NOx. Especially, the emissions of aromatic hydrocarbons in NCP and YRD displayed significant increase in recent years (Zhang et al., 2015). Qiu et al. (2014) established a new inventory of NMVOCs in China during 1980e2010. They found that processes using NMVOCbearing products were the main source of NMVOCs and the predominant driver force for the increase of NMVOCs. Wang et al. (2014a,b) reported that vehicle emission was a main source of NMVOCs in Beijing, with a relative contribution of 52e69%. Significant increase of NMVOCs, which could react with NOx, leads to an increased O3 concentration, although the NO2 concentration decreased slightly in recent years. In addition, the reduction of the NOx emission in NCP probably inhibited the titration reaction of NO and O3, resulting in an elevation of O3 concentration in this region (Xu et al., 2016a,b). It should be noted that Lhasa and surrounding areas displayed high O3 concentration, which was usually regarded as a clean site due to a lack of industrial activities. Higher O3 concentrations could be ascribed to a strong stratosphere-troposphere exchange process due to lower height of troposphere in Tibetan Plateau (Skerlak et al., 2014). Additionally, Lhasa generally showed strong solar radiation and long duration of sunshine, particularly in spring and summer, favoring a photochemical production of O3 (Ran et al., 2014).
3.2. The concentration levels of six pollutants in different seasons Averaged PM2.5 concentration exhibited great temporal variability with the highest value in winter and the lowest one in summer (Fig. 3). The mean concentration of PM2.5 in spring was approximate to that in autumn. Higher PM2.5 concentration in winter were probably contributed by coal combustion and/or biomass burning for residential heating (Che et al., 2014). The sharp increase of coal consumption in winter could lead to the elevation of the SO2 concentration, resulting in the increase of sulfate, which is one of the main components of PM2.5. In addition, less precipitation, lower temperature and boundary layer height, and weaker winds in winter may further exacerbate ambient pollution (Xu et al., 2016a,b). Secondary sulfate and nitrate were inclined to accumulate in winter due to lower boundary layer height and precipitation (Zheng et al., 2015). The PM2.5 concentration in winter peaked at NCP and Xinjiang Province. Industrial sources could play significant roles on the PM2.5 accumulations at some cities such as Beijing, Tianjin and Urumqi (Wang et al., 2012; Zhao et al., 2013). Besides, farmers’ periodic emissions probably lead to the elevation of PM2.5 in NCP (Liu et al., 2016a,b,c). The PM10 concentration followed in the order of winter (116.35 ± 41.78 mg/m3) > spring (100.18 ± 36.75 mg/m3) > autumn (86.27 ± 28.46 mg/m3) > summer (66.53 ± 22.19 mg/m3), which was similar to the seasonal variation of PM2.5 (Fig. 3). Apart from the impacts of heating, Taklamakan desert has the greatest wind erosion flux, of which emission could be transported by strong northwest wind to the east of China (Wang et al., 2015). Some researchers observed different kinds of dust events (floating dust, dust storm and blowing dust) were contributed to main PM10 pollution sources in spring in northwestern China (Wang et al., 2006). Wang et al. (2012) reported that the contribution ratio of crustal dust to PM10 reached 39.87% in the spring of northern city. Thus, the PM10 concentration in spring was markedly higher than that of other seasons. The mean concentrations of SO2 and CO exhibited similar seasonal variation along with PM2.5 and PM10, reflecting the effects of meteorological conditions and emission sources (Fig. 4). Fossil fuel
Fig. 3. Seasonal variation of PM2.5 and PM10 in China.
R. Li et al. / Atmospheric Environment 161 (2017) 235e246
241
Fig. 4. Seasonal variation of the gaseous pollutants in China.
combustion for heating probably contributed to the accumulation of SO2 and CO (Lu et al., 2011; Zhang et al., 2009). Additionally, stagnant meteorological conditions characterized by slow wind speed and shallow mixing layers appeared more frequently in winter, which could trap the pollutants and resulted in the accumulations of SO2 and CO (Tai et al., 2010). In contrast, great solar radiation, strong turbulent eddies and precipitation scavenging diluted the pollutants released at the surface and cause lower SO2 and CO concentrations in summer (Antony Chen et al., 2001). Although SO2 and CO displayed the similar seasonal distributions to PM2.5 and PM10, they showed remarkable spatial differentiation. SO2 and CO displayed high values in Northeast China, NCP, Loess Plateau, and Xinjiang province, suggesting that coal combustion for heating exert vital roles on the formation of SO2 and CO because North China is main coal producer and consumer in winter.
However, CO displayed high concentration in the winter of Tibetan Plateau, which was not in accordance with the spatial distribution of SO2. The possible reason for a high CO level but a low SO2 level might be due to the difference in the emission sources. Fossil fuel burning was the primary source of SO2, while CO was also originated from biomass burning. Despite lack of large-scale industrial activities, biomass burning including open crop straw burning and indoor fuel combustion is widespread in Tibetan Plateau (Li et al., 2015a,b). In general, the emission contribution of crop residual combustion is significantly higher than that of most industrial sectors. As a consequence, CO concentration displayed higher concentration in Tibetan Plateau (Wang et al., 2005; Tian et al., 2011). The averaged concentration of NO2 was in the order of winter (42.55 ± 12.66 ppb) > autumn (35.45 ± 10.86 ppb) > spring
242
R. Li et al. / Atmospheric Environment 161 (2017) 235e246
(34.76 ± 10.45 ppb) > summer (25.60 ± 7.86 ppb). Aside from NCP and Xinjiang province, NO2 also peaked at YRD. Recently, NO2 significantly elevated in the YRD region as a result of the rapid increase of industrial activities and the number of automobiles (Pan et al., 2015). However, the concentration of O3 displayed an opposite seasonal variation with the highest value in summer (114.30 ± 23.78 ppb) but the lowest one (57.18 ± 13.14 ppb) in winter. Weak sunlight in winter could inhibit the formation of O3 because the formation rate of O3 depended on the intensity of solar radiation (Zhao et al., 2016). In contrast, the elevated O3 levels in spring and summer should be affected by the stratospheretroposphere exchange process, as reported by Yamaji et al. (2006). Furthermore, high temperature and strong solar radiation tended to generate large amount of OH radical, resulting in the formation of O3 through the reaction of VOC and OH radical (Ou et al., 2015). Apart from the reasons mentioned above, biomass burning was regarded as another factor for the highest concentration of O3 in summer, which was more frequently emitted at major harvest season of wheat and corn in northern China (Sun et al., 2016). However, no marked enhancement of O3 was observed in winter, suggesting a slow photochemical activity in winter (Boynard et al., 2014). 3.3. Air quality in China The annual average AQI value inChina during 2014e2016 ranged from 38.12 (Sanya) to 143.73 (Baoding). AQI among all of the cities except Haikou, and Sanya, exceeded the Grade I standard (50), and 40 cities exceeded the Grade II standard (100). The mean concentrations of AQI in China decreased from 93.04 ± 22.94 in 2014 to 81.09 ± 18.80 in 2016, suggesting the air quality of China has improved significantly. The non-attainment ratios were defined as the ratios with AQI exceeding CAAQS (the Grade I and the Grade II). Mean non-attainment ratios of the Grade I and the Grade II for all of cities during 2014e2016 are summarized in Fig. S1. The nonattainment ratios of the Grade I ranged from 0.16 (Sanya) to 0.98 (Anshan), while non-attainment ratios of the Grade II increased from 0 (Huizhou) to 0.68 (Hengshui). Most of the cities possessed non-attainment ratios of the Grade I over 30% except Sanya and
Haikou, whereas 52 of 187 cities experienced severe air pollution and more than 30% of the days exceeding the Grade II standards. Non-attainment ratios of the Grade I in the Northwest region and NCP varied between 0.8 and 1.0, while they were generally lower than 0.7 in the most areas of Yangtze Plain and Northeast China. Non-attainment ratios of the Grade II peaked in the NCP region (>0.6) and did not display remarkable spatial variation among Northwest, Southeast, and the Northeast regions of China, the same as those of the Grade I. Thus, air quality in the NCP region was significantly worse than those of the Southeast and Northeast regions of China. Substantially seasonal differentiation of non-attainment ratios was observed in the various regions of China (Fig. 5). The best air quality in summer in China with non-attainment ratios of the Grade I lower than 25% and non-attainment ratios of the Grade II lower than 15%, while the most regions in China generally displayed worst air quality in winter with non-attainment ratios of the Grade I higher than 80% and the Grade II higher than 50%. Although all of the regions in China exhibited seasonal difference of nonattainment ratios, the coefficient of variation (CV) differed in various regions. The CV of the non-attainment ratios were in the order of Southeast (the Grade I: 0.86, the Grade II: 1.38) > Southwest (the Grade I: 0.83, the Grade II: 1.18) > Northeast (the Grade I: 0.81, the Grade II: 1.17) > Northwest (the Grade I: 0.78, the Grade II: 1.07) > NCP (the Grade I: 0.78, the Grade II: 1.06). In the Southeast region, all of the cities showed strong seasonal variation in the nonattainment ratios. The attainment ratios of the Grade I and the Grade II in winter were 4e15 and 10e25 times of those in summer, respectively, suggesting that air quality of the Southeast region is best in summer due to strong wind and high intensity of solar irradiation. Besides, biomass burning in winter could be a key factor contributing to air pollution in this region (Zha et al., 2013). In NCP, no obvious seasonal variation was observed in some cities such as Anyang and Baoding, although they showed severe air pollution in winter. This was attributed to the contribution of the periodic emissions from farmers’ activities in NCP. Strong windblown dust could contribute to the elevation of non-attainment ratios in spring and summer, thereby degrading the variation in the Northwest region (Wang et al., 2006).
Fig. 5. Seasonal variation of the non-attainment ratios of China.
R. Li et al. / Atmospheric Environment 161 (2017) 235e246
3.4. Correlation between the ambient pollutants and AQI 3.4.1. Relationship of six pollutants The Pearson correlation coefficients of the air pollutants were calculated on the basis of the data supplied by 187 monitoring sites. PM2.5, PM10, SO2, NO2, and CO displayed significantly pairwise positive correlation (p < 0.01), reflecting the similar origin of these pollutants (Table 1). The correlation coefficients of SO2/NO2 and PM2.5 were remarkably higher than those of CO and PM2.5, although all of variables displayed extremely significant relativity. It was supposed that SO2 and NO2 are important precursors of SO24 and NO 3 , respectively, which constituted of main components of PM2.5 (Huang et al., 2011a,b). One can see that O3 was significantly negatively correlated with other five pollutants (p < 0.01), which was ascribed to that high gaseous pollutants such as SO2, NO2, and CO, as well as aerosol particles probably weaken the solar radiation. O3 was inclined to forming in the clear day with high intensity of solar radiation so that it exhibited negative correlation with other pollutants (Pochanart, 2015). O3 displayed higher correlation coefficient with NO2 than that of other pollutants because of the O3 depletion during the transformation from NO to NO2 (Zhou et al., 2014). The correlations among the pollutants also displayed highly seasonal variations (Table S2). In spring, all of six pollutants except O3 were markedly correlated with each other in China, while O3 became significantly correlated with the other pollutants in other seasons, particularly in summer and winter. The coal emissions increased considerably in winter as a result of heating, thereby probably leading to severe haze in winter. The formation of O3 could be prohibited under the unfavorable condition characterized with low intensity of solar radiation. However, O3 exhibited significantly positive relationships to the other five pollutants in summer, which was in agreement with the previous studies (Wang et al., 2014a,b; Zhao et al., 2016). SO2 and NO2 generally displayed closer relationships to PM2.5/PM10 than those of CO and O3, which could be attributed to the heterogeneous reactions in the atmosphere (Li and Shao, 2009). However, CO exhibited higher interdependency on PM2.5 and PM10 than those of other gaseous pollutants in winter, which was related to the release of CO to the atmosphere due to the incomplete combustion of coal for heating (Duan et al., 2014). Furthermore, the CO emission factor from coal combustion was significantly higher compared with ones of SO2 and NOx (Wang et al., 2005; Fu et al., 2013).
Table 1 Correlations of the pollutants and AQI based on the daily data in 2014e2016.
AQI PM2.5 PM10 SO2 CO NO2 O3
AQI
PM2.5
PM10
SO2
CO
NO2
O3
1 0.93b 0.91b 0.60b 0.13b 0.93b 0.91b
0.93b 1 0.87b 0.62b 0.12b 0.74b 0.37b
0.91b 0.87b 1 0.61b 0.09b 0.68b 0.27b
0.60b 0.62b 0.61b 1 0.10b 0.55b 0.37b
0.13b 0.12b 0.09b 0.10b 1 0.12b 0.06b
0.68b 0.74b 0.68b 0.55b 0.12b 1 0.39b
0.16b 0.37b 0.27b 0.37b 0.06b 0.39b 1
a: p < 0.05, b: p < 0.01.
243
3.4.2. Determination of the major pollutant AQI was dependent on the combined effects of six pollutants. Therefore, it was difficult to seek out the key factors affecting AQI only through Pearson correlation analysis. However, GCA is an effective method to determine the dominant factor influencing AQI. PM10 showed higher grey correlation coefficient with AQI than that of other five pollutants in the whole year (Table 2), indicating the significant effects of coarse particles originated from dust on the regional air quality. Remarkable variation among the seasons was observed for other pollutants. Both SO2 and NO2 exhibited higher grey relationship with AQI than those of other gaseous pollutants in spring and autumn, which suggested that SO24 and NO3 were the main water-soluble ions that affecting the formation of fog-haze episode in spring and autumn. Li et al. (2013) found higher sulfur oxidation ratio (SOR) and nitrate oxidation ratio (NOR) in spring and concluded that the large proportions of SO24 and NO3 were the key factor leading to the formation of haze. CO displayed the highest grey correlation with AQI in winter, which was possibly associated with incomplete combustion of fossil fuels in the heating season (Liang et al., 2015). Besides, CO replaced other gaseous pollutants as the second major pollutant in the later autumn and the early winter because biomass burning occurs frequently in the harvest season. The grey correlation coefficients among all of the pollutants and AQI were lower in summer than the other seasons due to the stronger photochemical reaction and higher atmospheric boundary layer (Lyu et al., 2016). However, O3 was a major pollutant in summer, implying that the pollution was promoted by thephotochemical processes due to the strong solar radiation. 3.4.3. Spatial relevance of AQI and six pollutants The spatial variation of the pollutants was not considered, although the relationship between the pollutants and AQI could be deciphered through Pearson correlation analysis andGCA. Thus, the GWR model was applied to examine the relationship between the pollutants and AQI with regard of spatial variation. The R2 values and local regression coefficients are shown in Fig. 6, which provided a direct way to detect the spatially varying relationships. The GWR regression models improved the explanatory power of the Pearson correlation analysis declared by the much higher R2 values. The R2 value increased from the north to the south of China, indicating higher correlation between the pollutants and AQI in the south of China. Local B coefficients of the pollutants presented clear spatial variations, implying the different effects on AQI. The local B coefficient of PM2.5 displayed similar spatial distribution to the R2 value, which implied that PM2.5 played a significant role in the spatial distribution of AQI. The local B coefficient of PM10 and NO2 increased from the east to the west of China, while that of SO2 and O3 exhibited the opposite variation. Asian dust exerted great impacts on regional air quality because it could release PM10 to the atmosphere, which occured more frequently in the west of China (Wang et al., 2016a,b). Many eastern cities suffered from severe O3 pollution, indicating the impacts of highly oxidized secondary pollutants. The influence of CO peaked at NCP, suggesting that the emissions such as biomass burning and coal combustion could significantly contribute to the regional air quality. 4. Conclusions
Table 2 The grey correlation degrees between AQI and the other pollutants.
Year Spring Summer Autumn Winter
PM2.5
PM10
SO2
CO
NO2
O3
0.71 0.81 0.52 0.80 0.64
0.90 0.90 0.66 0.88 0.74
0.84 0.84 0.56 0.82 0.69
0.61 0.52 0.50 0.51 0.72
0.75 0.75 0.49 0.77 0.52
0.64 0.62 0.60 0.61 0.54
Air quality in China was analyzed on the basis of the data of PM2.5, PM10, SO2, CO, NO2, and O3 in 187 cities in China from January 2014 to November 2016. The annual mean concentrations of PM2.5 exceeded the Grade I of CAAQS for all of the cities except Haikou and Sanya, and more than 100 cities exceeded the Grade II standard of CAAQS. The concentrations of PM2.5, PM10, SO2, CO, and NO2 decreased from 2014 to 2016, whereas the O3 concentration level
244
R. Li et al. / Atmospheric Environment 161 (2017) 235e246
Fig. 6. Local R2 values and B coefficients for AQI.
increased dramatically during this period. PM2.5, PM10, SO2, CO, and NO2 displayed the highest levels in winter and the lowest level in summer, indicating the emissions from fossil fuels combustion and biomass burning. However, the O3 concentration peaked in spring and summer, which was associated with the strong stratospheretroposphere exchange process and solar radiation. The nonattainment ratios in the most cities were highest in winter, while high pollution days were also frequently observed in the Southeast region in autumn and in the Northwest region in spring. Pearson correlation analysis suggested that all of the pollutants exhibited the significant correlation with each other. PM10 is a major pollutant affecting the air quality of China in all of the seasons. SO2 and NO2 exerted significantly adverse effects on the air quality in spring and autumn, which implied that SO24 and NO3 were the main water-soluble ions that affecting the formation of fog-haze episode. However, CO played a significant role on the air quality in winter, which could be related to the incomplete combustion of fossil fuels. O3 displayed high level in summer, inferring the impacts promoted by photochemical process. The GWR model suggested that the correlation of the pollutants and AQI peaked in the south of China. The impacts of PM10 and NO2 on air quality increased from east to west of China, while SO2 and O3 exhibited the opposite variation, suggesting that China not only should reduce the atmospheric pollutant emissions by implementing pollutant emission control strategies, but also perform a synchronous control of the pollutants in the various regions and conduct the joint prevention and control of air pollution.
Acknowledgements This work was supported by National Natural Science Foundation of China (Nos. 21577022, 21190053, 40975074), Ministry of Science and Technology of China (2016YFC0202700), and International cooperation project of Shanghai municipal government (15520711200).
Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.atmosenv.2017.05.008.
References Anenberg, S.C., Schwartz, J., Shindell, D.T., Amann, M., Faluvegi, G.S., Klimont, Z., et al., 2012. Global air quality and health co-benefits of mitigating near-term climate change through methane and black carbon emission controls. NASA Tec. Rep. 120, 831e839. Antony Chen, L.W., Doddridge, B.G., Dickerson, R.R., Chow, J.C., Mueller, P.K., Quinn, J., Butler, W.A., 2001. Seasonal variations in elemental carbon aerosol, carbon monoxide and sulfur dioxide: implications for sources. Geophys. Res. Lett. 28, 1711e1714. Araujo, J.A., 2011. Particulate air pollution, systemic oxidative stress, inflammation, and atherosclerosis. Air Qual. Atmos. Hlth 4, 79e93. Boynard, A., Clerbaux, C., Clarisse, L., Safieddine, S., Pommier, M., Van Damme, M., Bauduin, S., Oudot, C., Hadji-Lazaro, J., Hurtmans, D., 2014. First simultaneous space measurements of atmospheric pollutants in the boundary layer from IASI: a case study in the NCP. Geophys. Res. Lett. 41, 645e651. Brunsdon, C., Fotheringham, A.S., Charlton, M.E., 1996. Geographically weighted regression: a method for exploring spatial nonstationarity. Geogr. Anal. 28, 281e298. Carslaw, K., Lee, L., Reddington, C., Pringle, K., Rap, A., Forster, P., Mann, G., Spracklen, D., Woodhouse, M., Regayre, L., 2013. Large contribution of natural aerosols to uncertainty in indirect forcing. Nature 503, 67e71. Chai, F., Gao, J., Chen, Z., Wang, S., Zhang, Y., Zhang, J., Zhang, H., Yun, Y., Ren, C., 2014. Spatial and temporal variation of particulate matter and gaseous pollutants in 26 cities in China. J. Environ. Sci-China 26, 75e82. Che, H., Xia, X., Zhu, J., Li, Z., Dubovik, O., Holben, B., Goloub, P., Chen, H., Estelles, V., , E., 2014. Column aerosol optical properties and aerosol radiative Cuevas-Agullo forcing during a serious haze-fog month over NCP in 2013 based on groundbased sunphotometer measurements. Atmos. Chem. Phys. 14, 2125e2138. Chen, T., He, J., Lu, X., She, J., Guan, Z., 2016. Spatial and temporal variations of pm2.5 and its relation to meteorological factors in the urban area of Nanjing, China. Inter. J. Env. Res. pub. Heal 13, 921. Duan, L., Sun, H., Zhao, C., Zhou, W., Chen, X., 2014. Coal combustion characteristics on an oxy-fuel circulating fluidized bed combustor with warm flue gas recycle. Fuel 127, 47e51. Fu, X., Wang, S.X., Zhao, B., Xing, J., Cheng, Z., Liu, H., 2013. Emission inventory of primary pollutants and chemical speciation in 2010 for the Yangtze River Delta region, China. Atmos. Environ. 70, 39e50. Fu, X., Wang, X., Guo, H., Cheung, K., Ding, X., Zhao, X., He, Q., Gao, B., Zhang, Z., Liu, T., 2014. Trends of ambient fine particles and major chemical components in the Pearl River Delta region: observation at a regional background site in fall and winter. Sci. Total Environ. 497, 274e281. Hu, J.L., Wang, Y.G., Ying, Q., Zhang, H.L., 2014. Spatial and temporal variability of PM2.5 and PM10 over the North China Plain and the Yangtze River Delta, China. Atmos. Environ. 95, 598e609. Hua, S.B., Tian, H.Z., Wang, K., Zhu, C.Y., Gao, J.J., Ma, Y.L., et al., 2016. Atmospheric emission inventory of hazardous air pollutants from China's cement plants: temporal trends, spatial variation characteristics and scenario projections. Atmos. Environ. 128, 1e9. Huang, K., Zhuang, G., Lin, Y., Fu, J., Wang, Q., Liu, T., Zhang, R., Jiang, Y., Deng, C., 2011a. Typical types and formation mechanisms of haze in an eastern Asia megacity, Shanghai. Atmos. Chem. Phys. Dis. 11, 21,713e721,767. Huang, C., Chen, C.H., Li, L., Cheng, Z., Wang, H.L., Huang, H.Y., 2011b. Emission inventory of anthropogenic air pollutants and VOC species in the Yangtze River
R. Li et al. / Atmospheric Environment 161 (2017) 235e246 Delta region, China. Atmos. Chem. Phys. 11, 4105e4120. Johansson, C., Norman, M., Gidhagen, L., 2007. Spatial & temporal variations of PM10 and particle number concentrations in urban air. Environ. Monit. Assess. 127, 477e487. €bbing, J.F., Patuzzi, F., Baratieri, M., Beckmann, V., Thevs, N., Zerbe, S., 2014. Ko Economic evaluation of common reed potential for energy production: a case study in Wuliangsuhai Lake (Inner Mongolia, China). Biomass Bioenergy 70, 315e329. Kalkavouras, P., Bossioli, E., Biskos, G., Mihalopoulos, N., Nenes, A., Tombrou, M., 2017. Impact of Chemical Composition on NPF, CCN and Droplet Formation at South Aegean Sea during Summertime Etesians, Perspectives on Atmospheric Sciences. Springer, pp. 875e881. Karar, K., Gupta, A., Kumar, A., Biswas, A.K., 2006. Seasonal variations of PM10 and TSP in residential and industrial sites in an urban area of Kolkata, India. Environ. Monit. Assess. 118, 369e381. Kim, K.-H., Kabir, E., Kabir, S., 2015. A review on the human health impact of airborne particulate matter. Environ. Int. 74, 136e143. Li, P., Yan, R., Yu, S., Wang, S., Liu, W., Bao, H., 2015a. Reinstate regional transport of PM2.5 as a major cause of severe haze in Beijing. P. Natl. Acad. Sci. U. S. A. 112, E2739eE2740. Li, W., Shao, L., 2009. Transmission electron microscopy study of aerosol particles from the brown hazes in Northern China. J. Geophys. Res. 114. Li, W.J., Chen, S.R., Xu, Y.S., Guo, X.C., Sun, Y.L., Yang, X.Y., 2015b. Mixing state and sources of submicron regional background aerosols in the northern QinghaieTibet Plateau and the influence of biomass burning. Atmos. Chem. Phys. 15, 13365e13376. Li, X., Wang, L., Ji, D., Wen, T., Pan, Y., Sun, Y., Wang, Y., 2013. Characterization of the size-segregated water-soluble inorganic ions in the Jing-Jin-Ji urban agglomeration: spatial/temporal variability, size distribution and sources. Atmos. Environ. 77, 250e259. Li, Y., Meng, J., Liu, J.F., Xu, Y., Guan, D.B., Tao, W., Huang, Y., Tao, S., 2016. Interprovincial reliance for improving air quality in China: a case study on black carbon aerosol. Environ. Sci. Technol. 50, 4118e4126. Liang, X., Zou, T., Guo, B., Li, S., Zhang, H., Zhang, S., Huang, H., Chen, S.X., 2015. Assessing Beijing's PM2.5 pollution: severity, weather impact, APEC and winter heating. Proc. R. Soc. A. R. Soc. 20150257. Lindner, S., Liu, Z., Guan, D., Geng, Y., Li, X., 2013. CO2 emissions from China's power sector at the provincial level: consumption versus production perspectives. Renew. Sust. Energ. Re 19, 164e172. rquez, S., Frank, A., Haas, C.N., 2016a. Liu, L., Yang, X., Liu, H., Wang, M., Welles, S., Ma spatialeTemporal analysis of air Pollution, climate change, and Total Mortality in 120 cities of China, 2012e2013. Front. Public Health 4. Liu, P.F., Zhang, C.L., Mu, Y.J., Liu, C.T., Xue, C.Y., Ye, C., 2016b. The possible contribution of the periodic emissions from farmers' activities in the NCP to atmospheric water-soluble ions in Beijing. Atmos. Chem. Phys. 16, 10097e10109. Liu, T., Wang, X., Hu, Q., Deng, W., Zhang, Y., Ding, X., 2016c. Formation of secondary aerosols from gasoline vehicle exhaust when mixing with SO2. Atmos. Chem. Phys. 16, 675e689. Liu, X., Tian, L., Zhang, X., 2004. Influence of spring dust activities over the Taklimakan Desert area on concentrations of atmospheric PM10 in east of Northwest China. China Environ. Sci. 24, 528e532. Lu, Z., Streets, D.G., Zhang, Q., Wang, S., Carmichael, G.R., Cheng, Y.F., Wei, C., Chin, M., Diehl, T., Tan, Q., 2010. Sulfur dioxide emissions in China and sulfur trends in East Asia since 2000. Atmos. Chem. Phys. 10, 6311e6331. Lu, Z., Zhang, Q., Streets, D.G., 2011. Sulfur dioxide and primary carbonaceous aerosol emissions in China and India, 1996e2010. Atmos. Chem. Phys. 11, 9839e9864. Lyu, X., Chen, N., Guo, H., Zeng, L., Zhang, W., Shen, F., Quan, J., Wang, N., 2016. Chemical characteristics and causes of airborne particulate pollution in warm seasons in Wuhan, central China. Atmos. Chem. Phys. 16, 10671e10687. Niu, H., Hu, W., Zhang, D., Wu, Z., Guo, S., Pian, W., Cheng, W., Hu, M., 2016. Variations of fine particle physiochemical properties during a heavy haze episode in the winter of Beijing. Sci. Total Environ. 571, 103e109. Ou, J., Zheng, J., Li, R., Huang, X., Zhong, Z., Zhong, L., Lin, H., 2015. Speciated OVOC and VOC emission inventories and their implications for reactivity-based ozone control strategy in the Pearl River Delta region, China. Sci. Total Environ. 530, 393e402. Pan, X., Kanaya, Y., Tanimoto, H., Inomata, S., Wang, Z., Kudo, S., Uno, I., 2015. Examining the major contributors of ozone pollution in a rural area of the YRD region during harvest season. Atmos. Chem. Phys. 15, 6101e6111. € a €l€ Paasonen, P., Asmi, A., Petaja, T., Kajos, M.K., Aij a, M., Junninen, H., et al., 2013. Warming-induced increase in aerosol number concentration likely to moderate climate change. Nature 6, 438e442. Pochanart, P., 2015. Residence time analysis of photochemical buildup of ozone in central eastern China from surface observation at Mt. Tai, Mt. Hua, and Mt. Huang in 2004. Environ. Sci. Pollut. R. 22, 14087e14094. Pui, D.Y., Chen, S.-C., Zuo, Z., 2014. PM2.5 in China: measurements, sources, visibility and health effects, and mitigation. Particuology 13, 1e26. Qiu, K.Q., Yang, L.X., Lin, J.M., Wang, P.T., Yang, Y., Ye, D.Q., Wang, L.M., 2014. Historical industrial emissions of non-methane volatile organic compounds in China for the period of 1980-2010. Atmos. Environ. 86, 102e112. Ran, L., Lin, W., Deji, Y., La, B., Tsering, P., Xu, X., Wang, W., 2014. Surface gas pollutants in Lhasa, a highland city of Tibetecurrent levels and pollution implications. Atmos. Chem. Phys. 14, 10721e10730. San Martini, F.M., Hasenkopf, C.A., Roberts, D.C., 2015. Statistical analysis of PM2.5
245
observations from diplomatic facilities in China. Atmos. Environ. 110, 174e185. Skerlak, B., Sprenger, M., Pfahl, S., Roches, A., Sodemann, H., Wernli, H., 2014. Rapid exchange between the stratosphere and the planetary boundary layer over the Tibetan Plateau, EGU General Assembly Conference Abstracts, p. 9903. Streets, D.G., Bond, T., Carmichael, G., Fernandes, S., Fu, Q., He, D., Klimont, Z., Nelson, S., Tsai, N., Wang, M.Q., 2003. An inventory of gaseous and primary aerosol emissions in Asia in the year 2000. J. Geophys. Res. 108. Sun, L., Xue, L., Wang, T., Gao, J., Ding, A., Cooper, O.R., Lin, M., Xu, P., Wang, Z., Wang, X., 2016. Significant increase of summertime ozone at Mount Tai in central Eastern China. Atmos. Chem. Phys. 16, 10637e10650. Tai, A.P., Mickley, L.J., Jacob, D.J., 2010. Correlations between fine particulate matter (PM2.5) and meteorological variables in the United States: implications for the sensitivity of PM2.5 to climate change. Atmos. Environ. 44, 3976e3984. Tian, G., Qiao, Z., Xu, X., 2014. Characteristics of particulate matter (PM10) and its relationship with meteorological factors during 2001e2012 in Beijing. Environ. Pollut. 192, 266e274. Tian, H., Zhao, D., Wang, Y., 2011. Emission inventories of atmospheric pollutants discharged from biomass burning in China. Acta Sci. Circumstantiae 31, 349e357. Tian, H.Z., Liu, K.Y., Hao, J.M., Wang, Y., Gao, J.J., Qiu, P.P., Zhu, C.Y., 2013. Nitrogen oxides Emissions from thermal power plants in China: current status and future predictions. Environ. Sci. Tech. 47, 11350e11357. Wang, H., Jia, X., Li, K., Li, Y., 2015. Horizontal wind erosion flux and potential dust emission in arid and semiarid regions of China: a major source area for East Asia dust storms. Catena 133, 373e384. Wang, K., Tian, H.Z., Hua, S.B., Zhu, C.Y., Gao, J.J., Xue, Y.F., Hao, J.M., 2016a. A comprehensive emission inventory of multiple air pollutants from iron and steel industry in China: temporal trends and spatial variation characteristics. Sci. Total Environ. 559, 7e14. Wang, L.T., Zhang, Q., Hao, J.M., He, K.B., 2005. Anthropogenic CO emission inventory of Mainland China. Acta Sci. Circumstantiae 25, 1580e1585. Wang, M., Shao, M., Chen, W., Yuan, B., Lu, S., Zhang, Q., Zeng, L., Wang, Q., 2014a. A temporally and spatially resolved validation of emission inventories by measurements of ambient volatile organic compounds in Beijing, China. Atmos. Chem. Phys. 14, 5871e5891. Wang, Q., Zhuang, G., Huang, K., Liu, T., Lin, Y., Deng, C., Fu, Q., Fu, J.S., Chen, J., Zhang, W., 2016b. Evolution of particulate sulfate and nitrate along the Asian dust pathway: secondary transformation and primary pollutants via long-range transport. Atmos. Res. 169, 86e95. Wang, S., Yuan, W., Shang, K., 2006. The impacts of different kinds of dust events on PM10 pollution in northern China. Atmos. Environ 40, 7975e7982. Wang, Y., Ying, Q., Hu, J., Zhang, H., 2014b. Spatial and temporal variations of six criteria air pollutants in 31 provincial capital cities in China during 2013e2014. Environ. Int. 73, 413e422. Wang, Z.S., Wu, T., Shi, G.L., Fu, X., Tian, Y.Z., Feng, Y.C., Wu, X.F., Wu, G., Bai, Z.P., Zhang, W.J., 2012. Potential source analysis for PM10 and PM2.5 in autumn in a northern city in China. Environ. Sci. Eco 12, 39e48. Wu, J., Li, J., Peng, J., Li, W., Xu, G., Dong, C., 2015. Applying land use regression model to estimate spatial variation of PM2.5 in Beijing, China. Environ. Sci. Pollut. R. 22, 7045e7061. Xie, Y.Y., Zhao, B., Zhang, L., Luo, R., 2015. Spatiotemporal variations of PM2.5 and PM10 concentrations between 31 Chinese cities and their relationships with SO2, NO2, CO and O3. Particulogy 20, 141e149. Xu, R.G., Tang, G.Q., Wang, Y.S., Tie, X.X., 2016a. Analysis of a long-term measurement of air pollutants (2007-2011) in North China Plain (NCP); Impact of emission reduction during the Beijing Olympic Games. Chemosphere 159, 647e658. Xu, W., Wu, Q., Liu, X., Tang, A., Dore, A.J., Heal, M.R., 2016b. Characteristics of ammonia, acid gases, and PM2.5 for three typical land-use types in the NCP. Environ. Sci. Pollut. R. 23, 1158e1172. Yamaji, K., Ohara, T., Uno, I., Tanimoto, H., Kurokawa, J.-i., Akimoto, H., 2006. Analysis of the seasonal variation of ozone in the boundary layer in East Asia using the Community Multi-scale Air Quality model: what controls surface ozone levels over Japan? Atmos. Environ. 40, 1856e1868. Yuan, B., Hu, W., Shao, M., Wang, M., Chen, W., Lu, S., Zeng, L., Hu, M., 2013. VOC emissions, evolutions and contributions to SOA formation at a receptor site in eastern China. Atmos. Chem. Phys. 13, 8815e8832. Zha, S., Zhang, S., Cheng, T., Chen, J., Huang, G., Li, X., Wang, Q., 2013. Agricultural fires and their potential impacts on regional air quality over China. Aerosol Air Qual. Res. 13, 992e1001. Zhang, Q., Streets, D.G., Carmichael, G.R., He, K., Huo, H., Kannari, A., Klimont, Z., Park, I., Reddy, S., Fu, J., 2009. Asian emissions in 2006 for the NASA INTEX-B mission. Atmos. Chem. Phys. 9, 5131e5153. Zhang, H., Li, J., Ying, Q., Yu, J.Z., Wu, D., Cheng, Y., et al., 2012. Source apportionment of PM2.5 nitrate and sulfate in China using a source-oriented chemical transport model. Atmos. Environ. 62, 228e242. Zhang, Z., Wang, X., Zhang, Y., Lü, S., Huang, Z., Huang, X., Wang, Y., 2015. Ambient air benzene at background sites in China's most developed coastal regions: exposure levels, source implications and health risks. Sci. Total Environ. 511, 792e800. Zhao, S., Yu, Y., Yin, D., He, J., Liu, N., Qu, J., Xiao, J., 2016. Annual and diurnal variations of gaseous and particulate pollutants in 31 provincial capital cities based on in situ air quality monitoring data from China National Environmental Monitoring Center. Environ. Int. 86, 92e106. Zhao, X., Zhao, P., Xu, J., Meng, W., Pu, W., Dong, F., He, D., Shi, Q., 2013. Analysis of a
246
R. Li et al. / Atmospheric Environment 161 (2017) 235e246
winter regional haze event and its formation mechanism in the NCP. Atmos. Chem. Phys. 13, 5685e5696. Zhao, Y., Wang, S., Duan, L., Lei, Y., Cao, P., Hao, J., 2008. Primary air pollutant emissions of coal-fired power plants in China: current status and future prediction. Atmos. Environ. 42, 8442e8452. Zheng, B., Zhang, Q., Zhang, Y., He, K., Wang, K., Zheng, G., Duan, F., Ma, Y., Kimoto, T., 2015. Heterogeneous chemistry: a mechanism missing in current models to
explain secondary inorganic aerosol formation during the January 2013 haze episode in North China. Atmos. Chem. Phys. 15, 2031e2049. Zheng, C.-X., Liu, B., 2012. Speciation analysis of Cd in PM10 and PM2. 5 during heating period in Urumqi. Spectrosc. Spect. Anal. 32, 235e238. Zhou, R., Wang, S., Shi, C., Wang, W., Zhao, H., Liu, R., Chen, L., Zhou, B., 2014. Study on the traffic air pollution inside and outside a road tunnel in Shanghai, China. PloS one 9, e112195.
