field
stringclasses 1
value | paper_idx
stringclasses 111
values | doi
stringclasses 111
values | type
stringclasses 3
values | table_or_image
imagewidth (px) 333
2.14k
⌀ | text_or_caption
stringlengths 1
115k
|
|---|---|---|---|---|---|
atmosphere
|
0018
|
10.1016/j.scitotenv.2014.08.008
|
image
|
Fig. 6. Annual variation of (a) sulfate $({\mathsf{S}}0_{4}^{2-})$ (red dots) and sulfur dioxide $(S0_{2})$ (black dot), and (b) nitrate $\left(\mathsf{N O}_{3}^{-}\right)$ (red dot) and nitrogen oxide $\left(\mathrm{NO_{x}}\right)$ (black dot) in fall and winter from 2007 to 2011.
|
|
atmosphere
|
0018
|
10.1016/j.scitotenv.2014.08.008
|
image
|
Fig. 7. Annual variation of (a) elemental carbon (EC) and (b) organic carbon (OC) in fall and winter from 2007 to 2011.
|
|
atmosphere
|
0018
|
10.1016/j.scitotenv.2014.08.008
|
image
|
Fig. 8. Annual variation of (a) primary organic carbon (POC) and (b) secondary organic carbon (SOC) in fall and winter from 2007 to 2011.
|
|
atmosphere
|
0018
|
10.1016/j.scitotenv.2014.08.008
|
text
| Not supported with pagination yet
|
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
Xiaoxin Fu a,b, Xinming Wang a,⁎, Hai Guo b,⁎⁎, Kalam Cheung b, Xiang Ding a, Xiuying Zhao a, Quanfu He a, Bo Gao , Zhou Zhang , Tengyu Liu , Yanli Zhang
a State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou 510640, China b Air Quality Studies, Department Civil and Environmental Engineering, The Hong Kong Polytechnic University, Hong Kong
H I G H L I G H T S
• The annual reduction trend of $\mathrm{PM}_{2.5}$ was $8.58\,\upmu\mathrm{g}\,\mathsf{m}^{-3}$ in fall and winter of 2007 to 2011.
• The reduction rate of sulfate $(\mathsf{S O}_{4}^{2\,-})$ was $1.72\,\upmu\mathrm{g}\,\mathsf{m}^{-3}\,\mathsf{y r}^{-1}$ .
• Nitrate $(\mathsf{N O}^{3-})$ presented a growth trend with a rate of $0.79\,\upmu\mathrm{g}\,\mathsf{m}^{-3}\,\mathsf{y r}^{-1}$ .
a r t i c l e i n f o
Article history:
Received 21 April 2014
Received in revised form 2 August 2014
Accepted 3 August 2014
Available online xxxx
Editor: P. Kassomenos
Keywords:
$\mathrm{PM}_{2.5}$
Sulfate
Nitrate
Carbonaceous aerosols
Pearl River Delta
a b s t r a c t
In the fall and winter of 2007 to 2011, 167 24-h quartz filter-based fine particle $\left(\mathsf{P M}_{2.5}\right)$ samples were collected at a regional background site in the central Pearl River Delta. The $\mathrm{PM}_{2.5}$ showed an annual reduction trend with a rate of $8.58\,\upmu\mathrm{g}\;\mathrm{m}^{-3}$ $(p<0.01)$ . The OC component of the $\mathsf{P M}_{2.5}$ reduced by $1.10\,\upmu\mathrm{g}\,\mathsf{m}^{-3}\,\mathsf{y r}^{-1}$ $(p<0.01)$ , while the reduction rates of sulfur dioxide $\left({\mathsf{S O}}_{2}\right)$ and sulfate $({\mathsf{S}}{\mathsf{O}}_{4}^{2\,-})$ were $10.2\,\upmu\mathrm{g}\,\upmu^{-3}\,\mathrm{yr}^{-1}$ $p<0.01)$ and $1.72\,\upmu\mathrm{g}$ $\mathsf{m}^{-3}\,\mathsf{y r}^{-1}$ $p<0.01)$ ), respectively. In contrast, nitrogen oxides $\left(\mathrm{NO_{x}}\right)$ and nitrate $(\mathsf{N O}^{3-})$ presented growth trends with rates of $6.73\,\upmu\mathrm{g}\,\mathrm{m}^{-3}\,\mathrm{yr}^{-1}$ $p<0.05)$ and $0.79\,\upmu\mathrm{g}\,\mathsf{m}^{-3}\,\mathsf{y r}^{-1}$ $\!\!\!/<0.05\!\!\!/$ , respectively. The $\mathrm{PM}_{2.5}$ reduction was mainly related to the decrease of primary OC and $S0_{4}^{2-}$ , and the enhanced conversion efficiency of $S0_{2}$ to $S0_{4}^{2-}$ was related to an increase in the atmospheric oxidizing capacity and a decrease in aerosol acidity. The discrepancy between the annual trends of $\mathrm{NO_{x}}$ and $\mathrm{NO}_{3}^{-}$ was attributable to the small proportion of $\mathrm{NO}_{3}^{-}$ in the total nitrogen budget.
Capsule abstract: Understanding annual variations of $\mathrm{PM}_{2.5}$ and its chemical composition is crucial in enabling policymakers to formulate and implement control strategies on particulate pollution.
$\mathcal{Q}\,2014$ Elsevier B.V. All rights reserved.
1. Introduction
Many cities in China currently suffer severe air pollution problems, in particular haze caused by fine particles $\left(\mathsf{P M}_{2.5}\right)$ , resulting in visibility degradation and adverse health effects (Zhang et al., 2012a). Numerous heavy haze episodes have been observed in megacities such as Beijing, Shanghai, and Guangzhou in recent years (Wu et al., 2005; Sun et al., 2006; Fu et al., 2008; Chang et al., 2009). During these episodes, ambient 24-h average $\mathsf{P M}_{2.5}$ levels up to $175\,\upmu\mathrm{g}\,\mathfrak{m}^{-3}$ have been recorded, well over the World Health Organization (WHO) daily Air Quality Guidelines of $25\,\upmu\mathrm{g}\;\mathrm{m}^{-3}$ . High $\mathsf{P M}_{2.5}$ levels are closely associated with long- and short-term health problems (Tie et al., 2009; van Donkelaar et al.,
2010; Chen R.J. et al., 2012; Shang et al., 2013). In an attempt to reduce particulate pollution, the Chinese government has recently implemented new national ambient air quality standards, which for the first time include $\mathsf{P M}_{2.5}$ . Moreover, the government has emphasized the control of particulate pollution at a regional scale, with the main focus on the three economically relevant and densely populated city clusters; the North China Plain (NCP), the Yangtze River Delta (YRD) region, and the Pearl River Delta (PRD) region.
The PRD region in southern China makes up less than $0.5\%$ of China's total land area but contributes about $10\%$ of the nation's GDP, and is home to around $10\%$ of its population. The ambient annual mean $\mathsf{P M}_{2.5}$ level in this highly urbanized and industrialized region exceeded $100\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in 2004 (Andreae et al., 2008). However, in recent years, the number of hazy days recorded a large drop from over 120 days in 2005 to less than 60 days in 2011 (http://www.gzepb.gov.cn/). Despite this reduction, average annual $\mathsf{P M}_{2.5}$ levels in the PRD still exceed the daily and annual guidelines of the WHO. A systematic, long-term investigation into the variations in the main components of $\mathsf{P M}_{2.5}$ and its mass concentrations will provide important information on sources and formation mechanisms, which will be useful in the formulating and implementing of particulate pollution control measures in the region, and also of value to other Chinese city clusters.
Over the last decade, studies have been conducted at different locations in the region on $\mathsf{P M}_{2.5}$ mass concentrations and their major components, such as water soluble ions and carbonaceous aerosols (e.g. Lai et al., 2007; Hu et al., 2008; Tan et al., 2009a,2009b; Yang et al., 2011), and on the aerosols' light extinction and visibility impairment (Andreae et al., 2008; Jung et al., 2009; Tao et al., 2012). However, the measurements were mainly carried out over short periods. Thus, the long-term variations in $\mathsf{P M}_{2.5}$ mass concentrations and compositions were not determined. In this study, $\mathsf{P M}_{2.5}$ filter samples were systematically collected from a background site in the region in fall and winter from 2007 and 2011 so that the annual trends of the mass concentrations and chemical components of $\mathsf{P M}_{2.5}$ could be obtained.
2. Experimental
2.1. Field sampling
The PRD region has a typical Asian monsoon climate — hot and humid in the summer, with prevailing southwesterly monsoon winds from the sea, and relatively cool and dry in the fall and winter, when northeasterly monsoon winds from northern China dominate (Ding and Chan, 2005). The region is often under the influence of high pressure ridges in the fall and winter, causing long periods of sunny days, with a low boundary layer and a high frequency of inversion. This stable meteorological condition facilitates the accumulation of pollutants and a resulting deterioration of air quality. As a result, high levels of air pollutants usually occur in fall and winter (Simpson et al., 2006; Fan et al., 2008; Liu et al., 2008; Cheng et al., 2010). Field measurements were thus collected in those two seasons each year.
The sampling site, Wanqingsha (WQS: $22.42^{\circ}\mathrm{~N},$ $113.32^{\circ}\mathrm{~E~}$ ), was located in a small town south of Guangzhou, in the center of the PRD (Fig. 1). The town was surrounded by farmland, has little traffic, and very few textile and clothing workshops. The local anthropogenic emissions were thus not significant, with most air pollutants originating from the surrounding cities. The site was $50\,\mathrm{km}$ southeast of Guangzhou center, $40\;\mathrm{km}$ southwest of Dongguan, $50\;\mathrm{km}$ northwest of Shenzhen, and $25\;\mathrm{km}$ northeast of Zhongshan, making it a good regional station to characterize the air pollution of the inner PRD (Guo et al., 2009). The $\mathsf{P M}_{2.5}$ high-volume samplers (Tisch Environmental Inc., USA) were placed on the rooftop of a building, about $30\,\mathrm{m}$ above the ground. Gas-phase pollutants, including $S0_{2}$ and $\mathsf{N O}_{\mathrm{x}},$ were also monitored.
The 24-h $\mathsf{P M}_{2.5}$ samples were collected by drawing air through an $8\times10$ inch quartz filter (QMA, Whatman, UK) at a rate of 1.1 $\mathrm{m}^{3}\,\mathrm{min}^{-1}$ . The filters were pre-baked at $450~^{\circ}\mathrm{C}$ for $4\;\mathrm{h}$ , wrapped in aluminum foil, zipped in Teflon bags, and stored at $-20\,^{\circ}\mathrm{C}$ before sampling. They were again stored in this way after sample collection. In 2007, 2008, 2009, 2010, and 2011, 32, 29, 25, 53, and 28 samples were collected, respectively. The meteorological parameters were measured by a mini weather station (Vantage Pro2TM, Davis Instruments Corp., USA) with wind speed/direction, relative humidity (RH), and temperature recorded every minute.
2.2. Chemical analysis
The $\mathsf{P M}_{2.5}$ filters were weighed before and after field sampling, after 24-h equilibrium, at a temperature of $20{-}23\ ^{\circ}\mathrm{C}$ and with a RH between 35 and $45\%$ The organic carbon (OC) and elemental carbon (EC) in the $\mathsf{P M}_{2.5}$ were determined by the thermo-optical transmittance (TOT) method (NIOSH, 1999) using an OC/EC analyzer (Sunset Laboratory Inc., USA), with a punch $(1.5\times1.0\,\mathrm{cm})$ of the sampled filters. For the water-soluble inorganic ions, a punch $\mathrm{'}5.06\,\mathrm{cm}^{2})$ of the filters was extracted twice with $10\;\mathrm{ml}$ ultrapure Milli-Q water $\left(18.2\:\mathrm{M}\Omega^{\cdot}\mathrm{cm}/25\ ^{\circ}\mathrm{C}\right)$ each for $15\;\mathrm{min}$ using an ultrasonic ice-water bath. The total water extracts $(20\ \mathrm{ml})$ ) were filtered through a $0.22~\upmu\mathrm{m}$ pore size filter and then stored in a pre-cleaned HDPE bottle. The cations (i.e. $\mathtt{N a}^{+}$ , $\mathrm{NH_{4}^{+}}$ , $\mathsf{K}^{+}$ , ${\mathrm{Mg}}^{2+}$ , and ${\mathsf{C}}{\mathsf{a}}^{2+}$ ) and anions (i.e. $\mathsf{C l}^{-}$ , $\mathtt{N O}_{3}^{-}$ , and $S0_{4}^{2-}$ ) were analyzed with an ion-chromatography system (Metrohm, 883 Basic IC plus). Cations were measured using a Metrohm Metrosep C4-100 column with $2\,\mathrm{mmol}\,\mathrm{L}^{-1}$ sulfuric acid as the eluent. Anions were measured using a Metrohm Metrosep A sup5-150 column equipped with a suppressor. The anion eluent was a solution of $3.2\;\mathrm{mmol}\;\mathrm{L}^{-1}\;\mathrm{Na}_{2}\mathrm{CO}_{3}$ and $1.0\;\mathrm{mmol}\;\mathrm{L}^{-1}\;\mathrm{NaHCO}_{3}$ .
2.3. Quality assurance/quality control (QA/QC)
Field and laboratory blank samples were analyzed in the same way as field samples. All the OC/EC and cation/anion data were corrected using the field blanks. The method detection limits (MDLs) were $0.01-$ $0.05\,\upmu\mathrm{g}\,\mathfrak{m}^{-3}$ for the OC, EC, cations, and anions. Ions balance was used as a quality control check in the cation/anion analysis. Nano-equivalents of cations and anions were calculated using their mass concentrations and molecular weights:
Cation nano‐equivalents CE
$$
\begin{array}{r l}{\centering}&{=\left(\mathrm{N}\mathsf{a}^{+}/23+\mathrm{N}\mathsf{H}_{4}^{\;\;+}/18+\mathrm{K}^{+}/39+\mathrm{M}\mathsf{g}^{2+}/24\times2+\mathsf{C}\mathsf{a}^{2+}/40\times2\right)}\\ &{\;\;\times\;1000}\end{array}
$$
Anion nano‐equivalentsðAEÞ
$$
=\left(\mathrm{Cl}^{-}/35.5+\mathrm{NO_{3}}^{-}/62+\mathrm{SO_{4}}^{2-}/96\times2\right)\times1000
$$
A significant linear correlation $\mathrm{R}^{2}=0.984\$ was observed between CE and AE (Fig. 2) with a slope of 1.14 for all $\mathsf{P M}_{2.5}$ samples. This slope was close to identity and indicated that all the significant ions were resolved. The AE/CE slope was slightly higher than 1.0, suggesting that the aerosols in WQS tended to be acidic (Seinfeld and Pandis, 2006).
3. Results and discussion
3.1. $P M_{2.5}$ mass concentrations
The 24-h average $\mathsf{P M}_{2.5}$ concentration in the fall and winter of 2007– 2011 ranged from 22.3 (December 2010) to $191~\upmu\mathrm{g}~\mathrm{m}^{-3}$ (November 2010) with an average of $95.2\pm4.49\,\upmu\mathrm{g}\;\mathrm{m}^{-3}$ (average $\pm\ 95\%$ Confidence Interval). Table 1 shows that the $\mathsf{P M}_{2.5}$ level significantly decreased from $112.5\pm8.2\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ in 2007 to $78.6\pm7.6\,\upmu\mathrm{g}\;\mathrm{m}^{-3}$ in 2011 $\.p<0.01)$ , with a slope of $-8.58\;{\upmu\mathrm{g}}\;{\mathrm{m}}^{-3}\;{\mathrm{yr}}^{-1}$ , or an average reduction rate of $8.6\%\,\mathrm{yr}^{-1}$ (Fig. 3). This reflected the efficient reduction of $\mathsf{P M}_{2.5}$ pollution in these years. The Guangdong government implemented various control measures, such as the increased use of nuclear and hydroelectric power; the phasing out of small coal-fired power generation units; prohibiting the building of new cement plants, ceramics factories, and glassworks; the establishment of stricter emission standards for industrial boilers, and improvements in the quality of vehicle fuel (http://www.gzepb.gov.cn/). The decreasing trend of $\mathsf{P M}_{2.5}$ is consistent with the yearly $\mathsf{P M}_{10}$ variations measured in the region. The 24-h average $\mathsf{P M}_{10}$ was measured at the same site by the Guangdong Environmental Monitoring center during fall and winter from 2007 to 2011, and fell from $147\,\upmu\mathrm{g}\,\mathfrak{m}^{-3}$ in 2007 to $91\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ in 2011, with an average reduction rate of $11.8~{\upmu\mathrm{g}}~\mathrm{m}^{-3}~\mathrm{yr}^{-1}$ or $-\,10.3\%{\mathrm{~yr}}^{-\,1}$ (http://www.epd. gov.hk/epd/english/resources_pub/publications/m_report.html). Comparable or higher $\mathsf{P M}_{2.5}$ concentrations were observed at urban sites in the same region. For instance, Tan et al. (2009a) found that 24-hr average $\mathsf{P M}_{2.5}$ concentration was $171.7~\upmu\mathrm{g}\,\:\mathrm{m}^{-3}$ in January 2008, Yang et al. (2011) recorded daily average $\mathsf{P M}_{2.5}$ level of $81.7\pm25.6\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ (average $\pm$ standard deviation) in December 2008 to February 2009, and Tao et al. (2012) reported 24-h average of $103.3\pm50.1\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ in January 2010.
The $\mathsf{P M}_{2.5}$ values in the PRD region were, however, much higher than those observed in central California (daily average: $13.5~\upmu\mathrm{g}\:\:\mathrm{m}^{-3}.$ ) (Rinehart et al., 2006), in Spain (daily average: $9.0~\upmu\mathrm{g}\mathrm{~m}^{-3}$ ), and in Germany (daily average: $10\,\upmu\mathrm{g}\;\mathrm{m}^{-3}$ ) (Cusack et al., 2012). In contrast, emission estimate studies conducted in the PRD region found the opposite change in $\mathsf{P M}_{2.5}$ emissions. Zheng et al. (2009, 2012a) reported that the $\mathsf{P M}_{2.5}$ emission was $205\,\mathrm{Gg}$ in 2006 and $303\,\mathrm{Gg}$ in 2009, for example.
Among all of the $\mathsf{P M}_{2.5}$ samples, only one was below the WHO 24-h guideline level of $25\,\upmu\mathrm{g}\,\mathfrak{m}^{-3}$ and three were below the US EPA 24-h standard of $35\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , and $75\%$ of the samples were above the Chinese daily standard of $75\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ (Fig. 4) (GB 3095-2012, http://kjs.mep.gov. cn/hjbhbz/bzwb/dqhjbh/dqhjzlbz/201203/t20120302_224165.htm). The maximum concentration of $191\,\upmu\mathrm{g}\,\mathfrak{m}^{-3}$ was on November 6, 2010, with the rehearsal of the large-scale firework display for the opening ceremony of the 16th Asia games. Elevated $\mathsf{P M}_{2.5}$ levels were also recorded during the opening ceremony of the 10th Asian Games for the Disabled $!163.9\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ on December 12, 2010), and on the day after the closing ceremony $^{?174.9\,\upmu\mathrm{g}\,\mathrm{m}^{-3}}$ on December 20, 2010), reflecting the significant effect of burning fireworks. Indeed, Wang et al. (2007) stated that during the Chinese Lantern Festival in Beijing, when many fireworks were set off, $S0_{4}^{2-}$ and $\mathrm{NO}_{3}^{-}$ levels were over five times higher than normal. The lowest $\mathsf{P M}_{2.5}$ concentration $(22.3\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ ) occurred on December 15, 2010, when a strong intrusion of cold air masses from the north caused a sudden temperature drop, and air pollutants were swept south out of the region. In recent years, ambient fine particle concentrations have significantly reduced in the PRD region, but further efforts are necessary to reduce $\mathsf{P M}_{2.5}$ emissions.
3.2. Chemical compositions of $P M_{2.5}$
The 24-h average concentrations of carbonaceous aerosols and water soluble ions in $\mathsf{P M}_{2.5},$ the ratios of OC/EC, $\mathrm{NH}_{4}^{+}/\mathrm{SO}_{4}^{2-}$ , and $\mathsf{N O}_{3}^{-}/$ $\mathsf{S0}_{4}^{2}$ , and the meteorological conditions over the five year period are listed in Table 1. The chemical compositions of $\mathsf{P M}_{2.5}$ in the same period are shown in Fig. 5. In the figure, the aerosol organic matter (OM) equals $2\times0C$ (Wang et al., 2012a). It was found that OM was the most abundant component over this period (Fig. 5). From Table 1, it can be seen that the average OC concentration was highest in 2008 $(22.7~\pm$ $2.93\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ ; average $\pm\ 95\%$ CI) and lowest in 2011 ( $15.2\pm2.06\,\upmu\mathrm{g}$ $\mathfrak{m}^{-3}$ ). For EC, the average concentration was highest in 2009 $^{5.5\pm}$ $0.90\,\upmu\mathrm{g}\,\mathrm{m}^{-3}.$ ) and lowest in 2011 $(3.1\pm0.38\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ ). These 24-h average $\mathsf{P M}_{2.5}$ component levels approximated those recorded in the winter in urban Guangzhou, i.e. daily average $26.8\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ for OC and $6.2~\upmu\mathrm{g}\,\textrm{m}^{-3}$ for EC in January 2008 (Tan et al., 2009a), $17.5\,\pm$ $7.6\,\upmu\mathrm{g}\mathrm{~m}^{-3}$ (average $\pm\ S\mathrm{D}$ ) for OC and $4.1\pm2.0\ensuremath{\,\upmu\mathrm{g\,m}^{-3}}$ for EC in the winter of 2008–2009 (Yang et al., 2011), and $11.8~\pm$ $7.3\ \upmu\mathrm{g}\mathrm{~m~}^{-3}$ for OC and $7.8\pm4.3\mathrm{\}\upmu\mathrm{g}\mathrm{\m}^{-3}$ for EC in January 2010 (Tao et al., 2012). However, the OC and EC concentrations measured in this study were much higher than those observed in urban Paris in 2009–2010 (24-h average OC: $3.0\,\pm\,1.7~\upmu\mathrm{g}\,\,\mathrm{m}^{-3}$ and EC: $1.4\pm0.7~{\upmu\mathrm{g}}~\mathrm{m}^{-3}$ (average $\pm\ S\mathrm{D}$ )) (Bressi et al., 2013), and in both residential and commercial areas of Incheon, Korea, (24-h average OC: $10.9\pm0.8\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ and EC: $1.8\pm0.1\mathrm{\}\upmu\mathrm{g}\mathrm{\m}^{-3}.$ in the winter of 2009–2010 (Choi et al., 2012).
Daily average concentrations of $S0_{4}^{2-}$ ranged from $22.7\pm2.3\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ (average $\pm\,95\%\,\mathrm{CI})$ in 2007 to $14.2\pm1.8\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in 2011, while the 24-h average $\mathrm{NO}_{3}^{-}$ concentrations increased from $6.7\pm1.1\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in 2007 to a peak of $11.5\,\pm\,1.9\,\upmu\mathrm{g}\,\,\mathrm{m}^{-3}$ in 2009, and then decreased to $9.6~\pm$ $1.5\,\upmu\mathrm{g}\mathrm{~m~}^{-3}$ in 2011. The $\mathsf{N H}_{4}^{+}$ concentrations did not show a significant change over the five year period. As with $\mathsf{P M}_{2.5}$ , the fireworks of November 6, 2010 resulted in $S0_{4}^{2}$ , $\mathtt{N O}_{3}^{-}$ , and $\mathsf{N H}_{4}^{+}$ concentrations reaching their maxima, with 24-h average levels of 40.2, 41.4, and $24.4\,\upmu\mathrm{g}\mathrm{~m}^{-3}$ , respectively. High $S0_{4}^{2-}$ , $\mathtt{N O}_{3}^{-}$ , and $\mathrm{NH_{4}^{+}}$ values have been recorded on hazy days in various Chinese megacities. For example, at an urban site in Beijing, $24\mathrm{-h}$ average levels reached 24.8, 49.3, and $26.2\,\upmu\mathrm{g}\,\mathsf{m}^{-3}$ , respectively in October 2010, and 28.11, 42.46 and
$18.32\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ in October 2011 (Sun et al., 2013). In Shanghai, 22-h average levels of 28.7, 32.9, and $19.3~\upmu\mathrm{g}\,\:\mathrm{m}^{-3}$ were recorded in May–June 2009 (Du et al., 2011). By contrast, recorded concentrations of $S0_{4}^{2-}$ , $\mathtt{N O}_{3}^{-}$ , and $\mathrm{NH_{4}^{+}}$ were much lower in US and European cities. The 24-h average concentrations in the southeastern US were over five times lower than those found in WQS (Chen Y. et al., 2012), and in Spain in 2002–2010 daily average levels as low as 2.4, 1.0 and $1.0\,\upmu\mathrm{g}\,\mathfrak{m}^{-3}$ were recorded (Cusack et al., 2012).
The $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ ratio could indicate the contribution of mobile and stationary sources to sulfur and nitrogen in the atmosphere (Arimoto et al., 1996). The mass ratio of $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ rose from $0.31\pm0.06$ (aver$\mathsf{a g e}\pm95\%\,\mathrm{CI})$ in 2007 to $0.58\pm0.10$ in 2008, and reached $0.69\pm0.11$ during 2009–2011. A previous study reported a $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ mass ratio of 2–5 in Los Angeles, and in Rubidoux in southern California, where very little coal burning occurred (Kim et al., 2000). The $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ mass ratios in this study increased from 2007 to 2011, but they were all less than 1.0, and therefore much lower than those of Los Angeles and Rubidoux, indicating the effect of stationary sources (coal combustion) in the PRD region (Yao et al., 2002; Wang et al., 2005; Cao et al., 2009). The mole ratio of $[\mathsf{N H}_{4}^{+}]$ to $(2\times[50_{4}^{2-}]+[\N0_{3}^{-}])$ increased from $0.64\pm0.04$ in 2007 to $0.80\pm0.02$ in 2011, suggesting that aerosol acidity decreased over the five year period.
3.3. Annual trends of major components in $P M_{2.5}$
3.3.1. Sulfate $(S O_{4}^{2}{}^{-})$
Fig. 6(a) shows that on average, $S0_{4}^{2-}$ decreased at a rate of $1.72\,\upmu\mathrm{g}$ $\mathfrak{m}^{-3}\,\mathrm{yr}^{-1}$ or $11.0\%\,\mathrm{yr}^{-1}$ $p<0.01\$ ), whereas for $S0_{2}$ the reduction was $10.2\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ or $18.8\%$ a year $g<0.01$ ). $S0_{2}$ concentrations thus decreased much more rapidly than $S0_{4}^{2-}$ . Our data showed that each $1\%$ reduction in $S0_{2}$ concentration resulted in a $0.59\%$ (i.e. $11.0\%$ divided by $18.8\%$ ) decrease in $S0_{4}^{2-}$ concentration in the PRD region (i.e. a $1\,\upmu\mathrm{g}\,\mathsf{m}^{-3}$ change in $S0_{2}$ caused a $0.17\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ change in $S0_{4}^{2-}$ ). The decreasing trends of $S0_{2}$ and $S0_{4}^{2-}$ found are in line with previous studies. Based on satellite retrieval data, Zhang et al. (2012b) found the yearly average tropospheric $S0_{2}$ vertical columns in the PRD region decreased from $0.223\pm\:0.135$ DU (average $\pm\,S\mathbf{D}\quad$ ) in 2006 to $0.144\pm\:0.064$ DU in 2009 with a reduction rate of $11.8\%\,\mathrm{yr}^{-1}$ , while Lu et al. (2013) reported that normalized $S0_{2}$ emissions significantly decreased between 2007 and 2009, at a rate of $15.4\%\;\mathrm{yr}^{-1}$ . Previous studies have also reported the relationship between decreased concentrations of $S0_{2}$ and $S0_{4}^{2-}$ . Holland et al. (1999) found that $S0_{2}$ concentrations decreased by $35\%$ and $S0_{4}^{2-}$ concentrations by $26\%$ in the eastern US from 1989 to 1995. In Finland, France, and Germany, observed $S0_{4}^{2-}$ concentrations decreased by $85{-}70\%$ as $S0_{2}$ concentrations decreased by $85{-}90\%$ , between 1980 and 2000 (Lovblad et al., 2004). Manktelow et al. (2007) used a global model to investigate changes in the regional sulfur budget from 1985 to 2000. Their findings were similar to ours, and for every $1\%$ decrease in $S0_{2}$ surface concentration, $S0_{4}^{2-}$ surface concentration decreased by $0.55\%$ across Western Europe, and by $0.58\%$ across the US. The different response was due to the fact that conversion efficiency of $S0_{2}$ to $S0_{4}^{2-}$ in clouds increased when $S0_{2}$ emissions decreased. The much higher reduction rate of $S0_{2}$ found in the PRD region implied that the control measures of the time were effective. The main source of $S0_{2}$ in China was coal-fired power plants (Zhao et al., 2008; Lu et al., 2010), and after the installation and operation of flue gas desulfurization (FGD) systems in thermal power units and the closure of small and less-efficient power plants, the total industrial $S0_{2}$ emission in
Guangdong dropped from $1203\,\mathrm{Gg}$ in 2007 to $848\,\mathrm{Gg}$ in 2011, with a decreasing rate of $7.4\%\;\mathrm{yr}^{-1}$ (GPBS, 2008, 2009, 2010, 2011, 2012). The faster rate of decrease was also related to the atmospheric chemistry of sulfur. $S0_{4}^{2-}$ is produced from the dry oxidation between $S0_{2}$ and the OH radical, and/or from the oxidation of $S0_{2}$ by $\mathrm{H}_{2}\mathrm{O}_{2}$ and $0_{3}$ through in-cloud processes. $\mathrm{H}_{2}\mathrm{O}_{2}$ is the most dominant oxidant of $S0_{2}$ in atmospheric aqueous phases, particularly when the ${\mathsf p}H$ is lower than 5 (Calvert et al., 1985). In the PRD region, $\mathrm{H}_{2}\mathrm{O}_{2}$ was significant in the formation of sulfate in the aerosol phase (Hua et al., 2008). The intensity of solar radiation is a significant factor, as it controls the atmospheric oxidizing capacity (Merkel et al., 2011; Wang et al., 2012b). Furthermore, $\mathrm{H}_{2}\mathrm{O}_{2}$ positively correlates with solar radiation (Acker et al., 2008; Marinoni et al., 2011). In recent years, $\mathsf{P M}_{2.5}$ concentrations have significantly decreased in the PRD, resulting in enhanced solar radiation and actinic flux in the troposphere. Hence, the conversion efficiency of $S0_{2}$ to $S0_{4}^{2-}$ in clouds over the region is even more rapid. The equilibriums of $S0_{2}$ dissolving, which lead to the formation of bisulfite and sulfite ions in the presence of particle phase, are sensitive to the pH value. The aerosol acidity (mole ratio of $[\mathrm{NH}_{4}^{+}]$ to $(2\times[50_{4}^{2-}]+[\N0_{3}^{-}]))$ in the five year period decreased by $25\%$ in 2011, compared to the ratio in 2007, where the solubility of $S0_{2}$ was enhanced and certain oxidation processes were accelerated (Jones and Harrison, 2011). In conclusion, the rapid reduction of $S0_{2}$ was caused by the decrease in the source emissions and by the enhanced conversion efficiency of $S0_{2}$ to $S0_{4}^{2-}$ through in-cloud processes, due to the increased oxidizing capacity and the drop in aerosol acidity in this period. Consequently, the combined effect of these factors led to the slow decreasing trend of $S0_{4}^{2-}$ in the region.
3.3.2. Nitrate $\left(N O_{3}^{-}\right)$
The observed $\mathtt{N O}_{3}^{-}$ levels increased at a rate of $0.79\,\upmu\mathrm{g}\,\mathsf{m}^{-3}\,\mathsf{y r}^{-1}$ or $9.5\%\,\mathrm{yr}^{-1}$ $p<0.05)$ , and $\Nu0_{\mathrm{x}}$ on average increased by $6.73\,\upmu\mathrm{g}\;\mathrm{m}^{-3}$ or
$9.8\%$ every year $(p<0.05)$ (Fig. 6(b)). The $\Nu0_{\mathrm{x}}$ concentrations increased more rapidly than those of the $\mathtt{N O}_{3}^{-}$ . Specifically, every $1\%$ increase in $\Nu0_{\mathrm{x}}$ concentration resulted in a $0.97\%$ increase in $\mathrm{NO}_{3}^{-}$ concentration in the PRD region.
It is well known that power plants, factories, and vehicles were major contributors of $\Nu0_{\mathrm{x}}$ emissions in China (Streets et al., 2003; Ohara et al., 2007; Gu et al., 2012). Electricity production in the PRD region grew at a rate of $12.7\%\,\mathrm{yr}^{-1}$ during 2007–2011 (GPBS, 2008, 2009, 2010, 2011, 2012), which led to an increase in $\Nu0_{\mathrm{x}}$ emission from $392\,{\mathrm{Gg}}$ in 2005 to $586\,{\mathrm{Gg}}$ in 2010; an increase rate of $9.9\%\,\mathrm{yr}^{-1}$ (Zhao et al., 2008). The number of vehicles in Guangdong increased from 5.07 million in 2007 to 9.12 million in 2011, a striking growth rate of $20\%\,\mathrm{yr}^{-1}$ (GPBS, 2008, 2009, 2010, 2011, 2012), which also contributed to the $\Nu0_{\mathrm{x}}$ emission increase. Power plants in Guangdong, however, were obliged to use low- $\cdot\mathrm{NO}_{\mathrm{x}}$ burner technologies and denitrification facilities after the implementation of emission standards for coal-fired power plants in 2009. Thus, the effort to control $\Nu0_{\mathrm{x}}$ emission from coal-fired power plants in the PRD region over the study period was counteracted by the rapid growth in power generation and in motor vehicle numbers.
Combustion sources emit $\mathrm{NO}_{\mathrm{x}},$ and involve a series of chemical reactions producing organic and inorganic nitrate compounds, including $\mathrm{NO}_{3}^{-}$ . The nitrogen chemistry in the atmosphere results in both $\mathrm{NO}_{3}^{-}$ and $\Nu0_{\mathrm{x}}$ generating organic nitrates (i.e. ${\tt R O N O}_{2}.$ ), peroxyacetyl nitrate (PAN), $\mathrm{HNO}_{3}$ (gas), nitrous acid (HONO), and reactive intermediates, which are difficult to detect but are extremely important for the nitrogen budget (Atkinson, 2000). The total level of $C_{1}..C_{5}$ alkyl nitrates $\left({\mathrm{RONO}}_{2}\right)$ reached about $0.35\,\upmu\mathrm{g}\,\mathsf{m}^{-3}$ at a coastal site of Hong Kong in November 2002 (Simpson et al., 2006), while the highest concentration of PAN in the PRD was $19.3\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in the summer of 2006, equal to the level of $\mathrm{NO}_{3}^{-}$ found in this study (Wang et al., 2010). The average concentrations of $\mathsf{H N O}_{3}$ and HONO in the PRD region in October–November 2004 were 6.3 and $2.9\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , respectively (Hu et al., 2008). In general, $\mathrm{NO}_{3}^{-}$ only accounted for a small proportion of $\Nu0_{\mathrm{x}}$ products, which is why this study found that the $\Nu0_{\mathrm{x}}$ concentrations increased more rapidly than $\mathrm{NO}_{3}^{-}$ .
In summary, the increase in $\Nu0_{\mathrm{x}}$ emissions from coal-fired power plants and vehicles in recent years suggests that future $\Nu0_{\mathrm{x}}$ reduction in the region will be a major challenge. As the precursor of ozone in the troposphere, $\Nu0_{\mathrm{x}}$ increase leads to an alteration in atmospheric oxidizing capacity, and subsequently affects the formation of secondary components of $\mathsf{P M}_{2.5}$ .
3.4. Elemental carbon (EC) and Organic carbon (OC)
Fig. 7(a) shows there was no clear decreasing trend in EC over this time $.p=0.06]$ ), perhaps due to the combined effect of residential, industrial, and vehicular emissions. The main EC sources in the PRD were residential and industrial emissions, transportation, and biomass burning (Cao et al., 2006; Lei et al., 2011; Qin and Xie, 2012). During
2007–2011, the total annual residential coal usage decreased, whereas the consumption of liquefied petroleum gas and household electricity increased. Moreover, industrial EC emission reduced from $27.3\:\mathrm{Gg}$ in 2007 to $26.4\:\mathrm{Gg}$ in 2011, with an annual reduction rate of $0.8\%$ (GPBS, 2008, 2009, 2010, 2011, 2012). In contrast, the rapid increase in vehicle numbers in the region increased EC emissions, offsetting the industrial and residential decrease.
A higher decreasing rate of OC (i.e. $1.10\,\upmu\mathrm{g}\,\mathsf{m}^{-3}\,\mathsf{y r}^{-1}$ or $5.9\%\,\mathrm{yr}^{-1}$ ) $(p<0.01)$ was found in this period (Fig. 7(b)). OC is composed of primary OC (POC) and secondary OC (SOC). The SOC was estimated using the EC-tracer method (Turpin and Huntzicker, 1995), and the POC was the difference between OC and SOC. Fig. 8(a) and (b) show that POC levels decreased at a rate of $0.74\,\upmu\mathrm{g}\,\mathsf{m}^{-3}\,\mathsf{y r}^{-1}$ ( $p<0.01\$ , whereas SOC did not show a significant decreasing trend $p=0.171$ ). The average proportion of POC and SOC in OC was $60.9\%$ and $39.2\%$ , respectively. Hence, POC was the major component of OC, and the OC reduction was mainly attributed to the decrease in POC emissions. The unchanged SOC levels during the study period might indicate a potential impediment to further $\mathsf{P M}_{2.5}$ reduction in the region.
4. Conclusions
$\mathsf{P M}_{2.5}$ mass concentrations and its chemical components were measured at a site in the central PRD region in fall and winter from 2007 to 2011. There was a significant annual reduction rate of $\mathsf{P M}_{2.5}$ of $8.58~\upmu\mathrm{g}\,\textrm{m}^{-3}\,\textrm{y r}^{-1}$ . In $\mathsf{P M}_{2.5}$ , OC and $S0_{4}^{2-}$ decreased $1.10\,\upmu\mathrm{g}\;\mathrm{m}^{-3}$ and $1.72\;\upmu\mathrm{g}\;\mathsf{m}^{-3}\,\mathsf{y r}^{-1}$ , respectively. By contrast, $\mathtt{N O}_{3}^{-}$ displayed an increasing rate of $0.79\,\upmu\mathrm{g}\,\,\mathrm{m}^{-3}\,\mathrm{yr}^{-1}$ . In general, $\mathsf{P M}_{2.5}$ reduction in the PRD region was mainly due to the reduction of OM and $S0_{4}^{2-}$ . $S0_{2}$ had a decreasing rate of $10.2~{\upmu\mathrm{g}}~{\mathrm{m}}^{-3}~{\mathrm{yr}}^{-1}$ , while $\Nu0_{\mathrm{x}}$ presented a growth rate of $6.73\ensuremath{~\upmu\mathrm{g}~\mathrm{m}^{-3}\,\mathrm{yr}^{-1}}$ . The precursors $S0_{2}$ and $\Nu0_{\mathrm{x}}$ concentrations obviously decreased and increased more rapidly than $S0_{4}^{2-}$ and $\mathrm{NO}_{3}^{-}$ . The faster reduction of $S0_{2}$ than $S0_{4}^{2-}$ was associated with the combined influence of decreased source emissions, increased oxidizing capacity with cloud processes, and reduced aerosol acidity. In contrast, the more rapid increase in $\Nu0_{\mathrm{x}}$ concentration than that of $\mathrm{NO}_{3}^{-}$ was likely due to increased power generation and vehicle numbers, which offset efforts to control coal-fired power plants, and $\Nu0_{\mathrm{x}}$ was converted into $\mathrm{NO}_{3}^{-}$ and other nitrogen compounds. Although air pollution caused by $\mathsf{P M}_{2.5}$ has been reduced in the PRD region in recent years, the reduction of fine particle emissions, particularly $\mathrm{NO}_{3}^{-}$ and SOC, will be extremely challenging in the future.
Acknowledgments
This study was supported by the National Natural Science Foundation of China (Project No. 41025012), the Strategic Priority Research Program of the Chinese Academy of Sciences (Grant No. XDB05010200), the Research Grants Council of the Hong Kong government (PolyU5154/13E), and the joint supervision scheme of the Hong Kong Polytechnic University (G-UB67).
References
Acker K, Kezele N, Klasinc L, Moller D, Pehnec G, Sorgo G, et al. Atmospheric $\mathrm{H}_{2}\mathrm{O}_{2}$ measurement and modeling campaign during summer 2004 in Zagreb, Croatia. Atmos Environ 2008;42:2530–42.
Andreae MO, Schmid O, Yang H, Chand D, Yu JZ, Zeng LM, et al. Optical properties and chemical composition of the atmospheric aerosol in urban Guangzhou, China. Atmos Environ 2008;42:6335–50.
Arimoto R, Duce RA, Savoie DL, Prospero JM, Talbot R, Cullen JD, et al. Relationships among aerosol constituents from Asia and the North Pacific during PEM-West A. J Geophys Res - Atmos 1996;101:2011–23.
Atkinson R. Atmospheric chemistry of VOCs and $\mathrm{NO}_{\mathrm{x}}.$ Atmos Environ 2000;34:2063–101.
Bressi M, Sciare J, Ghersi V, Bonnaire N, Nicolas JB, Petit J-E, et al. A one-year comprehensive chemical characterisation of fine aerosol $\left(\mathrm{PM}_{2.5}\right)$ at urban, suburban and rural background sites in the region of Paris (France). Atmos Chem Phys 2013;13:7825–44.
Calvert JG, Lazrus A, Kok GL, Heikes BG, Walega JG, Lind J, et al. Chemical mechanisms of acid generation in the troposphere. Nature 1985;317:27–35.
Cao GL, Zhang XY, Zheng FC. Inventory of black carbon and organic carbon emissions from China. Atmos Environ 2006;40:6516–27.
Cao JJ, Shen ZX, Chow JC, Qi GW, Watson JG. Seasonal variations and sources of mass and chemical composition for $\mathrm{PM}_{10}$ aerosol in Hangzhou, China. Particuology 2009;7: 161–8.
Chang D, Song Y, Liu B. Visibility trends in six megacities in China 1973–2007. Atmos Res 2009;94:161–7.
Chen RJ, Kan HD, Chen BH, Huang W, Bai ZP, Song GX, et al. Association of particulate air pollution with daily mortality: the China Air Pollution and Health Effects Study. Am J Epidemiol 2012;175:1173–81.
Chen Y, Zheng M, Edgerton ES, Ke L, Sheng G, Fu J. PM2.5 source apportionment in the southeastern U.S.: Spatial and seasonal variations during 2001–2005. J Geophys Res - Atmos 2012;117.
Cheng HR, Guo H, Saunders SM, Lam SHM, Jiang F, Wang XM, et al. Assessing photochemical ozone formation in the Pearl River Delta using a photochemical trajectory model. Atmos Environ 2010;44:4199–208.
Choi JK, Heo JB, Ban SJ, Yi SM, Zoh KD. Chemical characteristics of $\mathrm{PM}_{2.5}$ aerosol in Incheon, Korea. Atmos Environ 2012;60:583–92.
Cusack M, Alastuey A, Perez N, Pey J, Querol X. Trends of particulate matter $\left(\mathrm{PM}_{2.5}\right)$ and chemical composition at a regional background site in the Western Mediterranean over the last nine years (2002–2010). Atmospheric Chemistry and Physics 2012;12: 8341–57.
Ding YH, Chan JCL. The East Asian summer monsoon: an overview. Meteorol Atmos Phys 2005;89:117–42.
Du H, Kong L, Cheng T, Chen J, Du J, Li L, et al. Insights into summertime haze pollution events over Shanghai based on online water-soluble ionic composition of aerosols. Atmos Environ 2011;45:5131–7.
Fan SJ, Wang BM, Tesche M, Engelmann R, Althausen A, Liu J, et al. Meteorological conditions and structures of atmospheric boundary layer in October 2004 over Pearl River Delta area. Atmos Environ 2008;42:6174–86.
Fu Q, Zhuang GS, Wang J, Xu C, Huang K, Li J, et al. Mechanism of formation of the heaviest pollution episode ever recorded in the Yangtze River Delta, China. Atmos Environ 2008;42:2023–36.
Gu B, Ge Y, Ren Y, Xu B, Luo WD, Jiang H, et al. Atmospheric reactive nitrogen in China: sources, recent trends, and damage costs. Environ Sci Technol 2012;46:9420–7.
Guangdong Provincial Bureau of Statistics (GPBS). Guangdong Statistical Yearbook 2008. Beijing: China Statistics Press; 2008.
Guangdong Provincial Bureau of Statistics (GPBS). Guangdong Statistical Yearbook 2009. Beijing: China Statistics Press; 2009.
Guangdong Provincial Bureau of Statistics (GPBS). Guangdong Statistical Yearbook 2010. Beijing: China Statistics Press; 2010.
Guangdong Provincial Bureau of Statistics (GPBS). Guangdong Statistical Yearbook 2011. Beijing: China Statistics Press; 2011.
Guangdong Provincial Bureau of Statistics (GPBS). Guangdong Statistical Yearbook 2012. Beijing: China Statistics Press; 2012.
Guo H, Jiang F, Cheng HR, Simpson IJ, Wang XM, Ding AJ, et al. Concurrent observations of air pollutants at two sites in the Pearl River Delta and the implication of regional transport. Atmospheric Chemistry and Physics 2009;9:7343–60.
Holland DM, Principe PP, Sickles JE. Trends in atmospheric sulfur and nitrogen species in the eastern United States for 1989–1995. Atmos Environ 1999;33:37–49.
Hu M, $\mathrm{Wu}\,Z_{\mathrm{r}}$ Slanina J, Lin P, Liu S, Zeng L. Acidic gases, ammonia and water-soluble ions in $\mathrm{PM}_{2.5}$ at a coastal site in the Pearl River Delta, China. Atmos Environ 2008;42: 6310–20.
Hua W, Chen ZM, Jie CY, Kondo Y, Hofzumahaus A, Takegama N, et al. Atmospheric hydrogen peroxide and organic hydroperoxides during PRIDE-PRD'06, China: their concentration, formation mechanism and contribution to secondary aerosols. Atmos Chem Phys 2008;8:6755–73.
Jones AM, Harrison RM. Temporal trends in sulphate concentrations at European sites and relationships to sulphur dioxide. Atmos Environ 2011;45:873–82.
Jung J, Lee H, Kim YJ, Liu X, Zhang Y, Gu J, et al. Aerosol chemistry and the effect of aerosol water content on visibility impairment and radiative forcing in Guangzhou during the 2006 Pearl River Delta campaign. J Environ Manage 2009;90:3231–44.
Kim BM, Teffera S, Zeldin MD. Characterization of $\mathrm{PM}_{2.5}$ and $\mathrm{PM_{10}}$ in the South Coast Air Basin of southern California: Part 1 — Spatial variations. J Air Waste Manage Assoc 2000;50:2034–44.
Lai SC, Zou SC, Cao JJ, Lee SC, Ho KF. Characterizing ionic species in $\mathrm{PM}_{2.5}$ and $\mathrm{PM}_{10}$ in four Pearl River Delta cities, South China. J Environ Sci (China) 2007;19:939–47.
Lei Y, Zhang $\mathrm{0},$ He KB, Streets DG. Primary anthropogenic aerosol emission trends for China, 1990–2005. Atmospheric Chemistry and Physics 2011;11:931–54.
Liu Y, Shao M, Lu S, Chang C-C, Wang J-L, Fu L. Source apportionment of ambient volatile organic compounds in the Pearl River Delta, China: Part II. Atmos Environ 2008;42: 6261–74.
Lovblad G, Tarrason L, Torseth K. Chapter 2: Sulphur, EMEP Assessment Part 1: European Perspective. Norwegian Meteorological Institute; 2004. p. 25–6.
Lu Z, Streets DG, Zhang $\mathbb{Q},$ Wang S, Carmichael GR, Cheng YF, et al. Sulfur dioxide emissions in China and sulfur trends in East Asia since 2000. Atmospheric Chemistry and Physics 2010;10:6311–31.
Lu $\mathrm{0},$ Zheng JY, Ye SQ, Shen XL, Yuan ZB, Yin SS. Emission trends and source characteristics of $S0_{2},$ NOx, $\mathrm{PM}_{10}$ and VOCs in the Pearl River Delta region from 2000 to 2009. Atmos Environ 2013;76:11–20.
Manktelow PT, Mann GW, Carslaw KS, Spracklen DV, Chipperfield MP. Regional and global trends in sulfate aerosol since the 1980s. Geophys Res Lett 2007;34.
Marinoni A, Parazols M, Brigante M, Deguillaume $\mathrm{L},$ Amato P, Delort AM, et al. Hydrogen peroxide in natural cloud water: Sources and photoreactivity. Atmos Res 2011;101: 256–63.
Merkel AW, Harder JW, Marsh DR, Smith AK, Fontenla JM, Woods TN. The impact of solar spectral irradiance variability on middle atmospheric ozone. Geophys Res Lett 2011; 38.
Ohara T, Akimoto H, Kurokawa J, Horii N, Yamaji K, Yan X, et al. An Asian emission inventory of anthropogenic emission sources for the period 1980–2020. Atmospheric Chemistry and Physics 2007;7:4419–44.
Qin Y, Xie SD. Spatial and temporal variation of anthropogenic black carbon emissions in China for the period 1980–2009. Atmospheric Chemistry and Physics 2012;12: 4825–41.
Rinehart LR, Fujita EM, Chow JC, Magliano K, Zielinska B. Spatial distribution of $\mathrm{PM}_{2.5}$ associated organic compounds in central California. Atmos Environ 2006;40:290–303.
Seinfeld JH, Pandis SM. Atmospheric Chemistry and Physics: From Air Pollution to Climate Change. 2nd ed. New York: John Wiley and Sons; 2006.
Shang Y, Sun ZW, Cao JJ, Wang XM, Zhong LJ, Bi XH, et al. Systematic review of Chinese studies of short-term exposure to air pollution and daily mortality. Environ Int 2013;54:100–11.
Simpson IJ, Wang T, Guo H, Kwok YH, Flocke F, Atlas E, et al. Long-term atmospheric measurements of $C_{1}{-}C_{5}$ alkyl nitrates in the pearl river delta region of southeast China. Atmos Environ 2006;40:1619–32.
Streets DG, Bond TC, Carmichael GR, Fernandes SD, Fu Q, He D, et al. An inventory of gaseous and primary aerosol emissions in Asia in the year 2000. J Geophys Res - Atmos 2003;108.
Sun YL, Zhuang GS, Tang AH, Wang Y, An ZS. Chemical characteristics of $\mathrm{PM}_{2.5}$ and $\mathrm{PM}_{10}$ in haze-fog episodes in Beijing. Environ Sci Technol 2006;40:3148–55.
Sun ZQ, Mu YJ, Liu YJ, Shao LY. A comparison study on airborne particles during haze days and non-haze days in Beijing. Sci Total Environ 2013;456:1–8.
Tan JH, Duan JC, He KB, Ma YL, Duan FK, Chen Y, et al. Chemical characteristics of $\mathrm{PM}_{2.5}$ during a typical haze episode in Guangzhou. J Environ Sci (China) 2009a;21:774–81.
Tan JH, Duan JC, Chen DH, Wang XH, Guo SJ, Bi XH, et al. Chemical characteristics of haze during summer and winter in Guangzhou. Atmos Res 2009b;94:238–45.
Tao J, Cao JJ, Zhang RJ, Zhu L, Zhang T, Shi S, et al. Reconstructed light extinction coefficients using chemical compositions of $\mathrm{PM}_{2.5}$ in winter in urban Guangzhou, China. Advances in Atmospheric Sciences 2012;29:359–68.
Tie XX, Wu D, Brasseur G. Lung cancer mortality and exposure to atmospheric aerosol particles in Guangzhou, China. Atmos Environ 2009;43:2375–7.
Turpin BJ, Huntzicker JJ. Identification of secondary organic aerosol episodes and quantitation of primary and secondary organic aerosol concentrations during SCAQS. Atmos Environ 1995;29:3527–44.
Van Donkelaar A, Martin RV, Brauer M, Kahn R, Levy R, Verduzco C, et al. Global estimates of ambient fine particulate matter concentrations from satellite-based aerosol optical depth: development and application. Environ Health Perspect 2010;118:847–55.
Wang Y, Zhuang GS, Tang AH, Yuan H, Sun YL, Chen SA, et al. The ion chemistry and the source of $\mathrm{PM}_{2.5}$ aerosol in Beijing. Atmos Environ 2005;39:3771–84.
Wang Y, Zhuang G, ${\mathrm{Xu}}\,{\mathsf{C}},{\mathsf{A n}}\,{\mathsf{Z}}.$ The air pollution caused by the burning of fireworks during the lantern festival in Beijing. Atmos Environ 2007;41:417–31.
Wang B, Shao M, Roberts JM, Yang G, Yang F, Hu M, et al. Ground-based on-line measurements of peroxyacetyl nitrate (PAN) and peroxypropionyl nitrate (PPN) in the Pearl River Delta, China. Int J Environ Anal Chem 2010;90:548–59.
Wang XM, Ding X, Fu XX, He QF, Wang SY, Bernard F, et al. Aerosol scattering coefficients and major chemical compositions of fine particles observed at a rural site hit the central Pearl River Delta, South China. J Environ Sci (China) 2012a;24:72–7.
Wang X, Shen ZX, Cao JJ, Zhang LM, Liu L, Li JJ, et al. Characteristics of surface ozone at an urban site of Xi'an in Northwest China. J Environ Monit 2012b;14:116–26.
Wu D, Tie XX, Li CC, Ying ZM, Lau AKH, Huang J, et al. An extremely low visibility event over the Guangzhou region: A case study. Atmos Environ 2005;39:6568–77.
Yang F, Tan J, Zhao $\mathsf{Q},$ Du Z, He K, Ma Y, et al. Characteristics of $\mathrm{PM}_{2.5}$ speciation in representative megacities and across China. Atmos Chem Phys 2011;11:5207–19.
Yao XH, Chan CK, Fang M, Cadle S, Chan T, Mulawa P, et al. The water-soluble ionic composition of $\mathrm{PM}_{2.5}$ in Shanghai and Beijing, China. Atmos Environ 2002;36:4223–34.
Zhang Q, He KB, Huo H. Cleaning China's air. Nature 2012a;484:161–2.
Zhang X, van Geffen J, Liao H, Zhang P, Lou S. Spatiotemporal variations of tropospheric $S0_{2}$ over China by SCIAMACHY observations during 2004–2009. Atmos Environ 2012b;60:238–46.
Zhao Y, Wang S, Duan L, Lei Y, Cao P, Hao J. Primary air pollutant emissions of coal-fired power plants in China: Current status and future prediction. Atmos Environ 2008;42: 8442–52.
Zheng J, Zhang L, Che W, Zheng Z, Yin S. A highly resolved temporal and spatial air pollutant emission inventory for the Pearl River Delta region, China and its uncertainty assessment. Atmos Environ 2009;43:5112–22.
Zheng J, He M, Shen X, Yin S, Yuan Z. High resolution of black carbon and organic carbon emissions in the Pearl River Delta region, China. Sci Total Environ 2012;438:189–200.
|
atmosphere
|
0019
|
10.1016/j.envpol.2017.10.123
|
image
|
Fig. 1. Layout of sampling points of BTH region.
|
|
atmosphere
|
0019
|
10.1016/j.envpol.2017.10.123
|
image
|
Fig. 2. Temporal and spatial variations of mass concentration of $\mathrm{PM}_{2.5}$ in BTH region.
|
|
atmosphere
|
0019
|
10.1016/j.envpol.2017.10.123
|
table
|
Table 1 Seasonal average mass concentration of elements for $\mathrm{PM}_{2.5}$ in BTH region $(\upmu\mathrm{g}/\up m^{3})$ .
|
|
atmosphere
|
0019
|
10.1016/j.envpol.2017.10.123
|
image
|
Fig. 3. Enrichment factor of elements for $\mathrm{PM}_{2.5}$ at (a) BJ, (b) TJ, (c) LF and (d) BD sampling sites.
|
|
atmosphere
|
0019
|
10.1016/j.envpol.2017.10.123
|
image
|
Fig. 4. Seasonal distribution of WSIIs for $\mathrm{PM}_{2.5}$ at (a) BJ, (b) TJ, (c) LF and (d) BD sampling sites.
|
|
atmosphere
|
0019
|
10.1016/j.envpol.2017.10.123
|
image
|
Fig. 5. Mass concentration of OC, EC, and the ratio of OC/EC for $\mathrm{PM}_{2.5}$ in BTH region.
|
|
atmosphere
|
0019
|
10.1016/j.envpol.2017.10.123
|
image
|
Fig. 6. Seasonal correlation of OC and EC for $\mathrm{PM}_{2.5}$ at BJ site.
|
|
atmosphere
|
0019
|
10.1016/j.envpol.2017.10.123
|
image
|
Fig. 7. Mass balance of $\mathrm{PM}_{2.5}$ at BTH region.
|
|
atmosphere
|
0019
|
10.1016/j.envpol.2017.10.123
|
table
|
Table 2 PMF factor profiles for $\mathrm{PM}_{2.5}$ at BJ, TJ, LF, and BD sampling site.
|
|
atmosphere
|
0019
|
10.1016/j.envpol.2017.10.123
|
text
| Not supported with pagination yet
|
Temporal-spatial characteristics and source apportionment of PM2.5 as well as its associated chemical species in the Beijing-Tianjin-Hebei region of China
Jiajia Gao a, b, Kun Wang a, b, Yong Wang a, c, Shuhan Liu a, c, Chuanyong Zhu a, d, Jiming Hao e, Huanjia Liu a, c, Shenbing Hua a, c, Hezhong Tian a, c, e, \*
a State Key Joint Laboratory of Environmental Simulation & Pollution Control, School of Environment, Beijing Normal University, Beijing 100875, China
b Department of Air Pollution Control, Beijing Municipal Institute of Labour Protection, Beijing 100054, China
c Center for Atmospheric Environmental Studies, Beijing Normal University, Beijing 100875, China
d School of Environmental Science and Engineering, Qilu University of Technology, Jinan 250353, China
e State Environmental Protection Key Laboratory of Sources and Control of Air Pollution Complex, School of Environment, Tsinghua University, Beijing
10084, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 20 November 2016
Received in revised form
29 October 2017
Accepted 30 October 2017
Keywords:
Beijing-Tianjin-Hebei region
$\mathrm{PM}_{2.5}$
Chemical composition
Spatial and temporal characteristics
Source apportionment
$\mathsf{P M}_{2.5}$ and its major chemical compositions were sampled and analyzed in January, April, July and October of 2014 at Beijing (BJ), Tianjin (TJ), Langfang (LF) and Baoding (BD) in order to probe the temporal and spatial characteristics as well as source apportionment of $\mathrm{PM}_{2.5}$ in the Beijing-Tianjin-Hebei (BTH) region. The results showed that $\mathrm{PM}_{2.5}$ pollution was severe in the BTH region. The average annual concentrations of $\mathrm{PM}_{2.5}$ at four sampling sites were in the range of $126{-}180~\mathrm{\textmug/m}^{3}$ with more than $95\%$ of sampling days exceeding $35\;\upmu\mathrm{g}/\mathrm{m}^{3}$ , the limit ceiling of average annual concentration of $\mathsf{P M}_{2.5}$ regulated in the Chinese National Ambient Air Quality Standards (GB3095-2012). Additionally, concentrations of $\mathrm{PM}_{2.5}$ and its major chemical species were seasonally dependent and demonstrated spatially similar variation characteristics in the BTH region. Concentration of toxic heavy metals, such as As, Cd, Cr, Cu, Mn, Ni, Pb, Sb, Se, and Zn, were higher in winter and autumn. Secondary inorganic ions $(\mathsf{S O}_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ , and $\mathrm{{NH}_{4}^{+}}$ ) were the three-major water-soluble inorganic ions (WSIIs) of $\mathrm{PM}_{2.5}$ and their mass ratios to $\mathrm{PM}_{2.5}$ were higher in summer and autumn. The organic carbon (OC) and elemental carbon (EC) concentrations were lower in spring and summer than in autumn and winter. Five factors were selected in Positive Matrix Factorization (PMF) model analysis, and the results showed that $\mathrm{PM}_{2.5}$ pollution was dominated by vehicle emissions in Beijing, combustion emissions including coal burning and biomass combustion in Langfang and Baoding, and soil and construction dust emissions in Tianjin, respectively. The air mass that were derived from the south and southeast local areas around BTH regions reflected the features of short-distant and small-scale air transport. Shandong, Henan, and Hebei were identified the major potential sources-areas of secondary aerosol emissions to $\mathrm{PM}_{2.5}$ .
$\circledcirc$ 2017 Elsevier Ltd. All rights reserved.
1. Introduction
Nowadays, due to rapid economic, industrial expansion, and urbanization in China, haze or smog episodes occur frequently.
There has been a growing concern about the air pollution impact in China, and the emphasis is shifting from pollution on a local scale to that on a regional scale. The BTH region is the largest and most dynamic economic region in northern China, accounting for $9.7\%$ of national GDP and $8.1\%$ of national population in 2014 (NBS, 2015), as well as $9.0\%$ of national coal consumption and $22.6\%$ of national steel production in 2013 (NBS, 2014). Numerous energy-intensive and highly polluting industries, such as coal-fired power plants, cement, iron and steel, oil refining, and petro-chemical manufacturing, are gathered in this region. Owing to the emissions from fossil fuel combustion, industrial production processes as well as vehicle exhaust, more polluted and hazy days have appeared and become increasingly conspicuous (Wang et al., 2014; Xue et al., 2016). It has attracted wide attention from the government and public due to the adverse effects of fine particles pollution on human health and the ecological environment. Several studies have revealed the health effects from aerosols and the relationship between $\mathsf{P M}_{2.5}$ pollution and morbidity and mortality (Wu et al., 2010; Li et al., 2015a; Feng et al., 2016). On February 29, 2012, the third revision of “the National Ambient Air Quality Standards” (NAAQS) (GB 3095-2012) was issued, in which daily and average annual ambient $\mathsf{P M}_{2.5}$ concentrations were included in the NAAQS as one of six criterion pollutants for the first time.
In the BTH region, multiple studies about chemical species of $\mathsf{P M}_{2.5}$ have been carried out in Beijing. Several studies have discussed the general characteristics of $\mathsf{P M}_{2.5}$ chemical compositions and given their seasonal variations, correlations, or sources (Sun et al., 2014; Li et al., 2015b; Lv et al., 2016). Previous studies have focused mainly on the concentrations, formation, and sources of some specific chemical species of $\mathsf{P M}_{2.5}$ in Beijing, such as elements (Schleicher et al., 2011a; Duan and Tan, 2013; Chen et al., 2016), water-soluble inorganic ions (Liu et al., 2015; Wang et al., 2015a), and carbonaceous species (Schleicher et al., 2013; Sun et al., 2016). Moreover, aerosol optical and radiative characteristics or mixing state (Zhao et al., 2011; Bi et al., 2014; Wang et al., 2015b) and new particle formation processes (Sun et al., 2013a, b; Wang et al., 2015c) have also been discussed for Beijing and its surrounding region. The results showed that $\mathsf{P M}_{2.5}$ pollution not only came from local sources emissions but also was highly influenced by the surrounding cities in the BTH region. The introduced mitigation measures could reduce particle concentrations by $30–70\%$ (Schleicher et al., 2011b, 2012; Chen et al., 2016). Chen et al. (2014) showed that some element concentrations in Beijing correlated to the restrictiveness of relative measures, especially during different traffic restrictions (including after the Olympic Games). However, most reductions were only temporary and particle concentrations in Beijing increased again back to pre-event levels already in the following year (Schleicher et al., 2011a).
To date, $\mathsf{P M}_{2.5}$ seldom has been simultaneously sampled and chemically analyzed at several sites in different cities of the BTH region. Comprehensive investigations of $\mathsf{P M}_{2.5}$ chemical species for the BTH region outside Beijing are still quite limited. To control regional $\mathsf{P M}_{2.5}$ pollution and conduct further related investigations about the BTH region, it is necessary to obtain detailed information about regional concentrations of $\mathsf{P M}_{2.5}$ as well as its chemical compositions and to know their spatial and temporal variation characteristics and sources origination. In this study, $\mathsf{P M}_{2.5}$ samples were collected simultaneously at four sites in the BTH region over four seasons. We focused on characterizing the seasonal and spatial variations in $\mathsf{P M}_{2.5}$ mass concentrations and compositions. Results on chemical mass balance and source apportionment of $\mathsf{P M}_{2.5}$ are presented and commented.
2. Experimental section
2.1. Field observation and meteorological conditions
The detailed location of each sampling site is illustrated in Fig. 1. As shown in Fig. 1, $\mathsf{P M}_{2.5}$ samples were collected simultaneously at each sampling site for 15 consecutive days during four selected months (January, April, July, and October of 2014), which represented four seasons of four typical cities in the BTH region. (A) Beijing Normal University campus in Beijing (BJ), (B) Nankai University in Tianjin (TJ), (C) North China Institute of Science and Technology in Langfang (LF), and (D) North China Electric Power University in Baoding (BD). Four sampling sites were separately located in the urban areas of four cities. The details of location and characteristics of each sampling site were given in Table S1 in the appendix. It should be noted that only three groups of data (spring, summer, and autumn season) were obtained from TJ sites because of problems with the samplers. The collection and preservation methods of $\mathsf{P M}_{2.5}$ samples were described in detail in our previous study (Gao et al., 2014, 2015). $\mathsf{P M}_{2.5}$ were collected on quartz filters (Pallflex Tissuquartz™, $90\ \mathrm{mm}$ , USA) over a period of about $24~\mathrm{h}$ each day with TH-150C medium volume air samplers (Wuhan Tianhong Instruments Co., Ltd., flow rate: $100~\mathrm{L/min})$ . For each season, in each sampling site located in the four cities, 15 valid daily samples were collected consecutively and simultaneously. In total, 225 samples were collected and analyzed. All the procedures were strictly quality-controlled to avoid any possible contamination of the samples.
In BTH region, wind mainly blows from the south in the summer and from the north in the winter (Zhao et al., 2013). The topography not only governs the wind directions but also decides the regional transportation of air pollutants. In this study, meteorological parameters (temperature, relative humidity, wind speed, observed visibility, etc.) during sampling were obtained from Wunderground website (www.wunderground.com) and Langfang Municipal Environmental Protection Bureau website (http://www.lfhbj.gov. ${\mathsf{c n}}/{\mathsf{\Omega}}$ ). The average values of meteorological data for four sampling periods at the four sites were listed in Table S2 in the appendix, and the seasonal wind roses of four sampling periods at each site were illustrated in Fig. S1 in the appendix. In most cases, the four considered cities of BTH region are controlled under same weather system and the meteorological conditions showed an overall similar variation pattern, with some advance and lag in daily evolution among cities along with the direction of prevailing winds. For the four sites, temperature was highest at TJ, and wind speed was lowest in at BD; for the four seasons, temperature ranked in the order of winter $<$ autumn $<$ spring $<$ summer, wind speed was highest in the spring, relative humidity was higher and visibility was lower in autumn and winter than in spring and summer.
2.2. Chemical composition analysis
After weighing, half of the aerosol-loaded filters were placed in Teflon tubes and each filter was digested with a 3:1:1 mixture of $\mathsf{H N O}_{3}\mathrm{-HClO}_{4}$ -HF in Teflon vessels and heated in a microwave system. Blank filers, which was brought to the sampling site and installed in the sampler without sampling, were randomly inserted for quality control and elemental concentrations of the blank filter were subtracted from the samples. Then, the digested solution was diluted to $10\,\mathrm{mL}$ . In total,18 elements (Al, As, Ca, Cd, Co, Cr, Cu, Fe, K, Mg, Na, Ni, Pb, S, Sb, Se, V, and $Z\mathfrak{n}$ ) were measured by using inductively coupled plasma atomic emission spectrometry (ICPAES, SPECTRO Analytical Instruments GmbH, SPECTRO ARCOS EOP) (Zhao et al., 2013; Gao et al., 2015). An external calibration method was used to calibrate ICP-AES. Calibrants were prepared from multi-element standard solution $(50~\upmu\mathrm{g/L}$ , Teknolab $\mathsf{A}/\mathsf{S}$ , Norway). Standard Reference Material, SRM 1648 ‘Urban Particulate Matter’, from the National Institute of Standards and Technology (Gaithersburg, MD, USA) was used to validate the methods. The SRM was treated in the same manner as the samples. Reagent blanks were also routinely analyzed in between samples to check for contamination. Determination of each sample was repeated three times, the relative standard deviation (RSD) was less than $3\%$ . The detection limits (units: $\upmu\mathrm{g}/\mathrm{L})$ were Al (5.6), Ca (6.6), Fe (0.83), K (5.4), Mg (0.37), Na (1.9), S (3.4), Mn (0.32), As (6.3), Cd (1.3), Co (9.9), Cr (6.7), Cu (0.92), Ni (1.3), Pb (11), Sb (21), Se (18), V (0.76) and Zn (4.7).
One fourth of each quartz fiber filter was put into a glass tube, to which $10~\mathrm{mL}$ of deionized water was added. After a $40\ \mathrm{min}$ ultrasonic bath at room temperature $(20{-}25~\mathrm{~^{\circ}C})$ , the solution was drawn into a $5~\mathrm{mL}$ syringe, filtered by a syringe filter, and injected into a polymeric vial with $0.45~{\upmu\mathrm{m}}$ filter cap. The polymeric vials were put into a Dionex AS-DV auto sampler and eight species of WSIIs $(\mathsf{N}\mathsf{a}^{+}$ , NH4þ, $\mathsf{K}^{+}$ , $\mathrm{Mg}^{2+}$ , ${\mathsf{C a}}^{2+}$ , $C1^{-}$ , $\mathtt{N O}_{3}^{-}$ , and $\mathrm{SO}_{4}^{2-}$ ) were analyzed by an ion chromatograph (IC, Dionex 600) that was composed of a separation column (CS12A for cations and Dionex Ionpac AS11 for anions), a guard column (AG12A for cations and Dionex Ionpac AG11 for anions), a self-regenerating suppressed conductivity detector (Dionex Ionpac ED50), and a gradient pump (Dionex Ionpac GP50) (Gao et al., 2015).
Additionally, a $0.53~\mathrm{cm}^{2}$ punch from each quartz fiber filter was analyzed using a thermal optical carbon analyzer (DRI-2001A) for eight carbon fractions (OC1, OC2, OC3, OC4, EC1, EC2, EC3 and OPC), following the IMPROVE_A protocol (Chow et al., 2007; Zhao et al., 2013).
2.3. Mass balance of $P M_{2.5}$
The chemical mass balance of atmospheric particles usually comprises seven kinds of components, including heavy metals, mineral dust, $\mathsf{N H}_{4}^{+}$ , $\mathrm{NO}_{3}^{-}$ , $\mathrm{SO}_{4}^{2-}$ , POM (particulate organic matter), and EC. In this study, $C1^{-}$ was also incorporated into the mass balance system of atmospheric particles because of its relatively high concentration among the WSIIs. POM and mineral dust were determined by using the following equations (Taylor and Mclennan, 1985; Zhao et al., 2013; Cheng et al., 2016):
$$
P O M=1.4\times O C
$$
$$
\begin{array}{r l}&{l u s t=1.89\times A l+1.4\times C a+1.43\times F e+1.21\times K}\\ &{\qquad\qquad+1.66\times M g}\\ &{\qquad=1.89\times A l+1.4\times C a+1.43\times F e+[1.21}\\ &{\qquad\qquad\times\left(0.65\times F e\right)]+1.66\times M g}\\ &{\qquad=1.89\times A l+1.4\times C a+2.22\times F e+1.66\times M g}\end{array}
$$
2.4. Source appointment of $P M_{2.5}$
Enrichment Factor (EF) method with Al as the reference element was used to assess the enrichment characteristics of various elements in $\mathsf{P M}_{2.5}$ . EF of each element which was calculated relative to the average crustal rock composition with Al as the reference element: $\mathsf{\partial}\ F\,{=}(\mathsf{X}/\mathsf{A l})_{\mathsf{A e r o s o l}}\,/(\mathsf{X}/\mathsf{A l})_{\mathsf{C r u s t}}$ (Mason and Moore,1982; Gao et al., 2014).
The US Environmental Protection Agency's positive matrix factorization (PMF) receptor model (version 3.0) was adopted to identify and determine the source apportionment of $\mathsf{P M}_{2.5}$ . The algorithms used in the PMF model to compute profiles and contributions had been reviewed in detail in our previous study (Gao et al., 2014) and were certified to be scientifically robust (Chen et al., 2014; Chuang et al., 2016). In this study, 13 independent elements (Al, As, Cd, Cr, Cu, Fe, Mn, Ni, Pb, Sb, Se, V and Zn), 8 WSIIs, OC and EC were selected in the EPA PMF 3.0 model. The input data included chemical compositions and equation-based uncertainties. The equation-based uncertainty contained detection limits and error fractions $(5\%)$ . If the concentration of each chemical compositions was less than or equal to the method detection limit (MDL) provided, the uncertainty (Unc) was calculated using the equation as follows (Polissar et al., 1998; Tan et al., 2016), the detail descriptions about the process of EPA PMF 3.0 were presented in the appendix file after Fig. S5.
$$
U n c=5/6\times M D L
$$
If the concentration was greater than the MDL provided, the calculation was:
$$
\mathrm{Unc}=\sqrt{\left(E r r o r\:F r a c t i o n\times c o n c e n t r a t i o n\:\right)^{2}+\left(M D L\right)^{2}}
$$
2.5. The backward trajectory
A GIS-based software-TrajStat was used to group trajectories that have similar geographic origins and histories for identify the potential impact of different source regions on $\mathsf{P M}_{2.5}$ composition (Squizzato and Masiol, 2015; Liu et al., 2017). The $72\,\mathrm{~h~}$ backtrajectories arriving at BJ sampling site $39^{\circ}56^{\prime}\mathsf{N}$ and $116^{\circ}20^{\prime}\mathrm{E}$ ) in the BTH region were calculated at $1\,\mathrm{h}$ intervals during the sampling period and the arrival level of the air mass was set at $100~\mathrm{m}$ AGL. The Final Operational Global Analysis data were applied to backward trajectory model, which were produced from the National Center for Environmental Prediction's Global Data Assimilation System (GDAS) wind field reanalysis (http://www.arl.noaa.gov/).
2.6. Conditional probability function study
The conditional probability function (CPF) (Kim et al., 2005; Zhang et al., 2011) can be used to identify the direction where the chemical species came from based on their ambient concentrations and wind direction data. In this study, the CPF was applied to estimate the probability that a given species emission from a given wind direction would exceed a predetermined threshold criterion (Kim et al., 2005). CPF is defined as
$$
C P F={\frac{m_{\Delta\theta}}{n_{\Delta\theta}}}
$$
where ${\pmb m}_{\pmb4\theta}$ is the number of occurrences from wind sector $\Delta\theta$ that exceed the threshold criterion and $\pmb{n}_{\Delta\theta}$ is the total number of data from the same wind sector. In this study, $\Delta\theta$ was set to $22.5^{\circ}$ . The same daily concentration was assigned to each hour of a given day to match to the hourly wind data (Kim et al., 2005). Calm winds (less than $0.5\;\mathrm{s}^{-1}$ ) were excluded from this analysis because of the difficulty in defining wind direction under calm condition (Song et al., 2007). A threshold criterion of the upper $25\%$ was chosen (Song et al., 2007; Zhang et al., 2011).
3. Results and discussion
3.1. Variation of $P M_{2.5}$ concentrations
In this study, the average annual concentrations of $\mathsf{P M}_{2.5}$ at BJ, TJ, LF and BD sampling sites were about 126, 133, 150, and $180~\upmu g/$ $\bar{\mathfrak{m}}^{3}$ , and ranked in order as $\mathrm{BD}>\mathrm{LF}>\mathrm{TJ}>\mathrm{BJ}$ (Fig. 2). The average annual $\mathsf{P M}_{2.5}$ concentrations in the southern BTH region were higher than they were in the northern regions. Hebei province covered intensively industrial activities and densely populated areas, the mixed considerable pollutants emitted from this region could be transported from distant sources to Beijing area with the air masses and aggravated the pollution level of urban Beijing (Zhao et al., 2013). Additionally, $\mathsf{P M}_{2.5}$ concentrations at the four sampling sites were significantly above the level Ⅱceiling of the National Ambient Air Quality Standard (GB 3095-2012) of China for $\mathsf{P M}_{2.5}$ $(35~\upmu\mathrm{g}/\mathrm{m}^{3}$ over a one-year period). $\mathsf{P M}_{2.5}$ concentrations for more than $95\%$ of the sampling days exceeded $35~\upmu\mathrm{g}/\mathrm{m}^{3}$ at the BJ, TJ, LF, and BD sites, and even exceeded $75\;\upmu\mathrm{g}/\mathrm{m}^{3}$ on more than $84\%$ of the sampling days at the four sites, indicating that $\mathsf{P M}_{2.5}$ pollutions were quite severe in the BTH region. In addition, the concentrations of $\mathsf{P M}_{2.5}$ in the present study were almost on the same level in comparison to those observed in Beijing and Tianjin $:123\,\upmu\mathrm{g}/\mathrm{m}^{3}$ and $141~\bar{{\upmu}}\mathrm{g}/\mathrm{m}^{3}$ , respectively, Zhao et al., 2013). The impact of regional transportation and diffusion of aerosols from areas surrounding BTH on $\mathsf{P M}_{2.5}$ concentrations in Beijing played a significant role in some situations and cannot be ignored.
Furthermore, the concentrations of $\mathsf{P M}_{2.5}$ at the four sites exhibited similar seasonal variations (Fig. 2). $\mathsf{P M}_{2.5}$ concentrations peaked in autumn and again in winter. In these two seasons, manmade emissions related to heating demand were increased. Furthermore, a combination of persistent temperature inversions and low mixed boundary layer in these two seasons were unfavorable for the dispersion of pollutants, which resulted in the
accumulation of pollution species in the atmosphere (Sun et al., 2013b; Li et al., 2015a).
3.2. Concentration and composition of $P M_{2.5}$ species
3.2.1. Elemental species
The concentrations of elements in $\mathsf{P M}_{2.5}$ in the BTH region during the four seasons are presented in Table 1. As can be seen, the average annual total concentrations of the 18 elements in $\mathsf{P M}_{2.5}$ were 15.1, 13.5, 14.1, and $16.9~\upmu\mathrm{g}/\mathrm{m}^{3}$ , which accounted for $11.4\%$ , $9.7\%$ , $9.6\%$ , and $10.2\%$ of the total mass concentration of $\mathsf{P M}_{2.5}$ at the BJ, TJ, LF, and BD sites, respectively. Among these 18 elements, S was the most abundant element. The average annual concentrations of S ranged from 5.83 to $7.94~\upmu\mathrm{g}/\mathrm{m}^{3}$ at four sampling sites. Then, average annual total concentrations of crustal elements (Al, Ca, Fe, K, Mg, and Na) were calculated at about $5.70{-}8.53~\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , accounting for $42{-}55\%$ of total element mass. Finally, trace elements (As, Cd, Cr, Cu, Mn, Ni, Pb, Sb, Se, V and Zn) were detected with very low concentrations, only contributing about $4\mathrm{-}8\%$ to the total element mass at four sampling sites. Furthermore, total proportions of crustal elements for $\mathsf{P M}_{2.5}$ were highest in spring, mainly because of the low relative humidity and strong winds during spring in northern China (Du et al., 2009). In spring, strong winds from Northern China often go across a number of desert regions in Mongolia and Inner Mongolia, where entrained more of the crustal species and enhanced dust transport (Sun et al., 2013a). In particular, it is known that fossil fuel combustion is one of the dominant emissions sources of S (Cheng et al., 2014). In our study, substantially higher S concentrations appeared during the winter season due to coal burning for the increased heating needs in this season. However, presently no concentration limits for heavy metals in ambient $\mathsf{P M}_{2.5}$ exist. Therefore, it is of urgent importance that emissions limit of heavy metals in atmospheric particulate matter be set, to help plan and implement comprehensive air pollution mitigation policies in the BTH region as well as the whole country. The average annual concentrations of minor elements for each sampling site were in the order of autumn and winter $>$ spring $>$ summer. In the autumn and winter seasons, anthropogenic emissions sources like biomass burning, coal combustion for heating, and other human activities increased (Tian et al., 2015). Additionally, the mixed layer height was low and relatively stable in the winter season, which inhibited both the vertical and horizontal diffusion of the air pollutants, resulting in pollution accumulation. All of these reasons led to the highest concentrations of trace elements occurring in autumn and winter. In spring, a phenomenon of temperature inversion often appeared because of the dry, windy, and warmer weather, which inhibited air pollutants dispersion, thus easily allowing the accumulation of toxic elements in $\mathsf{P M}_{2.5}$ . Whereas in summer, atmospheric wet deposition was frequent and atmospheric convection was strong, which favored diffusion and removal of suspended air pollutants. Consequently, during the year, the concentrations of trace elements at the four sampling sites were lowest in summer.
To identify whether the presence of a certain element in the $\mathsf{P M}_{2.5}$ was due primarily to natural or anthropogenic processes, the enrichment factor (EF) of each element was determined. The EFs of elements in the $\mathsf{P M}_{2.5}$ at the four sampling sites are illustrated in Fig. 3 (see corresponding data in Table S3). Here, the mean EFs of Al, Ca, Fe, K, Mg, Mn, Na, and V in each season for the four sampling sites were normally below 10, and similar characteristics were also found in the study of Zhao et al. (2009). It suggested that these elements more likely originated from natural sources and had no obvious enrichment in the $\mathsf{P M}_{2.5}$ . In comparison, the average EFs of As, Cd, Cr, Cu, Ni, Pb, S, Sb, Se, and Zn were higher than 10, and even as high as 1000, indicating these elements originated mainly from anthropogenic sources and had high enrichment in the $\mathsf{P M}_{2.5}$ (Gao et al., 2014).
Moreover, the EFs of elements at the four sampling sites were highest in the autumn or winter seasons (Fig. 3), which revealed that in addition to the intensity of anthropogenic emissions sources, the atmospheric diffusion condition might play a significant role in the enrichment degree of elements in $\mathsf{P M}_{2.5}$ for these two seasons. As to the regional distribution, mean EFs of elements in $\mathsf{P M}_{2.5}$ at the BD site were higher than those at the three other sampling sites, which possibly suggested that higher anthropogenic emissions (coal burning, biomass burning, etc.) as well as enrichment of elements in the $\mathsf{P M}_{2.5}$ occurred at the BD site.
3.2.2. Water-soluble inorganic ionic species
WSIIs, including $S0_{4}^{2-}$ , $\Nu0_{3}^{-}$ , $C1^{-}$ , $\mathsf{N H}_{4}^{+}$ , $\mathsf{K}^{+}$ , $C a^{2+}$ , $\mathtt{N a}^{+}$ , and $\mathrm{Mg}^{2+}$ , comprised a large proportion of atmospheric particles and played an important role in ambient air quality and visibility. Fig. 4 illustrates the seasonal variations in WSIIs in $\mathsf{P M}_{2.5}$ at the four sampling sites (see corresponding data in Table S4). The average annual total concentrations of WSIIs in the $\mathsf{P M}_{2.5}$ were 62.6, 65.5, 62.8, and $77.7~\upmu\mathrm{g}/\mathfrak{m}^{3}$ , which accounted for $49.5\%$ , $47.2\%$ , $41.1\%,$ and $42.3\%$ of the total mass concentrations of the $\mathsf{P M}_{2.5}$ at the BJ, TJ, LF, and BD sites, respectively. As discussed in Zhao et al. (2013), average annual concentrations of WSIIs in $\mathsf{P M}_{2.5}$ at the BJ and TJ from 2009 to 2010 were 52.8 and $64.3~\upmu\mathrm{g}/\mathrm{m}^{3}$ , accounting for $42.7\%$ and $45.5\%$ of the total mass concentrations of $\mathsf{P M}_{2.5}$ respectively. This evidently indicated a significant growth trend in concentrations of WSIIs and their proportions of the $\mathsf{P M}_{2.5}$ in the BTH region in recent years.
Additionally, secondary inorganic ions $(\bar{\mathrm{S}}0_{4}^{2-}$ , $\mathtt{N O}_{3}^{-}$ , and $\mathrm{NH}_{4}^{+}$ ) were the three major WSIIs of the $\mathsf{P M}_{2.5}$ comprising $30.4\%{-}45.6\%$ of the average annual $\mathsf{P M}_{2.5}$ mass at the four sampling sites. In the study by Zhao et al. (2013), the concentration and proportion of secondary inorganic ions in $\mathsf{P M}_{2.5}$ at BJ were evidently lower than they were in our study. Since 2010, with the implementation of energy-saving and emissions-reduction policies, the concentrations of regional $S0_{2}$ and nitric oxides $(\mathsf{N O}_{\mathsf{X}})$ have decreased. However, at the same time, according to Lin's research, the atmospheric oxidation increased with the increase of hydroxyl radical concentration in the past ten years (Lin and Zhao, 2009), which promoted a remarkable increase in secondary inorganic ions levels. The mass ratios of secondary inorganic ions to $\mathsf{P M}_{2.5}$ at the BJ site were highest in summer, whereas they were highest in autumn for the three other sites. In summer, when photochemical oxidation was strongest, $S0_{2}$ was converted quickly to $S0_{4}^{2-}$ . In autumn, the harvest of maize around the BTH region occurred in October, and thus open stalk burning obviously affected the concentrations of secondary inorganic ions. Moreover, the atmosphere was relatively stable and the relative humidity was high in autumn, all of which favor hygroscopic growth of particles, leading to maintenance of secondary inorganic ions at higher concentration levels (Sun et al., 2013a).
In particular, one interesting observation was that the concentrations of $\mathsf{K}^{+}$ in autumn at the LF site were evidently about 1.5e2.3 times higher than they were in the three other seasons. It is known that biomass burning in autumn is one of the dominant sources of $\mathsf{K}^{+}$ (Duan et al., 2004; Schleicher et al., 2013). According to the daily satellite monitoring data of environmental straw burning points released by China's Ministry of Environmental Protection, three straw burning fire points were observed in the city of LF on October 17, 2014. These events occurred during our sampling period in autumn at the LF site. Thus, the pronounced increase in $K^{+}$ concentrations may have been caused by the fall straw burning emissions.
The ratio of WSIIs in particles could be used to evaluate the pollution characteristics and possible sources of WSIIs. In this study, the ratios of $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ and $\mathrm{Cl}^{-}/\mathrm{N}\mathbf{a}^{+}$ were characterized in order to determine the possible sources of these ions. The averages of annual $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ at the four sampling sites ranged from 1.3 to 1.7, which were about $27\%{-}62\%$ higher than those reported by Wang. (2013). These results indicated that a greater fraction of the $\mathsf{P M}_{2.5}$ came from motor vehicle exhaust (Arimoto et al., 1996). It might be explained by both the rapid increase in number of vehicles and a series of strict clean air actions (e.g., most coal-fired industrial boilers had been shifted to natural gas, oil, or electricity) for the reduction of atmospheric $S0_{2}$ and primary particle concentrations in the BTH region in recent years (Gao et al., 2015). Moreover, in areas with intensive human activities, fossil fuel burning (coal combustion) was also a primary source of $C1^{-}$ (He et al., 2001; Han et al., 2016). Thus, the ratio of $\mathrm{Cl}^{-}/\mathrm{N}\mathbf{a}^{+}$ has been used as a good tracer for fossil fuel burning in $\mathsf{P M}_{2.5}$ in urban cities. In this study, the averages of annual $\mathrm{Cl}^{-}/\mathrm{N}\mathbf{a}^{+}$ were 8.3, 10.6, 12.2, and 13.2 at the BJ, TJ, LF, and BD sites, respectively. In addition, it was worth noting that the ratios of $\mathrm{Cl}^{-}/\mathrm{N}\mathbf{a}^{+}$ in winter were 1.7e4.9 times higher than they were in summer at the four sampling sites, and the highest ratio of $\mathrm{Cl}^{-}/\mathrm{N}\mathbf{a}^{+}$ was found at the BD site. Therefore, coal burning was the dominant source of $C1^{-}$ in the BTH region, especially for the BD site in winter.
3.2.3. Carbonaceous species
Organic carbon (OC) and elemental carbon (EC) are two dominant carbonaceous species in atmospheric aerosols. As illustrated in Fig. 5, the average annual concentrations of OC and EC in the $\mathsf{P M}_{2.5}$ ranged from 24.3 to $50.4\,\upmu\mathrm{g}/\mathrm{m}^{3}$ and $5.9{-}9.3\;\upmu\mathrm{g}/\mathrm{m}^{3}$ at the four sampling sites, respectively. Total concentrations of carbonaceous species accounted for $28.6\%$ , $21.7\%$ , $27.3\%$ , and $40.6\%$ of the total mass concentration of $\mathsf{P M}_{2.5}$ at the BJ, TJ, LF, and BD sites, respectively. Furthermore, the carbonaceous species exhibited obvious seasonal variation characteristics. For all four sampling sites, the OC and EC concentrations in the autumn and winter were higher than those in the spring and summer (Fig. 5). Similar to seasonal variation trends in OC and EC, the ratios of OC/EC were also seasonal distributed. For the BJ and TJ sites, the OC/EC values were higher in the autumn and winter seasons than in spring and summer, whereas for the LF and BD sites, the ratios of OC/EC were higher in winter and summer than in spring and autumn. In autumn and winter, concentrations of OC and EC from domestic coal-burning sources for heating, especially for OC, were increased, which led to higher OC/EC values in this season. In summer, because of the high temperature, volatile organic compounds (VOCs) in the atmosphere increased and were likely converted into secondary organic particles through photochemical reactions, leading to the increased concentration of OC. However, during the photochemical reactions in summer, only secondary OC was produced, thus, the OC/EC values showed a pronounced increasing tendency in summer.
Research on the correlation between OC and EC can be used to analyze whether these two kinds of carbonaceous components of aerosols are correlative (Turpin et al., 1990). Seasonal correlations of OC and EC for $\mathsf{P M}_{2.5}$ at the BJ site are illustrated in Fig. 6. The seasonal correlation coefficients (CCs) at the BJ site ranged from 0.28 to 0.98, and ranked in order as summer $<$ spring and autumn $<$ winter. This pointed out that the CCs of OC and EC in the winter season was probably the best. In the study of Park et al. (2015), in the Seoul metropolitan area in 2014, the correlations between OC and EC were the best during the winter $(\mathbf{r}^{2}=0.88)$ , which was generally consistent with our research. In addition, in our study, the mean ratio of OC/EC for $\mathsf{P M}_{2.5}$ in winter and autumn were 5.3 and 4.3, respectively. According to the studies by Chen et al. (2006) and Zhang et al. (2007), if the OC/EC values were in the range of 2.5e5.0, vehicle exhaust emissions were considered as the main source of OC and EC in aerosols, whereas if the OC/EC values were in the range of 5.0e10.5, coal combustion was considered the main source of OC and EC in aerosols. Therefore, in the winter and autumn seasons, coal combustion and vehicle exhaust emissions, respectively, were the two dominant sources of OC and EC for $\mathsf{P M}_{2.5}$ at the BJ site.
However, the correlations between OC and EC were weak in spring and summer (Fig. 6). Due to the low relative humidity and strong wind in spring of BTH region in spring (Du et al., 2009), the components of particulate matter were complex. In summer, strong atmospheric photochemical reactions generated more secondary OC, leading to an increase in OC concentrations. Thus, the correlations between OC and EC in spring and summer season were not strong than those in autumn and winter.
3.3. Mass balance of $P M_{2.5}$ in the BTH region
The annual mass balance of $\mathsf{P M}_{2.5}$ in the BTH region is presented in Fig. 7. POM, EC, $\mathsf{N H}_{4}^{+}$ , $\Nu0_{3}^{-}$ , $\mathrm{SO}_{4}^{2-}$ , $C1^{-}$ , mineral dust and heavy metals were the main eight chemical compositions that included in the mass balance of $\mathsf{P M}_{2.5}$ . At each site, the POM and secondary inorganic ions $\ (\mathsf{N H}_{4}^{+}$ , $\Nu0_{3}^{-}$ , and $\mathrm{SO}_{4}^{2-}$ ) were the two kinds of predominant species of the $\mathsf{P M}_{2.5}$ . The proportion of POM to $\mathsf{P M}_{2.5}$ $(25.8\%)$ was highest at the BD site, and the highest proportion of secondary inorganic ions to $\mathrm{PM}_{2.5}\left(44.0\%\right)$ was found at the BJ site. In April 2014, the mass percentages of the main components for $\mathsf{P M}_{2.5}$ in Beijing from 2012 to 2013 were released by the Beijing Municipal Environmental Protection Bureau (BJEPB) (BJEPB, 2014). In comparison with our research, the proportions of EC and POM to the $\mathsf{P M}_{2.5}$ mass released by the BJEPB were almost at the same level. The proportions of $\mathtt{N O}_{3}^{-}$ to the total mass of $\mathsf{P M}_{2.5}$ were about $30.6\%$ higher than those in the official released data, whereas the proportions of the concentrations of $S0_{4}^{2-}$ and $\mathsf{N H}_{4}^{+}$ to the $\mathsf{P M}_{2.5}$ mass were about $16.9\%$ and $22.7\%$ lower, respectively, than those released by the BJEPB.
It was especially notable that the total concentrations of heavy metals in the $\mathsf{P M}_{2.5}$ at each sampling point were very small and exhibited obvious spatial variation characteristics. These heavy metals were not only the toxic chemical components of the aerosols, but also significant identification elements of various types of anthropogenic emissions sources. In the BTH region, the total mass proportions of heavy metals in $\mathsf{P M}_{2.5}$ were in the order of $\mathsf{B J}<\mathsf{L F}<\mathsf{B D}<\mathsf{T J}$ , and were higher in autumn and winter than in spring and summer at all four sampling sites.
3.4. Source apportionment of $P M_{2.5}$ in the BTH region
In this study, our main purpose is to investigate and reveal the seasonal and spatial variation characteristics of $\mathsf{P M}_{2.5}$ in the BTH region of China, so we only collected $\mathsf{P M}_{2.5}$ samplers on four sites which reflect the typical urban environment for each city. However, it should be acknowledged that the results of $\mathsf{P M}_{2.5}$ pollution characteristics and its source apportionment for each city will be more reliable and creditable upon much more samplers collected on multiple sites (rural, urban, suburban, etc.) in a city simultaneously with the whole year duration.
The $\mathsf{P M}_{2.5}$ source apportionment results of the PMF 3.0 model are illustrated in Figs. S2eS5 in the appendix. The PMF factor profiles for $\mathsf{P M}_{2.5}$ at the BJ, TJ, LF, and BD sampling sites are presented in Table 2, respectively. Five factors were identified for $\mathsf{P M}_{2.5}$ at the four sampling sites, respectively. For BJ site, the contribution rates of the five kinds of emission sources to $\mathsf{P M}_{2.5}$ were, in order, vehicle emissions $(25.2\%)$ , combustion emissions which included coal burning and biomass combustion $(24.0\%)$ , industrial emissions $(18.0\%)$ , soil and construction dust $(12.9\%)$ , and secondary aerosol emissions $(19.9\%)$ . Notably, vehicle emissions were the primary source of $\mathsf{P M}_{2.5}$ at the BJ site, which was in well accordance with the results of $\mathsf{P M}_{2.5}$ source apportionment released by the BJEPB in 2014 (BJEPB, 2014). As depicted in Table S5, multiple source apportionments of $\mathsf{P M}_{2.5}$ in Beijing city have been carried out during the past few years. It exhibited that vehicle emissions have been contributed a considerable proportion and played an increasingly important role in Beijing in recent years. VOCs and $\Nu0\mathrm{x}$ released from vehicles were the precursors of the secondary organic compounds and nitrate in the $\mathsf{P M}_{2.5}$ and were important catalysts for increased atmospheric oxidation, which could promote the formation of an air pollution complex and gray haze (Zhu et al., 2010). Furthermore, as published by the BJEPB (2014), due to the reduction in coal consumption and the sharp increase in ownerships of vehicles, the ratio of $\mathrm{NO}_{3}^{-}$ to $\mathrm{SO}_{4}^{2-}$ in Beijing had increased from 0.6 in 2003 to 1.05 in 2014. This revealed that vehicle emissions had become the main $\mathsf{P M}_{2.5}$ pollution problem in Beijing.
For TJ, LF, and BD sites, soil and construction dust were the primary source of $\mathsf{P M}_{2.5}$ at the TJ site, which contributed $26.4\%$ of the total $\mathsf{P M}_{2.5}$ emissions. According to the $\mathsf{P M}_{2.5}$ source apportionment results released by the Tianjin Municipal Environmental Protection Bureau (TJEPB) in 2014, dust, which explained $30\%$ of $\mathsf{P M}_{2.5}$ emissions, was the largest contributor of $\mathsf{P M}_{2.5}$ at the TJ site. This was slightly higher than the results obtained in the present study. However, the $\mathsf{P M}_{2.5}$ pollutions were dominated by combustion emissions at the LF and BD sites, with contribution rates of $40.7\%$ and $34.8\%$ , respectively, to the total $\mathsf{P M}_{2.5}$ . Nowadays, the Hebei province has the highest density of coal consumption and heavy industries (iron and steel, coal-fired power plants, cement, etc.) in China. In the processes of combusting fossil fuels and biomass, making industrial products, the Hebei province will discharge a large amount of particles as well as gaseous pollutants into the atmosphere. Consequently, combustion emissions played a more significant role in the $\mathsf{P M}_{2.5}$ pollutions at the LF and BD sites.
Additionally, seasonal contributions of five factors to $\mathsf{P M}_{2.5}$ at each sampling site were illustrated in Fig. S6. As can be seen, factor contributions to $\mathsf{P M}_{2.5}$ were seasonally dependent. For factor 1 (soil and construction dust), the contributions to $\mathsf{P M}_{2.5}$ were 63.4, 17.0, 42.2 and $72.6~\upmu\mathrm{g/m}^{3}$ at BJ, TJ, LF and BD sampling sites in spring season, respectively, significantly higher than the contribution in the other three seasons. For factor 2 (combustion emissions), factor 3 (vehicle emissions) and factor 4 (industrial emissions), the contributions to $\mathsf{P M}_{2.5}$ of these three factors were much higher in autumn and winter than that of in spring and summer, especially for factor 2 at LF and BD. In LF and BD sampling site, the average contributions of Factor 2 to $\mathsf{P M}_{2.5}$ in autumn and winter season were 112.7 and $104.5~\upmu\mathrm{g/m}^{3}$ , about 4.7 and 9.3 times higher than those in spring and summer season, respectively. For factor 5 (secondary aerosol emissions), the contribution to $\mathsf{P M}_{2.5}$ was much higher in autumn and winter at BJ and TJ site, whereas the contribution was much higher in summer and winter at LF and BD site. In winter, coal-burning for heating activities were increased, and in summer, photochemical reactions were enhanced due to the suitable weather conditions. The concentration of OC showed a pronounced increasing tendency in these two seasons, leading to a higher contribution to the concentration of $\mathsf{P M}_{2.5}$ .
Moreover, as we all know, it will take some time after emission of the NOx, $S0_{2}$ and $\mathsf{N H}_{3}$ to form secondary aerosols in the atmosphere. Hence, air mass trajectory analysis was also conducted to better located the sources of secondary aerosol emissions factor (factor 5). Backward trajectories of secondary aerosol emissions with the highest G-scores and lowest G-scores to $\mathsf{P M}_{2.5}$ during our sampling periods in BJ site were illustrated in Fig. S7. In winter, Trajectory (1) with highest G-scores of factor 5 to $\mathsf{P M}_{2.5}$ was originated from Mongolia and Inner Mongolia, across over Shandong and Hebei before arriving at Beijing, which showed a long transport-patterns. Trajectory (2) with lowest G-scores of factor 5 to $\mathsf{P M}_{2.5}$ was derived from desert regions in Mongolia and Inner Mongolia, which showed the extremely long transport-patterns. The average contributions of Factor 5 to $\mathsf{P M}_{2.5}$ in Trajectory (1) was $38.4\;\upmu\mathrm{g/m}^{3}$ , about 2.9 times higher than those in Trajectory (2). The north of Shandong, south of Hebei were identified as the major potential source-areas of secondary aerosol emissions to $\mathsf{P M}_{2.5}$ in winter. In spring, summer and autumn, Trajectory (3), (5) and (7) with highest G-scores of factor 5 to $\mathsf{P M}_{2.5}$ were all derived from the south and southeast local areas (i.e. Shandong, Henan, and Hebei), which showed the features of short-distant and small-scale air transport. These south-type trajectories mainly went across the heavily polluted areas around BTH region. These areas covered highly coal industrialized activities and densely populated areas. The mixed considerable pollutants emitted from these regions could be transported to Beijing area with the air masses and aggravated the pollution level of urban Beijing (Gao et al., 2014).
Furthermore, according to the results of $\mathsf{P M}_{2.5}$ source apportionment released by the BJEPB, TJEPB, and Hebei Provincial Environmental Protection Bureau (HBEPB), $28\%{-36\%},$ , $22\%{-34\%}$ , and $15\%{-30\%}$ of $\mathsf{P M}_{2.5}$ emission sources in BJ, TJ, and HB, respectively, were from trans-regional and long-distance transmissions. Therefore, there is an incredible need to strengthen regional collaborative pollution management among the BTH regions. Only with close cooperation among the three areas, adjustment of the industrial and energy structures, and strict control of the emissions from motor vehicles and coal-fired sources as well as open biomass burning, can regional atmospheric pollution be fundamentally solved.
3.5. Conditional probability function analysis
The CPF plots of factor's G-scores at each sampling site are illustrated in Fig. S8. According to Fig. S1, BTH region are controlled under same weather system and the meteorological conditions showed an overall similar variation pattern. For the four sampling sites, the CPF plots for factor 1 indicated that the source could be located to the north direction, where a number of desert regions with amount of the crustal species were located. The CPF plots of factor 2 was similar and clearly indicated that the sources were located to the south direction. The CPF plots of factor 3, factor 4 and factor 5 were similar and obviously showed that the sources were located to the north and south direction. See Fig. S1, autumn winds often come from the southwest, where manmade emissions related to coal combustion for heating were increased. Furthermore, in this season, the meteorological conditions were unfavorable for the dispersion of pollutants, which lead to the accumulation of pollution species in the atmosphere. The CPF analysis would be better identify the direction where the factor contribution to $\mathsf{P M}_{2.5}$ came from.
4. Conclusions
In this study, seasonal and spatial variations as well as their potential source origins of $\mathsf{P M}_{2.5}$ samples are investigated, which were collected simultaneously at the BJ, TJ, LF, and BD sampling sites in the BTH region during January, April, July, and October of 2014. The following conclusions may be drawn from this study.
$\mathsf{P M}_{2.5}$ pollutions were severe in the BTH region, with average annual concentrations ranged from 126 to $180~\mathrm{{\mug/m^{3}}}$ at the four sampling sites, and the maxima $\mathsf{P M}_{2.5}$ concentration appeared in Baoding site. The $\mathsf{P M}_{2.5}$ concentrations on more than $95\%$ of the sampling days at the four sampling sites exceeded $35~\upmu\mathrm{g}/\mathrm{m}^{3}$ . The concentrations of $\mathsf{P M}_{2.5}$ and its major chemical components showed a seasonal pattern with high values in winter and low values in summer and presented spatially similar variation characteristics in the plain region of BTH. The concentrations of crustal elements (Al, Ca, Fe, K, Mg, and Na) were higher in spring due to the low relative humidity and strong winds of northern China during that season. The sum of toxic heavy metals (As, Cd, Cr, Cu, Mn, Ni, Pb, Sb, Se, and Zn) were higher in winter and autumn due to the increased coal combustion for heating and other human activities, like biomass burning. The secondary inorganic ions, $S0_{4}^{2-}$ , $\mathsf{N O}_{3}^{-}$ , and $\mathsf{N H}_{4}^{+}$ , were the three major WSIIs in the $\mathsf{P M}_{2.5}.$ The mass ratios of secondary inorganic ions to $\mathsf{P M}_{2.5}$ were higher in summer and autumn. Additionally, the concentrations of OC and EC in the spring and summer were lower than they were in the autumn and winter seasons. The CCs of OC and EC in winter was probably the best, demonstrating the significant contribution of primary OC and EC emissions from coal burning.
The US EPA PMF 3.0 analysis showed that the $\mathsf{P M}_{2.5}$ pollutions were dominated by vehicle emissions in Beijing. However, in LF and
BD, combustion emissions played a significant role in $\mathsf{P M}_{2.5}$ pollutions. At the TJ site, soil and construction dust emissions were more important than other sources. Air mass trajectory analysis indicated that the major potential sources-areas of secondary aerosol emissions were likely passing across the south and southeast local areas around BTH regions (i.e. Shandong, Henan, and Hebei), showing the features of short-distant and small-scale air transport. According to this work, strengthening regional collaborative pollution management among the three BTH regions should be an effective means to fundamentally solve the regional atmospheric pollution.
Acknowledgments
This work was supported by the National Natural Science Foundation of China (21377012, 21177012 and 21607008), the Special Program on Public Welfare of the Ministry of Environmental Protection (201409022), State Environmental Protection Key Laboratory of Sources and Control of Air Pollution Complex (No. SCAPC201305) and the special fund of State Key Joint Laboratory of Environmental Simulation and Pollution Control (13L02ESPC). We are also indebted to Nankai University, North China Electric Power University, and North China Institute of Science and Technology for their contributions to the field sampling work.
Appendix A. Supplementary data
Supplementary data related to this article can be found at https://doi.org/10.1016/j.envpol.2017.10.123.
References
Arimoto, R., Duce, R.A., Savoie, D.L., Prospero, J.M., Talbot, R., Cullen, J.D., Tomza, U., Lewis, N.F., Ray, B.J., 1996. Relationships among aerosol constituents from asia and the North pacific during PEM-west A. J. Geophys. Res. 101, 2011e2023.
BJEPB (Beijing Environmental Protection Bureau), 2014. Source apportionment of $\mathrm{PM}_{2.5}$ in Beijing. http://www.bjepb.gov.cn/bjepb/413526/331443/331937/ 333896/396191/index.html.
Bi, J.R., Huang, J.P., Hu, Z.Y., Holben, B.N., Guo, Z.Q., 2014. Investigating the aerosol optical and radiative characteristics of heavy haze episodes in Beijing during January of 2013. J. Geophys. Res. Atmos. 119, 9884e9900.
Chen, Y., Schleicher, N., Chen, Y.Z., Chai, F.H., Norra, S., 2014. The influence of governmental mitigation measures on contamination characteristics of $\mathrm{PM}_{2.5}$ in Beijing. Sci. Total Environ. 490, 647e658.
Chen, Y., Schleicher, N., Cen, K., Liu, X.L., Yu, Y., Zibat, V., Dietze, V., Fricker, M., Kaminski, U., Chen, Y.Z., Chai, F.H., Norra, S., 2016. Evaluation of impact factors on $\mathrm{PM}_{2.5}$ based on long-term chemical components analyses in the megacity Beijing, China. Chemosphere 155, 234e242.
Chen, Y.J., Zhi, G.R., Feng, Y.L., Fu, J.M., Feng, J.J., Sheng, G.Y., Simoneit, Bernd R.T., 2006. Measurements of emission factors for primary carbonaceous particles from residential raw-coal combustion in China. Geophys. Res. Lett. 33, L20815eL20818.
Cheng, H.R., Gong, W., Wang, Z.W., Zhang, F., Wang, X.M., Lv, X.P., Liu, J., Fu, X.X., Zhang, G., 2014. Ionic composition of submicron particles $(\mathrm{PM}_{1.0})$ during the long-lasting haze period in January 2013 in Wuhan, central China. J. Environ. Sci. 26, 810e817.
Cheng, Z., Luo, L.N., Wang, S.X., Wang, Y.G., Sharma, S., Shimadera, H., Wang, X.L., Bressi, M., de Miranda, R.M., Jiang, J.K., Zhou, W., 2016. Status and characteristics of ambient $\mathrm{PM}_{2.5}$ pollution in global megacities. Environ. Int. 89e90, 212e221.
Chow, J.C., Watson, J.G., Antony Chen, L.-W., Oliver Chang, M.C., Robinson, N.F., Trimble, D., Kohl, S., 2007. The IMPROVE A temperature protocol for thermal/ optical carbon analysis: maintaining consistency with a long-term database. J. Air & Waste Manag. Assoc. 57, 1014e1023.
Chuang, M.T., Chen, Y.C., Lee, C.T., Cheng, C.H., Tsai, Y.J., Chang, S.Y., Su, Z.S., 2016. Apportionment of the sources of high fine particulate matter concentration events in a developing aerotropolis in Taoyuan, Taiwan. Environ. Pollut. 214, 273e281.
Du, W.P., Gao, Y.X., Wang, Y.S., Xin, J.Y., Lin, Y.Y., 2009. Influence of dust storms on urban atmospheric environmental quality in Northern China. Res. Environ. Sci. 22, 1021e1026 (in Chinese).
Duan, J.C., Tan, J.H., 2013. Atmospheric heavy metals and Arsenic in China: situation, sources and control policies. Atmos. Environ. 74, 93e101.
Duan, F.K., Liu, X.D., Yu, T., Cachier, H., 2004. Identification and estimate of biomass burning contribution to the urban aerosol organic carbon concentrations in Beijing. Atmos. Environ. 38, 1275e1282.
Feng, C., Li, J., Sun, W.J., Zhang, Y., Wang, Q., 2016. Impact of ambient fine particulate matter $(\mathrm{PM}_{2.5})$ exposure on the risk of influenza-like-illness: a time-series analysis in Beijing, China. Environ. Health 15, 17e28.
Gao, J.J., Tian, H.Z., Cheng, K., Lu, L., Wang, Y.X., Wu, Y., Zhu, C.Y., Liu, K.Y., Zhou, J.R., Liu, X.G., Chen, J., Hao, J.M., 2014. Seasonal and spatial variation of trace elements in multi-size airborne particulate matters of Beijing, China: mass concentration, enrichment characteristics, source apportionment, chemical speciation and bioavailability. Atmos. Environ. 99, 257e265.
Gao, J.J., Tian, H.Z., Cheng, K., Lu, L., Zheng, M., Wang, S.X., Wang, K., 2015. The variation of chemical characteristics of $\mathrm{PM}_{2.5}$ and $\mathrm{PM}_{10}$ and formation causes during two haze pollution events in urban Beijing, China. Atmos. Environ. 107, 1e8.
Han, B., Zhang, R., Yang, W., Bai, Z.P., Ma, Z.Q., Zhang, W.J., 2016. Heavy haze episodes in Beijing during January 2013: inorganic ion chemistry and source analysis using highly time-resolved measurements from an urban site. Sci. Total Environ. 544, 319e329.
He, K.B., Yang, F.M., Ma, Y.L., Zhang, Q., Yao, X.H., Chan, C.K., Cadle, S., Chan, T., Mulawa, P., 2001. The characteristics of $\mathrm{PM}_{2.5}$ in Beijing, China. Atmos. Environ. 35, 4959e4970.
Kim, E., Hopke, P.K., Kenski, D.M., Koerber, M., 2005. Source of fine particles in a rural Midwestern U.S. area. Environ. Sci. Technol. 39, 4953e4960.
Li, Y., Ma, Z.Q., Zheng, C.J., Shang, Y., 2015a. Ambient temperature enhanced acute cardiovascular e respiratory mortality effects of $\mathrm{PM}_{2.5}$ in Beijing, China. Int. J. Biometeorology 59, 1761e1770.
Li, X., Zhang, Q., Zhang, Y., Zheng, B., Wang, K., Chen, Y., Wallington, T.J., Han, W.J., Shen, W., Zhang, X.Y., He, K.B., 2015b. Source contributions of urban $\mathrm{PM}_{2.5}$ in Beijing-Tianjin-Hebei region: changes between 2006 and 2013 and relative impacts of emissions and meteorology. Atmos. Environ. 123 (Part A), 229e239.
Lin, Y.P., Zhao, C.S., 2009. The oxidation in the troposphere: a review. Advances in. Earth Sci. 24, 488e496.
Liu, B.S., Wu, J.H., Zhang, J.Y., Wang, L., Yang, J.M., Liang, D.N., Dai, Q.L., Bi, X.H., Feng, Y.C., Zhang, Y.F., Zhang, Q.X., 2017. Characterization and source apportionment of $\mathrm{PM}_{2.5}$ based on error estimation from EPA PMF 5.0 model at a medium city in China. Environ. Pollut. 222, 10e22.
Liu, X.G., Sun, K., Qu, Y., Hu, M., Sun, Y.L., Zhang, F., Zhang, Y.H., 2015. Secondary formation of sulfate and nitrate during a haze episode in megacity Beijing, China. Aerosol Air Qual. Res. 15, 2246e2257.
Lv, B.L., Zhang, B., Bai, Y.Q., 2016. A systematic analysis of $\mathrm{PM}_{2.5}$ in Beijing and its sources from 2000-2012. Atmos. Environ. 124 (Part B), 98e108.
Mason, B., Moore, C.B., 1982. Principles of Geochemistry, Fourthed. Wiley, New York.
National Bureau of Statistics of China (NBS), 2015. China Statistical Yearbook. China Statistics Press, Beijing, China, 2014 (in Chinese).
National Bureau of Statistics of China (NBS), 2014. China Statistical Yearbook. China Statistics Press, Beijing, China, 2013 (in Chinese).
Park, J.S., Song, I.H., Park, S.M., Shin, H., Hong, Y., 2015. The characteristics and seasonal variations of OC and EC for $\mathrm{PM}_{2.5}$ in Seoul metropolitan area in 2014. J. Environ. Impact Assess. 24, 578e592.
Polissar, A.V., Hopke, P.K., Paatero, P., Malm, W.C., Sisler, J.F., 1998. Atmospheric aerosol over Alaska 2. Elemental composition and sources. J. Geophys. Res. 103, 19045e19057.
Schleicher, N., Norra, S., Chai, F.H., Chen, Y.Z., Wang, S.L., Chen, K.Q., Yu, Y., Stuben, D., 2011a. Temporal variability of trace metal mobility of urban particulate matter from Beijing - a contribution to health impact assessments of aerosols. Atmos. Environ. 45, 7248e7265.
Schleicher, N., Norra, S., Chen, Y.Z., Chai, F.H., Wang, S.L., 2012. Efficiency of mitigation measures to reduce particulate air pollution-A case study during the Olympic Summer Games 2008 in Beijing, China. Sci. Total Environ. 427e428, 146e158.
Schleicher, N., Norra, S., Dietze, V., Yu, Y., Fricker, M., Kaminski, U., Chen, Y., Cen, K., 2011b. The effect of mitigation measures on distributed mass concentrations of atmospheric particles and black carbon concentrations during the Olympic Summer Games 2008 in Beijing. Sci. Total Environ. 412e413, 185e193.
Schleicher, N., Norra, S., Fricker, M., Kaminski, U., Chen, Y.Z., Chai, F., Wang, S.L., Yu, Y., Cen, K., 2013. Spatio-temporal variations of black carbon concentrations in the Megacity Beijing. Environ. Pollut. 183, 392e401.
Song, Y., Shao, M., Liu, Y., Lu, S., Kuster, W., Goldan, P., Xie, S., 2007. Source apportionment of ambient volatile organic compounds in Beijing. Environ. Sci. Technol. 41, 4348e4353.
Squizzato, S., Masiol, M., 2015. Application of meteorology-based methods to determine local and external contributions to particulate matter pollution: a case study in Venice (Italy). Atmos. Environ. 119, 69e81.
Sun, F.B., Lun, X.X., Liu, X.H., Mo, L., Li, R.N., Zhang, H.X., Chen, J.G., Cao, Y.Y., Shi, F.T., Yu, X.X., 2016. Analysis of organic and elemental carbon in heating and nonheating periods in four locations of Beijing. Environ. Technol. 37, 121e128.
Sun, Y.L., Jiang, Q., Wang, Z.F., Fu, P.Q., Li, J., Yang, T., Yin, Y., 2014. Investigation of the sources and evolution processes of severe haze pollution in Beijing in January 2013. J. Geophys. Res. Atmos. 119, 4380e4398.
Sun, Y.L., Wang, Z.F., Fu, P.Q., Jiang, Q., Yang, T., Li, J., Ge, X.L., 2013a. The impact of relative humidity on aerosol composition and evolution processes during wintertime in Beijing, China. Atmos. Environ. 77, 927e934.
Sun, Y.L., Wang, Z.F., Fu, Q.P., Yang, T., Dong, H.B., Li, T., Jia, J.J., 2013b. Aerosol composition, sources and processes during wintertime in Beijing, China. Atmos. Chem. Phys. 13, 4577e4592.
Tan, J.H., Duan, J.X., Ma, Y.L., He, K.B., Cheng, Y., Deng, S.X., Huang, Y.L., Si-Tu, S.P., 2016. Long-term trends of chemical characteristics and sources of fine particle in Foshan City, Pearl River Delta: 2008e2014. Sci. Total Environ. 565, 519e528.
Taylor, S.R., Mclennan, S.M., 1985. The Continental Crust: its Composition and Evolution. Blackwells. Oxford, England.
Tian, H.Z., Zhu, C.Y., Gao, J.J., Cheng, K., Hao, J.M., Wang, K., Hua, S.B., Zhou, J.R., 2015. Quantitative assessment of atmospheric emissions of toxic heavy metals from anthropogenic sources in China: historical trend, spatial distribution, uncertainties, and control policies. Atmos. Chem. Phys. 15, 10127e10147.
Turpin, B.J., Cary, R.A., Huntzicker, J.J., 1990. An in-situ, time resolved analyzed for aerosol organic and elemental carbon. Aerosol Sci. Technol. 12, 161e171.
Wang, Q.Q., Ma, Y.L., Tan, J.H., Zheng, N.J., Duan, J.C., Sun, Y.L., He, K.B., Zhang, Y.X., 2015a. Characteristics of size-fractionated atmospheric metals and watersoluble metals in two typical episodes in Beijing. Atmos. Environ. 119, 294e303.
Wang, H.B., Tian, M., Li, X.H., Chang, Q., Cao, J.J., Yang, F.M., Ma, Y.L., He, K.B., 2015b. Chemical composition and light extinction contribution of $\mathrm{PM}_{2.5}$ in urban Beijing for a 1-year period. Aerosol Air Qual. Res. 15, 2200e2211.
Wang, L., 2013. Characteristics of Water-soluble Inorganic Ions in Aerosol Particles in JingJinJi-mega Typical Cities. Master Thesis. Beijing University of Chemical Technology, Beijing, China (in Chinese).
Wang, Y.S., Zhang, J.K., Wang, L.L., Hu, B., Tang, G.Q., Liu, Z.R., Sun, Y., Ji, D.S., 2014. Researching significance, status and expectation of haze in Beijing-TianjinHebei region. Adv. Earth Sci. 29 (3), 388e396 (in Chinese).
Wang, Z.B., Hu, M., Pei, X.Y., Zhang, R.Y., Paasonen, P., Zheng, J., Yue, D.L., Wu, Z.J., Boy, M., Wiedensohler, A., 2015c. Connection of organics to atmospheric new particle formation and growth at an urban site of Beijing. Atmos. Environ. 103, 7e17.
Wu, S.W., Deng, F.R., Niu, J., Huang, Q.S., Liu, Y.C., Guo, X.B., 2010. Association of heart rate variability in taxi drivers with marked changes in particulate air pollution in Beijing in 2008. Environ. Health Perspect. 118, 87e91.
Xue, Y.F., Tian, H.Z., Yan, J., Zhou, Z., Wang, J.L., Nie, L., Pan, T., Zhou, J.R., Hua, S.B., Wang, Y., Wu, X.Q., 2016. Temporal trends and spatial variation characteristics of primary air pollutants emissions from coal-fired industrial boilers in Beijing, China. Environ. Pollut. 213, 717e726.
Zhang, Y.F., Xu, H., Tian, Y.Z., Shi, G.L., Zeng, F., Wu, J.H., Zhang, X.Y., Li, X., Zhu, T., Feng, Y.C., 2011. The study on vertical variability of $\mathrm{PM_{10}}$ and the possible sources on a 220m tower, in Tianjin, China. Atmos. Environ. 45, 6133e6140.
Zhang, Y.X., Shao, M., Zhang, Y.H., Zeng, L.M., He, L.Y., Zhu, B., Wei, Y.J., Zhu, X.L., 2007. Source profiles of particulate organic matters emitted from cereal straw burnings. J. Environ. Sci. 19, 167e175.
Zhao, P.S., Zhang, X.L., Xu, X.F., Zhao, X.J., 2011. Long-term visibility trends and characteristics in the region of Beijing, Tianjin, and Hebei, China. Atmos. Res. 101, 711e718.
Zhao, P.S., Dong, F., He, D., Zhao, X.J., Zhang, X.L., Zhang, W.Z., Yao, Q., Liu, H.Y., 2013. Characteristics of concentrations and chemical compositions for $\mathrm{PM}_{2.5}$ in region of Beijing, Tianjin, and Hebei, China. Atmos. Chem. Phys. 13, 4631e4644.
Zhao, Q., He, K.B., Ma, Y.L., Jia, Y.T., Cheng, Y., Liu, H., Wang, S.W., 2009. Regional PM pollution in Beijing and surrounding area during summertime. Environ. Sci. 30, 1873e1880 (in Chinese).
Zhu, T., Shang, J., Zhao, D.F., 2010. The roles of heterogeneous chemical processes in the formation of an air pollution complex and gray haze. Sci. Sin. Chim. 40 (12), 1731e1740 (in Chinese).
|
atmosphere
|
0020
|
10.1016/j.atmosenv.2011.07.041
|
image
|
Fig. 1. Scatter plots of major water-soluble ions from AIM and Filter-based methods.
|
|
atmosphere
|
0020
|
10.1016/j.atmosenv.2011.07.041
|
table
|
Table 1 Concentrations of water-soluble ions (mean concentrations $\pm$ standard deviation (SD)) in four seasons in Jinan $\left(\upmu\mathrm{g}\:\mathsf{m}^{-3}\right)$ .
|
|
atmosphere
|
0020
|
10.1016/j.atmosenv.2011.07.041
|
table
|
Table 2 Mass concentrations of $\mathsf{P M}_{2.5}$ and the major chemical components in Jinan and other cities over the world $(\upmu\mathrm{g}\textrm{m}^{-3}$ ).
|
|
atmosphere
|
0020
|
10.1016/j.atmosenv.2011.07.041
|
image
|
Fig. 2. Seasonal variations of SOR, NOR, $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ mass ratio and temperature.
|
|
atmosphere
|
0020
|
10.1016/j.atmosenv.2011.07.041
|
image
|
Fig. 3. The diurnal profiles of $50_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ , $\mathrm{NH_{4}^{+}}$ and $\mathrm{NO}_{2}^{-}$ in four seasons (a) Spring; (b) Summer; (c) Autumn; (d) Winter.
|
|
atmosphere
|
0020
|
10.1016/j.atmosenv.2011.07.041
|
image
|
Fig. 4. Mean clusters and the corresponding mean ions concentrations in four seasons.
|
|
atmosphere
|
0020
|
10.1016/j.atmosenv.2011.07.041
|
image
|
Fig. 5. RCF distribution for sulfate in Jinan in four seasons.
|
|
atmosphere
|
0020
|
10.1016/j.atmosenv.2011.07.041
|
image
|
Fig. 6. RCF distribution for nitrate in Jinan in four seasons.
|
|
atmosphere
|
0020
|
10.1016/j.atmosenv.2011.07.041
|
table
|
Table 3 Factor loadings from PCA in the four seasons.
|
|
atmosphere
|
0020
|
10.1016/j.atmosenv.2011.07.041
|
text
| Not supported with pagination yet
|
Semi-continuous measurement of water-soluble ions in $\mathsf{P M}_{2.5}$ in Jinan, China: Temporal variations and source apportionments
Xiaomei Gao a, Lingxiao Yang a,b,\*, Shuhui Cheng a, Rui Gao a, Yang Zhou a, Likun Xue a, Youping Shou a, Jing Wang a, Xinfeng Wang a, Wei Nie a, Pengju Xu a, Wenxing Wang a,c
a Environment Research Institute, Shandong University, Jinan 250100, China b School of Environmental Science and Engineering, Shandong University, Jinan 250100, China c Chinese Research Academy of Environmental Sciences, Beijing 100012, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 18 April 2011 Received in revised form 21 July 2011
Accepted 24 July 2011
Keywords:
semi-continuous
Water-soluble ions
PM2.5
Seasonal and diurnal variations
Transport patterns
Sources
Jinan
To better understand secondary aerosol pollution and potential source regions, semi-continuous measurement of water-soluble ions in $\mathrm{PM}_{2.5}$ was performed from December 2007 to October 2008 in Jinan, the capital of Shandong Province. The data was analyzed with the aid of backward trajectory cluster analysis in conjunction with redistributed concentration field (RCF) model and principal component analysis (PCA). $S0_{4}^{2-}$ , $\Nu0_{3}^{-}$ and $\mathrm{NH}_{4}^{+}$ were the most abundant ionic species with annual mean concentrations $\vdots$ standard deviations) of 38.33 $(\pm26.20)$ , 15.77 $(\pm12.06)$ and 2 $1.26~(\pm16.28)~\upmu\mathrm{g}~\mathsf{m}^{-3}$ respectively, which are among the highest levels reported in the literatures in the world. Well-defined seasonal and diurnal patterns of $S0_{4}^{2-}$ , $\Nu0_{3}^{-}$ and $\mathsf{N H}_{4}^{+}$ were observed. The fine sulfate and nitrate oxidation ratios (SOR and NOR) were much higher in summer (SOR: $0.47\pm0.13$ ; NOR: $0.28\pm0.03]$ than those in other seasons (SOR: 0.17e0.30; NOR: 0.12e0.14), indicating more extensive formations of $S0_{4}^{2-}$ and $\Nu0_{3}^{-}$ in summer. The most frequent air masses connected with high concentrations of $S0_{4}^{2-}$ , $\Nu0_{3}^{-}$ and $\mathrm{NH}_{4}^{+}$ originated from Shandong Province in spring, autumn and winter, while from the Yellow Sea in summer, and then slowly traveled in Shandong Province to Jinan. RCF model indicated that Shandong Province was the main potential source region for $S0_{4}^{2-}$ and $\Nu0_{3}^{-}$ and other potential source regions were also identified including the provinces of Hebei, Henan, Anhui and Jiangsu and the Yellow Sea. Principal component analysis indicated that the major sources contributing to $\mathrm{PM}_{2.5}$ pollution were secondary aerosols, coal/biomass burnings and traffic emissions.
Crown Copyright $\circledcirc$ 2011 Published by Elsevier Ltd. All rights reserved.
1. Introduction
The rapid industrialization and urbanization in China have inevitably led to remarkable increase of air pollutants emissions in the past two decades. Coal combustions and automobile exhausts are mainly responsible for the severe air pollution in large cities in China (Chen et al., 2004). Primary air pollutants such as $S0_{2}$ and dust have been successfully reduced due to the enforcement of control measurements in recent years (Chan and Yao, 2008), while $\mathsf{P M}_{2.5}$ has emerged as the biggest concern. $\mathsf{P M}_{2.5}$ is believed to be a predominant factor to scatter and absorb solar radiation and reduce visibility (Sloane et al.,1991), and it can easily penetrate into lungs and lead to the respiratory and mutagenic diseases (Hughes et al., 1998). Water-soluble ions account for about half of the $\mathsf{P M}_{2.5}$ mass (Zhang et al., 2007b; Chan and Yao, 2008). The major water-soluble ions such as sulfate, nitrate and ammonium have effects on the hydroscopic nature and acidity of aerosols (Ocskay et al., 2006), while their characteristics vary significantly with seasons and geographic locations.
Shandong Province is located on the central coast of China and adjacent to Korea and Japan. The area of Shandong constitutes only $1.6\%$ of total China area while anthropogenic emissions from Shandong contributed approximately $10\%$ for $S0_{2}$ , $8\%$ for $\mathsf{N O}_{x},$ and $9\%$ for $\mathsf{P M}_{2.5}$ to China emissions in 2006 (National Bureau of Statistics of China, 2009; Zhang et al., 2009). Regional transport of air pollutants from Shandong was found to contribute to the aerosol pollution in Beijing under prevailing south and southeast winds (Streets et al., 2007). Besides, Shandong was identified as a potential source region for secondary inorganic aerosol in Seoul, Korea (Heo et al., 2009). As the capital of Shandong Province, Jinan was listed in the group of large cities with the highest concentrations of $S0_{2}$ , $\Nu0_{x}$ and TSP in the world (Baldasano et al., 2003). Previous study showed that Jinan suffered serious $\mathsf{P M}_{2.5}$ pollution and $S0_{4}^{2-}$ and $\Nu0_{3}^{-}$ were major contributors to the visibility reduction (Yang et al., 2007). However, temporal variations (especially diurnal variation) and source apportionments of watersoluble ions in $\mathsf{P M}_{2.5}$ in Jinan are still unclear.
In order to better understand secondary aerosol pollution and potential source regions in Jinan, semi-continuous measurement of water-soluble ions in $\mathsf{P M}_{2.5}$ was performed, in conjunction with trace gases and meteorological parameters in 2008. This paper presents the overall results of water-soluble ions. We first show the seasonal and diurnal variations of major water-soluble ions, and then deploy backward trajectory cluster analysis and redistributed concentration field (RCF) model to allocate the potential source regions for secondary ions in Jinan. Finally, principal component analysis (PCA) is used to uncover the underlying factors contributing to the $\mathsf{P M}_{2.5}$ pollution in Jinan.
2. Experiments and methodologies
2.1. Sampling sites
Four intensive measurements were conducted from December 2007 to October 2008. In winter (Dec 1 2007eJan 3 2008) and spring $(\mathrm{Apr~}1\mathrm{-}18~2008)$ , the observation site was chosen at the rooftop of public teaching building in Hongjialou Campus of Shandong University (in brief “HJLC”; $36^{\circ}69^{\prime}\mathrm{N}$ , $117^{\circ}06^{\prime}{\mathrm{E}}$ ), and the detailed information about this site was given by Xu et al. (2010). In summer (Jun 5e17 2008) and autumn $(\mathrm{Sep}\,12{-}0\mathrm{ct}\,15\,2008)$ , the study site was set up on the fourth floor at the building of Environmental Science and Engineering in Central Campus of Shandong University (in brief $^{*}\!C\!C^{\prime\prime}$ ; $36^{\circ}40^{\prime}\mathsf{N}$ , $117^{\circ}03^{\prime}\mathrm{E})$ (Shou et al., 2010), 1 km away from the HJLC. The sampling inlet was $\sim\!15\mathrm{~m~}$ above the ground level at the two sites. These two sites are both located in the urban area in Jinan, being surrounded by residential or commercial districts (Xu et al., 2010).
2.2. Instruments
An ambient ion monitor (AIM; Model URG 9000B, URG Corporation) was used to measure hourly concentrations of water-soluble ions in $\mathsf{P M}_{2.5}$ , including $\mathsf{F}^{-}$ , $C1^{-}$ , $\Nu0_{2}^{-}$ , $\Nu0_{3}^{-}$ , $S0_{4}^{2-}$ , $\mathtt{N a}^{+}$ , $\mathsf{N H}_{4}^{+}$ , $\mathsf{K}^{+}$ $\mathrm{Mg}^{2+}$ and ${\mathsf{C a}}^{2+}$ . The instrument has been used in several field campaigns, and the details can refer to Zhou et al. (2010). To avoid positive interference from $S0_{2}$ to the $50_{4}^{2-}$ measurement (Wu and Wang, 2007; Zhou et al., 2010), a NaOH solution $\left(5\;\mathrm{mmol}\;\mathrm{L}^{-1}\right.$ ) was substituted for the original ultra-pure water as the denuder liquid to enhance the absorption of $S0_{2}$ .
$\mathsf{P M}_{2.5}$ samples were collected on Teflon membranes using a commercially available filter-based sampler (Reference Ambient Air Sampler, Model RAAS 2.5e400, Thermo Andersen) and 101 sets of samples were obtained during our observation. The collection of samples and analysis of water-soluble ions have been described elsewhere (Wu and Wang, 2007; Zhou et al., 2010). In this study, we compared the results obtained from AIM and traditional filterbased measurements in Section 3.1.1.
Other instruments for measuring $S0_{2}$ (TEI, Model 43C), $\Nu0_{x}$ (TEI Model 42i-TL), $0_{3}$ (TEI, Model 49C), CO (API Model 300E) and BC (Magee Scientific, Berkeley, California, USA, Model AE-21) have been described in our previous studies (Zhou et al., 2009; Wang et al., 2010). And the meteorological data were directly obtained from an automatic meteorological station (Xu et al., 2010).
2.3. Trajectories calculation and cluster analysis
Three-day backward trajectories, terminated at $50\;\mathrm{m}$ a.s.l., were computed every hour by the Hybrid Single-Particle Lagrangian
Integrated Trajectory model (HYSPLIT, version 4.9) with the Global Data Assimilation system (GDAS) meteorological data (Draxler and Rolph, 2003). A total of 2278 backward trajectories with 72 hourly trajectory endpoints in four seasons were used as input for further analysis. A K-means cluster approach was then used to classify the trajectories into several different clusters (Salvador et al., 2010) and five suitable clusters were chosen in four seasons.
2.4. Redistributed concentration field (RCF) model
In this study, $C_{k}$ is the concentration measured at the receptor site for trajectory $k$ . If $C_{i k}$ is the mean concentration of the grid cells which are hit by segment $i\,(i=1,N_{k})$ of trajectory $k$ (Salvador et al., 2010), then the distribution of air pollutants for trajectory $k$ is
$$
C_{i k}\,=\,C_{k}\,\frac{C_{i k}N_{k}}{\sum_{i\,=\,1}^{N_{k}}C_{i k}},\,i\,=\,1,N_{k}
$$
After the redistribution of all individual trajectories, the new concentration field $\overline{{C}}_{m n}$ is calculated by the redistributed concentration $C_{i k}$ :
$$
\log\overline{{C}}_{m n}\,=\,\frac{1}{\sum_{k=1}^{M}\sum_{i=1}^{N_{k}}\tau_{m n i k}}\,\,\sum_{k=1}^{M}\sum_{i=1}^{N_{k}}\log(C_{i k})\tau_{m n i k}
$$
In eq. (2), $\tau_{m n i l}$ is the residence time of segment i for trajectory $k$ in grid cell $(m,n)$ . The new concentration filed is repeated until the average difference for the concentration fields of two successive iterations is below a threshold value of $0.5\%$ . The geophysical regions passed by the trajectories were divided into $1.0^{\circ}~\times~1.0^{\circ}$ grids.
3. Results and discussions
3.1. Overall statistics of water-soluble ions in $P M_{2.5}$
3.1.1. Comparison of results from AIM and traditional filter-based measurements
To evaluate the performance of modified AIM, hourly data from AIM were averaged to match the collection time of filter samples for comparison. The results from AIM and traditional filter-based measurements are plotted in Fig. 1. Excellent correlations were found for major ionic species, namely, $S0_{4}^{2-}$ , $\Nu0_{3}^{-}$ , $\mathrm{NH_{4}^{+}}$ , $C1^{-}$ and $\mathsf{K}^{+}$ $(R^{2}=0.84–\bar{0}.95$ , RMA slope $=0.83–1.08)$ , and good correlations $'R^{2}=0.46–0.90$ , RMA slope $=0.83–1.08)$ were obtained for $\mathsf{F}^{-}$ , $\mathtt{N a}^{+}$ , ${\mathrm{Mg}}^{2+}$ and ${\mathsf{C}}{\mathsf{a}}^{2+}$ with relatively low concentrations. $\Nu0_{2}^{-}$ showed no correlation and far higher concentration measured by AIM than that by traditional filter-based measurement (RMA slope $:=42.90\$ ; $R^{2}=0.081$ ), and this difference could be due to the loss of $\mathrm{NO}_{2}^{-}$ from the filters (Chang et al., 2007; Zhang et al., 2007a). Overall, AIM worked well for measuring major water-soluble ions in $\mathsf{P M}_{2.5}$ .
3.1.2. Mass concentrations
The hourly mean concentrations and standard deviations of water-soluble ions in $\mathsf{P M}_{2.5}$ are summarized in Table 1. The concentrations of water-soluble ions followed the order of $\mathrm{SO_{4}^{2-}>N H_{4}^{+}>N O_{3}^{-}>C l^{-}\sim N O_{2}^{-}\sim K^{+}>N a^{+}>C a^{2+}>F^{-}>M g^{2+}}$ and this order changed slightly with seasons. $50_{4}^{2-}$ , $\mathsf{N H}_{4}^{+}$ and $\Nu0_{3}^{-}$ were the dominant ions, and contributed more than $80\%$ to the total measured water-soluble ions. $S0_{4}^{2-}$ was the most abundant watersoluble ion in Jinan and its annual mean concentration was $38.33\pm26.20~{\upmu\mathrm{g}}~{\mathrm{m}}^{-3}$ , accounting for $44.65\pm11.30\%$ of the total measured water-soluble ions. It is worth noting that the concentration of $50_{4}^{2-}$ alone was more than twice the annual US National
Ambient Air Quality Standards of $\mathsf{P M}_{2.5}$ $15\;\upmu\mathrm{g}\;\mathrm{m}^{-3},$ and the hourly maximum concentration of $S0_{4}^{2-}$ could be up to $227\ \upmu\mathrm{g}\ \mathrm{m}^{-3}$ . $\mathrm{NH_{4}^{+}}$ $(21.16\,\pm\,16.28\,\upmu\mathrm{g}\,\mathfrak{m}^{-3})$ and $\Nu0_{3}^{-}$ $15.77\,\pm\,12.06\,\mathrm{\ug\,\m^{-3}}]$ were another major components, accounting for $17.63~\pm~7.61\%$ and $23.07\pm5.85\%$ of the total water-soluble ions respectively.
Table 2 compares the concentrations of $50_{4}^{2-}$ , $\mathrm{NH}_{4}^{+}$ and $\Nu0_{3}^{-}$ in $\mathsf{P M}_{2.5}$ in Jinan with those measured in other cities over the world. Obviously, the levels of these compounds in Jinan were substantially (5e10 times) higher than those in cities of USA, Europe, Japan and Korea. Moreover, they were also higher than those of other Chinese cities (e.g. Beijing, Shanghai), which are well-known to suffer serious aerosol pollution. These results demonstrated the severity of secondary inorganic aerosol pollution in Jinan.
3.2. Temporal variations of water-soluble ions in $P M_{2.5}$
3.2.1. Seasonal variations of major water-soluble ions
Different seasonal variations were observed for individual ion due to their differences in emission sources and formation mechanisms (Table 1). Secondary ions, namely $S0_{4}^{2-}$ , $\Nu0_{3}^{-}$ , $\mathrm{NH_{4}^{+}}$ and $\mathsf{N O}_{2}^{-}$ showed higher values in summer $(64.27\,\pm\,31.00$ , $19.22\,\pm\,11.84.$ $28.01\pm16.27$ and $2.93\pm2.45~\upmu\mathrm{g}~\mathrm{m}^{-3};$ and winter $(42.84\pm31.72$ $21.77\,\pm\,15.05$ , $29.19\,\pm\,20.72$ and $2.48\pm1.94~\upmu\mathrm{g}~\mathrm{m}^{-3}.$ , and lower levels in spring $(27.11~\pm~11.41\$ , $10.19\,\pm\,6.00$ , $13.28~\pm~8.75$ and $1.50\,\pm\,1.27\,\upmu\mathrm{g}\,\:\mathrm{m}^{-3})$ and autumn $(30.99\,\pm\,14.15$ , $11.69\,\pm\,7.33$ $15.13\pm7.36$ and $1.75\pm1.61~\upmu\mathrm{g}~\mathrm{m}^{-3}.$ . The summertime peak could be attributed to more active photochemistry process which can facilitate formation of secondary species, while the higher levels in winter may associate with huge emissions of primary pollutants (such as $S0_{2}$ and $\Nu0_{x_{\star}}$ ) from coal combustion for heating and worsen atmospheric dispersion.
The fine sulfate and nitrate oxidation ratios (SOR and NOR) are defined as $50\mathrm{R}\,=\,\mathrm{nSO}_{4}^{2-}/(\mathrm{nSO}_{4}^{2-}+\mathrm{nSO}_{2})$ and $\mathsf{N O R}\,=\,\mathsf{n N O}_{3}^{-}/$ $(\mathrm{nNO}_{3}^{-}+\mathrm{nNO}_{x})$ Þ to indicate the process and extent of formations from $S0_{2}$ to $S0_{4}^{2-}$ and $\Nu0_{x}$ to $\Nu0_{3}^{-}$ (Wang et al., 2005). SOR and NOR are represented in Fig. 2 and their average values were both larger than 0.10, reflecting occurrence of secondary formation in Jinan (Wang et al., 2005). SOR in summer was $0.47\pm0.13$ , much larger than that in spring $(0.22\pm0.05)$ , autumn $(0.30\pm0.04)$ and winter $(0.17~\pm~0.02)$ , indicating stronger oxidation of $S0_{2}$ to $S0_{4}^{2-}$ in summer leading to the highest concentration of $50_{4}^{2-}$ in spite of relatively lower $S0_{2}$ concentrations among four seasons ( $\mathrm{{SO}_{2}}$ : Summer $=\ 26.25\ \pm\ 28.73$ ppb; Spring $=\;32.14\;\pm\;26.86$ ppb; Autumn $=22.31\pm16.41$ ppb; Winter $=58.59\pm32.98~\mathrm{ppb}$ ). The formation of $S0_{4}^{2-}$ from $S0_{2}$ mainly includes gas-phase reaction of $S0_{2}$ and OH radical affected by temperature and solar radiation (Seinfeld, 1986), and heterogeneous reaction which is a function of RH (metal catalyzed oxidation or $\mathrm{H}_{2}0_{2}/0_{3}$ oxidation) (Dlugi et al., 1981). The seasonal variation of SOR was consistent with temperature in Fig. 2, indicating that gas-phase oxidation of $S0_{2}$ played a major role in the formation of $\bar{\mathrm{S}}0_{4}^{2-}$ in the whole year (Wang et al., 2005). NOR showed the highest level in summer $(0.28\pm0.03)$ , the lowest level in winter $(0.12\pm0.01)$ , and comparable level in spring $(0.14~\pm~0.01)$ and autumn $(0.14~\pm~0.01)$ , indicating that high temperature and high RH promoted the faster formation of $\Nu0_{3}^{-}$ in spite of more dissociation of $\mathsf{N H}_{4}\mathsf{N O}_{3}$ at high temperature in summer.
The mass ratio of $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ has been used as an indicator of relative importance of mobile (e.g. vehicles) vs. stationary sources (e.g. power plant) in the air pollution (Yao et al., 2002; Wang et al., 2006). High $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ mass ratios have been measured in southern California, with 2 in downtown Los Angeles and 5 in
Rubidoux, which was due to less use of coal (Kim et al., 2000); However in Chinese cities (e.g. Beijing, Shanghai), lower ratios had been reported as a result of the wide use of sulfur-containing coal (Yao et al., 2002). In our study, the $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ mass ratio ranged from 0.03 to 1.52, with the annual mean of 0.44. These results indicated that like other cities in China, stationary sources were more important compared with vehicle emissions in Jinan. The $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ mass ratios showed clear seasonal variations with the highest ratio in winter $(0.53\,\pm\,0.05)$ , the lowest ratio in summer $(0.34\,\pm\,0.03)$ , and comparable ratio in spring $(0.40\,\pm\,0.06)$ and autumn $[0.42\pm0.06)$ (see Fig. 2). The reason is that in summer high temperature and high RH are more favorable for formation of $\mathrm{S}0_{4}^{\bar{2}-}$ compared to $\Nu0_{3}^{-}$ due to enhanced evaporation of $\mathsf{N H}_{4}\mathsf{N O}_{3}$ at high temperature.
$C1^{-}$ showed higher concentrations in winter $(8.75~\pm~6.37$ $\upmu\mathrm{g}\:\:\mathrm{m}^{-3})$ than that in other seasons (Table 1). $C1^{-}$ is a major component of sea-salt particle, and is also released from coal combustion (Sun et al., 2006). In winter, the air generally originated from the northwest (Xu et al., 2010) and thus the influence of seasalt would be minor. The mass ratio of $\mathrm{Cl}^{-}/\mathrm{N}\mathbf{a}^{+}$ was calculated as $6.18\pm2.96$ in winter, which is much higher than that detected for sea water (1.797) (Moller, 1990). Therefore, the elevated concentration of $C1^{-}$ in winter was due to the enhanced coal combustion. $K^{+}$ exhibited higher levels in summer $(4.62\,\pm\,3.08~\upmu\mathrm{g}~\mathrm{m}^{-3})$ and winter $(3.07\,\pm\,2.54\,\mathrm{\upmug}\,\mathrm{\m}^{-3})$ than that in spring and autumn $(1.32\pm1.14$ and $1.44\pm1.17\;\upmu\mathrm{g}\;\mathrm{m}^{-3};$ . In summer, extensive activities of biomass burning around Shandong was the main factor contributing to the elevated concentration of $\mathrm{K^{+}\,(h t t p://m a p s.g e o g_{\ast}}$ umd.edu/firms/). High concentrations of $K^{+}$ in winter may be associated with coal combustion as implied from the strong correlation ${\it r}=0.821$ ) between $\mathsf{K}^{+}$ with $C1^{-}$ (Westberg et al., 2003). Other water-soluble ions were not relevant for seasonal variations due to their low concentrations.
3.2.2. Diurnal variations of secondary water-soluble ions The diurnal variations of secondary water-soluble ions in $\mathsf{P M}_{2.5}$ in four seasons are shown in Fig. 3. In general, $50_{4}^{2-}$ exhibited similar diurnal profiles in spring, summer and autumn, with an evident increase as sun rising and a broad daytime maximum, which was consistent with those reported in other cities (e.g. Beijing, PRD) (Hu et al., 2008; Wu et al., 2009). This typical pattern can be explained by the fact that photochemical production is more extensive during the daytime with stronger solar radiation. Compared with other seasons, in winter $S0_{4}^{2-}$ had a little diurnal variation and showed two peaks in the morning and evening, which may be related with boundary layer height and photochemical reaction. The morning peak may be contributed to enhanced photochemical production and the evening peak was a result of accumulation of pollutants with reduced boundary layer height. The lower concentration in the early afternoon was because that the dilution of $50_{4}^{2-}$ by increase of boundary layer overwhelmed the production of $\dot{\bf S}0_{4}^{2-}$ caused by photochemical reactions.
$\Nu0_{3}^{-}$ had a more apparent diurnal profile in summer and autumn than that in spring and winter. In summer and autumn, $\Nu0_{3}^{-}$ peaked in the morning, and its lowest concentration appeared at 16:00e18:00 (Hu et al., 2008; Wu et al., 2009). The morning peak was synchronous with $\Nu0_{x}$ , indicating its relation to vehicle emissions (Park et al., 2005). The lowest concentration of $\Nu0_{3}^{-}$ in the afternoon was attributed to dissociation of $\mathrm{NH}_{4}\mathrm{NO}_{3}$ at high temperature and increase of boundary layer height. In winter and spring $\Nu0_{3}^{-}$ showed a little diurnal pattern, which was due to minor influence of thermodynamic equilibrium at low temperature. $\Nu0_{3}^{-}$ showed two peaks in the morning and evening in spring which was consistent with that of $\Nu0_{x}$ and may be associated with vehicle emissions (Park et al., 2005). In four seasons $\mathrm{NH_{4}^{+}}$ showed similar diurnal profiles with $50_{4}^{2-}$ or $\Nu0_{3}^{-}$ , indicating the existences of $(\mathrm{NH}_{4})_{2}\mathrm{SO}_{4}$ and $\mathsf{N H}_{4}\mathsf{N O}_{3}$ .
$\Nu0_{2}^{-}$ showed similar diurnal profiles in four seasons with higher concentrations at night and lower concentrations during daytime (Fig. 3), and similar profile was also observed in Beijing (Zhang et al., $2007{\bf a}_{.}$ . Higher $\mathrm{NO}_{2}^{-}$ concentration at night than that during daytime probably was related with lower oxidized extent due to weak solar radiation and nighttime accumulation of $\mathrm{NO}_{2}^{-}$ .
3.3. Sources
3.3.1. The potential source regions identification using trajectory statistical methods
Backward trajectory cluster analysis is a useful tool to evaluate the origins of air pollutants at the receptor sites (Salvador et al., 2010), and redistributed concentration field (RCF) model can estimate the potential source regions (Stohl, 1996). The combination of backward trajectory cluster analysis and RCF model can be better to provide a comprehensive view of the potential source regions for $\bar{\mathrm{S}}0_{4}^{2-}$ and $\Nu0_{3}^{-}$ in $\mathsf{P M}_{2.5}$ in Jinan. Three-day mean trajectories for clusters in spring, summer, autumn and winter and corresponding mean concentrations of water-soluble ions are expressed in Fig. 4. All the trajectories in four seasons can be classified into 4 main categories based on their origins, paths and latitudes: (1) the shortest/local transport pattern, (2) eastern airflow, (3) northeast air masses and (4) northwest/north air parcel with long transport path. From Fig. 4, it can be seen that the shortest/local transport pattern (cluster 1) was frequent and accounted for $53\%$ of total trajectories in spring, $49\%$ in summer, $60\%$ in autumn and $62\%$ in winter. Eastern airflow was dominant in summer and contributed for $21\%$ of total trajectories in spring (cluster 2). Northeast/north air masses were observed in spring (cluster 3, $14\%$ and in autumn (cluster 2, $17\%$ . Other clusters generally originated from northwest/ north of China and traveled fast at the highest altitude.
The highest concentrations of $\mathrm{SO}_{4}^{2-}$ , $\Nu0_{3}^{-}$ and $\mathrm{NH_{4}^{+}}$ were observed in the shortest cluster (cluster 1) in four seasons, indicating that secondary ions in $\mathsf{P M}_{2.5}$ were easy to be enriched in the short trajectories from upwind regional and local emission sources (Karaca and Camci, 2010). Cluster 1 generally originated from the middle of Shandong Province, moved southerly and finally turned westerly to Jinan in spring, autumn and winter, while in summer it derived from the Yellow Sea, moved northwesterly to Jinan. A major big petro-chemical corporation, power plants and cement production base are located in the middle (Zibo city) and southwest (Jining city and Zaozhuang city) of Shandong Province. Cluster 1 spent much time on passing over industrial zones with high emissions of primary pollutants (e.g. $S0_{2}$ , $\Nu0_{x,}$ ), leading to the high concentrations of $\bar{\mathrm{S}0_{4}^{2-}}$ , $\Nu0_{3}^{-}$ and $\mathsf{N H}_{4}^{+}$ .
Much higher concentrations of $\dot{\mathrm{S}}0_{4}^{2-}$ , $\Nu0_{3}^{-}$ and $\mathrm{NH}_{4}^{+}$ were associated with air masses from northeast or northwest, including cluster 3 in spring, clusters 2 in autumn and cluster 2 in winter. The flow patterns of cluster 3 in spring and cluster 2 in autumn were both typically originated from Inner Mongolia, flowed over Liaoning Province and Bohai Gulf before arriving at Jinan. While cluster 2 in winter derived from Outer Mongolia, passed through Inner Mongolia, Hebei Province and then to Jinan. These trajectories all passed over the Bohai economic zone, which is one of the most populated and industrial zones in China and has the highest $S0_{2}$ and $\Nu0_{x}$ emissions (Zhang et al., 2009).
Low concentrations of $50_{4}^{2-}$ , $\Nu0_{3}^{-}$ and $\mathrm{NH}_{4}^{+}$ occurred in cluster 4 and 5 in spring, autumn and winter. These clusters derived from northwest of China and moved faster at the higher latitudes compared to other clusters.
Compared to other seasons, in summer higher concentrations of $50_{4}^{2-}$ , $\Nu0_{3}^{-}$ and $\mathrm{NH}_{4}^{+}$ could be explained by the combination of Dimethyl Sulfide (DMS) oxidation (Salvador et al., 2010) and long residence time over industrial zones with large amount emissions of primary pollutants. It was worth noting that there was also a significant fraction of the flow from Korea in summer (cluster 4 and 5, $10\%$ ), which passed westerly over the Yellow Sea and Jiaodong Peninsula before arriving at Jinan.
Figs. 5 and 6 show the results of RCF analysis for $50_{4}^{2-}$ and $\Nu0_{3}^{-}$ , respectively. The high potential source region of $50_{4}^{2-}$ and $\Nu0_{3}^{-}$ was Shandong, and Hebei, Henan, Anhui, Jiangsu, Liaoning, and Inner Mongolia in spring, autumn and winter and eastern Jiangsu Province, South Korea and the Yellow Sea in summer were also identified as the potential source regions.
3.3.2. The sources identification by principal component analysis (PCA)
In order to identify the sources of water-soluble ions in $\mathsf{P M}_{2.5}$ principal component analysis (PCA) was applied. PCA is a widely used statistical technique to quantitatively identify a smaller number of independent factors among the compound concentrations, which can explain the variance of the data, by using the eigenvector decomposition of a matrix of pair-wise correlations (Johnson and Wichern, 1998; Miller et al., 2002). PCA is conducted using a commercially available software package (SPSS). Hourly values of Nss- $S0_{4}^{2-}$ , $\Nu0_{3}^{-}$ , $\mathsf{N H}_{4}^{+}$ , $C1^{-}$ , $\Nu0_{2}^{-}$ , $\mathsf{N a}^{+}$ , $\mathsf{K}^{+}$ , $\mathrm{Mg}^{2+}$ , $C a^{2+}$ , $\mathsf{N O}_{x},$ $S0_{2}$ , CO, BC and $0_{3}$ were used for PCA and the results are shown in Table 3.
In spring, four principal components were obtained and accounted for $82\%$ of the total variance. The principal component 1 accounted for $45\%$ of the total variance, and comprised $\Nu0_{x},\Nu0_{2}^{-}$ , BC and CO, while anti-correlated with $0_{3}$ , indicating its relation to traffic emissions. The principal component 2 could be explained by secondary aerosols due to the positive contribution from Nss- $S0_{4}^{2-}$ , $\Nu0_{3}^{-}$ and $\mathrm{NH}_{4}^{+}$ , and this factor accounted for $15\%$ of the total variance. The principal component 3 contained $C1^{-}$ and $S0_{2}$ with $13\%$ of the total variance, which was related with stationary source emissions such as coal combustion. The principal component 4 was primarily composed of ${\mathsf{C}}{\mathsf{a}}^{2+}$ , ${\mathrm{Mg}}^{2+}$ and $\mathtt{N a}^{+}$ , and was attributed to the crustal and soil dust from urban constructions.
In summer, three principal components were identified and accounted for $73\%$ of the totalvariance. The principal component 1 had highly positive contributions from $\mathsf{N S S}^{-\S0_{4}^{2-}}$ , $\Nu0_{3}^{-}$ , $\mathrm{NH_{4}^{+}}$ , $C1^{-}$ , ${\bf K}^{+}$ , BC and CO, which could be explained by secondary aerosols mixing with biomass burning. In summer, extensive activities of biomass burning around Shandong were observed (http://maps.geog.umd.edu/firms/) and biomass burning is likely to emit large amount of $C1^{-}$ , $\mathsf{K}^{+}$ , BC and CO. The principal component 2 was composed of $\mathsf{N O}_{x},$ , $\Nu0_{2}^{-}$ , ${\cal{C}}a^{2+}$ and ${\mathrm{Mg}}^{2+}$ , and anti-correlated with $0_{3}$ , which was identified as traffic emissions mixing with crustal and soil dust. The principal component 3 was composed of $S0_{2}$ , mainly from coal combustion.
In autumn, three principal components were identified and accounted for $69\%$ of the total variance. The principal component 1 accounted for $39\%$ of the total variance, which could be explained by secondary aerosols due to the positive contribution from $\bar{\mathrm{Nss}}{-}50_{4}^{2-}\mathrm{NO}_{3}^{\bar{-}}$ and $\mathsf{N H}_{4}^{+}$ . The principal component 2 was composed of $\Nu0_{x},$ , $\Nu0_{2}^{-}$ , CO and BC, and anti-correlated with $0_{3}$ , indicating its relation to traffic emissions which accounted for $18\%$ of the total variance. The principal component 3 was primarily composed of ${\mathsf{C a}}^{2+}$ , ${\mathrm{Mg}}^{2+}$ and $\mathtt{N a}^{+}$ , and was attributed to the crustal and soil dust from urban constructions.
In winter, three principal components also were obtained. The principal component 1 could be identified as secondary aerosols mixing with coal combustion due to the positive combustion from $\mathsf{N s s}\!-\!\mathsf{S}0_{4}^{2-}$ , $\Nu0_{3}^{-}$ , $\mathrm{NH}_{4}^{+}$ , $C1^{-}$ , $\Nu0_{2}^{-}$ , $\mathsf{N a}^{+}$ , $\mathsf{K}^{+}$ , BC and CO. The principal components 2 and 3 were composed of primary pollutants (e.g. $S0_{2}$ , $\Nu0_{x}$ , ${\mathsf{C}}{\mathsf{a}}^{2+}$ , $\mathrm{Mg}^{2+}.$ , mainly from coal combustion, crustal and soil dust and traffic emissions. These components accounted for $75\%$ of the total variance in winter.
All in all, the source apportionment indicated that secondary aerosols, coal/biomass burnings and traffic emissions were major contributors for $\mathsf{P M}_{2.5}$ loading in Jinan.
4. Summary
Hourly concentrations of water-soluble ions in $\mathsf{P M}_{2.5}$ were measured to investigate secondary aerosol pollution and potential source regions in Jinan, Shandong Province from Dec 2007 to Oct 2008.
The results verified that Jinan was suffering more serious fine particle pollution compared with other cities in the world. Accelerated photochemistry reaction under high temperature, $0_{3}$ concentrations and strong solar radiation in summer and high emissions of $S0_{2}$ and $\Nu0_{x}$ from coal combustion in winter led to higher concentrations of $50_{4}^{2-}$ , $\Nu0_{3}^{-}$ and $\mathrm{NH}_{4}^{+}$ in these two seasons than those in spring and autumn. The diurnal variations of $50_{4}^{2-}$ in spring, autumn and winter were dominated by photochemical process, while in winter controlled by boundary layer height and photochemical reaction. In summer and autumn thermodynamic reaction affected diurnal variation of $\Nu0_{3}^{-}$ , while it rarely occurred in spring and winter with a little diurnal variation. Production of secondary inorganic aerosol was more extensive in summer implied by higher SOR and NOR compared to other seasons.
Cluster analysis showed that the synoptic flows arriving at Jinan were dominated by the air masses originating and circulating locally in Shandong Province in spring, autumn and winter, while originating from the Yellow Sea in summer. RCF results indicated that the major potential source regions for secondary ions were concentrated in Shandong and partly from the provinces of Hebei, Henan, Anhui, Jiangsu, and Liaoning, as well as Inner Mongolia in spring, autumn and winter, while in summer Shandong, eastern Jiangsu, South Korea and the Yellow Sea were identified as the main potential source regions. Secondary aerosol dominated the variations of aerosol loading. Traffic emissions and coal combustions were major contributors for urban pollution. The influence of biomass burnings in summer was observed in Jinan.
Acknowledgments
This work was supported by the National Basic Research Program (973 Program) of China (2005CB422203), a key project of Shandong Provincial Environmental Agency (2006045), Promotive Research Fund for Young and Middle-aged Scientists of Shandong Province (BS2010HZ010) and Independent Innovation Foundation of Shandong University (2009TS024).
References
Baldasano, J.M., Valera, E., Jiménez, P., 2003. Air quality data from large cities. Science of Total Environment 307 (1e3), 141e165.
Chan, C.K., Yao, X.H., 2008. Air pollution in mega cities in China. Atmospheric Environment 42 (1), 1e42.
Chang, S.Y., Lee, C.T., Chou, C.C.K., Liu, S.C., Wen, T.X., 2007. The continuous field measurements of soluble aerosol compositions at the Taipei Aerosol Supersite, Taiwan. Atmospheric Environment 41 (9), 1936e1949.
Chen, B.H., Hong, C.J., Kan, H.D., 2004. Exposures and health outcomes from outdoor air pollutants in China. Toxicology 198 (1e3), 291e300.
Dlugi, R., Jordan, S., Lindemann, E., 1981. The heterogeneous formation of sulfate aerosols in the atmosphere. Journal of Aerosol Science 12, 185e197.
Draxler, R.R., Rolph, G.D., 2003. HYSPLIT (Hybrid Single-Particle Lagrangian Integrated Trajectory) Model Access via NOAA ARL READY Websiter. NOAA Air Resources Laboratory, Silver Spring, MD. Available at: http://www.arl.noaa.gov/ HYSPLIT.php (accessed 2009).
Fang, G.C., Chang, C.N., Wu, Y.S., Fu, P.P.C., Yang, C.J., Chen, C.D., Chang, S.C., 2002. Ambient suspended particulate matters and related chemical species study in central Taiwan, Taichung during 1998e2001. Atmospheric Environment 36 (12), 1921e1928.
Heo, J.B., Hopke, P.K., Yi, S.M., 2009. Source apportionment of $\mathrm{PM}_{2.5}$ in Seoul, Korea. Atmospheric Chemistry and Physics 9, 4957e4971.
Hu, M., Wu, Z.J., Slanina, J., Lin, P., Liu, S., Zeng, L.M., 2008. Acidic gases, ammonia and water-soluble ions in $\mathrm{PM}_{2.5}$ at a coastal site in the Pearl River Delta, China. Atmospheric Environment 42 (25), 6310e6320.
Hueglin, C., Gehrig, R., Baltensperger, U., Gysel, M., Monn, C., Vonmont, H., 2005. Chemical characterisation of $\mathrm{PM}_{2.5}$ , $\mathsf{P M}_{10}$ and coarse particles at urban, nearcity and rural sites in Switzerland. Atmospheric Environment 39 (4), 637e651.
Hughes, L.S., Cass, G.G., Gone, J., Ames, M., Olmez, I., 1998. Physical and chemical characterization of atmospheric ultrafine particles in the Los Angeles area. Environmental Science & Technology 32 (9), 1153e1161.
Hu, M., He, L.Y., Zhang, Y.H., Wang, M., Kim, Y.P., Moon, K.C., 2002. Seasonal variation of ionic species in fine particles at Qingdao, China. Atmospheric Environment 36 (38), 5853e5859.
Johnson, R.A., Wichern, D., 1998. Applied Multivariate Statistical Analysis. Prentice Hall, Englewood Cliffs, NJ.
Karaca, F., Camci, F., 2010. Distant source contributions to $\mathrm{PM}_{10}$ profile evaluated by SOM based cluster analysis of air mass trajectory sets. Atmospheric Environment 44 (7), 892e899.
Khan, M.F., Shirasuna, Y., Hirano, K., Masunaga, S., 2010. Characterization of $\operatorname{PM}_{2.5},$ $\mathrm{PM}_{2.5-10}$ and $\mathrm{PM}_{>10}$ in ambient air, Yokohama, Japan. Atmospheric Research 96 (1), 159e172.
Kim, B.M., Teffera, S., Zeldin, M.D., 2000. Characterization of $\mathrm{PM}_{2.5}$ and $\mathrm{PM}_{10}$ in the South Coast Air Basin of Southern California: part 1 e spatial variations. Journal of the Air and Waste Management Association 50 (12), 2034e2044.
Kim, H.S., Huh, J.B., Hopke, P.K., Holsen, T.M., Yi, S.M., 2007. Characteristics of the major chemical constituents of $\mathrm{PM}_{2.5}$ and smog events in Seoul, Korea in 2003 and 2004. Atmospheric Environment 41 (32), 6762e6770.
Lee, J.H., Hopke, P.K., 2006. Apportioning sources of $\mathsf{P M}_{2.5}$ in St. Louis, MO using speciation trends network data. Atmospheric Environment 40 (2), 360e377.
Lonati, G., Giugliano, M., Butelli, P., Romele, L., Tardivo, R., 2005. Major chemical components of $\mathrm{PM}_{2.5}$ in Milan (Italy). Atmospheric Environment 39 (10), 1925e1934.
Louie, P.K.K., Watson, J.G., Chow, J.C., Chen, A., Sin, D.W.M., Lau, A.K.H., 2005. Seasonal characteristics and regional transport of $\mathrm{PM}_{2.5}$ in Hong Kong. Atmospheric Environment 39 (9), 1695e1710.
Moller, D., 1990. The $\mathrm{Na}/\mathrm{Cl}$ ration in rainwater and the seasalt chloride cycle. Tellus B 423, 254e262.
Miller, S.L., Anderson, M.J., Daly, E.P., Milford, J.B., 2002. Source apportionment of exposures to volatile organic compounds. I. Evaluation of receptor models using simulated exposure data. Atmospheric Environment 36 (22), 3629e3641.
National Bureau of Statistics of China, 2009. China Statistical Yearbook. China Statistics Press, Beijing (in Chinese).
Ocskay, R., Salma, I., Wang, W., Maenhaut, W., 2006. Characterization and diurnal variation of size-resolved inorganic water-soluble ions at a rural background site. Journal of Environmental Monitoring 8 (2), 300e306.
Park, S.S., Ondov, J.M., Harrison, D., Nair, N.P., 2005. Seasonal and shorter-term variations in particulate atmospheric nitrate in Baltimore. Atmospheric Environment 39 (11), 2011e2020.
Qin, Y.J., Kim, E., Hopke, P.K., 2006. The concentrations and sources of $\mathrm{PM}_{2.5}$ in metropolitan New York City. Atmospheric Environment 40 (2), 312e332.
Querol, X., Alastuey, A., Moreno, T., Viana, M.M., Castillo, S., Pey, J., Rodríguez, S., Artiñano, B., Salvador, P., Sánchez, M., Garcia Dos Santos, S., Herce Garraleta, M.D., Fernandez-Patier, R., Moreno-Grau, S., Negral, L., Minguillón, M.C., Monfort, E., Sanz, M.J., Palomo-Marín, R., Pinilla-Gil, E., Cuevas, E., de la Rosa, J., Sánchez de la Campa, A., 2008. Spatial and temporal variations in airborne particulate matter $\mathrm{\left[PM_{10}\right.}$ and $\mathsf{P M}_{2.5},$ ) across Spain 1999e2005. Atmospheric Environment 42 (17), 3964e3979.
Salvador, P., Artíñano, B., Pio, C., Afonso, J., Legrand, M., Puxbaum, H., Hammer, S., 2010. Evaluation of aerosol sources at European high altitude background sites with trajectory statistical methods. Atmospheric Environment 44 (19), 2316e2329.
Seinfeld, J.H., 1986. Atmospheric Chemistry and Physics of Air Pollution. Wiley, New York.
Shen, Z.X., Cao, J.J., Arimoto, R., Han, Z.W., Zhang, R.J., Han, Y.M., Liu, S.X., Okuda, T., Nakao, S., Tanaka, S., 2009. Ionic composition of TSP and $\mathrm{PM}_{2.5}$ during dust storms and air pollution episodes at Xi’an, China. Atmospheric Environment 43 (18), 2911e2918.
Shou, Y.P., Gao, X.M., Wang, J., Yang, L.X., Wang, W.X., 2010. Online measurement of water-soluble ions in fine particles from the atmosphere in autumn in Jinan. Research of Environmental Sciences 23 (1), 41e47.
Sloane, C.S., Watson, J., Chow, J., Pritchett, L., Richards, L.W., 1991. Size-segregated fine particle measurements by chemical species and their impact on visibility impairment in Denver. Atmospheric Environment. Part A. General Topics 25 (5e6), 1013e1024.
Stohl, A., 1996. Trajectory statistics-a new method to establish source-receptor relationships of air pollutants and its application to the transport of particulate sulfate in Europe. Atmospheric Environment 30 (4), 579e587.
Streets, D.G., Fu, J.S., Jang, C.J., Hao, J.M., He, K.B., Tang, X.Y., Zhang, Y.H., Wang, Z.F., Li, Z.P., Zhang, Q., Wang, L.T., Wang, B.Y., Yu, C., 2007. Air quality during the 2008 Beijing Olympic Games. Atmospheric Environment 41 (3), 480e492.
Sun, Y.L., Zhuang, G.S., Tang, A.H., Wang, Y., An, Z.S., 2006. Chemical characteristics of $\mathsf{P M}_{2.5}$ and $\mathrm{PM}_{10}$ in haze-fog episodes in Beijing. Environmental Science & Technology 40 (10), 3148e3155.
Wang, T., Nie, W., Gao, J., Xue, L.K., Gao, X.M., Wang, X.F., Qiu, J., Poon, C.N., Meinardi, S., Blake, D., Wang, S.L., Ding, A.J., Chai, F.H., Zhang, Q.Z., Wang, W.X., 2010. Air quality during the 2008 Beijing Olympics: secondary pollutants and regional impact. Atmospheric Chemistry and Physics 10, 7603e7615.
Wang, Y., Zhuang, G.S., Tang, A.H., Yuan, H., Sun, Y.L., Chen, S., Zheng, A.H., 2005. The ion chemistry and the source of $\mathrm{PM}_{2.5}$ aerosol in Beijing. Atmospheric Environment 39 (21), 3771e3784.
Wang, Y., Zhuang, G.S., Zhang, X.Y., Huang, K., Xu, C., Tang, A.H., Chen, J.M., An, Z.S., 2006. The ion chemistry, seasonal cycle, and sources of $\mathrm{PM}_{2.5}$ and TSP aerosol in Shanghai. Atmospheric Environment 40 (16), 2935e2952.
Westberg, H.M., Byström, M., Leckner, B., 2003. Distribution of potassium, chlorine, and sulfur between solid and vapor phases during combustion of wood chips and coal. Energy & Fuels 17 (1), 18e28.
Wu, W.S., Wang, T., 2007. On the performance of a semi-continuous $\mathrm{PM}_{2.5}$ sulphate and nitrate instrument under high loadings of particulate and sulphur dioxide. Atmospheric Environment 41 (26), 5442e5451.
Wu, Z.J., Hu, M., Shao, K.S., Slanina, J., 2009. Acidic gases, $\mathrm{NH}_{3}$ and secondary inorganic ions in $\mathrm{PM}_{10}$ during summertime in Beijing, China and their relation to air mass history. Chemosphere 76 (8), 1028e1035.
Xu, J., Bergin, M.H., Yu, X., Liu, G., Zhao, J., Carrico, C.M., Baumann, K., 2002. Measurement of aerosol chemical, physical and radiative properties in the Yangtze delta region of China. Atmospheric Environment 36 (2), 161e173.
Xu, P.J., Wang, W.X., Yang, L.X., Zhang, Q.Z., Gao, R., Wang, X.F., Nie, W., Gao, X.M., 2010. Aerosol size distributions in urban Jinan: seasonal characteristics and variations between weekdays and weekends in a heavily polluted atmosphere. Environmental Monitoring and Assessment. doi:10.1007/s10661-010-1747-2.
Yang, L.X., Wang, D.C., Cheng, S.H., Wang, Z., Zhou, Y., Zhou, X.H., Wang, W.X., 2007. Influence of meteorological conditions and particulate matter on visual range impairment in Jinan, China. Science of Total Environment 383 (1e3), 164e173.
Yao, X.H., Chan, C.K., Fang, M., Cadle, S., Chan, T., Mulawa, P., He, K.B., Ye, B., 2002. The water-soluble ionic composition of $\mathrm{PM}_{2.5}$ in Shanghai and Beijing, China. Atmospheric Environment 36 (26), 4223e4234.
Zhang, K., Wang, Y.S., Wen, T.X., Liu, G.R., Xu, H.H., 2007a. On-line analysis the water-soluble chemical of $\mathrm{PM}_{2.5}$ in late summer and early autumn in Beijing. Acta Scientiae Circumstantiae 27 (3), 459e465.
Zhang, Q., Jimenez, J.L., Canagaratna, M.R., Allan, J.D., Coe, H., Ulbrich, I., Alfarra, M.R., Takami, A., Middlebrook, A.M., Sun, Y.L., Dzepina, K., Dunlea, E., Docherty, K., DeCarlo, P.F., Salcedo, D., Onasch, T., Jayne, J.T., Miyoshi, T., Shimono, A., Hatakeyama, S., Takegawa, N., Kondo, Y., Schneider, J., Drewnick, F., Borrmann, S., Weimer, S., Demerjian, K., Williams, P., Bower, K., Bahreini, R., Cottrell, L., Griffin, R.J., Rautiainen, J., Sun, J.Y., Zhang, Y.M., Worsnop, D.R., 2007b. Ubiquity and dominance of oxygenated species in organic aerosols in anthropogenicallyinfluenced Northern Hemisphere midlatitudes. Geophysical Research Letters 34, L13801. doi:10.1029/2007GL029979.
Zhang, Q., Streets, D.G., Carmichael, G.R., He, K.B., Huo, H., Kannari, A., Klimont, Z., Park, I.S., Reddy, S., Fu, J.S., Chen, D., Duan, L., Lei, Y., Wang, L.T., Yao, Z.L., 2009. Asian emissions in 2006 for the NASA INTEX-B mission. Atmospheric Chemistry and Physics 9, 5131e5153.
Zhou, X.H., Cao, J., Wang, T., Wu, W.S., Wang, W.X., 2009. Measurement of black carbon aerosols near two Chinese megacities and the implications for improving emission inventories. Atmospheric Environment 43 (25), 3918e3924.
Zhou, Y., Wang, T., Gao, X.M., Xue, L.K., Wang, X.F., Wang, Z., Gao, J., Zhang, Q.Z., Wang, W.X., 2010. Continuous observations of water-soluble ions in $\mathrm{PM}_{2.5}$ at Mount Tai (1534 ma.s.l.) in central-eastern China. Journal of Atmospheric Chemistry. doi:10.1007/s10874-010-9172-z.
|
atmosphere
|
0021
|
10.1016/j.partic.2012.08.004
|
image
|
Fig. 1. Location of the sampling site in Zhengzhou, Henan, China.
|
|
atmosphere
|
0021
|
10.1016/j.partic.2012.08.004
|
table
|
Table 1 $\mathrm{PM}_{2.5}$ concentration and its chemical composition, collected in 2010.
|
|
atmosphere
|
0021
|
10.1016/j.partic.2012.08.004
|
image
|
Fig. 2. Variation of $\mathrm{PM}_{2.5}$ mass concentration during the sampling period in Zhengzhou, China.
|
|
atmosphere
|
0021
|
10.1016/j.partic.2012.08.004
|
image
|
Fig. 3. Seasonal variations of concentrations $\langle\mathrm{in}\,\upmu\mathrm{g}/\mathrm{m}^{3}\rangle$ ) of $\mathrm{PM}_{2.5}$ (a), total elements (b), TC (c), and total soluble ions (d), with error bars representing standard deviation.
|
|
atmosphere
|
0021
|
10.1016/j.partic.2012.08.004
|
image
|
Fig. 4. Seasonal variations of contributions of total components (a), total elements (b), TC (c) and total ions (d) to $\mathrm{PM}_{2.5}\,(\%)$ , with error bars representing standard deviation.
|
|
atmosphere
|
0021
|
10.1016/j.partic.2012.08.004
|
image
|
Fig. 5. Seasonal variations of (a) water-soluble ions, EC, OC and (b) elements in $\mathrm{PM}_{2.5}$ (in $\upmu\mathrm{g}/\mathfrak{m}^{3}$ , with error bars representing standard deviations.
|
|
atmosphere
|
0021
|
10.1016/j.partic.2012.08.004
|
table
|
Table 2 Average concentrations of $\mathrm{PM}_{2.5}$ , OC and EC in Zhengzhou, Shenzhen, Beijing, Guangzhou and Shanghai, all using NIOSH TOT method.
|
|
atmosphere
|
0021
|
10.1016/j.partic.2012.08.004
|
image
|
Fig. 6. Proflies of factors (contributions of species, $\%$ ). Factor 1: industrial; Factor 2: soil dust; Factor 3: secondary aerosol; Factor 4: biomass burning plus incineration; Factor 5: vehicle and Factor 6: coal combustion.
|
|
atmosphere
|
0021
|
10.1016/j.partic.2012.08.004
|
table
|
Table3 PMF source contribution estimates (SCE) $(\upmu\mathrm{g}/\uppi^{3}$ and percentage) for $\mathrm{PM}_{2.5}$ .
|
|
atmosphere
|
0021
|
10.1016/j.partic.2012.08.004
|
text
| Not supported with pagination yet
|
$\mathsf{P M}_{2.5}$ in an industrial district of Zhengzhou, China: Chemical composition and source apportionment
Ningbo Geng a, Jia Wang a, Yifei Xu a, Wending Zhang a, Chun Chen b, Ruiqin Zhang a,∗
a Research Institute of Environmental Science, Department of Chemistry, Zhengzhou University, Zhengzhou 450001, China b Henan Environmental Monitoring Center, Zhengzhou 450004, Henan, China
article info
a b s t r a c t
Article history: Received 17 December 2011 Received in revised form 23 July 2012 Accepted 22 August 2012
Keywords:
$\mathrm{PM}_{2.5}$
Component
Seasonal variation
PMF
Source apportionment
Zhengzhou is a developing city in China, that is heavily polluted by high levels of particulate matter. In this study, fine particulate matter $(\mathsf{P M}_{2.5})$ was collected and analyzed for their chemical composition (soluble ions, elements, elemental carbon (EC) and organic carbon (OC)) in an industrial district of Zhengzhou in 2010. The average concentrations of $\mathrm{PM}_{2.5}$ were 181, 122, 186 and $211\,\upmu\mathrm{g}/\mathrm{m}^{3}$ for spring, summer, autumn and winter, respectively, with an annual average of $175\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , far exceeding the $\mathrm{PM}_{2.5}$ regulation of USA National Air Quality Standards $(15\,\upmu\mathrm{g}/\mathrm{m}^{3})$ . The dominant components of $\mathrm{PM}_{2.5}$ in Zhengzhou were secondary ions (sulphate and nitrate) and carbon fractions. Soluble ions, total carbon and elements contributed $41\%$ , $13\%$ and $3\%$ of $\mathrm{PM}_{2.5}$ mass, respectively. Soil dust, secondary aerosol and coal combustion, each contributing about $26\%$ , $24\%$ and $23\%$ of total $\mathsf{P M}_{2.5}$ mass, were the major sources of $\mathrm{PM}_{2.5}$ , according to the result of positive matrix factorization analysis. A mixed source of biomass burning, oil combustion and incineration contributed $13\%$ of $\mathrm{PM}_{2.5}$ . Fine particulate matter arising from vehicles and industry contributed about $10\%$ and $4\%$ of $\mathrm{PM}_{2.5}$ , respectively.
$\copyright$ 2012 Chinese Society of Particuology and Institute of Process Engineering, Chinese Academy of Sciences. Published by Elsevier B.V. All rights reserved.
1. Introduction
With the rapid development of economy, anthropogenic aerosol pollutions are gradually becoming the major environmental problems. High aerosol concentration can cause a wide range of impacts on human health and also natural ecosystem, agriculture, visibility and tropospheric oxidation capacity (Callen, de la Cruz, Lopez, Navarro, & Mastral, 2009; Li et al., 2011; Tao et al., 2009). For example, the incidence of lung cancer can be increased for people exposed to high concentration of small particles produced by industry, vehicles, and power plants (Tie, Wu, & Brasseur, 2009). Particulate matters especially $\mathsf{P M}_{2.5}$ enriched with toxicants are more toxic because they are derived from specific emission sources and have large surface area available for biogenic interactions (Delfino, Sioutas, & Malik, 2005). $\mathsf{P M}_{2.5}$ can be highly variable in concentration, chemical composition, formation mechanism and origin across space and time due to the influence of meteorological conditions.
Most previous studies on $\mathsf{P M}_{2.5}$ in developing countries have focused on metropolis such as Beijing in China (Pu, Zhao, Zhang, & Ma, 2011; Song et al., 2007), New Delhi in India (Apte et al., 2011; Sahu, Beig, & Parkhi, 2011), but little or no information has been obtained for other cities such as Zhengzhou where air pollution is particularly severe. Located in Midwest of Huanghe-Huaihe River Flood Plain, Zhengzhou is the capital city of Henan Province with a population of more than 7 million. Like other densely populated areas, it has notorious air pollution problems caused by the increase of fossil fuel burning, transportation and industrial activities. However, no research has been carried out on the components and sources of $\mathsf{P M}_{2.5}$ in this district.
In this study, a comprehensive composition and source apportionment database of $\mathsf{P M}_{2.5}$ in Zhengzhou has been established for the first time. The $\mathsf{P M}_{2.5}$ was chemically characterized not only for inorganic components (trace elements and soluble ions) but also for organic components (EC and OC). Seasonal trend of $\mathsf{P M}_{2.5}$ concentration and its chemical composition were also quantified. A receptor model of PMF (Positive Matrix Factorization) was applied to identify the pollution sources and their potential contributions to $\mathsf{P M}_{2.5}$ . This study provides a better understanding of the pollution level and main pollution sources of the highly polluted area in a developing country. The results can be useful for policy decision makers in formulating effective $\mathsf{P M}_{2.5}$ control strategies.
2. Materials and methods
2.1. Site description
$\mathsf{P M}_{2.5}$ was monitored at the new campus of Zhengzhou University, located northwest of Zhengzhou $\langle34^{\circ}48^{\prime}\Nu$ ; $113^{\circ}31^{\prime}\mathrm{E})$ (Fig. 1). The instruments used in this study were installed on the rooftop ( $13\,\mathrm{m}$ above ground) of a building and the sampling port was $1.5\,\mathrm{m}$ above the ground.
Zhengzhou has a temperate continental monsoon climate with four clearly distinct seasons characterized by windy-dry spring, hot-rainy summer, cool-sunny autumn and cold-dry winter. South wind prevails in summer and north wind in winter (Ma, Chu, Li, & Song, 2009). During the studied period (2010), the average (range) temperatures were $_{3}\circ C$ ( $_{.}^{-5.8}$ to 13.7), $14.3^{\circ}\mathrm{C}$ ( $-0.6$ to 27.7), $26.4^{\circ}\mathrm{C}$ (18.2–31.9) and $15.7^{\circ}\mathrm{C}$ (5.2–28.1) for winter, spring, summer, and autumn, respectively. Corresponding relative humidity is: $47\%$ (11–96), $59\%$ (18–91), $73\%$ (45–95), $62\%$ (19–96), with precipitation: 4.6 (0–5.1), 31.8 (0–47.0), 119.1 (0–168.8), $50.0\,\mathrm{mm}$ (0–55.9). The atmospheric pressure is higher in winter than that in summer with an annual average value of $1003.1\,\mathrm{hPa}$ . The annual precipitation is $616.5\,\mathrm{mm}$ .
The sampling station is a neighborhood-scale site representing an industrial region of Zhengzhou with a population of 200,000. It is in the junction of the Lian-Huo Freeway on the north and the city highway in the west. Yellow River is located $15\,\mathrm{km}$ north of the monitoring station, which will bring sand dust in the windy spring. Previous study also showed that windy-dry spring in Zhengzhou can facilitate the sandy/dusty weather process (Tian, Zheng, Chen, Deng, & Du, 2010). The main industries in these zones are power plants (including a coal-burning power plant and a gas-burning power plant), steel plant, food and chemical industries. Moreover, fossil fuels, coal and oil, widely used for residential heating in winter, are significant sources of $\mathsf{P M}_{2.5}$ .
2.2. $P M_{2.5}$ sample collection
The $\mathsf{P M}_{2.5}$ samples were collected on quartz membrane fliter $20.3\,\mathrm{cm}\times25.4\,\mathrm{cm}$ , Pall Corporation, USA) using a high-volume gravimetric air sampler (Tisch Environmental). It separates particles with aerodynamic diameters below $2.5\,\upmu\mathrm{m}$ using a pump to draw air at a rate of $1.13\,\mathrm{m}^{3}/\mathrm{min}$ . The quartz fliters were baked at $450\,^{\circ}\mathrm{C}$ for $^{5\,\mathrm{h}}$ to remove adsorbed organic vapors and then equilibrated in a desiccator before sampling. The sampling was carried out in spring, summer, autumn and winter in 2010. Sampling duration for each sample was $23.5\,\mathrm{h}$ , from $9{\mathrel{:}}00\,{\mathsf{a m}}$ to 8:30 am next day, and a total of $52\;\mathrm{PM}_{2.5}$ samples were collected. Prior to measurement, the flow rate of the sampler was calibrated. Blank fliters were also collected and treated using the same procedure of regular samples to subtract any effect caused by possible contamination. After sampling, the fliters were stored in a freezer to prevent possible volatilization and photo degradation of particles (Tao et al., 2009). Before and after sampling, fliters were equilibrated for $48\,\mathrm{h}$ atatemperatureof $25\pm1\,^{\circ}\mathrm{C}$ andthenweighedwithamicrobalance (Mettler Toledo XS205, Switzerland) to measure the particle
concentration. Samples were then analyzed for soluble ions, elements, EC and OC.
2.3. $P M_{2.5}$ sample analysis
Carbon fractions were analyzed by a Sunset offline OC/EC analyzer following the thermo-optical transmission (TOT) protocol of the US National Institute for Occupational Safety and Health (Song et al., 2006). A $1\,\mathrm{cm}^{2}$ punch from the fliter was placed in the slotted quartz insert and the nut on the photo-detector was hand-tightened to seal the connection of the insert for analysis.
The sample is first heated in steps under helium, and the carbon volatilized is considered as OC. During the entire process a laser acts on the quartz membrane; the transmission of light is weakened when OC is carbonized. Then a second automatic heating up program is carried out under $\mathrm{He}/0_{2}$ oxidizing conditions to detect EC. During this process, the temperature increases, EC is oxidized and decomposed, and the transmitted light of the laser beam is gradually strengthened. The point at which the light recovers to the initial transmission light intensity is considered the dividing point between OC and EC: carbon detected before this point is OC, and after this point is EC (Huang, Zeng, & Shao, 2005).
Two punch holes with an area of $10.7\,\mathrm{cm}^{2}$ from the fliter were used to determine the mass concentrations of water-soluble ions. The sample was extracted three times in $20\,\mathrm{mL}$ of ultrapure water (conductivity $18.2\,\mathrm{M}\Omega\,\mathrm{cm})$ for $20\,\mathrm{min}$ in an ultrasonic bath. The average extraction efficiency for all ions was estimated to be $98–100\%$ after the 3rd extraction (Perrone et al., 2010). $60\,\mathrm{mL}$ of the extract was then flitered by a $0.2\,\upmu\mathrm{m}$ polyether sulfone fliter to remove insoluble fractions and analyzed by ion chromatography (ICS-90, Dionex, USA) with an autosampler and a conductivity detector. Anions including $\mathsf{F}^{-}$ , $C1^{-}$ , $\mathsf{N O}_{3}^{\mathrm{~-~}}$ and $S0_{4}{}^{2-}$ were determined with an IonPac AS14 separation column and an AG14 guard column (Dionex), using $8\,\mathrm{mM\Na_{2}C O_{3}}/1\,\mathrm{mM\NaHCO_{3}}$ as eluent at the rate of $0.8\,\mathrm{mL/min}$ . Five cations $\cdot(\mathrm{N}\mathbf{a}^{+},\mathrm{N}\mathrm{H}_{4}^{+},\mathrm{K}^{+},\mathrm{M}\mathbf{g}^{2+}$ and ${\bf C}{\bf a}^{2+}$ ) in aqueous extracts were measured by an IonPac CS12A separation column and a CG12A guard column (Dionex), using $20\,\mathrm{mM}$ methanesulfonic acid as eluent at $1.0\,\mathrm{mL/min}$ . The method was used for separation and quantification of water-soluble ions. Each sample was analyzed within $20\,\mathrm{min}$ with a sample loop volume of $25\,\upmu\mathrm{L}.$ . External standard method was used for quantification by using concentration range standard mixtures prepared from single liquid standards. The regression coefficients $(R^{2})$ of the calibration curves are over 0.995 for all ions except $\mathsf{N H}_{4}^{\,\,+}$ which showed a quadratic response. The method detection limits (MDLs) of $\mathsf{N a}^{+}$ , $\mathsf{N H}_{4}^{\,\,+}$ , $K^{+}$ , ${\mathrm{Mg}}^{2+}$ , ${\bf C}{\bf a}^{2+}$ , $\mathsf{F}^{-}$ , $C1^{-}$ , $\mathsf{N O}_{3}^{\mathrm{~-~}}$ and $S0_{4}{}^{2-}$ are 0.002, 0.006, 0.005, 0.004, 0.007, 0.005, 0.009, 0.032 and $0.029\,\mathrm{mg/L}$ , respectively.
One-sixth of each fliter sample was used for the analyses of elements in $\mathsf{P M}_{2.5}$ . The fliter was cut into fragments and put into a polytetrafluoroethylene digestion vessel. A microwave digestion system (Berghof MWS- $.3^{+}$ , Germany) was used with $15\,\mathrm{mL}$ mixture of $5.6\%\ H N O_{3}/16.8\%$ HCl for $1\,\mathrm{h}$ . Digestion by microwave energy is suitable for closed vessel technique, in which solution can be heated in low conductivity Teflon vessels. Rapid heating and cooling can be obtained and the performance of acids can be vastly improved (Joshi & Balasubramanian, 2010).
After digestion the extracts were diluted to $30\,\mathrm{mL}$ and then flitered by a $0.45\,\upmu\mathrm{m}$ polyamide fliter to remove insoluble fractions before analysis by inductively coupled plasma-mass spectrometry (ICP-MS) (Agilent7500cx, USA). Quantification was made by using the internal standard method (Sc, Ge, Rh, In, Bi) to correct signal drift and matrix suppression and the standard was injected simultaneously with the sample solution. For quality control of metal analysis, the MDL was estimated using blank fliter according to the same analysis process of the sample, and the values ranged from 0.003 to 1.172 ppb. Metal recoveries were calculated by dividing the measured values by theoretical values, and the recovery value ranged from $81\%$ (Cd) to $110\%$ (K). The accuracy test shows that the RSD (relative standard deviation) is satisfactory (ranging from 0.7 to $3.4\%$ , indicating a high accuracy of determination results. $R^{2}$ values for all the 30 trace elements are better than 0.99.
2.4. PMF analysis
PMF is a multivariate factor analysis tool that decomposes a matrix of sample data into two matrices: factor contributions and factor proflies. With measured source proflie information and emission inventories, the source type is determined (Norris et al., 2008). In this study, the EPA PMF 3.0 program based on the multilinear engine (ME)-2 was used for the PMF analysis. The PMF model can be expressed as:
$$
x_{i j}=\sum_{k=1}^{p}{g_{i k}f_{k j}}+e_{i j},
$$
where $x_{i j}$ is matrix X of i by $j$ dimensions, i is the number of samples and $j$ is the number of chemical species, $p$ the number of factors, $f$ the species proflie of each source, $g$ the amount of mass contributed by each factor to each individual sample, and $e_{i j}$ is the residual for each sample/species:
$$
e_{i j}=x_{i j}-\hat{x}_{i j}=x_{i j}-\sum_{k=1}^{p}g_{i k}f_{k j}.
$$
$Q$ is an object function, and a criterion for the model, which is defined as:
$$
Q=\sum_{i=1}^{n}\sum_{j=1}^{m}\left[\frac{e_{i j}}{u_{i j}}\right]^{2},
$$
where $u_{i j}$ is the uncertainty of the jth component in the ith sample. In this study, the uncertainty is computed as follows:
$$
u_{i j}=\left\{\begin{array}{c l}{\frac{5}{6}\mathrm{MDL},}&{\mathrm{when}~~u_{i j}\le\mathrm{MDL}}\\ {\sqrt{\left(\mathrm{RSD}\cdot x_{i j}\right)^{2}+\mathrm{MDL}^{2}},}&{\mathrm{when}~~u_{i j}>\mathrm{MDL}}\end{array}\right.
$$
The missing data were replaced by the median value of that species and given an uncertainty four times the median value. Concentrations of trace metals below the MDLs were replaced with values of half the MDLs (Joshi & Balasubramanian, 2010). Species with more than $95\%$ of samples below MDL were not included for the PMF analysis. The robust mode was adopted in the calculation to eliminate the impact of extreme values. Only converged solutions should be investigated for further analysis. The minimum Q value from the multiple runs was selected to carry out further analysis with PMF3.0 (Norris et al., 2008). The results are constrained so that factor contributions could not be negative for any species (Eatough et al., 2008). It allows each data point to be individually weighed. This feature allows the analyst to adjust the influence of each data point, depending on the confidence in the measurement.
3. Results and discussion
3.1. $P M_{2.5}$ concentrations in Zhengzhou
The concentration and chemical composition of $\mathsf{P M}_{2.5}$ collected in 2010 and its mass concentration variation during the sampling period are summarized in Table 1 and Fig. 2. The annual average concentration of $\mathsf{P M}_{2.5}$ is $175\,\upmu\mathrm{g}/\mathrm{m}^{3}$ (ranging from 61 to $349\,\upmu\mathrm{g}/\mathrm{m}^{3}$ with a standard deviation of $71\,\upmu\mathrm{g}/\mathrm{m}^{3}.$ ), which is significantly higher than the annual $\mathsf{P M}_{2.5}$ regulation $(15\,\upmu\mathrm{g}/\mathfrak{m}^{3})$ of USA NAAQS (National Ambient Air Quality Standards). It demonstrates serious particle-related pollution near our sampling station. A one-way analysis of variance indicates that the concentration of $\mathsf{P M}_{2.5}$ shows significant seasonal variation with high level of $211\,\upmu\mathrm{g}/\mathrm{m}^{3}$ in winter and low level of $122\,\upmu\mathrm{g}/\mathrm{m}^{3}$ in summer (Fig. 3(a)).
3.2. Chemical composition and seasonal variations of $P M_{2.5}$
3.2.1. Water-solubleions
Water-soluble ionic species account for an average of $41\%$ (ranging from 7 to $64\%$ of $\mathsf{P M}_{2.5}$ and contribute the major fraction of atmospheric aerosols in the studied area in 2010. The concentrations of the most abundant ionic species show the order of S $\mathrm{\DeltaO_{4}}^{2-}>\mathrm{NO_{3}}^{-}>\mathrm{NH_{4}}^{+}>\mathrm{Cl}^{-}>\mathrm{Ca}^{2+}>\mathrm{K}^{+}>\mathrm{Na}^{+}>\mathrm{F}^{-}>\mathrm{Mg}^{2+}$ (Table 1). It is noted that the sum of $\mathrm{NO}_{3}{}^{-}$ , $S0_{4}{}^{2-}$ and $\mathsf{N H}_{4}^{+}$ accounts for more than $78\%$ of total analyzed water-soluble ions. As they are all from secondary origins, the high contribution of water-soluble ionic species to $\mathsf{P M}_{2.5}$ may indicate a serious secondary photochemical pollution, although dark reaction is responsible for nitrate formation in many areas, e.g. nitrate formation on sea-salt and mineral particles (Mamane & Gottlieb, 1992).
As for the seasonal distribution, high water-soluble ion contribution $(54\%)$ to total $\mathsf{P M}_{2.5}$ mass is observed in summer (Fig. 4(d)), followed by autumn $(48\%)$ . During spring, water-soluble ions account for $36\%$ of $\mathsf{P M}_{2.5}$ , lower than the average value $(55\%)$ reported for the spring of Guangzhou, China in 2007 (Tao et al., 2009). The contribution of water-soluble ions to $\mathsf{P M}_{2.5}$ mass is $36\%$ in winter. Zhengzhou has a typical monsoon climate, causing hot summer (average temperature $26.4\,^{\circ}\mathrm{C})$ with strong solar radiation. In autumn, cloud cover is low in most regions, the solar radiation is relatively strong and the absorption capability of particulate matter for gaseous contaminants decreased. Moreover, photochemical reaction occurs mainly in summer and autumn which is beneficial to formation of gaseous contaminants.
Among the water-soluble ions, sulfate is the most abundant species. It comprises $15\%$ of $\mathsf{P M}_{2.5}$ with an average concentration of $26\,\upmu\mathrm{g}/\mathrm{m}^{3}$ and a standard deviation of $14\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , ranging from 5 to $54\,\upmu\mathrm{g}/\mathrm{m}^{3}$ . Sulfate has a seasonal trend with high level of 35 and $32\,\upmu\mathrm{g}/\mathrm{m}^{3}$ in autumn and summer and low level of approximately $22\,\upmu\mathrm{g}/\mathrm{m}^{3}$ in spring and winter (Fig. 5(a)). Mean $S0_{4}{}^{2-}$ contributions to $\mathsf{P M}_{2.5}$ mass are $27\%$ and $19\%$ in summer and autumn, $12\%$ and $10\%$ in spring and winter, respectively. $\mathsf{P M}_{2.5}$ enriched with $S0_{4}{}^{2-}$ may result from secondary formation by oxidation of the gaseous precursor $S0_{2}$ with hydroxyl radicals (Lonati, Giugliano, & Ozgen, 2008).
$\mathsf{N O}_{3}^{\mathrm{~-~}}$ concentrations are measured to be 23 and $17\,\upmu\mathrm{g}/\mathrm{m}^{3}$ in autumn and winter, respectively. Relatively low level appears in summer with the values of $11\,\upmu\mathrm{g}/\mathrm{m}^{3}$ . $\mathsf{N O}_{3}^{\ensuremath{-}}$ contribution to total $\mathsf{P M}_{2.5}$ mass is $9\%$ on average. In seasonal distribution (Fig. 5(a)), it has a relatively high level in autumn, but no significantly seasonal trend in the other seasons. $\mathsf{N O}_{3}^{\mathrm{~-~}}$ is generally formed via the oxidation of $\Nu0_{x}$ , and the variation of $\mathsf{N O}_{3}^{\mathrm{~-~}}$ is related closely to $\mathsf{N O}_{2}$ and the meteorological factors. Generally, $\mathsf{N O}_{2}$ is mainly from traffic and industrial emissions in cities, which are generally evenly distributed over the whole year, although the variation of $\mathsf{N O}_{3}^{\mathrm{~-~}}$ might be strongly related to meteorological factors. Researches show that $\mathsf{N O}_{3}^{\ensuremath{-}}$ is mainly present as semi-volatile ammonium nitrate $\left(\mathrm{NH}_{4}\mathrm{NO}_{3}\right)$ ). Winter with low temperature and autumn with low atmospheric boundary layer favor the shift from the gas phase of nitric acid to the particle phase of nitrate, resulting in high $\mathsf{N O}_{3}^{\mathrm{~-~}}$ concentration (Wang et al., 2005; Zhao, Zhang, Xu, & Chen, 2011).
$\mathsf{N H}_{4}^{+}$ contributes $9\%$ of total $\mathsf{P M}_{2.5}$ mass on average, with a relatively high concentration in winter $(17\,\upmu\mathrm{g}/\mathfrak{m}^{3})$ and summer $(16\,\upmu\mathrm{g}/\mathrm{m}^{3})$ , as shown in Fig. 5(a). The seasonal variations of $\mathsf{N H}_{4}^{+}$ are different from Shanghai in which high concentrations are observed in spring and winter while low concentrations in summer and autumn (Wang, Bi, Sheng, & Fu, 2006; Wang, Zhuang, et al., 2006), but basically coincide with Beijing, where high concentrations occur in winter and summer (Wang et al., 2005). The difference from Shanghai might be due to different meteorological conditions and pollution sources, and partly influenced by the marine sources. Consistency of seasonal variation of $\mathsf{N H}_{4}^{+}$ with Beijing may be due to similar transformation mechanisms under the same meteorological conditions. High concentrations are observed in both summer and winter because secondary transformation is accelerated by high humidity and strong solar radiation in summer and poor dispersion and the lower removal rate in winter.
The average concentration of $C1^{-}$ is $10\,\upmu\mathrm{g}/\mathrm{m}^{3}$ during winter but $<2\,\upmu\mathrm{g}/\mathrm{m}^{3}$ during summer (Fig. 5(a)). The lowest levels of
$K^{+}$ , $\mathsf{F}^{-}$ , $\mathsf{N a}^{+}$ and ${\bf C}{\bf a}^{2+}$ all appear in summer and the highest concentrations of $K^{+}$ , $\mathsf{F}^{-}$ and $\mathsf{N a}^{+}$ all occur in winter while ${\bf C}{\bf a}^{2+}$ in spring (Fig. 5(a)).
3.2.2. Elements
All 30 analyzed elements comprise about $2.5\%$ of $\mathsf{P M}_{2.5}$ mass, and all elements together have an average concentration of
$4.6\,\upmu\mathrm{g}/\mathrm{m}^{3}$ ranging from 0.8 to $14.8\,\upmu\mathrm{g}/\mathrm{m}^{3}$ each. Total element contribution to $\mathsf{P M}_{2.5}$ has a significant seasonal variation (Fig. 5(b)). The concentrations of the most abundant species follow the order of $\mathsf{F e}>\mathsf{C a}>\mathrm{K}>\mathrm{Al}>\mathrm{Zn}>\mathrm{Mg}>\mathsf{P b}>\mathrm{Mn}>\mathrm{Na}>\mathrm{Ba}>\mathsf{C u}>\mathrm{Sr}>\mathrm{As}>\mathsf{C r}>\mathrm{Cd}>\mathrm{K}.$ $\scriptstyle\mathrm{Se}>\mathrm{V}>\mathrm{Ni}>\mathrm{Ce}>\mathrm{Nd}>\mathrm{Tl}>\mathrm{Co}>\mathrm{Ag}>\mathrm{Th}>\mathrm{Sm}>\mathrm{U}>\mathrm{Dy}>\mathrm{Be}>\mathrm{Er}>\mathrm{Ho}$ (Table 1). In this study, Fe is the most abundant element with an average concentration of $1.2\,\upmu\mathrm{g}/\mathrm{m}^{3}$ and a standard deviation of
$1.0\,\upmu\mathrm{g}/\mathrm{m}^{3}$ . The second most abundant element is Ca which has a concentration of $0.8\pm0.6\,\upmu\mathrm{g}/\mathrm{m}^{3}$ . The first four abundant species are crust elements quite possibly coming from soil dust (Song et al., 2006).
The seasonal variation of elemental concentrations is given in Fig. 5(b), all generally decreasing in summer except Pb. In addition, Fe, Al, Ca, Mg, Ba, Sr and V show highest values in spring followed by winter and autumn, and the low values in summer. This trend may be due to sandstorm weather which usually appears in spring in this area and, to a greater extent, original soil dust. This result is in accord with the research carried out in East Sea (Kang et al., 2011), where high concentrations of Al, Mg, Ca, Fe, Ti, and Co are measured in spring (March–May), when Asian dust transportation occurs. Quite differently, high concentrations of K, Na and Zn appear in winter, while Cu, As, Cr, Cd and Ni show high concentrations in autumn.
3.2.3. EC and OC
TheconcentrationsofECandOCinthestudiedareaaresummarized in Table 1. The EC concentration in $\mathsf{P M}_{2.5}$ is $3.9\pm1.7\,\upmu\mathrm{g}/\mathrm{m}^{3}$ and comprises about $2\%$ of $\mathsf{P M}_{2.5}$ mass. EC has a slightly high value in autumn, partially due to the location of the sampling station and the climate in Zhengzhou. Previous study shows that diesel emissions are rich in EC (Eatough et al., 2008). Two freeways on the north and west of the sampling station have a large number of heavy-duty diesel-powered vehicles which generally emit high EC compared to gasoline vehicles (Cheung et al., 2011). Under Chinese monsoon climate, northwest wind which prevails in autumn and winter can bring high concentration of EC from vehicles.
The corresponding OC concentration ranges from 5 to $58\,\upmu\mathrm{g}/\mathrm{m}^{3}$ contributing $11\%$ of $\mathsf{P M}_{2.5}$ mass. OC has the same seasonal distribution in parallel with $\mathsf{P M}_{2.5}$ concentrations (Table 1). High level of OC in cold wintertime near our monitoring station is caused by intense coal burning, as OC is often emitted from burning sources, e.g. coal combustion and biomass burning (Song et al., 2006). Incidentally, both EC and OC values in Zhengzhou are similar to several major Chinese cities such as Shenzhen, Beijing, Guangzhou and Shanghai (Feng et al., 2009; Hagler et al., 2006; Song et al., 2006), as shown in Table 2, all studies using the NIOSH TOT method. A given OC/EC ratio may serve to indicate its source (Malm, Schichtel, Pitchford, Ashbaugh, & Eldred, 2004; Watson et al., 2002). The average OC/EC ratio observed in Zhengzhou is 5.6, clearly indicative of potential multi-source contribution to $\mathsf{P M}_{2.5}$ .
3.2.4. Mass closure
Chemical mass closure was used to determine the accuracy of the measurements. In this study the chemical components were classified into seven major types, i.e. $S0_{4}{}^{2-}$ , $\mathsf{N O}_{3}^{\ensuremath{-}}$ , $\mathsf{N H}_{4}^{+}$ , EC, particulate organic matter (POM), soil dust and other elements (OE):
The contribution of soil dust in $\mathsf{P M}_{2.5}$ was assessed by using the concentrations of Na, Mg Al, K, Fe, Ca and Si. Elements Na, Mg, Al, K and Fe were calculated as metal oxides, Ca as $\mathsf{C a C O}_{3}$ (Brook, Dann, & Burnett, 1997), elemental Si was estimated by multiplying Al with a factor of 3.41 (Hueglin et al., 2005) since it was not analyzed by the ICP-MS method used in this study. The conversion factors for $S0_{4}{}^{2-}$ , $\mathsf{N O}_{3}^{\mathrm{~-~}}$ and $\mathsf{N H}_{4}^{+}$ were 1.29. Aerosols in this region were prone to be acidic, involving significant bound water in gravimetric mass (Rees, Robinson, Khlystov, Stanier, & Pandis, 2004). Harrison, Jones, and Lawrence (2003) applied hydration multiplication factor of 1.29 to convert these inorganic species (ammonium sulfate, ammonium nitrate) into hydrated species. Turpin and Lim (2001) recommended the POM-to-OC conversion factor of $2.1\pm0.2$ for non-urban aerosols and $1.6\pm0.2$ for urban aerosols after calculating the ratios of organic species at various measurement sites. In addition, 1.8 was usually used as OC-to-POM multiplier for aerosols in Chinese urban sites (Guinot, Cachier, & Oikonomou, 2007; Wang, Bi, et al., 2006; Wang, Zhuang, et al., 2006). In view of the above, we supposed a conversion factor of 1.8 in our study.
The results of chemical mass closure indicate that $73\%$ of $\mathsf{P M}_{2.5}$ mass were identified, which fall within the range of previous studies of $60{-}80\%$ (Andrews et al., 2000; Chan et al., 1997; Kim, Teffera, & Zeldin, 2000).
3.3. Source apportionment of $P M_{2.5}$ using PMF
In this study, 25 species including $\mathsf{P M}_{2.5}$ were used as input data for PMF model. The factor number performed ranged from 1 to 8; the six-factor with $F_{\mathrm{PEAK}}\!=\!0$ solution was found to provide the “optimal solution”. Under the solution, residuals of the majority of standardized species were between $^{-3}$ and $+3.\,{\sf G}$ -space plots show data points lying within the source axes. Twenty runs were made for each factor and the lowest $Q_{\mathrm{robust}}$ is 1424.2 with $Q_{\mathrm{robust}}/Q_{\mathrm{true}}$ ratios of 0.97. Note that $Q_{\mathrm{robust}}/Q_{\mathrm{true}}<2$ indicates an overall acceptable fit (Baumann, Jayanty, & Flanagan, 2008). A better goodness-of-fit $Q_{\mathrm{robust}}$ was achieved by uniformly adding $5\%$ of concentration to the uncertainties of all input species. The improvement was further enhanced by excluding several outliers from the input data (Norris et al., 2008), resulting in six appropriate source factors with clear proflie and physical meaning as shown in Fig. 6.
Factor 1 is significantly loaded on Ni, Pb and Cr. High loadings by the heavy metal element Ni suggested an industrial source (Song et al., 2006). For example, an oil-refinery catalytic cracker proflie was characterized by relatively high levels of Ni (Chow et al., 2004). Pb was used in paints, varnishes, pipes, storage batteries and was emitted from various related industrial activities. Factor 1 was thus identified as industrial pollution.
Factor 2 is highly loaded on Mg, Al, Ca, Fe, Sr, Ba, which is characteristic of elements of soil dust (Callen et al., 2009; Mazzei et al., 2008). It shows a seasonal distribution with maxima in spring due to sandstorm weathers that usually appear in Zhengzhou in spring.
Factor 3 is heavily weighted on $S0_{4}{}^{2-}$ , $\mathrm{NH_{4}}^{+}$ and $\mathsf{N O}_{3}^{\ensuremath{-}}$ , and could be identified as a mixture of secondary aerosols of nitrates and sulfates. The formation of secondary ions in the atmosphere is mainly from gaseous precursors $S0_{2}$ , $\mathsf{N H}_{3}$ and $\Nu0_{x}$ ) created by anthropogenic activities (Perrone et al., 2010). High factor contribution of secondary aerosol was in summer and autumn when photochemical reaction is obvious.
Factor 4 shows major loadings for EC, $C1^{-}$ , $K^{+}$ and $\mathsf{N a}^{+}$ , species which are primarily emitted from biomass burning (Baumann et al., 2008; Chow et al., 2004; Reff et al., 2009). Factor 4 also shows loadings for Ni, V and Mn, which are trace elements of fuel oil burning (Vecchi, Marcazzan, & Valli, 2007; Yatkin & Bayram, 2007). Cu, Zn and Cd in Factor 4 are derived from refuse incineration (Morawska & Zhang, 2002). Consequently, this factor is classified as a mixing source of biomass burning/oil combustion/incineration. It has a seasonal variation with maxima in autumn. Henan is a large agricultural province and ranks number 4 in $\mathsf{P M}_{2.5}$ emissions caused by biomass burning (Lu, Kong, Han, Wang, & Bai, 2011). Biomass straw burning was conducted in harvest autumn, causing air pollution by particulate matter to produce serious problem which always makes the airport shut, highway traffic suffocating.
Factor 5 is predominantly loaded on Zn, Pb, EC, Cu and Cd. Vehicle brake wear, tire wear and oil drip could result in greater abundance of Zn, Cu in paved road dust (Watson et al., 2002; Yatkin & Bayram, 2007). Zn is a marker element in addition to Pb for transportation because utilization of Pb as a fuel additive nowadays has been banned (Fang et al., 2003). EC is also the characteristic element of vehicle. So, Factor 5 represents the pollution of vehicles.
Factor 6 shows high loadings for $C1^{-}$ , OC, $\mathsf{N O}_{3}^{\ensuremath{-}}$ , Cd, Se, As and Cu. Both source proflies measured in the laboratory and the chemical analysis of ambient $\mathsf{P M}_{2.5}$ samples have indicated that $C1^{-}$ and OC can be considered as tracer elements for coal combustion (Duan et al., 2006; Zheng et al., 2005). Coal combustion emits HCl. The presence of $C1^{-}$ perhaps is due to short-lived HCl in the atmosphere (one to five days) since it is very soluble and reacts readily with ammonia or alkaline cations such as Ca or K to form particulate chloride (Harkov & Ross, 1999). Cd, Se and As are also used as a markers for coal-fired power plant emissions. Coal combustion emissionhasaseasonalvariationwithmaximainwinter,andthe result is consistent with a previous study (Shi et al., 2009). Factor 6 has been identified as coal combustion-related source.
Source contribution estimates of $\mathsf{P M}_{2.5}$ were calculated for each of the samples from the correspondingly modeled species’ absolute mass (Table 3). Also the results of Bootstrap run are shown in Table 3 with similar source contribution estimates. During the sampling period, soil dust represented the most abundant contribution $(47\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , $26\%)$ to the total $\mathsf{P M}_{2.5}$ mass, immediately followed by secondary aerosol pollution $(42\,\upmu\mathrm{g}/\mathsf{m}^{3},24\%)$ . Coal combustion contributes $23\%$ to $\mathsf{P M}_{2.5}$ mass with a contributing concentration of $42\,\upmu\mathrm{g}/\mathrm{m}^{3}$ . The contributions of other sources were mixed (biomass burning/oil combustion/incineration) $(24\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , $13\%$ ), vehicle $(17\,\upmu\mathrm{g}/\mathsf{m}^{3},\;10\%)$ , industrial $(7\,\upmu\mathrm{g}/\mathfrak{m}^{3},\,4\%)$ . These results appear reasonable, because the climate in Zhengzhou was characterized by windy-dry spring, hot summer, sunny autumn and cold-dry winter. Sandstorm weather usually happens in spring which causes soil dust pollution. Hot-sunny summer and autumn benefit the photochemical reactions which can lead to secondary pollution. In addition, the fertilizer plant in the region produces $\mathrm{NH}_{3}$ which also leads to the formation of secondary ions. Also, coal combustion is used for domestic heating in winter. The sampling campaign was conducted in an industrial district adjoined to Lian-Huo freeway which led to industrial activities and vehicles.
4. Conclusions
This study employed three relatively simple analytical techniques, ICP-MS, IC and TOT, for elements, soluble ions and EC/OC analysis, respectively, to investigate chemical composition of $\mathsf{P M}_{2.5}$ sampled in an industrial district of Zhengzhou. The annually averaged concentration of $\mathsf{P M}_{2.5}$ was $175\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , significantly higher than the $\mathsf{P M}_{2.5}$ regulation $(15\,\upmu\mathrm{g}/\mathfrak{m}^{3})$ of US NAAQs. Seasonal trend of $\mathsf{P M}_{2.5}$ follows the order of winter $>$ autumn $>$ spring $>$ summer. Soluble ions contribute the major fraction of $\mathsf{P M}_{2.5}$ and have high levels in summer. TC is the second major fraction with high concentration in winter.
PMF was successfully performed and a reasonable result involving six factors was obtained for $\mathsf{P M}_{2.5}$ . The three main sources were soil dust, secondary aerosol and coal combustion, which contribute about $26\%$ , $24\%$ and $23\%$ of total $\mathsf{P M}_{2.5}$ mass concentration. The other sources were biomass burning/oil combustion/incineration, vehicleandindustrialemissionswhichcontribute $13\%$ , $10\%$ and $4\%$ of $\mathsf{P M}_{2.5}$ , respectively. A major contribution of secondary aerosol and the high OC/EC ratios of 5.3 observed in $\mathsf{P M}_{2.5}$ samples show serious secondary organic aerosol pollution near our monitoring station. These results are consistent with the climate of Zhengzhou.
The data of $\mathsf{P M}_{2.5}$ concentration and composition, reported for the first time in this study, can assist regulatory agency to take appropriate actions to reduce $\mathsf{P M}_{2.5}$ pollution. For example, the stack emission standards for $\Nu0_{x}$ and $S0_{2}$ from coal burning plants should be regulated to alleviate secondary aerosol pollution by using well known technologies. Also, employing central heating system to replace the light-duty coal-fired boilers used for winter heating could be an important way to control the coal-related $\mathsf{P M}_{2.5}$ pollution, along with replacing the current coal combustion technology with some advanced coal gasification technologies. Strict ban for straw burning should be implemented. Further, the public should be encouraged and educated to use public transportation as well as car pools and energy saving vehicles to reduce mobile pollutant sources. Lastly, the inventory of air pollutant quantity for major industrial sources needs to be quantified and restrained.
Acknowledgments
This study is part of the Science and Technology Plan Project in Zhengzhou funded by Henan Administration of Foreign Experts Affairs and Science and Technology Bureau of Zhengzhou City (grant no. 094SYJH36069). Authors would like to thank the support from Peking University and Taiwan Yunlin University of Science and Technology.
References
Andrews, E., Saxena, P., Musarra, S., Hildemann, L. M., Koutrakis, P., McMurry, P. H., et al. (2000). Concentration and composition of atmospheric aerosols from the 1995 SEAVS experiment and a review of the closure between chemical and gravimetric measurements. Journal of the Air and Waste Management Association, 50(5), 648–664.
Apte, J. S., Kirchstetter, T. W., Reich, A. H., Deshpande, S. J., Kaushik, G., Chel, A., et al. (2011). Concentrations of fine, ultrafine, and black carbon particles in auto-rickshaws in New Delhi, India. Atmospheric Environment, 45(26), 4470–4480.
Baumann, K., Jayanty, R. K., & Flanagan, J. B. (2008). Fine particulate matter source apportionment for the chemical speciation trends network site at Birmingham, Alabama, using positive matrix factorization. Journal of the Air and Waste Management Association, 58(1), 27–44.
Brook, J. R., Dann, T. F., & Burnett, R. T. (1997). The relationship among TSP, $\mathrm{PM}_{10}$ , $\mathrm{PM}_{2.5}$ , and inorganic constituents of atmospheric particulate matter at multiple Canadian locations. Journal of the Air and Waste Management Association, 47(1), 2–19.
Callen, M. S., de la Cruz, M. T., Lopez, J. M., Navarro, M. V., & Mastral, A. M. (2009). Comparison of receptor models for source apportionment of the $\mathrm{PM}_{10}$ in Zaragoza (Spain). Chemosphere, 76(8), 1120–1129.
Chan, Y. C., Simpson, R. W., McTainsh, G. H., Vowles, P. D., Cohen, D. D., & Bailey, G. M. (1997). Characterisation of chemical species in $\mathrm{PM}_{2.5}$ and $\mathsf{P M}_{10}$ aerosols in Brisbane, Australia. Atmospheric Environment, 31(22), 3773–3785.
Cheung, K., Daher, N., Kam, W., Shafer, M. M., Ning, Z., Schauer, J. J., et al. (2011). Spatial and temporal variation of chemical composition and mass closure of ambient coarse particulate matter $\left(\mathsf{P M}_{10-2.5}\right)$ in the Los Angeles area. Atmospheric Environment, 45(16), 2651–2662.
Chow, J. C., Watson, J. G., Kuhns, H., Etyemezian, V., Lowenthal, D. H., Crow, D., et al. (2004). Source proflies for industrial, mobile, and area sources in the Big Bend Regional Aerosol Visibility and Observational study. Chemosphere, 54(2), 185–208.
Delfino, R. J., Sioutas, C., & Malik, S. (2005). Potential role of ultrafine particles in associations between airborne particle mass and cardiovascular health. Environmental Health Perspectives, 113(8), 934–946.
Duan, F. K., He, K. B., Ma, Y. L., Yang, F. M., Yu, X. C., Cadle, S. H., et al. (2006). Concentration and chemical characteristics of $\mathrm{PM}_{2.5}$ in Beijing, China: 2001–2002. The Science of the Total Environment, 355(1–3), 264–275.
Eatough, D. J., Grover, B. D., Woolwine, W. R., Eatough, N. L., Long, R., & Farber, R. (2008). Source apportionment of 1 h semi-continuous data during the 2005 study of organic aerosols in riverside (SOAR) using positive matrix factorization. Atmospheric Environment, 42(11), 2706–2719.
Fang, G. C., Chang, C. N., Chu, C. C., Wu, Y. S., Fu, P. P. C., Chang, C. C., et al. (2003). Characterization of particulate, metallic elements of TSP, $\mathrm{PM}_{2.5}$ and $\mathsf{P M}_{2.5-10}$ aerosols at a farm sampling site in Taiwan, Taichung. Science of the Total Environment, 308(1–3), 157–166.
Feng, Y., Chen, Y., Guo, H., Zhi, G., Xiong, S., Li, J., et al. (2009). Characteristics of organic and elemental carbon in $\mathrm{PM}_{2.5}$ samples in Shanghai, China. Atmospheric Research, 92(4), 434–442.
Guinot, B., Cachier, H., & Oikonomou, K. (2007). Geochemical perspectives from a new aerosol chemical mass closure. Atmospheric Chemistry and Physics, 7(6), 1657–1670.
Hagler, G. S. W., Bergin, M. H., Salmon, L. G., Yu, J. Z., Wan, E. C. H., Zheng, M., et al. (2006). Source areas and chemical composition of fine particulate matter in the Pearl River Delta region of China. Atmospheric Environment, 40(20), 3802–3815.
Harkov, R., & Ross, J. (1999). Hydrogen chloride (HCl) emissions from Maryland utility boilers. Annapolis, MD, USA: Environmental Resources Management, Inc./Maryland Power Plant Research Program. Technical report.
Harrison, R. M., Jones, A. M., & Lawrence, R. G. (2003). A pragmatic mass closure model for airborne particulate matter at urban background and roadside sites. Atmospheric Environment, 37(35), 4927–4933.
Huang, N., Zeng, L., & Shao, M. (2005). Review of measurement techniques about organic carbon and elemental carbon in atmospheric particles. Acta Scientiarum Naturalium Universitatis Pekinensis, (6), 957–964 (in Chinese).
Hueglin, C., Gehrig, R., Baltensperger, U., Gysel, M., Monn, C., & Vonmont, H. (2005). Chemical characterisation of $\mathrm{PM}_{2.5}\ \mathrm{PM}_{10}$ and coarse particles at urban, near-city and rural sites in Switzerland. Atmospheric Environment, 39(4), 637–651.
Joshi, U. M., & Balasubramanian, R. (2010). Characteristics and environmental mobility of trace elements in urban runoff. Chemosphere, 80(3), 310–318.
Kang, J., Choi, M., Yi, H., Song, Y., Lee, D., & Cho, J. (2011). A five-year observation of atmospheric metals on Ulleung Island in the East/Japan Sea: Temporal variability and source identification. Atmospheric Environment, 45(25), 4252–4262.
Kim, B. M., Teffera, S., & Zeldin, M. D. (2000). Characterization of $\mathrm{PM}_{2.5}$ and $\mathrm{PM}_{10}$ in the South Coast Air Basin of Southern California: Part 1: spatial variations. Journal of the Air and Waste Management Association, 50(12), 2034–2044.
Li, J., Wang, Z., Wang, X., Yamaji, K., Takigawa, M., Kanaya, Y., et al. (2011). Impacts of aerosols on summertime tropospheric photolysis frequencies and photochemistry over Central Eastern China. Atmospheric Environment, 45(10), 1817–1829.
Lonati, G., Giugliano, M., & Ozgen, S. (2008). Primary and secondary components of in Milan (Italy). Environment International, 34(5), 665–670.
Lu, B., Kong, S., Han, B., Wang, X., & Bai, Z. (2011). Inventory of atmospheric pollutants discharged from biomass burning in China continent in 2007. China Environmental Science, 31(2), 186–194 (in Chinese).
Ma, J., Chu, C., Li, J., & Song, B. (2009). Heavy metal pollution in soils on railroad side of Zhengzhou–Putian section of Longxi–Haizhou railroad, China. Pedosphere, 19(1), 121–128.
Malm, W. C., Schichtel, B. A., Pitchford, M. L., Ashbaugh, L. L., & Eldred, R. A. (2004). Spatial and monthly trends in speciated fine particle concentration in the United States. Journal of Geophysical Research, 109, D03306.
Mamane, Y., & Gottlieb, J. (1992). Nitrate formation on sea-salt and mineral particles—A single particle approach. Atmospheric Environment Part A: General Topics, 26(9), 1763–1769.
Mazzei, F., D’Alessandro, A., Lucarelli, F., Nava, S., Prati, P., Valli, G., et al. (2008). Characterization of particulate matter sources in an urban environment. Science of the Total Environment, 401(1–3), 81–89.
Morawska, L., & Zhang, J. (2002). Combustion sources of particles, 1. Health relevance and source signatures. Chemosphere, 49(9), 1045–1058.
Norris, G., Vedantham, R., Katie, W., Steve, B., Prouty, J., & Foley, C. (Eds.). (2008). EPA positive matrix factorization (PMF) 3.0 fundamentals $\delta\sigma$ user guide. U.S. Environmental Protection Agency Office of Research and Development.
Perrone, M. G., Gualtieri, M., Ferrero, L., Porto, C. L., Udisti, R., Bolzacchini, E., et al. (2010). Seasonal variations in chemical composition and in vitro biological effects of fine PM from Milan. Chemosphere, 78(11), 1368–1377.
Pu, W., Zhao, X., Zhang, X., & Ma, Z. (2011). Effect of meteorological factors on $\mathrm{PM}_{2.5}$ during July to September of Beijing. Procedia Earth and Planetary Science, 2(0), 272–277.
Rees, S. L., Robinson, A. L., Khlystov, A., Stanier, C. O., & Pandis, S. N. (2004). Mass balance closure and the Federal Reference Method for $\mathrm{PM}_{2.5}$ in Pittsburgh, Pennsylvania. Atmospheric Environment, 38(20), 3305–3318.
Reff, A., Bhave, P. V., Simon, H., Pace, T. G., Pouliot, G. A., Mobley, J. D., et al. (2009). Emissions inventory of $\mathrm{PM}_{2.5}$ trace elements across the United States. Environmental Science and Technology, 43(15), 5790–5796.
Sahu, S. K., Beig, G., & Parkhi, N. S. (2011). Emissions inventory of anthropogenic PM2.5 and PM10 in Delhi during Commonwealth Games 2010. Atmospheric Environment, 45(34), 6180–6190.
Shi, G. L., Li, X., Feng, Y. C., Wang, Y. Q., Wu, J. H., Li, J., et al. (2009). Combined source apportionment, using positive matrix factorization-chemical mass balance and principal component analysis/multiple linear regression-chemical mass balance models. Atmospheric Environment, 43(18), 2929–2937.
Song, Y., Tang, X., Xie, S., Zhang, Y., Wei, Y., Zhang, M., et al. (2007). Source apportionment of PM2.5 in Beijing in 2004. Journal of Hazardous Materials, 146(1–2), 124–130.
Song, Y., Xie, S. D., Zhang, Y. H., Zeng, L. M., Salmon, L. G., & Zheng, M. (2006). Source apportionment of $\mathrm{PM}_{2.5}$ in Beijing using principal component analysis/absolute principal component scores and UNMIX. Science of the Total Environment, 372(1), 278–286.
Tao, J., Ho, K., Chen, L., Zhu, L., Han, J., & Xu, Z. (2009). Effect of chemical composition of PM2.5 on visibility in Guangzhou, China, 2007 Spring. Particuology, 7(1), 68–75.
Tian, H., Zheng, Y., Chen, H., Deng, W., & Du, Z. (2010). Inversion and analysis of aerosol optical depth in Zhengzhou. Meteorological Science and Technology, 38(04), 515–520 (in Chinese).
Tie, X., Wu, D., & Brasseur, G. (2009). Lung cancer mortality and exposure to atmospheric aerosol particles in Guangzhou, China. Atmospheric Environment, 43(14), 2375–2377.
Turpin, B. J., & Lim, H. J. (2001). Species contributions to $\mathrm{PM}_{2.5}$ mass concentrations: Revisiting common assumptions for estimating organic mass. Aerosol Science and Technology, 35(1), 602–610.
Vecchi, R., Marcazzan, G., & Valli, G. (2007). A study on nighttime-daytime $\mathsf{P M}_{10}$ concentration and elemental composition in relation to atmospheric dispersion in the urban area of Milan (Italy). Atmospheric Environment, 41(10), 2136–2144.
Wang, X., Bi, X., Sheng, G., & Fu, J. (2006). Chemical composition and sources of PM10 and $\mathrm{PM}_{2.5}$ aerosols in Guangzhou, China. Environmental Monitoring and Assessment, 119(1), 425–439.
Wang, Y., Zhuang, G., Tang, A., Yuan, H., Sun, Y., Chen, S., et al. (2005). The ion chemistry and the source of $\mathrm{PM}_{2.5}$ aerosol in Beijing. Atmospheric Environment, 39(21), 3771–3784.
Wang, Y., Zhuang, G., Zhang, X., Huang, K., Xu, C., Tang, A., et al. (2006). The ion chemistry, seasonal cycle, and sources of $\mathrm{PM}_{2.5}$ and TSP aerosol in Shanghai. Atmospheric Environment, 40(16), 2935–2952.
Watson, J. G., Zhu, T., Chow, J. C., Engelbrecht, J., Fujita, E. M., & Wilson, W. E. (2002). Receptor modeling application framework for particle source apportionment. Chemosphere, 49(9), 1093–1136.
Yatkin, S., & Bayram, A. (2007). Elemental composition and sources of particulate matter in the ambient air of a metropolitan city. Atmospheric Research, 85(1), 126–139.
Zhao, J., Zhang, F., Xu, Y., & Chen, J. (2011). Characterization of water-soluble inorganic ions in size-segregated aerosols in coastal city, Xiamen. Atmospheric Research, 99(3–4), 546–562.
Zheng, M., Salmon, L. G., Schauer, J. J., Zeng, L., Kiang, C. S., Zhang, Y., et al. (2005). Seasonal trends in $\mathrm{PM}_{2.5}$ source contributions in Beijing, China. Atmospheric Environment, 39(22), 3967–3976.
|
atmosphere
|
0022
|
10.1016/j.partic.2011.03.001
|
image
|
Fig. 1. Location of the sampling site in Tianjin, China.
|
|
atmosphere
|
0022
|
10.1016/j.partic.2011.03.001
|
table
|
Table 1 Statistical summary of the measured species concentrations of $\mathrm{PM}_{2.5}$ in Tianjin.
|
|
atmosphere
|
0022
|
10.1016/j.partic.2011.03.001
|
table
|
Table 2 Comparison of TC, OC, and EC in Tianjin with those in other cities of China.
|
|
atmosphere
|
0022
|
10.1016/j.partic.2011.03.001
|
image
|
Fig. 2. Relationship between OC and EC of $\mathrm{PM}_{2.5}$ in Tianjin.
|
|
atmosphere
|
0022
|
10.1016/j.partic.2011.03.001
|
table
|
Table 3Comparison of element concentrations in major cities of China (ng/m3).
|
|
atmosphere
|
0022
|
10.1016/j.partic.2011.03.001
|
image
|
Fig. 3. Denary logarithm of element enrichment factors of $\mathrm{PM}_{2.5}$ in Tianjin.
|
|
atmosphere
|
0022
|
10.1016/j.partic.2011.03.001
|
image
|
Fig. 4. Average mass composition of 24-h $\mathrm{PM}_{2.5}$ samples in Tianjin.
|
|
atmosphere
|
0022
|
10.1016/j.partic.2011.03.001
|
table
|
Table 4 Pearson’s correlation coefficient matrix of elements in $\mathrm{PM}_{2.5}$
|
|
atmosphere
|
0022
|
10.1016/j.partic.2011.03.001
|
text
| Not supported with pagination yet
|
Chemical composition of $\mathsf{P M}_{2.5}$ during winter in Tianjin, China
Jinxia Gu a,b, Zhipeng Bai b,∗, Weifang Li b, Liping Wu a,b, Aixia Liu c, Haiyan Dong d, Yiyang Xie c
a Tianjin Institute of Urban Construction, Tianjin 300384, China
b State Environmental Protection Key Laboratory of Urban Ambient Air Particulate Matter Pollution Prevention and Control, College of Environmental Science and Engineering, Nankai
University, Tianjin 300071, China
c Tianjin Institute of Meteorological Instruments, Tianjin 300074, China
d Tianjin Environmental Monitoring Central Station, Tianjin 300191, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 December 2009 Received in revised form 1 July 2010 Accepted 3 March 2011
Keywords:
$\mathrm{PM}_{2.5}$
Water-soluble ions
Organic carbon (OC)
Elemental carbon (EC)
Crustal matter
$\mathrm{PM}_{2.5}$ samples for $24\,\mathrm{h}$ were collected during winter in Tianjin, China. The ambient mass concentration and chemical composition of the $\mathsf{P M}_{2.5}$ were determined. Ionic species were analyzed by ion chromatography, while carbonaceous species were determined with the IMPROVE thermal optical reflectance (TOR) method, and inorganic elements were measured by inductively coupled plasma-atomic emission spectrometer. The daily $\mathrm{PM}_{2.5}$ mass concentrations ranged from 48.2 to $319.2\,\upmu\mathrm{g}/\mathrm{m}^{3}$ with an arithmetic average of $144.6\,\upmu\mathrm{g}/\mathrm{m}^{3}$ . The elevated $\mathrm{PM}_{2.5}$ in winter was mostly attributed to combustion sources such as vehicle exhaust, heating, cooking and industrial emissions, low wind speeds and high relative humidity (RH), which were favorable for pollutant accumulation and formation of secondary pollutants. By chemical mass balance, it was estimated that about $89.1\%$ of the $\mathrm{PM}_{2.5}$ mass concentrations were explained by carbonaceous species, secondary particles, crustal matters, sea salt and trace elements. Organic material was the largest contributor, accounting for about $32.7\%$ of the total $\mathsf{P M}_{2.5}$ mass concentrations. $S0_{4}{}^{2-}$ , $\mathsf{N O}_{3}^{\mathrm{~-~}}$ , $C1^{-}$ and $\mathsf{N H_{4}}^{+}$ were four major ions, accounting for $16.6\%$ , $11.5\%$ , $4.7\%$ and $6.0\%$ , respectively, of the total mass of $\mathrm{PM}_{2.5}$ .
$\copyright$ 2011 Chinese Society of Particuology and Institute of Process Engineering, Chinese Academy of Sciences. Published by Elsevier B.V. All rights reserved.
1. Introduction
Ambient particulate matters (PM), whether from anthropogenic or natural sources, have a pronounced effect on atmospheric chemistry, air quality and climate change. Especially, atmospheric fine particles $\mathrm{PM}_{2.5}$ , particles with an aerodynamic diameter of $2.5\,\upmu\mathrm{m}$ or less), have been found to play an important role in global climate change (Charlson et al., 1992; Wexler & Ge, 1998), human health problem (Dockery & Stone, 2007; Norris et al., 1999; Ostro, Broadwin, Green, Feng, & Lipsett, 2006; Schwartz, Dockery, & Neas, 1996) and visibility degradation (Chan et al., 1997; Chow et al., 1996; Sisler & Malm, 1994). Major components of $\mathsf{P M}_{2.5}$ , reported as sulfate $(\mathsf{S}0_{4}{}^{2-})$ , nitrate $\left(\mathsf{N O}_{3}^{\mathrm{~-}}\right)$ , ammonium $\left(\mathrm{NH}_{4}^{+}\right)$ , organic carbon (OC) and elemental carbon (EC), were the significant causes of visibility degradation (Cao et al., 2007; Chan et al., 1997; Chow et al., 1996; Conner, Bennett, Weathers, & Wilson, 1991; Tanner, Parkhurst, Valente, & Phillips, 2004).
In China, a combination of rapid industrialization and high population density has inevitably made the air pollution problem deteriorate, of which PM has been frequently observed as the principal pollutant in most urban area. Tianjin is the largest coastal city of north China, located about $120\,\mathrm{km}$ southeast of Beijing. It has a total population of over 10 million with major industries including automobile, petrochemical, metallurgy, energy, electronics and medicine. Like many other well-developed cities, such as Beijing, Shanghai, Guangzhou, and Hong Kong (Cao et al., 2003, 2004; He et al., 2001; Louie, Chow, et al. 2005; Louie, Watson, et al., 2005; Ye et al., 2003), Tianjin is also faced with serious problems of particulate matter pollution and poor visibility. What are the major components of $\mathsf{P M}_{2.5}$ in Tianjin? What are the main reasons for poor visibility? These problems call for solution. So the $\mathsf{P M}_{2.5}$ mass and its components are collected in order to identify the main sources influencing the $\mathsf{P M}_{2.5}$ pollution level during winter in Tianjin.
2. Sampling and analysis
2.1. Sampling site
The 24-h (09:00 to 09:00 local time) $\mathsf{P M}_{2.5}$ samples were collected at an urban sampling site (Fig. 1) situated at the Atmospheric Boundary-layer Observation Station of Tianjin, which is located in a commercial-residential area and approximately $200\,\mathrm{m}$ away from a major roadway. There are no high buildings and factories around, and it has natural ventilation and no special contamination. The sampling devices of particulate matter were situated on the second floor (about $10\,\mathrm{m}$ above the ground) of the meteorological observation tower (height $225\,\mathrm{m}$ ).
2.2. Sample collection
Daily $\mathsf{P M}_{2.5}$ samples were respectively collected on $90\,\mathrm{mm}$ quartz-fiber and $90\,\mathrm{mm}$ polypropylene-fiber fliters with middleflow impact samplers (TH 150AII,Wuhan Tianhong) operating at a flow rate of $1001/\mathrm{min}$ from January 6 to January 24 in 2008. The quartz fliters were pre-heated in a muffle furnace at $600\,^{\circ}\mathrm{C}$ for $^{3\,\mathrm{h}}$ while the polypropylene-fiber fliters in a oven at $60\,^{\circ}\mathrm{C}$ for $1\,\mathrm{h}$ before sampling to remove volatile components. Before and after sampling, the fliters were equilibrated in a dessicator (temperature $20{-}23\,^{\circ}\mathrm{C}$ and relative humidity $35\mathrm{-}45\%)$ for $48\,\mathrm{h}$ , and then weighed on an electronic microbalance with a $\pm1\,\upmu g$ sensitivity (Mettler Toledo Inc., Switzerland) to determine the PM mass. Each fliter was weighed at least three times before and after sampling, and the net mass was obtained by subtracting the pre-sampling weight from the post-sampling weight. Difference among replicate weights were ${<}10\,\upmu g$ for blanks and ${<}20\,\upmu g$ for samples. After weighing, the samples were placed in a refrigerator at $4\,^{\circ}\mathrm{C}$ until analysis.
The meteorological data (wind speed and RH) were obtained from Tianjin Institute of Meteorological Instruments, and the data of $S0_{2}$ (MODEL 4208, Beijing Zhong Sheng Tai Ke) and $\mathsf{N O}_{2}$ (MODEL 2208, Beijing Zhong Sheng Tai Ke) from Tianjin Environmental Monitoring Central Station, respectively.
2.3. Chemical analysis
2.3.1. Ionic composition
One-fourth of each quartz-fiber fliter sample was extracted with $10\,\mathrm{ml}$ ultra-pure water in an ultrasonic bath for $30\,\mathrm{min}$ . The extracts were stored at $4\,^{\circ}\mathbf{C}$ in a pre-cleaned tube before analysis. Ion chromatography (ICS-1500, Dionex) was used to analyze the three anions $\left(\mathbf{Cl}^{-}\right)$ , $\mathrm{NO}_{3}{}^{-}$ and ${\mathrm{SO}}_{4}{}^{2-}\,.$ ) and five cations $(\boldsymbol{\mathrm{N}}\boldsymbol{\mathrm{a}}^{+}$ $\mathsf{N H}_{4}^{\,\,+}$ , $K^{+}$ , ${\mathrm{Mg}}^{2+}$ and ${\bf C}{\bf a}^{2+}$ ). The analytical and guard columns were AS14 and AG14 for anions, and CS12A and CG12A for cations, respectively. The eluent used for anion was a $3.5\,\mathrm{mM}$ carbonate $(\mathrm{Na}_{2}\mathrm{CO}_{3})/1.0\,\mathrm{mM}$ bicarbonate $\mathrm{(NaHCO_{3}}$ ) solution, while for cation a $20\,\mathrm{mM}$ methanesulfonic acid solution. The limits of detection were less than $0.05\,\mathrm{mg/l}$ for both anions and cations. Uncertainties were $\pm10\%$ for all ions.
2.3.2. Carbon species
OC and EC were analyzed by Desert Research Institute (DRI) Model 2001 Thermal/Optical Carbon Analyzer (Chow et al., 1993; Chow et al., 2004; Chow, Watson, Crow, Lowenthal, & Merrifield, 2001; Chow, Watson, Louie, Chen, & Sin, 2005; Fung, Chow, & Watson, 2002). In the analysis, a $0.5\,\mathrm{cm}^{2}$ punch from the quartzfiber fliter sample was analyzed and the four OC fractions (OC1, OC2, OC3, and OC4) were respectively obtained at 120, 250, 450, and $550\,^{\circ}\mathrm{C}$ in a helium atmosphere; the pyrolyzed carbon fraction (OP) was determined when a reflected laser light attained its original intensity after $0_{2}$ was added to the analysis atmosphere; and three EC fractions (EC1, EC2, and EC3) were respectively obtained at 550, 700, and $800\,^{\circ}\mathrm{C}$ in a $2\%\,0_{2}/98\%$ He atmosphere. Each day the analyzer was calibrated with known quantities of $\mathsf{C H}_{4}$ . One sample per group of 10 samples was repeated. The difference determined from replicate analyses was smaller than $5\%$ for TC, and $10\%$ for OC and EC. In all the procedure, the blanks fliters were also analyzed to get the blank OC and EC concentrations. The average blank concentrations were used to correct the sample results.
2.3.3. Inorganic elements
One-fourth of each polypropylene-fiber fliter sample was extracted with $6\,\mathrm{ml}\,\mathrm{HNO}_{3}$ and $2\,\mathrm{ml}\,\mathrm{HCl}$ in a microwave laboratory system (ETHOS, Milestone) for $30\,\mathrm{min}$ , with power of 1400 W, maximal temperature of $170\,^{\circ}\mathrm{C}$ and ultimate pressure of 20 bar. After element extraction from the polypropylene-fiber fliter sample, 14 elements (Si, Mg, Fe, Al, Mn, Ca, V, Co, Ni, Cu, Zn, Cd, Ba and Pb) were measured by using inductively coupled plasma-atomic emission spectrometer (ICP-AESIRIS, Intrepid II, Thermo Electron). The sample solutions were measured in triplicates, and quality controls and blanks were inserted for every 10 samples. The relative standard deviations of the measured element concentrations were typically $<\!5\%$ . Precision and bias were $<\!10\%$ . Element concentrations of the procedural blanks were generally $<\!5\%$ of the samples.
3. Results and discussion
3.1. $P M_{2.5}$ mass concentrations
The $24\mathrm{-h}$ integrated $\mathsf{P M}_{2.5}$ mass concentrations ranged from 48.2 to $319.2\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , with an arithmetic average of $144.6\,\upmu\mathrm{g}/\mathrm{m}^{3}$ in winter. Because State Environmental Protection Agency of China (SEPA) has not established the standard for $\mathsf{P M}_{2.5}$ , the concentrations of $\mathsf{P M}_{2.5}$ in Tianjin were compared to the Ambient Air Quality Standards (AAQS) of Class II $(35\,\upmu\mathrm{g}/\mathrm{m}^{3})$ promulgated by United States Environmental Protection Agency (USEPA) for $\mathsf{P M}_{2.5}$ .
We found that the $19\;\mathrm{PM}_{2.5}$ samples all exceeded the standard of USEPA, and furthermore, the pollution level of $\mathsf{P M}_{2.5}$ during winter in Tianjin was $1.4{-}9.1$ times the standard of USEPA.
Auto-monitoring data of gaseous pollutants $S0_{2}$ and $\mathsf{N O}_{2}$ ) for the same sampling period were available. By the linear regression analysis, the correlation coefficient was 0.64 between $\mathsf{P M}_{2.5}$ and $S0_{2}$ , suggesting an important influence of coal combustion on $\mathsf{P M}_{2.5}$ during winter. The high level of $S0_{2}$ coincided with the so-called “heating season” in Tianjin as well as most cities in northern China (November 15 to March 15 each year) when the emissions of coalrelated secondary sulfate and coal fly ash from residential heating to the atmosphere increased dramatically. On the other hand, the correlation coefficient was 0.58 between $\mathrm{PM}_{2.5}$ and $\mathrm{NO}_{2}$ . $\mathsf{N O}_{2}$ was generally considered to be mobile-source emissions related. Mobile-source emissions were likely to increase somewhat in the winter due to cold start emissions.
Moreover, the concentrations of $\mathrm{PM}_{2.5}$ varied significantly in different sampling days due to influence of meteorological conditions. The occurrences of terrifically elevated concentration episodes were always associated with low wind speed and high relative humidity. This coincided with the former results when wind speed was low, the atmosphere was stable and slow dispersion occurred, leading to accumulation of pollutants, and high RH, too, might be favorable for the secondary particle formation (Yao et al., 2002).
3.2. Ionic components
The analytical results of the major ionic components $(\mathrm{Cl}^{-},\mathrm{NO}_{3}{}^{-}$ , $S0_{4}{}^{2-}$ , $\mathsf{N a}^{+}$ , $\mathsf{N H}_{4}^{+}$ , $K^{+}$ , ${\mathrm{Mg}}^{2+}$ and ${\bf C}{\bf a}^{2+}$ ) are presented in Table 1.
The $S0_{4}{}^{2-}$ was the most abundant ionic species, ranging in concentrations from 11.2 to $37.9\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , with an arithmetic average of $24.1\,\upmu\mathrm{g}/\mathrm{m}^{3}$ . Because the $S0_{4}{}^{2-}$ came from sea salts or was formed by oxidation of sulfur dioxide, using $\mathsf{N a}^{+}$ as a tracer of sea salt, excess sulfate (non-sea-salt sulfates $(\mathrm{nss}{-}\mathrm{SO}_{4}{}^{2-})$ ) can be calculated as nss- $\mathrm{SO}_{4}^{\,2-}=[\mathrm{SO}_{4}^{\,2-}]-[\mathrm{Na}^{+}]\times0.2516$ . According to this calculation, the non-sea-salt sulfates over the total sulfate in $\mathsf{P M}_{2.5}$ were over $90\%$ in winter, indicating that although Tianjin was located in coastal areas, the contribution of sea salts to sulfate in fine particles was small and a substantial anthropogenic origin might contribute to that. The sulfur oxidation ratio $\mathrm{SOR}=[S\mathrm{O}_{4}^{\mathrm{~2-}}]/([S\mathrm{O}_{2}]+[S\mathrm{O}_{4}^{\mathrm{~2-}}])$ could be used to assess the transformation degree of gaseous $S0_{2}$ to aqueous sulfate. The value of SOR was lower than 0.1 in primary pollutants (Wang et al., 2005). Higher SOR suggested the oxidation of gaseous species and more secondary aerosols in the atmosphere. In this study, the average SOR was 0.19 in Tianjin winter, indicating that gas-particle conversion and further condensation or adsorption on the particle surface was likely the major source of secondary sulfate particles in Tianjin (Wang, Huang, Gao, Gao, & Wang, 2002; Yao et al., 2002).
The $\mathsf{N O}_{3}^{\ensuremath{-}}$ is the second ionic species of $\mathsf{P M}_{2.5}$ in Tianjin, ranging in concentration from 5.0 to $29.7\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , with an arithmetic average of $16.6\,\upmu\mathrm{g}/\mathrm{m}^{3}$ . The nitrogen oxidation ratio $\mathrm{NOR}\,{=}\,[\mathrm{NO_{3}}^{-}]/([\mathrm{NO_{2}}]\,{+}\,[\mathrm{NO_{3}}^{-}])$ could be used to assess the transformation degree of gaseous $\mathsf{N O}_{2}$ to aqueous nitrate. In this study, the NOR value ranged between 0.08 and 0.29 with a mean of 0.19, also indicating the formation of $\mathsf{N O}_{3}^{\mathrm{~-~}}$ from $\mathsf{N O}_{2}$ . It was also found that the gas-particle partition of ambient $\mathsf{N O}_{3}^{\ensuremath{-}}$ strongly depended on such meteorological conditions as temperature and RH. In this study, the average temperature and RH were $-3.1\,^{\circ}\mathsf{C}$ and $50\%$ , respectively during winter in Tianjin. The meteorology was expected to result in a large variation of $\mathsf{N O}_{3}^{\ensuremath{-}}$ in $\mathsf{P M}_{2.5}$ .
The $C1^{-}$ concentrations ranged from 2.3 to $11.4\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , with an arithmetic average of $6.7\,\upmu\mathrm{g}/\mathrm{m}^{3}$ . The concentration ratio of $\mathrm{{Cl}^{-}/\mathrm{{Na}^{+}}}$ was about 1.2 in sea water (Tsitouridou & Samara, 1993). But in this study, the average ratio of $\mathrm{{Cl}^{-}/\mathrm{{Na}^{+}}}$ was 2.2, indicating that most of the chloride in $\mathsf{P M}_{2.5}$ may not be present as sea salt particles. Possibly, it originated from industrial contamination sources, or else, possibly the chloride originated from large sea salt particles, but was displaced by reaction with nitric acid and followed by reaction with ammonia to form small particles (Wang & Hu, 2000).
The $\mathsf{N H}_{4}^{+}$ concentrations ranged from 5.4 to $15.0\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , with an arithmetic average of $8.7\,\upmu\mathrm{g}/\mathrm{m}^{3}$ . The neutralization ratio $\mathrm{NR}=[\mathrm{NH_{4}}^{+}]/([\mathrm{nss}-\mathrm{SO_{4}}^{2-}]+[\mathrm{NO_{3}}^{-}])$ was employed to express the degree of neutralization of aerosol acidity (Yuan, Sau, & Chen, 2004). In this study, the average NR value of 0.3 indicated that the aerosol particles collected in this study were acidic.
The arithmetic average concentrations of $\mathsf{N a}^{+}$ , ${\mathrm{Mg}}^{2+}$ and ${\bf C}{\bf a}^{2+}$ were 3.4, 1.0 and $1.8\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , respectively. Sea salt emission was usually regarded as the major contribution to $\mathsf{N a}^{+}$ , ${\mathrm{Mg}}^{2+}$ and ${\bf C}{\bf a}^{2+}$ at coastal areas. On the other hand, soil and dust particles in north China generally contributed to those ions (Wang et al., 2002). In this study, $\mathsf{N a}^{+}$ , ${\mathrm{Mg}}^{2+}$ and ${\bf C}{\bf a}^{2+}$ only accounted for a total of $4.2\%$ in $\mathsf{P M}_{2.5}$ , suggesting that the contributions of sea salt, soil and dust to $\mathsf{P M}_{2.5}$ were very small during winter in Tianjin. The arithmetic average concentration of $K^{+}$ was the lowest. Because the agriculture field vegetation burning mainly occurred during harvest season in north China, in winter, the lowest abundance $(0.6\%)$ of soluble potassium for $\mathsf{P M}_{2.5}$ was presented to indicate minimal impact of biomass burning.
3.3. Carbonaceous species
Measurement results of $24\mathrm{-h}$ average concentrations of OC and EC in Tianjin are given in Table 1. The concentrations of OC ranged between 11.7 and $77.6\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , with an average of $33.8\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , while EC between 3.4 and $19.6\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , with an average of $9.1\;\upmu\mathrm{g}/\mathrm{m}^{3}$ , respectively.
The OC and EC concentrations in this study were compared with those measured in other cities of China in Table 2. In winter, the OC concentration was higher than that at Guangzhou, Shenzhen, Zhuhai, Hong Kong and Shanghai, Beijing, Taiyuan but lower than that at Xi’an. The severity of organic aerosol pollution was likely to be related to coal burning for heating in winter. On the other hand, the temperature was low and favorable for adsorption and coagulation of organic gases on the particle surface; moreover, the low mixing layer height in winter could result in stagnation of the SOC (secondary organic carbon) precursors and SOC formation (Strader, Lurman, & Pandis, 1999). However, the EC concentration was higher than that at Guangzhou, Shenzhen, Shanghai, Zhuhai, Hong Kong and Taiyuan, but lower than that at Beijing and Xi’an,
The relationship between OC and EC could help identify the origins of carbonaceous $\mathsf{P M}_{2.5}$ (Chow et al., 1996; Turpin & Huntzicker, 1991). Therefore, the mass ratio of OC to EC (OC/EC) has been used to study emission and transformation characteristics of carbonaceous aerosol. The OC/EC ratio exceeding 2.0–2.2 has been used for identification and evaluation of secondary organic aerosols (Chow et al., 1996; Turpin & Huntzicker, 1991). In this study, the average OC/EC ratio was 3.7, indicating the possible presence of secondary organic carbon (SOC). As shown in Fig. 2, the good correlation coefficient (R) of OC/EC was 0.92 during winter in Tianjin, showing that the sources of OC and EC in $\mathsf{P M}_{2.5}$ were relatively simple: possibly, commercial coal combustion and motor-vehicle exhaust were responsible for that in winter.
Because the organic carbon might have been derived from emitted particles as well as secondary organic aerosol, it was important to confirm the contributions of the primary and secondary organic carbon to carbonaceous aerosol for controlling particulate pollution. Because there was no simple direct analytical technique to analyze secondary organic carbon, several indirect methodologies have been applied to evaluation of secondary organic carbon in ambient aerosols (Castro, Pio, Harrison, & Smithd, 1999; Pandis, Harley, Cass, & Seinfeld, 1992; Turpin & Huntzicker, 1991). Following the Castro’s equation, the concentration of SOC could be calculated by the experimentally derived equation:
$$
\mathsf{S O C}=\mathsf{O C}-\mathsf{E C}\left(\frac{0\mathsf{C}}{\mathsf{E C}}\right)_{\mathrm{min}}.
$$
In this study, the observed minimum ratio of OC/EC was 2.1 in Tianjin winter, from which the calculated concentrations of SOC are shown in Table 1, with an average of $14.6\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , accounting for $10.1\%$ of $\mathsf{P M}_{2.5}$ in Tianjin winter. These results suggested that secondary organic aerosol might be a significant contributor to fine organic particles in Tianjin winter. This might be attributed to the enhanced emission of organic precursors, as well as the low mixing layer height that resulted in stagnation of the SOC precursors and SOC formation (Dan, Zhuang, Li, Tao, & Zhuang, 2004; Duan et al., 2005; Strader et al., 1999).
3.4. Inorganic elements
While the average concentrations of 14 elements (Si, Fe, Al, Ca, Mg, Mn, V, Co, Ni, Cu, Zn, Cd, Ba and Pb) at sampling site were already listed in Table 1, to better understand the pollution level of elements associated on $\mathsf{P M}_{2.5}$ in Tianjin, some studies in other cities of China were listed in Table 3, indicating that the crustal elements, such as Al, Si and Fe were lower in Tianjin than those measured in other cities. This might be due to less surface soil exposure and also greater snow coverage during winter in Tianjin. While other elements derived from anthropogenic pollution, such as Cu, Zn, Cd and Pb, were in the same level among the different cities. The elevated level of elements during winter might be due to heating and low wind speed, which may lead to an accumulation of exhaust emissions. Although the amount of wind-blown particles from crustal origin was likely to be reduced when the wind speed was low enough, crustal elements might be increased owing to the accumulation of re-suspended road dust.
Table 4 provides Pearson’s correlation coefficients between elements in Tianjin (in all cases $[P(r,n)\!<\!0.01]$ ). The Pearson’s correlation coefficients were 0.90 and 0.76 between Al and Si, Al and Ca, respectively, indicating that they mainly originated from the crustal sources, for they were typical crustal elements. The element Cd had good relationship with V, Co, Ni, Cu, Zn and Ba, indicating that these elements were likely related to urban anthropogenic sources, such as fuel combustion and waste incineration. The good correlation between $\mathsf{P b}$ and $Z\mathfrak{n}$ proved they mainly came from the traffic sources and non-ferrous metal smelting.
Enrichment factors (EF) of trace elements in $\mathsf{P M}_{2.5}$ relative to the earth’s crust were calculated to indicate the contribution of anthropogenic sources and natural sources to the air pollution (Gao et al., 2002; Zhang et al., 2002). Then the method was applied to evaluate the contribution of crustal and non-crustal sources. In this study, Al was selected as the reference element. For each element, the EF was calculated according to the equation given below:
$$
\mathrm{EF}=\frac{({E/A l})_{\mathrm{sample}}}{({E/A l})_{\mathrm{crust}}},
$$
$(E/\mathsf{A l})_{\mathsf{s a m p l e}}$ and $(E/\mathsf{A l})_{\mathrm{crust}}$ were the concentration ratio of element $E$ to Al in the atmospheric particulate and in Earth’s crust, respectively. The average values of denary logarithm of EFs for $\mathsf{P M}_{2.5}$ were shown in Fig. 3. In this study, the EF values of Si and Fe were both below 10, indicating these elements were mainly contributed by crustal sources. The EF values of Ca, Mg, Mn and V were between 10 and 100, suggesting that both natural emissions and anthropogenic sources were important for these elements. The EF values of Co, Ni, Cu, Zn, Cd, Ba and $\boldsymbol{\mathrm{Pb}}$ were larger than 100, indicating they mainly originated from non-crustal sources such as vehicular exhaust and industrial emission.
3.5. $P M_{2.5}$ mass balance
Normally, $\mathsf{P M}_{2.5}$ could be classified into six major types: major water soluble ions, elemental carbon, organic matter, sea salt, crustal matter and trace species. Calculation of mass balance was described in detail elsewhere (Chan et al., 1997; Christoforou, Salmon, Hannigan, Solomon, & Cass, 2000). In brief, organic matter was obtained by multiplying OC by a factor of 1.4, as is commonly used to estimate the unmeasured hydrogen and oxygen in organic compounds. Soluble Na was assumed to come solely from sea salt, thus the mass of sea salt was estimated as $2.54\times\mathsf{N a}^{+}$ . The mass of crustal matter was estimated on the basis of the oxides of Al, Ca, Si, Ti, Fe, Mg, Na and K as follows:
Crustal matter $\dot{}=1.16(1.90\mathrm{Al}+1.41\mathrm{Ca}+2.09\mathrm{Fe}+2.15\mathrm{Si}+1.67\mathrm{Ti}),$ where the factor 1.16 was to compensate for the exclusion of $\tt M g O$ , $\tt N a_{2}O$ , $\mathtt{K}_{2}0$ and $\mathrm{H}_{2}\mathrm{O}$ from the crustal mass calculation. Sum of all other elements was defined as trace species, including Ni, Co, V, Cu, Zn, Ba, Mg, Mn and Pb in this study. So the $\mathsf{P M}_{2.5}$ mass balance was shown in Fig. 4.
Approximately $89.1\%$ of the $\mathsf{P M}_{2.5}$ mass concentrations were accounted for by nine components during winter in Tianjin. As noted earlier, of all the species, organic matter was the most predominant fraction accounting for about $32.7\%$ of $\mathsf{P M}_{2.5}$ . The higher OC contribution may indicate the impact of residential cooking and heating on $\mathsf{P M}_{2.5}$ , while EC accounted for about $6.0\%$ of $\mathsf{P M}_{2.5}$ . Four major secondary ionic species (sulfate, nitrate, ammonium and chloride) contributed $16.6\%,11.5\%,6.0\%$ and $4.7\%$ , respectively. Crustal matter fraction was $4.2\%$ . The unidentified portion averaging $10.9\%$ of the $\mathsf{P M}_{2.5}$ mass might have been due to systematic weighing errors. Although the trace metal elements typically constituted a much smaller fraction of $\mathsf{P M}_{2.5}$ than the other components such as carbon and ions, their potential adverse harm to human should deserve great attention in future studies.
4. Conclusions
(1) The average $\mathsf{P M}_{2.5}$ mass concentration was $144.6\,\upmu\mathrm{g}/\mathrm{m}^{3}$ in Tianjin winter, indicating that fine particulate pollution was serious. The high $\mathsf{P M}_{2.5}$ was attributed to the combined action of combustion, low wind speed and high relative humidity.
(2) $S0_{4}{}^{2-}$ , $\mathsf{N O}_{3}^{\mathrm{~-~}}$ , $C1^{-}$ and $\mathsf{N H}_{4}^{\,\,+}$ were the major ionic components, accounting for $16.6\%$ , $11.5\%$ , $4.7\%$ and $6.0\%$ , respectively of the total $\mathsf{P M}_{2.5}$ mass.
(3) The average SOC concentration was $14.6\,\upmu\mathrm{g}/\mathrm{m}^{3}$ in $\mathsf{P M}_{2.5}$ , and the average ratio of SOC/OC was $61.7\%$ , indicating that SOC was an important contributor to $\mathsf{P M}_{2.5}$ in Tianjin.
(4) Analysis based on Pearson correlation and enrichment factors indicated that the associated contributions of crustal and noncrustal sources were to the inorganic elements.
(5) By chemical mass balance, approximately $89.1\%$ of the $\mathsf{P M}_{2.5}$ mass concentrations were contributed by nine components. Organic matter was the most predominant fraction accounting for about $32.7\%$ of $\mathsf{P M}_{2.5}$ .
Acknowledgements
This project is supported by National Natural Science Foundation of China (Grant No. 20677030), and Tianjin Science and Technology Development Commission (Grant No. 06YFSYSF02900).
References
Cao, J. J., Lee, S. C., Chow, J. C., Watson, J. G., Ho, K. F., Zhang, R. J., et al. (2007). Spatial and seasonal distributions of carbonaceous aerosols over China. Journal of Geophysical Research, 112, D22S11. doi:10.1029/2006JD008205
Cao, J. J., Lee, S. C., Ho, K. F., Zhang, X. Y., Zou, S. C., Fung, K. K., et al. (2003). Characteristics of carbonaceous aerosol in Pearl River Delta region China during 2001 winter period. Atmospheric Environment, 37, 1451–1460.
Cao, J. J., Lee, S. C., Ho, K. F., Zou, S. C., Fung, K. K., Li, Y., et al. (2004). Spatial and seasonal variations of atmospheric organic carbon and elemental carbon in Pearl River Delta Region China. Atmospheric Environment, 38, 4447–4456.
Cao, J. J., Wu, F., Chow, J. C., Lee, S. C., Li, Y., Chen, S. W., et al. (2005). Characterization and source apportionment of atmospheric organic and elemental carbon during fall and winter of 2003 in Xi’an China. Atmospheric Chemistry and Physics, 5, 3127–3137.
Castro, L. M., Pio, C. A., Harrison, R. M., & Smithd, R. T. (1999). Carbonaceous aerosol in urban and rural European atmospheres: Estimation of secondary organic carbon concentrations. Atmospheric Environment, 33, 2771–2781.
Chan, Y. C., Simpson, R. W., Mctainsh, G. H., Vowles, P. D., Cohen, D. D., & Bailey, G. M. (1997). Characterization of chemical species in $\mathrm{PM}_{2.5}$ and $\mathrm{PM}_{10}$ aerosols in Brisbane Australia. Atmospheric Environment, 31, 2061–2080.
Charlson, R. J., Schwartz, S. E., Hales, J. M., Cess, R. D., Coakley, J. A., Jr., Hansen, J. E., et al. (1992). Climate forcing by anthropogenic aerosols. Science, 255, 423–430.
Chow, J. C., Watson, J. G., Chen, L.-W. A., Arnott, W. P., Moosmüller, H., & Fung, K. K. (2004). Equivalence of elemental carbon by thermal/optical reflectance and transmittance with different temperature protocols. Environmental Science $\delta\mathcal{T}$ Technology, 38, 4414–4422.
Chow, J. C., Watson, J. G., Crow, D., Lowenthal, D. H., & Merrifield, T. M. (2001). Comparison of IMPROVE and NIOSH carbon measurements. Aerosol Science $\delta\mathcal{T}$ Technology, 34, 23–34.
Chow, J. C., Watson, J. G., Louie, P. K. K., Chen, L.-W. A., & Sin, D. (2005). Comparison of $\mathrm{PM}_{2.5}$ carbon measurement methods in Hong Kong, China. Environmental Pollution, 137, 334–344.
Chow, J. C., Watson, J. G., Lu, Z., Lowenthal, D. H., Frazier, C. A., Solomon, P. A., et al. (1996). Descriptive analysis of $\mathrm{PM}_{2.5}$ and $\mathrm{PM}_{10}$ at regionally representative locations during SJVAQS/AUSPEX. Atmospheric Environment, 30, 2079–2112.
Chow, J. C., Watson, J. G., Pritchett, L. C., Pierson, W. R., Frazier, C. A., & Purcell, R. G. (1993). The DRI thermal/optical reflectance carbon analysis system: Description, evaluation and applications in U.S. air quality studies. Atmospheric Environment, 27A, 1185–1201.
Christoforou, C. S., Salmon, L. G., Hannigan, M. P., Solomon, P. A., & Cass, G. R. (2000). Trends in fine particle concentration and chemical composition in southern California. Journal of the Air and Waste Management Association, 50, 43–53.
Conner, W. D., Bennett, R. L., Weathers, W. S., & Wilson, W. E. (1991). Particulate characteristics and visual effects of the atmosphere at Research Triangle Park. Journal of the Air and Waste Management Association, 41, 154–160.
Dan, M., Zhuang, G. S., Li, X. X., Tao, H. R., & Zhuang, Y. H. (2004). The characteristics of carbonaceous species and their sources in $\mathrm{PM}_{2.5}$ in Beijing. Atmospheric Environment, 38, 3443–3452.
Dockery, D. W., & Stone, P. H. (2007). Cardiovascular risks from fine particulate air pollution. The New England Journal of Medicine, 356, 511–513.
Duan, F. K., He, K. B., Ma, Y. L., Jia, Y. T., Yang, F. M., Lei, Y., et al. (2005). Characteristics of carbonaceous aerosols in Beijing, China. Chemosphere, 60, 355–364.
Fung, K. K., Chow, J. C., & Watson, J. G. (2002). Evaluation of OC/EC speciation by thermal manganese dioxide oxidation and the IMPROVE method. Journal of the Air and Waste Management Association, 52, 1333–1341.
Gao, Y., Nelson, E. D., Field, M. P., Ding, Q., Li, H., Sherrell, R. M., et al. (2002). Characterization of atmospheric trace elements on $\mathrm{PM}_{2.5}$ particulate matter over the New York—New Jersey harbor estuary. Atmospheric Environment, 36, 1077–1086.
He, K., Yang, F., Ma, Y., Zhang, Q., Yao, X., Chan, C. K., et al. (2001). The characteristics of $\mathrm{PM}_{2.5}$ in Beijing, China. Atmospheric Environment, 35, 4959–4970.
Ho, K. F., Lee, S. C., Cao, J. J., Chow, J. C., Watson, J. G., & Chan, C. K. (2006). Seasonal variations and mass closure analysis of particulate matter in Hong Kong. Science of the Total Environment, 355, 276–287.
Louie, P. K. K., Chow, J. C., Chen, L.-W. A., Watson, J. G., Leung, G., & Sin, D. W. M. (2005). $\mathrm{PM}_{2.5}$ chemical composition in Hong Kong: Urban and regional variations. Science of the Total Environment, 338, 267–281.
Louie, P. K. K., Watson, J. G., Chow, J. C., Chen, L.-W. A., Sin, D. W. M., & Lau, A. K. H. (2005). Seasonal characteristics and regional transport of $\mathrm{PM}_{2.5}$ in Hong Kong. Atmospheric Environment, 39, 1695–1710.
Meng, Z. Y., Jiang, X. M., Yan, P., Lin, W. L., Zhang, H. D., & Wang, Y. (2007). Characteristics and sources of PM2.5 and carbonaceous species during winter in Taiyuan China. Atmospheric Environment, 41, 6901–6908.
Norris, G., YoungPong, S. N., Koenig, J. Q., Larson, T. V., Sheppard, L., & Stout, J. W. (1999). An association between fine particles and asthma emergency department visits for children in Seattle. Environmental Health Perspectives, 107, 489–493.
Ostro, B., Broadwin, R., Green, S., Feng, W. Y., & Lipsett, M. (2006). Fine particulate air pollution and mortality in nine California counties: Results from CALFINE. Environmental Health Perspectives, 114, 29–33.
Pandis, S. N., Harley, R. A., Cass, G. R., & Seinfeld, J. H. (1992). Secondary organic aerosol formation and transport. Atmospheric Environment, 26A, 2269–2282.
Schwartz, J., Dockery, D. W., & Neas, L. M. (1996). Is daily mortality associated specifically with fine particles? Journal of the Air and Waste Management Association, 46, 927–939.
Sisler, J. F., & Malm, W. C. (1994). The relative importance of aerosols to spatial and seasonal trends of impaired visibility in the United States. Atmospheric Environment, 28, 851–862.
Strader, R., Lurman, F., & Pandis, S. N. (1999). Evaluation of secondary organic aerosol formation in winter. Atmospheric Environment, 33, 4849–4863.
Tanner, R. L., Parkhurst, W. J., Valente, M. L., & Phillips, W. D. (2004). Regional composition of $\mathrm{PM}_{2.5}$ aerosols measured at urban, rural and “background” sites in the Tennessee valley. Atmospheric Environment, 38, 3143–3153.
Turpin, B. J., & Huntzicker, J. J. (1991). Secondary formation of organic aerosol in the Los Angeles Basin: A descriptive analysis of organic and elemental carbon concentrations. Atmospheric Environment, 25A, 207–215.
Tsitouridou, R., & Samara, C. (1993). First results of acidic and alkaline constituents determination in air particulates of Thessaloniki Greece. Atmospheric Environment, 27B, 313–319.
Wang, M., & Hu, M. (2000). Sea salt source contribution to the aerosol in Qingdao seashore area. Environmental Science, 21, 83–85 (in Chinese).
Wang, G. H., Huang, L. M., Gao, S. X., Gao, S. T., & Wang, L. S. (2002). Characterization of water-soluble species of $\mathsf{P M}_{10}$ and $\mathrm{PM}_{2.5}$ aerosols in urban area in Nanjing China. Atmospheric Environment, 36, 1299–1307.
Wang, Y., Zhuang, G. S., Tang, A. H., Yuan, H., Sun, Y. L., Chen, S., et al. (2005). The ion chemistry and the source of $\mathrm{PM}_{2.5}$ aerosol in Beijing. Atmospheric Environment, 39, 3771–3784.
Wei, F., Teng, E., Wu, G., Hu, W., Wilson, W. E., Chanpman, R. S., et al. (1999). Ambient concentrations and elemental compositions of $\mathsf{P M}_{10}$ and $\mathrm{PM}_{2.5}$ in four Chinese cities. Environmental Science and Technology, 33, 4188–4193.
Wexler, A. S., & Ge, Z. (1998). Hydrophobic particles can activate at lower relative humidity than slightly hygroscopic ones: A Kohler theory incorporating surface fixed charge. Journal of Geophysical Research, 103, 6083–6088.
Yao, X. H., Chan, C. K., Fang, M., Cadle, S., Chan, T., Mulawa, P., et al. (2002). The watersoluble ionic composition of $\mathrm{PM}_{2.5}$ in Shanghai and Beijing, China. Atmospheric Environment, 36, 4223–4234.
Ye, B., Ji, X., Yang, H., Yao, X., Chan, C. K., Cadle, S. H., et al. (2003). Concentration and chemical composition of $\mathrm{PM}_{2.5}$ in Shanghai for a 1-year period. Atmospheric Environment, 37(4), 499–510.
Yuan, C.-S., Sau, C.-C., & Chen, M.-C. (2004). Influence of Asian dusts on the physicochemical properties of atmospheric aerosols in Taiwan district—Using the Penghu Islands as an example. China Particuology, 2, 144–152.
Zhang, X. Y., Cao, J. J., Li, L. M., Arimoto, R., Cheng, Y., Huebert, B., et al. (2002). Characterization of atmospheric aerosol over XiAn in the South Margin of the Loess Plateau, China. Atmospheric Environment, 36, 4189–4199.
|
atmosphere
|
0023
|
10.1007/s11869-013-0232-9
|
image
|
Fig 1 The sampling locations in Jinan, China
|
|
atmosphere
|
0023
|
10.1007/s11869-013-0232-9
|
table
|
Table 1 Summary statistics of $\operatorname{PM}_{2.5}$ , OC, and EC mass concentration at two sampling sites in Jinan, China
|
|
atmosphere
|
0023
|
10.1007/s11869-013-0232-9
|
image
|
Fig. 2 The distribution of $\operatorname{PM}_{2.5}$ mass concentration at two sampling sites in Jinan, China
|
|
atmosphere
|
0023
|
10.1007/s11869-013-0232-9
|
table
|
Table 2 Average mass concentrations of inorganic elements at two sampling sites in Jinan, China
|
|
atmosphere
|
0023
|
10.1007/s11869-013-0232-9
|
image
|
Fig. 3 The seasonal average concentrations of major watersoluble ions for $\operatorname{PM}_{2.5}$ in Jinan, China (in micrograms per cubic meter). a Spring. b Summer. c Fall. d Winter
|
|
atmosphere
|
0023
|
10.1007/s11869-013-0232-9
|
image
|
Fig. 4 The element enrichment factors of inorganic elements for $\operatorname{PM}_{2.5}$ in Jinan, China
|
|
atmosphere
|
0023
|
10.1007/s11869-013-0232-9
|
image
|
Fig. 5 The contribution percentages of the chemical species for $\operatorname{PM}_{2.5}$ in Jinan, China. a At EMS. b At SJU
|
|
atmosphere
|
0023
|
10.1007/s11869-013-0232-9
|
image
|
Fig. 6 PMF source profiles of secondary sulfate, soil dust, secondary nitrate and vehicle emissions, coal combustion, and biomass burning and in Jinan, China. a At EMS. b At SJU
|
|
atmosphere
|
0023
|
10.1007/s11869-013-0232-9
|
image
|
Fig. 6 (continued) Mass closure analysis
|
|
atmosphere
|
0023
|
10.1007/s11869-013-0232-9
|
text
| Not supported with pagination yet
|
Major chemical compositions, possible sources, and mass closure analysis of $\mathbf{PM}_{2.5}$ in Jinan, China
Jinxia Gu $\cdot$ Shiyong Du $\cdot$ Daowen Han $\cdot$ Lujian Hou $\cdot$ Jing Yi $\cdot$ Ja Xu $\cdot$ Guanghui Liu $\cdot$ Bin Han & Guangwu Yang $\cdot$ Zhi-Peng Bai
Received: 6 August 2013 /Accepted: 30 October 2013
$\copyright$ Springer Science+Business Media Dordrecht 2014
Abstract The fine particulate matter samples for $24~\mathrm{h}$ were carried out at the Environment Monitoring Station (EMS) and Shandong Jianzhu University (SJU) sites during 2010 in Jinan City, China. Eight water-soluble ion species were analyzed by ion chromatography, while organic carbon (OC) and elemental carbon (EC) were determined with the IMPROVE thermal optical reflectance method, and 20 inorganic elements were measured by inductively coupled plasma-atomic emission spectrometer and inductively coupled plasma-mass spectroscopy. The annual average mass concentration of $\operatorname{PM}_{2.5}$ was $168.85~\upmu\mathrm{g}\,\mathrm{~m}^{-3}$ at EMS and $148.67~\upmu\mathrm{g}\,\textrm{m}^{-3}$ at SJU. The coefficient of divergence was 0.14, 0.19, 0.23, and 0.23 in spring, summer, fall, and winter, respectively, indicating that there was no obvious spatial difference at the two sampling sites. The highest $\operatorname{PM}_{2.5}$ , OC, and OC/EC ratio were in winter because of the enhanced emissions from coal combustion for heating and poor atmospheric dispersion. By the method of enrichment factors, the 20 inorganic elements were divided into three types owing to their sources. Al, Si, and Ti were mainly contributed by crustal sources. Na, Mg, P, K, Ca, V, Cr, Mn, Fe, Co, Ni, Ba, and Sr were from both natural emissions and anthropogenic sources. Cu, Zn, Pb, and Sn mainly originated from anthropogenic sources such as vehicular exhaust and industrial emission. Chemical mass closure calculation estimated that $\mathrm{SO}_{4}^{\ 2-}$ was the largest contributor and explained $29.66\,\%$ of $\operatorname{PM}_{2.5}$ mass at EMS, while $31.64\,\%$ was at SJU. The organic matter, crustal matter, and $\mathrm{NO}_{3}^{\mathrm{~-~}}$ , respectively, accounted for 15.12, 12.87, and $13.77\ \%$ to $\operatorname{PM}_{2.5}$ at EMS, while it accounted for 13.46, 13.96, and $14.93\ \%$ at SJU, respectively. By the positive matrix factorization analysis, the coal combustion and biomass burning, secondary sulfate, soil dust, secondary nitrate, and vehicle emissions were identified as the major emission sources.
Keywords $\operatorname{PM}_{2.5}$ . Chemical species $\cdot$ Mass closure $\cdot$ Coefficient of divergence $\cdot$ Positive matrix factorization
Introduction
Atmospheric aerosol particles play an important role in air quality and climate system as well as the control of physical and chemical processes of the atmosphere. In recent decades, concerns had been raised and it indicated that atmospheric fine particles $\mathrm{PM}_{2.5}$ , particles with an aerodynamic diameter of $2.5\ \upmu\mathrm{m}$ or less) and their poisonous components (heavy metals, PAHs, etc.) were responsible for adverse health and environmental effects (Schwartz et al. 1996; Vedal 1997; Norris et al. 1999; Menon et al. 2002; Ostro et al. 2006; Dockery and Stone 2007; Duan et al. 2007; 2013). Airborne fine particles led to smog and haze formation that reduced visibility (Watson 2002; Kim et al. 2007). Major chemical components of $\mathrm{PM}_{2.5}$ had been reported as organic and element carbon, sulfate, and nitrate (Harrison 2004; Walker et al. 2004; Reiss et al. 2007) and were the significant causes of visibility degradation (Chow et al. 1996; Chan et al. 1997; Cao et al. 2007).
In China, studies for $\operatorname{PM}_{2.5}$ had been gradually carried out since 2000 and mainly focused on the urban areas of the Pearl River Delta, Beijing, Tianjin, and Xi’an. Some studies gave the general characteristics of $\operatorname{PM}_{2.5}$ chemical compositions and discussed their seasonal variations, correlations, or sources (He et al. 2001; Yao et al. 2002, 2003; Dan et al. 2004; Sun et al. 2004; Huang et al. 2006; Song et al. 2006; Cheng et al. 2009; Wang et al. 2009; Guo et al. 2010; Ianniello et al. 2011; Gu et al. 2011; Li et al. 2012; Tao et al. 2012). However, $\mathrm{PM}_{2.5}$ had seldom been sampled and chemically analyzed in Jinan City, and related published data for $\operatorname{PM}_{2.5}$ in urban areas of Jinan are also short. To better control the local aerosol pollution and implement the New National Ambient Air Quality Standard promulgated on February 2012, obtaining the information of local $\operatorname{PM}_{2.5}$ mass concentration, chemical composition, pollution source, and spatial and temporal variation is quite necessary.
Jinan is the capital of Shandong Province and one of the important city agglomerations in China. The total amount of population had exceeded 6.0 million inhabitants on a $153.5–\mathrm{km}^{2}$ area in 2010. Although the structures of fuel consumption have always been changing as result of the rapid economic development, coal is still the primary fuel in Jinan and widely used for industrial processes and daily life, especially more coal is combusted for heating between November and March due to the cold winter. The amount of motor vehicles now exceeds 1.3 million and is annually growing at a rate of nearly $19\,\%$ . The Jinan City experiences a continental monsoon climate, with hot, humid summer and dry, cold winter. The mean annual precipitation is $619\ \mathrm{mm}$ , and nearly half of the annual rainfall occurs in July and August. Rapid urbanization and economic development are deteriorating Jinan’s environmental air quality. Previous works about atmospheric fine particles in Jinan tended to the particle size fractions (Gao et al. 2007) and limited to chemical compositions. The main objectives of this study are to (1) characterize mass concentration variation for $\operatorname{PM}_{2.5}$ and major chemical compositions; (2) determine the relative contribution of chemical compositions by the method of mass closure; (3) discriminate the spatial difference for $\mathrm{PM}_{2.5}$ and its major chemical compositions; and (4) estimate the source contribution by using positive matrix factorization (PMF).
Materials and methods
Sampling sites
The locations of the two sampling sites were shown in Fig. 1
EMS
The sampling location is on the top floor of Jinan Environment Monitoring Station $\mathrm{[117^{\circ}2^{\prime}55^{\prime\prime}E}$ , $36^{\circ}39^{\prime}44^{\prime\prime}\mathrm{N})$ , about $15\,\mathrm{\,m}$
Springer
high above ground level in downtown area. It is close to residential buildings, so the pollutant from cooking emission will not be ignored. In addition, the concentration of contaminants in gaseous and condensed phase can be assumed to be strongly affected by the traffic due to the two major roadways of the district. The sampling site represents a mixed residential/ commercial/industrial area.
SJU
The sampling site is located on the rooftop of the science and technology building in Shandong Jianzhu University $\langle117^{\circ}11^{\prime}$ $9^{\prime\mathrm{{p}}}\mathrm{{E}}$ , $36^{\circ}40^{\prime}53^{\prime\prime}\mathrm{N})$ . It is approximately $15\,\mathrm{m}$ above ground and surrounded by three major roadways. But the traffic volume of these roadways is not extremely high compared with that of Environment Monitoring Station (EMS). The sampling site is experiencing a rapid economic development and abundant construction activities. The sampling site represents a newly formed urban area.
Sample collection
Atmospheric particle sampling was respectively performed in 2010. The 24-h $\mathrm{PM}_{2.5}$ samples were synchronously collected on $90{\-}\mathrm{mm}$ quartz fibers and $90{-}\mathrm{mm}$ polypropylene fiber filters at the two sampling sites using medium-volume samplers (model TH-150S, manufactured by Tianhong Instrument Co., Ltd. Wuhan, China) operating at a flow rate of $100\ \mathrm{L\min}^{-1}$ with a $2.5{-}\upmu\mathrm{m}$ cutpoint impactor. To remove volatile components that might be present on the filter before sampling, the quartz fiber filters were preheated in the muffle furnace at $600\,^{\circ}\mathrm{C}$ for $^{3\,\mathrm{h}}$ , while the polypropylene fiber filters were in the oven at $60\ ^{\circ}\mathrm{C}$ for $1\,\mathrm{h}$ . Before and after sampling, the filters were equilibrated in a dessicator (temperature $20-$ $23\ ^{\circ}\mathrm{C}$ and relative humidity $\left(R H\right)35{-}45\;\%)$ for $48\,\mathrm{h}$ and then weighed on an electronic microbalance with $\mathbf{a}\pm1-\upmu\mathbf{g}$ sensitivity (Mettler Toledo Inc., Switzerland). After that, the samples were stored in a refrigerator at $-18\mathrm{~}^{\circ}\mathrm{C}$ before chemical analysis.
Eliminating the invalid atmospheric particle samples due to the rain, sampler’s malfunction, filter’s fracture, or other unexpected accidents, the available $\operatorname{PM}_{2.5}$ quartz fiber and polypropylene fiber filter samples were synchronously collected, 18, 11, 17, and 14 in spring, summer, fall, and winter, respectively, at EMS and Shandong Jianzhu University (SJU) in 2010.
Chemical analysis
Organic carbon (OC) and elemental carbon (EC) were measured by the IMPROVE thermal optical reflectance method with Desert Research Institute Model 2001 Thermal Optical Carbon Analyzer (Chow et al. 2001, 2004, 2005).
Details of the analytical procedure as well as QA/QC were given by Cao et al. (2005).
Ion chromatography (ICS-1500, Dionex Ltd., USA) was used to analyze the three anions (Cl–, $\mathrm{NO}_{3}^{\mathrm{~-~}}$ , and $\mathrm{SO}_{4}^{\ 2-}$ ) and five cations $(\mathrm{Na}^{+},\mathrm{NH}_{4}^{+},\mathrm{K}^{+},\mathrm{Mg}^{2+})$ , and $C\mathrm{a}^{2+}$ ). Principal and details of the analytical procedure were given by Gu et al. (2011).
Inductively coupled plasma-atomic emission spectrometer (Baldwin et al. 1994) (IRIS Intrepid II, Thermo Electron) and inductively coupled plasma-mass spectroscopy (Chio et al. 2004; Kong et al. 2011) (Agilent 7500a, Agilent Co. USA) analyses were employed for the determination of Na, Mg, Al, Si, P, K, Ca, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Pb, Ba, Sr, and Sn in samples collected on polypropylene fiber filters.
Other data collection
The data of $\mathrm{SO}_{2}$ and $\mathrm{NO}_{2}$ were all obtained from Jinan Environment Monitoring Station. The corresponding meteorological data (temperature, wind speed, RH, etc.) were obtained from Jinan Meteorological Bureau.
Results and discussion
$\mathrm{PM}_{2.5}$ mass concentration
Statistics for particulate mass concentration at the two sampling sites were shown in Table 1 during the observation period. The annual arithmetic mean of $\operatorname{PM}_{2.5}$ was $168.85~{\upmu\mathrm{g}}~\mathrm{m}^{-3}$ at EMS and $147.33~\upmu\mathrm{g}\mathrm{~m}^{-3}$ at SJU, respectively. The lowest mass concentrations were 65.45 and $44.10\ensuremath{~\upmu\mathrm{g}\,\mathrm{m}^{-3}}$ , which occurred on April 23 and April 29, while the highest concentrations were $563.16~\mathrm{\textmug}$ and $559.13~\upmu\mathrm{g}\,\,\mathrm{m}^{-3}$ , which occurred on January 29 and January 19 at EMS and SJU, respectively. It was noted that the minimum mass concentration was all presented at low wind speed $(1.5\,\mathrm{\m/s})$ and after rain in spring, while the maximum concentration was also presented at low wind speed ( $\,\!\!\lfloor1.7\;\!\mathrm{m}/$ s) but before rain or snow in winter at both sampling sites. It indicated that the meteorological conditions were important to particulate convergence and diffusion in Jinan.
The diurnal mass concentrations of $\operatorname{PM}_{2.5}$ were compared to the Ambient Air Quality Standards (GB3095-2012) of class II $(75\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ promulgated by the Ministry of Environmental Protection of the People’s Republic of China for $\operatorname{PM}_{2.5}$ . It was found that there were $23\;\mathrm{PM}_{2.5}$ samples at EMS and $19\,\mathrm{PM}_{2.5}$ samples at SJU, whose concentration exceeded the standard of GB3095-2012; furthermore, the $\mathrm{PM}_{2.5}$ mass concentration level was the standard by 1.4–9.1 times during the sampling period in Jinan, indicating that the $\operatorname{PM}_{2.5}$ pollution was quite serious and could not be ignorable in the future.
From the seasonal variation of $\operatorname{PM}_{2.5}$ , the highest $\operatorname{PM}_{2.5}$ mass concentration was in winter at both sampling sites. When many heating systems were run by burning coal, particulate emission increased during the winter. Moreover, in winter, the weather in Jinan was usually influenced by high pressure systems, which caused formation of inversion layers that could inhibit pollutant dispersion. In the present study, the average wind speed was 3.75, 2.90, 2.92, and $2.99~\mathrm{m/s}$ , RH 46.6, 68.4, 56.4, and $44.6\,\%$ , and temperature 13.7, 26.4, 15.6, and $1.5\,^{\circ}\mathrm{C}$ in spring, summer, fall, and winter, respectively. In spring, the weather was windy and dry and favorable for dispersion of PM, at the same time the low relative humidity might not favor to secondary particle formation. In summer, the rainfall was very plentiful and the PM could be efficiently removed by wet scavenging. But dust events and high RH frequently occurred in spring and summer, respectively, so the $\mathrm{PM}_{2.5}$ mass concentration did not have obvious seasonal fluctuation in spring and summer. It should be noted that the $\mathrm{PM}_{2.5}$ mass concentration was higher at SJU than that at EMS only in fall. Agrarian cultivation in SJU suburbs was still extensive. Both the burning crop remnants and fertilizing the soil were contributed to the higher $\operatorname{PM}_{2.5}$ mass concentration. Except in fall, the arithmetic means of $\operatorname{PM}_{2.5}$ in other seasons were all higher at EMS than that at SJU, which might have the high traffic flows and the emissions from the nearby industrial and restaurants cooking at EMS. Figure 2 showed the spatial variation of $\operatorname{PM}_{2.5}$ mass concentration during the observation period. When the $\mathrm{PM}_{2.5}$ mass concentration was lower than $100~\upmu\mathrm{g}~\mathrm{m}^{-3}$ , there was no obvious difference in $\operatorname{PM}_{2.5}$ mass concentration at two sampling sites. However, the spatial difference was more obvious with the increase of the diurnal $\operatorname{PM}_{2.5}$ mass concentration, especially when the mass concentration was larger than $200\ \upmu\mathrm{g}\ \mathrm{m}^{-3}$
Coefficient of divergence (CD) (Wongphatarakul et al. 1998; Zhang and Friedlander 2000; Park and $\mathrm{Kim}~2004\rangle$ was used to test the extent of spatial difference. The CD was defined as follows:
$$
\mathrm{CD}_{j k}={\sqrt{{\frac{1}{p}}\sum_{i=1}^{p}\left({\frac{x_{i j}-x_{i k}}{x_{i j}+x_{i k}}}\right)^{2}}}
$$
where $j$ and $k$ represented two sampling sites, and $p$ was the number of chemical components. $x_{i j}$ was the average mass concentration for a chemical component $i$ at site $j$ . The CD approached zero, indicating the little difference between them, and approached one, indicating the big difference.
In order to analyze the spatial difference of $\mathrm{PM}_{2.5}$ and chemical species at the two sampling sites, the CDs of $\mathrm{PM}_{2.5}$ and chemical species were calculated and the values were 0.14 in spring, 0.19 in summer, 0.23 in fall, and 0.23 in winter, respectively. These CDs were little and adjacent zero, indicating that there were no obvious difference for the $\mathrm{PM}_{2.5}$ and chemical species at the two sampling sites. So, the regional pollution character of the fine particles was present in Jinan, China.
Chemical compositions
Carbonaceous components
The overall average mass concentration of the OC and EC components in $\operatorname{PM}_{2.5}$ was 17.69 and $5.53~\upmu\mathrm{g}\,\,\mathrm{m}^{-3}$ at EMS, while it was 14.29 and $5.12\ \upmu\mathrm{g}\ \mathrm{m}^{-3}$ at SJU, respectively. The carbonaceous pollution exhibited obvious seasonal characteristics. Table 1 presented the order of OC seasonal variation: winter $>$ autumn $>$ spring $>$ summer at both sampling sites. This seasonal variation could be attributed to the cooperative effects of changes in emission rate and meteorological conditions. In winter, the highest concentration of OC could be attributed to the enhanced emission from coal combustion heating and unfavorable atmospheric dispersion (low mixing layer height, frequent inversion, etc.). The smallest concentration of OC was in summer for the decreasing consumption of coal, and more rainy days aided the dispersion and mitigated the carbonaceous pollution. The seasonal variation of EC at EMS site was autumn $>$ winter $>$ spring $>$ summer, while at the SJU site, the variation was autumn $>$ winter $>$ summer $>$ spring, respectively. These variations were distinct from what was observed for OC. The differences of seasonal variation between OC and EC suggested that emission sources for OC and EC were probably different.
At EMS, the proportion of OC for $\mathrm{PM}_{2.5}$ was 9.75, 8.77, 13.56, and $12.45\;\%$ in spring, summer, fall, and winter, while it was 10.17, 7.21, 11.28, and $12.10~\%$ at SJU, respectively.
The proportions of EC for $\operatorname{PM}_{2.5}$ were all less than $5.00\,\%$ in all the seasons at both sampling sites. It could indicate that coal combustion heating and biomass fuels led to more carbonaceous species enriched in fine particles. Because the relative abundances of OC and EC, respectively, determined the relative amounts of scattering and absorption, the OC abundance was higher than EC abundance in all the seasons at two sampling sites in this study, indicating that the light scattering of carbonaceous aerosol should be the more important factor causing visibility degradation in Jinan.
The ratio of OC to EC (OC/EC) can be used to interpret the emission and transformation characteristics of carbonaceous aerosol. As shown in Table 1, the diurnal ratio of OC/EC was the highest in winter at both sampling sites. This seasonal pattern of higher wintertime OC/EC ratio was also observed in Beijing (Dan et al. 2004), Guangzhou and Hong Kong (Duan et al. 2007), and Tianjin (Gu et al. 2011). The EC tracer method was used to calculate the seasonal mass concentration of secondary organic carbon (SOC) (Turpin and Huntzicker 1995; Castro et al. 1999; Cao et al. 2004, 2005, 2007). The seasonal SOC mass concentration and the ratio of SOC/OC were listed in Table 2 using the EC tracer method in this study. Similar to OC, SOC was also highest in winter and lowest in summer. This might be attributed to several reasons. First, coal consumption for winter heating contributed more to OC than EC and also increased the emission of volatile organic precursors. Second, low temperature led to the adsorption and condensation of semi-volatile organic compounds onto existing solid particles. Third, the low mixing layer height in winter would enhance the SOC formation.
Inorganic water-soluble ions
Figure 3 presents the mass concentration of major watersoluble ions $\left(\mathbf{Na}^{+}$ , $\mathrm{NH_{4}}^{+}$ , $\mathbf{K}^{+}$ , $\mathrm{Mg}^{2+}$ , $C\mathrm{a}^{2+}$ , $\mathrm{Cl}^{-}$ , $\mathrm{NO}_{3}^{\phantom{\mathrm{\,-}}}$ , $\mathrm{SO}_{4}^{\ 2-}$ ) in $\operatorname{PM}_{2.5}$ . The average concentration of $\mathrm{SO}_{4}^{\ 2-}$ was higher in summer, fall, and winter than that in spring at both sampling sites. Although the concentration of $\mathrm{SO}_{2}$ was relatively lower than that in winter, the sulfur oxidation ratio SOR $([\mathrm{SO}_{4}^{\;\;2-}]/$ $(\mathrm{[SO_{2}]}+\mathrm{[SO_{4}}^{2-}\mathrm{])})$ ) was higher than 0.1 in primary pollutants (Wang et al. 2005) for the strong photochemical reaction. Higher SOR suggested the oxidation of gaseous species and more secondary aerosols in the atmosphere. In this study, the SOR was 0.22, 0.55, 0.34, and 0.16 in spring, summer, fall, and winter at EMS, while it was 0.25, 0.55, 0.37, and 0.14 at SJU, respectively, indicating that SOR was highest in summer and lowest in winter. In summer, the gas particle conversion and further condensation or adsorption on the particle surface were likely the major sources of the secondary sulfate particles for the high temperature, and long photoperiod created the favorable photoreaction conditions for photochemical oxidation. At the same time, high relative humidity in summer could increase the heterogeneous oxidation reactions involving $\mathrm{SO}_{2}$ in clouds and fog (Yao et al. 2002; Zhou et al. 2012). Although the meteorological conditions (such as stable atmospheric boundary layer and the low humid elimination efficiency) completely led to the lower rate of the second conversion of $\mathrm{SO}_{2}$ to $\mathrm{SO}_{4}^{\ 2-}$ , the union effects of the sufficient $\mathrm{SO}_{2}$ and liquid-phase reaction and substantial anthropogenic sources jointly led to the relatively high concentration of $\mathrm{SO}_{4}^{\ 2-}$ in winter.
The highest seasonal average of $\mathrm{NO}_{3}^{\mathrm{~-~}}$ was 38.35 and $33.00\,\ \upmu\mathrm{g}\:\mathrm{\m}^{-3}$ in winter at EMS and SJU, respectively. Compared to $\mathrm{SO}_{4}^{\ 2-}$ , the $\mathrm{NO}_{3}^{\mathrm{~-~}}$ had not evidently fluctuated. The nitrate had weak thermal stability and was more sensitive to temperature. Higher temperature did not favor the formation of nitrate. The nitrogen oxidation ratio $\mathrm{(NOR=}[\mathrm{NO}_{3}^{\,-}]/$ $\left(\left[\mathrm{NO}_{2}\right]+\left[\mathrm{NO}_{3}^{\;-}\right]\right)$ could be used to assess the transformation degree of gaseous $\mathrm{NO}_{2}$ to aqueous nitrate. In this study, the NOR was 0.27, 0.20, 0.17, and 0.27 in spring, summer, fall, and winter at EMS, while it was 0.28, 0.23, 0.18, and 0.25 at SJU, respectively, indicating that the formation degree of $\mathrm{NO}_{3}^{\mathrm{~-~}}$ from $\mathrm{NO}_{2}$ was very weak. And that the annual average ratios of $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{\,\,2-}$ were 0.89 and 0.45 at EMS and SJU, respectively. The lower value of $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{\;2-}$ at SJU could also reflect the dominated coal combustion sources for particles. The higher value of $\mathrm{NO}_{3}^{\mathrm{~-}}/\mathrm{SO}_{4}^{\mathrm{~2-}}$ at EMS depicted that more fractions of particles resulted from motor vehicle exhaust.
Compared to $\mathrm{SO}_{4}{}^{2-}$ and $\mathrm{NO}{_3}^{-}$ , the concentration of $\mathrm{Cl}^{-}$ was relatively low in all the seasons. The ratio of $\mathrm{{Cl}^{-}/{N a}^{+}}$ was about 1.2 in seawater (Tsitouridou and Samara 1993). But in this study, the highest ratio of $\mathrm{Cl}^{-}/\mathrm{Na}^{+}$ was in winter and the value was 10.07 and 9.59, much larger than 1.2, at EMS and SJU, respectively, indicating that most of the chloride might not be present as sea salt particles. Possibly, it originated from industrial contamination sources or, else, the chloride which originated from large sea salt particles, but was displaced by reaction with nitric acid and followed by reaction with ammonia to form small particles (Wang and $\mathrm{Hu}\ 2000\$ ).
Agricultural activities had been considered to be the major emission source for $\mathrm{NH}_{3}$ in atmosphere (Sutton et al. 1995). Ammonium ion was usually transformed from ammonia in atmosphere. In the present study, the highest concentration of $\mathrm{NH_{4}}^{+}$ was observed in summer, likely because of the increasing of the high temperature, increased animal activities, and more frequent agricultural activities. The mass concentration of $\mathrm{{Ca}}^{2+}$ was relatively higher in spring and winter and was related to dry weather, low precipitation, construction activities, and strong winds, which resulted in both local and long-distance transport dust events. The seasonal variation of other three cations $(\mathrm{Na}^{+},\mathrm{K}^{+}$ , and $\mathrm{Mg}^{2+}$ ) was very little.
Inorganic elements
Average mass concentrations of 20 inorganic elements at EMS and SJU were listed in Table 2. At both sampling sites, the sum of element mass concentration was higher in winter and spring than that in fall and summer. Moreover, of these elements, Al, Si, K, Ca, and Fe exhibited especially high concentration, indicating that the mineral dust pollution for fine particles was very serious. Table 2 also presented the obvious seasonal variation for elements. The highest mass concentration appearing in spring at SJU was likely ascribed to the soil, non-mineral dust, or dust storm contribution. But at EMS, the highest concentrations of Si, Al, Ca, and Fe appeared in winter, indicating that there were no significant contributions of soil and non-mineral dust; probably, this was due to in part to fly ash. Additional influences from cement dust and other construction-related activities would also appear to be possible contributors owing to the rapid urbanization. For the other trace elements, their concentrations were similar and there was no obvious variation in spring, summer, and fall, while the anthropogenic sources contributed to the maximum concentration in winter.
In order to evaluate the contributions of crustal and noncrustal sources, one way to estimate the possible source of the trace elements associated with the $\mathrm{PM}_{2.5}$ could be carried out by calculating the enrichment factors (EF). For each element, the EF was calculated according to the equation given below:
$$
\mathrm{EF}_{\mathrm{crust},X}=(X/Y)_{\mathrm{sample}}/\Big(X\Big/Y\Big)_{\mathrm{crust}}
$$
Where $X$ was the trace element of interest, $Y$ was the reference element, while $(X/Y)_{\mathrm{sample}}$ and $(X/Y)_{\mathrm{crust}}$ were the ratio between the concentration of the trace element of interest in the air and geological material, respectively. In this study, Al was used as the reference element for abundance in the upper continental crust. The EF was approximate to the unit, indicating that the soil and geological materials were the dominant sources for $\mathrm{PM}_{2.5}$ . The EF was respectively 1–2, 2–5, 5–20, and 20–40, not only indicating that both natural emissions and anthropogenic sources were important for these elements, but also indicating that the enrichment degrees were slight, medium, notable, and intensity, respectively. The EF was larger than 40, indicating that the enrichment degree was pole strength. In this study, Al was selected as the reference element because it was relatively stable and was not affected by most anthropogenic contaminants. The denary logarithm of EFs was shown in Fig. 4. At both sampling sites, the EF values of Si and Ti were approximate to the unit, indicating that they were mainly contributed by crustal source. The EF values of Na, Mg, P, K, Ca, V, Cr, Mn, Fe, Co, Ni, Ba, and Sr were mainly between 1 and 40, suggesting that both natural emissions and anthropogenic sources were important for these elements. The EF values of Cu, Zn, Pb, and Sn were larger than 40, indicating that they mainly originated from noncrustal sources such as vehicular exhaust and industrial emission.
The method of mass closure for particles had become the accepted method for the analysis of aerosols (Chan et al. 1997; Christoforou et al. 2000). Normally, $\operatorname{PM}_{2.5}$ could be classified into six major types: organic matter, major water soluble ions, element carbon, crustal matter, sea salt, and trace species. Since the chemical composition of organic fraction for aerosol was largely unknown. Organic matter was obtained by multiplying OC by a factor of 1.4, as was commonly used to estimate the unmeasured hydrogen and oxygen in organic compounds. Soluble Na in the aerosol samples was assumed to come solely from sea salt; thus, the mass of sea salt was estimated by $2.54\!\times\!\mathrm{Na}^{+}$ . The mass of crustal matter was estimated on the basis of the oxides of Al, Ca, Si, Ti, Fe, Na, $\mathrm{Mg}$ , and K as follows:
$$
{\begin{array}{r}{{\mathrm{Crustal~matter}}=1.89\,{\mathrm{~Al}}+1.40\,{\mathrm{~Ca}}+2.14\,{\mathrm{~Si}}+1.67\,{\mathrm{~Ti}}}\\ {+}&{1.36\,{\mathrm{~Fe}}+1.35\,{\mathrm{Na}}+1.67\,{\mathrm{Mg}}+1.2\,{\mathrm{K}}}\end{array}}
$$
The sum of all other elements was defined as trace species, including P, V, Cr, Mn, Co, Ni, Ba, Sr, Cu, Zn, Pb, and Sn in this study.
The average contribution percentages of the major types for $\mathrm{PM}_{2.5}$ were given in Fig. 5. The sum of contribution percentage for the nine chemical components was approximately 86.77 and $89.25\;\%$ of the $\operatorname{PM}_{2.5}$ mass concentration at EMS and SJU, respectively. At the both sampling sites, the $\mathrm{SO}_{4}^{\ 2-}.$ , $\mathrm{NO}_{3}^{\mathrm{~-~}}$ , organic carbon, and crustal matter were the main contribution, and the sums of their percentage abundance were 71.44 and $73.99\,\%$ at EMS and SJU, respectively. This indicated that it was very pivotal to control the four chemical components in order to reduce the pollution of $\operatorname{PM}_{2.5}$ in Jinan. The higher percentage abundance of organic carbon should be at EMS than that at SJU. The percentage abundance of crustal matter was similar to that in Beijing (Yang et al. 2004) and higher than that in Tianjin (Gu et al. 2011). Maybe it indicated that there were many construction activities and special geographical situation to the disadvantage of diffusion. The unidentified average portion was 13.23 and $10.25\,\%$ of the $\mathrm{PM}_{2.5}$ at EMS and SJU, respectively. Their components probably included unmeasured crustal matter, trace elements, and moisture absorbed by particles or might have been due to systematic weighing errors.
Source apportionment by PMF
PMF (v3.0, USEPA, 2008) was described in detail in Pattero and Tapper (1994) and Pattero (2004) and used to identify the contribution of various emission sources. As suggested by several recent source apportionment studies (Lee et al. 1999; Liu et al. 2005; Song et al. 2006; Raman and Hopke 2007), the mass fraction distribution of species was used to identify the sources, which included soil dust, vehicle emissions, sea salt, industrial emissions, and secondary aerosols.
By the PMF analysis, the source profiles were displayed in Fig. 6. The source enriched in $\mathrm{NH_{4}}^{+}$ and $\mathrm{SO}_{4}^{\ 2-}$ was classified as secondary sulfate in the previous studies (Kim et al. 2007; Raman and Hopke 2007). The secondary sulfates formed by photochemical and other chemical processes were known to be major constituents of fine particulate matter measured in most industrialized areas. In this study, $\mathrm{NH_{4}}^{+}$ and $\mathrm{SO}_{4}^{\ 2-}$ were composed of higher mass fractions and indicated secondary sulfate source. Soil dust, including airborne soil and road dust, included most of the crustal elements and had high concentrations of Al, Ca, Fe, Si, Na, $\mathrm{Mg}$ , and K. Coal combustion and biomass burning were presented by high OC and EC (Zheng et al. 2005; Watson 2002). Coal is still a major energy source in Jinan. Moreover, agrarian cultivation in Jinan suburbs is still extensive. Secondary nitrate is characterized by high nitrate. The precursor gas of nitrate, $\mathrm{NO}_{x}$ , is emitted by traffic and stationary source such as electricity-generating plants. Motor vehicle emissions are characterized by high $\mathrm{Pb}$ , Sn, and EC. The source concentrations to $\operatorname{PM}_{2.5}$ were as follows: coal combustion and biomass burning $(38.00\ \%)$ , secondary sulfate $(34.98\ \%)$ ), secondary nitrate and vehicle emissions $(16.85\,\%)$ , and soil dust $(10.17\,\%)$ at EMS and coal combustion and biomass burning $(46.23\ \%)$ , soil dust ( $[16.91\;\%)$ , and secondary sulfate $(36.86\;\%)$ at SJU.
Conclusions
A 1-year-long observation of $\operatorname{PM}_{2.5}$ in Jinan was carried out in 2010. The $\mathrm{PM}_{2.5}$ mass concentration and major chemical components were synchronously investigated at two sampling sites. The annual mean mass concentration of $\mathrm{PM}_{2.5}$ was $168.85~{\upmu\mathrm{g}}~\mathrm{m}^{-3}$ at EMS and $148.67~\upmu\mathrm{g}\mathrm{~m}^{-3}$ at SJU, respectively. The $\mathrm{PM}_{2.5}$ mass concentration with obvious seasonal variation character resulted from varying emission sources and meteorological conditions. The coefficients of dispersion were calculated in each season and the values were all adjacent zero, showing that there was no obvious spatial difference and character of regional air pollution in Jinan. The highest carbon concentration and OC/EC ratios were in winter and resulted from enhanced emissions from coal combustion for heating and poor atmospheric dispersion. On the contrary, the abundant rainfall removed carbonaceous aerosol by wet scavenging and led to the lowest carbon concentration and OC/EC ratio in summer. $\mathrm{SO}_{4}{}^{2-}$ , $\mathrm{NO}_{3}^{\phantom{\,+}}$ , and $\mathrm{NH_{4}}^{+}$ were the main chemical components of $\mathrm{PM}_{2.5}$ in Jinan. $\mathrm{SO}_{4}{}^{2-}$ and $\mathrm{NH_{4}}^{+}$ concentrations were highest in summer, while $\mathrm{NO}_{3}^{\mathrm{~-~}}$ was highest in autumn. The seasonal concentration variation for different ions resulted from the source variation, secondary aerosol ion formation mechanisms, and meteorological conditions. By the method of chemistry mass closure, approximately 86.77 and $89.25~\%$ of the $\operatorname{PM}_{2.5}$ mass concentrations were explained by nine chemical components and sulfate was the most predominant fraction accounting for about 29.66 and $31.64\;\%$ of $\operatorname{PM}_{2.5}$ at EMS and SJU site, respectively. By the PMF analysis, the coal combustion and biomass burning, secondary aerosol, and soil dust were identified as the major emission sources in Jinan.
Acknowledgments This study was projected by Tianjin Science and Technology Development Commission (11JCYBJC05200), Ministry of Environmental Protection of China (2010467007), and Jinan Science and Technology Development Commission (201101090).
References
Baldwin AT, Schmelz EA, Ohnmeiss TE (1994) Wound-induced changes in root and shoot jasmonic acid pools correlate with induced nicotine synthesis in Nicotiana sylvestris Spegazzini and Comes. J Chem Ecol 20(8):2139–2157
Cao JJ, Lee SC, Ho KF, Zou SC, Fung K, Li Y, Watson JG, Chow JC (2004) Spatial and seasonal variations of atmospheric organic carbon and elemental carbon in Pearl River Delta Region, China. Atmos Environ 38(27):4447–4456
Cao JJ, Wu F, Chow JC, Lee SC, Li Y, Chen SW, An ZS, Fung KK, Watson JG, Zhu CS, Liu SX (2005) Characterization and source apportionment of atmospheric organic and elemental carbon during fall and winter of 2003 in Xi’an, China. Atmos Chem Phys 5:3127– 3137
Cao JJ, Lee SC, Chow JC, Watson JG, Ho KF, Zhang RJ, Jin ZD, Shen ZX, Chen GC, Kang YM, Zou LZ, Zhang LZ, Qi Sh, Dai MH, Cheng Y, Hu K (2007) Spatial and seasonal distributions of carbonaceous aerosols over China. J Geophys Res 112 (D22), doi:10. 1029/2006JD008205.
Castro LM, Pio CA, Harrison RM, Smith DJT (1999) Carbonaceous aerosol in urban and rural European atmospheres: estimation of secondary organic carbon concentrations. Atmos Environ 33(17): 2771–2781
Chan YC, Simpson RW, Mctainsh GH, Vowle PD (1997) Characterization of chemical species in $\operatorname{PM}_{2.5}$ and $\mathrm{PM_{10}}$ aerosols in Brisbane, Australia. Atmos Environ 31(22):2061–2080
Cheng YF, Berghof M, Garland RM, Wiedensohler A, Wehner B, Müller T, Su H, Zhang YH, Achter P, Nowak A, Pöschl U, Zhu T, Hu M, Meng LM (2009) Influence of soot mixing state on aerosol light absorption and single scattering albedo during air mass 15 aging at a polluted regional site in northeastern China. J Geophys Res 114(D2):D00G10, doi:10.1029/2008JD010883
Chio C, Cheng M, Wang C (2004) Source apportionment to PM in different air quality conditions for Taichung urban and coastal areas, Taiwan. Atmos Environ 38(39):6893–6905
Chow JC, Watson JG, Lu ZQ, Lowenthal DH, Frazier CA, Solomon PA, Thuillier RH, Magliano K (1996) Descriptive analysis of $\operatorname{PM}_{2.5}$ and $\mathrm{PM}_{10}$ at regionally representative locations during SJVAQS/ AUSPEX. Atmos Environ 30(12):2079–2112
Chow JC, Watson JG, Crow D, Lowenthal DH, Merrifield T (2001) Comparison of IMPROVE and NIOSH carbon measurements. Aerosol Sci Technol 34(1):23–34
Chow JC, Watson JG, Chen LWA, Arnott WP, Moosmüller H (2004) Equivalence of elemental carbon by thermal/optical reflectance and transmittance with different temperature protocols. Environ Sci Technol 38(16):4414–4422
Chow JC, Watson JG, Louie PK, Chen LW, Sin D (2005) Comparison of $\mathrm{PM}_{2.5}$ carbon measurement methods in Hong Kong, China. Environ Pollut 137(2):334–344
Christoforou CS, Salmon LG, Hannigan MP, Solomon PA, Cass GR (2000) Trends in fine particle concentration and chemical composition in southern California. J Air Waste Manag Assoc 50(1):43–53
Dan M, Zhuang GS, Li XX (2004) The characteristics of carbonaceous species and their sources in $\mathrm{PM}_{2.5}$ in Beijing. Atmos Environ 38(21):3443–3452
Dockery DW, Stone PH (2007) Cardiovascular risks from fine particulate air pollution. N Engl J Med 356:511–513 Duan JC, Tan JH, Cheng DX, Bi XH, Deng WJ, Sheng GY, Fu JM, Wong MH (2007) Sources and characteristics of carbonaceous aerosol in two largest cities in Pearl River Delta Region, China. Atmos Environ 41(14):2895–2903 Duan J, Tan J (2013) Atmospheric heavy metals and arsenic in China: situation, sources and control policies. Atmos Environ 74:93–101 Gao J, Wang J, Cheng SH, Xue LK, Hz Y, Hou LJ, Jiang YQ, Wang WX (2007) Number concentration and size distribution of submicron particles in Jinan urban area: characteristics in summer and winter. J Environ Sci 19(12):1466–1473 Gu JX, Bai ZP, Li AX, Wu LP, Xie YY, Lei WF, Dong HY, Zhag X (2011) Chemical composition of $\operatorname{PM}_{2.5}$ during winter in Tianjin, China. Particuology 9(3):215–221 Guo S, Hu M, Wang ZB, Wang ZB, Slanina J, Zhao YL (2010) Sizeresolved aerosol water-soluble ionic compositions in the summer of Beijing: implication of regional f secondary ormation. Atmos Chem Phys 10:947–959 Harrison RM (2004) Key pollutants—airborne particles. Sci Total Environ 334–335:3–8 He KB, Yang FM, Ma YL, Ma YL, Zhang Q, Yao XH, Chan CK, Cadle S, Chan T, Mulawa P (2001) The characteristics of $\mathrm{PM}_{2.5}$ in Beijing, China. Atmos Environ 35(29):4959–4970 Huang XF, He LY, Hu M, Zhang YH (2006) Annual variation of particulate organic compounds in $\mathrm{PM}_{2.5}$ in the urban atmosphere of Beijing. Atmos Environ 40(14):2449–2458 Ianniello A, Spataro F, Esposito G, Allegrini I, Hu M, Zhu T (2011) Chemical characteristics of inorganic ammonium salts in $\operatorname{PM}_{2.5}$ in the atmosphere of Beijing (China). Atmos Chem Phys 11:10803–
10822 Kim HS, Huh JB, Hopke PK, Holsen TM, Yi SM (2007) Characteristics of the major chemical constituents of $\operatorname{PM}_{2.5}$ and smog events in Seoul, Korea in 2003 and 2004. Atmos Environ 41(32):6762–6770 Kong SF, Lu B, Bai ZP, Zhao XY, Han B, Li ZY, Ji YQ, XuYH LY, Jiang H (2011) Potential threat of heavy metals in re-suspended dusts on building surfaces in oilfield city. Atmos Environ 45(25):4192–4204 Lee E, Chan CK, Paatero P (1999) Application of Positive matrix factorization in source apportionment of particle pollutants in HongKong. Atmos Environ 33:3201–3212 Li XR, Wang LL, Wang YS, Wen TX, Yang YJ, Zhao YN, Wang YF (2012) Chemical composition and size distribution of airborne particulate matters in Beijing during the 2008 Olympics. Atmos Environ 50:278–286 Liu W, Wang Y, Russell A, Edgerton LES (2005) Atmospheric aerosol over two urban–rural pairs in the southeastern United States: chemical composition and possible sources. Atmos Environ 39:4453–4470 Menon S, Hansen J, Nazarenko L, Luo YF (2002) Climate effects of black carbon aerosols in China and India. Science 297(5590):2250–
2253 Norris G, YoungPong SN, Koenig JQ, Lrson TV, Sheppard L, Stout JW (1999) An association between fine particles and asthma emergency department visits for children in Seattle. Environ Health Perspect
107(6):489–493 Ostro B, Broadwin R, Green S, Feng WY, Lipsett M (2006) Fine particulate air pollution and mortality in nine California counties: results from CALFINE. Environ Health Perspect 114(1):29–33 Park SS, Kim YJ (2004) $\operatorname{PM}_{2.5}$ particles and size-segregated ionic species measured during fall season in three urban sites in Korea. Atmos Environ 38(10):1459–1471 Paatero P, Tapper U (1994) Positive matrix factorization: a non-negative factor model with optimal utilization of error estimates of data values. Environmetrics 5:111–126 Paatero, P (2004) User’s guide for positive matrix factorization programs PMF2 and PMF3, part 1: tutorial. Raman RS, Hopke PK (2007) Source apportionment of fine particles utilizing partially speciated carbonaceous aerosol data at two rural locations in New York state. Atmos Environ 41:7923–7939
Reiss R, Anderson EL, Cross CE, Hidy DH, Clellan RM, Moolgavkar S (2007) Evidence of health impacts of sulfate- and nitrate-containing particles in ambient air. Inhal Toxicol 19(5):419–449
Schwartz J, Dockery DW, Neas LM (1996) Is daily mortality associated specifically with fine particles? J Air Waste Manag Assoc 46(10): 927–939
Song Y, Zhang YH, Xie SD, Zeng LM, Zheng M, Salmon LG, Shao M, Slanina S (2006) Source apportionment of $\operatorname{PM}_{2.5}$ in Beijing by positive matrix factorization. Atmos Environ 40(8):1526–1537
Sun YL, Zhuang GS, Wang Y, Han L, Guo J, Dan M, Zhang WJ, Wang ZF, Hao ZP (2004) The air-borne particulate pollution in Beijing— concentration, composition, distribution and sources. Atmos Environ 38(35):5991–6004
Sutton MA, Place CJ, Eager M, Fowler D, Smith RI (1995) Assessment of the magnitude of ammonia emissions in the United Kingdom. Atmos Environ 29(12):1393–1411
Tao J, Cao JJ, Zhang RJ, Zhu LH, Zhang T, Shi S, Chan CY (2012) Reconstructed light extinction coefficients using chemical compositions of $\mathrm{PM}_{2.5}$ in winter in urban Guangzhou, China. Adv Atmos Sci 29(2):359–368
Turpin BJ, Huntzicker JJ (1995) Identification of secondary aerosol episodes and quantification of primary and secondary organic aerosol concentrations during SCAQS. Atmos Environ 29(23):3527– 3544
Tsitouridou R, Samara C (1993) First results of acidic and alkaline constituents determination in air particulates of Thessaloniki, Greece. Atmos Environ, Part B Urban Atmos 27(3):313–319
Vedal S (1997) Critical review: ambient particles and health: lines that divide. J Air Waste Manag Assoc 47(5):551–581
Walker JT, Whitall DR, Robarge W, Paerl HW (2004) Ambient ammonia and ammonium aerosol across a region of variable ammonia emission density. Atmos Environ 38(9):1235–1246
Wang M, Hu M (2000) Sea salt source contribution to the aerosol in Qingdao seashore area. J Environ Sci-China 21(5):83–85
Wang Q, Shao M, Zhang Y, Wei Y, Hu M, Guo S (2009) Source apportionment of fine organic aerosols in Beijing. Atmos Chem Phys 9:8573–8585
Wang Y, Zhuang GS, Tang AH, Yuan H, Sun YL, Chen S, Zheng AH (2005) The ion chemistry and the source of $\operatorname{PM}_{2.5}$ aerosol in Beijing. Atmos Environ 39(21):3771–3784
Watson JG (2002) Critical review discussion. Visibility: science and regulation. J Air Waste Manag Assoc 52:628–713
Wongphatarakul V, Friedlander SK, Pinto JP (1998) A comparative study of $\mathrm{PM}_{2.5}$ ambient aerosol chemical databases. Environ Sci Technol 32(24):3926–3934
Yang FM, He KB, Ma YL, Zhnag Q (2004) Characterization of mass balance of $\mathrm{PM}_{2.5}$ chemical speciation in Beijing. Environ ChemChina 23(3):326–333
Yao XH, Chan CK, Fang M, Cadle S, Chan T, Mulawa P, He K, Ye B (2002) The water-soluble ionic composition of $\operatorname{PM}_{2.5}$ in Shanghai and Beijing, China. Atmos Environ 36(26):4223–4234
Yao XH, Lau APS, Fang M, Chan CK, Hu M (2003) Size distributions and formation of ionic species in atmospheric particulate pollutants in Beijing, China: 1-inorganic ions. Atmos Environ 37(21):2991–3000
Zhang Z, Friedlander SK (2000) A comparative study of chemical databases for fine particle Chinese aerosols. Environ Sci Technol 34(22): 4687–4694
Zheng M, Salmon LG, Schauer JJ, Zeng L, Kiang CS, Zhang Y, Cass GR (2005) Seasonal trends in $\operatorname{PM}_{2.5}$ source contributions in Beijing, China. Atmos Environ 39:3967–3976
Zhou JM, Zhang RJ, Cao JJ, Chow JC, Watson JG (2012) Carbonaceous and ionic components of atmospheric fine particles in Beijing and their impact on atmospheric visibility. Aerosol Air Qual Res 12: 492–502
|
atmosphere
|
0024
|
10.1016/j.atmosenv.2006.02.032
|
table
|
Table 1 Sampling sites in the Pearl River Delta
|
|
atmosphere
|
0024
|
10.1016/j.atmosenv.2006.02.032
|
image
|
Fig. 1. Location of sites used for the PRD study.
|
|
atmosphere
|
0024
|
10.1016/j.atmosenv.2006.02.032
|
table
|
Table 2 Mean fine particulate matter concentrations across the PRD
|
|
atmosphere
|
0024
|
10.1016/j.atmosenv.2006.02.032
|
table
|
Table 3 Sampling days categorized into southerly, northerly or mixed flow
|
|
atmosphere
|
0024
|
10.1016/j.atmosenv.2006.02.032
|
image
|
Fig. 2. Measurements of wind speed, wind direction and precipitation at the Shenzhen meteorology site for selected days categorized into northerly, southerly and mixed flow. Hourly wind magnitude and direction are indicated by the arrows and $24\,\mathrm{{h}}$ rainfall is indicated by the square markers.
|
|
atmosphere
|
0024
|
10.1016/j.atmosenv.2006.02.032
|
image
|
Fig. 3. Normalized concentrations and standard error of species measured at seven sites in the Pearl River Delta, categorized by wind pattern. Site labels are as follows: Tap Mun (TM), Tung Chung (TC), central/western (CW), Shenzhen (SZ), Zhongshan (ZS). Guangzhou (GZ) and Conghua (CH).
|
|
atmosphere
|
0024
|
10.1016/j.atmosenv.2006.02.032
|
table
|
Table 4 Average concentrations of measured species during southerly, northerly and mixed flow
|
|
atmosphere
|
0024
|
10.1016/j.atmosenv.2006.02.032
|
text
| Not supported with pagination yet
|
Source areas and chemical composition of fine particulate matter in the Pearl River Delta region of China
G.S.W. Haglera, , M.H. Bergina,b, L.G. Salmonc, J.Z. Yud, E.C.H. Wand, M. Zhengb, L.M. Zenge, C.S. Kiange, Y.H. Zhange, A.K.H. Lauf, J.J. Schauerg
aSchool of Civil and Environmental Engineering, Georgia Institute of Technology, 311 Ferst Drive, Atlanta, GA 30332, USA bSchool of Earth and Atmospheric Science, Georgia Institute of Technology, 311 Ferst Drive, Atlanta, GA 30332, USA cEQL, California Institute of Technology, Pasedena, CA 91125, USA $\mathrm{_d}$ Department of Chemistry, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, People’s Republic of China eState Key Joint Laboratory of Environmental Simulation and Pollution Control, College of Environmental Sciences, Peking University, Beijing, People’s Republic of China
fCenter for Coastal and Atmospheric Research, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, People’s Republic of China
gCivil and Environmental Engineering, University of Wisconsin-Madison, 660 North Park Street, Madison, WI 53706, USA
Received 18 August 2005; accepted 22 February 2006
Abstract
Fine particulate matter $(\mathbf{PM}_{2.5})$ was measured for 4 months during 2002–2003 at seven sites located in the rapidly developing Pearl River Delta region of China, an area encompassing the major cities of Hong Kong, Shenzhen and Guangzhou. The 4-month average fine particulate matter concentration ranged from 37 to $71\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in Guangdong province and from 29 to $34\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in Hong Kong. Main constituents of fine particulate mass were organic compounds $(24-35\%$ by mass) and sulfate $(21\mathrm{-}32\%)$ ). With sampling sites strategically located to monitor the regional air shed patterns and urban areas, specific source-related fine particulate species (sulfate, organic mass, elemental carbon, potassium and lead) and daily surface winds were analyzed to estimate influential source locations. The impact of transport was investigated by categorizing 13 (of 20 total) sampling days by prevailing wind direction (southerly, northerly or low windspeed mixed flow). The vicinity of Guangzhou is determined to be a major source area influencing regional concentrations of $\operatorname{PM}_{2.5}$ , with levels observed to increase by $18{-}34\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ (accounting for $46–56\%$ of resulting particulate levels) at sites immediately downwind of Guangzhou. The area near Guangzhou is also observed to heavily impact downwind concentrations of lead. Potassium levels, related to biomass burning, appear to be controlled by sources in the northern part of the Pearl River Delta, near rural Conghua and urban Guangzhou. Guangzhou appears to contribute $5{-}6\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ of sulfate to downwind locations. Guangzhou also stands out as a significant regional source of organic mass (OM), adding $8.5–14.5\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ to downwind concentrations. Elemental carbon is observed to be strongly influenced by local sources, with highest levels found in urban regions. In addition, it appears that sources outside of the Pearl River Delta contribute a significant fraction of overall fine particulate matter in Hong Kong and Guangdong province. This is evident in the
relatively high $\mathrm{PM}_{2.5}$ concentrations observed at the background sites of $18\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ (Tap Mun, southerly flow conditions) and $27\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ (Conghua, northerly flow conditions).
$\copyright$ 2006 Elsevier Ltd. All rights reserved.
Keywords: Pearl River Delta region; $\mathrm{PM}_{2.5}$ ; Chemical composition; Source regions; Transport
1. Introduction
While the rapid development of the Pearl River Delta Region (PRD) of China has elevated the living conditions of many Chinese citizens in both Hong Kong and Guangdong province, the fastpaced growth in population and energy use has been paralleled by the degradation of local air quality. One pollutant strongly impacted by growth in anthropogenic sources is fine particulate matter, $\mathrm{PM}_{2.5}$ . Atmospheric fine particulate pollution, produced primarily through combustion processes, has received the attention of governments and research programs around the world for its impact on human respiratory health (Dockery et al., 1993; Englert, 2004; Pope et al., 2002; Pozzi et al., 2003; Schwartz et al., 1996), visibility and climate (Chameides and Bergin, 2002; Chameides et al., 1999; Charlson et al., 1992; Haywood and Shine, 1995; Schwartz, 1996). With a lifetime in the lower atmosphere of days to weeks, $\mathrm{PM}_{2.5}$ can be transported thousands of kilometers, complicating policy development to alleviate fine particle pollution in a target region.
Home to a population of 46 million, the PRD encompasses the floodplains region of the Pearl River (Zhujiang) in the southern province of Guangdong as well as the southward mountainous New Territories and island regions of Hong Kong. With Hong Kong established as a service sector (Cullinane and Cullinane, 2003) and Guangdong province focused on manufacturing and energy production (Warren-Rhodes and Koenig, 2001), the intertwined economies have proven to be mutually beneficial to all members of the delta, where the regional GDP grew at a rate of $\sim\!17\%$ per year from 1980 to 2000 (Cullinane and Cullinane, 2003). Such phenomenal growth, however, has also yielded high concentrations of $\mathrm{PM}_{2.5}$ in many areas of the PRD, an issue of concern to local governments and citizens.
Without well-established information on major sources and source locations in the PRD, the majority of past studies have used ground-level monitoring to study the nature of regional fine particulate pollution, with the majority of these field campaigns limited to the Hong Kong area. Measurements by a number of research groups (Cao et al., 2003, 2004; Chan et al., 2001; Chao and Wong, 2002; Ho et al., 2002, 2003; Louie et al., 2005; Wei et al., 1999) have observed fine particle concentrations at various sites in the PRD consistently exceeding the annual US National Ambient Air Quality Standard of $15\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ . A recent field campaign conducted by Cao et al. (2003, 2004) sampled for $\mathrm{PM}_{2.5}$ and carbonaceous species (EC and OC) during the Winter and Summer of 2002 in four cities of the PRD, finding highest $\mathrm{PM}_{2.5}$ concentrations in Guangzhou $(106\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ , followed by Shenzhen $(61\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ , Zhuhai $(59\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ and Hong Kong $(55\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ .
Though several researchers have found evidence that sulfate concentrations in Hong Kong are influenced by regional sources (Ho et al., 2003; Louie et al., 2005; Qin et al., 1997), with more localized sources impacting the carbonaceous component (Louie et al., 2005), the location of major $\mathrm{PM}_{2.5}$ sources in the PRD has yet to be determined. In order to guide regional air quality management, an in-depth study linking regional $\mathrm{PM}_{2.5}$ chemical characteristics with influencing factors (sources, source locations and meteorology) is needed. This paper presents an overview of a large-scale monitoring study that took place for 4 months (1 month per season) over the time span of October 2002 to June 2003, with simultaneous measurements of $\mathrm{PM}_{2.5}$ mass and chemical composition at seven sites in the PRD. Measurements of local meteorology are combined with the daily concentrations of sourcerelated fine particulate species to approximate the location of influential sources and to better understand the impact of specific wind patterns on the resulting regional $\mathrm{PM}_{2.5}$ concentrations.
2. Experiment methodology
Measurements were conducted for 1 month per season at seven sites located in Hong Kong Special Administrative Region and Guangdong province.
The sites, three in Hong Kong and four in Guangdong, were selected to represent background concentrations, urban sources and receptor areas downwind of the major urban sources. The specific sites and their characteristics are listed in Table 1 with locations shown in a regional map displayed in Fig. 1. Also shown on the regional map are meteorological stations providing our project with measurements of hourly wind speed and direction as well as $24\,\mathrm{{h}}$ precipitation.
At each selected sampling site, $24\,\mathrm{h}\;\mathrm{PM}_{2.5}$ samples intended for later analysis of mass concentration and composition were collected every sixth day (five samples per month) for the months of October 2002, December 2002, March 2003 and June 2003. The sampling sites of Tap Mun (TM), Tung Chung (TC), central and western (CW), Guangzhou (GZ) and Shenzhen (SZ) used ThermoAndersen RAAS $\mathrm{PM}_{2.5}$ chemical speciation samplers, while Conghua (CH) and Zhongshan (ZS) used Caltech Gray Box samplers that served as the prototype design for the RAAS $\mathrm{PM}_{2.5}$ sampler. The Caltech Gray Box prototype, which has been described previously (Chowdhury et al., 2001), removes coarse particles via a cyclone separator (John and Reischl, 1980). Both the ThermoAndersen RAAS and Caltech Gray Box $\mathrm{PM}_{2.5}$ samplers collect fine particles onto filters in four channels (two $47\,\mathrm{mm}$ quartz filters, two $47\,\mathrm{mm}$ Teflon filters) via two separate flows of $24\,\mathrm{L}\,\mathrm{min}^{-1}$ passing through a $\mathrm{PM}_{2.5}$ cyclone, followed by a manifold splitting each flow line into a quartz filter channel $(16.7\,\bar{\mathrm{L}}\,\mathrm{min}^{-1}$ for RAAS $\mathrm{PM}_{2.5}$ and $14.0\,\mathrm{L}\,\mathrm{min}^{-1}$ for Caltech Gray Box) and a Teflon filter channel $(7.3\,\mathrm{L}\,\mathrm{min}^{-1}$ for RAAS $\mathrm{PM}_{2.5}$ and $10.0\,\mathrm{L}\,\mathrm{min}^{-1}$ for Caltech Gray Box). Flow rates were controlled by critical orifices located upstream of a vacuum pump and were measured periodically throughout each sampling month with a calibrated dry gas meter (NIST Traceable-ID $\#\mathbf{C}{\cdot}0701$ ).
Intercomparison sampling performed between colocated Caltech Gray Box and ThermoAndersen RAAS samplers demonstrated measurements of $\mathrm{PM}_{2.5}$ mass concentration to be within $5\%$ of a reference ThermoAndersen RAAS sampler and thus suitable for joint use in the PRD field measurement campaign. Operators at each of the seven sampling sites were carefully trained in filter handling and storage. After each sampling period, filter samples were sealed in Petri dishes and stored under freezing temperatures to minimize the loss of volatile species. Samples were later transported in ice-packed coolers to their eventual destinations in Hong Kong and California for mass measurement and chemical analyses. To track any contamination due to handling, four field blanks (one per sampling month) were taken at each site. The blank filters were stored and transported alongside the $24\,\mathrm{h}$ samples.
3. Chemical analyses
Following sample collection at each of the seven locations, the Teflon filters were analyzed for mass concentration, major ions (sulfate, nitrate, chloride, ammonium) and trace elements. Quartz filters were used to perform detailed speciation of organics as well as to measure elemental carbon and organic carbon. Teflon filter mass measurements were conducted via a microbalance (Mettler Instruments) following the guidelines of the EPA Quality Assurance Document 2.12 (EPA, 1998) and repeat measurements were performed on all samples to ensure accurate results. Major ion concentrations $(\mathrm{SO}_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ and $\mathrm{Cl}^{-},$ ) were determined via ion chromatography, comparing sampled concentrations with laboratory standards prepared from
ACS grade analytical reagents. For the measurement of ammonium ion $(\mathrm{NH}_{4}^{+})$ , indophenol colorimetric analysis was performed with a rapid flow analyzer (RFA-300 TM, Alpkem Corporation) (Bolleter et al., 1961). Trace elements were determined using X-ray fluorescence (XRF) analysis (Watson et al., 1996). Quartz fiber filters were analyzed with a carbon analyzer (Sunset Laboratory) for elemental carbon (EC) and organic carbon (OC) using the established NIOSH protocol of thermal evolution and combustion, initially developed by Birch and Cary (1996). A correction factor of 1.4 is applied to measured organic carbon to estimate overall organic mass (OM). While the OM to OC ratio is not necessarily a constant value in an ambient aerosol population, the value of 1.4 has been suggested by recent research as an appropriate adjustment factor to reconcile OC values to the original mass of organic compounds (Russell, 2003).
4. Results and discussion
4.1. Consolidated averages
Consolidated averages of 20 filter samples resulted in $\mathrm{PM}_{2.5}$ concentrations in Guangdong province ranging from $37\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ at rural Conghua to $71\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ at urban Guangzhou, a much smaller range is observed in Hong Kong from $29\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ at rural Tap Mun to $34\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ at urban Central/ Western. Averages of chemical composition, presented in Table 2, show that fine particulate mass within the PRD is dominated by organic compounds $(24-35\%)$ and sulfate $(21\!-\!32\%)$ . Other important measured constituents include crustal material $(7\!-\!13\%)$ , ammonium $(6–8\%)$ , elemental carbon $(3{-}8\%)$ and nitrate $(1{-}6\%)$ .
Particulate OM, formed by both direct emissions and through gas-to-particle conversion, ranged annually from 6.9 to $9.3\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ at sites in Hong Kong, and from 12.7 to $24.6\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in Guangdong. As shown in Table 2, annual mean elemental carbon (EC) concentrations were low at background sites Conghua $(1.4\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ and Tap Mun $\bar{(0.8\,\upmu\mathrm{g}\,\mathrm{m}^{-3})}$ , compared with a range of $1.9{-}4.4\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ for the five sites in more developed locations. Annual average OM/EC ratios had maximum values at background sites Conghua (9.0) and Tap Mun (8.4), with lower values of 4.0–5.9 for the remaining areas of the region. Previous studies have also measured the carbonaceous component to $\mathrm{PM}_{2.5}$ in Hong Kong and Guangdong, although different OC and EC detection methods were used and thus the data cannot be directly compared. Measurements in Hong Kong by Louie et al. (2005) and throughout both Hong Kong and Guangdong by Cao et al. (2003, 2004) utilized thermal evolution analysis based on laser reflectance (IMPROVE) with a different temperature program than the laser transmittance method (NIOSH) that is used in this study. Chow et al. (2001) verified experimentally that the two methods have similar results for total carbon but variability exists in the division of carbon between EC and OC. Ho et al. (2003) measured EC and OC in Hong Kong using thermal manganese dioxide oxidation (TMO method), which has been found to closely compare to results from the IMPROVE protocol (Fung et al., 2002).
Average particulate sulfate concentrations, generated primarily through the oxidation of emitted gaseous $\mathrm{SO}_{2}$ , had lower values and a smaller range in Hong Kong $(9.0{-}9.3\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ compared with that in Guangdong $(10.0–14.7\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ ). The close range in measured sulfate at sites representing both urban and background locations in Hong Kong supports the notion that sulfate has regional and perhaps longer range sources. Our measurements at sites in Hong Kong are similar to those by Louie et al. (2005), who measured average $\mathrm{PM}_{2.5}$ sulfate concentrations at three locations in Hong Kong in the range of $8.7–9.4\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ during 2000–2001. In comparing sulfate and ammonium ion concentrations, it is observed that the 4-month average ammonium/sulfate molar charge ratio ranges from 0.60 to 0.85 among the seven sites in the PRD. This indicates that the aerosol phase is acidic, with ammonium principally associated with sulfate. A recent emissions inventory by Streets et al. (2003) found agricultural activities to be the main source of ammonia in China, which is the likely explanation for the higher levels of ammonium at the sites in Guangdong compared with Hong Kong.
With overall average concentrations of crustal and trace species shown in Table 2, two specific species related to anthropogenic activities, lead (fuel combustion, industrial sources) and potassium (biomass burning), will be discussed in further detail. While fine particulate potassium can also be derived from crustal material, the high portion of water soluble potassium $(\sim\!90\%)$ in Hong Kong’s $\mathrm{PM}_{2.5}$ reported by Louie et al. (2005) indicates that crustal sources have a limited influence on resulting potassium levels. Lead and potassium are both at trace levels in samples, though are frequently linked with other particulate species (OM, EC) that constitute a larger portion of fine particulate mass. It is of note that consolidated averages of lead in the PRD have lower levels in Hong Kong $(0.05{-}0.06\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ compared with Guangdong $(0.08–0.26\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ . Lead concentrations in Hong Kong are similar to past measurements by Louie et al. (2005), who found a range of $0.06{-}0.07\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ at three sites in Hong Kong during year-long monitoring in 2000–2001. Overall lead concentrations in the PRD are similar to elsewhere in
Asia, with wintertime levels at a background site in Korea measuring 0.03 and $0.20\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ observed at urban Seoul (Mishra et al., 2004). He et al. (2004) reported high average lead concentrations in Beijing, with an annual average at a residential area in the city $(0.33\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ exceeding our consolidated average at Guangzhou $(0.26\,\upmu\mathrm{g}\,\mathrm{m}^{-\bar{3}})$ ). As with lead, potassium concentrations in Hong Kong $(0.56{-}0.57\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ are lower than that measured northward in Guangdong $(0.74{-}1.60\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ . The presented potassium values in Hong Kong are comparable with the range of $0.49–0.58\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ measured during 2000–2001 by Louie et al. (2005). Linked with biomass burning, measured potassium concentrations in the PRD are within the range of values measured during biomass burning events in Korea $(0.49\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ for the burning of rice straw, $4.19\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ for burning of barley) and more than double levels observed during non-burning periods $(0.25\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ (Ryu et al., 2004).
4.2. Meteorology case studies
In order to provide insight into source area locations as well as meteorological influences (i.e. wind speed and precipitation) on fine particulate concentrations, daily surface wind patterns are compared with concentrations of fine particulate species at the seven sites in the PRD. In the four months of sampling, three unique meteorological cases were identified which can be summarized as follows: (1) ‘‘southerly flow’’ characterized by low to moderate winds from the South; (2) ‘‘northerly flow’’ having moderate to strong winds from the North, and (3) ‘‘mixed flow’’ associated with weak winds (wind speed $<\!3\,\mathrm{m}\,\mathrm{s}^{-1}.$ ) shifting in direction throughout the day. Table 3 lists the 13 sampling days (of 20 total) categorized into each meteorological case, with the remaining 7 days excluded for not clearly fitting into one of the three identified categories or for inconsistency in wind measurements among monitoring sites. Measured hourly wind speed, wind direction and daily precipitation for each categorized day are shown in Fig. 2. Though some variability exists among the meteorology sites, the presented measurements at the Shenzhen meteorology station (Lat: 22.5500, Long: 114.1000) represent general observed trends in wind and precipitation at the five other meteorological sites shown in Fig. 1. It should be noted that surface wind measurements do not necessarily represent large-scale flow patterns, as local topography can affect surface measurements. However, the selected meteorological sites in the PRD are strategically placed to represent regional rather than local winds.
To relate upwind source regions with downwind concentrations of fine particulate matter, specific species linked with sources are examined for each meteorological category, including sulfate (coal combustion), organic compounds (combustion of fossil fuels, biomass burning, industrial sources, local cooking), elemental carbon (poor coal combustion, fuel oil combustion, combustion of diesel gasoline), potassium (biomass burning) and lead (combustion of leaded gasoline, industrial sources). In order to compare the grouped series of days, each daily measured species is normalized by the concentration of the identical species on the same day at the Guangzhou site and then the relative concentrations of the grouped days are averaged. Guangzhou was selected as a reference site because it is centrally located, has the highest average fine particulate matter concentration among the seven monitoring sites and is hypothesized to be a major source area contributing to downwind concentrations. The normalization lessens any bias due to precipitation events and seasonally changing source strengths, such as the variability of biomass burning events throughout the year. Assuming that source locations are remaining constant, the normalization allows a clear view into impacts of wind patterns on relative concentrations among the seven sites. Shown in Fig. 3, the average relative concentration and standard error of the selected species at each site are categorized into southerly, northerly and mixed flows. Also, non-normalized average concentrations for each meteorological case are shown in Table 4, though it should be noted that looking at a single site’s change between meteorological cases may be biased by an uneven distribution of precipitation events within each category.
4.2.1. Southerly flow
Relative concentrations of $\mathrm{PM}_{2.5}$ and specific chemical species for southerly flow conditions can be seen in the leftmost column of graphs in Fig. 3. Average values for each of the flow conditions are given in Table 4. During days with southerly winds transporting air masses northward from the ocean, a large difference is seen between $\mathrm{PM}_{2.5}$ average concentrations at the southernmost five sites (Tap Mun, Tung Chung, and central/western in Hong Kong; Shenzhen and Zhongshan in Guangdong) and the northern two (Guangzhou and Conghua in Guangdong). As shown in Table 4, mean values from the southern five sites range from 18 to $25\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , as compared to $47\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ measured at the more northerly urban Guangzhou site. The mean $\mathrm{PM}_{2.5}$ concentration at the northern background site in Conghua is $40\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , nearly twice as high as the sites south of Guangzhou. It is apparent that a source area must be located near Guangzhou to cause the observed accumulation of fine particulate mass over a relatively short distance. Comparing rural Conghua located North of Guangzhou with the Zhongshan site placed South of Guangzhou, an $18\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ increase representing a near doubling of downwind $\mathrm{PM}_{2.5}$ is observed and can be attributed to sources located in the region between the two sites, including the city of
Guangzhou. In addition to the contribution to downwind $\mathrm{PM}_{2.5}$ by the Guangzhou area, it should be noted that the levels at the upwind background site of Tap Mun $(18\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ indicate significant region-wide background $\mathrm{PM}_{2.5}$ . To investigate further the origin of the particulate matter, $24\,\mathrm{{h}}$ back-trajectory modeling was performed on the 4 days of southerly flow, using the HYbrid SingleParticle Lagrangian Integrated Trajectory (HYSPLIT) model (Draxler and Rolph, 2003) with NCEP/GDAS FNL reanalysis meteorological data. Modeled trajectories were calculated for the two background sites, Tap Mun and Conghua, at elevations of 100, 500 and $1500\,\mathrm{m}$ . Although the rugged terrain of the region imparts uncertainty to meteorological modeling, the HYSPLIT model did confirm that the 4 southerly-flow days had air parcels transported inland from the ocean area to the South. This implies that southerly flow fine particulate levels at rural Tap Mun may be due to long-range transport. Local shipping emissions may also affect overall levels at Tap Mun, but the relative influence of this source requires further investigation.
Sulfate is a dominant component of fine particulate matter in the PRD, on average contributing $21{-}32\%$ of overall mass, as shown in Table 2. The relative concentrations of sulfate during southerly wind are highest at the Guangdong background site at Conghua (1.09), shown in Fig. 3. As with overall $\mathrm{PM}_{2.5}$ , relative values of sulfate at the southern five sites are about half that measured in Conghua and are within a close range of one another (0.44–0.53), with absolute concentrations of $6.0–7.1\;\upmu\mathrm{g}\,\mathrm{m}^{-3}$ . It is apparent that a source of sulfate lies in the Guangzhou vicinity, leading to a doubling of concentrations from sites South of Guangzhou to northernmost Conghua. However, though levels of sulfate at the southern sites are far less than northern areas, the sulfate concentrations at the southernmost five sites are still substantial. With sulfate levels at remote Tap Mun similar to that at urban Shenzhen, it is expected that sulfate has regional background levels during southerly flow that results in approximately half of the PRD’s sulfate mass. This background sulfate may be due to long-distance transport from outside of the PRD region.
Both primary and secondary in origin, relative concentrations of organic compounds at Conghua are more than double the values measured at the southern five sites during southerly wind patterns, as shown in Fig. 3. The change between the southern five sites and Guangzhou is even more extreme, with a near tripling of relative concentrations. Comparing the southerly flow distributions of OM and sulfate, a higher range of OM concentrations is observed among the southernmost five sites $(3.3–5.8\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ , with mean normalized concentrations also showing a wider range (0.20–0.35). Some localized influence on OM is thus predicted in the Hong Kong vicinity, with a tripling of OM concentrations in Guangzhou attributed to both transport of precursors from the South and locally emitted organic species in the Guangzhou area. The relatively high OM concentrations observed in Guangzhou are perhaps not surprising given the size of the city and intense traffic congestion.
Elemental carbon during southerly flow has a unique pattern compared with $\mathrm{PM}_{2.5}$ , sulfate and OM. Among the southern five sites, background Tap Mun has a low relative concentration of 0.13 while more developed sites at Shenzhen and Zhongshan are nearly three-fold higher with relative concentrations of 0.55 and 0.47, respectively. With normalized concentrations at the Hong Kong sites of Tung Chung and central/western more than doubling background Tap Mun and even higher increases at urban sites within Guangdong, local sources appear to strongly affect EC levels throughout the delta. Though EC concentrations seem to be mainly dominated by local sources, some impact of transport is apparent with high relative levels observed at Conghua (0.56), located downwind of Guangzhou. As with OM, the relatively high EC concentration in urban Guangzhou strongly suggests the importance of local sources on particulate matter concentrations.
As shown in Fig. 3, the southerly flow case for potassium shows extraordinarily high relative concentrations at rural site Conghua, more than doubling urban Guangzhou and higher than sites South of Guangzhou by a factor of five. A tracer for biomass burning, the significant increases moving northward throughout the delta lead to the conclusion that biomass burning sources are distributed within the northern section of the monitoring area, both near Guangzhou and North of the city toward Conghua. Comparatively low measured potassium at sites south of Guangzhou points to a lack of biomass burning in close proximity to Hong Kong and southern Guangdong.
Of all observed species, lead appears to be most significantly dominated by sources in Guangdong, with homogeneously low relative concentrations South of Guangzhou (ranging from .09 to .11) and levels at Guangzhou and Conghua higher by more than six-fold. Shown in Fig. 3, the sudden jump in lead levels moving from sites Zhongshan and Shenzhen to nearby Guangzhou indicates a localized source area of lead within the vicinity of Guangzhou and perhaps North of the city. Assuming no local production of lead near Conghua, the high lead concentrations at Conghua appear to be caused by transport from upwind Guangzhou.
4.2.2. Northerly flow
As seen in Fig. 3, northerly flow relative levels of $\mathrm{PM}_{2.5}$ at sites South of Guangzhou more than double that observed during southerly flow, while normalized concentration at northernmost Conghua decreases by 0.2. A spatial gradient is seen among sites downwind of Guangzhou, with highest relative concentrations at Zhongshan (1.4) and Shenzhen (1.2) and lower levels in the Hong Kong area (0.75–0.93). Comparing upwind Conghua and downwind Zhongshan, an increase of $34\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ can be linked to the Guangzhou region located inbetween the two sites. With an attenuation of impact related to distance from Guangzhou, the increase in concentration at the background site of Tap Mun relative to Conghua is $6.3\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ . Overall, the doubling increase in relative concentrations at the three sites in Hong Kong as compared to southerly flow conditions points to the significant impact of the Guangzhou area on levels of fine particulate matter in Hong Kong. In addition to increases observed downwind of Guangzhou, northerly flow $\mathrm{PM}_{2.5}$ measured upwind at rural Conghua is significantly high $({\sim}27\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ , indicating a regional background concentration that may be due to long-range transport from northern areas.
Similar to the reversal observed in $\mathrm{PM}_{2.5}$ concentrations when comparing cases of northerly and southerly flows, particulate sulfate levels likewise increase at sites downwind of Guangzhou and decrease at Conghua, located upwind of Guangzhou. As observed in Fig. 3, Zhongshan receives the heaviest dose of sulfate, with concentrations relative to Guangzhou at 1.34 compared with 0.53 under southerly flow. Conghua now has lowest relative concentrations in the region (0.78) compared with ranking highest when downwind of Guangzhou. Assuming negligible local impact on particulate sulfate concentrations at Conghua, its northerly flow average sulfate concentration of $7.8\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ indicates significant background particulate sulfate advected into the PRD that constitutes over half of the $13.3\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ of sulfate measured at Zhongshan. Examining Table 4, it is interesting to note that the average difference between the maximum at Conghua and upwind Zhongshan under southerly flow is $5.3\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , while the northerly flow difference between the same two sites is $5.5\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ . Thus, the direct contribution of the Guangzhou vicinity to particulate sulfate can be roughly estimated at 5–6 mg m 3.
Having a nearly identical distribution as the northerly flow case of $\mathrm{PM}_{2.5}$ , OM concentrations appear to be influenced by a source area near Guangzhou. Impact based on proximity to Guangzhou is again observed, with Zhongshan and Shenzhen having much higher average OM levels relative to that seen during southerly flow and lesser increases in OM concentrations at sites in Hong Kong. Though OM concentrations appear to have a regional increase at sites in Hong Kong, a localized influence is still evident with normalized OM levels observed at rural Tap Mun ${\sim}20\%$ lower than nearby Tung Chung and central/western, as shown in Fig. 3. With expected biomass burning sources near Conghua, as indicated by high potassium levels during southerly flow, OM concentrations at Conghua may be likewise influenced by nearby sources and thus not indicative of regional background levels. Even with the possible presence of OM sources near Conghua, the impact of transported organic particulate species from Guangzhou is significant. Absolute OM concentrations displayed in Table 4 show a northerly flow difference between downwind Zhongshan and upwind Conghua of 14.5 and an $8.5\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ increase from Zhongshan to Conghua during southerly flow.
With much lower relative concentrations at background sites (Tap Mun, Conghua) compared with urban areas (Shenzhen, Guangzhou), the distribution of EC during northerly flow appears to be dominated by local sources. However, it should be pointed out that some degree of transport is seen in concentrations at the background sites. Comparing the case of northerly winds to that of southerly winds, relative EC concentrations at Tap Mun increase by 0.21 during flow from the North, while normalized levels at Conghua decrease by 0.30, as shown in Fig. 3. Despite the observed transport of EC, local influence appears to remain significant at the three sites in Hong Kong, with much higher EC concentrations at central/western and Tung Chung (2.3 and $2.7\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , respectively) as compared with background Tap Mun $(1.3\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ .
While normalized potassium at Conghua is seen to double levels at Guangzhou during southerly flow, as shown in Fig. 3, Conghua ranks lowest among all sites during wind from the North, indicating significant biomass burning occurring South of Conghua. The reversal of flow causes highest relative levels at Zhongshan (1.9) with lessening impact moving southward to the sites in Hong Kong (.78–.91). Though Conghua has the lowest potassium concentration during the case of northerly flow, the level is still relatively high $(0.85\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ and indicates a background contribution that constitutes more than a third of the peak level observed at Zhongshan. The background potassium during northerly flow may be caused by biomass burning located near Conghua or due to transport from North of the PRD region.
Normalized particulate lead concentrations are observed to dramatically increase at sites south of Guangzhou when comparing the case of northerly flow to southerly flow, with a maximum 20-fold increase observed at Zhongshan and a minimum five-fold increase observed at central/western. With relative concentrations spiking at Zhongshan (2.0), attenuation is again seen moving southward to Shenzhen (1.0) and the sites in Hong Kong (0.56–0.71). While absolute lead concentrations, shown in Table 3, are observed to increase from 0.01 to $0.10\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ at the background site of Tap Mun, comparing cases of flow from the North and South, only a slight decrease of $0.02\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in concentration is seen at Conghua. It is expected that particulate lead is regionally advected into the PRD during flow from the North, maintaining high concentrations of lead at Conghua. However, local sources of lead at the background site in Guangdong cannot be ruled out. Even given a rise in background levels of lead during northerly flow, a tripling in absolute lead concentrations from Conghua to Zhongshan indicates lead emissions local to Guangzhou.
4.2.3. Mixed flow
Even though larger rainfall was observed during days within the mixed flow category in comparison with northerly and southerly flows, the stagnant conditions result in extremely high $\mathrm{PM}_{2.5}$ concentrations at Guangzhou, with levels more than $60\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ higher than that observed for the other two flow categories, as presented in Table 4. In comparison to the dramatic change observed in Guangzhou, a more muted increase is seen at the remainder of sites in the delta, resulting in relative fine particulate levels at Guangzhou more than $40\%$ higher than any other sampling site. This observed maximum at Guangzhou is similarly observed for all presented species (sulfate, OM, EC, potassium and lead). With limited transport of fine particulate concentrations, it is observed that sources within the vicinity of Guangzhou heavily impact local pollution during mixed flow. Throughout the delta, stagnant winds result in an accumulation of fine particulate matter from both local and regional sources, leading to higher $\mathrm{PM}_{2.5}$ concentrations at both background sites at Conghua $(46\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ and Tap Mun $(39\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ than observed during southerly or northerly flows. The relatively low wind speeds likely favor more local influences at the sampling stations.
Compared with all other examined species, the mixed flow distribution of normalized sulfate concentrations indicates the most significant degree of regional impact, with high relative concentrations (0.53–0.74) observed at the six sites surrounding Guangzhou. In contrast, the same six sites have lower relative levels of $\mathrm{PM}_{2.5}$ (0.36–0.58) as shown in Fig. 3. Though sulfate sources appear to have a regional influence on concentrations in the PRD, sources in the vicinity of Guangzhou city cause sulfate concentrations at Guangzhou to measure more than $25\%$ higher than any other site in the region.
With a nearly identical distribution as relative concentrations of $\mathrm{PM}_{2.5}$ , normalized OM has a maximum at Guangzhou during stagnant conditions, significantly higher than the remainder of sites (ranging from 0.25 to 0.55). In comparison to the distribution of sulfate among the seven monitoring sites during mixed flow, OM levels appear to have a more localized impact. Major sources of OM are expected to be located near Guangzhou, causing a doubling of average OM concentrations at Guangzhou in comparison with southerly and northerly flows, despite higher precipitation during mixed flow days. Although a doubling in absolute OM is observed at Guangzhou, comparing mixed flow to northerly flow, a $10–20\%$ decrease is observed at the southernmost five sites, indicating an isolation of high OM levels to Guangzhou.
As observed during northerly and southerly flows, a localized influence on EC is again apparent during mixed flow, with relative concentrations at the urban areas of Guangzhou and Shenzhen (0.69) much higher than background levels at Conghua (0.29) and Tap Mun (0.18). Localized sources are also apparent within the Hong Kong region, with normalized EC at central/western and Tung Chung more than doubling that measured at background Tap Mun.
The distribution of potassium concentrations during mixed flow supports a source area located near Guangzhou and Conghua. Ranked highest in the region during stagnant conditions, levels of potassium at Guangzhou $(2.8\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ are more than double concentrations measured during both southerly and northerly flows, in spite of higher rainfall during mixed flow days. Though potassium levels increase at Conghua by $130\%$ , comparing the case of mixed flow to northerly flow, a decrease of $40–60\%$ is measured among the five sites South of Guangzhou. The stagnant winds appear to isolate high concentrations of potassium to the northern PRD region and indicate a source area affecting concentrations at both Guangzhou and Conghua.
With a nearly identical distribution as particulate potassium, levels of lead during mixed flow similarly appear to be dominated by sources in the northern PRD and have little transport to sites south of Guangzhou. Compared with northerly flow, the southernmost five sites have a $30–60\%$ reduction in absolute levels of lead while Guangzhou and Conghua increase by $140\%$ and $160\%$ , respectively. With mixed flow lead concentrations at remote Tap Mun $(0.07\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ similar to that at urban Shenzhen $(0.08\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ , particulate lead in Hong Kong vicinity appears to be regionally controlled.
5. Conclusions
Development of an effective fine particulate management plan in the Pearl River Delta has been hindered by a lack of information about the regional nature of $\mathrm{PM}_{2.5}$ , with regional chemical composition, influencing source areas and meteorological impacts yet unknown. To assess fine particulate pollution throughout the PRD, simultaneous $24\,\mathrm{{h}}$ filter measurements were conducted at seven sites during October 2002, December 2002, March 2003 and June 2003. Combining the 4 months of sampling, overall average $\mathrm{PM}_{2.5}$ concentrations at all sites far exceed the United States national ambient air quality standard of $15\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ (annual average), with levels in Hong Kong ranging from 29 to $34\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ and even higher average concentrations in Guangdong ranging from $37{-}71\;\upmu\mathrm{g}\,\mathrm{m}^{-3}$ . Despite the variability in concentrations throughout the region, the general chemical make-up of the fine particulate matter is very similar among the seven sites, with organic mass and sulfate dominating fine particulate mass at $24–35\%$ and $21{-}32\%$ , respectively. The significant levels of organic mass and sulfate, both related to fossil fuel combustion, points to the impact of anthropogenic activities on local air quality. Other measured species include crustal matter $(7-13\%)$ , ammonium $(6{-}8\%)$ , elemental carbon $(3{-}8\%)$ and nitrate $(1{-}6\%)$ . To our knowledge, these are the first reported measurements of overall chemical composition of fine particulate matter in Guangdong.
Combined analysis of local meteorology (wind speed, wind direction and precipitation) and fine particulate levels illustrates the significant influence wind patterns have on regional air quality. The impact of transport was investigated by categorizing 13 (of 20 total) sampling days by prevailing wind direction (southerly, northerly and low-speed mixed flows). Comparison of $\mathrm{PM}_{2.5}$ levels at sites immediately upwind and downwind of Guangzhou during northerly and southerly flow conditions indicates an estimated contribution of $18{-}34\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ to downwind $\mathrm{PM}_{2.5}$ by sources in the vicinity of Guangzhou city. A gradient effect was observed, with the most extreme increases in fine particulate matter during northerly winds occurring at Zhongshan, located close to Guangzhou and lesser change observed at the more distant sites.
Looking into the impacts of wind patterns on the spatial distribution of specific fine particulate species, contributing source regions can be assessed in greater detail. Sulfate, related to the burning of coal, has a high regional background level estimated at $6{-}8\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , over half of the total measured sulfate. Analysis of sulfate concentrations at sites upwind and downwind of Guangzhou indicates a direct input of $5{-}6\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ from sources near Guangzhou city. Guangzhou also stands out as a prominent regional source of organic mass (OM), with observed increases of $8.5–1\bar{4}.5\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ at sites immediately downwind and a disproportionate elevation in organic mass at the Guangzhou site during stagnant conditions. Local OM sources are also evident in the Hong Kong region, with rural Tap Mun consistently lower than urban central/ western. regional levels of elemental carbon (EC) are highest during all flow conditions at urban Shenzhen and Guangzhou. Local sources of EC are evident within Hong Kong, with more developed sites at Tung Chung and central/western having nearly double the EC concentrations measured at rural Tap Mun. In contrast, the distribution of potassium (biomass burning) and lead (industrial sources, combustion of fossil fuels) indicate significant sources in northern area of the delta influencing concentrations downwind. The regional distribution of potassium points to sources in the vicinity of both Guangzhou and Conghua, with strikingly high levels observed at Conghua during southerly flow. Regional levels of lead appear to be controlled by sources in the vicinity of Guangzhou.
Acknowledgments
This research was funded by Civic Exchange and through a National Science Foundation Graduate Fellowship to the author (G.H.). This study would not have been possible without the sampling coordination provided by Tao Liu of the Guangzhou Environmental Monitoring Center, Jianjun Chen of the Conghua Environmental Monitoring Center and Wendong Yang of the Zhongshan Environmental Monitoring Center. Generous assistance with sampling coordination in Hong Kong and general project oversight was provided by the Hong Kong Environmental Protection Department (HKEPD), with Dr. Peter Louie especially appreciated for his investment of time and expertize in guiding the involvement of the HKEPD. In addition, we are grateful to the HKEPD and the Chinese Meterological Association for access to measurements made at meterological monitoring stations throughout the Pearl River Delta. We also want to thank Dr. Tao Wang of Hong Kong Polytechnic University for his insight and assistance throughout this study.
References
Birch, M.E., Cary, R.A., 1996. Elemental carbon-based method for monitoring occupational exposures to particulate diesel exhaust. Aerosol Science And Technology 25 (3), 221–241.
Bolleter, W.T., Bushman, C.J., Tidwell, P.W., 1961. Spectrophotometric determination of ammonia as indophenol. Analytical Chemistry 33 (4), 592.
Cao, J.J, et al., 2003. Characteristics of carbonaceous aerosol in Pearl River Delta region, China during 2001 Winter period. Atmospheric Environment 37 (11), 1451–1460.
Cao, J.J., et al., 2004. Spatial and seasonal variations of atmospheric organic carbon and elemental carbon in Pearl River Delta region, China. Atmospheric Environment 38 (27), 4447–4456.
Chameides, W.L., Bergin, M.H., 2002. Soot takes center stage. Science 297, 2214–2215.
Chameides, W.L., et al., 1999. Case study of the effects of atmospheric aerosols and regional haze on agriculture: an opportunity to enhance crop yields in China through emission controls? Proceedings from the National Academy of Sciences 96 (24), 13626–13633.
Chan, L.Y., Kwok, W.S., Lee, S.C., Chan, C.Y., 2001. Spatial variation of mass concentration of roadside suspended particulate matter in metropolitan Hong Kong. Atmospheric Environment 35 (18), 3167–3176.
Chao, C.Y., Wong, K.K., 2002. Residential indoor $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ in Hong Kong and the elemental composition. Atmospheric Environment 36 (2), 265–277.
Charlson, R.J., et al., 1992. Climate forcing by anthropogenic aerosols. Science 255, 423–430.
Chow, J.C., Watson, J.G., Crow, D., Lowenthal, D.H., Merrifield, T., 2001. Comparison of IMPROVE and NIOSH carbon measurements. Aerosol Science and Technology 34 (1), 23–34.
Chowdhury, Z., Hughes, L.S., Salmon, L.G., Cass, G.R., 2001. Atmospheric particle size and composition measurements to support light extinction calculations over the Indian Ocean. Journal Of Geophysical Research—Atmospheres 106 (D22), 28597–28605.
Cullinane, S., Cullinane, K., 2003. City profile: Hong Kong. Cities 20 (4), 279–288.
Dockery, D.W., et al., 1993. An association between air pollution and mortality in 6 United States cities. New England Journal of Medicine 329 (24), 1753–1759.
Draxler, R.R., Rolph, G.D., 2003. HYSPLIT (HYbrid SingleParticle Lagrangian Integrated Trajectory) Model. http:// www.arl.noaa.gov/ready/hysplit4.html, NOAA Air Resources Laboratory, Silver Spring, MD (access via NOAA ARL READY website).
Englert, N., 2004. Fine particles and human health—a review of epidemiological studies. Toxicology Letters 149, 235–242.
EPA, 1998. Quality Assurance Guidance Document 2.12. US Environmental Protection Agency.
Fung, K., Chow, J.C., Watson, J.G., 2002. Evaluation of OC/EC speciation by thermal manganese dioxide oxidation and the IMPROVE method. Journal of the Air & Waste Management Association 52 (11), 1333–1341.
Haywood, J.M., Shine, K.P., 1995. The effect of anthropogenic sulfate and soot aerosol on the clear-sky planetary radiation budget. Geophysical Research Letters 22 (5), 603–606.
He, K.B., et al., 2004. Concentration level of fine airborne lead in Beijing, People’s Republic of China. Bulletin of Environmental Contamination and Toxicology 72 (2), 233–239.
Ho, K.F., Lee, S.C., Yu, J.C., Zou, S.C., Fung, K., 2002. Carbonaceous characteristics of atmospheric particulate matter in Hong Kong. Science of the Total Environment 300 (1-3), 59–67.
Ho, K.F., et al., 2003. Characterization of chemical species in $\mathrm{PM}_{2.5}$ and $\mathrm{{PM}_{10}}$ aerosols in Hong Kong. Atmospheric Environment 37 (1), 31–39.
John, W., Reischl, G., 1980. A cyclone for size-selective sampling of ambient air. Journal of the Air Pollution Control Association 30 (8), 872–876.
Louie, P.K.K., et al., 2005. $\mathrm{PM}_{2.5}$ chemical composition in Hong Kong: urban and regional variations. Science of the Total Environment 338, 267–281.
Mishra, V.K., Kim, K.H., Kang, C.H., Choi, K.C., 2004. Wintertime sources and distribution of airborne lead in Korea. Atmospheric Environment 38 (17), 2653–2664.
Pope III, C.A., et al., 2002. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. Journal of the American Medical Association 287 (9), 1132–1141.
Pozzi, R., De Berardis, B., Paoletti, L., Guastadisegni, C., 2003. Inflammatory mediators induced by coarse $(\mathbf{PM}_{2.5-10})$ and fine $(\mathbf{PM}_{2.5})$ urban air particles in RAW 264.7 cells. Toxicology 183, 243–254.
Qin, Y., Chan, C.K., Chan, L.Y., 1997. Characteristics of chemical compositions of atmospheric aerosols in Hong Kong: spatial and seasonal distributions. Science of the Total Environment 206, 25–37.
Russell, L.M., 2003. Aerosol organic-mass-to-organic-carbon ratio measurements. Abstracts of Papers of the American Chemical Society 225, U817–U818.
Ryu, S.Y., Kim, J.E., Zhuanshi, H., Kim, Y.J., Kang, G.U., 2004. Chemical composition of post-harvest biomass burning aerosols in Gwangju, Korea. Journal of the Air & Waste Management Association 54 (9), 1124–1137.
Schwartz, S.E., 1996. The whitehouse effect—shortwave radiative forcing of climate by anthropogenic aerosols: an overview. Journal of Aerosol Science 27 (3), 359–382.
Schwartz, J., Dockery, D.W., Neas, L.M., 1996. Is daily mortality associated specifically with fine particles? Air & Waste Management Association 46, 927–939.
Streets, D.G., Bond, T.C., Carmichael, G.R., et al., 2003. An inventory of gaseous and primary aerosol emissions in Asia in the year 2000. Journal of Geophysical Research—Atmospheres 108 (D21).
Warren-Rhodes, K., Koenig, A., 2001. Ecosystem appropriation by Hong Kong and its implications for sustainable development. Ecological Economies 39, 347–359.
Watson, J.G., Chow, J.C., Frazier, C.A., 1996. X-ray fluorescence analysis of ambient air samples. Elemental Analysis of airBorne Particles. Gordon and Breach, Newark, pp. 1–31.
Wei, F., et al., 1999. Ambient concentrations and elemental compositions of $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ in four Chinese cities. Environmental Science & Technology 33 (23), 4188–4193.
|
atmosphere
|
0025
|
10.5194/acp-12-1377-2012
|
table
|
Table 1. Annual mass concentrations of $\mathrm{PM}_{2.5}$ , its major ionic species and their equivalent ratios at Beijing and Chongqing from March 2005 to February 2006.
|
|
atmosphere
|
0025
|
10.5194/acp-12-1377-2012
|
table
|
Table 2. Seasonal $\mathrm{R_{C/A}}$ at multiple sites in the northern and southern China $(\mathrm{R}_{\mathrm{C/A}}\!=\!(\mathrm{NH}_{4}^{+}+\mathrm{Ca}^{2+})/(\mathrm{SO}_{4}^{2-}+\mathrm{NO}_{3}^{-})$ (µeq/µeq), spring $=$ MAM, summer $=\mathrm{JJA}$ , $\mathrm{{fall}=S O N}$ , winter $=\mathrm{D}\mathbf{F}_{\mathrm{.}}$ ).
|
|
atmosphere
|
0025
|
10.5194/acp-12-1377-2012
|
image
|
Fig. 1. Seasonal variations of (a) $\mathrm{R_{C/A}}$ (uncertainty $=$ standard deviation) and (b) normalized $\mathrm{R_{C/A}}$ of $\mathrm{PM}_{2.5}$ at Beijing and Chongqing.
|
|
atmosphere
|
0025
|
10.5194/acp-12-1377-2012
|
image
|
Fig. 2. Seasonal variations of in situ $\operatorname{PM}_{2.5}\operatorname{pH}$ , $[\mathrm{H^{+}}]_{\mathrm{Ins}}$ , $[\mathrm{H}_{2}\mathrm{O}]$ and RH at (a) Beijing and (b) Chongqing.
|
|
atmosphere
|
0025
|
10.5194/acp-12-1377-2012
|
image
|
Fig. 3. (a) Daily variation in the subtropical high over the northwestern Pacific between $110{-}130^{\circ}\,\mathrm{E}$ (Lu et al., 2007); (b) the geopotential heights of $500\,\mathrm{hPa}$ at UTC 08:00, 23 June 2005 over East Asia; (b) the geopotential heights of $500\,\mathrm{hPa}$ at UTC 12:00, 29 June 2005 over East Asia.
|
|
atmosphere
|
0025
|
10.5194/acp-12-1377-2012
|
image
|
Fig. 4. Clusters of air mass backward trajectories arriving at $500\,\mathrm{m}$ above ground level at Beijing for (a) 4 March–6 May, (c) $7{-}31\ \mathrm{May},$ (d) 1–27 June in 2005 and (b) 3 March–5 May in 2006.
|
|
atmosphere
|
0025
|
10.5194/acp-12-1377-2012
|
image
|
Fig. 5. Clusters of air mass backward trajectories arriving at $500\,\mathrm{m}$ above ground level at Chongqing for (a) 4 March–6 May, (c) 7–31 May, (d) 1–27 June in 2005 and (b) 3 March–5 May in 2006.
|
|
atmosphere
|
0025
|
10.5194/acp-12-1377-2012
|
image
|
Fig. 6. Differences of $\mathrm{R_{C/A}}$ , $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ , concentrations of associated ionic species and meteorological factors between the two springs of 2005 and 2006 for (a) Beijing (TH and MY) and (b) Chongqing (JB, DDK and BB), which were calculated based on $(2006{-}2005)/2005$ and (2005–2006)/2006, respectively.
|
|
atmosphere
|
0025
|
10.5194/acp-12-1377-2012
|
image
|
3.5 Formation of $\mathbf{No}_{3}^{-}$ at different levels of aerosol acidity Fig. 7. Monthly variations of (a) amount of precipitation vs. concentration of $\mathrm{NH}_{4}^{+}$ in $\mathrm{PM}_{2.5}$ and (b) equivalent charge ratios of $\mathrm{NH}_{4}^{+}/\mathrm{SO}_{4}^{2-}$ in $\mathrm{PM}_{2.5}$ vs. precipitation at Chongqing from February 2005 to April 2006.
|
|
atmosphere
|
0025
|
10.5194/acp-12-1377-2012
|
image
|
Fig. 8. Molar ratios of $[\mathrm{NO}_{3}^{-}]/[\mathrm{SO}_{4}^{2-}]$ vs. $[\mathrm{NH}_{4}^{+}]/[\mathrm{SO}_{4}^{2-}]$ at different (a) acidity and (b) water content $\left(\mathrm{[H}_{2}\mathrm{O}\right)\right)$ in $\mathrm{PM}_{2.5}$ at Beijing (TH and MY) and Chongqing (JB, DDK and BB). LA, less acidic. MA, more acidic.
|
|
atmosphere
|
0025
|
10.5194/acp-12-1377-2012
|
image
|
Fig. 9. Relationships between molar concentrations of $[\mathrm{NO}_{3}^{-}]$ and $[\mathrm{NH}_{4}^{+}]_{\mathrm{Excess}}$ in $\mathrm{PM}_{2.5}$ at Beijing (TH and MY) and Chongqing (JB, DDK and BB).
|
|
atmosphere
|
0025
|
10.5194/acp-12-1377-2012
|
text
| Not supported with pagination yet
|
Spatial and seasonal variability of $\mathbf{PM}_{2.5}$ acidity at two Chinese megacities: insights into the formation of secondary inorganic aerosols
K. $\mathbf{H}\mathbf{e}^{1}$ , Q. Zhao1, Y. Ma1, F. Duan1, F. Yang2, Z. Shi3, and G. Chen4
1State Key Joint Laboratory of Environment Simulation and Pollution Control, School of Environment, Tsinghua University, Beijing 100084, China
2Key Laboratory of Computational Geodynamics, College of Earth Science, Graduate University of Chinese Academy of Sciences, Beijing 100049, China
3School of Geography, Earth and Environmental Science, University of Birmingham, Edgbaston Birmingham B15 2TT, UK 4Chongqing Environmental Protection Bureau, Chongqing 401147, China
Correspondence to: K. He (hekb $@$ tsinghua.edu.cn)
Received: 30 August 2011 – Published in Atmos. Chem. Phys. Discuss.: 13 September 2011
Revised: 11 January 2012 – Accepted: 24 January 2012 – Published: 6 February 2012
Abstract. Aerosol acidity is one of the most important parameters influencing atmospheric chemistry and physics. Based on continuous field observations from January 2005 to May 2006 and thermodynamic modeling, we investigated the spatial and seasonal variations in $\operatorname{PM}_{2.5}$ acidity in two megacities in China, Beijing and Chongqing. Spatially, $\mathrm{PM}_{2.5}$ was generally more acidic in Chongqing than in Beijing, but a reverse spatial pattern was found within the two cities, with more acidic $\mathrm{PM}_{2.5}$ at the urban site in Beijing whereas the rural site in Chongqing. Ionic compositions of $\mathrm{PM}_{2.5}$ revealed that it was the higher concentrations of $\mathrm{NO}_{3}^{-}$ at the urban site in Beijing and the lower concentrations of $C\mathrm{a}^{2+}$ within the rural site in Chongqing that made their $\mathrm{PM}_{2.5}$ more acidic. Temporally, $\operatorname{PM}_{2.5}$ was more acidic in summer and fall than in winter, while in the spring of 2006, the acidity of $\mathrm{PM}_{2.5}$ was higher in Beijing but lower in Chongqing than that in 2005. These were attributed to the more efficient formation of nitrate relative to sulfate as a result of the influence of Asian desert dust in 2006 in Beijing and the greater wet deposition of ammonium compared to sulfate and nitrate in 2005 in Chongqing. Furthermore, simultaneous increase of $\mathrm{PM}_{2.5}$ acidity was observed from spring to early summer of 2005 in both cities. This synoptic-scale evolution of $\operatorname{PM}_{2.5}$ acidity was accompanied by the changes in air masses origins, which were influenced by the movements of a subtropical high over the northwestern Pacific in early summer. Finally, the correlations between $[\mathrm{NO}_{3}^{-}]/[\mathrm{SO}_{4}^{2-}]$ and $[\mathrm{NH}_{4}^{+}]/[\mathrm{SO}_{4}^{2-}]$ suggests that under conditions of high aerosol acidity, heterogeneous reactions became one of the major pathways for the formation of nitrate at both cities. These findings provided new insights in our understanding of the spatial and temporal variations in aerosol acidity in Beijing and Chongqing, as well as those reported in other cities in China.
1 Introduction
Acidic aerosols can increase the risks to human health by direct inhalation and indirectly by activating hazardous particulate materials (Amdur and Chen, 1989; Health Effects Institute, 2002). Wet/dry deposition of acidic aerosols also lead to severe degradation of ecosystems (Larssen et al., 2006). Most acidic aerosols are hygroscopic, and as such act to reduce atmospheric visibility (Watson, 2002) as well as disturbing the radiative balance of the atmosphere (Boucher and Anderson, 1995; Crumeyrolle et al., 2008). They are also of great importance to atmospheric chemistry through their influence on many heterogeneous reactions and the behaviors of reactants and oxidants (Seinfeld and Pandis, 1998; Jang et al., 2002). Aerosol acidity can also affect the solubility of iron and phosphorus in the atmospheric aerosols (Meskhidze et al., 2005; Shi et al., 2011; Nenes et al., 2011), which has important implications for ocean biogeochemistry and global climate change (Jickells et al., 2005).
Acidic aerosol species in cities are usually dominated by sulfate $(\mathrm{SO}_{4}^{2-})$ and nitrate $(\mathrm{NO}_{3}^{-})$ , mostly converted from the precursors $S O_{2}$ and $\mathrm{NO_{x}}$ , respectively, and are partly or fully neutralized by ammonium $\left(\mathrm{NH}_{4}^{+}\right)$ and basic cations such as $\mathrm{{Ca}}^{2+}$ and ${\mathrm{Mg}}^{2+}$ . $\mathrm{Na^{+}}$ and $\mathrm{Cl^{-}}$ may also be important species influencing aerosol acidity in coastal area where sea salt plays a role. Aerosol acidity cannot be directly measured due to its low water content (Meng et al., 1995; Nenes et al., 1998), and is generally assessed using three different kinds of parameters, namely, strong acidity, ion-balanced acidity and in situ acidity.
Strong acidity, measured from the aqueous extracts of aerosol samples, represents the absolute acidity of the aerosols, but it cannot show any in situ characteristics due to the large excesses of water (Pathak et al., 2004). Ionbalanced acidity refers to the estimation of $\mathrm{H^{+}}$ concentration by subtracting the equivalent cations, other than $\mathrm{H^{+}}$ , from anions (Zhang et al., 2007a). It is more widely used in a relative way to indicate the neutralizing level with the equivalent ratio of cations/anions (Adams et al., 1999; Zhang et al., 2002, 2007a; Chu et al., 2004; Sun et al., 2010; Johansen et al., 1999; Takami et al., 2007; Chou et al., 2008). In situ aerosol acidity, in the form of the concentration of free $\mathrm{H^{+}}$ or pH in the deliquesced particles at the ambient condition, is most likely to influence the chemical behavior of aerosols. It can be estimated from a variety of thermodynamic models, such as E-AIM, SCAPE and GFEMN (Pathak et al., 2004, 2009; Yao et al., 2006; Takahama et al., 2006; Zhang et al., 2007a). However, it should be noted that ion-balanced and in situ aerosol acidity are empirical approaches that both depend on the choice of ion species. For example, due to their low abundance relative to ammonium in fine particles, basic $\mathrm{{Ca}}^{2+}$ and ${\mathrm{Mg}}^{2+}$ are usually ignored in the estimation of aerosol acidity, which might be less appropriate during dust events (Ziemba et al., 2007).
The characteristics of aerosol acidity may vary from region to region due to the spatiotemporal variability in the emission of primary aerosols and gaseous precursors, as well as regional differences in the climatic driving forces. The earliest observations on aerosol acidity in China were initiated in the 1980s in regions in the south and southwest (Huang et al., 1988; Shen et al., 1992; Zhao et al., 1994), but they generally focused on the acidification of fog and cloud in respect to severe acid rain, with most of the sites located in rural and remote areas. In recent years in China, there have been many field observations on aerosol acidity in the megacities of different regions, such as Beijing (Yao et al., 2002; Dillner et al., 2006; Sun et al., 2010), Shanghai (Yao et al., 2002; Xiu et al., 2005; Wang et al., 2006), Hong Kong (Pathak et al., 2003, 2004a, b) and Chongqing (Quan and Zhang, 2008; Aas et al., 2007).
While these studies suggest a general pattern of higher acidity in southern China than in the north, only a few of them presented parallel inter-region comparisons, with little information on seasonal variation. For example, Wang et al. (2006) and Pathak et al. (2009) investigated aerosol acidity at different regions in China, in spring and summer, respectively. Moreover, even for a specific region, there are large discrepancies between the studies. For Shanghai, Xiu et al. (2005) found that aerosols were almost completely neutralized, a finding that is contrary to the results reported by Yao et al. (2002) and Wang et al. (2006) (Chan and Yao, 2008). The discrepancies might be attributed to a variety of factors, including the procedures of sampling analysis and changes in emission strength and meteorological condition. Increasingly there is a need to understand how these factors have influenced the variability of aerosol acidity.
Along with the increase in domestic $\mathrm{NO_{x}}$ emissions in recent years (Zhang et al., 2007b), the concentration and proportion of $\mathrm{NO}_{3}^{-}$ in aerosols have been found to have increased significantly in most Chinese megacities (Richter et al., 2005; Chan and Yao, 2008; Shen et al., 2008), and become a major concern in the acidity of aerosols as well as their wet/dry deposition (Han et al., 2006; Larssen et al., 2006; Song et al., 2008). The formation pathway of $\mathrm{NO}_{3}^{-}$ depends on not only the availability of $\mathrm{NH}_{4}^{+}$ and meteorological condition (such as temperature), but also the characteristics of the preexisting particles, such as aerosol acidity, water content and alkaline mineral composition (Pakkanen et al., 1996; Hu and Abbatt, 1997; John et al., 1990; Zhuang et al., 1999). In particular, Pathak et al. (2009) recently reported high concentration of $\mathrm{NO}_{3}^{-}$ with high acidity and low $\mathrm{NH_{4}^{+}}$ at Beijing and Shanghai, with their formation being largely attributed to the hydrolysis of $\mathrm{N}_{2}\mathrm{O}_{5}$ on the preexisting particles. This differs from the findings of most previous field observations, which indicated that high concentration of $\mathrm{NO}_{3}^{-}$ are found in association with high $\mathrm{NH_{4}^{+}}$ (Pathak et al., 2009). However, the observations of Pathak et al. (2009) were limited to summertime.
In this study, the aerosol acidity at Beijing and Chongqing, two megacities in northern and southwestern China, respectively, was examined in parallel during a 15-month period of field observation, and the characteristics of fine particles $(\mathrm{PM}_{2.5}$ , particles of aerodynamic diameter $<\!2.5\,\upmu\mathrm{m})$ were investigated in detail. Based on the measurements of ionbalanced and in situ acidity, we investigated the spatial and seasonal patterns of $\operatorname{PM}_{2.5}$ acidity at these two cities. We also discussed the factors that determined these spatial and temporal variations.
2 Experimental method and model description
2.1 Sampling and analysis
Weekly $\mathrm{PM}_{2.5}$ samples were collected at both urban and rural sites of Beijing and Chongqing using a three-channel lowflow sampler (Aerosol Dynamics Inc., Berkeley, CA). Details of the sampling sites have been provided previously (He et al., 2001; Zhao et al. 2010). In brief, the urban and rural sites at Beijing were inside Tsinghua University (TH, $40^{\circ}19^{\prime}\,\mathrm{N}$ , $116^{\circ}19^{\prime}\,\mathrm{E})$ and near the Miyun Reservoir (MY, $40^{\circ}29^{\prime}\,\mathrm{N}$ , $116^{\circ}47^{\prime}\,\mathrm{E})$ , respectively, with a distance of $70\,\mathrm{km}$ between them. Three sites were selected at Chongqing, including a residential urban site in Jiangbei District (JB, $29^{\circ}34^{\prime}\,\mathrm{N}$ , $106^{\circ}32^{\prime}\,\mathrm{E})$ , an industrial site in Dadukou District (DDK, $29^{\circ}29^{\prime}\,\mathrm{N}$ , $106^{\circ}29^{\prime}\,\mathrm{E}$ ), and a rural site near the Jinyun Mountain in the Beibei District (BB, $29^{\circ}50^{\prime}\,\mathrm{N}.$ , $106^{\circ}25^{\prime}\,\mathrm{E})$ , which is $30\,\mathrm{km}$ and $40\,\mathrm{km}$ away from JB and DDK, respectively.
The sampling procedure was also given by He et al. (2001). Operating at a flow rate of $0.41\ensuremath{\mathrm{min}^{-1}}$ , one of the three channels collected $\mathrm{PM}_{2.5}$ on a Teflon filter with a Teflon impactor followed by a glass denuder. The glass denuder is coated with a $2\,\%$ carbonate solution prepared in 50:50 water:methanol to remove the acidic gases. $\mathrm{HNO}_{3}$ volatilized from the Teflon filter is collected on a nylon filter. Hence water soluble ions are determined from this Teflon filter but the reported particulate $\mathrm{NO}_{3}^{-}$ is the sum of $\mathrm{NO}_{3}^{-}$ on both the Teflon and nylon filters. The other two channels collected $\mathrm{PM}_{2.5}$ with a single Teflon filter and quartz filter for measuring elements and carbonaceous components, respectively, which were not used in this study. Each sample was collected continuously for a week. From 28 January 2005 to 5 May 2006, 106 and $180\ \mathrm{PM}_{2.5}$ samples were collected at Beijing and Chongqing, respectively.
Mass concentrations of $\mathrm{PM}_{2.5}$ were obtained by weighing on an analytical balance (Mettler Toledo AG285), after stabilizing under constant temperature $(20\pm5\,^{\circ}\mathrm{C})$ and humidity $(40\pm5\,\%)$ . Eight ions, including $\mathrm{SO}_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ , $\mathrm{Cl^{-}}$ , $\mathrm{NH_{4}^{+}}$ , $\ K^{+}$ , $C\mathrm{a}^{2+}$ , $\mathrm{Na^{+}}$ and ${\mathrm{Mg}}^{2+}$ , were measured by ion chromatography (Dionex 600, details in Wang et al., 2005).
Hourly meteorological data for both Beijing and Chongqing were obtained from the website http://www. wunderground.com, including temperature, dewpoint, wind speed, visibility and precipitation. The spatial distribution of geopotential height was derived from the archived meteorological data of NOAA’s Air Resources Laboratory (ARL, http://ready.arl.noaa.gov/).
2.2 Indicators of $\mathbf{PM}_{2.5}$ acidity
2.2.1 Ratio of cation/anion
In this study, the equivalent charge ratio $(\mathrm{eq/eq})$ of the major cations $\mathrm{NH}_{4}^{+}$ and $\mathrm{{Ca}}^{2+}$ to anions $\mathrm{SO}_{4}^{2-}$ and $\mathrm{NO}_{3}^{-}$ was used to indicate the neutralizing level of $\mathrm{PM}_{2.5}$ , as the other ions generally had little influence on the acidity at Beijing and Chongqing (to be discussed in Sect. 3.1). The equivalent charge ratio was defined as following (Adams et al., 1999; Zhang et al., 2002):
$$
\mathrm{R_{C/A}}\mathrm{=}\frac{\mathrm{NH_{4}^{+}\mathrm{+}C a^{2+}}}{\mathrm{SO_{4}^{2-}\mathrm{+}N O_{3}^{-}}}
$$
where all the species denote the concentrations of their equivalent charges (likewise for all the ratios of species without brackets in the following text). In this equation, $\mathrm{R_{C/A}}\geq1$ indicates that most of the acids can be neutralized, while $R<1$ indicates the aerosol is acidic.
2.2.2 In situ aerosol acidity
Both free $\mathrm{H^{+}}$ concentration ( $[\mathrm{H^{+}}]_{\mathrm{Ins}}$ , the square brackets indicate the molar concentration of the species inside, used here and henceforth) and in situ $\mathsf{p H}$ in the deliquesced particles were used as indicators of aerosol acidity, which can be estimated from a chemical thermodynamic model (EAIM2, http://www.aim.env.uea.ac.uk/aim/). E-AIM2 is a state-of-the-art model that can accurately simulate the liquid and solid phase of ionic compositions in the mixing system $\mathrm{H^{+}{-N H_{4}^{+}{-S O_{4}^{2-}{-N O_{3}^{-}{-H_{2}O}}}}}$ at a given temperature and relative humidity (Clegg et al., 1998). The model input of E-AIM2 includes weekly averaged temperature, relative humidity, $[\mathrm{SO}_{4}^{2-}]$ , $[\mathrm{NO}_{3}^{-}]$ , $[\mathrm{NH}_{4}^{+}]$ and total $\mathrm{H^{+}}$ $(\mathrm{[H^{+}]_{T o t a l}})$ , which is estimated from the ionic balance of the relevant species (Yao et al., 2006; Zhang et al., 2007; Pathak et al., 2009):
$$
[\mathrm{H}^{+}]_{\mathrm{Total}}\,{=}\,2\times[\mathrm{SO}_{4}^{2-}]\,{+}[\mathrm{NO}_{3}^{-}]\,{-}[\mathrm{NH}_{4}^{+}]
$$
The aerosol $\mathrm{pH}$ was calculated as:
$$
\mathrm{pH}\,{=}\,{-}\mathrm{log}(\gamma\times[\mathrm{H^{+}}]_{\mathrm{Frac}})
$$
where $\gamma$ is the activity coefficient on mole fraction basis and $[\mathrm{H^{+}}]_{\mathrm{Frac}}$ is the molar fractions of aqueous phase $\mathrm{H^{+}}$ (Zhang et al., 2007a). In addition to these two parameters, $[\mathrm{H^{+}}]_{\mathrm{Ins}}$ and the concentration of water content $\langle\mathrm{H}_{2}\mathrm{O}\mathrm{I}\rangle$ were derived from E-AIM2.
The lack of information about the organic acids generally has little influence on the estimation of aerosol acidity due to their low abundance in aerosols (Zhang et al., $2007\mathrm{a}$ ; Pathak et al., 2009). However, larger bias may exist because of a lack of information about basic $C\mathrm{a}^{2+}$ and ${\mathrm{Mg}}^{2+}$ , especially in samples containing high concentrations of mineral dust (Ziemba et al., 2007). Although there are models, such as SCAPE, that take into account a system with these basic ions included, they cannot be used in the current study because the required gaseous $\mathrm{HNO}_{3}$ and $\mathrm{NH}_{3}$ , the input parameters for the models, were not measured here. It should also be noted that the E-AIM2 model only simulates the average results over the whole week without considering the influence from temporal variations in aerosol composition, temperature and relative humidity (Yao et al., 2006).
2.3 Trajectory computation and clustering
Backward trajectories of air masses arriving at Beijing and Chongqing were calculated using the HYSPLIT model (Version 4.8) to investigate the influence of different air masses on aerosol composition and acidity. The meteorological data fields used to run the model are 6-hourly FNL archived data, which are available at NOAA’s ARL archives. For single trajectory calculation, the model was run 4 times per day (UTC 00:00, 06:00, 12:00 and 18:00) with the arrival level at $500\,\mathrm{m}$ (below the boundary layer) or $2000\,\mathrm{m}$ (above the boundary layer).
As the typical errors of individual trajectories were estimated to be $20\,\%$ of the traveled distance (Stohl, 1998), the trajectories over the whole sampling period were classified into seasonal transport patterns using the HYSPLIT model. A detailed procedure of the clustering analysis is provided in the supplementary material according to the model description (Draxler et al., 2006). The percentage change in total spatial variance (TSV) was used to determine what is the reasonable number of clusters in each season: a large increase in TSV indicates that different clusters are being paired and therefore that the cluster process should stop.
3 Results and discussions
3.1 Abundance of ionic species in $\mathbf{PM}_{2.5}$
The annual concentrations of $\mathrm{PM}_{2.5}$ and ionic species were averaged from March 2005 to February 2006 for Beijing and Chongqing, as shown in Table 1. $\mathrm{PM}_{2.5}$ mass concentration was similar at all the three sites in Chongqing $(\sim\!130\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ , all of which were higher than those at Beijing $(118\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ for TH and $68\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ for MY). This indicates high regional background levels of $\mathrm{PM}_{2.5}$ in the surrounding area in Chongqing. As with $\mathrm{PM}_{2.5}$ mass, higher concentrations of total ionic species were found in Chongqing $(41.3–45.0\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ ) than in Beijing $(28.3-$ $39.1\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ . The proportion of ionic species in $\mathrm{PM}_{2.5}$ at the urban site in Beijing $(33.0\,\%)$ was close to those at Chongqing $(31.8{-}34.9\,\%)$ , but the higher fraction in MY $(41.4\,\%)$ suggests a more important role of ionic species in rural areas of Beijing. Components other than ionic species contributed similar amounts to $\mathrm{PM}_{2.5}$ at Beijing and Chongqing, i.e. carbonaceous species and crustal dust accounted for $36{-}40\,\%$ and $6{-}8\,\%$ of $\mathrm{PM}_{2.5}$ mass, respectively (Zhao et al., 2010; He et al., 2011).
$\mathrm{SO}_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ and $\mathrm{NH_{4}^{+}}$ dominated the ionic species, with a contribution of up to $85{-}90\,\%$ at both Beijing and Chongqing. As with $\mathrm{PM}_{2.5}$ at Chongqing, the three species when considered in combination had a small spatial variation, while at Beijing their relatively small difference between MY and TH (MY/TH: $76.8\,\%$ ) compared to $\mathrm{PM}_{2.5}$ (MY/TH: $57.7\,\%$ ) indicates that they are of greater regional significance than the other aerosol species. Concentrations of $\bar{\mathrm{SO}_{4}^{2-}}$ for all the sites were higher at Chongqing than at Beijing, whereas $\mathrm{NO}_{3}^{-}$ showed the opposite spatial pattern with higher concentrations at Beijing than at Chongqing. This is mainly attributed to regional differences in energy structure and meteorology (see Sect. 3.2). At the mean time, it should be noted that a higher proportion of $\mathrm{SO}_{4}^{2-}$ in $\mathrm{PM}_{2.5}$ was found for MY (0.19) than TH (0.13), but similar as that for the sites in Chongqing (0.18–0.20). This was probably due to the more homogeneous spatial distribution of sulfate in Beijing than the other aerosol species, as has been found by other studies (Zhao et al., 2009; Guo et al., 2010). It also indicates that the regional influence of sulfate in Beijing was as important as in Chongqing.
Compared to the above major ionic species, $C\mathrm{a}^{2+}$ , ${\mathrm{Mg}}^{2+}$ , $\mathrm{Na^{+}}$ , $\mathbf{K}^{+}$ and $\mathrm{Cl^{-}}$ constituted a minor fraction $(10{-}15\,\%)$ of ionic species at Beijing and Chongqing. The annual concentrations of basic $\mathrm{{Ca}}^{2+}$ and ${\mathrm{Mg}}^{2+}$ were higher at Chongqing (totaling $0.9–1.5\,\upmu\mathrm{g}\,\mathrm{m}^{-3},$ ) than at Beijing (totaling $0.7–1.1\,\upmu\mathrm{g}\,\mathrm{m}^{-3},$ ). Their low abundance relative to $\mathrm{NH_{4}^{+}}$ suggests that they have only a weak influence on neutralizing the acidic species, as found in most studies on aerosol acidity (Yao et al., 2006; Zhang et al., 2007a; Pathak et al., 2009). However, as an indicator of mineral dust, which was found in high concentrations during the spring and winter at Beijing and Chongqing (Zhao et al., 2010), $\mathrm{{Ca}}^{2+}$ was included in the acidity analysis to get an idea of the regional influence of alkaline dust.
$\mathrm{Cl^{-}}$ was relatively abundant at all the sites (0.4– $2.2\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ , and primarily comes from coal combustion (Yao et al., 2002) and biomass burning (Li et al., 2007, 2009). Similarly, $\mathrm{Na^{+}}$ $(0.3–0.7\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ and $\Chi^{+}$ $(1.4{-}3.1\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ mainly come from coal combustion and biomass burning, respectively, as revealed by our previous study (Zhao et al., 2010). These three ions were excluded in the following acidity analysis due to their neglectable effect on aerosol acidity.
3.2 Spatiotemporal variations in cation/anion ratio
3.2.1 Spatial distribution
The ratio of cation/anion $(\mathrm{R_{C/A}})$ in $\mathrm{PM}_{2.5}$ was calculated for all the sites according to Eq. (1). As shown in Table 1, annual $\mathrm{R_{C/A}}$ from March 2005 to February 2006 were 0.97 and 1.04 at TH and MY, respectively, whereas those in Chongqing were substantially lower, ranging from 0.76 to 0.86, respectively. This indicates that the aerosols were much more acidic at Chongqing than at Beijing. This pattern is consistent with the findings of Wang et al. (2006), who reported that aerosols over southern China were less neutralized during springtime than aerosols over northern China.
The regional distribution of $\operatorname{PM}_{2.5}$ acidity was mainly caused by the regional differences in the abundance of $\mathrm{NH_{4}^{+}}$ relative to $\mathrm{NO}_{3}^{-}$ and $\mathrm{SO}_{4}^{2-}$ , as revealed from the low ratio of $\mathrm{Ca}^{2+}/\mathrm{NH}_{4}^{+}$ at all sites (0.06–0.08, equivalent charge ratio);
however, $\mathrm{NO}_{3}^{-}$ and $\mathrm{SO}_{4}^{2-}$ are identified to play different roles in contributing to the acidity of the two cities. As shown in Table 1, $\mathrm{N}\bar{\mathrm{O}}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ in $\mathrm{PM}_{2.5}$ in Beijing (0.47–0.60) was ${\sim}3{-}4$ times higher than that in Chongqing (0.15–0.17). This is consistent with the estimated higher contribution from vehicular emissions to $\mathrm{PM}_{2.5}$ in Beijing (Arimoto et al., 1996; Yao et al., 2002). This also agrees with the larger vehicle populations in Beijing (2.1 million) than Chongqing (0.47 million) in 2005, where coal consumption was both close to 30 million tons (Beijing Statistic Bureau, 2006; Chongqing Statistic Bureau, 2006). Although $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ may also be influenced by sampling artifacts due to a long sampling duration (Pathak et al., 2004b), the charge ratios were quite close to those previously reported using daily samples for Beijing (0.52 in 2001–2003; Wang et al., 2005) and Chongqing (0.16 in 2003; Aas et al., 2007).
The urban-rural distributions of $\mathrm{R_{C/A}}$ for $\mathrm{PM}_{2.5}$ in Beijing and Chongqing were opposite to each other, which can be explained by their compositional difference in $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ and $\mathrm{Ca}^{2+}/\mathrm{NH}_{4}^{+}$ . Both of these two ratios were higher at urban sites, possibly due to more vehicle sources and construction activities, however they showed different spatial gradients within the two cities. Compared to the difference of $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ at Chongqing $(<\!10\,\%$ ; urban: $0.16-$ 0.17, rural: 0.15), the ratio was found ${\sim}30\,\%$ higher at urban TH (0.60) than rural MY (0.47) at Beijing. At the meantime, $\mathrm{Ca}^{2+}/\mathrm{NH}_{4}^{+}$ showed a larger urban-rural difference at Chongqing $(\sim\!25\,\%)$ than at Beijing $(\sim\!10\,\%)$ . Therefore, it was the higher concentrations of $\mathrm{NO}_{3}^{-}$ within the urban site in Beijing and the lower concentrations of $C\mathrm{a}^{2+}$ within the rural site in Chongqing that made their $\mathrm{PM}_{2.5}$ more acidic. This differs from findings for Pittsburgh where $\mathrm{NH}_{4}^{+}$ levels determined the spatial distribution of aerosol acidity at urban and semi-rural sites (Liu et al., 1996).
3.2.2 Seasonal variation
The seasonal averages of $\mathrm{R_{C/A}}$ of $\mathrm{PM}_{2.5}$ in Beijing and Chongqing are shown in Table 2. In Beijing, $\mathrm{PM}_{2.5}$ at TH was more acidic in the summer of 2005 and spring of 2006 than other seasons $(\mathrm{R_{C/A}}\leq1$ for all seasons). Similar seasonal variation in $\mathrm{R_{C/A}}$ was also observed at rural MY, but its higher $\mathrm{R_{C/A}}$ indicates that $\mathrm{PM}_{2.5}$ was almost neutral in all seasons $(\mathrm{R_{C/A}}\geq1)$ , except for the spring of 2006 $(\mathrm{R_{C/A}}\,{=}\,0.80)$ . These results are similar to previous findings that aerosols were more acidic in warm seasons than in cold seasons at Beijing (Wang et al., 2000; Pathak et al., 2009), as indicated from $\mathrm{R_{C/A}}$ reproduced from their reported datasets shown in Table 2. However, no consistent pattern in interannual trends can be discerned for each season. During the summer, $\mathrm{R_{C/A}}$ was ${>}1$ in 2001–2003 (Wang et al., 2005) and 2006 (Sun et al., 2010), and ${<}1$ in 1999–2000 (Wang et al., 2000; He et al., 2001) and 2005 (Pathak et al., 2009; this study); for spring, $\mathrm{R_{C/A}}$ was ${>}1$ in 1994–1995 (Wang et al., 2000) and 2001–2003 (Wang et al., 2005), and ${<}1$ in 1999– 2000 (He et al., 2001; Dillner et al., 2006) and 2005–2006 (this study); for winter, $\mathrm{R_{C/A}}$ was ${>}1$ in most years except 1999–2000 when it was only 0.63 (He et al., 2001), indicating the aerosols were much more acidic during spring than during other seasons. Similar characteristics in seasonal variation of aerosols are also observed in other northern cities.
As shown in Table 2, while they were neutralized in most cases $(\mathrm{R_{C/A}}>1)$ ), very acidic aerosols were also observed at Yungang (summer of 1988; Wang et al., 2000), Xi’an (winter of 1996–1997; Zhang et al., 2002) and Jinan (spring of 2004–2005; Yang et al., 2007).
As in Beijing, $\mathrm{R_{C/A}}$ was high in winter and low in summer and fall of 2005 at both urban and rural sites of Chongqing. However, the ratios for the two springs were not the same, with much more acidic aerosols being observed in 2005 $(\mathrm{R}_{\mathrm{C/A}}=0.58–0.62)$ than in 2006 $(\mathrm{R}_{\mathrm{C/A}}=0.94{-}1.00)$ ). No consistent inter-annual trend in seasonal acidity could be found at Chongqing, either. For example, Liu et al. (1988) found $\mathrm{R_{C/A}}$ of $\mathrm{PM}_{2}$ to be 1.77 in the fall of 1980s, much higher than the ratios from our observation (0.79–1.02). The large difference between the two studies cannot be simply explained by the increased acidity of aerosol over the past 20 yr because similarly low $\mathrm{R_{C/A}}$ (0.78–1.01) were also measured in TSP by Zhao et al. (1994) during 1987–1988 in Chongqing (the ratio was even lower for fine particles). This phenomenon shifts the likely explanation to factors other than the changes in the emission of acidic aerosols and their precursors. As at Chongqing, aerosol $\mathrm{R_{C/A}}$ at other southern cities have also exhibited an inconsistent inter-annual trend in recent years. The $\mathrm{R_{C/A}}$ of $\mathrm{PM}_{2.5}$ in Shanghai was significantly higher in the summer and winter of 2004 (1.38 and 0.78, respectively; Wang et al., 2006) than in summer of 2005 and winter of 2001 (0.4–0.5; Xiu et al., 2004; Pathak et al., 2009), while for Hong Kong the ratio in winter was also distinctively higher in 2002 (1.06; Cheung et al., 2005) than 2001 (0.73; Louie et al., 2005).
In spite of the uncertainties in using $\mathrm{R_{C/A}}$ to compare the aerosol acidity between different studies, such as the variable sampling methods, the representativeness of sampling periods, and analytical procedures, the above findings collectively suggest that the seasonal variation of aerosol acidity in northern and southern China may be influenced by a variety of factors (emission strength, meteorological condition and characteristics of preexisting particles and precursors, etc.). For either long-term or short-term field observations, it is consequently risky to attribute the inter-annual changes of aerosol acidity to any of these factors alone. For example, based on a comparison with the acidic aerosols reported by Dillner et al. (2006) for spring 2001, Sun et al. (2010) simply attributed their fully neutralized aerosols in Beijing during the summer of 2006 to the reduced $\mathrm{SO}_{2}$ emissions or increased $\mathrm{NH}_{3}$ emissions in this region.
It is interesting to observe that the inter-annual variation in spring $\mathrm{PM}_{2.5}$ acidity for Chongqing was opposite to that for Beijing. For both cities, their covariation at urban and rural sites indicates that the inter-annual trend was of regional scale. Thus, the weekly $\mathrm{R_{C/A}}$ for each city was averaged to investigate the short-period variation in aerosol acidity within each season.
As shown in Fig. 1a, $\mathrm{R_{C/A}}$ showed extensive weekly fluctuation for both cities, with larger variation at Beijing (0.39– 1.60) than at Chongqing (0.51–1.13). However, $\mathrm{R_{C/A}}$ at Beijing was higher in the spring of 2005 $(>\!1)$ than in 2006 ( $^{<1}$ , from the week of 24–31 March), while at Chongqing continuously higher $\mathrm{R_{C/A}}$ was observed in the spring of 2006 $\left(0.8-\right.$ 1.1) than in 2005 (0.6–0.8). This pattern is better presented as normalized $\mathrm{R_{C/A}}$ $\mathrm{(R_{C/A}}$ minus the averaged ratio during the whole sampling period, as shown in Fig. 1b), indicating a weak intra-seasonal variation of aerosol acidity for both Beijing and Chongqing during the two springs. Also of note is that $\mathrm{R_{C/A}}$ at both Beijing and Chongqing showed a similar decreasing trend during February–June 2005 with a sharp increase at the end, which implies that the aerosol acidity of both cities had been influenced by large-scale driving forces, as discussed in Sect. 4.
3.3 Seasonal variation of in situ aerosol acidity
In situ aerosol $\mathrm{pH}$ , $[\mathrm{H^{+}}]_{\mathrm{Ins}}$ and $[\mathrm{H}_{2}\mathrm{O}]$ of $\mathrm{PM}_{2.5}$ at Beijing and Chongqing were shown in Fig. 2. $[\mathrm{NH}_{4}^{+}]$ , $[\mathrm{SO}_{4}^{2-}]$ and $[\mathrm{NO}_{3}^{-}]$ were averaged for urban and rural sites and used as the input data to simplify the comparison between the two cities.
The in situ $\operatorname{PM}_{2.5}$ acidity showed similar seasonal variation as previously indicated by $\mathrm{R_{C/A}}$ , but gave additional insight into the hygroscopic properties of aerosols. As shown in Fig. 2a, it was only in summer and fall of 2005 and spring of 2006 that deliquescent aerosols were found to be abundant at Beijing with free $\mathrm{H^{+}}$ , while most of them remained in solid phase during the spring and winter of 2005. The spring of 2006 at Beijing had the most acidic aerosols, with an in situ $\mathrm{pH}$ of only $-0.618$ to 0.404, while there were only two weeks in the spring of 2005 when $\mathrm{PM}_{2.5}$ was found acidic. Moreover, although high $[\mathrm{H^{+}}]_{\mathrm{Ins}}$ existed in $\mathrm{PM}_{2.5}$ during the summer of 2005 (average: $0.030\,\upmu\mathrm{mol}\,\mathrm{m}^{-3}$ , range: $0.002{-}0.126\,\mathrm{\mumol\,m}^{-3})$ and spring of 2006 (average: $0.034\,\upmu\mathrm{mol}\,\mathrm{m}^{-3}$ , range: $0.006{-}0.122\,\upmu\mathrm{mol}\,\mathrm{m}^{-3}\rangle$ , the former was less acidic because of its much higher $\left[\mathrm{H}_{2}\mathrm{O}\right]$ (average: $3.074\,\upmu\mathrm{mol}\,\mathrm{m}^{-3}$ , range: $0.100{-}8.466\,\upmu\mathrm{mol}\,\mathrm{m}^{-3})$ ) than the latter (average: $0.186\,\upmu\,\mathrm{mol}\,\mathrm{m}^{-3}$ , range: 0.034– $0.468\,\upmu\mathrm{mol}\,\mathrm{m}^{-3}.$ ).
Contrasting with Beijing, $\mathrm{PM}_{2.5}$ at Chongqing was deliquescent throughout the year with high $[\mathrm{H^{+}}]_{\mathrm{Ins}}$ and $\mathrm{[H}_{2}\mathrm{O]}$ , while a significant variation of in situ acidity between the two springs was also clearly evident, as shown in Fig. 2b. The most acidic aerosols were found during February–June 2005, when in situ $\mathrm{pH}$ remained at its lowest level (0.52– 1.38) due to a faster increase in $[\mathrm{H^{+}}]_{\mathrm{Ins}}$ (20 times) than in $[\mathrm{H}_{2}\mathrm{O}]$ (5 times). Interestingly, both parameters decreased simultaneously decreased to their lowest level of the whole observation period in the week of 25 June to 1 July 2005, resulting in a significant increase of in situ $\mathrm{pH}$ and thus much less acidic $\operatorname{PM}_{2.5}$ . This was also coincident with the week when $\mathrm{R_{C/A}}$ showed a remarkable increase (Fig. 1).
In contrast to results revealed by $\mathrm{R_{C/A}}$ , a noteworthy finding for the variation of in situ $\mathrm{pH}$ is that $\operatorname{PM}_{2.5}$ was more acidic at Beijing than at Chongqing during the springs. This was mainly due to the drier climatology and lower water content in aerosols at Beijing that favored high in situ acidity, even though there might be less free $\mathrm{H^{+}}$ . Similar finding was also reported in Hong Kong where variation of in situ $\mathrm{PM}_{2.5}$ acidity was a function of relative humidity (RH) and even the more neutralized particles could have a high acidity under the influence of dry air masses from the Chinese mainland (Pathak et al., 2004a). This result highlighted the importance of the in situ acidity relative to other parameters (Pathak et al., 2004b).
In addition, as one of the most important parameters determining the in situ acidity, RH clearly exhibited opposite trends from winter into spring at Chongqing during 2005 and 2006, which can partly explain the inter-annual variation of $\mathrm{PM}_{2.5}$ acidity. As shown in Fig. 2b, for the period from February to May, it increased from ${\sim}60$ to ${\sim}80\,\%$ in 2005, but decreased from ${\sim}80$ to ${\sim}50\,\%$ in 2006. These long-playing reverse seasonal trends were likely to have been influenced by large-scale synoptic system anomalies.
3.4 Factors influencing the spring-summer variation of $\mathbf{PM}_{2.5}$ acidity
As a case study, we examined in the following the covariation of $\mathrm{PM}_{2.5}$ acidity from spring into summer 2005 for Beijing and Chongqing, as well as the opposite inter-annual variation in $\mathrm{PM}_{2.5}$ acidity in these cities during the springs of 2005 and 2006.
3.4.1 Asian summer monsoon
The covariation of $\mathrm{R_{C/A}}$ for $\mathrm{PM}_{2.5}$ at Beijing and Chongqing from February to June 2005, with a sharp increase at the end of June, indicates a synoptic-scale influence (Roger and Andrew, 2002).
The anomaly of the Asian summer monsoon in June 2005 was the abnormal behavior of the subtropical high over the Northwestern Pacific (Pacific High, in short) and the trough/ridge systems over mid- and high latitudes (Lu et al., 2007; Mu et al., 2008). The northward movement of the Pacific High, which is one of the most important parameters indicating the evolution of spring into summer in East Asia, was delayed until the end of June 2005. As shown in Fig. 3a (Lu et al., 2007), the ridge of the Pacific High remained around $13{-}16^{\circ}\,\mathrm{N}$ before 26 June 2005, $3{-}5^{\circ}$ to the south of the normal position. During 26–28 June, however, it suddenly moved from 17 to $28^{\circ}\,\mathrm{N}$ at a speed of $3{-}4^{\circ}$ per day, and its representative positions before and after the movement are indicated by the geopotential heights of $500\,\mathrm{hPa}$ in Fig. 3b (23 June) and Fig. 3c (29 June), respectively.
The evolution of air mass sources at Beijing and Chongqing was investigated before and after the northward movement of the Pacific High. The period from 4 March to 27 June 2005 was divided into three phases, including 4 March–6 May (to be compared with the same period in the spring of 2006), 7–31 May and 1–27 June. For each phase, backward trajectories of air masses at $500\,\mathrm{m}$ above ground level of Beijing and Chongqing were classified according to the procedures in Sect. 2.3. Trajectories of longer duration were calculated for Chongqing $(120\,\mathrm{h})$ than for Beijing $(72\,\mathrm{h})$ because of their different spread of travel.
The increasing aerosol acidity from March to June 2005 in the two cities was closely associated with the contribution of air masses from areas between the Northern China Plain to the south of Beijing and from central China to the east of Chongqing. As shown in Fig. 4, Beijing was gradually dominated by air masses originating from south of the city, which increased from $13\,\%$ in March–April (cluster 5 in Fig. 4a) to $41\,\%$ in May (clusters 4 and 5 in Fig. 4c) and $59\,\%$ in June (clusters 3 and 4 in Fig. 4d). At the same time, Chongqing was dominated by air masses from the east of the city, which increased from $50\,\%$ in March–April (cluster 4 in Fig. 5a) to $65\,\%$ in May (clusters 1 and 3 in Fig. 5c) and $73\,\%$ in June (clusters 1, 4 and 5 in Fig. 5c). As indicated by the aerosol optical depth (AOD) shown in the Supplement Fig. S1a–c, these source regions were found to have high aerosol loading from March to June in 2005, which clearly suggested that the aerosol acidity was increasing over a broad region of mainland China, with a stronger influence in the south than in the north.
Along with the northward movement of the Pacific High at the end of June 2005, the acidic aerosols over Beijing and Chongqing were replaced by the cleaner air from the northwest and southeast, respectively, which coincided with the simultaneous decrease of $\mathrm{PM}_{2.5}$ acidity at both cities. This effect was also evident from significantly weakened AOD during 29–30 June over a wide region that used to be covered by highly acidic aerosols (Supplement Fig. S1d). In July, Beijing was again dominated by air masses from the south with a monthly contribution $65\,\%$ of all air masses) and $[\mathrm{H^{+}}]_{\mathrm{Ins}}$ level comparable to those in June. However, the Asian summer monsoon was found to have a much greater significance in Chongqing, where the air masses in July were dominated by those having been transported over long distances from the south of China and from southeastern Asia, with high monthly contribution ( $74\,\%$ of all air masses), but low aerosol acidity.
These lines of evidence collectively suggest the major role of the Asian summer monsoon in determining the regional evolution of $\operatorname{PM}_{2.5}$ acidity from the spring to the summer of 2005 for Beijing and Chongqing. However, it can not explain the inter-annual variation of $\mathrm{PM}_{2.5}$ acidity during the springs of 2005 and 2006, since no obvious difference was found between the transport patterns of their air mass trajectories. As shown in Figs. 4b and 5b, respectively, the dominant air masses for Beijing and Chongqing from 3 March to 5 May 2006 were also from the northwest and east of China, a situation that was similar to that from 4 March to 6 May 2005 (Figs. 4a and 5a, respectively). Therefore, there must be other factors that caused the inter-annual variation of $\mathrm{PM}_{2.5}$ acidity in the springs of Beijing and Chongqing.
3.4.2 Asian desert dust
Mineral dust can affect aerosol acidity by either directly neutralizing the acidic aerosol or increasing the surface area of heterogeneous reaction for the acids. As an indicator of mineral dust, higher $\mathrm{{Ca}}^{2+}$ as well as higher ratios of $\mathrm{Ca}^{2+}/\mathrm{NH}_{4}^{+}$ were observed in spring of 2005 and 2006 for Chongqing and Beijing, respectively. This is consistent with the findings of Wu et al. (2009) who reported that emission of Asian desert dust was more active in the spring of 2006 than in the spring of 2005 for Beijing, and our related study (Zhao et al., 2010) found Asian dust to be more active in the spring of 2005 than of 2006 for Chongqing.
In order to compare the $\mathrm{R_{C/A}}$ and ionic species of the two springs in parallel, data of weekly samples from 4 March to 6 May 2005 and 3 March to 5 May 2006 were averaged to represent the spring of 2005 and 2006 for each site, respectively. As shown in Fig. 6a, the $C\mathrm{a}^{2+}$ concentration at Beijing was 23.8 to $30.6\,\%$ higher in the spring of 2006 than of 2005, while the concentrations of $\mathrm{SO}_{4}^{2-}$ and $\mathrm{NO}_{3}^{-}$ also increased by 11.3 to $23.3\,\%$ and 1.9 to $8.8\,\%$ , respectively, with little variation for $\mathrm{NH}_{4}^{+}$ . Due to the small contribution of neutralization from mineral components, the increased $\mathrm{SO}_{4}^{2-}$ and $\mathrm{NO}_{3}^{-}$ in the spring of 2006 remained acidic. Meanwhile, a significant increase $(45.5\,\%)$ was also found for $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ at urban TH in the spring of 2006 compared to 2005, which strongly suggests the influence from enhanced production of $\mathrm{NO}_{3}^{-}$ on the surface of mineral particles. It is well known that the reactions with alkaline mineral components are of several magnitudes faster for gaseous $\mathrm{HNO}_{3}$ than for $\mathrm{NO}_{2}$ and $\mathrm{SO}_{2}$ (Ooki and Uematsu, 2005; Vlasenko et al., 2006), which were all abundant in the atmosphere of Beijing (Bergin et al., 2001). When Asian dust was transported to Beijing, $\mathrm{CaCO_{3}}$ could react with $\mathrm{HNO}_{3}$ to form $\mathrm{Ca}(\mathrm{NO}_{3})_{2}$ , providing more hygroscopic surfaces for the heterogeneous reaction with $S O_{2}$ and $\mathrm{NO}_{2}$ , as has been directly observed by single particle analysis during dust storms at Beijing (Li and Shao, 2009). However, $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ showed little increase $(\sim\!1.5\,\%)$ at MY in the spring of 2006 compared to 2005, as seen in Fig. 6a. This is perhaps due to the lack of precursors of $\mathrm{NO}_{3}^{-}$ , which was more concentrated in urban area of Beijing, and the unstable nature of $\mathrm{NH}_{4}\mathrm{NO}_{3}$ , which could easily be decomposed into gaseous $\mathrm{NH}_{3}$ and $\mathrm{HNO}_{3}$ during transport from the urban area to MY.
On the other hand, compared to that in the spring of 2005 a higher increase in $\mathrm{SO}_{4}^{2-}$ was observed at MY $(23.3\,\%)$ than at TH $(11.3\,\%)$ in the spring of 2006. The elevated $\mathrm{SO}_{4}^{2-}$ concentration at the rural MY than at the urban TH (a pattern not observed for $\mathrm{NO}_{3}^{-}$ or $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ ) can be explained by coupling the $\mathrm{SO}_{4}^{2-}$ formation with the inter-annual variation in transport pathways of air masses during spring. Compared to the transport pattern in the spring of 2005 (Fig. 4a), polluted air masses were more frequently transported from the west and south of Beijing in the spring of 2006 (Fig. 4b), which favored a higher regional contribution of $\mathrm{SO}_{4}^{2-}$ at MY than during periods when other transport pathways were in play (Jia et al., 2008; Zhao et al., 2009). Moreover, faster transformation of local $\mathrm{SO}_{2}$ to $\mathrm{SO}_{4}^{2-}$ at Beijing could also lead to higher increase of $\mathrm{SO}_{4}^{2-}$ in MY due to the more acidic and hygroscopic aerosols in the southern and southwestern air masses. However, it should be noted that the difference in $\mathrm{NH_{4}^{+}}$ concentrations between the two springs was nearly the same for the two sites at Beijing, perhaps because of the recapture of decomposed $\mathrm{NH}_{3}$ from $\mathrm{NH}_{4}\mathrm{NO}_{3}$ by the unneutralized $\mathrm{SO}_{4}^{2-}$ or $\mathrm{HSO_{4}^{-}}$ during transport.
Although the influence of Asian dust at Beijing may partly explain the inter-annual variation of $\mathrm{PM}_{2.5}$ acidity for the springs of 2005 and 2006, this does not seem to be an explanatory factor at Chongqing. Firstly, it was a significant decrease in $\mathrm{NH_{4}^{+}}$ concentration in the spring of 2005, which was 28.0 to $30.2\,\%$ lower than that for the spring of 2006 (Fig. 6b), that essentially led to the elevation of aerosol acidity. The increased concentration of mineral dust might have changed the gas–particle equilibrium of $\mathrm{NH}_{3}/\mathrm{NH}_{4}^{+}$ by limiting the transfer of $\mathrm{NH}_{3}$ to $\mathrm{NH}_{4}^{+}$ in fully neutralized aerosols (Luo et al., 2007), but it could hardly influence the highly acidic aerosols at Chongqing during the spring of 2005. Secondly, significant monthly variation in the transport of northwestern air masses was found from March through June 2005 at either the boundary layer (Fig. 5) or the higher atmosphere (Supplement Fig. S2), but the $\operatorname{PM}_{2.5}$ acidity at Chongqing (as indicated by $\mathrm{R_{C/A}}$ in Fig. 1 and $[\mathrm{H^{+}}]_{\mathrm{Ins}}$ in Fig. 2) remained at a consistently high level, and indeed showed a slight increase.
3.4.3 Wet deposition of $\mathbf{NH}_{4}^{+}$ in Chongqing
Particulate $\mathrm{NH_{4}^{+}}$ mainly comes from the gaseous $\mathrm{NH}_{3}$ and has a residence time of 4–6 days compared to only 1 day for gaseous $\mathrm{NH}_{3}$ (Adams et al., 1999). The variation in $\mathrm{NH_{4}^{+}}$ concentration at Chongqing for the two springs can be influenced by many factors, including the emission strength of precursor $\mathrm{NH}_{3}$ , the gas-particle equilibrium of $\mathrm{NH}_{3}/\mathrm{NH}_{4}^{+}$ , and patterns of atmospheric transport, diffusion and deposition (Asman et al., 1998).
$\mathrm{NH}_{3}$ emissions from natural sources, including animal waste, natural and fertilized soils, and vegetation, usually depend on temperature, which showed little difference between the two springs in Chongqing, as shown in Fig. 6b. Anthropogenic sources, such as industrial process, are considered to be stable during all seasons. The gas-particle equilibrium of $\mathrm{NH}_{3}/\mathrm{NH}_{4}^{+}$ is usually related to $\mathrm{NH}_{4}\mathrm{NO}_{3}$ and $\mathrm{NH_{4}C l}$ , which are unstable; however, these were not the major form of $\mathrm{NH_{4}^{+}}$ in Chongqing due to the high $\mathrm{PM}_{2.5}$ acidity dominated by $\mathrm{SO}_{4}^{2-}$ .
The influence of atmospheric transport, diffusion and deposition on $\mathrm{NH_{4}^{+}}$ concentration at Chongqing can be assessed from the variation in air mass trajectories and meteorological factors. For the two springs at Chongqing, little difference was observed in the patterns of air mass backward trajectories, as previously discussed. Meanwhile, as shown in Fig. 6b, surface temperature, wind speed and relative humidity also showed weak variations. Together they suggest that atmospheric transport and diffusion at Chongqing play a minor role in explaining the significant inter-annual variation of $\mathrm{NH}_{4}^{+}$ during the springs of 2005 and 2006.
In contrast to all the above factors, the amount of precipitation was $35.5\,\%$ higher in the spring of 2005 than of 2006, which is comparable to the differences for $\mathrm{NH}_{4}^{+}\,(28.0-$ $30.2\,\%)$ and $\mathrm{R_{C/A}}$ $(26.3{-}30.7\,\%)$ . As shown in Fig. 7a, the precipitation was negatively correlated with $\mathrm{NH}_{4}^{+}$ in $\mathrm{PM}_{2.5}$ at the JB site from February 2005 to April 2006 $\mathit{\Delta}R=-0.63$ , $p=0.01$ ), indicating that the wet removal of $\mathrm{NH}_{4}^{+}$ was favored by the increase in precipitation. In fact, southwestern China experienced a long drought from the fall of 2005 to the spring of 2006, and the number of days on which rain fell during spring 2005 was 20 to $50\,\%$ more than in the spring of 2006 for most cities in the Sichuan Basin (Supplement Fig. S3; meteorological data from http://www. wunderground.com). Although the precipitation in spring at Beijing also showed large inter-annual variation (Fig. 6a), a similar effect of increased precipitation was not evidence due to the much smaller rain volumes during both springs ( $40\,\mathrm{mm}$ and $11\,\mathrm{mm}$ in 2005 and 2006, respectively).
Along with these lines of evidence, the chemistry of wet deposition at Chongqing, derived from the Acid Deposition Monitoring Network in East Asia (http://www.eanet.cc/ product/index.html), also suggests the significant influence of precipitation on the variation of $\mathrm{NH}_{4}^{+}$ in $\mathrm{PM}_{2.5}$ . Since $\mathrm{NH}_{4}^{+}$ and $\mathrm{SO}_{4}^{2-}$ were the major species determining $\mathrm{PM}_{2.5}$ acidity at Chongqing, the ratio of $\mathrm{NH}_{4}^{+}/\mathrm{SO}_{4}^{2-}$ (eq/eq) was used to better present the chemical behavior of $\mathrm{NH_{4}^{+}}$ in determining the acidity of $\mathrm{PM}_{2.5}$ and precipitation. As shown in Fig. 7b, during February–November 2005 when precipitation was relatively abundant, the ratios of $\mathrm{NH_{4}^{+}/S O_{4}^{2-}}$ in the precipitation and $\mathrm{PM}_{2.5}$ were significantly negatively correlated with each other at Chongqing ( $R=-0.88$ , $p<0.001]$ ). However, only a weak correlation ( $\ R=0.44$ , $p=0.38)$ ) was found for the dry seasons from November 2005 to April
2006, possibly due to the fact that the two datasets became less suitable for comparison as the period covered by the rain samples was much shorter than that of the $\mathrm{PM}_{2.5}$ samples.
Interestingly, the influence of the Asian summer monsoon on wet deposition of $\mathrm{NH_{4}^{+}}$ is also evident in Fig. 7b. The ratio of $\mathrm{NH}_{4}^{+}/\mathrm{SO}_{4}^{2-}$ increased from February to June in 2005, decreased and remained at a low level after the arrival of the prevailing summer monsoon in July, and returned to a higher level in October when the winter monsoon started to prevail. All of these collectively suggest that precipitation was one of the key factors that dominated the partition of $\mathrm{NH_{4}^{+}}$ in $\mathrm{PM}_{2.5}$ and rain water at Chongqing, and the enhanced wet deposition of $\mathrm{NH_{4}^{+}}$ was responsible for the lower $\mathrm{NH_{4}^{+}}$ and higher acidity of $\operatorname{PM}_{2.5}$ during the spring of 2005 compared to that of 2006. Our findings have important implications for the interpretation of large-scale variability of airborne $\mathrm{NH}_{3}/\mathrm{NH}_{4}^{+}$ . For example, large inter-annual variation of $\mathrm{NH}_{3}$ was recently observed at a rural site in southwestern China, the cause of which remained undetermined (Meng et al., 2010).
As $\mathrm{SO}_{4}^{2-}$ competes with $\mathrm{NO}_{3}^{-}$ for $\mathrm{NH_{4}^{+}}$ during its formation, the relationship between $\mathrm{NO}_{3}^{-}$ and $\mathrm{NH_{4}^{+}}$ at different levels of $\mathrm{SO}_{4}^{2-}$ , which are expressed as $[\mathrm{NO}_{3}^{-}]/[\mathrm{SO}_{4}^{2-}]$ and $[\mathrm{NH}_{4}^{+}]/[\mathrm{SO}_{4}^{2-}]$ , is indicative of the pathway of $\mathrm{NO}_{3}^{-}$ formation (Pathak et al., 2004a, 2009). For a variety of cities worldwide, linear correlation between $[\mathrm{NO}_{3}^{-}]/[\mathrm{SO}_{4}^{2-}]$ and $[\mathrm{NH_{4}^{+}}]/[\mathrm{SO_{4}^{2-}}]$ in $\mathrm{NH}_{4}^{+}$ -rich conditions $([\mathrm{NH_{4}^{+}}]/[\mathrm{SO_{4}^{2-}}]\geq1.5$ , molar ratio) suggested the homogenous formation of $\mathrm{NO}_{3}^{-}$ :
$$
\mathrm{HNO}_{3}(\mathrm{g})+\mathrm{NH}_{3}(\mathrm{g})\rightleftharpoons\mathrm{NH}_{4}\mathrm{NO}_{3}(\mathrm{s},\mathrm{a}\mathrm{q})
$$
while no relationship was observed in $\mathrm{NH}_{4}^{+}$ -poor conditions $(\mathrm{[NH_{4}^{+}]/[S O_{4}^{2-}]<1.5)}$ and the high level of $\mathrm{NO}_{3}^{-}$ was attributed to its formation from the hydrolysis of $\mathrm{N}_{2}\mathrm{O}_{5}$ on the preexisting aerosols (Pathak et al., 2009):
$$
\mathrm{N}_{2}\mathrm{O}_{5}(\mathrm{aq})+\mathrm{H}_{2}\mathrm{O}(\mathrm{aq})\rightleftharpoons2\mathrm{NO}_{3}^{-}(\mathrm{aq})+2\mathrm{H}^{+}(\mathrm{aq})
$$
However, our study found significant correlations between $[\mathrm{NO}_{3}^{-}]/[\mathrm{SO}_{4}^{2-}]$ and $[\mathrm{NH_{4}^{+}}]/[\mathrm{\bar{S}O_{4}^{2-}}]$ during both $\mathrm{NH_{4}^{+}}$ -rich and $\mathrm{NH}_{4}^{+}$ -poor conditions at Beijing and Chongqing. Thus aerosol acidity, in terms of the ratio $\mathrm{R_{C/A}}$ , was used instead as the key parameter to investigate the relationship between $[\mathrm{NO}_{3}^{-}]/[\dot{\mathrm{S}}\dot{\mathrm{O}_{4}^{2-}}]$ and $[\mathrm{NH_{4}^{+}}]/[\mathrm{SO_{4}^{2-}}]$ , as well as the formation pathways of $\mathrm{NO}_{3}^{-}$ . Considering the uncertainties in representing the neutralization level of $\mathrm{PM}_{2.5}$ , a $\mathrm{R_{C/A}}$ ratio of 0.9 was used to divide the samples into a group of more acidic aerosols $(\mathrm{R_{C/A}}<0.9)$ and a group of less acidic aerosols $(\mathrm{R_{C/A}}\geq0.9)$ , which gave the prospect of a good fit in the regression analysis as discussed below.
$[\mathrm{NO}_{3}^{-}]/[\mathrm{SO}_{4}^{2-}]$ is plotted against $[\mathrm{NH}_{4}^{+}]/[\mathrm{SO}_{4}^{2-}]$ at different acidities for both Beijing and Chongqing $\operatorname{PM}_{2.5}$ in Fig. 8a. For less acidic samples $(\mathrm{R_{C/A}}\geq0.9)$ , although their $[\mathrm{NO}_{3}^{-}]/[\mathrm{SO}_{4}^{2-}]$ showed a clear intercity variation, with higher ratios at Beijing $(\geq\!0.6)$ than at Chongqing $(\le\!0.6)$ , together they were significantly correlated with $[\bar{\mathrm{NH}}_{4}^{+}]/[\mathrm{SO}_{4}^{2-}]$ ( $R^{2}\!=\!0.71$ , $p<0.001]$ ) with the regression function:
$$
\frac{[\mathrm{NO}_{3}^{-}]}{[\mathrm{SO}_{4}^{2-}]}\,{=}\,0.80\times\frac{[\mathrm{NH}_{4}^{+}]}{[\mathrm{SO}_{4}^{2-}]}\,{-}\,1.33
$$
The intercept of the regression line with the axis of $[\mathrm{NH_{4}^{+}}]/[\mathrm{SO_{4}^{2-}}]$ (1.66) was close to that (1.5) found by Pathak et al. (2009), indicating that $\mathrm{NO}_{3}^{-}$ in these less acidic samples was mainly formed from Eq. (4) between $\mathrm{HNO}_{3}$ and the excess $\mathrm{NH}_{3}$ , which became available after neutralizing most of the $\mathrm{SO}_{4}^{2-}$ and $\mathrm{HSO_{4}^{-}}$ . Excess $\mathrm{NH_{4}^{+}}$ associated with the formation of $\mathrm{NO}_{3}^{-}$ can be derived from the following equation:
$$
[\mathrm{NH}_{4}^{+}]_{\mathrm{Excess}}=\left(\frac{[\mathrm{NH}_{4}^{+}]}{[\mathrm{SO}_{4}^{2-}]}-1.66\right)\times[\mathrm{SO}_{4}^{2-}]
$$
For both Beijing and Chongqing, significant correlation $R^{2}\,{=}\,0.70$ , $p<0.001]$ ) was found between the excess $\mathrm{NH_{4}^{+}}$ and $\mathrm{NO}_{3}^{-}$ when $[\mathrm{NH}_{4}^{+}]_{\mathrm{Excess}}\geq30\,\mathrm{nmol}\,\mathrm{m}^{-3}$ (Fig. 9). The slope of regression line for $\mathrm{NO}_{3}^{-}$ against excess $\mathrm{NH}_{4}^{+}$ at Beijing equaled to 1.0, which is consistent with the molar ratio for the reaction between $\mathrm{HNO}_{3}$ and $\mathrm{NH}_{3}$ . However, a shallower slope (0.65) was found for Chongqing, which indicates that in $\mathrm{PM}_{2.5}$ there was approximately $35\,\%$ excess $\mathrm{NH_{4}^{+}}$ bounded to species other than $\mathrm{NO}_{3}^{-}$ . Some of which might exist in the form of $\mathrm{NH_{4}C l}$ , while others could be associated with acidic $\mathrm{SO}_{4}^{2-}$ or $\mathrm{HSO_{4}^{-}}$ , which could recapture the decomposed $\mathrm{NH}_{3}$ from $\mathrm{NH}_{4}\mathrm{NO}_{3}$ , as shown by the following equation:
$$
\mathrm{NH}_{4}\mathrm{NO}_{3}(\mathrm{s},\mathrm{a}\mathrm{q})+\mathrm{H}^{+}(\mathrm{a}\mathrm{q})\rightleftharpoons\mathrm{HNO}_{3}(\mathrm{g},\mathrm{a}\mathrm{q})+\mathrm{NH}_{4}^{+}(\mathrm{a}\mathrm{q})
$$
For the more acidic samples $(\mathrm{R_{C/A}}<0.9)$ ), a significant correlation $?R^{2}=0.59$ , $p<0.001)$ also existed between $[\mathrm{NO}_{3}^{-}]/[\mathrm{SO}_{4}^{2-}]$ and $[\mathrm{NH}_{4}^{+}]/[\mathrm{SO}_{4}^{2-}]$ , although $R^{2}$ was slightly lower than that of the less acidic samples. Its regression equation was:
$$
\frac{[\mathrm{NO}_{3}^{-}]}{[\mathrm{SO}_{4}^{2-}]}\,{=}\,0.59\times\frac{[\mathrm{NH}_{4}^{+}]}{[\mathrm{SO}_{4}^{2-}]}\,{-}0.42
$$
Its intercept with the axis of $[\mathrm{NH_{4}^{+}}]/[\mathrm{SO_{4}^{2-}}]$ (0.71) was much smaller than that for the less acidic samples (1.66). However it is notable that the two lines approximated each other as $[\mathrm{NH_{4}^{+}}]/[\mathrm{SO_{4}^{2-}}]$ increased over 1.66, indicating that the homogeneous reaction for $\mathrm{HNO}_{3}$ and $\mathrm{NH}_{3}$ was also favored with abundant $\mathrm{NH_{4}}$ in the acidic samples. On the other hand, along with decreasing $[\mathrm{NH}_{4}^{+}]/[\mathrm{SO}_{4}^{2-}]$ , a large $\mathrm{NH}_{4}^{+}$ deficit $(\mathrm{[NH_{4}^{+}]_{E x c e s s}}$ of up to $-350\,\mathrm{nmol}\,\mathrm{m}^{-3}$ , calculated according to Eq. 7) was evident in these more acidic samples (Fig. 9), and the reaction of Eq. (4) was supposed to be constrained. However, markedly high concentrations of $\mathrm{NO}_{3}^{-}$ were still found at Chongqing (up to $169\,\mathrm{nmol}\,\mathrm{m}^{-3}$ ) and Beijing (up to $318\,\mathrm{nmol}\,\mathrm{m}^{-3};$ ), which suggests the dominance of heterogeneous reactions without involving $\mathrm{NH}_{3}$ , most likely the hydrolysis of $\mathrm{N}_{2}\mathrm{O}_{5}$ on the preexisting aerosols (Pathak et al., 2009).
The close correlation between $[\mathrm{NO}_{3}^{-}]/[\mathrm{SO}_{4}^{2-}]$ and $[\mathrm{NH_{4}^{+}}]/[\mathrm{SO_{4}^{2-}}]$ for the more acidic aerosols (which is in contrast to the weak correlation reported by Pathak et al., 2009) can be explained in at least two ways. Firstly, due to the long sampling duration of our study $\mathord{\left\langle{\sim}1\right\rangle}$ week), $\mathrm{NO}_{3}^{-}$ in each sample had formed from both homogenous and heterogeneous reactions, and thus a good correlation between $[\bar{\mathrm{NO}}_{3}^{-}]/[\bar{\mathrm{SO}}_{4}^{2-}]$ and $[\mathrm{NH_{4}^{+}}]/[\mathrm{SO_{4}^{2-}}]$ might exist even for those dominated by the latter pathway. Secondly, with an increase in aerosol acidity and a decrease in $[\mathrm{NH}_{4}^{+}]/[\mathrm{SO}_{4}^{2-}]$ , $\mathrm{NO}_{3}^{-}$ tends to partition into coarse particles with abundant alkaline mineral components (Pakkanen et al., 1996; Zhuang et al., 1999). Consequently there is a decrease in $[\mathrm{NO}_{3}^{-}]/[\mathrm{SO}_{4}^{2-}]$ that is positively correlated with $[\mathrm{NH}_{4}^{+}]/[\mathrm{SO}_{4}^{2-}]$ in fine particles (Sun et al., 2006).
These hypotheses are further suggested by coupling the variation of aerosol water content with the two compositional ratios. As shown in Fig. 8b, aerosol samples with $[\mathrm{NH}_{4}^{+}]/[\mathrm{SO}_{4}^{2-}]<1.66$ contained more water, indicating the vital role of the liquid phase reactions of Eq. (5), while less water was observed in most of the other samples $(\mathrm{[NH_{4}^{+}]/[S O_{4}^{2-}]\geq1.66})$ , in which $\mathrm{NO}_{3}^{-}$ formation was dominated by the gaseous reaction of Eq. (1). However, relatively high water content was also found in some of the less acidic samples with high $[\mathrm{NH}_{4}^{+}]/[\mathrm{SO}_{4}^{2-}]$ , which were mainly collected during the winter in Chongqing and the summer in Beijing. $\mathrm{NO}_{3}^{-}$ in these samples may be formed on the existing particles and/or to in-cloud processes as $\mathrm{NH}_{4}\mathrm{NO}_{3}$ (Yao et al., 2003).
It should be noted that there were, respectively, 5 and 2 outliers excluded in the regression analysis of $[\mathrm{NO}_{3}^{-}]/[\mathrm{SO}_{4}^{2-}]$ against $[\mathrm{NH_{4}^{+}}]/[\mathrm{SO_{4}^{2-}}]$ for the less acidic and the more acidic samples, as shown in Fig. 8a. These outliers, all collected during spring and winter in Beijing, were found to have significantly higher $[\mathrm{Ca}^{2+}]/[\mathrm{NH}_{4}^{+}]$ than other samples with low water content (Fig. 8b), which suggests that heterogeneous reactions on the dry surface of fine mineral particles was also an important pathway during these periods at Beijing.
4 Conclusions and atmospheric implications
The spatial and seasonal variations of $\mathrm{PM}_{2.5}$ acidity were investigated at both rural and urban sites of Beijing and Chongqing from January 2005 to May 2006. With similar levels of $\mathrm{NH_{4}^{+}}$ at each site, $\mathrm{PM}_{2.5}$ was generally more acidic at Chongqing than at Beijing. $\mathrm{SO}_{4}^{2-}$ concentrations in $\mathrm{PM}_{2.5}$ was higher in Chongqing but lower in Beijing, indicating a more important contribution to $\mathrm{PM}_{2.5}$ from coal combustion in southwestern China and more influence to $\mathrm{PM}_{2.5}$ from vehicular emissions in Beijing. The intra-city comparison of $\operatorname{PM}_{2.5}$ acidity showed a reverse pattern for Beijing and Chongqing, with higher levels of $\mathrm{NO}_{3}^{-}$ and lower levels of $\mathrm{{Ca}}^{2+}$ making $\mathrm{PM}_{2.5}$ more acidic in urban areas of Beijing and in rural areas of Chongqing.
$\mathrm{PM}_{2.5}$ was more acidic in the summer and fall than in winter of 2005 at Beijing and Chongqing, but large inter-annual variation was found during the springs of 2005 and 2006, with two cities exhibiting opposite trends. The higher acidity of $\operatorname{PM}_{2.5}$ in the spring of 2006 at Beijing was attributed to the influence of Asian desert dust with significant enhancement of the formation of $\mathrm{NO}_{3}^{-}$ relative to $\mathrm{\bar{so}}_{4}^{2-}$ , both of which were not completely neutralized by the increase in alkaline dust, however. For Chongqing, the higher acidity of $\mathrm{PM}_{2.5}$ in the spring of 2005 was mainly due to increased wet deposition of $\mathrm{NH}_{4}^{+}$ . As revealed by a variety of previous studies, significant inter-annual variation in aerosol acidity was also found during other seasons at Beijing, Chongqing and many other cities in China, with no consistent long-term trend. These variations may be influenced by a variety of factors, such as emission strength, meteorological conditions and the characteristics of preexisting particles and precursors.
The Asian monsoon systems were found to be related to the synoptic-scale evolution of $\operatorname{PM}_{2.5}$ acidity at Beijing and Chongqing from spring to early summer in 2005. For both cities, $\mathrm{PM}_{2.5}$ acidity increased from spring to early summer of 2005, a trend that was closely associated with an increased contribution of air masses from between the Northern China Plain to the south of Beijing and from central China to the east of Chongqing. The regionally acidic aerosols were replaced by more neutralized aerosols at the end of June 2005, coupled with the northward movement of a subtropical high over the northwestern Pacific, which is a major element of the Asian summer monsoon. Previous studies have found a seasonal influence of the Asian monsoon on the concentrations of aerosol and gaseous pollutants in China (e.g. He et al., 2001; Ye et al., 2003; Wai and Turner, 2005; Xin et al., 2007; Zhang et al., 2010), but few of them, if any, have related its behavior to large-scale variability in aerosol acidity. Moreover, a recent study using modeling suggests that the strength of the Asian monsoon could influence the inter-annual variation in aerosols in eastern China mostly by altering wet deposition and aerosol transport (Zhang et al., 2010), which we believe is also likely to explain the inter-annual variation of aerosol acidity during the springs of 2005 and 2006. For example, RH, one of most important parameters influencing in situ aerosol acidity, exhibited opposite trends from winter into spring at Chongqing during 2005 and 2006, a situation that was likely influenced by large-scale synoptic system anomalies. In the meantime, more air masses from northwestern Asian deserts were transported to the south of the country in the spring of 2005 than of 2006, while the reverse trend was found for the air masses influencing the Northern China Plain (not shown). This may also have been related to variation of the strength of the Asian monsoon, which was found to have greater bearing on the transport pathway of dust to the Asian subcontinent than to dust production itself (Gong et al., 2006). Clearly, these are important subjects of future work.
$\mathrm{PM}_{2.5}$ acidity was closely related to the formation of $\mathrm{NO}_{3}^{-}$ at both Beijing and Chongqing. $\mathrm{NO}_{3}^{-}$ formation in more neutralized $\mathrm{PM}_{2.5}$ was favored by the homogeneous reaction of $\mathrm{HNO}_{3}$ and $\mathrm{NH}_{3}$ , while heterogeneous reactions (such as the hydrolysis of $\mathrm{N}_{2}\mathrm{O}_{5}$ on preexisting aerosols with higher water content) may become major pathways when particulates are more acidic. In addition, the formation of $\mathrm{NO}_{3}^{-}$ on the relatively dry surface of mineral dust could also be an important pathway during winter and spring at Beijing.
Aerosol acidity has also frequently been linked to the formation of secondary organic aerosols (Jang et al., 2002; Takahama et al., 2006; Zhang et al., 2007a). In addition to the inorganic aerosols discussed in this study, we also observed significantly higher ratios of Organic carbon to Elemental carbon (OC/EC) in the spring of 2005 than of 2006 at the rural sites of both Beijing and Chongqing (unpublished data); a pattern which was not in evidence at the urban sites, however. This phenomenon probably indicates the different levels of oxidation of organic aerosols in the background air masses, and more detailed investigation is required.
Supplementary material related to this article is available online at: http://www.atmos-chem-phys.net/12/1377/2012/ acp-12-1377-2012-supplement.pdf.
Acknowledgements. This research was supported by the National Natural Science Foundation of China (20625722). We would like to thank Yingtao Jia, Yuan Cheng and the staff of the Chongqing Research Academy of Environmental Sciences for their help in sample collection and chemical analysis. We also gratefully acknowledge the NOAA Air Resources Laboratory (ARL) for provision of the HYSPLIT trajectory model and meteorological data, and Simon Clegg, Peter Brimblecombe and Anthony Wexler for sharing the E-AIM model.
Edited by: X. Tie
References
Aas, W., Shao, M., Jin, L., Larssen, T., Zhao, D. W., Xiang, R. J., Zhang, J. H., Xiao, J. S., and Duan, L.: Air concentrations and wet deposition of major inorganic ions at five non-urban sites in China, 2001–2003, Atmos. Environ., 41, 1706–1716, doi:10.1016/j.atmosenv.2006.10.030, 2007.
Adams, P. J., Seinfeld, J. H., and Koch, D. M.: Global concentrations of tropospheric sulfate, nitrate, and ammonium aerosol simulated in a general circulation model, J. Geophys. Res.-Atmos., 104, 13791–13823, 1999.
Amdur, M. O. and Chen, L. C.: Furnace-generated acid aerosols – speciation and pulmonary effects, Environ. Health Perspect., 79, 147–150, 1989.
Arimoto, R., Duce, R. A., Savoie, D. L., Prospero, J. M., Talbot, R., Cullen, J. D., Tomza, U., Lewis, N. F., and Jay, B. J.: Relationships among aerosol constituents from Asia and the North Pacific during PEM-West A, J. Geophys. Res.-Atmos., 101, 2011–2023, 1996.
Asman, W. A. H., Sutton, M. A., and Schjorring, J. K.: Ammonia: emission, atmospheric transport and deposition, New Phytol., 139, 27–48, 1998.
Beijing Municipal Bureau of Statistics: Beijing statistical Yearbook, http://www.bjstats.gov.cn/tjnj/2006-tjnj, 2006 (in Chinese).
Bergin, M. H., Cass, G. R., Xu, J., Fang, C., Zeng, L. M., Yu, T., Salmon, L. G., Kiang, C. S., Tang, X. Y., Zhang, Y. H., and Chameides, W. L.: Aerosol radiative, physical, and chemical properties in Beijing during June 1999, J. Geophys. Res.-Atmos., 106, 17969–17980, 2001.
Boucher, O. and Anderson, T. L.: General circulation model assessment of the sensitivity of direct climate forcing by anthropogenic sulfate aerosols to aerosol size and chemistry, J. Geophys. Res.- Atmos., 100, 26117–26134, 1995.
Chan, C. K. and Yao, X.: Air pollution in mega cities in China, Atmos. Environ., 42, 1–42, doi:10.1016/j.atmosenv.2007.09.003, 2008.
Cheung, H. C., Wang, T., Baumann, K., and Guo, H.: Influence of regional pollution outflow on the concentrations of fine particulate matter and visibility in the coastal area of southern China, Atmos. Environ., 39, 6463–6474, doi:10.1016/j.atmosenv.2005.07.033, 2005.
Chongqing Municipal Bureau of Statistics: Chongqing statistical Yearbook, http://www.cqtj.gov.cn/tjnj/2006/, 2006 (in Chinese).
Chou, C. C. K., Lee, C. T., Yuan, C. S., Hsu, W. C., Lin, C. Y., Hsu, S. C., and Liu, S. C.: Implications of the chemical transformation of Asian outflow aerosols for the long-range transport of inorganic nitrogen species, Atmos. Environ., 42, 7508–7519, doi:10.1016/j.atmosenv.2008.05.049, 2008.
Chu, S. H.: $\mathrm{PM}_{2.5}$ episodes as observed in the speciation trends network, Atmos. Environ., 38, 5237–5246, doi:10.1016/j.atmosenv.2004.01.055, 2004.
Clegg, S. L., Brimblecombe, P., and Wexler, A. S.: Thermodynamic model of the system $\mathrm{H^{+}{-N H_{4}^{+}{-S O_{4}^{2}{-N O_{3}{-H_{2}O}}}}}$ at tropospheric temperatures, J. Phys. Chem. A, 102, 2137–2154, 1998.
Crumeyrolle, S., Gomes, L., Tulet, P., Matsuki, A., Schwarzenboeck, A., and Crahan, K.: Increase of the aerosol hygroscopicity by cloud processing in a mesoscale convective system: a case study from the AMMA campaign, Atmos. Chem. Phys., 8, 6907–6924, doi:10.5194/acp-8-6907-2008, 2008.
Dillner, A. M., Schauer, J. J., Zhang, Y. H., Zeng, L. M., and Cass, G. R.: Size-resolved particulate matter composition in Beijing during pollution and dust events, J. Geophys. Res.-Atmos., 111, D05203, doi:10.1029/2005jd006400, 2006.
Draxler, R. R., Stunder, B., Rolph, G., and Taylor, A.: HYSPLIT4 user’s guide (online version), NOAA Tech. Memo. ERL ARL230., available at: http://www.arl.noaa.gov/data/web/models/ hysplit4/win95/user guide.pdf, 2006.
Feng, J. L., Guo, Z. G., Chan, C. K., and Fang, M.: Properties of organic matter in $\mathrm{PM}_{2.5}$ at Changdao Island, China – A rural site in the transport path of the Asian continental outflow, Atmos. Environ., 41, 1924–1935, doi:10.1016/j.atmosenv.2006.10.064, 2007.
Gong, S. L., Zhang, X. Y., Zhao, T. L., Zhang, X. B., Barrie, L. A., McKendry, I. G., and Zhao, C. S.: A simulated climatology of Asian dust aerosol and its trans-Pacific transport. Part II: Interannual variability and climate connections, J. Climate, 19, 104–122, 2006.
Guo, S., Hu, M., Wang, Z. B., Slanina, J., and Zhao, Y. L.: Sizeresolved aerosol water-soluble ionic compositions in the summer of Beijing: implication of regional secondary formation, Atmos. Chem. Phys., 10, 947–959, doi:10.5194/acp-10-947-2010, 2010.
Han, J. S., Moon, K. J., Lee, S. J., Kim, Y. J., Ryu, S. Y., Cliff, S. S., and Yi, S. M.: Size-resolved source apportionment of ambient particles by positive matrix factorization at Gosan background site in East Asia, Atmos. Chem. Phys., 6, 211–223, doi:10.5194/acp-6-211-2006, 2006.
He, K. B., Yang, F. M., Ma, Y. L., Zhang, Q., Yao, X. H., Chan, C. K., Cadle, S., Chan, T., and Mulawa, P.: The characteristics of $\mathrm{PM}_{2.5}$ in Beijing, China, Atmos. Environ., 35, 4959–4970, 2001.
He, K. B., Zhao, Q., Ma, Y. L., Duan, F. K., and Yang, F. M.: Seasonal composition and long-term trend of $\mathrm{PM}_{2.5}$ at Beijing and Chongqing, China, Atmos. Environ., submitted, 2011.
Health Effects Institute: Understanding the health effects of components of the particulate matter mix: progress and next steps, HEI Perspectives, 2002.
Hu, J. H. and Abbatt, J. P. D.: Reaction probabilities for $\mathrm{N}_{2}\mathrm{O}_{5}$ hydrolysis on sulfuric acid and ammonium sulfate aerosols at room temperature, J. Phys. Chem. A, 101, 871–878, 1997.
Huang, M. Y., Shen, Z. L., Wu, Y. X., Luo, Q. R., Xu, Y., Zhang, D. B., and Chen, S. L.: Observation on the acidity and chemical compositions of cloudwater and rainwater over Chongqing, Scientia Atmospherica Sinica, 12, 389–395, 1988.
Jang, M. S., Czoschke, N. M., Lee, S., and Kamens, R. M.: Heterogeneous atmospheric aerosol production by acid-catalyzed particle-phase reactions, Science, 298, 814–817, 2002.
Jia, Y. T., Rahn, K. A., He, K. B., Wen, T. X., and Wang, Y. S.: A novel technique for quantifying the regional component of urban aerosol solely from its sawtooth cycles, J. Geophys. Res.-Atmos., 113, D21309, doi:10.1029/2008jd010389, 2008.
Jickells, T. D., An, Z. S., Andersen, K. K., Baker, A. R., Bergametti, G., Brooks, N., Cao, J. J., Boyd, P. W., Duce, R. A., Hunter, K. A., Kawahata, H., Kubilay, N., LaRoche, J., Liss, P. S., Mahowald, N., Prospero, J. M., Ridgwell, A. J., Tegen, I., and Torres, R.: Global iron connections between desert dust, ocean biogeochemistry, and climate, Science, 308, 67–71, 2005.
Johansen, A. M., Siefert, R. L., and Hoffmann, M. R.: Chemical characterization of ambient aerosol collected during the southwest monsoon and intermonsoon seasons over the Arabian Sea: Anions and cations, J. Geophys. Res.-Atmos., 104, 26325– 26347, 1999.
John, W., Wall, S. M., Ondo, J. L., and Winklmayr, W.: Modes in the size distributions of atmospheric inorganic aerosol, Atmos. Environ. A-Gen., 24, 2349–2359, 1990.
Larssen, T., Lydersen, E., Tang, D. G., He, Y., Gao, J. X., Liu, H. Y., Duan, L., Seip, H. M., Vogt, R. D., Mulder, J., Shao, M., Wang, Y. H., Shang, H., Zhang, X. S., Solberg, S., Aas, W., Okland, T., Eilertsen, O., Angell, V., Liu, Q. R., Zhao, D. W., Xiang, R. J., Xiao, J. S., and Luo, J. H.: Acid rain in China, Environ. Sci. Technol., 40, 418–425, 2006.
Li, W. J. and Shao, L. Y.: Observation of nitrate coatings on atmospheric mineral dust particles, Atmos. Chem. Phys., 9, 1863– 1871, doi:10.5194/acp-9-1863-2009, 2009.
Li, X. G., Wang, S. X., Duan, L., Hao, J., Li, C., Chen, Y. S., and Yang, L.: Particulate and trace gas emissions from open burning of wheat straw and corn stover in China, Environ. Sci. Technol., 41, 6052–6058, doi:10.1021/es0705137, 2007.
Li, X. H., Wang, S. X., Duan, L., Hao, J. M., and Nie, Y. F.: Carbonaceous Aerosol Emissions from Household Biofuel Combustion in China, Environ. Sci. Technol., 43, 6076–6081, doi:10.1021/es803330j, 2009.
Liu, L. J. S., Burton, R., Wilson, W. E., and Koutrakis, P.: Comparison of aerosol acidity in urban and semirural environments, Atmos. Environ., 30, 1237–1245, 1996.
Liu, S. R. and Huang, M. Y.: Numerical simulation on the acidification of rainwater under cloud and formation of acid rain in Chongqing, Scientia Atmospherica Sinica, 12, 245–257, 1988.
Louie, P. K. K., Chow, J. C., Chen, L. W. A., Watson, J. G., Leung, G., and Sin, D. W. M.: $\mathrm{PM}_{2.5}$ chemical composition in Hong Kong: Urban and regional variations, Sci. Total Environ., 338, 267–281, 2005.
Lu, X. Y., Zhang, X. Z., and Chen, J. N.: Study on the anomaly of the East Asian summer monsoon in June 2005 and the affecting mechanism, J. Trop. Meteorol., 23, 553–562, 2007 (in Chinese with abstract in English).
Luo, C., Zender, C. S., Bian, H. S., and Metzger, S.: Role of ammonia chemistry and coarse mode aerosols in global climatological inorganic aerosol distributions, Atmos. Environ., 41, 2510–2533, doi:10.1016/j.atmosenv.2006.11.030, 2007.
Meng, Z. Y., Seinfeld, J. H., Saxena, P., and Kim, Y. P.: Atmospheric gas-aerosol equilibrium .4. thermodynamics of carbonates, Aerosol Sci. Technol., 23, 131–154, 1995.
Meng, Z. Y., Xu, X. B., Wang, T., Zhang, X. Y., Yu, X. L., Wang, S. F., Lin, W. L., Chen, Y. Z., Jiang, Y. A., and An, X. Q.: Ambient sulfur dioxide, nitrogen dioxide, and ammonia at ten background and rural sites in China during 2007–2008, Atmos. Environ., 44, 2625–2631, doi:10.1016/j.atmosenv.2010.04.008, 2010.
Meskhidze, N., Chameides, W. L., and Nenes, A.: Dust and pollution: A recipe for enhanced ocean fertilization?, J. Geophys. Res., 110, D03301, doi:10.1029/2004JD005082, 2005.
Mu, J. L., Wang, J. J., and Li, Z. C.: A study of environment and mesoscale convective systems of continuous heavy rainfall in the South of China in June 2005, Acta Meteorol. Sin., 66, 437–451, 2008.
Nenes, A., Pandis, S. N., and Pilinis, C.: ISORROPIA: A new thermodynamic equilibrium model for multiphase multicomponent inorganic aerosols, Aquat. Geochem., 4, 123–152, 1998.
Nenes, A., Krom, M. D., Mihalopoulos, N., Van Cappellen, P., Shi, Z., Bougiatioti, A., Zarmpas, P., and Herut, B.: Atmospheric acidification of mineral aerosols: a source of bioavailable phosphorus for the oceans, Atmos. Chem. Phys., 11, 6265–6272, doi:10.5194/acp-11-6265-2011, 2011.
Ooki, A. and Uematsu, M.: Chemical interactions between mineral dust particles and acid gases during Asian dust events, J. Geophys. Res.-Atmos., 110, D03201, doi:10.1029/2004jd004737, 2005.
Pakkanen, T. A., Kerminen, V. M., Hillamo, R. E., Makinen, M., Makela, T., and Virkkula, A.: Distribution of nitrate over seasalt and soil derived particles – Implications from a field study, J. Atmos. Chem., 24, 189–205, 1996.
Pathak, R. K., Yao, X. H., Lau, A. K. H., and Chan, C. K.: Acidity and concentrations of ionic species of $\mathrm{PM}_{2.5}$ in Hong Kong, Atmos. Environ., 37, 1113–1124, doi:10.1016/s1352- 2310(02)00958-5, 2003.
Pathak, R. K., Louie, P. K. K., and Chan, C. K.: Characteristics of aerosol acidity in Hong Kong, Atmos. Environ., 38, 2965–2974, doi:10.1016/j.atmosenv.2004.02.044, 2004a.
Pathak, R. K., Yao, X. H., and Chan, C. K.: Sampling artifacts of acidity and ionic species in $\mathrm{PM}_{2.5}$ , Environ. Sci. Technol., 38, 254–259, doi:10.1021/es0342244, 2004b.
Pathak, R. K., Wu, W. S., and Wang, T.: Summertime $\mathrm{PM}_{2.5}$ ionic species in four major cities of China: nitrate formation in an ammonia-deficient atmosphere, Atmos. Chem. Phys., 9, 1711– 1722, doi:10.5194/acp-9-1711-2009, 2009.
Quan, H. N. and Zhang, X. S.: Assessing the role of ammonia in sulfur transformation and deposition in China, Atmos. Res., 88, 78–88, doi:10.1016/j.atmosres.2007.10.006, 2008.
Richter, A., Burrows, J. P., Nuss, H., Granier, C., and Niemeier, U.: Increase in tropospheric nitrogen dioxide over China observed from space, Nature, 437, 129–132, doi:10.1038/nature04092, 2005.
Roger, G. B. and Andrew, M. C.: Synoptic and dynamic climatology, New York, Routledge, 735–876, 2002.
Seinfeld, J. H. and Pandis, S. N.: Atmos. Chem. Phys.: From Air Pollution to Climate Change, New York, Wiley, 1998.
Shen, Z. L., Wu, Y. X., Xiao, H., and Huang, M. Y.: Annual variation of chemical characteristics of cloudwater in Chongqing area, Acta Scientiae Circumstantiae, 12, 193–199, 1992.
Shen, Z. X., Arimoto, R., Cao, J. J., Zhang, R. J., Li, X. X., Du, N., Okuda, T., Nakao, S., and Tanaka, S.: Seasonal Variations and Evidence for the Effectiveness of Pollution Controls on WaterSoluble Inorganic Species in Total Suspended Particulates and Fine Particulate Matter from Xi’an, China, J. Air Waste Manage., 58, 1560–1570, doi:10.3155/1047-3289.58.12.1560, 2008.
Shi, Z., Bonneville, S., Krom, M. D., Carslaw, K. S., Jickells, T. D., Baker, A. R., and Benning, L. G.: Iron dissolution kinetics of mineral dust at low pH during simulated atmospheric processing, Atmos. Chem. Phys., 11, 995–1007, doi:10.5194/acp11-995-2011, 2011.
Song, C. H., Park, M. E., Lee, K. H., Ahn, H. J., Lee, Y., Kim, J. Y., Han, K. M., Kim, J., Ghim, Y. S., and Kim, Y. J.: An investigation into seasonal and regional aerosol characteristics in East Asia using model-predicted and remotely-sensed aerosol properties, Atmos. Chem. Phys., 8, 6627–6654, doi:10.5194/acp8-6627-2008, 2008.
Stohl, A.: Computation, accuracy and applications of trajectories – A review and bibliography, Atmos. Environ., 32, 947–966, 1998.
Sun, J. Y., Zhang, Q., Canagaratna, M. R., Zhang, Y. M., Ng, N. L., Sun, Y. L., Jayne, J. T., Zhang, X. C., Zhang, X. Y., and Worsnop, D. R.: Highly time- and size-resolved characterization of submicron aerosol particles in Beijing using an Aerodyne Aerosol Mass Spectrometer, Atmos. Environ., 44, 131–140, doi:10.1016/j.atmosenv.2009.03.020, 2010.
Sun, Y. L., Zhuang, G. S., Wang, Z. F., Wang, Y., Zhang, W. J., Tang, A. H., Zhao, X. J., and Jia, L.: Regional characteristics of spring Asian dust and its impact on aerosol chemistry over northern China, Atmos. Chem. Phys. Discuss., 6, 12825–12864, doi:10.5194/acpd-6-12825-2006, 2006.
Takahama, S., Davidson, C. I., and Pandis, S. N.: Semicontinuous measurements of organic carbon and acidity during the Pittsburgh air quality study: Implications for acid-catalyzed organic aerosol formation, Environ. Sci. Technol., 40, 2191–2199, doi:10.1021/es050856+, 2006.
Takami, A., Miyoshi, T., Shimono, A., Kaneyasu, N., Kato, S., Kajii, Y., and Hatakeyama, S.: Transport of anthropogenic aerosols from Asia and subsequent chemical transformation, J. Geophys. Res.-Atmos., 112, D22S31, doi:10.1029/2006jd008120, 2007.
Vlasenko, A., Sjogren, S., Weingartner, E., Stemmler, K., Ga¨ggeler, H. W., and Ammann, M.: Effect of humidity on nitric acid uptake to mineral dust aerosol particles, Atmos. Chem. Phys., 6, 2147– 2160, doi:10.5194/acp-6-2147-2006, 2006.
Wai, K. M. and Tanner, P. A.: Relationship between ionic composition in $\mathrm{PM_{10}}$ and the synoptic-scale and mesoscale weather conditions in a south China coastal city: A 4-year study, J. Geophys. Res.-Atmos., 110, D18210, doi:10.1029/2004jd005385, 2005.
Wang, W., Jiang, Z. Y., Zhang, M. H., and Wang, W. X.: Pollution characteristic of atmospheric aerosol and relationship between aerosol and acid precipitation in south China, J. Environ. Sci., 12, 7–15, 1992 (in Chinese with abstract in English).
Wang, W., Tang, D. G., Liu, H. J., Yue, X., Pan, Z., and Ding, Y.: Research on Corrent Pollution Sataus and Pollution, Characteristic s of $\mathrm{PM}_{2.5}$ in China, Res. Environ. Sci., 13, 1–5, 2000 (in Chinese with abstract in English).
Wang, Y., Zhuang, G. S., Tang, A. H., Yuan, H., Sun, Y. L., Chen, S. A., and Zheng, A. H.: The ion chemistry and the source of $\mathrm{PM}_{2.5}$ aerosol in Beijing, Atmos. Environ., 39, 3771–3784, doi:10.1016/j.atmosenv.2005.03.013, 2005.
Wang, Y., Zhuang, G. S., Zhang, X. Y., Huang, K., Xu, C., Tang, A. H., Chen, J. M., and An, Z. S.: The ion chemistry, seasonal cycle, and sources of $\mathrm{PM}_{2.5}$ and TSP aerosol in Shanghai, Atmos. Environ., 40, 2935–2952, doi:10.1016/j.atmosenv.2005.12.051, 2006.
Watson, J. G.: Visibility: Science and Regulation, J. Air Waste Manage., 52, 628–713, 2002.
Wu, D., Tie, X. X., and Deng, X. J.: Chemical characterizations of soluble aerosols in southern China, Chemosphere, 64, 749–757, doi:10.1016/j.chemosphere.2005.11.066, 2006.
Wu, Z. J., Cheng, Y. F., Hu, M., Wehner, B., Sugimoto, N., and Wiedensohler, A.: Dust events in Beijing, China (2004– 2006): comparison of ground-based measurements with columnar integrated observations, Atmos. Chem. Phys., 9, 6915–6932, doi:10.5194/acp-9-6915-2009, 2009.
Xin, J. Y., Wang, Y. S., Li, Z. Q., Wang, P. C., Hao, W. M., Nordgren, B. L., Wang, S. G., Liu, G. R., Wang, L. L., Wen, T. X., Sun, Y., and Hu, B.: Aerosol optical depth (AOD) and Angstrom exponent of aerosols observed by the Chinese Sun Hazemeter Network from August 2004 to September 2005, J. Geophys. Res.-Atmos., 112, D05203, doi:10.1029/2006jd007075, 2007.
Xiu, G. L., Zhang, D. N., Chen, J. Z., Huang, X. J., Chen, Z. X., Guo, H. L., and Pan, J. F.: Characterization of major watersoluble inorganic ions in size-fractionated particulate matters in Shanghai campus ambient air, Atmos. Environ., 38, 227–236, doi:10.1016/j.atmosenv.2003.09.053, 2004.
Yang, H., Yu, J. Z., Ho, S. S. H., Xu, J. H., Wu, W. S., Wan, C. H., Wang, X. D., Wang, X. R., and Wang, L. S.: The chemical composition of inorganic and carbonaceous materials in $\mathrm{PM}_{2.5}$ in Nanjing, China, Atmos. Environ., 39, 3735–3749, doi:10.1016/j.atmosenv.2005.03.010, 2005.
Yang, L. X., Wang, D. C., Cheng, S. H., Wang, Z., Zhou, Y., Zhou, X. H., and Wang, W. X.: Influence of meteorological conditions and particulate matter on visual range impairment in Jinan, China, Sci. Total Environ., 383, 164–173, doi:10.1016/j.scitotenv.2007.04.042, 2007.
Yao, X. H., Chan, C. K., Fang, M., Cadle, S., Chan, T., Mulawa, P., He, K. B., and Ye, B. M.: The water-soluble ionic composition of $\mathrm{PM}_{2.5}$ in Shanghai and Beijing, China, Atmos. Environ., 36, 4223–4234, 2002.
Yao, X. H., Lau, A. P. S., Fang, M., Chan, C. K., and Hu, M.: Size distributions and formation of ionic species in atmospheric particulate pollutants in Beijing, China: 1 – inorganic ions, Atmos. Environ., 37, 2991–3000, doi:10.1016/s1352-2310(03)00255-3, 2003.
Yao, X. H., Ling, T. Y., Fang, M., and Chan, C. K.: Comparison of thermodynamic predictions for in situ $\mathrm{pH}$ in $\mathrm{PM}_{2.5}$ , Atmos. Environ., 40, 2835–2844, doi:10.1016/j.atmosenv.2006.01.006, 2006.
Ye, B. M., Ji, X. L., Yang, H. Z., Yao, X. H., Chan, C. K., Cadle, S. H., Chan, T., and Mulawa, P. A.: Concentration and chemical composition of $\mathrm{PM}_{2.5}$ in Shanghai for a 1-year period, Atmos. Environ., 37, 499–510, 2003.
Zhang, L., Liao, H., and Li, J. P.: Impacts of Asian summer monsoon on seasonal and interannual variations of aerosols over eastern China, J. Geophys. Res.-Atmos., 115, D00K05, doi:10.1029/2009jd012299, 2010.
Zhang, Q., Jimenez, J. L., Worsnop, D. R., and Canagaratna, M.: A case study of urban particle acidity and its influence on secondary organic aerosol, Environ. Sci. Technol., 41, 3213–3219, doi:10.1021/es061812j, 2007a.
Zhang, Q., Streets, D. G., He, K., Wang, Y., Richter, A., Burrows, J. P., Uno, I., Jang, C. J., Chen, D., Yao, Z., and Lei, Y.: $\mathrm{NO_{X}}$ emission trends for China, 1995–2004: The view from the ground and the view from space, J. Geophys. Res.-Atmos., 112, D22306, doi:10.1029/2007jd008684, 2007b.
Zhang, X. Y., Cao, J. J., Li, L. M., Arimoto, R., Cheng, Y., Huebert, B., and Wang, D.: Characterization of Atmospheric Aerosol over XiAn in the South Margin of the Loess Plateau, China, Atmos. Environ., 36, 4189–4199, 2002.
Zhao, D. W., Seip, H. M., and Zhang, D. B.: Pattern and cause of acidic deposition in the chongqing region, sichuan-province, china, Water Air Soil Poll., 77, 27–48, 1994.
Zhao, Q., He, K. B., Ma, Y. L., Jia, Y. T., Cheng, Y., Liu, H., and Wang, S. W.: Regional PM Pollution in Beijing and Surrounding Area During Summertime, Environ. Sci., 30, 1873–1880, 2009.
Zhao, Q., He, K., Rahn, K. A., Ma, Y., Jia, Y., Yang, F., Duan, F., Lei, Y., Chen, G., Cheng, Y., Liu, H., and Wang, S.: Dust storms come to Central and Southwestern China, too: implications from a major dust event in Chongqing, Atmos. Chem. Phys., 10, 2615– 2630, doi:10.5194/acp-10-2615-2010, 2010.
Zhuang, H., Chan, C. K., Fang, M., and Wexler, A. S.: Formation of nitrate and non-sea-salt sulfate on coarse particles, Atmos. Environ., 33, 4223–4233, 1999.
Ziemba, L. D., Fischer, E., Griffin, R. J., and Talbot, R. W.: Aerosol acidity in rural New England: Temporal trends and source region analysis, J. Geophys. Res.-Atmos., 112, D10S22, doi:10.1029/2006jd007605, 2007.
|
atmosphere
|
0026
|
10.1016/j.atmosenv.2015.11.055
|
image
|
Fig. 1. $\mathrm{PM}_{2.5}$ samples were taken in two southern China cities: Hong Kong (HK) and Xiamen (XM); and two northern China cities: Beijing (BJ) and Xi'an (XA).
|
|
atmosphere
|
0026
|
10.1016/j.atmosenv.2015.11.055
|
table
|
Table 1 Average $\pm$ standard error) mass concentration and chemical composition of $\mathrm{PM}_{2.5}$ collected at four megacities in China during the haze events.
|
|
atmosphere
|
0026
|
10.1016/j.atmosenv.2015.11.055
|
image
|
Fig. 2. (a) Oxidative potentials, IL-6, IFN- $\cdot Y$ and TNF- $\cdot\mathfrak{x}$ production of the haze $\mathrm{PM}_{2.5}$ collected from BJ, XA, XM and HK. (b) Correlation of oxidative potentials of the haze $\mathrm{PM}_{2.5}$ to IL-6, $\mathrm{IFN-}Y$ and TNF- $\cdot a$ . Carbon black (CB) was served as negative control. $^{*}\mathfrak{p}<0.05$ compared with control; $^{\#}\mathfrak{p}<0.05$ compared between groups.
|
|
atmosphere
|
0026
|
10.1016/j.atmosenv.2015.11.055
|
image
|
Fig. 3. Correlations of DCFH (oxidative potential), IL-6, $\mathrm{IFN-}Y$ and TNF- $\cdot\mathfrak{a}$ to the chemical compositions (sulfate, nitrate, ammonium, OC, EC, amine, urea and levoglucosan). $^{*}\mathtt{p}<0.05$ .
|
|
atmosphere
|
0026
|
10.1016/j.atmosenv.2015.11.055
|
text
| Not supported with pagination yet
|
Chemical composition and bioreactivity of PM2.5 during 2013 haze events in China
Kin-Fai $\mathsf{H o}^{\;\mathsf{a},\,\mathsf{b},\;^{*}}$ , Steven Sai Hang Ho b, c, Ru-Jin Huang b, d, e, Hsiao-Chi Chuang f , g, \*\* Jun-Ji Cao b, h, Yongming Han b, Ka-Hei Lui a, Zhi Ning i, Kai-Jen Chuang j, k, Tsun-Jen Cheng l, m, Shun-Cheng Lee n, Di Hu o, Bei Wang p, Renjian Zhang q
a The Jockey Club School of Public Health and Primary Care, The Chinese University of Hong Kong, Hong Kong, China
b Key Laboratory of Aerosol Chemistry and Physics, SKLLQG, Institute of Earth Environment, Chinese Academy of Sciences, Xi'an, 710061, China
c Division of Atmospheric Sciences, Desert Research Institute, Reno, NV 89512, USA
d Laboratory of Atmospheric Chemistry, Paul Scherrer Institute, 5232 Villigen PSI, Switzerland
e Centre for Climate and Air Pollution Studies, Ryan Institute, National University of Ireland Galway, University Road, Galway, Ireland
f School of Respiratory Therapy, College of Medicine, Taipei Medical University, Taipei, Taiwan
g Division of Pulmonary Medicine, Department of Internal Medicine, Shuang Ho Hospital, Taipei Medical University, Taipei, Taiwan
h Institute of Global Environmental Change, Xi’an Jiaotong University, Xi'an, 710061, China
i School of Energy and Environment, City University of Hong Kong, Hong Kong, China
j School of Public Health, College of Public Health and Nutrition, Taipei Medical University, Taipei, Taiwan
k Department of Public Health, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
l Institute of Occupational Medicine and Industrial Hygiene, College of Public Health, National Taiwan University, Taipei, Taiwan
m Institute of Environmental Health, College of Public Health, National Taiwan University, Taipei, Taiwan
n Department of Civil and Environmental Engineering, The Hong Kong Polytechnic University, Hong Kong, China
o Department of Chemistry, Hong Kong Baptist University, Kowloon Tong, Kowloon, Hong Kong, China
p Faculty of Science and Technology, Technological and Higher Education Institute of Hong Kong, Hong Kong, China
q Key Laboratory of Regional Climate-Environment for Temperate East Asia(RCE-TEA), Institute of Atmospheric Physics, Chinese Academy of Sciences, China
h i g h l i g h t s
Significant increases of sulfate, nitrate and ammonium were observed during episodes.
High contributions of biomass burning emissions to organic carbon (OC) were estimated in this study.
BJ $\mathrm{PM}_{2.5}$ samples has the highest bioreactivity although $\mathrm{PM}_{2.5}$ levels are not the highest.
The OC, urea and levoglucosan are associated with oxidative-inflammatory responses.
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 10 September 2015
Received in revised form
24 November 2015
Accepted 24 November 2015
Available online 2 December 2015
Keywords:
$\mathrm{PM}_{2.5}$
Oxidative potential
Haze event
Amines
China
Chemical composition and bioreactivity of $\mathrm{PM}_{2.5}$ samples collected from Beijing (BJ), Xi'an (XA), Xiamen (XM) and Hong Kong (HK) in China during haze events were characterized. $\mathrm{PM}_{2.5}$ mass concentrations in BJ, XA, XM and HK in the episodes were found to be $258\pm100\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ , $233\pm52~\upmu\mathrm{g}~\mathrm{m}^{-3}$ , $46\pm9~\upmu\mathrm{g}~\mathrm{m}^{-3}$ and $48\pm13~\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , respectively. Significant increase of sulfate, nitrate and ammonium concentrations in northern cities were observed. High contributions of biomass burning emissions to organic carbon (OC) in northern cities were estimated in this study implying frequent biomass burning during the haze periods. The urea concentrations in $\mathrm{PM}_{2.5}$ were $1855\,\pm\,755\,\mathrm{\ng\m}^{-3}$ (BJ), $1124\,\pm\,243\,\mathrm{\ng\m}^{-3}$ (XA), $543\pm104\mathrm{\,ng\,m}^{-3}$ (XM) and $363\pm61\mathrm{{\;ng\,m^{-3}}}$ (HK) suggesting higher or close to upper limits compared to other regions in the world. Dose-dependent alterations in oxidative potential, IL-6, IFN- $\cdot\gamma$ and TNF- $\cdot\alpha$ levels were also investigated. The oxidative potential levels are $\mathrm{BJ}>\mathrm{XM}>\mathrm{XA}>\mathrm{HK}$ , whereas levels of IL6, IFN- $\cdot\gamma$ and TNF- $^\circ\cdot$ were $\mathrm{BJ}>\mathrm{XA}>\mathrm{XM}>\mathrm{HK}.$ The sulfate, nitrate, ammonium, OC, urea and levoglucosan are associated with oxidative-inflammatory responses. These experimental results are crucial for the policymakers to implement cost-effective abatement strategies for improving air quality. $\circledcirc$
2015 Elsevier Ltd. All rights reserved.
1. Introduction
Very recent studies report that more than 2 million premature deaths around the world each year are associated with anthropogenic $\mathsf{P M}_{2.5}$ (particulate matter with an aerodynamic diameter ${<}2.5\ \upmu\mathrm{m}\rangle$ related cardiopulmonary diseases and lung cancer (Silva et al., 2013). Numerous epidemiological and toxicological studies have also shown correlation between exposure to particulate matter (PM) and adverse health effects (Brunekreef and Holgate, 2002; Pope et al., 2002; Pun et al., 2014). The underlying mechanism of particle-induced health effects is believed to be driven by the production of reactive oxygen species (ROS, e.g. superoxide $(\bullet0^{2-})$ , hydrogen peroxide $\left({\mathrm{H}}_{2}0_{2}\right)$ , and hydroxyl radical ( OH)) and the interaction of PM with epithelial cells and macrophages in the lung environment (Li et al., 2003; Nel, 2005). Adverse human health effects occur when an overproduction of ROS impacts on the body's anti-oxidative defenses (oxidative stress, OS) leading to cell dysfunction, inflammation and cardio-pulmonary disease (BeruBe et al., 2007). The oxidative stress can activate signaling pathways leading to the release of pro-inflammatory mediators (e.g. interleukin 6 (IL-6), interleukins 8 (IL-8), tumor necrosis factor a (TNF-a) and interferon $\gamma\ (\mathrm{IFN-\gamma})$ ) (Mitschik et al., 2008). Previous studies have focused on airborne particles with an aerodynamic diameter smaller than $2.5\ \upmu\mathrm{m}$ $(\mathsf{P M}_{2.5})$ because these fine particles can penetrate into the airways of the respiratory tract, reaching the alveoli and diffusing to other extrapulmonary target organs (Semmler et al., 2004). It is found that the size of particles is an important parameter in inducing cardiovascular and respiratory effects (Boldo et al., 2011). The ultrafine and fine particles are more potent than coarse particles on per mass basis (Cho et al., 2005; Ntziachristos et al., 2007). Besides particle size, there is growing evidence that the chemical composition of particles is also an important factor mediating cellular oxidative stress (Daher et al., 2014; Yang et al., 2014).
The $\mathsf{P M}_{2.5}$ concentrations observed in Chinese urban air are often one to two orders of magnitude higher than those observed in urban areas in the US and European countries (Huang et al., 2014b). Despite such high level of particulate pollution, studies related to particle toxicities are still very scarce in China (Deng et al., 2013; Huang et al., $20144$ ; Wei et al., 2011; Xu and Zhang, 2004). Especially, the toxicity of particles during haze events (visibility range is less than $10\;\mathrm{km}$ , $\mathrm{RH}<80\%$ ) is not clear. The objective of this study was to investigate the toxicity of particles collected during severe haze events in China during JanuaryeFebruary 2013. These haze events received worldwide media attention, with daily $\mathsf{P M}_{2.5}$ concentration higher than $700~\upmu\mathrm{g}~\:\mathrm{m}^{-3}$ . The $\mathsf{P M}_{2.5}$ samples were collected in four Chinese megacities with the total mass, organic and elemental carbon (OC and EC) content being analyzed in order to determine the $\mathsf{P M}_{2.5}$ cytotoxicity. Certain organic compounds such as levoglucosan, water soluble organic nitrogen (WSON) species (e.g. amino acids, amines and urea) were also measured in the aerosol samples. WSON consists of a broad array of nitrogencontaining organic species that are derived from both anthropogenic and natural emissions (Cornell et al., 2003; Neff et al., 2002). There are concerns about toxicity of organic nitrogen compounds such as nitrophenols (Natangelo et al., 1999), nitrated polycyclic aromatic hydrocarbons and other N-containing combustion products or industrial emissions which are present in the atmosphere (Albinet et al., 2008; Cheng et al., 2006) and how all of these compounds have effects on human health. Therefore, the $\mathsf{P M}_{2.5}$ toxicity was tested in vitro in human alveolar epithelial A549 cells that are considered as the relevant target cells. PM biological activities were characterized by measuring the expression of a panel of biomarkers. The IL-6, TNF- $\cdot\alpha$ and IFN- $\cdot\gamma$ were further used as effective biomarkers for investigating PM exposure to the occurrence of oxidative stress and pro-inflammatory responses. The above biological results were combined with chemical analysis to elucidate the difference in aerosol bioactivities during haze episodes in four Chinese megacities.
2. Materials and methods
2.1. $P M_{2.5}$ collection
Beijing (BJ), Xi'an (XA), Xiamen (XM) and Hong Kong (HK) were chosen for sample collection and were classified as northern (BJ and XA) and southern (XM and HK) Chinese cities in order to represent high and low $\mathsf{P M}_{2.5}$ exposure levels during the haze periods in this study (Fig. 1). The $\mathsf{P M}_{2.5}$ samples were collected over a range of six to eight days sampling campaign during the haze pollution period from the end of January until the beginning of February in 2013 (26th January to 2nd Feburary, 2013). The $\mathsf{P M}_{2.5}$ samples were collected on quartz fiber filters (QM/A, Whatman Inc., Clifton, NJ, USA, $8\,\mathrm{inch}\times10\,\mathrm{i}$ nch) using a high-volume sampler at a flow rate of $1.05{-}1.16\;\mathrm{m}^{3}\;\mathrm{min}^{-1}$ for bulk chemical analysis. Quartz filters were pre-baked at $800\,^{\circ}\mathrm{C}$ for $^{3\,\mathrm{h}}$ prior to sampling. The $\mathsf{P M}_{2.5}$ particulates were collected on Teflon filters (Pall Life Sciences, Ann Arbor, MI, $47{-}\mathrm{mm}^{\prime}$ ) using mini-volume samplers equipped with $\mathsf{P M}_{2.5}$ impactors (Airmetrics, OR, USA) at a flow rates of $5\,\mathrm{L/min}$ and underwent the biological testings described below. A preliminary sampling test was conducted using $\mathsf{P M}_{2.5}$ high-volume and $\mathsf{P M}_{2.5}$ mini-volume samplers in parallel at Xi'an city sampling location prior to the field study. The test results showed no significant mass differences in using high-volume and mini-volume samplers. The mass concentration data collected by the mini-volume samplers were used throughout this study. Samples were collected starting at 10:00 am. each day over a $24~\mathrm{h}$ interval at the four sampling locations. All $\mathsf{P M}_{2.5}$ Teflon filters were equilibrated on a temperature of $25\pm1\,^{\circ}\mathrm{C}$ and $40\pm5\%$ relative humidity for $48\,\mathrm{h}$ before and after $\mathsf{P M}_{2.5}$ mass concentration analysis. A microbalance (Sartorius Model MC5 Microbalance, Go€ttingen, Germany) with $1\,\upmu\mathrm{g}$ precision was used for the $\mathsf{P M}_{2.5}$ mass concentration measurements. All of the filters were stored at $-20~^{\circ}\mathrm{C}$ and in the dark prior to the analysis. Sample blanks were collected and analyzed along with the samples.
2.2. Extraction procedures
Teflon filters were extracted with methanol for biological analysis and quartz filters were extracted with either methanol or ultrapure water for different chemical analysis.
2.3. Methanol extraction
Each quartz sample filter $(4.3\ \mathrm{cm}^{2})$ ) was extracted with $20~\mathrm{ml}$ methanol and ultrasonicated in a constant temperature $(25\ ^{\circ}\mathrm{C})$ water bath for $20~\mathrm{min}$ . A few drops of toluene were added to the extractant as a volatile compound trapping agent subsequently after the methanol addition. The extractant was transferred to a round bottom flask and evaporated by the rotary evaporator until $5\;\mathrm{ml}$ of sample was conserved. The aerosol extractants were stored at $-20\,^{\circ}\mathrm{C}$ before amine analysis.
2.4. Water extraction
Each quartz sample filter $(4.3\;\mathrm{cm}^{2})$ ) was extracted with $5~\mathrm{mL}$ of Milli-Q water $.18\ \mathrm{M\Omega\cm})$ and ultrasonicated in a constant temperature $(25\ ^{\circ}\mathrm{C})$ water bath for $30\ \mathrm{min}$ . The extraction procedure was repeated. The first and second extractant was combined and subsequently filtered by a syringe filters $(0.45\,\upmu\mathrm{m}$ , Pall Corporation, NY, USA) to remove insoluble materials. The filtered solution was stored at $4\ {}^{\circ}\mathbf{C}$ before water-soluble ions and anhydrosugars analysis. The filtered water extractant was pre-concentrated to $0.5\ \mathrm{ml}$ before amino acids and urea analysis (Yang et al., 2005).
2.5. Methanol extraction for biological analysis
Each Teflon sample filter was extracted with $20~\mathrm{ml}$ methanol and ultrasonicated in a constant temperature $(25\,^{\circ}\mathrm{C})$ water bath for $20\,\mathrm{\min}$ . The procedure was repeated and the first and second extractant were combined. The combined methanol extractant was purged by nitrogen gas $(\mathsf{N}_{2}\,\geq\,99.995\%)$ for $60\,\mathrm{\min}$ until all the methanol solvent was removed (Lee et al., 2014). The residual particulate matter was re-dissolved in dimethyl sulfoxide (DMSO) $[<\!0.01\%$ vol in phosphate-buffered saline (PBS)] at 0 (control), 50 and $150~\upmu\mathrm{g/ml}$ for biological assays. Near-pure, manufactured, carbon black (CB), with an average diameter of $65~\mathrm{nm}$ (Monarch 120; Cabot Corporation, UK), was selected as a control particle (Chiang et al., 2013; Chuang et al., 2011b). The chemical characteristics of CB have been described previously (Chuang et al., 2011a; Zhu et al., 2004).
2.6. Chemical analysis
Water-soluble ions and anhydrosugars (e.g., levoglucosan and mannosan) were separated and identified by high-performance anion exchange chromatography coupled with pulsed amperometric detection (HPAEC-PAD). The detailed description of the analytical method is found in (Engling et al., 2006; Iinuma et al., 2009). The urea compounds in the water extractant was analyzed by the HPLC (Aglient 1200 system) coupled with a photodiode array detector (DAD). The absorbance wavelength $210\,\mathrm{nm}$ was applied for the identification and quantification of the urea samples. After using dansyl chloride (in acetone) as the derivatization reagent, twenty-two amines derivatized samples were determined by HPLC coupled to an ion trap mass spectrometer for the detection.
OC and EC were analyzed (on a $0.526~\mathrm{cm}^{2}$ punch) by thermal analysis with optical detection following the IMPROVE protocol on a Desert Research Institute (DRI) Model 2001 Thermal/Optical Carbon Analyzer (Atmoslytic Inc., Calabasas, CA, USA) (Cao et al., 2003; Chow et al., 2005). The MDL for the carbon analysis were 0.8 and $0.4~{\upmu\mathrm{g}}{\mathsf{C}}~{\mathrm{cm}}^{-2}$ for the OC and EC, respectively, with a precision better than $10\%$ for total carbon (TC).
2.7. DCFH assay
ROS production was determined by using a $2^{\prime}.7^{\prime}$ -dichlorodihydrofluorescein diacetate (DCFH-DA; SigmaeAldrich, UK) probe with serum-containing DMEM, which resulted in the cleavage of the diacetate groups by esterase enzymes to produce the relatively lipid insoluble and non-fluorescent dichlorodihydrofluorescein (DCFH). The fluorescent $^{2^{\prime},7^{\prime}}$ -dichlorofluorescein (DCF) was produced after oxidation of DCFH with ROS, which was measured using a FLUOstar fluorescence plate reader (BMG, Germany) at an excitation wavelength of $485~\mathrm{nm}$ and an emission wavelength of $530~\mathrm{nm}$ . The result was presented as fluorescence intensity (AU) (Chuang et al., 2011b).
2.8. Cell culture and treatment
A549 cells were obtained from the American Type Culture Collection and cultured in RPMI ( $10\%$ foetal bovine serum, penicillin and streptomycin) under the conditions of $37\,^{\circ}\mathrm{C},$ $95\%$ humidity and $5\%$ $C0_{2}$ . A549 cells were seeded onto surface-treated, 24-well transwells $(1~\times~10^{5}~\mathrm{cells/ml}$ ; BD Biosciences, UK) and incubated for $24~\mathrm{h}$ . The cells were then incubated with $300~\upmu\mathrm{l}$ of sample at particle concentrations of 0 (control), 50 and $150\,\upmu\mathrm{g/ml}$ for $4\,\mathrm{h}$ . Each experiment was conducted in quadruplicate. The concentrations were chosen to produce oxidative and inflammatory effects $(>\!80\%$ cell viability), according to criteria described previously (Chuang et al., 2012; Wilson et al., 2002).
2.9. Determination of cytokines
An ELISA (BD OptEIATM set, BD Biosciences, USA) was used to determine IL-6, TNF- $\cdot\alpha$ and IFN- $\cdot\gamma$ levels according to the manufacturer's instructions.
2.10. Statistical analysis
The samples analyzed in the experimental Sections 2.2e2.6 were repeated in triplicate. The data were expressed in the mean $\pm$ standard deviation (SD). One-way analysis of variance (ANOVA) was applied for multiple values comparison. Pearson's correlation coefficient analysis was performed to identify the correlation between (1) oxidative stress and inflammation and (2) chemical species with oxidative-inflammation cytokines. The statistical analyses were conducted using GraphPad Prism software (Version 5 for Windows). The significance level was set at ${\tt p}<0.05$ .
3. Results and discussion
3.1. Major components of $P M_{2.5}$
Table 1 shows the chemical components of the $\mathsf{P M}_{2.5}$ samples collected during the haze events from the 26th January to 2nd Feburary in 2013. Significant differences in $\mathsf{P M}_{2.5}$ concentration was observed between north and south China (4e5 fold). The average $\mathsf{P M}_{2.5}$ mass concentrations in BJ, XA, XM and HK were found to be $258~\pm~100~\upmu\mathrm{g}~\:\mathrm{m}^{-3}$ , $233\;\pm\;52\;\upmu\mathrm{g}\;\;\mathrm{m}^{-3}$ , $46~\pm~9~\upmu\mathrm{g}~\:\mathrm{m}^{-3}$ and $48\,\pm\,13\,\mathrm{\\upmug\,\m}^{-3}$ , respectively. Most of the results measured at northern cities are higher than observable mass concentration in the previous roadside and ambient studies in China (Cao et al., 2007; Jahn et al., 2013; Yang et al., 2011). Sulfate, nitrate and OC are the major constituents of $\mathsf{P M}_{2.5}$ at the BJ, XA and XM sampling locations, contributing up to $74\%$ of total $\mathsf{P M}_{2.5}$ mass. In HK, sulfate, OC and EC are the major constituents of $\mathsf{P M}_{2.5}$ (Ho et al., 2003). The detailed OC and EC results were reported previously at elsewhere. The sulfate concentrations in BJ and XA are on average 3.3 to 9.5 times higher than those in XM and HK, indicating a significant sulfate source in the northern cities (e.g., residential coal combustion). Similar trends were observed for the nitrate and ammonium between the northern and southern cities. The concentrations of nitrate in BJ and XA are $3.3{-}24\$ times higher than in XM and HK, while the concentrations of ammonium in BJ and XA are 5e83 times higher than in XM and HK. The gas-to-particle partition of nitrate and ammonium depends strongly on temperature (T), relative humidity (RH), and gaseous precursor concentrations. Relative low concentrations of nitrate and ammonium in the southern cities could be a consequence of thermal decomposition of ammonium nitrate at relatively high temperature and hence the release of gaseous nitric acid and ammonium.
3.2. Levoglucosan and organic nitrogen species in $P M_{2.5}$
Levoglucosan, the thermal degradation product of cellulose (Simoneit et al., 1999), has been used as an organic tracer for biomass burning derived aerosols for many years (Ho et al., 2014). The concentrations of levoglucosan range from 6 to $870\,\mathrm{ng}\,\mathrm{m}^{-3}$ and follow an ascending order $\mathrm{HK}<\mathrm{XM}<\mathrm{BJ}<\mathrm{XA}$ at these four cities. High levoglucosan concentrations at the northern cities (i.e., $653\pm191~\mathrm{\bar{n}g}\,\mathrm{m}^{-3}$ at XA and $359\pm119\,\mathrm{ng\,m}^{-3}$ at BJ) indicate strong local biomass burning activities and/or regional transport of biomass burning particles. The relatively low values in Hong Kong could be explained by the lack of such activities in the city. The levoglucosan to OC ratio (LG/OC) is used to estimate contributions from biomass burning to OC mass (Ho et al., 2014; Puxbaum et al., 2007; Zhang et al., 2010). A simplified receptor-based approach was used to estimate biomass burning contributions at these four cities:
contribution of biomass burning emissions to OC
$$
=(\mathrm{LG}/\mathrm{OC})_{\mathrm{measured}}\Big/(\mathrm{LG}/\mathrm{OC})_{\mathrm{refe}}
$$
while “ $\mathrm{\Delta[LG/OC)}$ reference” refers to the emission factor of levoglucosan based on OC emissions derived from biomass burning chamber studies (Zhang et al., 2007). The contributions of biomass burning emissions to OC at XA $(28.2\%)$ and BJ $(14.5\%)$ were significantly higher than those at XM $(9.0\%)$ and HK $(5.5\%)$ , consistent with the more common and intensive biomass burning activities in northern cities. In previous studies, the effects of short-term exposure to biomass burning emission on respiratory symptoms, lung function, asthma and daily mortality were observed (Boman et al., 2003; Saffari et al., 2013).
Table 1 shows the concentrations of 21 organic nitrogen species in $\mathsf{P M}_{2.5}$ samples. The concentrations of urea are much higher in northern China $^{\,\prime}1855\pm755~\mathrm{ng}~\mathrm{m}^{-3}$ at BJ and $1124\pm243~\mathrm{ng~m}^{-3}$ at XA) than in southern China $(543~\pm~104~\mathrm{{\scriptsize~ng}~}\mathrm{{\scriptsize~m}}^{-3}$ at XM and $363\pm61~\mathrm{ng~m}^{-3}$ at HK). The levels of urea measured in this study are consistent with previous studies. For example, Shi et al. (2010) measured $1188~\mathrm{ng~m}^{-3}$ and $636\;\mathrm{ng}\;\mathrm{m}^{-3}$ of urea in total suspended particulate during dust episodes and non-dust periods at Qingdao in North China. This shows a worrying high level of urea in urban atmosphere of China when compared with other countries, e.g. the average concentration of urea in TSP at Hawaii was $78{-}378~\mathrm{ng}~\mathrm{m}^{-3}$ in dirty samples and $42{-}132\,\mathrm{ng}\,\mathrm{m}^{-3}$ in clean samples (Cornell et al., 2001). The use of agricultural fertilizers could be a key anthropogenic source of urea (Glibert et al., 2005) with high urea concentration being found in BJ and XA sampling locations. China is a global player of manufacturing and consuming urea with production of $1.9\,\times\,10^{7}$ tons in 2004. However, only $30{-}35\%$ of the urea fertilizer was efficiently utilized compared to $70–80\%$ in Europe and in North America (Wang, 2004). Urea of agricultural origin can enter the atmosphere by wind-blown and biomass burning. The possible primary emission source of urea in China could be due to active agricultural activities, soil dust and the sea surface through cavitation processes (Shi et al., 2010; Violaki and Mihalopoulos, 2010). The secondary production of urea from atmospheric reactions could also be additional major contributing sources. All of these possible pathways could lead to high urea concentrations in the four sampling locations.
The total concentration of 20 amines measured in $\mathsf{P M}_{2.5}$ range from 35 to $212~\mathrm{ng~m}^{-3}$ . The BJ samples contain the highest levels $(133\,\pm\,53\,\mathrm{\ng\m}^{-3})$ ) whereas the HK samples contain the lowest concentrations $(46\;\pm\;10\;\mathrm{\ng\m}^{-3})$ during the haze periods. Nmethylformamide, methylamine and ethylamine are the three most abundant amines in all of the $\mathsf{P M}_{2.5}$ samples at the four locations. Amines are strong bases in the atmosphere and can rapidly react with nitric acid $\left(\mathrm{HNO}_{3}\right)$ and sulfuric acid $\mathrm{(H_{2}S O_{4})}$ via acidebase reactions leading to secondary aerosol formation. The atmospheric reaction pathways of amines include gas-phase reactions with oxidants such as OH , $\mathsf{N O}_{3}$ and $0_{3}$ , to form amides, nitramines, and imines, which can also partition to the particle phase (Murphy et al., 2007; Schade and Crutzen, 1995; Silva et al., 2008). For example, N-methylformamide is formed from the OH and $0_{3}$ reaction with dimethylamine, and possibly OH reaction with methyldiethanolamine. More than one hundred amine species have been observed in the aerosol particles (Ozel et al., 2010), contributing $10{-}20\%$ of the organic content in ambient particles (Docherty et al., 2011; Hildebrandt et al., 2011). Amines originate from various primary sources including animal husbandry, industry and combustion, composting operations, automobiles, and natural sources such as biomass burning and biodegradation of organic matter that contains proteins or amino acids (Ge et al., 2011). The total emissions of methylamine, dimethylamine and trimethylamine are estimated to be ${\sim}300\mathrm{\Gg}\mathrm{\N\,}{\mathsf{a}}^{-1}$ globally (Ge et al., 2011).
Although large numbers of organic nitrogen species in the atmosphere were determined, only limited studies of their health effects were investigated, e.g. the potential symptoms of exposure to aliphatic and aromatic amines include: respiratory irritation; drowsiness and headache; cough, sneezing, wheezing, and dyspnea etc (Ge et al., 2011). Moreover, amines 2-naphthylamine, benzidine, and 4-aminobiphenyl are proven human carcinogens (Pinches and Walker, 1980).
3.3. Oxidative potential and inflammatory response
potential, IL-6, $\mathrm{IFN-}Y$ and TNF- $\cdot\alpha$ levels. The levels of oxidative potential are in descending order B $\mathrm{\DeltaJ>XM>XA>HK}$ at $150\,\upmu\mathrm{g/ml}$ . The levels of IL-6, IFN- $\cdot\gamma$ and TNF- $\cdot\alpha$ is in descending order $\mathsf{B J}>\mathsf{X A}>\mathsf{X M}>\mathsf{H K}$ in at $150\ \upmu\mathrm{g/ml}$ . The results show that BJ $\mathsf{P M}_{2.5}$ samples have the highest bioreactivity although the $\mathsf{P M}_{2.5}$ levels are not the highest among the four cities. Pearson's correlation coefficient test was applied to elucidate the correlation between oxidative potential and inflammation induced by the $\mathsf{P M}_{2.5}$ exposure. The test shows the oxidative potential is highly correlated with the IL-6 $\mathrm{r}^{2}=0.78$ , $\mathsf{p}<0.05\mathrm{;}$ (Figure 2b) and TNF- $\cdot\alpha$ production $(\mathbf{r}^{2}=0.78$ $\mathsf{p}<0.05\rangle$ (Fig. 2 (b)), while the correlation between oxidative stress and IFN- $\cdot\gamma$ is moderate $(\mathfrak{r}^{2}\,=\,0.47,\,\mathfrak{p}\,<\,0.05)$ . Oxidative stress is recognized to be an important factor in the regulation of inflammatory response (Danielsen et al., 2010; Luo et al., 2009). The oxidative stress induced by the PM depends on the physical properties and chemical composition of the particles. For example, reactive oxygen species (ROS) can be generated from the surface of particles or by the particle itself. The PM can provide a platform for intermixing with various ROS produced, for example by the organics (Chuang et al., 2012) and metals. Such exogenous ROS from PM can alter functions of mitochondria or NADPH-oxidase, leading to inflammation (Risom et al., 2005).
3.4. Correlation between chemicals and oxidative-inflammatory responses
The correlations between chemical composition per unit $\upmu g$ of the haze $\mathsf{P M}_{2.5}$ and oxidative-inflammatory responses were calculated using the Pearson's correlation coefficient test and the results are shown in Fig. 3. Good correlations are observed for the oxidative potential (determined by DCFH) against sulfate $(\mathsf{r}^{2}\!=\!0.70,\mathsf{p}<0.05)$ , nitrate $(\mathsf{r}^{2}=0.57,\,\mathsf{p}<0.05)$ , ammonium $\left[\mathrm{r}^{2}=0.59\right]$ , $\mathsf{p}<0.05]$ , OC $(\mathrm{r}^{2}=0.76$ , $\mathsf{p}<0.05\mathrm{)}$ , urea $\mathrm{T}^{2}=0.61$ , $\mathsf{p}<0.05$ ) and levoglucosan $\mathrm{~\,~r~}^{2}\,=\,0.60$ , $\mathsf{p}\,<\,0.05^{}.$ . IL-6 is positively correlated with sulfate $\operatorname{r}^{2}=0.68$ , $\mathsf{p}\,<\,0.05\mathrm{)}$ , nitrate $\mathrm{~\,~r~}^{2}\,=\,0.69$ , $\mathsf{p}\,<\,0.05\mathrm{;}$ , ammonium $\mathrm{r}^{2}=0.75$ , $\mathsf{p}<0.05\mathrm{;}$ and levoglucosan $\mathrm{r}^{2}=0.76$ , $\mathsf{p}<0.05^{\prime}$ ). IFN- $\cdot\gamma$ is positively correlated with sulfate $\mathbf{\epsilon}(\mathbf{r}^{2}=0.72$ , $\mathsf{p}\,<\,0.05\mathrm{~}$ , nitrate $\bar{(\mathbf{r}^{2}}=0.\dot{6}0$ , $\textsf{p}<\,0.05)$ , ammonium $(\mathbf{r}^{2}\,=\,0.58$ , $\textsf{p}<\,0.05)$ , urea $(\mathsf{r}^{2}\,{=}\,0.60,\mathsf{p}\,{<}\,0.05)$ and levoglucosan $(\mathsf{r}^{2}\,{=}\,0.62,\mathsf{p}<0.05)$ . TNF- $\cdot\alpha$ is positively correlated with sulfate $(\mathsf{r}^{2}=0.58,\,\mathsf{p}<0.05)$ , ammonium $\mathrm{\dot{r}}^{2}=0.54$ , $\mathsf{p}<0.05\mathrm{;}$ ) and OC $\mathrm{f}^{2}=0.59$ , $\mathsf{p}<0.05^{\prime}$ ). According to the Windrose plot in Fig. 3, the sulfate, nitrate, ammonium, OC, urea and levoglucosan are statistically associated with oxidativeinflammatory responses. Since $\mathsf{P M}_{2.5}$ have the capability of penetrating deep into the lung and even entering the blood circulation the chemical components carried by the $\mathsf{P M}_{2.5}$ are important parameters in revealing the cytotoxicity and bioreactivity (Chuang et al., 2012). Atmospheric carbonaceous PM is an important component of $\mathsf{P M}_{2.5}$ , and its major fractions are EC and OC. EC is emitted directly from combustion sources and has been considered an indicator for the primary anthropogenic air pollutants. However, OC is produced from volatile organic compound gas-to-particle conversion processes occurring in the atmosphere. Previous studies have shown that diesel exhaust particles contain a solid carbon core (soot/black carbon) that can intermix with chemical compounds and lead to adverse health effects. Physicochemicalspecific relationships in inflammatory initiation were observed. A blunting of the endothelium-dependent and -independent vasodilatation has been reported in relation to black carbon and sulfate (O'Neill et al., 2005). Sulfate and OC have also been reported to be capable of initiating endothelial dysfunction (Lei et al., 2005). Changes in the cardiovascular reaction in response to $\mathrm{PM}_{2.5},\,\mathrm{NO}_{3}^{-}$ , OC and EC have been revealed in human (Chuang et al., 2007) and in vivo studies (Rohr et al., 2011), which could be associated with the oxidative-inflammatory reaction resulted from the particle. Urea is associated with oxidative stress and $\mathrm{IFN-}Y$ production, which is in line with previous findings (D'Apolito et al., 2010; Zhang et al., 1999). Urea-induced ROS has been associated with mitochondrial ROS production and the mitochondrial manganese superoxide dismutase (MnSOD) pathway (D'Apolito et al., 2010). Levoglucosan has been used as a urinary biomarker after the exposure to wood smoke (Hinwood et al., 2008). However, its potential toxicity in response to pulmonary exposure remains to be determined. The haze $\mathsf{P M}_{2.5}$ -containing levoglucosan correlates with oxidative stress, IL-6 and $\mathrm{IFN-}Y$ production. A previous study demonstrated that ROS-induced by $\mathsf{P M}_{2.5}$ emitted from biomass burning in related to levoglucosan levels (Saffari et al., 2013), which is in agreement with our results. Levoglucosan is commonly generated from biomass burning with other co-emitted species; therefore, the toxicity of levoglucosan requires further investigation.
The chemical composition of $\mathsf{P M}_{2.5}$ during the haze events in 2013 at four Chinese megacities and the relevant $\mathsf{P M}_{2.5}$ toxicity with the different chemical species were investigated. Oxidative potential and inflammation induced by the haze $\mathsf{P M}_{2.5}$ exposure demonstrates that oxidative potential is highly correlated with different biomarkers of pro-inflammatory responses. Several inorganic and organic species are associated with oxidativeinflammatory responses. Further research should focus on identifying the cytotoxic and carcinogenic mechanisms of $\mathsf{P M}_{2.5}$ in human organs (e.g. lung). Our study therefore supports the development of more effective and source-specific regulations for mitigating PM production.
Acknowledgments
This study is partially supported by projects from the Research Grants Council of the Hong Kong Special Administrative Region China (Project No. 412612), “Strategic Priority Research Program” of the Chinese Academy of Science (XDA05100401) and Ministry of Science & Technology (201209007). The help of Mr. Chi Sing CHAN (The Chinese University of Hong Kong) on map drawing is acknowledged.
References
Albinet, A., Leoz-Garziandia, E., Budzinski, H., Villenave, E., Jaffrezo, J.L., 2008. Nitrated and oxygenated derivatives of polycyclic aromatic hydrocarbons in the ambient air of two French alpine valleys e Part 1: concentrations, sources and gas/particle partitioning. Atmos. Environ. 42, 43e54.
BeruBe, K., Balharry, D., Sexton, K., Koshy, L., Jones, T., 2007. Combustion-derived nanoparticles: mechanisms of pulmonary toxicity. Clin. Exp. Pharmacol. Physiology 34, 1044e1050.
Boldo, E., Linares, C., Lumbreras, J., Borge, R., Narros, A., Garcia-Perez, J., FernandezNavarro, P., Perez-Gomez, B., Aragones, N., Ramis, R., Pollan, M., Moreno, T., Karanasiou, A., Lopez-Abente, G., 2011. Health impact assessment of a reduction in ambient $\mathrm{PM}_{2.5}$ levels in Spain. Environ. Int. 37, 342e348.
Boman, B.C., Forsberg, A.B., Jarvholm, B.G., 2003. Adverse health effects from ambient air pollution in relation to residential wood combustion in modern society. Scand. J. Work Environ. Health 29, 251e260.
Brunekreef, B., Holgate, S.T., 2002. Air pollution and health. Lancet 360, 1233e1242.
Cao, J.J., Lee, S.C., Chow, J.C., Watson, J.G., Ho, K.F., Zhang, R.J., Jin, Z.D., Shen, Z.X., Chen, G.C., Kang, Y.M., Zou, S.C., Zhang, L.Z., Qi, S.H., Dai, M.H., Cheng, Y., Hu, K., 2007. Spatial and seasonal distributions of carbonaceous aerosols over China. J. Geophys. Res.-Atmos. 112.
Cao, J.J., Lee, S.C., Ho, K.F., Zhang, X.Y., Zou, S.C., Fung, K., Chow, J.C., Watson, J.G., 2003. Characteristics of carbonaceous aerosol in Pearl River Delta Region, China during 2001 winter period. Atmos. Environ. 37, 1451e1460.
Cheng, Y., Li, S.M., Leithead, A., 2006. Chemical characteristics and origins of nitrogen-containing organic compounds in PM2.5 aerosols in the Lower Fraser Valley. Environ. Sci. Technol. 40, 5846e5852.
Chiang, L.L., Chen, H.C., Lee, C.N., Chuang, K.J., Chen, T.T., Yeh, C.T., Wang, L.S., Lee, W.H., Lin, L.Y., Tseng, H.E., Chuang, H.C., 2013. Serum protein oxidation by diesel exhaust particles: effects on oxidative stress and inflammatory response in vitro. Chemico-Biological Interact. 206, 385e393.
Cho, A.K., Sioutas, C., Miguel, A.H., Kumagai, Y., Schmitz, D.A., Singh, M., EigurenFernandez, A., Froines, J.R., 2005. Redox activity of airborne particulate matter at different sites in the Los Angeles Basin. Environ. Res. 99, 40e47.
Chow, J.C., Watson, J.G., Chen, L.W.A., Paredes-Miranda, G., Chang, M.C.O., Trimble, D., Fung, K.K., Zhang, H., Yu, J.Z., 2005. Refining temperature measures in thermal/optical carbon analysis. Atmos. Chem. Phys. 5, 2961e2972.
Chuang, H.C., Fan, C.W., Chen, K.Y., Chang-Chien, G.P., Chan, C.C., 2012. Vasoactive alteration and inflammation induced by polycyclic aromatic hydrocarbons and trace metals of vehicle exhaust particles. Toxicol. Lett. 214, 131e136.
Chuang, H.C., Jones, T., Chen, Y., Bell, J., Wenger, J., BeruBe, K., 2011a. Characterisation of airborne particles and associated organic components produced from incense burning. Anal. Bioanal. Chem. 401, 3095e3102.
Chuang, H.C., Jones, T.P., Lung, S.C., BeruBe, K.A., 2011b. Soot-driven reactive oxygen species formation from incense burning. Sci. Total Environ. 409, 4781e4787.
Chuang, K.J., Chan, C.C., Su, T.C., Lee, C.T., Tang, C.S., 2007. The effect of urban air pollution on inflammation, oxidative stress, coagulation, and autonomic dysfunction in young adults. Am. J. Respir. Crit. Care Med. 176, 370e376.
Cornell, S., Mace, K., Coeppicus, S., Duce, R., Huebert, B., Jickells, T., Zhuang, L.Z., 2001. Organic nitrogen in Hawaiian rain and aerosol. J. Geophys. Res.-Atmos. 106, 7973e7983.
Cornell, S.E., Jickells, T.D., Cape, J.N., Rowland, A.P., Duce, R.A., 2003. Organic nitrogen deposition on land and coastal environments: a review of methods and data. Atmos. Environ. 37, 2173e2191.
D'Apolito, M., Du, X.L., Zong, H.H., Catucci, A., Maiuri, L., Trivisano, T., PettoelloMantovani, M., Campanozzi, A., Raia, V., Pessin, J.E., Brownlee, M., Giardino, I., 2010. Urea-induced ROS generation causes insulin resistance in mice with chronic renal failure. J. Clin. Investigation 120, 203e213.
Daher, N., Saliba, N.A., Shihadeh, A.L., Jaafar, M., Baalbaki, R., Shafer, M.M., Schauer, J.J., Sioutas, C., 2014. Oxidative potential and chemical speciation of size-resolved particulate matter (PM) at near-freeway and urban background sites in the greater Beirut area. Sci. Total Environ. 470, 417e426.
Danielsen, P.H., Loft, S., Jacobsen, N.R., Jensen, K.A., Autrup, H., Ravanat, J.L., Wallin, H., Moller, P., 2010. Oxidative stress, inflammation, and DNA damage in rats after intratracheal instillation or oral exposure to ambient air and wood smoke particulate matter. Toxicol. Sci. 118, 574e585.
Deng, X.B., Zhang, F., Rui, W., Long, F., Wang, L.J., Feng, Z.H., Chen, D.L., Ding, W.J., 2013. $\mathrm{PM}_{2.5}$ -induced oxidative stress triggers autophagy in human lung epithelial A549 cells. Toxicol. In Vitro 27, 1762e1770.
Docherty, K.S., Aiken, A.C., Huffman, J.A., Ulbrich, I.M., DeCarlo, P.F., Sueper, D., Worsnop, D.R., Snyder, D.C., Peltier, R.E., Weber, R.J., Grover, B.D., Eatough, D.J., Williams, B.J., Goldstein, A.H., Ziemann, P.J., Jimenez, J.L., 2011. The 2005 Study of Organic Aerosols at Riverside (SOAR-1): instrumental intercomparisons and fine particle composition. Atmos. Chem. Phys. 11, 12387e12420.
Engling, G., Carrico, C.M., Kreldenweis, S.M., Collett, J.L., Day, D.E., Malm, W.C., Lincoln, E., Hao, W.M., Iinuma, Y., Herrmann, H., 2006. Determination of levoglucosan in biomass combustion aerosol by high-performance anion-exchange chromatography with pulsed amperometric detection. Atmos. Environ. 40, S299eS311.
Ge, X.L., Wexler, A.S., Clegg, S.L., 2011. Atmospheric amines e Part I. A review. Atmos. Environ. 45, 524e546.
Glibert, P.M., Trice, T.M., Michael, B., Lane, L., 2005. Urea in the tributaries of the chesapeake and coastal bays of Maryland. Water Air Soil Pollut. 160, 229e243.
Hildebrandt, L., Kostenidou, E., Lanz, V.A., Prevot, A.S.H., Baltensperger, U., Mihalopoulos, N., Laaksonen, A., Donahue, N.M., Pandis, S.N., 2011. Sources and atmospheric processing of organic aerosol in the Mediterranean: insights from aerosol mass spectrometer factor analysis. Atmos. Chem. Phys. 11, 12499e12515.
Hinwood, A.L., Trout, M., Murby, J., Barton, C., Symons, B., 2008. Assessing urinary levoglucosan and methoxyphenols as biomarkers for use in woodsmoke exposure studies. Sci. Total Environ. 402, 139e146.
Ho, K.F., Engling, G., Ho, S.S.H., Huang, R.J., Lai, S.C., Cao, J.J., Lee, S.C., 2014. Seasonal variations of anhydrosugars in $\mathrm{PM}_{2.5}$ in the Pearl River Delta Region, China. Tellus Ser. B-Chemical Phys. Meteorology 66.
Ho, K.F., Lee, S.C., Chan, C.K., Yu, J.C., Chow, J.C., Yao, X.H., 2003. Characterization of chemical species in $\mathrm{PM}_{2.5}$ and $\mathrm{PM_{10}}$ aerosols in Hong kong. Atmos. Environ. 37, 31e39.
Huang, Q.Y., Zhang, J., Peng, S.Y., Tian, M.P., Chen, J.S., Shen, H.Q., 2014a. Effects of water soluble PM2.5 extracts exposure on human lung epithelial cells (A549): a proteomic study. J. Appl. Toxicol. 34, 675e687.
Huang, R.J., Zhang, Y.L., Bozzetti, C., Ho, K.F., Cao, J.J., Han, Y.M., Daellenbach, K.R., Slowik, J.G., Platt, S.M., Canonaco, F., Zotter, P., Wolf, R., Pieber, S.M., Bruns, E.A., Crippa, M., Ciarelli, G., Piazzalunga, A., Schwikowski, M., Abbaszade, G., Schnelle-Kreis, J., Zimmermann, R., An, Z.S., Szidat, S., Baltensperger, U., El Haddad, I., Prevot, A.S.H., 2014b. High secondary aerosol contribution to particulate pollution during haze events in China. Nature 514, 218e222.
Iinuma, Y., Engling, G., Puxbaum, H., Herrmann, H., 2009. A highly resolved anionexchange chromatographic method for determination of saccharidic tracers for biomass combustion and primary bio-particles in atmospheric aerosol. Atmos. Environ. 43, 1367e1371.
Jahn, H.J., Kraemer, A., Chen, X.C., Chan, C.Y., Engling, G., Ward, T.J., 2013. Ambient and personal PM2.5 exposure assessment in the Chinese megacity of Guangzhou. Atmos. Environ. 74, 402e411.
Lee, K.Y., Wong, C.K.C., Chuang, K.J., Bien, M.Y., Cao, J.J., Han, Y.M., Tian, L.W., Chang, C.C., Feng, P.H., Ho, K.F., Chuang, H.C., 2014. Methionine oxidation in albumin by fine haze particulate matter: an in vitro and in vivo study. J. Hazard. Mater. 274, 384e391.
Lei, Y.C., Hwang, J.S., Chan, C.C., Lee, C.T., Cheng, T.J., 2005. Enhanced oxidative stress and endothelial dysfunction in streptozotocin-diabetic rats exposed to fine particles. Environ. Res. 99, 335e343.
Li, N., Sioutas, C., Cho, A., Schmitz, D., Misra, C., Sempf, J., Wang, M.Y., Oberley, T., Froines, J., Nel, A., 2003. Ultrafine particulate pollutants induce oxidative stress and mitochondrial damage. Environ. Health Perspect. 111, 455e460.
Luo, J.C.J., Hsu, K.H., Shen, W.S., 2009. Inflammatory responses and oxidative stress from metal fume exposure in automobile welders. J. Occup. Environ. Med. 51, 95e103.
Mitschik, S., Schierl, R., Nowak, D., Jorres, R.A., 2008. Effects of particulate matter on cytokine production in vitro: a comparative analysis of published studies. Inhal. Toxicol. 20, 399e414.
Murphy, S.M., Sorooshian, A., Kroll, J.H., Ng, N.L., Chhabra, P., Tong, C., Surratt, J.D., Knipping, E., Flagan, R.C., Seinfeld, J.H., 2007. Secondary aerosol formation from atmospheric reactions of aliphatic amines. Atmos. Chem. Phys. 7, 2313e2337.
Natangelo, M., Mangiapan, S., Bagnati, R., Benfenati, E., Fanelli, R., 1999. Increased concentrations of nitrophenols in leaves from a damaged forestal site. Chemosphere 38, 1495e1503.
Neff, J.C., Holland, E.A., Dentener, F.J., McDowell, W.H., Russell, K.M., 2002. The origin, composition and rates of organic nitrogen deposition: a missing piece of the nitrogen cycle? Biogeochemistry 57, 99e136.
Nel, A., 2005. Air pollution-related illness: effects of particles. Science 308, 804e806.
Ntziachristos, L., Froines, J.R., Cho, A.K., Sioutas, C., 2007. Relationship between redox activity and chemical speciation of size-fractionated particulate matter. Part. Fibre Toxicol. 4.
O'Neill, M.S., Veves, A., Zanobetti, A., Sarnat, J.A., Gold, D.R., Economides, P.A., Horton, E.S., Schwartz, J., 2005. Diabetes enhances vulnerability to particulate air pollution e associated impairment in vascular reactivity and endothelial function. Circulation 111, 2913e2920.
Ozel, M.Z., Ward, M.W., Hamilton, J.F., Lewis, A.C., Raventos-Duran, T., Harrison, R.M., 2010. Analysis of organic nitrogen compounds in urban aerosol samples using GCxGC-TOF/MS. Aerosol Sci. Technol. 44, 109e116.
Pinches, M.A., Walker, R.F., 1980. Determination of atmospheric contaminants using a continuous paper-tape personal monitor .1. Analysis of aromatic-amines. Ann. Occup. Hyg. 23, 335e352.
Pope, C.A., Burnett, R.T., Thun, M.J., Calle, E.E., Krewski, D., Ito, K., Thurston, G.D., 2002. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. Jama-Journal Am. Med. Assoc. 287, 1132e1141.
Pun, V.C., Yu, I.T.S., Qiu, H., Ho, K.F., Sun, Z.W., Louie, P.K.K., Wong, T.W., Tian, L.W., 2014. Short-term associations of cause-specific emergency hospitalizations and particulate matter chemical components in Hong Kong. Am. J. Epidemiol. 179, 1086e1095.
Puxbaum, H., Caseiro, A., Sanchez-Ochoa, A., Kasper-Giebl, A., Claeys, M., Gelencser, A., Legrand, M., Preunkert, S., Pio, C., 2007. Levoglucosan levels at background sites in Europe for assessing the impact of biomass combustion on the European aerosol background. J. Geophys. Res.-Atmos. 112.
Risom, L., Moller, P., Loft, S., 2005. Oxidative stress-induced DNA damage by particulate air pollution. Mutat. Res.-Fundamental Mol. Mech. Mutagen. 592, 119e137.
Rohr, A.C., Kamal, A., Morishita, M., Mukherjee, B., Keeler, G.J., Harkema, J.R., Wagner, J.G., 2011. Altered Heart rate Variability in Spontaneously Hypertensive rats is associated with specific particulate matter components in Detroit, Michigan. Environ. Health Perspect. 119, 474e480.
Saffari, A., Daher, N., Samara, C., Voutsa, D., Kouras, A., Manoli, E., Karagkiozidou, O., Vlachokostas, C., Moussiopoulos, N., Shafer, M.M., Schauer, J.J., Sioutas, C., 2013. Increased biomass burning due to the economic crisis in Greece and its adverse impact on wintertime air quality in Thessaloniki. Environ. Sci. Technol. 47, 13313e13320.
Schade, G.W., Crutzen, P.J., 1995. Emission of aliphatic-amines from animal husbandry and their reactions e potential source of $\mathrm{N}_{2}\mathrm{O}$ and HCN. J. Atmos. Chem. 22, 319e346.
Semmler, M., Seitz, J., Erbe, F., Mayer, P., Heyder, J., Oberdorster, G., Kreyling, W.G., 2004. Long-term clearance kinetics of inhaled ultrafine insoluble iridium particles from the rat lung, including transient translocation into secondary organs. Inhal. Toxicol. 16, 453e459.
Shi, J.H., Gao, H.W., Qi, J.H., Zhang, J., Yao, X.H., 2010. Sources, compositions, and distributions of water-soluble organic nitrogen in aerosols over the China Sea. J. Geophys. Res.-Atmos. 115.
Silva, P.J., Erupe, M.E., Price, D., Elias, J., Malloy, Q.G.J., Li, Q., Warren, B., Cocker, D.R., 2008. Trimethylamine as precursor to secondary organic aerosol formation via nitrate radical reaction in the atmosphere. Environ. Sci. Technol. 42, 4689e4696.
Silva, R.A., West, J.J., Zhang, Y.Q., Anenberg, S.C., Lamarque, J.F., Shindell, D.T., Collins, W.J., Dalsoren, S., Faluvegi, G., Folberth, G., Horowitz, L.W., Nagashima, T., Naik, V., Rumbold, S., Skeie, R., Sudo, K., Takemura, T., Bergmann, D., Cameron-Smith, P., Cionni, I., Doherty, R.M., Eyring, V., Josse, B., MacKenzie, I.A., Plummer, D., Righi, M., Stevenson, D.S., Strode, S., Szopa, S., Zeng, G., 2013. Global premature mortality due to anthropogenic outdoor air pollution and the contribution of past climate change. Environ. Res. Lett. 8.
Simoneit, B.R.T., Schauer, J.J., Nolte, C.G., Oros, D.R., Elias, V.O., Fraser, M.P., Rogge, W.F., Cass, G.R., 1999. Levoglucosan, a tracer for cellulose in biomass burning and atmospheric particles. Atmos. Environ. 33, 173e182.
Violaki, K., Mihalopoulos, N., 2010. Water-soluble organic nitrogen (WSON) in sizesegregated atmospheric particles over the Eastern Mediterranean. Atmos. Environ. 44, 4339e4345.
Wang, W.S., 2004. General situation of global supply and demand of chemical fertilizers and present status of domestic chemical fertilizer industry (in Chinese). J. Chem. Fertilizer Industry 32, 1e5.
Wei, H., Wei, D., Yi, S., Zhang, F., Ding, W.J., 2011. Oxidative stress induced by urban fine particles in cultured EA.hy926 cells. Hum. Exp. Toxicol. 30, 579e590.
Wilson, M.R., Lightbody, J.H., Donaldson, K., Sales, J., Stone, V., 2002. Interactions between ultrafine particles and transition metals in vivo and in vitro. Toxicol. Appl. Pharmacol. 184, 172e179.
Xu, D.Q., Zhang, W.L., 2004. Monitoring of pollution of air fine particles $(\mathrm{PM}_{2.5})$ and study on their genetic toxicity. Biomed. Environ. Sci. 17, 452e458.
Yang, A., Jedynska, A., Hellack, B., Rooter, I., Hoek, G., Brunekreef, B., Kuhlbusch, T.A.J., Cassee, F.R., Janssen, N.A.H., 2014. Measurement of the oxidative potential of $\mathrm{PM}_{2.5}$ and its constituents: the effect of extraction solvent and filter type. Atmos. Environ. 83, 35e42.
Yang, F., Tan, J., Zhao, Q., Du, Z., He, K., Ma, Y., Duan, F., Chen, G., Zhao, Q., 2011. Characteristics of $\mathrm{PM}_{2.5}$ speciation in representative megacities and across China. Atmos. Chem. Phys. 11, 5207e5219.
Yang, H., Yu, J.Z., Ho, S.S.H., Xu, J.H., Wu, W.S., Wan, C.H., Wang, X.D., Wang, X.R., Wang, L.S., 2005. The chemical composition of inorganic and carbonaceous materials in $\mathrm{PM}_{2.5}$ in Nanjing, China. Atmos. Environ. 39, 3735e3749.
Zhang, Y.X., Shao, M., Zhang, Y.H., Zeng, L.M., He, L.Y., Zhu, B., Wei, Y.J., Zhu, X.L., 2007. Source profiles of particulate organic matters emitted from cereal straw burnings. J. Environ. Sciences-China 19, 167e175.
Zhang, Z., Yang, X.Y., Cohen, D.M., 1999. Urea-associated oxidative stress and Gadd153/CHOP induction. Am. J. Physiology-Renal Physiology 276, F786eF793.
Zhang, Z.S., Engling, G., Lin, C.Y., Chou, C.C.K., Lung, S.C.C., Chang, S.Y., Fan, S.J., Chan, C.Y., Zhang, Y.H., 2010. Chemical speciation, transport and contribution of biomass burning smoke to ambient aerosol in Guangzhou, a mega city of China. Atmos. Environ. 44, 3187e3195.
Zhu, W.Z., Miser, D.E., Chan, W.G., Hajaligol, M.R., 2004. HRTEM investigation of some commercially available furnace carbon blacks. Carbon 42, 1841e1845.
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
image
|
Fig. 1. Location of monitoring sites: Hong Kong Polytechnic University Campus [PolyU]; Kwun Tong (KT); Hok Tsui (HT).
|
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
table
|
Table 1 Method detection limit (MDL) of selected species determined
|
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
table
|
Table 2 Average concentrations and standard deviations of selected species at PolyU and KT stations
|
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
table
|
Table 3 Comparison of concentrations of metal levels at the major cities of China
|
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
table
|
Table 4 Summary of correlation coefficients of selected species in $\mathrm{PM}_{2.5}$ at PolyU stations
|
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
table
|
Table 5 Summary of correlation coefficients of selected species in $\mathrm{PM}_{2.5}$ at KT stations
|
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
table
|
Table 6 Enrichment factors of major elements and heavy metals at three sampling sites
|
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
table
|
Table 7 The results of principal component analysis for selected species in $\mathrm{PM}_{2.5}$ at PolyU stations (varimax with Kaiser normalization)
|
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
table
|
Table 8 The results of principal component analysis for selected species in $\mathrm{PM}_{2.5}$ at KT stations (varimax with Kaiser normalization)
|
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
table
|
Table 9 The Results of APCA estimated concentration for selected species in $\mathrm{PM}_{2.5}$ at PolyU stations
|
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
table
|
Table 10 The results of APCA estimated concentration for selected species in $\mathrm{PM}_{2.5}$ at KT stations
|
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
image
|
Fig. 2. Estimated source contribution to $\mathrm{PM}_{2.5}$ at PolyU.
|
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
image
|
Fig. 3. Estimated source contribution to $\mathrm{PM}_{2.5}$ at KT.
|
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
image
|
Dendrogramusing Ward Method RescaledDistanceClusterCombine Fig. 4. Cluster analysis of selected species in $\mathrm{PM}_{2.5}$ of PolyU station.
|
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
image
|
Fig. 5. Cluster analysis of selected species in $\mathrm{PM}_{2.5}$ of KT station.
|
|
atmosphere
|
0027
|
10.1016/j.jhazmat.2006.05.047
|
text
| Not supported with pagination yet
|
Source apportionment of $\mathrm{PM}_{2.5}$ in urban area of Hong Kong
K.F. Ho a,∗, J.J. Cao b, S.C. Lee a, Chak K. Chan c
a Department of Civil & Structural Engineering The Hong Kong Polytechnic University, Hong Kong, China b State Key Laboratory of Loess & Quaternary Geology Institute of Earth Environment Chinese Academy of Sciences, China c Department of Chemical Engineering, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China
Received 3 March 2006; received in revised form 15 May 2006; accepted 15 May 2006 Available online 20 May 2006
Abstract
A monitoring program for $\mathbf{PM}_{2.5}$ had been performed at two urban monitoring stations in Hong Kong from November 2000 to February 2001 and June 2001 to August 2001. $\mathrm{PM}_{2.5}$ samples were collected once every 6 days at PolyU and KT stations with the sampling duration of 24-h. A sum of 25 chemical species in $\mathrm{PM}_{2.5}$ were determined and selected for receptor models. Enrichment factors relative to earth crust abundances were evaluated and it was noted that most crustal elements including Al, Ti, Mg, Ca and K have small enrichment factors. Correlation and multivariate analysis technique, such as principal components analysis (PCA)/absolute principal components analysis (APCA) and cluster analysis (CA) are used for source apportionment to identify the possible sources of $\mathrm{PM}_{2.5}$ and to determine their contribution. Six factors at each site were isolated by using PCA/APCA and cluster analysis. Similar sources (crustal matter, automobile emission, diesel emission, secondary aerosols, tire wear, and non-ferrous smelter) are identified by the PCA/APCA and cluster analysis.
$\copyright$ 2006 Elsevier B.V. All rights reserved.
Keywords: Source apportionment; Enrichment factors; Absolute principal components analysis (APCA); Cluster analysis
1. Introduction
Receptor models provide a theoretical and mathematical framework for quantifying source contributions. They interpret measurement of physical and chemical properties taken at different times and places to infer the possible or probable sources and to quantify the contributions from those sources [1]. The purpose of a receptor source apportionment model is to estimate thecontributionsofspecificsourcetypestopollutantlevelsinthe atmosphere at a sampling (or receptor) site. The contributions of each source are distinguished through differences in their physical and chemical properties. Computer-generated source apportionment results must be interpreted by those with knowledge of the site and the associated potential sources.
Principal component analysis (PCA) is one of the oldest and most widely used multivariate statistical techniques in the atmospheric sciences [2]. Usually the data for atmospheric aerosols exhibit many large correlations among parameters and PCA results in a much more compact representation of their variations. By using PCA, Saucy (1991) was able to identify three major sources that contributed to the atmospheric aerosol (near Phoenix, Arizona), namely crustal material, copper smelters and marine air [3]. Fung and Wong (1995) sampled total suspended particulates (TSP) in the western part of the New Territories in Hong Kong between 1986 and 1987 and analyzed various trace metals (e.g. Se, As, Sr, V, etc) as markers [4]. Then PCA was applied to identify the sources and the mass contributions of each source obtained. In many recent source apportionment studies, quantitative aerosol source apportionment was performed using absolute principal component analysis [5–8]. The APCA model can determine: (1) the number of relevant source types influencing the receptor site, (2) the source profiles of these sources in absolute numbers and, finally (3) the impact that each source type has on the concentration levels of the measured air pollutants at the receptor site [5].
The cluster analysis is another effective multivariate statistical method. Although cluster analysis is a potentially useful technique for grouping samples, its application to atmospheric studies has not been attempted broadly. One of the reasons might lie in the difficult interpretation related to the dendrograms. Environmental variables may force unclear sample groups such that the dendrogram is difficult to interpret. In spite of this, it is considered that cluster analysis should be performed, at the very least, to confirm the sample score groups [9]. Saucy (1991) also coupled cluster analysis with principal component analysis to examine compositions and time-dependent concentrations of aerosol particles; they revealed 15 chemically distinct particle types from the samples [3]. Most source identification/apportionment applications have been based on inorganic aerosol components, primarily trace elements often combined with ionic components [10,11]. Moreover, previous work on receptor modeling study of aerosols in East Asia is rather limited [4,12–15]. The objective of this study was to isolate, identify and quantify possible sources that contributed to fine particulate matter $(\mathbf{PM}_{2.5})$ in urban area of Hong Kong.
2. Methodology
2.1. Sampling sites
Two sampling sites including The Hong Kong Polytechnic University (PolyU campus) and Kwun Tong (KT) were selected for $\mathrm{PM}_{2.5}$ monitoring. The field descriptions were given as follows (Fig. 1).
PolyU campus: It situated at about $6\,\mathrm{m}$ above ground level and about $8\,\mathrm{m}$ away from the main traffic road. The station is adjacent to Hong Chong Road, which leads to the Cross Harbour Tunnel. The traffic volume of the road is extremely high which is more than 170,000 vehicles per day.
Kwun Tong (KT): It is close to the residential buildings and most of vehicles are light- and heavy-duty vehicles. Kwun Tong belongs to one of the EPD air quality monitoring stations, which were chosen for data comparison; it represents as mixed residential/commercial/industrial area. The samples were collected on the rooftop of $25\,\mathrm{m}$ .
2.2. Sampling method
The monitoring program for $\mathbf{PM}_{2.5}$ had been performed at two urban monitoring stations in Hong Kong during the two studies periods ((1) November 2000 to February 2001 and (2) June 2001 to August 2001). $\mathrm{PM}_{2.5}$ samples were collected once every 6 days (24-h sampling duration) at PolyU and KT stations.
The high volume (hi-vol) samplers manufactured by Andersen Instruments/GMW were used for $\mathrm{PM}_{2.5}$ sampling at two monitoring stations. The hi-vol samplers were operated at flow rates of $1.\dot{1}3{-}1.41\,\mathrm{m}^{3}\,\mathrm{min}^{-1}$ . $\mathrm{PM}_{2.5}$ samples were collected on $20.3\,\mathrm{cm}\times25.4\,\mathrm{cm}$ Whatmanquartzmicrofiberfliters.Thefilters were pre-heated before sampling at $900\,^{\circ}\mathrm{C}$ for $^{3\,\mathrm{h}}$ . A balance for hi-vol filters (Sartorius, analytic) with accuracy of $0.1\,\mathrm{mg}$ was used to weigh the filter paper which was conditioned in an electronic desiccator before and after sample collection for $24\,\mathrm{{h}}$ . After collection, loaded filters were stored in a refrigerator at about $4\,^{\circ}\mathrm{C}$ before chemical analysis to limit the evaporation of volatile components. Field blank filters were also collected to indicate the artifacts collected onto the filter before/during/after sampling.
2.3. Chemical analysis
After sampling, the filters were conditioned and weighted again to determine the mass concentration of the loaded particles. The filters were then cut into four portions for individual analysis. The filters are then analyzed with different analytical methods: (1) atomic absorption spectrophotometer (AAS) for sodium and potassium, (2) inductively coupled plasma–mass spectrometry (ICP–MS) for elements. The sample solutions were measured in triplicates, and quality controls and blanks were inserted at every 10 samples. The relative standard deviations of the measured element concentrations were typically ${<}5\%$ . Precision and bias were ${<}10\%$ . Element concentrations of the procedural blanks were generally ${<}5\%$ of the samples. (3) Ion chromatography (IC) for water-soluble inorganic ions. Uncertainties were $\pm6\%$ for $\mathrm{Cl^{-}}$ and were $\pm12\%$ for $\mathrm{NO}{_3}^{-}$ , $\mathrm{SO}_{4}{}^{2-}$ and $\mathrm{NH_{4}}^{+}$ . (4) Thermal/optical reflectance (TOR) method for organic carbon (OC) and elemental carbon (EC). The difference determined from replicate analyses was smaller than $5\%$ for TC (total carbon), and $10\%$ for OC and EC. Method detection limit (MDL) of each species was shown in Table 1. The chemical analyses were carried out by the Air Laboratory of PolyU (OC/EC and elements), Department of Chemical Engineering of Hong Kong University of Science and Technology (water-soluble ions), and Department of Chemistry of The Chinese University of Hong Kong (heavy metals).
2.4. Receptor models
Enrichment factor analysis and multivariate techniques such as correlations, absolute principal component analyses (APCA), and cluster analysis (CA) were used to define a relationship between the sources and the receptor. These analytical methods were combined to assist the identification of sources and the apportionment of the observed pollutant concentrations to those sources in the urban area of Hong Kong.
3. Results and discussion
To maximize the source-identification power of factor analysis, only 25 species were selected (after eliminating species with missing data or below detection limit). In general, one should exclude those elements from the analyses with many missing values due to poor detectability, but this must be balanced by including as many elements as possible to increase the degrees of freedom for tuning the model [12]. The 25 species are OC, EC, $\mathrm{Cl^{-}}$ , $\mathrm{NO}{_3}^{-}$ , $\mathrm{SO}_{4}{}^{\bar{2}-}$ , $\mathrm{NH_{4}}^{+}$ , $\mathrm{Na^{+}}$ , $\mathbf{K}^{+}$ , Al, As, Ca, Cr, Cu, Fe, $\mathbf{M}\mathrm{g}$ , Mn, Ni, Pb, Sr, Ti, V, Zn, Ba, Cd and Ga. Median values were substituted for missing data. Values below detection limit (limit of detection) will be replaced by half of the minimum value reported. The average concentrations and standard deviations of the selected species were shown in Table 2. Generally speaking, the concentrations of metal levels determined in this study (both sites) were lower than other major cities in China (Table 3).
3.1. Correlations of selected species in $P M_{2.5}$
Correlations of selected species in $\mathrm{PM}_{2.5}$ at PolyU and KT stations were determined by regression analysis. Their correlation coefficients $(r)$ are shown in Tables 4 and 5, respectively. Nitrate $(\mathrm{NO}_{3}{}^{-})$ , sulfate $(\mathrm{SO}_{4}{}^{2-})$ and ammonium $\mathrm{(NH_{4}^{+})}$ are secondary pollutants, and arise from the oxidation of anthropogenic gases $(\mathrm{NO}_{X}$ , $\mathrm{SO}_{2}$ and $\mathrm{NH}_{3}$ , respectively). Correlation coefficients among $\mathrm{NO}{_3}^{-}$ , $\mathrm{SO}_{4}{}^{2-}$ and $\mathrm{NH_{4}}^{+}$ in both stations were from 0.71 to 0.91 $^{**}P\!<\!0.01)$ . Ammonium nitrate and ammonium sulfate are the possible compounds in the fine aerosols due to the secondary formation from anthropogenic origin. Good correlations $(r\!=\!0.52\!\-\!0.97$ , $^{**}P\!<\!0.01)$ were also observed among the marking species for the crustal matter (Al, Ca, Fe, $\mathbf{M}\mathrm{g}$ and Ti) at both stations (especially in KT) even in fine particles. It implies that they should come from the same sources. At PolyU, the major sources of crustal elements came from paved road dust and they mainly appeared in the large particles (TSP or $\mathbf{PM}_{10}$ ). Therefore, the correlation coefficients among the crustal elements were lower relative to the KT site, where the major sources of crustal elements were due to resuspension and transportation of fine crustal matter. Some heavy metals, such as $\mathrm{Cr}$ and $\mathrm{Cu}$ were also observed to have good correlations with crustal elements in KT.
At PolyU, the major emission sources of carbonaceous species came from vehicular exhaust and fuel evaporation. A fairly good correlation was observed between OC and EC $(r\!=\!0.42)$ because gasoline vehicle exhaust and fuel evaporation were the major sources of OC, however, diesel vehicle exhaust was the major source of EC. At KT, the correlation of OC and EC was slightly higher with $r\!=\!0.65$ $(^{**}P\!<\!0.01)$ . From previous studies [4,7,16,17], heavy metal could be used as marker species. The correlation coefficients between $\mathrm{Ni}$ and V are 0.52 $(^{\ast\ast}P\!<\!0.01)$ in PolyU and 0.68 $(^{**}P\!<\!0.01)$ in KT. It is a clear indication for combustion of oil in previous studies [16,18]. Also Pb and Cd are well correlated $(r\!=\!0.61$ , $^{**}P\!<\!0.01)$ at PolyU site as they mainly came from the traffic sources. However, poor correlation $(r\!=\!0.24)$ ) was observed at KT because Cd might come from other industrial activities also [19]. As is a tracer for coal combustion and $Z\mathfrak{n}$ is a good marker for tire wear and non-ferrous smelters. It implies that the combination of carbonaceous compounds (OC and EC) with inorganic species would give more detail information.
3.2. Enrichment factor of elements PM2.5
Enrichment factors (EF) of trace elements in $\operatorname{PM}_{2.5}$ relative to the earth’s crust were calculated to indicate the extent of contribution of sources other than natural crust to the ambient elemental levels [20,22]. In this study, Fe was used as reference, and the compositions of the earth’s crust were taken from Mason and Moore (1982) [21]. Trace elements EFs include some degree of uncertainty related to the natural variations of the earth crustal composition. For this reason, it is usually assumed that the EFs should be more than an order of magnitude higher than unity to suggest an anthropic origin [22,23].
Results given in Table 6 showed that Al, Ti, $\mathbf{M}\mathrm{g}$ , Ca and K have small enrichment factors since they are mostly crustal. However, Cd is the most enriched elements in $\mathrm{PM}_{2.5}$ , (aver$\mathrm{age}\!=\!1794$ and 1624, at PolyU and KT, respectively) followed by Pb, Zn, As and $\mathrm{Cu}$ . For these elements non-crustal sources such as vehicular exhaust and industrial emission may be suggested. Nevertheless, enrichment of crustal components in fine particles can also occur as a result of their transport to some distance before being removed from the atmosphere by deposition processes [10]. Mn, $\mathrm{Cr}$ and Sr showed low EF which suggested that these crustal sources also predominate in $\operatorname{PM}_{2.5}$ . For these elements, natural emissions are very important and normally exceed anthropogenic sources. Lower EF values with increasing particle size have been reported for Cd, Pb, $Z\mathfrak{n}$ and Ni [24–26]. On the other hand, Eleftheriadis and Colbeck (2001), found increasing EFs with size for coarse V, Cu and $\mathrm{Cr}$ with highest enrichment at around $10\,\upmu\mathrm{m}$ and at the very large sizes [27].
3.3. Principal component analysis (PCA)/absolute principal component analysis (APCA) of selected species in $P M_{2.5}$
PCA has been used generally as an exploratory tool to identify the major sources of aerosol emissions and to statistically select independent source tracers [16]. Result of varimax rotated factor analysis carried out on various selected ambient air components at PolyU and KT and the corresponding possible sources are depicted in Tables 7 and 8, respectively. Six factors at each site were isolated based on the following criteria. Firstly, the number of factors were selected such that the cumulative percentage variance explained by all the chosen factors is more than $80\%$ . Secondly, only the factors with eigenvalue more than one were chosen. Since higher factor loading of particular elements (marker elements) in a factor can help identify the possible sources [28], the number of factors selected (sources identified) should represent the sources which are relevant in the receptor domain [29].
At PolyU, six factors were determined, summing $80\%$ of the totalvarianceinthefineparticledataset.Factor1explains $21.9\%$ of the variance and presented high loading for Al, Ca, Fe, Mg and Ti; thus it can be interpreted as crustal contribution. This can be proved by the low enrichment factors (EF) of these elements. A significant amount of road dust is present near PolyU and is also kept in suspension by vehicular movement. OC, Pb, Zn and Cr were observed in paved road dust profiles of PolyU [30] which have moderate to minor loading in factor 1. Hence, this factor can be identified as the road dust component. Factor 2 explains $18.5\%$ of the variance and contains high loading of $\mathbf{K}^{+}$ , As, Pb and Cd. These elements are used as markers for non-ferrous metal smelter and automobile emission [4,5,12]. Moderate loading of OC, $\mathrm{NO}{_3}^{-}$ , $\mathrm{SO}_{4}{}^{2-}$ and $\mathrm{NH_{4}}^{+}$ were also observed in factor 2. They are secondary aerosols and the byproducts of combustion. Hence, this factor can be identified as automobile emission plus secondary aerosol. Factor 3 is heavily loaded with Zn, Sr and Ga with percentage variance of 13.5. Harrison et al. (1996) and Manoli et al. (2002) used Zn as the marker for tire wear [7,16]. Incineration is also the possible source for Zn [31]. However, there are no significant sources for incineration around PolyU. Therefore, this factor can be determined as tire wear. Factor 4 explains $9.3\%$ of the variance and was highly loaded with V and Ni. The correlation coefficient between Ni and $\mathrm{v}$ is high (see previous section). As discussed before, V and Ni are marker species for oil combustion [16,18]. Thus, factor 4 is identified as oil combustion source. Factor 5 has a high factor loading for OC, EC and $\mathrm{Cu}$ with percentage variance of 9.1. It is obvious that OC and EC come from vehicular exhaust (especially at PolyU site). Also diesel engines are the source of $\mathrm{Cu}$ [7]. This factor was interpreted as representing emissions from diesel vehicles. Finally, high to moderate loading for $\mathrm{Cl^{-}}$ , $\mathrm{NO}{_3}^{-}$ , $\mathrm{SO}_{4}{}^{2-}$ with percentage variance of 7.6 which indicated the presence of secondary or transportation aerosols in factor 6. Six factors were also obtained at KT station with eigenvalues ${>}1$ , explaining $90.9\%$ of the total variance. Basically the six factors are very similar to that obtained at PolyU; but in each factor it may contain more than one source. Factor 1 explains $33.0\%$ of the variance which represented crustal matter and some secondary aerosols $\left(\mathrm{NO}_{3}-\right.$ and $\mathrm{SO}_{4}{}^{2-}$ ). Hence, this factor can be identified as the crustal matter component plus secondary aerosols. Factor 2 explains $22.3\%$ of the variance and contains high loading of $\mathbf{K}^{+}$ , As, $\mathbf{P}\mathbf{b}$ and $\mathrm{Cd}$ , which are markers for automobile emission and non-ferrous metal smelter. Moreover, high loading of OC, $\mathrm{NO}{_3}^{-}$ , $\mathrm{SO}_{4}{}^{2-}$ and $\mathrm{NH_{4}}^{+}$ was also observed in factor 2. Other than combustion, this source type could be the result of the continuous oxidation of primary pollutants (VOCs, $S O_{2}$ , $\mathrm{NO}_{X}$ and $\mathrm{NH}_{3}$ ) taking place during atmospheric transport. Hence, this factor can be identified as automobile emission plus secondary aerosols. Factors 3 and 4 explain 10.8 and $9.4\%$ of the variance, respectively. As in PolyU, factor 3 is identified as tire wear, while factor 4 for oil combustion source. Factor 5 has a high factor loading for EC and Ba with percentage variance of $9.1\%$ . This factor was identified as emissions from diesel vehicles. Finally, the last factor explains $6.8\%$ of the variance and contains high to moderate loading of crustal elements (Sr, Ca and Fe) and heavy metals (Cr, Cu, Pb and V). This factor was interpreted as emissions from several metallurgical activities along with crustal matter occurring in the industrial area.
APCA can provide a quantitative elemental source profile, instead of just a qualitative factor-loading matrix as in PCA. UsingthegroupingresultsofPCAintheprevioussection,source contributions were then calculated using multiple regression of particle mass concentration on absolute principal component scores (APCS). Regressing the gravimetric data on these APCS can give estimates of the coefficients that convert the APCS into mass contributions from each source for each sample. For each source identified by the APCA, the weighted regression of each element’s concentration on the predicted mass contributions yields estimates of the content of that element in each source. Tables 9 and 10 present the APCA source apportionment of $\operatorname{PM}_{2.5}$ in PolyU and KT given in the original concentration units $(\mathrm{ng}/\mathrm{m}^{3})$ ), respectively. The estimated quantitative source apportionments for $\mathrm{PM}_{2.5}$ are shown in Figs. 2 and 3, respectively. Diesel emission is responsible for the most of the $\mathrm{PM}_{2.5}$ mass at PolyU station (about $47\%$ ); and automobile emission plus non-ferrous metal smelter also composed about $15\%$ of the $\mathrm{PM}_{2.5}$ fraction. Secondary aerosols are the second most important factor (about $18\%$ ) at PolyU. Crustal matter also dominates the $\mathrm{PM}_{2.5}$ fraction (about $6\%$ at PolyU). However, at KT, some factor represents more than one source. For example, automobile emission plus non-ferrous metal smelter mixed with secondary aerosols, which composed about $44\%$ of the total $\operatorname{PM}_{2.5}$ mass. Secondary aerosols formed during transportation were due to the primary emission of $\mathrm{SO}_{2}$ and $\mathrm{NO}_{X}$ from vehicular emission. About $30\%$ of $\mathrm{PM}_{2.5}$ was also calculated as crustal matter and secondary aerosols. Diesel emission contributed $14\%$ of the total $\operatorname{PM}_{2.5}$ mass. The large unexplained fractions in PolyU and KT maybe ascribed to the water uptake in the quartz filters [32] and some sources that have not been identified in the analysis. This is reasonable because of the large surface area and moisture absorption property of $20.3\,\mathrm{cm}\times25.4\,\mathrm{cm}$ Whatman quartz microfiber fliters were used in this study [32]. This unexplained fractions were also observed in the mass closure analysis at both stations.
3.4. Cluster analysis of selected species in $P M_{2.5}$
In order to achieve a greater confidence in the final classification, cluster analysis was used for comparison purpose. Resulting dendograms (obtained through Ward’s method) are observed in Figs. 4 and 5, for the PolyU and KT sites, respectively, as obtained using SPSS 8.0 software. The results of cluster analysis in both sites agree with those of PCA. At PolyU site, there is a clear grouping of the elements $\mathbf{K}^{+}$ , Pb, $\mathrm{NH_{4}}^{+}$ , As, $\mathrm{NO}{_3}^{-}$ , $\mathrm{SO}_{4}{}^{2-}$ and Cd which are commonly associated with non-ferrous metal smelter and gasoline emission. $\mathrm{NH_{4}}^{+}$ , $\mathrm{NO}{_3}^{-}$ , and $\mathrm{SO}_{4}{}^{2-}$ are also the major secondary aerosol components in the atmosphere. The clusters are similar to the factors 2 and 6 of PCA result at PolyU. Moreover, the elements Sr, Zn, Ga, Ba and $\mathrm{Cl^{-}}$ also form a group, suggesting a tire wear or incineration origin (like factor 3 in PCA). On the other hand, a cluster typically identified as crustal matter is formed by Fe, Ti, Cr, Na, Al, Mg, Ca and Mn. These elements are associated with pave road dust (like factor 1 in PCA). In addition, Ni and V form a
Dendrogram using Ward Method group which is suggested as oil combustion source (like factor 4 in PCA). A final group, easily identified with traffic (especially diesel engine), is based on OC, EC and $\mathrm{Cu}$ (like factor 5 in PCA).
The KT site dendogram presents a grouping of crustal elements, such as Ca, Fe, Ti, Mg, Cu, Cl, Na, and Al, which is similar to the PCA result for crustal matter in factor 1. To a less extent, Cr and Sr are also correlated to this group. It means the emitted heavy metal $(\mathbf{Cr})$ from industrial area may be attached to the crustal matter and be transported to the receptor site (like factor 6 in PCA). On the other hand, a cluster typically identified as oil combustion is formed by Ni, V and Cd (like factor 4 in PCA). Moreover, the elements OC, As, Pb, $\mathrm{NH_{4}}^{+}$ , $\mathbf{K}^{+}$ , $\mathrm{NO}{_3}^{-}$ and $\mathrm{SO}_{4}{}^{2-}$ also form a group, which is identified as automobile and non-ferrous metal smelter emission mixed with secondary aerosols (like factor 2 in PCA). EC and Ba form a group that is suggested to be as diesel engine related species (like factor 5 in PCA). Finally, the elements $Z\mathfrak{n}$ , Ga and Mn form a group, (undetermined), suggesting a tire wear origin (like factor 3 in PCA). It has been shown that how cluster analysis match with PCA for the identification of pollutants sources in $\operatorname{PM}_{2.5}$ .
source apportionment. Twenty-five chemical species in $\operatorname{PM}_{2.5}$ were determined and selected for receptor models. Six factors at each site were isolated by using PCA/APCA analysis. Fine particulate matter produced by the transport activities is the main air pollution problem in Hong Kong. At PolyU, automobile emission plus secondary aerosol and diesel vehicles are responsible for $62\%$ of the $\mathrm{PM}_{2.5}$ mass. Basically the six factors in KT station are very similar to that obtained at PolyU, but in each factor it may contains not only one source. Factor 2 identified as automobile emission plus secondary aerosols which composed about $44\%$ of the total $\mathrm{PM}_{2.5}$ mass. Also $30\%$ of $\mathrm{PM}_{2.5}$ was calculated ascrustalmatterandsecondaryaerosolsinfactor1.However,the large unexplained fractions in PolyU and KT maybe ascribed to the water uptake in the quartz filters and some sources that have not been identified in the analysis. Cluster analysis was used for comparison purpose in source apportionment. The results in both sites agree with the PCA results for the identification of pollutants sources in $\mathrm{PM}_{2.5}$ .
Acknowledgement
4. Conclusion
A monitoring program for $\mathrm{PM}_{2.5}$ had been performed at two urban monitoring stations in order to identify the main sources influencing the $\mathrm{PM}_{2.5}$ quality in Hong Kong. Enrichment factors and correlation analysis were used as the first step to gain insight on the data and to simplify the chemical interpretation. Then PCA/(APCA and cluster analysis, were used to carry out the
ThisprojectissupportedbyResearchGrantsCouncilofHong Kong (PolyU 5197/05E and PolyU 5145/03E).
References
[1] J.G. Watson, J.C. Chow, Particulate pattern recognition, in: B.L. Murphy, R.T. Morrison (Eds.), Introduction of Environmental Forensics, Academic Press, 2002, pp. 429–460.
[2] P.K. Hopke, E.S. Gladney, G.E. Gordon, W.H. Zoller, A.G. Jones, The use of multivariate analysis to identify sources of selected elements in the Boston aerosol, Atmos. Environ. 10 (1976) 1015–1025.
[3] D.A. Saucy, Aerosol particle characteristics determined by combined cluster and principal component analysis, J. Geophys. Res. 96 (1991) 7407–7414.
[4] Y.S. Fung, L.W.Y. Wong, Apportionment of air pollution sources by receptor models in Hong Kong, Atmos. Environ. 29 (1995) 2041–2048.
[5] E. Swietlicki, R. Krejci, Source characterisation of the Central European atmospheric aerosol using multivariate statistical methods, Nucl. Instrum. Meth. B 109–110 (1996) 519–525.
[6] P. Artaxo, P. Oyola, R. Martinez, Aerosol composition and source apportionment in Santiago de Chile, Nucl. Instrum. Meth. B 150 (1999) 409–416. [7] E. Manoli, D. Voutsa, C. Samara, Chemical characterization and source identification/apportionmentoffineandcoarseairparticlesinThessaloniki, Greece, Atmos. Environ. 36 (2002) 949–961.
[8] W. Maenhaut, M.-T. Fern´andez-Jim´enez, I. Rajta, P. Artaxo, Two-year study of atmospheric aerosols in Alta Floresta Brazil: multielemental composition and source apportionment, Nucl. Instrum. Meth. B 189 (2002) 243–248.
[9] P. Prendes, J.M. Andrade, P. Lopez-Mahia, D. Prada, Source apportionment of inorganic ions in airborne urban particles from Coruna city (N.W. of Spain) using positive matrix factorization, Talanta 49 (1999) 165–178.
[10] A.S. Kao, S.K. Friedlander, Frequency distribution of $\mathrm{{PM}_{10}}$ chemical components and their sources, Environ. Sci. Technol. 29 (1995) 19–28.
[11] J. Miranda, E. Andrade, A. Lopez-Suarez, R. Ledesma, A receptor model for atmospheric aerosols from a southwestern site in Mexico city, Atmos. Environ. 30 (1996) 3471–3479.
[12] Eddie Lee, C.K. Chan, P. Paatero, Application of positive matrix factorization in source apportionment of particulate pollutants in Hong Kong, Atmos. Environ. 33 (1999) 3201–3212.
[13] M. Zheng, L.G. Salmon, J.J. Schauer, L. Zeng, C.S. Kiang, Yuanhang Zhang, G.R. Cass, Seasonal trends in $\mathrm{PM}_{2.5}$ source contributions in Beijing, China, Atmos. Environ. 39 (2005) 3967–3976.
[14] S.S. Park, Y.J. Kim, Source contributions to fine particulate matter in an urban atmosphere, Chemosphere 59 (2005) 217–226.
[15] C.J. Ma, Y. Oki, S. Tohno, M. Kasahara, Assessment of wintertime atmospheric pollutants in an urban area of Kansai, Japan, Atmos. Environ. 38 (2004) 2939–2949.
[16] R.M. Harrison, D.T.J. Smith, L. Luhana, Source apportionment of atmospheric polycyclic aromatic hydrocarbons collected from an urban location in Birmingham, UK, Environ. Sci. Technol. 30 (1996) 825–832.
[17] M. Yatin, S. Tuncel, N.K. Aras, I. Olmez, S. Aygun, G. Tuncel, Atmospheric trace elements in Ankara Turkey. Part 1. Factors affecting chemical composition of fine particles, Atmos. Environ. 34 (2000) 1305–1318.
[18] G.S. Kowalczyk, G.E. Gordon, S.W. Rheingrover, Identification of atmospheric particulate sources in Washington, DC, using chemical element balances, Environ. Sci. Technol. 16 (1982) 79–90.
[19] C. Samara, Th. Kouimtzis, G. Katsoulos, Characterization of airborne particulate matter in Thessaloniki, Greece. Part II. A multivariate modelling approach for the source apportionment of heavy metal concentrations within total suspended particles, Toxicol. Environ. Chem. 41 (1994) 221–232.
[20] Y. Gao, E.D. Nelson, M.P. Field, Q. Ding, H. Li, R.M. Sherrell, C.L. Gigliotti, D.A. Van Ry, T.R. Glenn, S.J. Eisenreich, Characterization of atmospheric trace elements on $\mathrm{PM}_{2.5}$ particulate matter over the New York–New Jersey harbor estuary, Atmos. Environ. 36 (2002) 1077–1086.
[21] B. Mason, C.B. Moore, Principles of Geochemistry, Wiley, New York, 1982.
[22] X.Y. Zhang, J.J. Cao, L.M. Li, R. Arimoto, Y. Cheng, B. Huebert, D. Wang, Characterization of atmospheric aerosol over XiAn in the South Margin of the Loess Plateau, China, Atmos. Environ. 36 (2002) 4189–4199.
[23] K.D. Perry, T.A. Cahill, R.C. Schnell, J.M. Cahill, Long-range transport of anthropogenic aerosols to the National Oceanic and Atmospheric administration baseline station at Mauna Loa Observatory, Hawaii, J. Geophys. Res. 104 (D15) (1999) 18521–18533.
[24] W. Bayens, H. Dedeurwaerder, Particulate trace metals above the southern bight of the North Sea. Part II. Origin and behaviour of the trace metals, Atmos. Environ. 25A (5/6) (1991) 1077–1092.
[25] J. Santamaria, M. Fernandez, J. Mendez, M.T. Bomboi, Particle size distribution of metals in the atmosphere of Madrid (Spain), Fresen. J. Anal. Chem. 337 (1990) 362–365.
[26] Y.C. Chan, R.W. Simpson, G.H. Mctainsh, P.D. Vowles, Characterization of chemical species in $\mathrm{PM}_{2.5}$ and $\mathrm{{PM}_{10}}$ aerosols in Brisbane, Australia, Atmos. Environ. 31 (1997) 3773–3785.
[27] K. Eleftheriadis, I. Colbeck, Coarse atmospheric aerosol: size distributions of trace elements, Atmos. Environ. 35 (2001) 5321–5330.
[28] R.C. Henry, C.W. Lewis, P.K. Hopke, H.J. Williamson, Review of receptor model fundamentals, Atmos. Environ. 18 (1984) 1507–1515.
[29] A.V. Kumar, R.S. Patil, K.S.V. Nambi, Source apportionment of suspended particulate matter at two traffic junctions in Mumbai, India, Atmos. Environ. 35 (2001) 4245–4251.
[30] K.F. Ho, S.C. Lee, J.C. Chow, J.G. Watson, Characterization of $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ source profiles for fugitive dust in Hong Kong, Atmos. Environ. 37 (2003) 1023–1032.
[31] S. Huang, R. Arimoto, K.A. Rahn, Sources and source variations for aerosol at Mace Head, Ireland, Atmos. Environ. 35 (2001) 1421–1437.
[32] EMEP, Measurement of aerosol particle mass and chemical speciation of aerosol particles. EMEP/CCC-Note 1/2000 (2000).
[33] T. Okuda, J. Kato, J. Mori, M. Tenmoku, Y. Suda, S. Tanaka, K. He, Y. Ma, F. Yang, X. Yu, F. Duan, Y. Lei, Daily concentrations of trace metals in aerosols in Beijing, China, determined by using inductively coupled plasmamassspectrometryequippedwithlaserablationanalysis,andsource identification of aerosols, Sci. Total Environ. 330 (2004) 145–158.
[34] C.X. Wang, W. Zhu, An Peng, R. Guichreit, Comparative studies on the concentration of rare earth elements and heavy metals in the atmospheric particulate matter in Beijing, China, and in Delft, the Netherlands, Environ. Int. 26 (2001) 309–313.
[35] J. Zheng, M. Tan, Y. Shibata, A. Tanaka, Y. Li, G. Zhang, Y. Zhang, Z. Shan, Characteristics of lead isotope ratios and elemental concentrations in $\mathrm{{PM}_{10}}$ fraction of airborne particulate matter in Shanghai after the phase-out of leaded gasoline, Atmos. Environ. 38 (2004) 1191–1200.
[36] F. Wei, E. Teng, G. Wu, W. Hu, W.E. Wilson, R.S. Chapman, J.C. Pau, J. Zhang, Ambient concentrations and elemental compositions of $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ in four Chinese cities, Environ. Sci. Technol. 33 (1999) 4188– 4193.
|
atmosphere
|
0028
|
10.1016/j.atmosenv.2015.03.046
|
image
|
Fig. 1. Locations of field observation sites. $\,^{*}\!\mathbf{N}\!]$ , SZ, SH, SAES, HZ and NB represent Nanjing, Suzhou, Shanghai, Shanghai Academy of Environmental Sciences, Hangzhou and Ningbo.
|
|
atmosphere
|
0028
|
10.1016/j.atmosenv.2015.03.046
|
image
|
Fig. 2. Particle concentration and visibility during the haze episode.
|
|
atmosphere
|
0028
|
10.1016/j.atmosenv.2015.03.046
|
image
|
Fig. 3. Concentrations of gaseous pollutants during the haze episode.
|
|
atmosphere
|
0028
|
10.1016/j.atmosenv.2015.03.046
|
image
|
Y. Hua et al. / Atmospheric Environment xxx (2015) 1e12 Fig. 4. Chemical composition of $\mathrm{PM}_{2.5}$ in the YRD during the haze episode. \*Chemical composition of $\mathrm{PM}_{2.5}$ in SAES was obtained from continuous monitoring and $K^{+}$ concentration is the original data instead of non-soil $K^{+}$ due to lack of Fe concentration.
|
|
atmosphere
|
0028
|
10.1016/j.atmosenv.2015.03.046
|
image
|
Fig. 5. Wind direction and speed during the haze episode. $^{*}({\sf a})$ is the wind direction of the four cities. Directions of arrows represent wind directions. The arrow direction in the legend represent $0^{\circ}$ (north direction). Meanwhile, in the clockwise direction, 90, 180, $270^{\circ}$ represent east, south and west direction, respectively. Data of wind direction in Nanjing is lost. (b) shows the wind speed.
|
|
atmosphere
|
0028
|
10.1016/j.atmosenv.2015.03.046
|
image
|
Fig. 6. Relative humidity and temperature during the haze episode.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.