Introduction

Urban air pollution receives increasing attention due to its remarkable environmental, climate, and health effects (Du and Li 2016). As the largest developing country, economic growth in China over the past decades has been one of the strongest in the world history (Kan et al. 2012). However, such a rapid economic expansion is greatly driven by the use of fossil fuels such as coal, petroleum, and natural gases, which results in considerable increases of both gaseous and particulate matter pollutants in ambient air. Seventy percent of the most polluted cities in the world are located in China, and air quality of only 5 cities met World Health Organization Air Quality Guidelines in China (Bapna 2012). Many laws and regulations have been formulated by Chinese central and local governments in recent years to reduce primary emissions and improve urban air quality (GB3095-1996, GB13223-2003). Additionally, the new air quality standards were released in late February 2012 (GB3095-2012), and O3, PM2.5 (particulate matter with aerodynamic diameter less than 2.5 μm), and CO were included in the newly revised standards. However, some air quality degradation events have been extensively reported in many Chinese cities (Chan and Yao 2008; Cheng et al. 2013; Qu et al. 2010; Chai et al. 2014; Li et al. 2014; Wang et al. 2008a, 2008b, 2014a, 2014b, 2014c, 2014d; Chen and Xie 2014), especially extreme haze-fog episodes in North China Plain (Che et al. 2009; Zhao et al. 2013; Guo et al. 2014; Zhang et al. 2014; Wang et al. 2014a, 2014b, 2014c, 2014d; Han et al. 2014; Huang et al. 2014). The reasons that caused the phenomenon were investigated using different methods such as satellite data and air quality model. Some researchers indicated that the frequent extreme haze-fog episodes in China were caused by a large amount of emissions of trace gaseous pollutants, aerosols, and ozone precursors due to frequent human activities (Tan et al. 2016; Wang et al. 2013), while other studies showed that weather conditions such as large-scale circulation and local meteorological conditions (wind speed and direction, temperature, and relative humidity) become increasingly worse for diffusion of air pollutants due to barrier effect of Qinghai-Tibet Plateau to central and eastern China (Cao et al. 2015; Jia et al. 2015; Xu et al. 2016; Zheng et al. 2015). To understand current fine particulate matter pollution situations and its temporal and spatial variations in China, the central government decided to monitor PM2.5 in the most polluted provincial capital cities including three key regions, i.e., the Yangtze River Delta, the Pearl River Delta, and the North China Plain. Since January 2013, the hourly real-time concentrations of PM2.5, PM10, SO2, NO2, CO, and O3 for 74 major cities have been released online (http://113.108.142.147:20035/emcpublish/). These data with high resolutions can be used to study air pollution levels in some major Chinese cities and investigate formation mechanism of extreme haze events in central or eastern China using some statistical methods or numerical models (Hu et al. 2014).

The spatial–temporal distributions of air pollutants concentrations in China have been reported by some researchers based on the limited observation data (Wang et al. 2008a; Wang et al. 2008b; Han et al. 2009; Meng et al. 2009; Sun et al. 2013; Zheng et al. 2013; Chai et al. 2014; Chen and Xie 2014; Li et al. 2014; Wang et al. 2014a, 2014b; Wang et al. 2014c, 2014d; Zhang et al. 2014; Martini et al. 2015). Most of the studies are limited in one single megacity such as urban areas of Guangzhou (Li et al. 2014), Beijing (Wang et al. 2008a, 2008b) and Tianjin (Han et al. 2009), or one single air pollutant such as PM10 (Johansson et al. 2007; Karar et al. 2006; Qu et al. 2010), PM2.5 (Martini et al. 2015) and NO2 (Lewne et al. 2004). PM2.5 mass concentration levels were assessed by Martini et al. (2015) using some statistical methods based on the air quality data from US Consulates in Beijing, Chengdu, Guangzhou, Shanghai, and Shenyang. It was found that the diurnal variations of PM2.5 concentrations were significant across seasons and between cities. The previous studies pay much less attention to larger spatial coverage of multiple air pollutants due to scarce air quality monitoring stations. Recently, some studies were conducted on spatial and temporal variations of nationwide multiple pollutants including PM2.5, PM10, SO2, NO2, CO, and O3 based on hourly mean data published by the Ministry of Environmental Protection (MEP) of China (Chai et al. 2014; Wang et al. 2014a, 2014b, 2014c, 2014d). Their studies found that concentrations of SO2, CO, PM2.5, and PM10 were much higher in northern cities than those in southern cities while NO2 and O3 did not show significant differences between northern and southern cities. The number of days exceeded Chinese Ambient Air Quality Standards (CAAQS) was the highest in the winter, but highly polluted days also frequently occurred in the southeastern China during autumn due to biomass burning and in the west region during spring due to dust events. The previous studies gave a basic understanding of current air pollution situations in China, but few of them cover long-term fine particle mass concentrations data (PM2.5, particulate matter with aerodynamic diameter less than 2.5 μm) and most of them divide pollution regions only based on geographical location of the cities.

The objective of this study is to investigate nationwide spatial and seasonal variations of PM2.5, PM10, SO2, NO2, CO, and O3 using the air quality data published by MEP for 31 provincial capital cities from April 2014 to March 2015. This study provides useful information regarding the status of nationwide air pollution. The paper was organized as follows. Data and methods used in this study such as monitoring systems and locations were introduced in detail in “Data and methods.” Results and discussion were given in “Results and discussion” and concluded in “Conclusions.”

Data and methods

The data used in this study were from China National Environmental Monitoring Center (http://113.108.142.147:20035/emcpublish/). The arrangement rules of state controlling air sampling sites, the instruments measured gaseous and particulate matter pollutants, the quality assurance, and controls of data and analytical methods such as cluster analysis were introduced in detail by Zhao et al. (2016), which can be referred to if one wanted to know that.

Results and discussion

Overview of air pollution situations in China

The Chinese capital cities experienced severe air pollution in recent years such as long-lasting haze events in North China Plain due to more primary emissions from human activities and poor diffusion conditions. To better see air pollution levels in China, the annual mean concentrations of gaseous (SO2, NO2, CO, and O3) and particulate matter pollutants (PM2.5 and PM10) and the ratios of PM2.5 to PM10 in the 31 provincial capital cities were summarized in Table 1. The cities were divided into two classes, i.e., one with winter heating and the other without winter heating to know effect of domestic heating on urban air quality in China. The annual mean PM2.5 concentrations varied from 42.3 μg/m3 (Huhehaote) to 102.1 μg/m3 (Shijiazhuang) for the cities with home heating in winter, which were much higher than those for the cities without winter heating with lowest concentration of 22.4 μg/m3 in Haikou. No cities in China were up to Chinese Ambient Air Quality Standards (CAAQS) Grade I standard (15 μg/m3) of PM2.5, and only four cities including Fuzhou, Haikou, Kunming, and Lasa met the CAAQS Grade II standard (35 μg/m3). The annual mean concentrations of PM10 were between 40.3 μg/m3 (Haikou) and 176.1 μg/m3 (Shijiazhuang). PM10 exceeded the CAAQS Grade I standard (40 μg/m3) in all cities and exceeded the CAAQS Grade II standard (70 μg/m3) with the exception of part cities without winter heating including Fuzhou, Guangzhou, Guiyang, Haikou, Kunming, and Lasa. CO and SO2 concentrations in Taiyuan were the highest among the provincial capital cities partly due to the amount mining of coal in Shanxi. SO2 concentrations exceeded Grade I standard (20 μg/m3) in the cities with winter heating, and the cities exceeded Grade II standard (60 μg/m3) have domestic heating in winter. NO2 concentration differences between the cities with winter heating and those without winter heating were less evident. NO2 was higher in Beijing than the other cities, which may be attributed to more vehicle numbers in Beijing. No clear spatial difference was found for O3 among the cities, which was accordance with the results of Chai et al. (2014). As a typically secondary pollutant, O3 is formed by some photochemical reactions between NO x and volatile organic compounds (VOC) in the atmosphere (Atkinson 2000). To some extent, the spatial patterns of O3 were in contrast to other pollutants, such as the concentrations of PM2.5, PM10, CO, NO2, and SO2 were low in Lasa, while O3 (77.1 ± 31.6 μg/m3) was the highest among 31 provincial capital cities in China.

Table 1 Annual mean concentrations of the six criteria pollutants and the ratios of PM2.5 to PM10 in the 31 capital cities
Full size table

Higher concentrations of PM2.5, PM10, CO, and SO2 were generally observed in the cities with winter heating, indicating effect of home heating on urban air quality in the northern China. Xiao et al. (2015) analyzed the impact of winter heating on aerosol loadings over China using the MODIS-Aqua Collection 6 aerosol product from 2004 to 2012 and found that domestic heating had a large contribution to elevated PM2.5 concentrations in more than three quarters of central and eastern China. Furthermore, some studies showed that high concentrations in the North China were due to the emissions from fossil fuel combustion, coal-based industries, and biomass burning (Chai et al. 2014; Pui et al. 2014; Qu et al. 2010; Yang et al. 2013; Zheng et al. 2005; Zhang, et al. 2009; Zhao et al. 2011).

Cluster analysis was used to classify the annual variations of each pollutant in the 31 cities as several groups with similar pollutant concentrations and variation trends within groups to better see air pollution situations in major Chinese cities. SO2 and PM2.5 were divided into five and three clusters, respectively, while the remaining air pollutants formed four groups each. The clustering results from annual variations of each pollutant in 31 provincial capital cities and the corresponding cities to each cluster are shown in Figs. 1 and 2, respectively. The red, gray, and blue symbols, respectively, represented severely, moderately, and slightly polluted cities in Fig. 2. The air pollution in winter was more severe than that in other seasons mainly due to more primary emissions such as domestic heating (Xiao et al. 2015), poor diffusional conditions, and shallow boundary layer in winter. Some studies revealed local meteorology largely impacted on air pollution levels with the negative correlations between meteorological factors such as temperature, relative humidity, precipitation, wind speed and mixing layer height (MLH), and air pollution index (API) (Han et al. 2009; Xu et al. 2011). High NO x concentration in ambient air can also depress producing of O3 and lead to low O3 concentration near the ground surface (Han et al. 2009). The nitrogen monoxide (CO) from combustion is a major sink for O3, and thus, the concentrations of O3 are negatively correlated with the concentrations of all the combustion emitted pollutants.

Fig. 1
figure 1

Clustering results for annual variations of monthly mean concentrations of six criteria air pollutants in 31 provincial capital cities

Full size image
Fig. 2
figure 2

The corresponding cities to each cluster in Fig. 1. The red, gray, and blue symbols represented severely, moderately, and slightly polluted cities, respectively

Full size image

The differences between clusters for each pollutant were less in spring and summer than those in fall and winter, especially SO2 due to more coal combustion for domestic heating in northern China and adverse diffusion conditions. The mean annual variation of PM2.5 in northern cities (Cluster A1) was bimodal with peaks in October 2014 and January 2015, which mainly appears in North China Plain (Beijing, Jinan, Shijiazhuang, Tianjin, and Zhenzhou) and Northeast China (Changchun, Haerbin, and Shenyang), while that in southern cities (Cluster A2) was less evident with slightly high PM2.5 in winter (see Figs. 1 and 2). The annual variations of O3 were generally in contrast to that of other pollutants with the highest value in summer (Fig. 1). However, O3 and PM2.5 have significantly high values in October 2014 for the cities of North China Plain and Northeast China (Clusters A1 and E3 of Fig. 1). O3 is formed by some photochemical reactions between NO x and VOCs in the low atmosphere (Atkinson 2000). Liu et al. (2005) estimated that vehicular exhaust was the major contributor of VOC (on average 57.7 %), followed by painting operations, gasoline vapor, and liquefied petroleum gas (LPG) at 12.4, 11.3, and 5.8 %, respectively. The high O3 concentration in October 2014 in North China Plain indicated that high volatile organic compounds in the atmosphere due to more motor vehicles in the two regions. Therefore, the high PM2.5 concentration in October 2014 in northern cities (Cluster A1) may be due to more secondary organic aerosols (SOA) from chemical reactions of VOCs. The study of Chai et al. (2014) also showed that monthly variation of PM2.5 had a weak peak in October 2011 in northern cities, which was consistent with the results from our study. High PM2.5 concentration in January 2015 for Clusters A1 and A3 may be due to integrated effects of primary emissions such as domestic heating and adverse diffusional conditions. The annual variation of PM10 concentrations in northern cities also had a remarkable peak in April–May 2014 with the exception of the peaks in October 2014 and January 2015, and the variations of PM2.5 and PM10 were inconsistent in April–May 2014 due to larger impact of dust events on PM10 than on PM2.5 (Clusters A1 and B4) (Zhao et al. 2015c). However, the variations of PM2.5 and PM10 concentrations were synchronous in southern China (Clusters A2 and B2). Furthermore, elevated CO (Clusters C1 and C2) and SO2 (Clusters F2 and F5) concentrations all appeared in northern cities (see Fig. 2). For NO2, Cluster D1 had significant low value in November and only Wulumuqi was assigned to the cluster. Particulate matter (PM2.5 and PM10) and other gaseous pollutants (SO2 and CO) also had significant low values in November 2014 in Wulumuqi, especially PM10 and SO2, which is in accordance with NO2 (data not shown). Therefore, the phenomenon that Cluster D1 has significant low value in November is true and the NO2 concentration is valid. The low concentration of five criteria air pollutants (PM2.5, PM10, SO2, CO, and NO2) in November 2014 in Wulumuqi was maybe related to favorable weather conditions. The precipitation was 31.6 mm in November 2014 in Wulumuqi, which was much higher than other months with the exception of April and May 2014 (data not shown). The air pollutants were partly scavenged from the atmosphere by raindrops or snow particles. Clusters F2 and F5, which mainly appears in North China Plain (Tianjin, Zhenzhou, Jinan, and Taiyuan) and Northeast China (Shenyang, Haerbin, and Changchun), had marked seasonal variation with maximal SO2 in winter season due to domestic heating, while other clusters have much weaker seasonal variation in SO2 concentration and mainly appears in southern cities without winter heating in those cities. The weak peak of SO2 concentration in Clusters F1, F3, and F4 was mainly related to poor diffusional conditions in winter season in southern cities. Stagnant air with weak winds and low boundary layer height appeared frequently in winter, and thus, the air pollutants were trapped near the surface layer and local air pollution was aggravated (Zhao et al. 2015b). The above analyses indicated that SO2 concentrations in the cities of North China Plain and Northeast China in winter season were much higher than other regions of China, indicating that there were apparent spatial feature of air pollution in China. However, the regional feature of air pollution cannot be better showed in our study due to only used the data from provincial capital cities and the distance among the cities is relatively far. Additionally, the O3 was significantly higher in Fuzhou, Lasa, and Shanghai than that in other cities, while other primarily emitted pollutants such as SO2 and CO were low in the three cities.

The cluster analysis was also used to divide mean diurnal variations of each air pollutant in all cities in the winter and clustering results were shownin Fig. 3. As can be seen from the figure, the diurnal variations of O3 concentration were unimodal with a peak at 15:00∼16:00 (Beijing time, hereafter the same), while other air pollutants (PM2.5, PM10, SO2, NO2, and CO) had two obvious peaks at morning and evening rush hours due to effect of on-road motor vehicles. Furthermore, it is worth noting that the air pollutants from primary emissions also had a peak in winter early morning which were comparable with or sometimes more significant than the morning and evening peaks due to stagnant weather and the shallow planetary boundary layer in winter morning causing high concentration of air pollutants near surface layer in these cities. The above analyses indicated that local adverse weather conditions such as more stagnant air and the lower planetary boundary layer in winter played a major role in winter air pollution. Additionally, the pulses of some air pollutant concentrations such as CO (Clusters DC2 and DC3) and O3 (Cluster DE2) can be seen in Fig. 3. After careful analyzing, we confirmed the phenomenon was primarily related to underlying discharges from some industries during nighttime in some cities (Zhao et al. 2016).

Fig. 3
figure 3

Clustering results for diurnal variations of hourly mean concentrations of six criteria air pollutants in 31 provincial capital cities in the winter

Full size image

Spatial and seasonal variations of six criteria pollutants

PM2.5 mass concentrations have large temporal variation with the highest in the winter (December 2014–February 2015) and the lowest in the summer (June–August 2014) for the Chinese provincial capital cities (Fig. 4), which was mainly related to more emissions and more adverse meteorological conditions in winter as source emissions and meteorological field are two major factors influencing concentrations of air pollutants. Furthermore, less emissions and favorable diffusion and dilution conditions in summer lead to low air pollution in China. The spring (April–May 2014) and fall (September–November 2014) concentrations almost were comparable, but PM2.5 concentrations were much higher in fall than those in spring for the cities located at northeastern China probably due to biomass burning near the cities in fall. The elevated PM2.5 in winter were caused by combined effect of both domestic heating and poor meteorological conditions for air pollution dilution and diffusion (Chai et al. 2014). The PM2.5 mass concentrations were even higher in eastern China than those in other areas in four seasons, especially North China Plain (Beijing, Tianjin, Shijiazhuang, Jinan, and Zhenzhou) due to large urban agglomeration in NCP. Additionally, industrial sources can also lead to high PM2.5 in the cities with heavy industry. For example, Cheng et al. (2012) indicated that about 27 % of PM2.5 can be attributed to industrial emissions in Wuhan.

Fig. 4
figure 4

Spatial distribution of seasonal mean PM2.5 mass concentrations (μg/m3) during the study period (the gray-shaded areas represent desert and desertified land in China)

Full size image

PM10 reached maximum in the spring in some cities of Northwestern China such as Wulumuqi, Xining, Lanzhou, and Yinchuan (Fig. 5). Dust plumes came from Asian desert areas were main pollution sources of particulate matter in northwestern cities in spring. The dust plumes, originated from Taklimakan and Gobi Deserts, were transported by strong winds to downstream cities and then affected urban air quality (Qu et al. 2010; Zhao et al. 2015a; Chen et al. 2015). The study of Ta et al. (2004) found that the “high PM10 episodes” in northern China were mostly due to desert dust invasions. In addition, Wang et al. (2006) studied the effects of floating dust, dust storm, and blowing dust on PM10 mass concentration in four north Chinese cities including Beijing, Huhehaote, Xi’an, and Lanzhou and found that floating dust had the largest contribution to PM10 followed by dust storm and blowing dust. As can also be seen from Fig. 6, PM2.5-10 (particles with size between 2.5 and 10 μm) accounted for a large proportion of PM10 in Northern China in spring (67 %), indicating effects of dust events and domestic heating on urban air quality. For the other cities, PM10 concentrations show a similar seasonal variation as PM2.5 and especially in the southern cities. PM2.5 concentration accounted for PM10 was high in Southern China (>55 %), resulting in the synchronously seasonal variations between PM2.5 and PM10 mass concentrations (Fig. 6).

Fig. 5
figure 5

Spatial distributions of seasonal mean PM10 mass concentrations (μg/m3) during the study period (the gray-shaded areas represent desert and desertified land in China)

Full size image
Fig. 6
figure 6

Relationships between annual and seasonal mean PM2.5 and PM10 concentrations for the cities with (black dots) or without winter heating (gray dots)

Full size image

CO (Fig. 7), NO2 (Fig. 8), and SO2 (Fig. 9) also exhibited the synchronously seasonal trends with PM2.5 mass concentrations, which reflected combined effects of weather conditions and primary emissions. Stagnant air with weak winds and low boundary layer height appeared frequently in winter, and thus, the air pollutants were trapped near the surface layer and local air pollution was aggravated (Zhao et al. 2015b). Furthermore, frequent air pollution episodes in winter were partly attributed to home coal combustion for heating (Zhang et al. 2009; Zhao et al. 2011). Furthermore, the difference of SO2 concentrations between summer and winter was more obvious than the two other gases pollutants in the Northern China, suggesting large impact of the domestic home heating on urban air quality. The domestic heating also is an important pollution source in winter and spring for the northern cities, which can lead to significant increases of some air pollutants such as SO2 and CO. The seasonal variation of O3 was opposite to the five other pollutants (PM2.5, PM10, CO, NO2, and SO2) with a minimum in the winter and a maximum in the summer (Fig. 10), which may be due to dependence of O3 formation rate on the intensity of solar radiation (Atkinson 2000).

Fig. 7
figure 7

Spatial distributions of seasonal mean CO concentrations (mg m−3) during the study period (the gray-shaded areas represent desert and desertified land in China)

Full size image
Fig. 8
figure 8

Spatial distributions of seasonal mean NO2 concentrations (μg/m3) during the study period (the gray-shaded areas represent desert and desertified land in China)

Full size image
Fig. 9
figure 9

Spatial distributions of seasonal mean SO2 concentrations (μg/m3) during the study period (the gray-shaded areas represent desert and desertified land in China)

Full size image
Fig. 10
figure 10

Spatial distributions of seasonal mean O3 concentrations (μg/m3) during the study period (the gray-shaded areas represent desert and desertified land in China)

Full size image

In the spatial scale, the air pollutants from primary emissions such as CO and SO2 (Figs. 7 and 9) in northern China were significantly higher than those in southern China, especially North China Plain (NCP). NCP was one of the most polluted areas in China due to more local primary emissions from coal-based industries and strong regional transportation between the cities. The study by Hu et al. (2014) indicated regional characteristics of air pollution were significant in North China Plain.

Conclusions

This study has reported the spatial and seasonal variations of PM2.5, PM10, CO, NO2, SO2, and O3 using air quality monitoring data observed in 31 Chinese provincial cities during April 2014 to March 2015, which will help to obtain a better understanding of the current air pollution situation in China and provide useful information regarding the status of nation-wide air pollution. The annual mean concentrations of PM2.5 and PM10 for all cities exceeded CAAQS Grade I standard. Only Fuzhou, Haikou, Kunming, and Lasa met Grade II standard for PM2.5 and PM10. Higher SO2 concentration was observed in northern cities and especially in winter, indicating large emissions from coal combustion and biomass burning during the heating period. Additionally, PM2.5-10 (particles with size between 2.5 and 10 μm) accounted for a large proportion of PM10 in northern China in spring (67 %), indicating the effect of dust events on urban air quality in northern China. The seasonal variation of O3 was opposite to other pollutants with the lowest concentrations in the winter and the highest in the summer.

With rapid increases of vehicle counts, urban expansion and economic development in the last two decades in China, severe air pollution problems have significantly negative effects on human health. Almost all the Chinese cities are suffering from a high concentration of primary or secondary pollutants. This suggests that China should conduct region-oriented air pollution management plans. This study only analyzed spatial and seasonal variations of six criteria pollutants in China. In order to better understand main factors influencing urban air quality, future study is necessary to investigate the associations between air quality and weather conditions and transportation of air pollutants in both intra- and inter-region.