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
|
0009
|
10.1016/j.atmosres.2012.12.004
|
table
|
Table 3 Seasonal mean mass concentrations of $\mathrm{PM}_{2.5}$ in other cities in China.
|
|
atmosphere
|
0009
|
10.1016/j.atmosres.2012.12.004
|
table
|
Table 4 Seasonal mean mass concentrations and contributions of WSIs and PAHs in $\mathrm{PM}_{2.5}$ in the suburb of Shenzhen.
|
|
atmosphere
|
0009
|
10.1016/j.atmosres.2012.12.004
|
image
|
Fig. 4. Ion balance of WSIs in $\mathrm{PM}_{2.5}$ in the suburb of Shenzhen.
|
|
atmosphere
|
0009
|
10.1016/j.atmosres.2012.12.004
|
image
|
Fig. 5. Factor profiles of $\mathrm{PM}_{2.5}$ in the suburb of Shenzhen.
|
|
atmosphere
|
0009
|
10.1016/j.atmosres.2012.12.004
|
image
|
Fig. 6. Seasonal relative contributions of factors to $\mathrm{PM}_{2.5}$ in the suburb of Shenzhen.
|
|
atmosphere
|
0009
|
10.1016/j.atmosres.2012.12.004
|
text
| Not supported with pagination yet |
Chemical composition and source identification of $\mathsf{P M}_{2.5}$ in the suburb of Shenzhen, China
Wei Dai, Jiaqi Gao, Gang Cao, Feng Ouyang
School of Civil and Environmental Engineering, Harbin Institute of Technology Shenzhen Graduate School, Shenzhen 518055, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 3 May 2012
Received in revised form 28 October 2012
Accepted 3 December 2012
Keywords:
PM2.5
Inorganic water-soluble ions
Polycyclic aromatic hydrocarbons
Ion balance
Positive matrix factorization
In this study, $\mathsf{P M}_{2.5}$ has been measured in the suburb of Shenzhen. The mean mass concentration of $\mathsf{P M}_{2.5}$ was $101.6\pm27.5~\upmu\mathrm{g}~\mathrm{m}^{-3}$ in winter and $32.7\pm19.7\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ in summer. The high $\mathrm{PM}_{2.5}$ concentration level is probably due to the serious emissions from the Pearl River Delta where Shenzhen is located. The sum contribution of water-soluble ions was an average of $53.1\pm7.6\%$ in winter and $56.3\pm15.2\%$ in summer, while the sum contribution of polycyclic aromatic hydrocarbons was an average of $7.5\pm1.4\%$ in winter and $20.5\pm7.0\%$ in summer. Obvious seasonal variations of $\mathrm{PM}_{2.5}$ and chemical composition were partly because of the different origins of air mass reaching Shenzhen. $S0_{4}^{2-}$ , $\mathrm{NH}_{4}^{+}$ and $\Nu0_{3}^{-}$ were the major water-soluble ions and fluorene and pyrene were the major polycyclic aromatic hydrocarbons in $\mathrm{PM}_{2.5}$ . Additionally, six factors including combustion emission factor, fresh sea-salt factor, secondary nitrate factor, airborne dust factor, secondary aerosol factor and aged sea-salt factor were identified by EPA positive matrix factorization.
$\circledcirc$ 2012 Elsevier B.V. All rights reserved.
1. Introduction
$\mathsf{P M}_{2.5},$ also known as fine particulate matter (generally defined as those particles with an aerodynamic diameter of $2.5\;\upmu\mathrm{m}$ or less) is the major cause of haze in most areas (Dougle et al., 1996; Lee et al., 2005; Yuan et al., 2006). Primary particles are emitted directly from a source, such as construction sites, unpaved roads, fields, smokestacks or fires. Secondary particles form in complicated reactions in the atmosphere of chemicals, such as $S0_{2}$ and $\Nu0_{\mathrm{x}}$ that are emitted from power plants, industries and vehicles (EPA, 2012). Fine particles can cause serious health problems, because they can penetrate into the gas exchange regions of the lung. Moreover, fine particles lead to high plaque deposits in arteries, causing vascular inflammation and atherosclerosis (Lee et al., 2000; Pope et al., 2002; Dockery and Stone, 2007).
Fine particulate matter can be made up of hundreds of different chemicals. Water-soluble ions (WSIs) are important components of particles and can account for one-third or more of the particulate mass in Chinese urban regions (Hu et al., 2002; Wang et al., 2002, 2006). WSIs in particles have both natural and anthropogenic sources. Natural sources include wind-blown dust, sea salt and lightning-induced biomass burning. Anthropogenic sources are vehicle exhaust, industrial emissions, fossil fuel and biomass burning (Gobbi et al., 2007; Karar and Gupta, 2007). Polycyclic aromatic hydrocarbons (PAHs) are particularly harmful components of particles due to the carcinogenic and mutagenic effects on humans (Qiu et al., 2009; Ohura et al., 2004; Lu et al., 2008). Low molecular PAHs consisting of two or three aromatic rings exist in the gas phase, whereas high molecular PAHs are incorporated into atmospheric particles (Seinfeld and Pandis, 1998). The major sources of PAHs in China were demonstrated to be biomass burning, coal combustion and industrial emissions (Lee et al., 2002; Wang et al., 2010).
Shenzhen $22^{\circ}33^{\prime}\mathrm{N},$ , $114^{\circ}06^{\prime}{\mathrm{E}}$ is a coastal city in the south of Guangdong province in China, located in the Pearl River Delta. Shenzhen has been experiencing elevated levels of particulate matter pollution in recent years because of the rapid economic development. However, there is no comprehensive study on particulate matter in Shenzhen up to date. This research is aimed to better understand the chemical composition and potential sources of fine particulate matter in Shenzhen.
In this study, $\mathsf{P M}_{2.5}$ was measured in the suburb of Shenzhen from the summer of 2009 to the winter of 2010. Then, seasonal variations of chemical compositions including WSIs and PAHs in $\mathsf{P M}_{2.5}$ were also discussed. Additionally, a multivariate factor analysis tool, positive matrix factorization (PMF), was used to identify the potential sources of $\mathsf{P M}_{2.5}$ in different seasons.
2. Methods
2.1. Sampling
The sampling site was located in the suburb of Shenzhen $22^{\circ}35^{\prime}\mathsf{N}.$ $113^{\circ}58^{\prime}\mathrm{E}$ , Fig. 1), surrounded by hills except in the southwest direction. The hills can prevent the transport of air pollutants and thus the local emission sources should mainly originate from the southwest. To the southwest of the sampling site, there are the second largest business district and the two largest power stations in Shenzhen. Additionally, there are several residential areas and construction sites near the sampling site. Therefore, the potential local sources of $\mathsf{P M}_{2.5}$ could be construction dust, vehicle exhaust, biomass burning and fossil fuel combustion emissions.
The sampling was conducted in summer (July and August in 2009 and 2010) and winter (November and December in 2009 and 2010) in a campus in the university town. During the sampling periods, 24-hour $\mathsf{P M}_{2.5}$ samples (from 8:00, local time) were collected on $90\;\mathrm{mm}$ diameter quartz fiber filters (Model AQFA9050, Millipore) using a medium-volume $\mathsf{P M}_{2.5}$ sampler (Model KC6120, LaoShan) operating continuously at a flow rate of $100\;\mathrm{L}\;\mathrm{min}^{-1}$ on the roof of a building about $20~\mathrm{m}$ above the ground. All quartz fiber filters were pre-baked at $500~^{\circ}\mathrm{C}$ for $^{8\mathrm{~h~}}$ to remove trace organic compounds and restored at $0.5~^{\circ}\mathrm{C}$ to avoid volatilization before and after the sampling respectively. Additionally, field blank samples were collected to check the background contamination during the analysis.
2.2. Chemical composition
$\mathsf{P M}_{2.5}$ mass was determined gravimetrically with an electronic microbalance $\vdots10\,\upmu\mathrm{g}$ sensitivity, Model BT25s, Sartorius). The sample filters were equilibrated at a temperature of $25{-}30\ ^{\circ}\mathrm{C}$ and relative humidity of $40–50\%$ for $24~\mathrm{h}$ before the weighing.
After weighing, the filter was cut into small pieces. One half of the filter was extracted in $50~\mathrm{mL}$ ultrapure water for $60\;\mathrm{min}$ using an ultrasonicator. The extract was filtered through $0.45\;\upmu\mathrm{m}$ filter paper to remove insoluble matters. Then the extract was analyzed using an ion chromatography (Model ICS-3000, DIONEX) with 100 mmol $\mathsf{L}^{-1}\,\mathsf{N a O H}$ as an eluent at a flow rate of $0.25~\mathrm{mL~min}^{-1}$ to determine the mass concentrations of inorganic water-soluble anions including chlorine ion $(\mathbf{Cl}^{-})$ , nitrate ion $\left(\mathsf{N O}_{3}^{-}\right)$ and sulfate ion $(\mathsf{S}0_{4}^{2-})$ ). The ion chromatography was equipped with an electric conductivity detector, an AS19 analytical column and a GP40 gradient pump. The mass concentrations of calcium ion $\big(\mathsf{C a}^{2+}\big)$ , magnesium ion $(\mathsf{M}\mathsf{g}^{2+})$ , potassium ion $(\mathsf{K}^{+})$ and sodium ion $(\mathbf{Na}^{+})$ were determined using a flame atomic absorption spectrophotometer (Model 6300C, Shimadzu). The mass concentration of ammonium ion $\left(\mathsf{N H}_{4}^{+}\right)$ was determined by Nessler's reagent method using a UV–Vis spectrophotometer (Model UV-2450, Shimadzu).
The other half of the filter was extracted in $50~\mathrm{mL}$ dichloromethane for $30\;\mathrm{min}$ using an ultrasonicator. The extract was concentrated to about $1~\mathrm{mL}$ by a rotary evaporator and cleaned up by a $\mathrm{SiO_{2}-A l_{2}O_{3}}$ column. The volume of the extract was fixed to $0.5~\mathrm{mL}$ by nitrogen gas. Then the extract was analyzed using a high performance liquid chromatography (Model LC2000, TECHCOMP) with acetonitrile/water $(8/2,\,\mathrm{v/v})$ as a mobile phase at a flow rate of $1\mathrm{\mL\,min}^{-1}$ to determine the mass concentrations of PAHs including acenaphthene, acenaphthylene, anthracene, benzo(a)anthracene, benzofluoranthene, benzo(a)pyrene, chrysene, dibenz(a,h)anthracene, indeno(1,2,3-cd)pyrene, fluoranthene, fluorene, naphthalene, phenanthrene and pyrene.
To ensure the quality of the analysis, the detection limit, precision, recovery ratio and field blank concentrations of WSIs and PAHs in $\mathsf{P M}_{2.5}$ were calculated and the results were listed in Table 1.
2.3. Positive matrix factorization
EPA PMF is one of the source receptor models and the latest version is 3.0. PMF decomposes a matrix of speciated sample data into two matrices (factor contributions and factor profiles)
which then need to be interpreted by an analyst as to what source types are represented using measured source profile information and emission inventories (Paatero and Tapper, 1994; Paatero, 1997).
Variability in the PMF solution can be estimated using a bootstrapping technique, which is a re-sampling method in which “new” data sets are generated that are consistent with the original data. Each data set is decomposed into profile and contribution matrices, and the resulting profile and contribution matrices are compared with the base run (Eberly, 2005). Instead of inspecting point estimates, this method allows the analyst to review the distribution for each species to evaluate the stability of the solution.
2.4. Meteorological condition
Shenzhen has a warm, monsoon-influenced, humid subtropical climate. Winters are mild and dry and summers are hot and humid. The meteorological data with a resolution of $24~\mathrm{h}$ at the sampling site, obtained from the weather station of Shenzhen national climate observatory are shown in Table 2.
Additionally, 72-hour air mass backward trajectories of air mass reaching Shenzhen during the sampling periods were calculated by the NOAA HYbrid Single-Particle Lagrangian Integrated Trajectory model to trace the origin of particulate matters. The calculated height of the backward trajectories was $100~\mathrm{m}$ above ground-level to reduce the influence of the Earth's surface on air mass.
3. Results and discussion
3.1. Seasonal variation of $P M_{2.5}$
The seasonal mean mass concentration of $\mathsf{P M}_{2.5}$ in the suburb of Shenzhen is shown in Fig. 2. It was $101.6\pm$ $27.5\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ in winter and $32.7\pm19.7\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ in summer. WHO air quality guideline (AQG) which is designed to reduce the health impacts of air pollution was chosen as a reference in this study, because there is no quality standard for $\mathsf{P M}_{2.5}$ in China. A mean mass concentration of $25.0\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ was selected as the AQG value (WHO, 2006). It is the lowest level at which total, cardiopulmonary and lung cancer mortality have been shown to increase with more than $95\%$ confidence in response to exposure to $\mathsf{P M}_{2.5}$ . The mean mass concentration of $\mathsf{P M}_{2.5}$ in the suburb of Shenzhen was 4.06 times in winter and 1.31 times in summer higher than the AQG value. Moreover, there were $100\%$ of the samples in winter and $46.8\%$ of the samples in summer over the AQG. The Pearl River Delta where Shenzhen is located is the largest industrial and commercial center in China and it can be a serious emission source of $\mathsf{P M}_{2.5}$ Additionally, because of the relatively backward technology of air pollution control in China, $\mathsf{P M}_{2.5}$ has a much higher concentration level than AQG.
The result shows an obvious seasonal variation of $\mathsf{P M}_{2.5}$ . In winter, low temperature accelerated the gas-to-particle conversion of air pollutants. Additionally, low boundary layer height and low wind speed limited dispersion of air pollutants in winter (Duan et al., 2004; Guinot et al., 2007). Thus, the concentration of $\mathsf{P M}_{2.5}$ significantly increased in winter. Conversely, more precipitation in summer can effectively remove the air pollutants and reduce the concentration of $\mathsf{P M}_{2.5}$ .
The backward trajectories during the sampling periods (Fig. 3) indicate that the seasonal variation of $\mathsf{P M}_{2.5}$ could partly result from the different sources of air mass. The air mass mainly originated from the continent in winter and the sea in summer. Generally, the continent is more polluted than the sea due to human activities. Thus, the air mass in winter brought more air pollutants to local atmosphere after long range transport and led to a higher concentration level of $\mathsf{P M}_{2.5}$ than in summer.
Compared to other cities in China, similar variation of $\mathsf{P M}_{2.5}$ was observed (Table 3). The mass concentrations of $\mathsf{P M}_{2.5}$ in all these cities were higher than the AQG value. It means high particulate matter pollution was frequent in China. The concentrations were high in winter because of the cold and dry climate. Additionally, the inland cities had higher concentrations of $\mathsf{P M}_{2.5}$ than the coastal cities. The result can be used to prove that inland air is more polluted than costal air.
3.2. Seasonal variations of WSIs and PAHs in $P M_{2.5}$
The seasonal mean mass concentrations and contributions of WSIs and PAHs in $\mathsf{P M}_{2.5}$ in the suburb of Shenzhen are listed in Table 4. The sum mass concentration of WSIs was an average of
$51.2\pm8.3\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ in winter and $14.4\pm2.1~\upmu\mathrm{g}~\mathrm{m}^{-3}$ in summer, while the sum mass concentration of PAHs was an average of $10.2\pm1.1~\upmu\mathrm{g}~\mathrm{m}^{-3}$ in winter and $5.0\pm1.2\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ in summer. The mass concentrations of WSIs and PAHs had similar seasonal variations to $\mathsf{P M}_{2.5}$ but the contributions were different.
The sum contribution of WSIs was an average of $53.1\pm$ $7.6\%$ in winter and $56.3\pm15.2\%$ in summer. $S0_{4}^{2-}$ , $\mathrm{NH}_{4}^{+}$ and $\mathrm{NO}_{3}^{-}$ were the major ions in $\mathsf{P M}_{2.5}$ in both seasons. These ions are considered as secondary particulate components which are converted from air pollutants such as $S0_{2}$ , $\Nu0_{\mathrm{x}}$ and $\mathsf{N H}_{3}$ (Amodio et al., 2010). $S0_{2}$ and $\Nu0_{\mathrm{x}}$ are mainly from fossil fuel combustion emissions and $\mathrm{NH}_{3}$ is mainly from biomass burning emissions (Deshmukh et al., 2011). The business district and power stations to the southwest of the sampling site could be the main local sources of secondary particulate components. The high contributions were also attributed to the industrial emissions from the PRD regions. Additionally, the secondary particulate ions contributed more to $\mathsf{P M}_{2.5}$ in winter than in summer. According to the backward trajectories, air mass was mainly from the north continent in winter. Heating devices were widely used in the north of China because of the cold climate in winter. It significantly increased the emissions of power stations. Furthermore, straw burning in the inland village in winter increased biomass burning emissions. Thus, more secondary particles from the north were brought to Shenzhen in winter. In summer, the contributions of $C a^{2+}$ , $\mathrm{Cl}^{-}$ , $\mathsf{K}^{+}$ , ${\mathrm{Mg}}^{2+}$ and $\mathsf{N a}^{+}$ to $\mathsf{P M}_{2.5}$ increased. These ions are considered to be from sea salt and airborne dust (Hellebust et al., 2010). Air mass mainly originated from the sea in summer and brought marine particles to Shenzhen. In addition, more airborne dust was probably due to high wind speed in summer.
The sum contribution of PAHs was an average of $7.5\pm1.4\%$ in winter and $20.5\pm7.0\%$ in summer. Fluorene and pyrene were the major PAHs in $\mathsf{P M}_{2.5}$ . These components are mainly originated from fossil fuel combustion emissions. Contrary to WSIs, PAHs contributed more to $\mathsf{P M}_{2.5}$ in summer than in winter. It can be inferred that PAHs were mainly from the local sources such as power station emissions and vehicle exhaust. Thus, PAHs were hardly affected by air mass from different regions. Summer is a hot but a tourist season in Shenzhen. More refrigeration equipment and higher traffic flow increased the emissions of power stations and vehicles respectively.
3.3. Ion balance analysis of WSIs in $P M_{2.5}$
Ion balance analysis has been commonly used to identify the potential sources of $\mathsf{P M}_{2.5}$ (Verma et al., 2010). The ionic mass concentrations were converted into anion and cation equivalents for ion balance analysis as follows:
$$
\begin{array}{r l}&{\mathrm{Anion\;equivalent}=\mathrm{c}(\mathrm{Cl}^{-})/35.5+\mathrm{c}(\mathrm{NO}_{3}^{-})/62.0}\\ &{\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,+\mathrm{c}(\mathrm{SO}_{4}^{2^{-}})/48.0}\end{array}
$$
$$
\begin{array}{r l}&{\mathrm{Cation\;equivalent=c(NH^{+})/18.0+c(N a^{+})/23.0}}\\ &{\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,+\,\mathrm{c}(\mathbf{Mg}^{2+})/12.2+\mathrm{c}(\mathbf{K}^{+})/39.1}\\ &{\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,+\,\mathrm{c}(\mathbf{Ca}^{2+})/20.0}\end{array}
$$
The correlations between the anion and cation equivalents in the suburb of Shenzhen are shown in Fig. 4. The slopes of both regression equations are less than 1. It means that $\mathsf{P M}_{2.5}$ was alkali in the suburb of Shenzhen during the entire sampling periods. Crustal dust and sea salt could affect the atmospheric nitrogen and sulfur cycles because reactions with the alkaline dust and salt neutralize the acidity (Yuan et al., 2004; Shen et al., 2008). Crustal dust and sea salt contributed more to $\mathsf{P M}_{2.5}$ in summer Shenzhen and thus it shows less slope value and more alkali. Strong correlations $(\mathsf{r}\!>\!0.7)$ ) indicate that the measured anions and cations kept quite constant relationship in neutralization. Additionally, the correlation coefficient in summer was larger than it in winter. $\mathsf{P M}_{2.5}$ was mainly from the local sources in summer and north continent in winter. Thus, the ionic components were more complex and unstable after long range transport in winter than in summer.
3.4. PMF analysis of $P M_{2.5}$
The mass concentrations of WSIs and PAHs in $\mathsf{P M}_{2.5}$ were used to identify the potential sources with EPA PMF model. The mass concentration of individual PAHs was too low to analyze and thus PAHs were considered as a whole group of compounds in the analysis. Six factors of $\mathsf{P M}_{2.5}$ have been identified and each factor represents a potential source. The factors are combustion emission factor, fresh sea-salt factor, secondary nitrate factor, airborne dust factor, secondary aerosol factor and aged sea-salt factor.
3.4.1. Factor profile
The factor profiles of $\mathsf{P M}_{2.5}$ in the suburb of Shenzhen are shown in Fig. 5.
Factor 1 is combustion emission factor, including high contributions of ${\mathrm{Mg}}^{2+}$ , $\mathsf{K}^{+}$ and PAHs. ${\mathrm{Mg}}^{2+}$ and $\mathsf{K}^{+}$ were mainly from the biomass burning emissions in the countryside around Shenzhen (Kim et al., 2010). PAHs were formed by incomplete combustion of carbon-containing fuels from the power station and traffic emissions (Roy, 1995).
Factor 2 and factor 6 are determined to be sea-salt factors. Factor 2 is fresh sea-salt factor including high contributions of $\mathsf{C l}^{-}$ and $\mathsf{N a}^{+}$ . Factor 6 is aged sea-salt factor including high contributions of $\mathsf{K}^{+}$ , $\mathsf{N a}^{+}$ and $S0_{4}^{2-}$ . The lack of $\mathsf{C l}^{-}$ in aged sea-salt factor is caused by chloride depletion for long retention time of particles in the air. Chloride depletion is resulted when the acidic species, mainly nitrate, sulfate and some organic acids, react with NaCl in fresh sea-salt particles and replace $\mathsf{C l}^{-}$ in the form of HCl gas (Zhuang et al., 1999).
Factor 3 and factor 5 are secondary aerosol factors, including high contributions of $\mathsf{N H}_{4}^{+}$ and $S0_{4}^{2-}$ . $\mathrm{NH}_{4}^{+}$ and $S0_{4}^{2-}$ were mainly from biomass burning and fossil fuel combustion emissions respectively. The main difference between the two factors is that $\mathtt{N O}_{3}^{-}$ contributed more to factor 3 than to factor 5. $\mathtt{N O}_{3}^{-}$ was mainly converted from $\Nu0_{\mathrm{x}}$ which was emitted from vehicle exhaust.
Factor 4 is identified as airborne dust factor, including high contributions of ${\mathsf{C}}{\mathsf{a}}^{2+}$ and ${\mathrm{Mg}}^{2+}$ . This factor was mainly affected by the construction sites around the sampling site.
3.4.2. Seasonal variation
The seasonal relative contributions of factors to $\mathsf{P M}_{2.5}$ in the suburb of Shenzhen are shown in Fig. 6. Factor 1 contributed more to $\mathsf{P M}_{2.5}$ in summer than in winter. As discussed in Section 3.2, the large power consumption and traffic flow in summer Shenzhen increased the air pollutants emissions. Thus, the contribution of the components from combustion emissions increased.
On the contrary, the other factors contributed more to $\mathsf{P M}_{2.5}$ in winter than in summer. It is because of the different climatic conditions which have significantly affected the dispersion of $\mathsf{P M}_{2.5}$ . The dispersion was limited by the cold and dry climate in winter and was promoted by the hot and humid climate in summer. However, the seasonal variations of factor 3 and factor 5 were more obvious than factor 2, factor 4 and factor 6. The secondary particles were mainly from the inland air mass after long range transport in winter. Compared with the secondary particles, sea salt and airborne dust were affected by local sources in both seasons.
4. Conclusions
$\mathsf{P M}_{2.5}$ has been measured in the suburb of Shenzhen. High concentration level of $\mathsf{P M}_{2.5}$ was observed, especially in winter. It can be attributed to the serious emissions from industrial and commercial activities in the Pearl River Delta.
$S0_{4}^{2-}$ , $\mathrm{NH_{4}^{+}}$ and $\Nu0_{3}^{-}$ were the major WSIs and fluorene and pyrene were the major PAHs in $\mathsf{P M}_{2.5}$ . Different origins of air mass led to obvious seasonal variations of chemical composition in $\mathsf{P M}_{2.5}$ . Air mass mainly from the north continent in winter and the sea in summer brought more secondary particles and sea salt to $\mathsf{P M}_{2.5}$ respectively.
The result of ion balance analysis has shown that $\mathsf{P M}_{2.5}$ samples in the suburb of Shenzhen were alkali. It is mainly due to high crustal dust and sea salt loads which neutralize the acidity in the atmosphere. Additionally, $\mathsf{P M}_{2.5}$ in summer was more alkali and stable than in winter.
Six factors of $\mathsf{P M}_{2.5}$ in the suburb of Shenzhen, including combustion emission factor, fresh sea-salt factor, secondary nitrate factor, airborne dust factor, secondary aerosol factor and aged sea-salt factor have been identified by EPA PMF. Combustion factor contributed more to $\mathsf{P M}_{2.5}$ in summer than in winter. The large power consumption and traffic flow in summer Shenzhen increased the combustion emissions. Contrarily, the other factors contributed more in winter than in summer because the cold and dry climate limited the dispersion of PM2.5.
Acknowledgments
This work was supported by the Shenzhen National Climate Observatory.
References
Amodio, M., Andriani, E., Cafagna, I., Caselli, M., Daresta, B.E., de Gennaro, G., Di Gilio, A., Placentino, C.M., Tutino, M., 2010. A statistical investigation about sources of PM in South Italy. Atmos. Res. 98, 207–218.
Deshmukh, D.K., Deb, M.K., Tsai, Y.I., Mkoma, S.L., 2011. Water soluble ions in $\mathrm{PM}_{2.5}$ and $\mathrm{PM}_{1}$ aerosols in Durg City, Chhattisgarh, India. Aerosol Air Qual. Res. 11, 696–708.
Dockery, D.W., Stone, P.H., 2007. Cardiovascular risks from fine particulate air pollution. N. Engl. J. Med. 356, 511–513.
Dougle, P.G., Vlasenko, A.L., Veefkind, J.P., Brink, H.M.T., 1996. Humidity dependence of the light scattering by mixtures of ammonium nitrate, ammonium sulfate and soot. J. Aerosol Sci. 27, 513–514.
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, 1275–1282.
Eberly, S., 2005. EPA PMF 1.1 User's Guide. National Exposure Research Laboratory, Research Triangle Park, NC.
EPA, 2012. http://www.epa.gov/pm/basic.html.
Feng, Y.L., Chen, Y.J., Guo, H., Zhi, G.R., Xiong, S.C., Li, J., Sheng, G.Y., Fu, J.M., 2009. Characteristics of organic and elemental carbon in $\mathrm{PM}_{2.5}$ samples in Shanghai, China. Atmos. Res. 92, 434–442.
Gobbi, G.P., Barneba, F., Ammanato, L., 2007. Estimating the impact of Saharan dust on the year 2001 $\mathrm{PM}_{10}$ record of Rome, Italy. Atmos. Environ. 41, 261–275.
Guinot, B., Cachier, H., Sciare, J., et al., 2007. Beijing aerosol: atmospheric interactions and new trends. J. Geophys. Res. 112, 14314.
Hellebust, S., Allanic, A., Oconnor, I.P., Jourdan, C., Healy, D., Sodeau, J.R., 2010. Sources of ambient concentrations and chemical composition of PM(2.5–0.1) in Cork Harbour, Ireland. Atmos. Res. 95, 136–149.
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. Atmos. Environ. 36, 5853–5859.
Karar, K., Gupta, A.K., 2007. Source apportionment of $\mathrm{PM}_{10}$ at residential and industrial sites of an urban region of Kolkata, India. Atmos. Res. 84, 30–41.
Kim, E., Turkiewicz, K., Sylvia, Z.A., Magliano, K.L., 2010. Sources of fine particles in the South Coast area, California. Atmos. Environ. 44, 3095–3100.
Lai, S.C., Zou, S.C., Cao, J.J., Lee, S.C., Ho, K.F., 2007. Characterizing ionic species in $\mathrm{PM}_{2.5}$ and $\mathrm{PM}_{10}$ in four Pearl River Delta cities, South China. J. Environ. Sci. 19, 939–947.
Lee, J.T., Kim, H., Hong, Y.C., Kwon, H.J., Schwartz, J., Christiani, D.C., 2000. Air pollution and daily mortality in seven major cities of Korea, 1991–1997. Environ. Res. 84, 247–254.
Lee, S.C., Chiu, M.Y., Ho, K.F., Zou, S.C., Wang, X.M., 2002. Volatile organic compounds (VOCs) in urban atmosphere of Hong Kong. Chemosphere 48, 375–382.
Lee, C.G., Yuan, C.S., Chang, J.C., Yuan, C., 2005. Effects of aerosol species on atmospheric visibility in Kaohsiung City, Taiwan. J. Air Waste Manage. Assoc. 55, 1031–1041.
Lu, H., Zhu, L.Z., Chen, S.G., 2008. Pollution level, phase distribution and health risk of polycyclic aromatic hydrocarbons in indoor air at public places of Hangzhou, China. Environ. Pollut. 152, 569–575.
Ohura, T., Amagai, T., Sugiyama, T., Fusaya, M., Matsushita, H., 2004. Characteristics of particle matter and associated polycyclic aromatic hydrocarbons in indoor and outdoor air in two cities in Shizuoka, Japan. Atmos. Environ. 38, 2045–2054.
Paatero, P., 1997. Least squares formulation of robust non-negative factor analysis. Chemom. Intell. Lab. Syst. 37, 23–35.
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.
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. J. Am. Med. Assoc. 287, 1132–1141.
Qiu, Y.W., Zhang, G., Liu, G.Q., Guo, L.L., Li, X.D., Wai, O., 2009. Polycyclic aromatic hydrocarbons (PAHs) in the water column and sediment core of Deep Bay, South China. Estuarine Coastal Shelf Sci. 83, 60–66.
Roy, G.M., 1995. Activated Carbon Applications in the Food and Pharmaceutical Industries. CRC Press, New York, p. 125.
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. 427, 146–158.
Seinfeld, J.H., Pandis, S.N., 1998. Atmospheric Chemistry and Physics, Third ed. John Wiley & Sons, USA, p. 748.
Shen, Z.X., Arimoto, R., Cao, J.J., Zhang, R.J., Li, X.X., Du, N., Okuda, T., Nakao, S., Tanaka, S., 2008. Seasonal variations and evidence for the effectiveness of pollution controls on water-soluble inorganic species in TSP and $\mathrm{PM}_{2.5}$ from Xi'an, China. J. Air Waste Manag. Assoc. 58, 1560–1570.
Verma, S.K., Deb, M.K., Suzuki, Y., Tsai, Y.I., 2010. Ion chemistry and source identification of coarse and fine aerosols in an urban area of eastern central India. Atmos. Res. 95, 65–76.
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. Atmos. Environ. 36, 1299–1307.
Wang, Y., Zhuang, G.S., Sun, Y.L., An, Z.S., 2006. The variation of characteristics and formation mechanisms of aerosols in dust, haze, and clear days in Beijing. Atmos. Environ. 40, 6579–6591.
Wang, Y., Li, P.H., Li, H.L., Liu, X.H., Wang, W.X., 2010. PAHs distribution in precipitation at Mount Taishan China. Identification of sources and meteorological influences. Atmos. Environ. 95, 1–7.
WHO, 2006. WHO Air Quality Guidelines for Particulate Matter, Ozone, Nitrogen Dioxide and Sulfur Dioxide. World Health Organization, p. 9.
Yuan, C.S., Sau, C.C., Chen, M.C., Huang, M.H., Chang, S.W., Lin, Y.C., Lee, C.G., 2004. Mass concentration and size-resolved chemical composition of atmospheric aerosols sampled at Pescadores Islands during Asian dust storm periods in the years of 2001 and 2002. Terr. Atmos. Ocean. Sci. 15, 857–879.
Yuan, C.S., Lee, C.G., Liu, S.H., Chang, J.C., Yuan, C., Yang, H.Y., 2006. Correlation of atmospheric visibility with chemical composition of Kaohsiung aerosols. Atmos. Res. 82, 663–679.
Zhuang, H., Chan, C.K., Fang, M., Wexler, A.S., 1999. Formation of nitrate and non-sea-salt sulfate on coarse particles. Atmos. Environ. 33, 4223–4233.
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
image
|
Fig. 1. Map of the sampling sites in this study.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
table
|
Table 1. Mass concentrations of aerosols $(\upmu\mathrm{g}\,\\mathsf{m}^{-3})$ at the summit of MT and other sampling sites.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
table
|
Table 2. Mass concentrations $(\upmu\mathrm{g}\,\\\mathrm{m}^{-3})$ and size distributions of aerosols at the summit of MT and other sampling sites.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
table
|
Table 3. Meteorological conditions and concentrations of gases $(\mathrm{O}_{3}$ , CO, Peroxide) in ambient air at MT.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
image
|
Fig. 2. Daily variations of TSP and $\mathrm{PM}_{2.5}$ from 14 March to 30 June in 2006 at MT.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
table
|
Table 4. Water-soluble ions in $\mathrm{PM}_{2.5}$ and TSP and the corresponding ratios of summer/spring at MT.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
table
|
Table 5. SOR, NOR of ambient air and main water-soluble ions in $\mathrm{PM}_{2.5}$ and TSP at MT and Beijing.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
table
|
Table 6. Concentrations $(\upmu\mathrm{g}\,\\mathsf{m}^{-3})$ ) of elements in $\mathrm{PM}_{2.5}$ and TSP and the corresponding ratios of summer/spring at MT.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
image
|
Fig. 3. Enrichment factors (EFs) of elements at MT in spring and summer, 2006.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
table
|
Table $7.\,\mathrm{pH}$ of aqueous filtrates of $\mathrm{PM}_{2.5}$ and TSP aerosols at different sampling sites. 3.2 Sources and formation mechanisms of aerosols at summit of MT
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
image
|
Fig. 4. Daily variations of pH of aqueous filtrates of aerosols at MT, 2006.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
image
|
Fig. 5. Daily variations of $\mathrm{K^{+}}$ , $\mathrm{Na^{+}}$ , and Al in $\mathrm{PM}_{2.5}$ from 14 March to 30 June in 2006 at MT.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
image
|
Fig. 6. Concentrations of biomass burning derived $\ K^{+}$ calculated with two methods in $\mathrm{PM}_{2.5}$ in 2006 at MT.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
image
|
Fig. 7. Fire spot data derived from MODIS Global Fire Mapping during 2006: (a) 1–29 March, (b) 1–29 April, (c) 1–29 May, (d) 1–29 June, (e) 1–9 June and (f) 10–19 June at MT.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
image
|
Fig. 8. Daily contributions of biomass burning, crustal dust, and others in $\operatorname{PM}_{2.5}$ at MT.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
table
|
Table 8. Concentrations of $\ K^{+}$ in $\mathrm{PM}_{2.5}$ at different sampling sites.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
image
|
Fig. 9. Seasonal variations of $\mathrm{SO}_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ , and $\mathrm{NH}_{4}^{+}$ in aerosols at different sampling sites.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
table
|
Table 9. Correlation coefficients between $\mathsf{K}^{+}$ and other species in aerosols in summer at MT.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
table
|
Table 10. Ratio of $\mathrm{Ca/Al}$ in aerosol or surface soil in different sampling sites over China.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
image
|
Fig. 10. Scatter plots of $\mathrm{SO}_{4}^{2-}$ vs. decides (sum of $\mathrm{CH}_{2}\mathrm{C}_{2}\mathrm{O}_{4}^{2-}$ , $\mathrm{CH}_{4}\mathrm{C}_{2}\mathrm{O}_{4}^{2-}$ , and $C_{2}0_{4}^{2-}$ ).
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
image
|
Fig. 11. Daily variation of mineral elements (Ca, Al and Fe) at different sampling sites in spring, 2007.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
table
|
Table 11. Correlation coefficients among the certain species in spring and summer at MT.
|
|
atmosphere
|
0010
|
10.5194/acp-11-7319-2011
|
text
| Not supported with pagination yet |
Chemical characterization of aerosols at the summit of Mountain Tai in Central East China
C. Deng1, G. Zhuang1, K. Huang1, J. Li1, R. Zhang1, Q. Wang1, T. Liu1, Y. $\mathbf{Sun}^{2}$ , Z. Guo1, J. S. $\mathbf{F}\mathbf{u}^{3}$ , and Z. Wang2
1The Center for Atmospheric Chemistry Study, Department of Environmental Science & Engineering, Fudan University, Shanghai 200433, China
2The Institute of Atmospheric Physics, Chinese Academy of Science, LAPC, Beijing 100029, China
3Department of Civil and Environmental Engineering, The University of Tennessee, Knoxville, TN 37996, USA
Received: 13 July 2010 – Published in Atmos. Chem. Phys. Discuss.: 2 September 2010
Revised: 18 March 2011 – Accepted: 16 July 2011 – Published: 25 July 2011
Abstract. $\mathrm{PM}_{2.5}$ and TSP samples were collected at the summit of Mountain Tai (MT) ( $1534\,\mathrm{m}\,\mathrm{a.s.l.})$ in spring 2006/2007 and summer 2006 to investigate the characteristics of aerosols over central eastern China. For comparison, aerosol samples were also collected at Tazhong, Urumqi, and Tianchi in Xinjiang in northwestern China, Duolun and Yulin in northern China, and two urban sites in the megacities, Beijing and Shanghai, in 2007. Daily mass concentrations of TSP and $\mathrm{PM}_{2.5}$ ranged from $39.6–287.6\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ and $17.2–235.7\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ respectively at the summit of MT. Averaged concentrations of $\operatorname{PM}_{2.5}$ showed a pronounced seasonal variation with higher concentration in summer than spring. 17 water-soluble ions $(\mathrm{SO}_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ , $\mathrm{Cl^{-}}$ , $\mathrm{F}^{-}$ , $\mathrm{PO}_{4}^{3-}$ , $\mathrm{NO}_{2}^{-}$ , $\mathrm{CH}_{3}\mathrm{COO}^{-}$ , $\mathrm{CH}_{2}\mathrm{C}_{2}\mathrm{O}_{4}^{2-}$ , $\mathrm{C}_{2}\mathrm{H}_{4}\mathrm{C}_{2}\mathrm{O}_{4}^{2-}$ , $\mathrm{HCOO^{-}}$ , MSA, $\mathrm{C}_{2}\mathrm{O}_{4}^{2-}$ , $\mathrm{NH_{4}^{+}}$ , $\mathrm{{Ca}}^{2+}$ , $\mathbf{K}^{+}$ , ${\mathrm{Mg}}^{2+}$ , $\mathrm{Na}^{+})$ , and 19 elements of all samples were measured. $\mathrm{SO}_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ , and $\mathrm{NH_{4}^{+}}$ were the major water-soluble species in $\mathrm{PM}_{2.5}$ , accounting for $61.50\,\%$ and $72.65\,\%$ of the total measured ions in spring and summer, respectively. The average ratio of $\mathrm{PM}_{2.5}/\mathrm{TSP}$ was 0.37(2006) and 0.49(2007) in spring, while up to 0.91 in summer, suggesting that aerosol particles were primarily comprised of fine particles in summer and of considerable coarse particles in spring. Crustal elements (e.g., Ca, Mg, Al, Fe, etc.) showed higher concentration in spring than summer, while most of the pollution species $(\mathrm{SO}_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ , $\Chi^{+}$ , $\mathrm{NO}_{2}^{-}$ , $\mathrm{NH_{4}^{+}}$ , $\mathrm{Cl^{-}}$ , organic acids, $\mathbf{P}\mathbf{b}$ , Zn, Cd, and $\mathrm{Cr}$ ) from local/regional anthropogenic emissions or secondary formation presented higher concentration in summer. The ratio of $\mathrm{Ca/Al}$ suggested the impact of Asian dust from the western deserts on the air quality in this region.
The high concentration of $\Chi^{+}$ in $\mathrm{PM}_{2.5}$ $(4.41\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ ) and its good correlation with black carbon $(r=0.90)$ and oxalic acid $(r\,{=}\,0.87)$ suggested the severe pollution from biomass burning, which was proved to be a main source of fine particles over central eastern China in summer. The contribution of biomass burning to the fine particle at MT accounted for $7.56\,\%$ in spring and $36.71\,\%$ in summer, and even reached to $81.58\,\%$ on a day. As and $\mathbf{P}\mathbf{b}$ were two of the most enriched elements. The long-range transport of aerosols spread the heavy pollution from coal-mining/coal-ash to everywhere over China. Anthropogenic air-pollution was evidently rather severe at MT, though it has been declared by UNESCO to be a World Heritage site.
1 Introduction
Aerosols have potential impact on the global atmospheric chemistry, cloud properties and precipitation development (Tegen et al., 1996; Arimoto, 2001; Rastogi and Sarin, 2005, 2006). Anthropogenic aerosols, including the primarily emissions and the secondary aerosols, are mainly in fine mode, which has much more adverse impact on climate and hydrologic cycling (Kaufman et al., 2002), visibility (Chan et al., 1999), and human health (Dockery et al., 1993).
Eastern China, including provinces of Hebei, Shandong, Jiangsu, Zhejiang, and the mega-city, Shanghai, is the area with rapidest growth in economy, e.g. Shandong is one of the two provinces in China, whose GDP exceeded $3\times10^{11}$ RMB in 2008 (http://finance.people.com.cn/), which resulted in the increasing emissions of $S\mathrm{O}_{2}$ , $\mathrm{NO}_{2}$ , and particulate matter, and, in turn, the severe acidic precipitation (Wang et al., 2008). Mountain Tai $200\times50$ square kilometers), with the highest altitude ( $1534\,\mathrm{m}$ high) in central-eastern China, is located in Shandong and surrounded by Jiangsu, Anhui, Henan, and Hebei provinces. Aerosols, from the summit of MT could be the representative of the regional pollution. Previous study found that CO and ${\mathrm{O}}_{3}$ at the top of MT exhibited high in summer and low in winter, which was attributed to the seasonal changes of meteorological conditions and seasonal variations of sources (Gao et al., 2005; Wang et al., 2001). In addition, VOCs, $\mathrm{O}_{3}$ , and CO were all higher than those observed at other rural mountainous sites (Suthawaree, et al., 2010), which might be due to the strong sources surrounded. Average concentration of peroxides at MT was much lower than that measured at some rural mountain sites, suggesting that significant removal processes took place in this region (Ren et al., 2009). ${\mathrm{O}}_{3}$ and CO play key roles in determining the oxidizing capacity of the atmosphere in the presence of sunlight, and they are ideal tracers for anthropogenic pollutions (Novelli et al., 1994, 1998). VOC, $\mathrm{O}_{3}$ , CO, and peroxide are all related to the formations of secondary aerosols in ambient air, which suggested that the characteristics of the aerosols at MT might be different from those at other sites.
Mineral aerosols through long-range transport may directly or/and indirectly affect the properties of air mass by providing surfaces for many chemical and physical processes and serving as carriers of anthropogenic substances, which would affect the global climate/environmental change (Guo et al., 2004; Dentener et al., 1996; Sun et al., 2004; Liu et al., 2002). Northwestern China is one of the main source areas of Asian dust that can be transported to hundred and thousand miles away, passing through central and eastern China and even to the Pacific. The composition of mineral aerosols would subject to transform due to adsorbing gaseous species, surface reactions, and coagulation with anthropogenic aerosol on the pathway during transport.
Little information on aerosols at the summit of MT is available. The studies on aerosols reported in literatures mostly represented those samples collected from ground level, and limited knowledge has been acquired on the aerosols at high elevation over the world. MT is just in the downstream of Asian Dust from northwestern China. Therefore, the summit of MT is an ideal site to examine the longrange transport of Asian dust and to observe the mixing of dust with anthropogenic aerosol. This paper presents the characteristics, sources, formation processes, and the relation with the long-range transport of aerosols collected at MT, which would reveal the air quality in PBL (Planetary Boundary Layer) over central east China.
2 Experimental
2.1 Sampling
TSP and $\mathrm{PM}_{2.5}$ aerosol samples were simultaneously collected at the meteorological observation station located at the summit of MT $36.25^{\circ}\,\mathrm{N}$ , $117.10^{\circ}\,\mathrm{E}$ ) in spring (14 March–
6 May) and summer (2–30 June) in 2006 and in spring (26 March–18 May) in 2007. The sampling duration of each sample was generally $24\,\mathrm{{h}}$ , except a few samples, which was $48\,\mathrm{h}$ . All of the samples were collected on Whatman $_{\mathcal{\registered41}}$ filters (Whatman Inc., Maidstone, UK) by mediumvolume samplers (model: $(\mathrm{TSP}/\mathrm{PM}_{10}/\mathrm{PM}_{2.5})-2$ 2, flow rate: $77.59\,\mathrm{L}\,\mathrm{min}^{-1},$ ). The samples were put in polyethylene plastic bags right after sampling and reserved in a refrigerator. All of these filters were weighed before and after sampling with an analytical balance (Sartorius 2004MP, reading precision $10\,\upmu\mathrm{g})$ after stabilizing under constant temperature $(20\pm1\,^{\circ}\mathrm{C})$ and humidity $(40\pm1.5\,\%)$ for over $24\,\mathrm{{h}}$ . All the procedures were strictly quality-controlled to avoid any possible contamination of the samples. For comparison, aerosol samples were also collected at Tazhong, Urumqi, Tianchi in Xinjiang in northwestern China, Duolun and Yulin in northern China, and in two urban sites in the megacities, Beijing and Shanghai (Fig. 1). Information of all samples is list in Table 1. The detailed analytical procedures were given elsewhere (Zhuang et al., 2001).
2.2 Chemical analysis
2.2.1 Ion analysis
One-fourth of each sample and blank filter was extracted ultrasonically by $10\,\mathrm{mL}$ deionized water $(18\,\mathrm{M}\Omega\mathrm{cm}^{-1})$ ). After passing through microporous membranes (pore size, $0.45\,\upmu\mathrm{m}$ ; diameter, $25\,\mathrm{mm}$ ; made by the affiliated plant of Beijing chemical school), the filtrates were determined for $\mathrm{pH}$ with a $\mathsf{p H}$ meter (model, Orion 818). Each filtrate was stored at $4\,^{\circ}\mathbf{C}$ in a clean tube for IC analysis. 12 anions $(\mathrm{SO}_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ , $\mathrm{Cl^{-}}$ , $\mathrm{F}^{-}$ , $\mathrm{PO}_{4}^{3-}$ , $\mathrm{NO}_{2}^{-}$ , $\mathrm{CH}_{3}\mathrm{COO}^{-}$ $\mathrm{{HCOO^{-}}}$ , MSA, $\mathrm{C}_{2}\mathrm{O}_{4}^{2-}$ , $\mathrm{CH}_{2}\mathrm{C}_{2}\mathrm{O}_{4}^{2-}$ , $\mathrm{C}_{2}\mathrm{H}_{4}\mathrm{C}_{2}\mathrm{O}_{4}^{2-})$ and 5 cations $\mathrm{(NH_{4}^{+}}$ , $\mathrm{{Ca}}^{2+}$ , $\Chi^{+}$ , ${\mathrm{Mg}}^{2+}$ , $\mathrm{Na^{+}}$ ) were analyzed by
Ion Chromatography (Model: Dionex 3000), which consists of a separation column (Dionex Ionpac AS11 for anion and CS12A for cation), a guard column (Dionex Ionpac AG 11 for anion and AG12A for cation), a self-regenerating suppressed conductivity detector (Dionex Ionpac ED50), and a gradient pump (Dionex Ionpac GP50). The gradient mobile phase generated by EG-3000 was used for anion detection, while the weak acid eluent ( $20\,\mathrm{mM}$ MSA) for cation detection. The recovery of each ion was in the range of $80–120\,\%$ . The relative standard deviation of each ion was less than $5\,\%$ for reproducibility test. The limits of detection $(\mathrm{S}/\mathrm{N}\!=\!3)$ were less than $0.04\,\mathrm{mg\,L^{-1}}$ for anions and $0.006\,\mathrm{mg\,L^{-1}}$ for cations. The quality assurance was routinely carried out by using Standard Reference Materials (GBW 08606) produced by National Research Center for Certified Reference Materials, China. Blank values were subtracted from sample determinations. The details were given elsewhere (Yuan et al., 2003).
2.2.2 Element analysis
Half of each sample filter and blank filter was digested at $170\,^{\circ}\mathrm{C}$ for $4\,\mathrm{{h}}$ in high-pressure Teflon digestion vessel with $3\,\mathrm{mL}$ concentrated $\mathrm{HNO}_{3}$ , $1\,\mathrm{mL}$ concentrated HCl, and $1\,\mathrm{mL}$ concentrated HF. After cooling, the solutions were dried, and then added $0.1\,\mathrm{mL}$ concentrated $\mathrm{HNO}_{3}$ , and diluted to $10\,\mathrm{mL}$ with deionized water (resistivity of $18\,\mathrm{M}\Omega\mathrm{cm}^{-1}$ ). Total 19 elements (Al, Fe, Mn, Mg, Ti, Na, Sr, Ca, Co, Cr, Ni, Cu, Pb, Zn, Cd, V, S, As and P) were determined by Inductively Coupled Plasma Atomic Emission Spectroscopy (ICPAES, model: ULTIMA, made by JOBIN-YVON Company, France). All the reagents used were of the highest grade. All the preparation was carried out in a Class-100 clean bench. The recovery rates were measured with standard addition, and the recoveries of each element were in the range of $95\,\%$ to $105\,\%$ . The relative standard deviations of each element were less than $2\,\%$ for reproducibility test. The geochemistry reference matter (TBW07401) made by the Center for National Standard Matter was also analyzed simultaneously to check the reliability of analysis (Han et al., 2005). The detection limits $(3\,{\mathrm{s}})$ for typical elements of Al, As, Ca, Cd, Co, Fe, Pb, Cu, and $Z\mathfrak{n}$ were 1.5, 5, 0.03, 0.35, 0.6, 0.5, 5, 0.6, and $0.3\,\upmu\mathrm{g}\,\mathrm{L}^{-1}$ , respectively. Black carbon (BC) was analyzed with Smokerstain Reflectometer (UK, Model, M43D). The detailed analytical procedures were given elsewhere (Zhuang et al., 2001).
2.3 Meteorological data, fire spot map and trace gases
The meteorological data, including temperature, relative humidity (RH), dew point, wind speed, wind direction, atmospheric pressure, visibility etc., were collected from http://www.wunderground; Data of $\mathrm{SO}_{2}$ , $\mathrm{NO}_{2}$ in Shanghai and MT were collected from http://www.envir.gov.cn and http://www.tahb.gov.cn; Fire spot data were got from MODIS Global Fire Mapping Service (http://firefly.geog. umd.edu/firemap/); $\mathrm{O}_{3}$ and CO were detected with a commercial UV photometric analyzer (Thermo Environment Instruments Inc., Model 49) that had a detection limit of 2 ppbv and a 2-sigma (2-s) precision of 2 ppbv for a 2- min average. CO was measured with a gas filter correlation, a nondispersive infrared analyzer (Advanced Pollution Instrumentation Inc., Model 300) with a heated catalytic scrubber for baseline determination, which was conducted every 2h. The detection limit was 30 ppbv for a 2-min average, with a 2-s precision of about $1\,\%$ for a level of 500 ppbv (2-min average). The overall uncertainty was estimated to be $10\,\%$ .
3 Results and discussion
3.1 Overview of particle matters at summit of MT
3.1.1 Mass concentrations of PM and size distribution
Temporal variations of mass concentrations and the corresponding deviations of $\operatorname{PM}_{2.5}$ and TSP at MT and other sampling sites are summarized in Fig. 2 and Tables 1–2. TSP were relatively seasonal stable $(\dot{1}28.1\,\upmu\mathrm{g}\,\mathrm{m}^{-3}/143.8\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in spring 2006/2007 and $135.0\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in summer 2006), while $\mathrm{PM}_{2.5}$ had highly seasonal variation with much higher concentration in summer 2006 $(123.1\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ ) than in spring 2006/2007 $(46.6\,\upmu\mathrm{g}\,\mathrm{m}^{-3}/70.1\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ at MT. $\mathrm{PM}_{2.5}$ was relatively higher in spring 2007 (26 March–18 May) compared to that in spring 2006 (14 March–6 May), which was likely attributed to the different sampling periods and it will be interpreted in Sect. 3.2.1. TSP from MT was comparable with those from other sampling sites in ground level, and there was no obvious decrease at a height of $1534\,\mathrm{m}$ compared to those at ground level, as expected. Both TSP and $\mathrm{PM}_{2.5}$ at MT were much higher than those at Tianchi that is also located at a high land $(1900\,\mathrm{m})$ . Aerosol pollutions at MT were as severe as or even worse than those at ground level, e.g. $\mathrm{PM}_{2.5}$ at MT were much higher than those at other sites, including the sites in megacities, Beijing and Shanghai, at ground level in summer. The average ratio of $\mathrm{PM}_{2.5}/\mathrm{TSP}$ was 0.37 in spring and 0.91 in summer in 2006, indicating that fine particles dominated in summer while coarse particles in spring. The seasonal variation of fine particles should be firstly attributed to the seasonally different meteorology at MT (Table 3), which showed typical seasonal variations with higher temperature, lower windy speed, lower atmospheric pressure and more solar radiation in summer than those in spring. The meteorology at MT could strengthen the vertical convection of the air, and the regional anthropogenic pollutants on the ground surface could easily transport upward, resulting in the increase of the height of PBL, that was even higher than the summit of MT. On the contrary, the height of the PBL in spring was so compressed that it used to be below the summit of MT, and the Mountain-valley breezes could not fully develop due to the weak solar radiation and the strong regional winds in ground level. Hence, the transport upward of the regional pollutants to the summit was less frequent in spring than in summer, which resulted in lower concentrations of fine particles in spring. Besides, dust storms mostly occur in spring and MT is located on the very pathway of its long-range transport from central Asia to the northern America (Arimoto et al., 1996; Zhang et al., 1997; Sun et al., 2010). Thus, in those days with high wind speed in spring, higher ratio of those coarse particles to the total aerosol mass was often observed even at elevation of more than $1500\,\mathrm{m}$ .
3.1.2 Ionic and elemental composition of the particles at MT
Mass concentrations of ions in $\mathrm{PM}_{2.5}$ and TSP at MT are listed in Table 4. Water soluble ions contributed $10.82\,\%$ of TSP and $23.99\,\%$ of $\mathrm{PM}_{2.5}$ in mass in spring, while $40.89\,\%$ and $41.08\,\%$ in summer. This result revealed evidently that the secondary aerosol possessed much larger part in summer than in spring. The ratios of total ions to the total mass in $\mathrm{PM}_{2.5}$ were higher than those in TSP both in summer and spring, suggesting that the pollution components, such as $\bar{\mathrm{SO}}_{4}^{2-}$ , $\mathrm{NH}_{4}^{+}$ , $\mathrm{NO}_{3}^{-}$ , existed more in fine mode. These three major water soluble ions accounted for $61.50\,\%$ in spring and $72.65\,\%$ in summer of the total ions measured in $\mathrm{PM}_{2.5}$ , while $69.20\,\%$ and $71.47\,\%$ in TSP, respectively (Table 5). Also, higher concentration of $\Chi^{+}$ , the tracer of the biomass source, in summer was observed, which could account for $8.26\,\%$ of total ions in TSP. Whereas, higher concentration of $C\mathrm{a}^{2+}$ , likely more from the Asian dust source, in spring, accounted for $21.93\,\%$ of the total ions in TSP.
Water-soluble ions are proved to play key roles in many atmospheric processes, such as cloud formation, visibility degradation, solar radiation, acidification of cloud, rain, and fog, and haze formation because of their affinity with water (Tsai et al., 1999; Novakov et al., 1993; Matsumoto et al., 1997; Facchini et al., 2000). Haze mostly occurred in stable, warm, and moist air, and was characterized by the high level of fine particles with high percentage of certain water soluble ions, such as $\mathrm{SO}_{4}^{2-}$ , $\mathrm{NH}_{4}^{+}$ , and $\mathrm{NO}_{3}^{-}$ , in the aerosols. Certain characteristics of the aerosols collected at MT in summer and spring and at Beijing in dust, haze, and clear days are summarized in Table 5. $\mathrm{C_{IC}/C_{P}}$ , $\mathrm{C_{(S+N+A)}/C_{I C}}$ , SOR and NOR at MT in summer were very similar to those in haze days in Beijing, suggesting that haze could occur frequently in summer at the summit of MT, which could be proved by the lower visibility $(9.0\,\mathrm{km})$ on average at MT.
The concentrations of nineteen elements and black carbon (BC) in aerosols at MT are listed in Table 6, which showed clearly that crustal element (Ca, Ma, Al, Mn, Ti, Sr, and Na) were higher in spring, while pollution elements (Pb, Cr, Cd, Zn, Ni, S, BC), except As and $\mathrm{Cu}$ , were higher in summer. The elements could be classified into four groups according to their enrichment factors $(\mathrm{EFs}\,{=}\,(\mathrm{X/Al})_{\mathrm{aerosol}}/(\mathrm{X/Al})_{\mathrm{crust}})$ : high enriched pollution elements ( $\mathrm{Pb}$ and As), medium enriched (S and $Z\mathfrak{n}$ ), slightly enriched (Ni, Cu and $\mathrm{Cr}$ ), and non-enriched crustal metals (Ca, Ma, Al, Mn, Ti, Sr and Na). EFs of all crustal elements (see Fig. 3) were higher in TSP and lower in $\mathrm{PM}_{2.5}$ , while pollution elements were higher in $\mathrm{PM}_{2.5}$ and lower in TSP. EFs of As and $\mathbf{P}\mathbf{b}$ exhibited different seasonal variations with higher EF of $\mathbf{P}\mathbf{b}$ in summer and higher EF of As in spring.
3.1.3 Acidity of the aerosols at MT
Figure 4 shows the variations of $\mathrm{pH}$ of the filtrates of aerosols at MT, and it indicated that $\mathrm{pH}$ decreased obviously from spring to summer. The mean $\mathsf{p H}$ values of the aqueous filtrates of aerosols collected at MT and other sampling sites are compared in Table 7. The aerosols at MT showed higher acidity for both $\mathrm{PM}_{2.5}$ $\left(\mathrm{pH}\!=\!4.62\right)$ and TSP $\mathrm{(pH}\!=\!4.92)$ in summer in comparison with the weak acidity of $\mathrm{PM}_{2.5}$ $({\mathrm{pH}}\!=\!5.92)$ and slight alkalinity $({\mathrm{pH}}\!=\!7.22)$ of TSP in spring. The $\mathsf{p H}$ of aerosols at MT in summer was the lowest, while TSP in spring showed slight alkalinity, which was similar to those from Tazhong $({\mathrm{pH}}\!=\!7.39\$ , original source of dust storm) and from Beijing in supper dust day $(\mathrm{pH}\,{=}\,7.25\$ , Wang et al., 2005). The higher acidity of aerosols at MT in summer was further supported by the fact that the $\mathrm{pH}$ of rain samples in summer at MT was obviously lower than those in other seasons (Wang et al., 2006c). The measured acidity of aerosols is determined by the presence and the proportions of the cations and anions in the filtrates of aerosols. In addition, the ratio of equivalent concentrations of total cations to total anions (C/A) should be equal to 1, if all the cations and anions in the aerosol were measured. However, $\mathrm{CO}_{3}^{2-}$ and ${\mathrm{HCO}}_{3}^{-}$ were not measured in this study due to the limitation of ion chromatography, hence, the difference of the total anions and cations could be used to roughly estimate the amount of $\mathrm{CO}_{3}^{2-}$ and ${\mathrm{HCO}}_{3}^{-}$ unmeasured (Wang et al., 2005). C/A of aerosols in spring at MT reached 1.6, while only 1.1 in summer, suggesting that much more carbonate or bicarbonate presented in the aerosols at MT in spring. The remarkable increase of $\mathrm{SO}_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ and organic acids in summer shown in Table 4 indicated that in summer much more aerosols at MT was from secondary pollution, which would lead to the lower $\mathsf{p H}$ of the filtrates. Compared to the $\mathsf{p H}$ of those aerosols collected from Shanghai in summer (5.29 in $\mathrm{PM}_{2.5}$ and 6.37 in TSP), the aerosols at MT showed even higher acidity, suggesting that the secondary air pollution at MT in summer was even severer than that in Shanghai.
3.2.1 Influence of biomass burning
Biomass burning includes those from grassland, forest, and crop residue, among which more than $60\,\%$ are from crop residue burning in the world (Streets et al., 2003). With the rapid development of economy in China, crop residue is increasingly being burned openly in the field. In harvest every year $\sim\!5.182\times10^{7}$ ton crop residues (accounted for $40.0\,\%$ of the total crop residues in China) were directly burnt openly, especially in central eastern China, including Shandong, Jiangsu, Henan, Hebei provinces. Shandong, where MT is located, was in number 1 among those provinces mentioned above with the largest amount of crop residues $(1.798\times10^{7}\,\mathrm{t})$ to be burned openly (Cao et al., 2007). Biomass burning emissions are known to contribute a considerable amount of PM and gaseous pollutants to the air, and crop residues contribute more fine particles and gaseous than other biomaterial residues (Cao et al., 2005; Zhu et al., 2005).
Water soluble potassium is a good tracer for aerosols formed from biomass burning (Andreae, 1983). Sources of total potassium ion $(\mathrm{K}_{\mathrm{Total}}^{+}$ ) in aerosols might be attributed to soot from biomass burning, sea salt, and crustal dust. Elemental Al and $\mathrm{Na^{+}}$ could be the good markers of crustal dust and sea salt respectively, as the composition of Na, K, and Al in sea water and in crustal dust were found to be: Na: $31\,\%$ , $K{:}1.1\,\%$ , $\operatorname{Al}{\operatorname{:}}0\,\%$ , and Na: $2.6\,\%$ , $K{:}2.9\,\%$ , $A1:7.7\,\%$ , respectively (Wedepohl, 1995), which showed that the percentage of $\ K^{+}$ were much lower than $\mathrm{Na^{+}}$ in sea salt and also lower than Al in crustal dust. This means that the concentrations of $\mathrm{Na^{+}}$ and Al in the aerosols would increase more than $\mathbf{K}^{+}$ if the aerosols were from sea salt and crustal dust. Figure 5 showed that $\Chi^{+}$ on certain days in summer was significant higher than $\mathrm{Na^{+}}$ and Al with remarkable increase from the normal days, which suggested that the high concentrations of $\mathrm{PM}_{2.5}$ in summer could likely from biomass burning rather than sea water or crustal dust. It was reasonably assumed that the total potassium $(\mathrm{K}_{\mathrm{Total}}^{+})$ is tphoet assusimu mof $(\mathbf{K}_{\mathrm{SS}}^{+})$ d,e rainvde d bpiootmasassisu bmu $(\mathbf{K}_{\mathrm{Crust}}^{+})$ e, rsiveea-ds aplto-tdaesrsiivuemd $(\mathbf{K}_{\mathrm{BB}}^{+})$ ni. ec. $\mathrm{K_{Total}^{+}=K_{C r u s t}^{+}+K_{S S}^{+}+K_{B B}^{+}}$ (g Vpiroktaksusliau emt 6a)s, $(\mathrm{K}_{\mathrm{BB}}^{+})$ $\mathrm{K_{BB=}^{+}\ K_{T o t a l}^{+}{-K_{C r u s t}^{+}{-K_{S S}^{+}}}}$ . Assuming $\mathrm{Na^{+}}$ in aerosol is from $\mathrm{Na_{Total}^{+}}$ is the concentration $\mathrm{Na}_{\mathrm{Total}}^{+}\,{=}\,\mathrm{Na}_{\mathrm{Crust}}^{+}\,{+}\,\mathrm{Na}_{\mathrm{SS}}^{+}$ $\mathrm{Na^{+}}$ sample, and Al in aerosol is originated from crust only, thus, $\mathbf{K}_{\mathrm{Crust}}^{+}$ and KS+S can be estimated through the ratios of $\mathrm{K}^{+}/\mathrm{Al}$ and $\mathrm{Na^{+}/\bar{A l}}$ in the aerosol sample and the ratio of $\mathrm{K^{+}/N a^{+}}$ in sea salt, i.e. $\mathrm{K_{Crust}^{+}\!=\!(K^{+}/A l)_{c r u s t}\times A l_{c r u s t}}$ and $\mathrm{K}_{\mathrm{SS}}^{+}\,{=}\,(\mathrm{Na}_{\mathrm{Total}}^{+}-$ $(\mathrm{Na}^{+}/\mathrm{Al})_{\mathrm{crustl}}\times\mathrm{Al}_{\mathrm{crustl}})\times(\mathrm{K}^{+}/\mathrm{Na}^{+})_{\mathrm{sea}-\mathrm{salt}}$ . The ratio of $\mathrm{K^{+}/N a^{+}}$ in those particles from sea salt is constant (0.037) (Chester, 1990), whereas the ratios of $\mathrm{K}^{+}/\mathrm{Al}$ and $\mathrm{Na^{+}/A l}$ in crustal dust were different between the samples collected from different sites because of the different background and the different influence of the human activities on each site. We now have two methods to estimate the biomass burning derived potassium $(\mathbf{K}_{\mathrm{BB}}^{+})$ : (1) Using the minimum ratios of $\mathrm{K}^{+}/\mathrm{Al}$ of 0.152 and $\mathrm{Na^{+}/A l}$ of 0.240 in the $\mathrm{PM}_{2.5}$ aerosol sample collected at MT on 26 April 006 among all samples collected to calculate the background value of $\ K^{+}$ and $\mathrm{Na^{+}}$ in crustal dust, then the biomass burning derived potassium $(\mathbf{K}_{\mathrm{BB}}^{+})$ is given from $\mathrm{K_{BB=}^{+}~K_{T o t a l}^{+}}^{-}$ $\mathrm{K}_{\mathrm{Crust}}^{+}{-}\mathrm{K}_{\mathrm{SS}}^{+}$ where $\mathrm{K_{Crust}^{+}}\,{=}\,(\mathrm{K^{+}/A l)_{A e r o s o l}\,{\times}\,A l_{A e r o s o l}}$ l, AlAerosol is the Al concentration measured in $\mathrm{PM}_{2.5}$ , i.e. $\mathrm{K_{Crust}^{+}}\,{=}\,0.152\times\mathrm{Al_{Aerosol}}$ , and K+ S = (NaT+otal– $(\mathrm{Na^{+}/A l)_{A e r o s o l}\times A l_{A e r o s o l})\times(K^{+}/N a^{+})_{s e a-s a l t},}$ i.e. $\mathrm{K_{SS}^{+}=(N a_{T o t a l}^{+}-0.240\times A l_{A e r o s o l})\times0.037,}$ (2) Using the ratios of $\mathrm{K}^{+}/\mathrm{Al}$ of 0.107 and $\mathrm{Na^{+}/A l}$ of 0.031 measured in the soil samples collected at MT to calculate the background value of $\mathbf{K}^{+}$ and $\mathrm{Na^{+}}$ in crustal dust, then the biomass burning derived potassium is given from
$$
\mathrm{K_{BB=}^{+}K_{T o t a l}^{+}-K_{C r u s t}^{+}-K_{S S}^{+},w h e r e\;K_{C r u s t}^{+}}
$$
$=(\mathrm{K^{+}/A l})_{\mathrm{soil}}\times\mathrm{Al_{Aerosol}},$ ,i.e. $\mathrm{K_{Crust}^{+}{=}0.107\times A l_{A e r o s o l}}$ , and
$$
\mathrm{K_{SS}^{+}\!=\!(N a_{T o t a l}^{+}\mathrm{-}(N a^{+}/A l)_{s o i l}\times A l_{A e r o s o l})\times(K^{+}/N a^{+})_{s e a-s a l t},}
$$
The concentrations of $\mathrm{K_{BB}^{+}}$ (biomass burning-derived $\ K^{+}$ ) calculated with two methods mentioned above in $\mathrm{PM}_{2.5}$ collected at MT in 2006 are shown in Fig. 6. It can be seen clearly that very similar concentrations of $\mathrm{K_{BB}^{+}}$ were acquired by the two methods. The averaged $\mathrm{K_{BB}^{+}}$ B were 0.40 and $4.30\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in spring and summer, respectively, with the method of using the minimum ratios of $\mathrm{K}^{+}/\mathrm{Al}$ and $\mathrm{Na^{+}/A l}$ in the aerosol sample to calculate the background value of $\Chi^{+}$ and $\mathrm{Na^{+}}$ in crustal dust, while the $\mathrm{K_{BB}^{+}}$ were 0.32 and $4.30\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ with the method of using the ratios of $\mathrm{K}^{+}/\mathrm{Al}$ and Na+/Al in the soil samples. The high level of KB+B $(4.30\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ ) in summer indicated that there were the significant part of the aerosols from biomass burning emissions at MT during the summer time. This result could also be confirmed by the fire spot data in this region, as shown in Fig. 7a–f. The amount of those fire spots distributed in the region surrounding MT increases obviously in May (Fig. 7c), and highly active fire disturbance appeared in June (Fig. 7d). The fire spots spread mainly to the south to MT during $1-$ 9 June (Fig. 7e), and then extended to the northern places to MT during 10–19 June (Fig. 7f), which was in accordance to the harvest time of wheat/rice from south to north in this region.
The contributions of biomass burning, crustal dust, and other sources to the fine particles, $\mathrm{PM}_{2.5}$ , in spring and summer 2006 at MT are shown in Fig. 8. To assess quantitatively the contributions of biomass burning openly from the agriculture residues to the formation of $\mathrm{PM}_{2.5}$ at MT, we used the ratio of $\mathrm{K}^{+}/\mathrm{PM}_{2.5}$ of $9.56\,\%$ $(\mathrm{wt}\,\%)$ from the agriculture residues, i.e. biomass burning-derived $\mathrm{PM}_{2.5}\,{=}\,\mathrm{K}_{\mathrm{BB}}^{+}/0.0956,$ to calculate the mass of $\operatorname{PM}_{2.5}$ derived from biomass burning, for both Li et al. (2007) and Cao et al. (2008a) reported a very similar content of $\Chi^{+}$ in $\mathrm{PM}_{2.5}$ emitted from agriculture residues: $9.94\pm11.8$ or $9.56\pm9.01(\mathrm{wt}\,\%)$ from wheat straw, and $11.38\pm8.49(\mathrm{wt}\,\%)$ from rice straw. The crustal dust-derived $\mathrm{PM}_{2.5}$ was calculated with the formula of $\mathrm{Al/0.08}$ . The results revealed that the contribution of biomass burning to the fine particle at MT accounted for $7.56\,\%$ in spring and $36.71\,\%$ in summer, and even reached to $81.58\,\%$ on the day of 12 June. In addition, the concentrations of $\mathbf{K}^{+}$ at different sampling sites are summarized in Table 8. The concentration of $\Chi^{+}$ at MT was much higher than those at other sites, and it showed strong seasonal variation with $4.41\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in summer and $0.48\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in spring in $\mathrm{PM}_{2.5}$ . Also, $\Chi^{+}$ in $\mathrm{PM}_{2.5}$ correlated well to other species that are related to biomass burning, such as BC, $\bar{\mathrm{C}_{2}}0_{4}^{2-}$ , etc. (see Table 9). All of these results demonstrated evidently that biomass burning was one of major contributor to the aerosol pollution in summer over central eastern China, where MT is located.
3.2.2 Secondary components: $\mathbf{S0}_{4}^{2-}$ , $\mathbf{NO}_{3}^{-}$ , and $\mathbf{NH}_{4}^{+}$
The concentrations of the main secondary species $(\mathrm{SO}_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ , and $\mathrm{NH}_{4}^{+}$ ) in aerosols in spring and summer of 2006/2007 at different sites (MT, Urumqi, Beijing, and Shanghai) are illustrated in Fig. 9. The results showed that these secondary ions in aerosols in summer at MT were ${\sim}5\,$ times higher of those in spring, while there was no big change in Beijing, less change in Shanghai between the two seasons, and they were evidently greater in spring than those in summer at Urumqi. The sum of the three ions at MT in summer was higher than that at any other site, indicating much heavy secondary pollution at this region. Besides, at MT in summer the concentration of $\mathrm{SO}_{4}^{2-}$ was much greater than $\mathrm{NO}_{3}^{-}$ , while in Beijing and Shanghai the concentrations of $\mathrm{NO}_{3}^{-}$ was close to that of $\mathrm{SO}_{4}^{2-}$ . Average concentrations of $\mathrm{SO}_{4}^{2-}$ and $\mathrm{NO}_{3}^{-}$ in TSP at MT were 4.47 and $3.61\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in spring, 20.73 and $8.82\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in summer respectively, while in Shanghai 2.28 and $1.42\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in spring, 7.34 and $5.50\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , in summer 2006, respectively. During the study period, the average concentrations of $S O_{2}$ and $\mathrm{NO}_{2}$ were 46.0 and $24.0\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in spring, 34.0 and $26.0\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in summer over MT, while 51.0 and $41.0\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in spring and 33.0 and $19.0\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in summer in Shanghai. Mineral aerosols reacted with $S O_{2}$ or $\mathrm{NO}_{2}$ to form a layer of sulfate or nitrate on the mineral surfaces through the heterogeneous reactions (Yaacov et al., 1989), and soil particles would be coated by solutions contained sulfate and nitrate. The reaction of $S O_{2}$ on calcium-rich mineral aerosol was likely to play an important role in the downwind regions (Dentener et al., 1996), as many studies confirmed that the composition and morphology of dust particles would be changed during their transport (Underwood et al., 2001; Song et al., 2001). The conversion of $S O_{2}$ and $\mathrm{NO}_{2}$ to be $\mathrm{SO}_{4}^{2\bar{-}}$ and $\mathrm{NO}_{3}^{-}$ in ambient air could be their major source in aerosols. Sulfur oxidation ratio (SOR) and nitrogen oxidation ratio (NOR) can indicate the efficiency of these transformations. If SOR is smaller than 0.10, the $\mathrm{SO}_{4}^{2-}$ could be from the primary emissions (Pierson et al., 1979; Truex et al., 1980), otherwise, $\mathrm{SO}_{4}^{2-}$ was produced through the photochemical oxidation from $S O_{2}$ (Ohta et al., 1990). Though concentrations of the gases were basically in the same levels, the SOR and NOR were significantly different between MT and Shanghai, especially in summer. SOR and NOR in TSP increased from 0.09 and 0.10 in spring to 0.32 and 0.26 in summer at MT, while 0.06 and 0.10 to 0.12 and 0.16 in Shanghai, respectively, in the same sampling time. This result indicated that in summer at MT the transformation efficiency of $S O_{2}$ and $\mathrm{NO}_{2}$ to be $\mathrm{SO}_{4}^{2-}$ and $\mathrm{NO}_{3}^{-}$ was much higher than that in Shanghai, and it would result in much higher concentrations of sulfate and nitrate at MT than that in Shanghai.
Many factors were likely attributed to the more effective conversion of $S\mathrm{O}_{2}$ and $\mathrm{NO}_{2}$ to be $\mathrm{SO}_{4}^{2-}$ and $\mathrm{NO}_{3}^{-}$ on the summit of the mountain. Firstly, humidity plays a key role in the formation of sulfate from $S O_{2}$ , for $80{-}90\,\%$ of the global sulfate was produced in the aqueous-phase, (Jill et al.,
2001). At the summit of MT, total cloud days were 5.4 and 7.0 in spring and summer, respectively, and monthly average foggy days could reach 26, sometimes even 30 in summer, while only 10 in spring. The relatively high humidity at the summit in summer is in favor of the formation of sulfate. Secondary, oxidation of $\mathrm{SO}_{2}$ occurs potentially via three important pathways: oxidation by hydrogen peroxide or ozone, and auto-oxidation catalyzed by Fe (III) and Mn (II), and the former two were proved to be the dominative oxidation processes under certain conditions (Jill et al., 2001). High level of the biomass burning of agriculture residues resulted in increase of those gaseous pollutants, such as $\mathrm{O}_{3}$ , CO and VOCs in summer. Averaged concentration of total peroxide $(0.55\pm0.67$ ppbv) was significantly higher in summer than that $(0.17\pm0.26$ ppbv) in spring at MT (Ren et al., 2009). The ${\mathrm{O}}_{3}$ in 2003 (Gao et al., 2005) and 2006 (Table 3) at MT showed that $\mathrm{O}_{3}$ was also higher in summer than that in spring. High levels of ${\mathrm{O}}_{3}$ and peroxide in ambient air in summer accelerated the conversion of $S O_{2}$ and $\mathrm{NO}_{2}$ to be $\mathrm{SO}_{4}^{2-}$ and $\mathrm{NO}_{3}^{-}$ . Lastly, at MT there is abundance of hydrocarbon emitted by the abundant foliage that covers more than $90\,\%$ of the area of MT (Suthawaree, et al., 2010). $\mathrm{O}_{3}$ accelerated the formation of $\mathrm{SO}_{4}^{2-}$ by directly reacting with $S\mathrm{O}_{2}$ and generating OH radical that further to be transformed to peroxide via cooperating with hydrocarbon under light radiation. Hydrocarbon can react with OH radical to produce $\mathrm{HO}_{2}$ and $\mathrm{RO_{2}}(\mathrm{OH}+\mathrm{RH}\xrightarrow{[O_{2}]}\mathrm{RO_{2}}+\mathrm{H_{2}O},\mathrm{NO}+\mathrm{RO_{2}}\rightarrow\mathrm{RCHO}+$ ${\mathrm{HO}}_{2}+{\mathrm{NO}}_{2})$ ) (Ariel et al., 2000), and $\mathrm{HO}_{2}$ further react with $\mathrm{SO}_{2}$ and $\mathrm{NO}_{2}$ . Higher CO level produced from biomass burning in summer also contributed to the oxidation of $\mathrm{SO}_{2}$ by generating peroxide with OH radical $\mathrm{(OH\mathrm{+}C O\,\frac{\mathrm{[O_{2}]}}{\mathrm{\Delta}}H O_{2}\mathrm{+}C O_{2})}$ .The reactive processes above would benefit the formation of organic acids: $\mathrm{HO}_{2}+\mathrm{RO}_{2}\rightarrow\mathrm{ROOH}+\mathrm{O}_{2}$ . The high concentrations of diacids and the good correlations of diacids with $\mathrm{SO}_{4}^{2-}$ , as shown in Fig. 10, further suggested the formation mechanism of high level of the secondary aerosol in summer at MT. Besides, $\mathrm{NH_{4}^{+}}$ increased significantly from $1.48\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in spring to $10.4\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in summer, which was likely due to more pesticide sprayed in the large area of farming fields in late May and the stock-raise at MT region, which could produce much more $\mathrm{NH}_{3}$ in summer than in spring. Also, higher temperature in summer could lead to the higher emission of $\mathrm{NH}_{3}$ from animal’s excrements (Sacoby et al., 2007).
3.2.3 Mineral dust in spring time
Concentrations of mineral elements (such as Al, Fe, and Ca) are good indicators for crustal aerosols. Daily variation of mineral elements (Al, Fe, and Ca) at different sampling sites in spring, 2007 is shown in Fig. 11. It can be seen clearly that a strong dust storm occurred from 30 March to 2 April and resulted in sharp increase of the three mineral elements at all of these monitoring sites, especially at Tazhong, Yulin,
Duolun, MT, and Beijing. The ratio of Ca/Al has been proved to be a good tracer for different dust original areas (Wang et al., 2005; Sun et al., 2005), for $\mathrm{Ca/Al}$ showed remarkable dependence of the source regions for both dust aerosol and soil samples. Ratios of $\mathrm{Ca/Al}$ in aerosol or soil from different sampling sites are shown in Table 10, and it could be seen clearly that Taklimakan desert, the western source of Asian dust, is characterized by high ratio of $\mathrm{Ca/Al}$ of greater than 1.5, while the northern source of Asian dust, such as Duolun and Hunshandake Sandland, is characterized by low ratio of $\mathrm{Ca/Al}$ of around 0.5. The ratio of $\mathrm{Ca/Al}$ $(1.37\pm0.22)$ in the aerosol at MT indicated that the dust aerosol at MT in spring could be from the long-range transport of the dust from western or northwestern high-dust sources. The facts mentioned above that the coarse particles dominated in spring $\left(\mathrm{PM}_{2.5}/\mathrm{TSP}\right)$ of 0.37) and the alkalinity $({\mathrm{pH}}\!=\!7.22)$ of the TSP in spring further confirmed that the long-range transport was the major source of the aerosols in spring at MT.
3.2.4 The pollution elements As and Pb
The pollution elements, As and $\mathbf{P}\mathbf{b}$ , were highly enriched in the aerosols with the EF of 1541 and 679, respectively, in spring, while 1470 and 1969 in summer in $\mathrm{PM}_{2.5}$ at MT. As and $\mathbf{P}\mathbf{b}$ were of different seasonal variations with summerhigh/spring-low of $\mathbf{P}\mathbf{b}$ , while spring-high/summer-low of As. Correlation analysis (see Table 11) revealed that both $\mathbf{P}\mathbf{b}$ and As were highly correlated to those crustal elements in spring with the correlation coefficients of 0.701 for $\mathbf{P}\mathbf{b}$ to Al and 0.873 for $\mathbf{P}\mathbf{b}$ to Fe, while of 0.837 for As to Al and 0.778 for As to Fe. Furthermore, the correlation coefficient between $\mathbf{P}\mathbf{b}$ and As was as high as 0.949. These results indicated that both $\mathbf{P}\mathbf{b}$ and As were highly associated with the mineral components, which was long-range transported from northwestern China to MT in spring. Those primary dust aerosols from northwestern China would mix with As and $\mathbf{P}\mathbf{b}$ , which were emitted from coal-mining/coal-ash and widely distributed over northwestern China, and act as a carrier for As and Pb. The pollution elements, As and Pb, would gradually be enriched in the dust aerosol during its longrange transport. The results revealed that the long-range transport of aerosols spread the heavy pollution from coal burning everywhere over China. However, in summer the correlation coefficients of $\mathbf{P}\mathbf{b}$ to Al and $\mathbf{P}\mathbf{b}$ to Fe decreased to $-0.184$ and $-0.194$ , while of As to Al and As to Fe to 0.469 and 0.456, respectively, and the correlation coefficient between $\mathbf{P}\mathbf{b}$ and As decreased to 0.494. These results indicated that in summer the source of $\mathbf{P}\mathbf{b}$ and As was not mainly from long-range transport, instead, they could be from those local/regional sources. It could be seen that $\mathbf{P}\mathbf{b}$ and As were both correlated well to Cr, Cu, and $Z\mathfrak{n}$ in summer, suggesting that the source of $\mathbf{P}\mathbf{b}$ and As in summer would likely from the same local/reginal anthropogenic discharge as the pollution elements, Cr, Cu, and $Z\mathfrak{n}$ , did. The strong convections of air mass would result in the elevation to the summit of MT from the ground level of the pollutants. The emission of $\mathbf{P}\mathbf{b}$ from the local/regional anthropogenic source in summer could be much more than that from the long-range transport in spring, which led to the higher enrichment of $\mathbf{P}\mathbf{b}$ in summer than that in spring.
4 Summary
Aerosols over central eastern China showed significantly season variation, with fine particles dominated in summer while coarse particles in spring. The summit of MT was suffering from the invasion of long-range transported dust from northwestern China, the heavy local/regional anthropogenic emissions from the surrounding areas, and the severe secondary pollutions. The contribution of biomass burning to the fine particle at MT accounted for $7.56\,\%$ in spring and $36.71\,\%$ in summer, and even reached to $81.58\,\%$ on a day. The high levels of peroxides and ozone, as well as the favourable meteorological conditions in the ambient air were in favour of the transformation of $S\mathrm{O}_{2}$ and $\mathrm{NO_{x}}$ to sulfate and nitrate, which resulted in the even severe secondary pollution in this region. As and $\mathbf{P}\mathbf{b}$ were two of the most enriched elements. The long-range transport of aerosols spread the heavy pollution from coal-mining/coal-ash to everywhere over China. Anthropogenic air-pollution was evidently rather severe at MT.
Acknowledgements. This work was supported by the great international collaboration project of MOST, China (2010DFA92230), the National Key Project of Basic Research of China (Grant Nos. 2006CB403704), and National Natural Science Foundation of China (Grant Nos. 20877020, 20977017). The authors would like to acknowledge Zhihua Wang of Shangdong University for providing the data of ${\mathrm{O}}_{3}$ and CO.
Edited by: S. C. Liu
References
Andreae, M. O.: Soot carbon and excess fine potassium: long-range transport of combustion-derived aerosols, Science, 220, 1148– 1151, 1983.
Ariel, F. S. and Lamb, D.: The sensitivity of sulfer wet deposition to atmospheric oxidants, Atmos, Environ., 34, 1681–1690, 2000.
Arimoto, R.: Eolian dust and Climate: Relationship to sources, tropospheric chemistry, transport and deposition, Earth Sci. Rev., 54, 29–42, 2001.
Arimoto R., Duce R. A., and Savoie D. L.: Relationships among aerosol constituents from Asia and the North Pacific during PEM-West, J. Geophys. Res. Atmos., 101(D1), 2011–2023, 1996.
Cao, G., Zhang, X., Wang, D., and Zheng, F.: Inventory of atmospheric pollutants discharged from biomass burning in China continent, China Environ. Sci., 25(4), 389–393, 2005 (in Chinese).
Cao, G., Zhang, X., Wang, Y., and Zheng, F.: Estimation of the emission from open burning of crop residues in China continent, Chin. Sci. Bull., 52(18), 1826–1831, 2007 (in Chinese).
Cao, G., Zhang, X., Gong, S., and Zheng, F.: Investigation on emission factors of particulate matter and gaseous pollutants from crop residue burning, Environ. Sci., 20, 50–55, 2008a.
Cao, J. J., Chow, J. C., Watson, J. G., Wu, F., Han, Y. M., Jin, Z. D., Shen, Z. X., and An, Z. S.: Size-differentiated source profiles for fugitive dust in the Chinese Loess Plateau, Atmos. Environ., 42(10), 2261–2275, 2008b.
Chan, Y. C., Simpson, R. W., Mctainsh, G. H., Vowles, P. D., Cohen, D. D., and Bailey, G. M.: Source apportionment of visibility degradation problems in Brisbane (Australia) using the multiple linear regression technique, Atmos. Environ., 33, 3237–3250, 1999.
Chester, R.: Marine Geochemistry, Unwin Hyman, London, 698 pp., 1990.
Dentener, F. J., Carmichael, G. R., and Zhang, Y.: Role of mineral aerosol as a reactive surface in the global troposphere, J. Geophys. Res. Atmos., 101(D17), 22869–22889, 1996.
Dockery, D. W., Pope, A. C., Xu, X., Spengler, J. D., Ware, J. H., Fay, M. E., Ferris Jr., B. G., and Speizer, F. E.: An association between air pollution and mortality in six US cities, New Engl. J. Med., 1753–1759, 1993.
Facchini, M. C. S., Decesari, M., Mircea, S., Fuzzi, and Loglio, G.: Surface tension of atmospheric wet aerosol and cloud/fog droplets in relation to their organic carbon content and chemical composition, Atmos. Environ., 34, 4853–4857, 2000.
Gao, J., Wang, T., Ding, A., and Liu, C.: Observational study of ozone and carbon monoxide at the summit of mount Tai ( $1534\,\mathrm{m}$ a.s.l.) in central-eastern China, Atmos. Environ., 39(20), 4779–4791, 2005.
Guo, J., Kenneth, A., Rahn, and Zhuang, G.: A mechanism for the increase of pollution elements in dust storms in Beijing, Atmos. Environ., 38, 855–862, 2004.
Han, L., Zhuang G., Sun Y., and Wang, Z. F.: Local and non-local sources of airborne particulate pollution at Beijing, Science in China Ser. B Chemistry, 48(4), 1–12, 2005.
Jill, E. R., Oliver, V. R., Katharine, F. M., Christopher J., Eli, D. S., Vincent, A. D., Sonia, M. K., Liaquat, H., and Collett Jr., J.: Drop size-dependent S (IV) oxidation in chemically heterogeneous radiation fogs, Atmos. Environ., 35, 5717–5728, 2001.
Kaufman, Y. J., Tanre, D., and Boucher, O.: A satellite view of aerosols in the climate system, Nature, 419, 215–223, 2002.
Li, X., Wang, S., Duan, L., Hao, J., Li, C., Chen, Y., and Yang, L.: Particulate and Trace Gas Emissions from Open Burning of Wheat Straw and Corn Stover in China, Environ. Sci. Technol., 41(17), 6052–6058, 2007.
Liu, C. L., Zhang, J., and Liu, S. M.: Physical and chemical characters of materials from several mineral aerosol sources in China, Environ. Sci., 23, 28–32, 2002.
Matsumoto, K., Tanaka, H., Nagao, I., and Ishizaka, Y.: Contribution of particulate sulfate and organic carbon to cloud condensation nuclei in the marine atmosphere, Geophys. Res. Lett., 24, 655–658, 1997.
Novakov, T. and Penner, J. E.: Large condensation of organic aerosols to cloud-condensation nuclei concentration, Nature, 365, 823–826, 1993.
Novelli, P. C., Masarie, K. A., Tans, P. P., and Lang, P. M.: Recent changes in atmospheric carbon monoxide, Science, 263, 1587– 1590, 1994.
Novelli, P. C., Masarie, K. A., and Lang, P. M.: Distributions and recent changes of carbon monoxide in the lower troposphere, J. Geophys. Res., 103(D51), 19015–19033, 1998.
Ohta, S. and Okita, T.: A chemical characterization of atmospheric aerosol in Sapporo, Atmos. Environ., 24A, 815–822, 1990.
Pierson, W. R., Brachaczek, W. W., and Mckee, D. E.: Sulfate emissions from catalyst equipped automobiles on the highway, J. Air Pollut. Control Assoc., 29, 255–257, 1979.
Rastogi, N. and Sarin M. M.: Long-term characterization of ionic species in aerosols from urban and high altitude sites in western India: Role of mineral dust and anthropogenic sources, Atmos. Environ, 39, 5541–5554, 2005.
Rastogi, N. and Sarin, M. M.: Chemistry of aerosols over a semiarid region: Evidence for acid neutralization by mineral dust, J. Geophys. Res. Lett., 33, L23815, doi:10.1029/2006GL027708, 2006.
Ren, Y., Ding, A., Wang, T., Shen, X., Guo, J., Zhang, J., Wang, Y., Xu, P., Wang, X., Gao, J., Jeffrey, L., and Collett, J.: Measurement of gas-phase total peroxides at the summit of Mount Tai in China, Atmos. Environ., 43, 1702–1711, 2009.
Sacoby, M. W. and Marc, L.: Examination of atmospheric ammonia levels near hog CAFOs, homes, and schools in Eastern North Carolina, Atmos. Environ., 41, 4977–4987, 2007.
Shen, Z. X., Cao, J. J., Li, X. X., Wang, Y. Q., Jie, D. M., and Zhang, X. Y.: Chemical characteristics of aerosol particles $(\mathbf{PM}_{2.5})$ at a site of Horqin Sand-land in northeast China, Environmental Sciences-China, 18(4), 701–707, 2006.
Song, C. H. and Carmichael, G. R.: A three-dimensional modeling investigation of the evolution processes of dust and sea-salt particles in East Asia, J. Geophys. Res.-Atmos., 106(D16), 18131– 18154, 2001.
Streets, D. G., Yarber, K. F., Woo, J. H., and Carmichael, G. R.: Biomass burning in Asia: Annual and seasonal estimates and atmospheric emissions, Global Biogeochem. Cy., 17(4), 1099, doi:10.1029/2003GB002040, 2003.
Sun, Y., Zhuang, G., Wang, Y., Han, L., Guo, J., Dan, M., Zhang, W., Wang, Z., and Hao, Z.: The air-borne particulate pollution in Beijing-concentration, composition, distribution and sources, Atmos. Environ., 38, 5991–6004, 2004.
Sun, Y., Zhuang, G., Wang, Y., Zhao, X., Li, J., Wang, Z., and An, Z.: Chemical composition of dust storms in Beijing and implications for the mixing of mineral aerosol with pollution aerosol on the pathway, J. Geophys. Res., 110, D24209, doi:10.1029/2005JD006054, 2005.
Sun, Y., Zhuang, G., Huang K., Li J., Wang, Q., Wang Y., Lin, Y., Fu, J., Zhang, W., Tang, A., and Zhao, X.: Asian dust over northern China and its impact on the downstream aerosol chemistry in 2004, J. Geophys. Res., 115, D00K09, doi:10.1029/2009JD012757, 2010.
Suthawaree, J., Kato, S., Okuzawa, K., Kanaya, Y., Pochanart, P., Akimoto, H., Wang, Z., and Kajii, Y.: Measurements of volatile organic compounds in the middle of Central East China during Mount Tai Experiment 2006 (MTX2006): observation of regional background and impact of biomass burning, Atmos. Chem. Phys., 10, 1269–1285, doi:10.5194/acp-10-1269-2010, 2010.
Tegen, I., Lacis, A. A., and Fung, I.: The influence of climate forcing of mineral aerosols from disturbed soils, Nature, 380(4), 419–422, 1996.
Truex, T. J., Pierson, W. R., and Mckee, D. E.: Sulfate in diesel exhaust. Environ. Sci. Technol., 14, 1118–1121, 1980.
Tsai, Y. I. and Cheng, M. T.: Visibility and aerosol chemical compositions near the coastal area in central Taiwan, Sci. Total Environ., 231, 37–51, 1999.
Underwood, G. M., Li, P., Al-Abadleh, H., and Grassian, V. H.: A Knudsen cell study of the heterogeneous reactivity of nitric acid on oxide and mineral dust particles, J. Phys. Chem. A., 105(27), 6609–6620, 2001.
Virkkula, A., Teinila¨, K., Hillamo, R., Kerminen, V.-M., Saarikoski, S., Aurela, M., Viidanoja, J., Paatero, J., Koponen, I. K., and Kulmala, M.: Chemical composition of boundary layer aerosol over the Atlantic Ocean and at an Antarctic site, Atmos. Chem. Phys., 6, 3407–3421, doi:10.5194/acp-6-3407-2006, 2006.
Wang, T., Cheung, A., Vincent, T. F., Anson, M., and Li, Y. S.: Ozone and related gaseous pollutants in the boundary layer of eastern China: overview of the recent measurements at a rural site, Geophys. Res. Lett., 28(12), 2373–2376, 2001.
Wang, Y., Zhuang, G., Sun, Y., and An, Z.: Water-soluble part of the aerosol in the dust storm season – evidence of the mixing between mineral and pollution aerosols, Atmos. Environ., 39, 7020–7029, 2005.
Wang, Y., Zhuang, G., Sun Y., and An, Z.: The variation of characteristics and formation mechanisms of aerosols in dust, haze, and clear days in Beijing, Atmos. Environ., 40, 6579–6591, 2006a.
Wang, Y., Zhuang, G., Zhang, X., Huang, K., Xu, C., Tang, A., Chen, J., and An, Z.: The ion chemistry, seasonal cycle, and sources of $\mathrm{PM}_{2.5}$ and TSP aerosol in Shanghai, Atmos. Environ., 40, 2935–2952, 2006b.
Wang, Y., Ge, F., Liu, X., Wang, W., Jia, H., and Wang, D.: Analysis of the ionic characteristics for wet droplets at Mountain Tai, China Environ. Sci., 26(4), 422–426, 2006c (in Chinese).
Wang, Y., Wai, K., Gao, J., Liu, X., Wang, T., and Wang, W.: The impacts of anthropogenic emissions on the precipitation chemistry at an elevated site in North-eastern China, Atmos. Environ., 42, 2959–2970, 2008.
Wedepohl, K. H.: The compositions of the continental crust, Geochim. Cosmochim. Acta, 59, 1217–1232, 1995.
Yaacov, M. and Judith, G.: Heterogeneous reactions of minerals with sulfur and nitrogen oxides, J. Aerosol Sci., 20(3), 303–311, 1989.
Yuan, H., Wang, Y., and Zhuang, G.: The simultaneous determination of organic acid, MSA with inorganic anions in aerosol and rainwater by ion chromatography, J. Inst. Anal., 6, 12–16, 2003 (in Chinese).
Zhang, X. Y., Zhuang, G. Y., Zhu, G. H., Zhang, D., An, Z. S., Chen, T., and Huang, X. P.: Element tracers for Chinese source dust, Science in China (Series D), 39(5), 512–521, 1996.
Zhang, X. Y., Arimoto, R., and An, Z. S.: Dust emission from Chinese desert sources linked to variations in atmospheric circulation, J. Geophys. Res.-Atmos., 102(D23), 28041–28047, 1997.
Zhang, D., Zang, J., Shi, G., Iwasaka, Y., Matsuki, A., and Trochkine, D.: Mixture state of individual Asian dust particles at a coastal site of Qingdao, China, Atmos. Environ., 37(28), 3895– 3901, 2003.
Zhu, B., Zhu, X., Zhang, Y., Zeng, L., and Zhang, Y.: Emission factors of $\mathrm{PM}_{2.5}$ from crop straw burning, Res. Environ. Sci., 18(2), 29–33, 2005 (in Chinese).
Zhuang, G., Guo, J., Yuan, H., and Zhao, C.: The compositions, sources, and size distribution of the dust storm from China in spring of 2000 and its impact on the global environment, Chin. Sci. Bull., 46(11), 895–900, doi:10.1007/BF02900460, 2001.
|
atmosphere
|
0011
|
10.1016/j.atmosres.2017.11.017
|
image
|
Fig. 1. Location of the Lin'an regional atmospheric background station.
|
|
atmosphere
|
0011
|
10.1016/j.atmosres.2017.11.017
|
image
|
Fig. 2. Concentrations of $\mathrm{PM}_{2.5}$ and its major components at LA during the study period.
|
|
atmosphere
|
0011
|
10.1016/j.atmosres.2017.11.017
|
image
|
Fig. 3. Correlations between measured and reconstructed $\mathrm{PM}_{2.5}$ concentrations with the three models.
|
|
atmosphere
|
0011
|
10.1016/j.atmosres.2017.11.017
|
table
|
Table 1 Correlations between measured and reconstructed concentrations of major $\mathrm{PM}_{2.5}$ chemical components by the three models.
|
|
atmosphere
|
0011
|
10.1016/j.atmosres.2017.11.017
|
table
|
Table 2 Results of the PCA analysis for $\mathrm{PM}_{2.5}$
|
|
atmosphere
|
0011
|
10.1016/j.atmosres.2017.11.017
|
table
|
Table 3 UNMIX source composition profile for $\mathrm{PM}_{2.5}$
|
|
atmosphere
|
0011
|
10.1016/j.atmosres.2017.11.017
|
text
| Not supported with pagination yet |
Source apportionment of $\mathbf{PM}_{2.5}$ at the Lin'an regional background site in China with three receptor models
Junjun Denga,b,⁎, Yanru Zhanga,b,c, Yuqing $\mathrm{\boldmath~Qiu^{a,b,c}~}$ , Hongliang Zhangd, Wenjiao Dua,b,c, Lingling $\mathrm{{Xu^{a,b}}}$ , Youwei Honga,b, Yanting Chena,b, Jinsheng Chena,b,⁎
a Center for Excellence in Regional Atmospheric Environment, Institute of Urban Environment, Chinese Academy of Sciences, Xiamen 361021, China b Key Laboratory of Urban Environment and Health, Institute of Urban Environment, Chinese Academy of Sciences, Xiamen 361021, China c University of Chinese Academy of Sciences, Beijing 100086, China d Department of Civil and Environmental Engineering, Louisiana State University, Baton Rouge, LA 70803, USA
A R T I C L E I N F O
A B S T R A C T
Keywords:
PM2.5
Source apportionment
Background area
PCA-MLR
UNMIX
PMF
Source apportionment of fine particulate matter $(\mathbf{PM}_{2.5})$ were conducted at the Lin'an Regional Atmospheric Background Station (LA) in the Yangtze River Delta (YRD) region in China from July 2014 to April 2015 with three receptor models including principal component analysis combining multiple linear regression (PCA-MLR), UNMIX and Positive Matrix Factorization (PMF). The model performance, source identification and source contribution of the three models were analyzed and inter-compared. Source apportionment of $\mathbf{PM}_{2.5}$ was also conducted with the receptor models. Good correlations between the reconstructed and measured concentrations of $\mathbf{PM}_{2.5}$ and its major chemical species were obtained for all models. PMF resolved almost all masses of $\mathrm{{PM}}_{2.5}.$ while PCA-MLR and UNMIX explained about $80\%$ . Five, four and seven sources were identified by PCA-MLR, UNMIX and PMF, respectively. Combustion, secondary source, marine source, dust and industrial activities were identified by all the three receptor models. Combustion source and secondary source were the major sources, and totally contributed over $60\%$ to $\mathbf{PM}_{2.5}$ . The PMF model had a better performance on separating the different combustion sources. These findings improve the understanding of $\mathbf{PM}_{2.5}$ sources in background region.
1. Introduction
Due to the rapid growth of economy and urbanization and lack of emission control in recent decades, air pollution becomes a severe environmental problem in China (Chan and Yao, 2008; He et al., 2002; Huo et al., 2014). In particular, particulate matter with dynamic diameter $\leq2.5\,\upmu\mathrm{m}\,\left(\mathrm{PM}_{2.5}\right)$ has become the dominant air pollutant in developed and industrialized areas, such as the Beijing-Tianjin-Hebei (BTH) region, the Yangtze River Delta (YRD) and the Guanzhong Plain (Cao et al., 2012; Ding et al., 2013; Niu et al., 2016; Shen et al., 2014; Wang et al., 2014; Wang et al., 2015). $\mathrm{PM}_{2.5}$ has significant adverse influences on visibility, climate change and human health (Brunekreef and Holgate, 2002; Chen et al., 2014; Deng et al., 2016; Poeschl, 2005). Source apportionment of $\mathbf{PM}_{2.5}$ is an important research topic to provide scientific supporting for emissions control to reduce $\mathrm{PM}_{2.5}$ and improve air quality.
Contributions of different sources to $\mathrm{PM}_{2.5}$ have been frequently reported to help control anthropogenic emissions for urban areas in China during recent years (Chen et al., 2015, 2017; Dai et al., 2013; Liu et al., 2017; Song et al., 2006; Tao et al., 2016, 2017; Wang et al., 2012; Zhang et al., 2013). In addition, source apportionment of $\mathrm{PM}_{2.5}$ at background areas can provide deep insights into the influences of emission sources and formation mechanism of $\mathrm{PM}_{2.5}$ on a regional scale. However, in rural or remote regions with different $\mathbf{PM}_{2.5}$ characteristics from urban areas, rare source apportionment studies have been conducted. For example, Positive Matrix Factorization (PMF) linked with radiocarbon analysis was applied to identify the contribution of biomass burning to $\mathrm{PM}_{2.5}$ at a regional background site in North China, and revealed that coal combustion, biomass burning and vehicle emission were the main contributors of $\mathrm{PM}_{2.5}$ (Zong et al., 2016). Liu et al. (2013) investigated the sources of carbonaceous aerosols in $\mathrm{PM}_{2.5}$ at a regional background site in Ningbo in East China using levoglucosan and radiocarbon, and found that biomass burning and biogenic sources were the major contribution to the water-insoluble organic carbon, whereas fossil fuel was the dominant contributor to the refractory elemental carbon.
Receptor models are widely used to perform source apportionment of $\mathrm{PM}_{2.5}$ by utilizing information from measurements of $\mathrm{PM}_{2.5}$ chemical components (Belis et al., 2013; Watson and Chow, 2005). Chemical Mass Balance (CMB) model is one class of receptor models, which requires the input of both concentrations of chemical species in $\mathbf{PM}_{2.5},$ and chemical fingerprints of the sources (source profiles). The other class is multivariate factor analysis models, such as Principal Component Analysis-Multiple Linear Regression (PCA-MLR), UNMIX, and PMF (Banerjee et al., 2015; Belis et al., 2013, 2015; Contini et al., 2010; Hopke, 2003; Hopke et al., 2006; Masiol et al., 2017; Ogundele et al., 2016; Shi et al., 2014; Tao et al., 2017; Viana et al., 2008; Zhang et al., 2017). The factor analysis models only use the concentrations of chemical species as input and do not need the input of source profiles. Due to the differences in theoretical approaches and/or local specificity of monitoring sites, the results from different receptor models have large variability as the number of sources and source types are determined by experiences. To provide more robust source contributions and better interpret the results, studies usually simultaneously apply different receptor models on same datasets (Cesari et al., 2016a, 2016b; Hopke et al., 2006; Ogundele et al., 2016; Pekney et al., 2006; Poirot et al., 2001; Shi et al., 2014; Song et al., 2006). For example, Cesari et al. (2016b) estimated the source contributions in an industrial area by performing inter-comparison of PMF and CMB outputs, and found that the inter-comparison of PMF and CMB gave significant differences for secondary nitrate, biomass burning and industrial sources since the source profiles had local specificities.
To take advantage of multiple receptor models, source apportionment of $\mathrm{PM}_{2.5}$ was performed at the Lin'an (LA) site in the regional background area of the YRD. Three multivariate factor receptor models including PCA-MLR, UNMIX and PMF were adopted to the same dataset for contributions of different sources to $\mathrm{PM}_{2.5}$ . This study is aiming to evaluate and inter-compare the three receptor models by comparing their results. Another goal of this work is to identify the contributions of the sources to $\mathrm{PM}_{2.5}$ at the background site with the receptor models. Knowledge obtained from this research would provide valuable information for designing emission control policies to improve local and regional air quality.
2. Experimental methods
2.1. Site description
$\mathrm{PM}_{2.5}$ sampling was conducted on the rooftop at the LA background site $.119^{\circ}44^{\prime}\,\mathrm{~.~}$ E, $30^{\circ}18^{\prime}\,\mathrm{N}$ , $138.6\;\mathrm{m}$ a.s.l.). The site is suited in the Lin'an city (with a population of $\sim\!57$ thousand) in Zhejiang province in the southern edge of the YRD region (Fig. 1). Located in eastern China and on the western coast of the Pacific Ocean, the YRD covers about $100,000\,\mathrm{km}^{2}$ with a population of 75 million. As the largest estuary delta in China, the YRD has the fastest economy growing rate in China and experiences serious $\mathrm{PM}_{2.5}$ pollution due to the resultant large emissions of pollutants (Li et al., 2011; Shen et al., 2014; Tie et al., 2006; Wang et al., 2016). LA site is one of the World Meteorological Organization Global Atmosphere Watch (WMO/GAW) network stations in China, which is $53\,\mathrm{km}$ west of Hangzhou and $210\,\mathrm{km}$ southwest of Shanghai, respectively. To the west and farther south are mountainous regions with small populations. This background site is surrounded by hills covered by vegetation. There are only a few small villages within $3\,\mathrm{km}$ . Within $10\,\mathrm{km}$ , there are some small industrial plants such as cement kilns and brick kilns, and agriculture-related biomass burning sources have been reported (Wang et al., 2004). The potential distant emission source regions are Ningbo-Shaoxing region (southeast of Lin'an city), a developed and industrialized area in Zhejiang province, and the Suzhou-Wuxi-Changzhou region, located within $100\,\mathrm{km}$ and the most developed/industrialized region in Jiangsu province (Feng et al., 2015). Accordingly, the LA site well represents a background region of $\mathbf{PM}_{2.5}$ .
2.2. $P M_{2.5}$ sampling and analysis
The sampling was conducted during four seasons in 2014–2015: summer (July–August 2014), autumn (October–November 2014), winter (December 2014–January 2015), and spring (March–April 2015). Detailed sampling dates of each sample are presented in Table S1 in Zhang et al. (2017). Daily $\mathrm{PM}_{2.5}$ samples were collected from $9{\cdot}00\;\mathrm{a.m}$ . to $8{\cdot}30\;\mathrm{a.m}$ . the next day with two mini-volume air samplers (Airmetrics, USA) at the flow rates of $5\,\mathrm{L}\mathrm{min}^{-1}$ . One was collected on $47{-}\mathrm{mm}$ quartz-fiber filters (Whatman, UK) for water-soluble inorganic ions and carbonaceous components analysis and the other was collected on $47{-}\mathrm{mm}$ poly‑carbonate filters (Whatman, UK) for elemental analysis. Totally 80 pairs of samples were collected during the sampling period. All quartz filters were previously annealed for $^{5\,\mathrm{h}}$ in a muffle furnace with the temperature of $450~^{\circ}\mathrm{C}$ to remove the organic matters before sampling. All filters were equilibrated in a chamber with controlled temperature $(25\ ^{\circ}\mathrm{C})$ and relative humidity $(50\%)$ for $24\,\mathrm{{h}}$ and then weighted twice on an electronic microbalance with $\pm\ 0.01\ \mathrm{mg}$ sensitivity (BT-125D, Sartorius, Germany) before and after sampling. The samples and blank filters were then stored in a refrigerator at $-\,15\,^{\circ}\mathrm{C}$ before chemical analysis to prevent evaporation of volatile components.
Elemental carbon (EC) and organic carbon (OC) in $\mathrm{PM}_{2.5}$ were analyzed with a carbon analyzer (Sunset Model-4, USA) with the thermal-optical transmittance (TOT) following the National Institute for Occupational Safety and Health (NIOSH) protocol (Birch and Cary, 1996). A $1.5\,\mathrm{cm}^{2}$ filter punch was placed in an oven and heated in a pure helium (He) atmosphere up to $850~^{\circ}\mathrm{C}$ to convert all organic compounds into $\mathrm{CO}_{2}$ by a manganese dioxide $\mathrm{(MnO_{2})}$ catalyst. $\mathrm{CO}_{2}$ was then detected by a self-contained non-dispersive infrared (NDIR) system. In this stage, some organic compounds were pyrolytically converted to EC. The oven was cooled to $550~^{\circ}\mathrm{C}$ to evolve EC and the pyrolytically generated EC (or char), before being heated until the temperature steps up back to $850~^{\circ}\mathrm{C}$ under a mixed atmosphere $.98\%$ ${\mathrm{He}}\,+\,2\%\,{\mathrm{O}}_{2})$ . OC and EC split was corrected as the laser transmittance returned to the initial value. The analyzer was calibrated with standard sucrose solutions every day. The method detection limits (MDLs) for OC and EC were $0.1\ \upmu\mathrm{g}\ \mathrm{m}^{-\,3}$ and $0.05\,\upmu\mathrm{g}\,\up m^{\,-\,3}$ , respectively. For quality assurance and quality control (QA/QC), blank samples were analyzed according to the same procedures to correct ambient samples by subtracting the blank values from sample concentrations.
Nine water-soluble inorganic ions, including $\mathrm{NH_{4}}^{+}$ , $\mathbf{Na}^{+},\ \mathbf{K}^{+}$ , ${\mathrm{M}}{\mathrm{g}}^{2\,+}$ , ${\mathrm{Ca}}^{2\,+}$ , ${\mathrm{SO}}_{4}^{\ 2-}$ , $\mathrm{NO}_{3}{}^{-}$ , $\mathrm{Cl}^{-}$ , and $\mathrm{F}^{-}$ , in aqueous extracts of the $\mathrm{PM}_{2.5}$ samples were detected by an ion chromatography (IC) analyzer (ICS-3000, Dionex Inc., USA). One-fourth of each filter was extracted with $10\;\mathrm{mL}$ ultrapure water $.18.2\:\mathrm{M}\Omega{\cdot}\mathrm{cm}$ , Milli-Q Advantage, Millipore, USA) in $15\,\mathrm{mL}$ polypropylene tube via ultrasonic bath for $^{1\,\mathrm{h}}$ . The extracts were then filtered with a $0.22\,\upmu\mathrm{m}$ filter and stored in a $2\,\mathrm{mL}$ pre-cleaned sample vials for analysis. Cations were analyzed with an IonPac CSRS-4 suppressor with an IonPac $\mathrm{CS12A\times250\;mm}$ analytical column and anions were analyzed with an IonPac ASRS-4 suppressor with an IonPac $\mathrm{AS11HC}\,\times\,250\,\mathrm{mm}$ analytical column. Eluents for cations and ions were methane sulfonic acid (MSA) with concentration of $20\;\mathrm{mM}$ and potassium hydroxide (KOH) with concentration of $10\,\mathrm{Mm}_{\mathrm{}}$ , respectively. MDLs were 0.001 to $0.02\,\upmu\mathrm{g}\,\upmu^{-\,3}$ for anions and 0.004 to $0.014\,\upmu\mathrm{g}\,\mathfrak{m}^{-\,3}$ for cations. The analysis of ions had achieved recovery rates of $80–110\%$ . Blank values were also subtracted from sample concentrations.
Major elements including Al, Si, Ti, Mn, Fe, Cu, Zn and As were determined by wave dispersive X-ray fluorescence (WD-XRF) spectrometry (Axios-MAX, Panalytical, Holland). QA/QC of WD-XRF measurement was guaranteed by the analysis of a certified standard, Standard Reference Material (SRM 2783) of National Institute of Standards & Technology (NIST). A field blank consisting of a poly‑carbonate filter sample was also analyzed to evaluate analytical bias. Generally, MDLs for elements were 0.001 to $0.04\,\upmu\mathrm{g}\,\upmu^{-\,3}$ .
2.3. Receptor models
Three statistically driven receptor models, including PCA-MLR, UNMIX, as well as PMF, were adopted to identify the sources of $\mathrm{PM}_{2.5}$ at LA. The fundamental principle of receptor modeling is that mass conservation can be assumed and a mass balance analysis can be used to identify and apportion sources of atmospheric particulate matter (Hopke, 2003). The major advantage of these three receptor models used in this study is that sources and source profiles need not to be known or predetermined.
2.3.1. PCA-MLR
PCA-MLR was performed with the SPSS software. PCA-MLR is a traditional factor analysis receptor model. PCA will represent the total variability of the original $\mathrm{PM}_{2.5}$ data in a minimum number of factors, i.e., principal components (PCs) (Thurston and Spengler, 1985). Each factor is independent on all others, leading to the smallest possible covariance. The first factor represents the weighted linear combination of the original variables that accounts for the greatest variability, and each subsequent factor accounts for less variability than the previous. Data of $\mathrm{PM}_{2.5}$ and its chemical species were applied on PCA analysis using Varimax rotation, retaining PCs with eigenvalues $>\,1$ . Relationship between PC and the chemical compounds is indicated by the factor loadings and relates to the source composition. MLR analysis was then applied on the PCA factor scores to determine the contribution of each source. The detailed processes of PCA-MLR analysis is briefly listed below.
First, mass concentrations go through the standardized processing:
$$
Z_{i j}=\frac{C_{i j}-C_{j}}{\sigma_{j}}
$$
where $Z_{i j}$ the standardized value of the jth species in the ith sample, $C_{i j}$ is the concentration of the jth species in the ith sample, $C_{j}$ is the average concentration of the jth species, and $\sigma_{j}$ is the standardized deviation of jth species.
Then, the standardized data are used to obtain the major contributing factors:
$$
F_{k}=\sum_{j=1}^{p}{\mathrm{a}_{i j}\times Z_{i j}}
$$
where $F_{k}$ is the factor score of the kth source, $p$ is the number of sources, and $a_{i j}$ is the characteristic vector of the jth species of the ith sample.
MLR is then adopted to quantitatively estimate source contributions based on the PCA factor scores:
$$
Z_{\mathrm{PM}_{2.5}}=\sum_{k=1}^{p}\beta_{k}\times F_{k}+D
$$
where $\beta_{k}$ is the partial regression coefficient of kth source which can convert the factor scores to source contributions, and $D$ is the constant term.
2.3.2. UNMIX
UNMIX model was conducted with the US EPA UNMIX6.0 software also by importing dataset of $\mathrm{PM}_{2.5}$ and its major chemical species. UNMIX model is also a receptor model based on factor analysis (Hopke, 2003). UNMIX model has a strict data screening process based on an eigenvalue analysis. The missing data or data below minimum detection limits will be eliminated. Results from the UNMIX model are constrained to nonnegative values since geometrical concepts of selfmodeling curve resolution are used (Song et al., 2006). By using the singular value decomposition (SVD) method, the model estimates the source number by reducing the dimensionality of data space $m$ to $p$ (Henry, 2003; Song et al., 2006). The equation is listed below (Paatero et al., 2005):
$$
C_{i j}=\sum_{k=1}^{p}G_{j k}S_{i k}+E
$$
where, the $C_{i j}$ is the concentration of the jth species in the ith sample, $G_{j k}$ is the percentage of jth species in the kth source, $S_{i k}$ is the contribution of the kth source to ith sample, and $E$ is the standard deviation of analysis.
The data matrix will be normalized to ensure that all the species are on the same scale with a mean of 1, and then the data are projected to a plane perpendicular to the first axis of $p$ -dimensional space, which can reduce the normalized source compositions (Larsen and Baker, 2003; Song et al., 2006). Find edges, which represent the samples characterizing the source, in sets of points in a space of arbitrary dimension is the most important method in the UNMIX model (Henry, 2003).
2.3.3. PMF
PMF analysis was performed with the US EPA PMF5.0 model. A dataset with $n$ samples and $m$ chemical species will be treated as a matrix $X$ with dimensions $n\times m$ when it is imported into PMF model.
The PMF model works by decomposing the matrix into two matrices, i.e., factor contributions and factor profiles (Paatero et al., 2013). For the purpose of identifying the number of factors $p$ , the species profile $f_{k}$ of each factor $k$ , and the proportion of mass $g_{k}$ contributed by each factor $k$ to each sample, the PMF method is described as the following equation (Paatero and Tapper, 1994; Paatero, 1997):
$$
x_{i j}=\sum_{k=1}^{p}g_{i k}f_{k j}\,+\,e_{i j}=c_{i j}+\,e_{i j}
$$
where $x_{i j},\ c_{i j}$ , and $e_{i j}$ are the concentration, the modeled part and the residual part for each sample/species, respectively.
Factor elements are constrained, ensuring that no sample can have a negative factor contribution (Paatero, 1997). The PMF solution minimizes the object function $Q$ based on the estimated uncertainties $u_{i j}$ and with factor matrix elements $g_{i k}$ and $f_{k j}$ subject to non-negativity constraints (Paatero et al., 2013):
$$
Q=\sum_{j=1}^{n}\sum_{i=1}^{m}\left[\frac{x_{i j}-\sum_{k=1}^{p}g_{i k}f_{k j}}{u_{i j}}\right]^{2}
$$
The uncertainty was estimated based on the EPA PMF 5.0 User Guide (Norris et al., 2014). If the concentration is below or equal to the MDL, the uncertainty was set at $5/6\,\times\,\mathrm{MDL}$ . If the concentration is greater than the $\mathrm{MDL}$ , the uncertainty calculation was based on a fraction of the concentration and MDL:
$$
y={\sqrt{({\mathrm{ErrorFraction}}\times{\mathrm{concentration}})^{2}+(0.5\times M D L)^{2}}}
$$
Uncertainty of each chemical component was estimated as $0.286\,\upmu\mathrm{g}\,\mathsf{m}^{-\,3}$ $2.044\,\upmu\mathrm{g}\,\mathsf{m}^{-\,3}$ $1.737\,\upmu\mathrm{g}\,\mathsf{m}^{-\,3}$ $0.966\;\upmu\mathrm{g}\;\mathrm{m}^{-\,3},$ $0.609\,\upmu\mathrm{g}\,\mathfrak{m}^{-\,3}$ $0.090\,\upmu\mathrm{g}\,\mathsf{m}^{-\,3},$ $0.206\,\upmu\mathrm{g}\,\mathsf{m}^{-\,3}$ $0.082\,\upmu\mathrm{g}\,\mathsf{m}^{-\,3},$ $0.013\,\upmu\mathrm{g}\textnormal{m}^{-\,3}$ $0.003\,\upmu\mathrm{g}\textnormal{m}^{-\,3}$ $0.012\,\upmu\mathrm{g}\,\mathsf{m}^{-\,3}$ $0.032\;\upmu\mathrm{g}\;\mathbf{m}^{-\frac{3}{3}},$ $0.002\,\upmu\mathrm{g}\,\mathsf{m}^{-\,3}$ $0.007\,\upmu\mathrm{g}\textnormal{m}^{-\,3}$ $0.045\,\upmu\mathrm{g}\,\mathrm{m}^{-\,3}$ $0.005\,\upmu\mathrm{g}\,\mathsf{m}^{-\,3},$ $0.017\,\upmu\mathrm{g}\,\mathfrak{m}^{-\,3}$ and $0.002\,\upmu\mathrm{g}\textnormal{m}^{-\,3}$ for EC, OC, ${\mathsf{S O}}_{4}^{\ 2-}$ , $\mathrm{NO}_{3}^{\mathrm{~-~}}$ , $\mathrm{NH_{4}}^{+}$ , $\mathrm{Cl}^{-}$ , $\mathrm{Na}^{+}$ , $\mathrm{K}^{+}$ , ${\mathrm{Ca}}^{2\,+}$ , ${\mathrm{M}}\mathfrak{g}^{2\,+}$ , Al, Si, Ti, Mn, Fe, Cu, $Z\mathbf{n}$ and As, respectively.
2.3.4. Comparison of the three receptor models
The three receptor models have been widely used in many source apportionment studies. All those statistical based techniques are useful in quantifying source contributions and identifying the chemical profiles of sources without any prior knowledge of the number or types of sources. However, significant differences exist among the results of the models, mostly due to the different mechanisms on which the models based. For example, PCA may generate negative source contributions which will lead to an inaccurate result and fail to fully determine the source profiles (Henry, 1987; Larsen and Baker, 2003). Conversely, UNMIX and PMF are constrained to positive values to address the PCA's problem and will get positive results (Cesari et al., 2016a; Hopke, 2010; Larsen and Baker, 2003). As for PMF and UNMIX, the approach of weighting individual data points, which provides for treatments of data that are missing or below the method detection limit, is allowed in PMF but not in UNMIX (Poirot et al., 2001). This method allows inclusion of measurement uncertainties (Anttila et al., 1995; Poirot et al., 2001). Although major factors resolved by PMF and UNMIX are generally the same, UNMIX does not always resolve as many factors as PMF (Pekney et al., 2006; Poirot et al., 2001).
3. Results and discussion
3.1. Overview of $P M_{2.5}$
Mean mass concentrations of $\mathrm{PM}_{2.5}$ and its major chemical species at LA are presented in Fig. 2 Average $\mathrm{PM}_{2.5}$ concentration was $68.9\,\upmu\mathrm{g}\,\up m^{-\,3}$ , which is almost twice of the Chinese National Ambient
Air Quality Standard of $\mathrm{PM}_{2.5}$ $(35\,\upmu\mathrm{g}\,\mathrm{m}^{-\,3})$ . It suggests serious $\mathrm{PM}_{2.5}$ pollution at the background site and heavy regional air pollution in the YRD region. Mean concentration of EC was $2.7\,\upmu\mathrm{g}\,\mathrm{m}^{-\,3}$ , and it accounted for $4.0\%$ of $\mathrm{PM}_{2.5}$ . Mean concentration of OC was $10.9\,\upmu\mathrm{g}\textnormal{m}^{-\,3}$ . The secondary water-soluble inorganic ions, namely, SO42− $(16.9\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ , $\mathrm{NO}_{3}^{\mathrm{~-~}}$ $(9.2\,\upmu\mathrm{g}\,\mathrm{m}^{-\,3})$ and $\mathrm{NH_{4}}^{+}$ $(5.8\,\upmu\mathrm{g}\,\mathrm{m}^{-\,3})$ , were the predominant species of $\mathbf{PM}_{2.5}$ . The secondary ions totally comprised $45.2\%$ of $\mathbf{PM}_{2.5},$ indicating the intense influences of secondary sources at the background region. Elements are also important components in $\mathrm{PM}_{2.5}$ . Al, Si, Ti, Mn and Fe indicate dust emissions (Cao et al., 2008), As indicates coal combustion (Luo et al., 2004), Zn indicates fossil fuel combustion, industrial metallurgical processes, and waste incineration (Henry et al., 1984; Huggins and Huffman, 2004; Moffet et al., 2008), and Cu indicates vehicle emissions (Stechmann and Dannecker, 1990). The total concentration of eight elements accounted for $1.7\%$ of $\mathrm{PM}_{2.5}$ mass. Fe and Si were the dominant elements, accounting for $62.3\%$ of all elemental concentrations. Detailed results and more discussion of characteristics of $\mathrm{PM}_{2.5}$ and its major components at LA were presented in our previous study (Zhang et al., 2017).
3.2. Model performance
Performances of the three receptor models in this study were evaluated by the ability of the models to reproduce the measured $\mathrm{PM}_{2.5}$ concentrations. Fig. 3 presents the linear correlations between measured and reconstructed $\mathrm{PM}_{2.5}$ concentrations. Overall, the three models all achieved reasonable results regarding their ability to reproduce the measured $\mathrm{PM}_{2.5}$ mass. Comparing to PCA-MLR and UNMIX models, the PMF model provided an optimal reconstruction with the highest correlation coefficient $(\mathrm{R}^{2}=0.922)$ and lowest dispersion. The PMF case $(\mathrm{Y}=1.08\mathrm{X}\mathrm{~-~}3.65)$ with a slope close to 1 and a low intercept indicated a nice fitting of the model. In the case of PCA-MLR $(\mathrm{Y}\,=\,0.86{X}\,+\,9.56)$ , the correlation coefficient was high $(\mathrm{R}^{2}=0.863)$ , suggesting that the results of PCA-MLR can also be trusted. Although the dispersion of UNMIX was highest, the correlation coefficient $(\mathrm{R}^{2}=0.746)$ still showed a reliable fitting.
Correlation coefficients between measured and reconstructed concentrations of major chemical species are listed in Table 1. In the case of PMF, the reconstructed concentrations of the chemical species in $\mathrm{PM}_{2.5}$ had high correlation with the measured values, and the lowest $\mathrm{\mathbf{R}^{2}}$ was 0.549. Similarly, the PCA-MLR model had a nice fitting for most species. In addition, major chemical species had high correlations in the UNMIX model with $\mathrm{R}^{2}\,>\,0.5$ . Such results showed that the reconstructed $\mathrm{PM}_{2.5}$ mass over the receptor models were relatively good mass closures, which was essential for meaningful source identification and apportionment (Cohen et al., 2004).
Inferior correlation between measured and reconstructed concentrations of $\mathbf{PM}_{2.5}$ and its major chemical components were obtained by UNMIX than the other two models, likely due to the different requirements of data quality and volume. According to the good fitting results, the performances of the three receptor models were reliable and viable. However, PMF and PCA-MLR had better fitting effects comparing to UNMIX.
3.3. Source identification
The matrix of loadings of PCA analysis are presented in Table 2 (only loadings larger than 0.3 are presented) together with the indication of the variance explained by each component. The first five principal components totally accounted for $77.50\%$ of the total variance. The first principal component (PC1) with high loadings of OC, EC, Zn, As and $\mathsf{K}^{+}$ explained $38.18\%$ of the total variance. OC and EC are usually regarded as the tracers of combustion sources such as fossil fuel combustion and biomass burning (Cao et al., 2005; Viana et al., 2008; Zong et al., 2016). Vehicular traffic is generally a highly relevant source of EC and the majority of EC emissions in urban environments originate from diesel exhaust engines (Cyrys et al., 2003; Gray and Cass, 1998). As is the tracer of coal burning (Baek et al., 1997; Luo et al., 2004), and $Z\mathfrak{n}$ is emitted from fossil fuel combustion and industrial metallurgical processes (Han et al., 2017; Huggins and Huffman, 2004; Moffet et al., 2008). $\mathtt{K}^{+}$ can be treated as an excellent marker of biomass burning (Boreddy et al., 2017; Cheng et al., 2013; Chuang et al., 2013; Zhu et al., 2017). Therefore, PC1 could be interpreted as combustion source of fuel and biomass. PC2 explaining $12.81\%$ of the total variance had remarkably high loadings of the secondary inorganic ions including ${\mathrm{so}_{4}}^{2\mathrm{~-~}}$ , $\mathrm{NO}_{3}{}^{-}$ and $\mathrm{NH_{4}}^{+}$ . Therefore, PC2 was related with secondary aerosols. Certain identified sources including biomass burning, coal combustion and vehicle exhausts can contribute to secondary aerosols through the emission of their precursor gases such as $S0_{2}$ and $\mathrm{NO_{X}}$ (Zhang et al., 2013; Zhang et al., 2014). $\mathrm{Na}^{+}$ , $\mathtt{K}^{+}$ , ${\mathrm{M}}{\mathrm{g}}^{2\,+}$ and $\mathrm{Cl}^{-}$ were highly loaded in PC3, which accounted for $11.20\%$ of the total variance. Thus, PC3 could be explained as the marine source (Finlayson-Pitts and Hemminger, 2000). PC4 had high loadings of Mn, Fe and Cu. The industrial processes contribute to Mn pollution, traffic emissions contribute to Cu pollution, and Fe can be both metallurgicalderived and traffic-emitted (Flament et al., 2008; Lough et al., 2005; Yatkin and Bayram, 2008). Thus, PC4 can be related to the industrial and traffic sources, which accounted for $8.10\%$ of the total variance. High loadings of Al and Si were presented in PC5, which explained $7.22\%$ of the total variance. PC5 may be viewed as soil dust sources since Al and Si are mostly from soil dust (Almeida et al., 2008; Tullio et al., 2008).
As shown in Table 3, the UNMIX model identified four sources. The minimum signal-to-noise ratio was 2.82 and the smallest $\scriptstyle\mathbf{R}^{2}$ value was 0.81, suggesting that the four sources can account for $81.0\%$ of the total variance. The first source (S1) had extremely high loadings in OC, EC, $^{\mathrm{SO}_{4}^{\ 2-}}$ , and relatively high loadings in $\mathtt{K}^{+}$ , $\mathrm{Na}^{+}$ , ${\mathrm{M}}\mathfrak{g}^{2\,+}$ , ${\mathrm{Ca}}^{2\,+}$ , Fe and Cu. Firstly, OC, EC, $^{\mathrm{SO}_{4}}^{2\mathrm{~-~}}$ and $\mathsf{K}^{+}$ can be caused by the emission of the coal/biomass combustion (Cheng et al., 2013; Viana et al., 2008; Zong et al., 2016). Secondly, Fe and Cu usually caused by industrial source, while ${\mathrm{Ca}}^{2\,+}$ can be the tracer of construction. What's more, ${\mathrm{Ca}}^{2\,+}$ had a higher loading in S1 than other high loading species. Thus, S1 may be the mixture of coal/biomass combustion, industrial emissions and construction dust. The second source (S2) may relate to secondary source for its high loadings in secondary ions ${\mathrm{so}}_{4}^{\ 2-}$ , $\mathrm{NO}_{3}^{\mathrm{~-~}}$ and $\mathrm{NH_{4}}^{+}$ . Apart from secondary ions, high loadings of $\mathrm{Cl}^{-}$ , Zn and Mn as well as some loadings of OC and EC were also found in S2. EC, $Z\mathfrak{n}$ and Mn have been verified as the indicators of vehicle source (Cyrys et al., 2003; He et al., 2008; Tao et al., 2017). From this respect, S2 can be the mixture of secondary sources and traffic emissions. The third source (S3) suggested a marine source on account of the relatively high levels of $\mathrm{Na}^{+}$ and $\mathrm{Cl}^{-}$ . The fourth source (S4) was significantly characterized by the soil dust sources since its high loadings of Al, Si and Ti. It is also affected by coal combustion since its relatively high loading in As.
Elements Fe and As were removed in the PMF model because of their poor reconstructions (Table 1). As shown in Fig. 4, seven source factors were achieved by PMF. The first factor showed high loadings of $\mathrm{Cl}^{-}$ , OC and EC. EC and OC can be considered as tracers for coal combustion (Duan et al., 2006; Sun et al., 2004; Zheng et al., 2005). Previous studies confirmed that coal combustion processes could directly emit $\mathrm{Cl}^{-}$ and coal combustion was an important source of global emissions of $\mathrm{Cl}^{-}$ (McCulloch et al., 1999; Morawska and Zhang, 2002; Watson et al., 2001). Although $\mathrm{Cl}^{-}$ may also be emitted from marine sources, but no enrichment of $\mathrm{Na}^{+}$ was found in the first factor. Therefore, the first factor was designated as coal combustion. The second factor loading high concentrations of ${\mathrm{so}_{4}}^{2\mathrm{~-~}}$ , $\ K^{+}$ , OC, EC, ${\mathrm{Ca}}^{2\,+}$ and ${\mathrm{M}}\mathfrak{g}^{2\,+}$ was identified as the mixed source of biomass burning and construction dust. In addition to high loading of the $\mathtt{K}^{+}$ tracer, the emission OC was more than that of EC in the second factor, also indicating the contribution of biomass burning (Lewis et al., 2003). The third factor which was dominated by Ti, Cu, Zn, OC and EC was characterized by the source of traffic emission, because the metal brake wear in the traffic process can release Cu, and Zn has been linked to engine oils and brake linings (Lough et al., 2005; Heo et al., 2008). Combining with Cu and Zn, OC and EC may also be emitted from traffic process. In the fourth factor, secondary species such as $\mathrm{NH_{4}}^{+}$ , $\mathrm{NO}_{3}{}^{-}$ and ${\mathrm{so}}_{4}^{\ 2-}$ were highly loaded. Therefore, the fourth factor was viewed as the secondary aerosols. The fifth factor contributing to most of the Ti, Mn and $Z\mathbf{n}$ can be treated as the source of industrial emission. The sixth factor could be the marine source for the reason that it showed a high contribution to $\mathrm{Na}^{+}$ , and was best represented by $\mathrm{K}^{+}$ and ${\mathrm{M}}{{g}^{2}}^{\mathrm{~+~}}$ . The crustal elements Al and Si were mostly contributed by the seventh factor. Thus, this factor could be the source of dust.
3.4. Source contribution
Source contributions to $\mathrm{PM}_{2.5}$ obtained by the three different receptor models are presented in Fig. 5. Overall, source contribution calculated by each model had something in common but not all the same. The solutions of PCA-MLR and UNMIX explained about $80\%$ $(78.1\%$ for PCA-MLR and $81.0\%$ for UNMIX) of all the sources. While PMF almost resolved out $100\%$ of the sources contributing to $\mathrm{PM}_{2.5}$ . Such results suggested that more chemical components such as organic molecular compounds should be included in the receptor models in order to achieve a solution with higher efficiency.
In the MLR analysis of the PCA results, the largest source was the secondary source $(36.7\%)$ followed by combustion $(27.3\%)$ , marine source $\left(8.6\%\right)$ , mixed sources of industrial and traffic $(4.7\%)$ , and soil dust $\left(0.8\%\right)$ . In the UNMIX model, the largest contribution $(30.2\%)$ was from the mixed sources of combustion, industrial and construction dust. The secondary source accounting for $28.7\%$ of $\mathrm{PM}_{2.5}$ was with the second largest contribution. The mixed sources of soil dust and coal combustion contributed $13.9\%$ and marine source contributed to $8.2\%$ to $\mathrm{PM}_{2.5}$ , respectively. While according to the PMF results, like the PCAMLR model, the largest source was also the secondary source with the contribution of $24.8\%$ . The contribution from industrial source was significant in the PMF results, which accounted for about $20.0\%$ . The contributions of the traffic source $(16.6\%)$ and the mixed sources of biomass burning and construction dust $(15.9\%)$ were comparable. The contribution of coal combustion source was $8.6\%$ and that of marine source was $10.2\%$ , while the contribution of soil dust was only $3.9\%$ . The source apportionment results determined by the three receptor models had common results. The three receptor models all identified that combustion, secondary aerosols, traffic emission, industrial emission and marine source were the major sources contributing to $\mathrm{PM}_{2.5}$ at the LA background site. Common sources were achieved by the three models, indicating that the source apportionment results were convictive to a certain extent. PCA-MLR, UNMIX, and PMF models are preferable in identifying the $\mathbf{PM}_{2.5}$ sources under large data volume. However, all of them have some difficulty in distinguishing some sources with common tracers. To be specific in this study, the solutions have some differences among the three models. Five, four and seven sources were obtained by PCA-MLR, UNMIX, and PMF, respectively. UNMIX model can only get some mixed sources such as the mixed sources of combustion, industrial and construction dust due to its relatively low resolutions. PCA-MLR could not separate industrial source and traffic source. PMF model was sensitive to distinguish different sources of combustion including coal combustion, biomass burning and industrial emission, while PCA-MLR and UNMIX only can get the mixed sources of combustions.
Based on the source contribution results of the three receptor models, combustion and secondary aerosols were the two dominating sources of $\mathbf{PM}_{2.5}$ at the LA background site. These two sources dominated over $60\%$ of total $\mathbf{PM}_{2.5}$ . Industrial source, marine source and dust source were the remaining sources achieved by source apportionment. The PMF model had a higher estimation of combustion source comparing with PCA-MLA and UNMIX. Marine source contributed $8.6\%$ , $8.2\%$ and $10.2\%$ to $\mathrm{PM}_{2.5}$ under the estimations of the PCA-MLR, UNMIX and PMF models, respectively. The contributions of marine source to $\mathrm{PM}_{2.5}$ at the LA site obtained in the three models were comparable and similar. According to direct estimation based on concentrations of $\mathrm{Na}^{+}$ and other ions (Wang and Shooter, 2001), the contribution of marine source to $\mathrm{{PM}}_{2.5}$ was approximately $9.6\%$ , which was in agreement with the results obtained by the receptor models. Although the LA site is not close to the sea, marine source originated from East China Sea could have influence on the site by regional transport (Feng et al., 2015). Considering that the LA site is located in the regional background area with less influences of local human activities, high contributions of combustion source and secondary aerosols may result from the regional transport of air pollutants in the YRD region. The finding of large contribution of secondary aerosols to $\mathrm{PM}_{2.5}$ in the YRD region is in accordance with the previous studies (Chen et al., 2015, 2017). The fraction associated to secondary ions was larger than the contributions $(25{-}37\%)$ found with source apportionment for all models. There might be at least two reasons which can explain the discrepancies. One reason is that some secondary sources might be unidentified by the receptor models due to limitations of the models. There were still about $22\%$ and $19\%$ unidentified sources in the PCAMLR and UNMIX models, respectively. These unidentified sources might contain certain secondary sources. Another reason is that some secondary aerosols might be attributed to other sources rather than the secondary source. For example, some $\mathrm{NO}_{3}{}^{-}$ and ${\mathrm{so}}_{4}^{\ 2-}$ were attributed to the marine source by receptor model since aged marine aerosols were usually found to contain those two components (Laskin et al., 2002; Yao et al., 2003). Above discussion have been implemented in the revision.
The source apportionment results conducted by these three models all have its viability although some divergences existed in either the source identification or the source contribution. To a certain extent, the three results can support each other and find out more reasonable sources of the $\mathrm{PM}_{2.5}$ at LA. Firstly, both secondary source and combustion source were identified as the major source of $\mathrm{PM}_{2.5}$ at LA with the three models, which suggested that PCA-MLR, PMF, and UNMIX reached an agreement on the identification of the major sources. Secondly, in spite that more sources were identified by PMF, other two models still have their merits and demerits as different models can be chosen for different cases. For example, when the sample size is not large enough to support PMF and UNMIX, PCA-MLR is more applicable. PMF is more effective in distinguishing the sources that have alike markers. In order to get a more convincing result with UNMIX, more indicative species should be added for a better identification of the sources.
4. Conclusions
In this study, $\mathrm{PM}_{2.5}$ samples were collected at the LA regional background site in the YRD region in East China during 2014–2015 and source apportionment of $\mathrm{PM}_{2.5}$ were conducted with PCA-MLR, UNMIX and PMF receptor models. The results of source apportionment of the three models were also inter-compared by evaluating model performance and identifying source profiles and contributions. Average $\mathrm{PM}_{2.5}$ mass concentration during the sampling period was $68.9~\pm~28.3\,\upmu\mathrm{g}\,\mathfrak{m}^{-3}$ . PMF resolved almost all of the sources, while the PCA-MLR and UNMIX models explained about $80\%$ . Combustion source, secondary source, marine source, dust source and industrial source were achieved. Combustion source and secondary source contributed over $60\%$ to $\mathbf{PM}_{2.5},$ indicating the impact of regional transport on particle pollution over the YRD region. Each receptor model has its specific advantage when solving the source of $\mathrm{PM}_{2.5}$ . The PMF model tended to have a better performance on distinguishing the different kinds of combustion. In future work, more chemical components may be imported into the receptor models to reduce the uncertainty of the model results.
Acknowledgments
This study was supported by National Key R&D Program of China (2016YFC0200500), National Natural Science Foundation of China (21607148, U1405235, 41575146 & 21507127), Key Research Program of the Chinese Academy of Sciences (KJZD-EW-TZ-G06-02), Fujian Natural Science Foundation (2017J01082), Youth Innovation Promotion Association CAS (2016279), and the Chinese Academy of Sciences Interdisciplinary Innovation Team Project.
References
Almeida, S.M., Freitas, M.C., Pio, C.A., 2008. Neutron activation analysis for identification of African mineral dust transport. J. Radioanal. Nucl. Chem. 276, 161–165.
Anttila, P., Paatero, P., Tapper, U., Jarvinen, O., 1995. Source identification of bulk wet deposition in Finland by positive matrix factorization. Atmos. Environ. 29, 1705–17018.
Baek, S.O., Choi, J.S., Hwang, S.M., 1997. A quantitative estimation of source contributions to the concentrations of atmospheric suspended particulate matter in urban, suburban, and industrial areas of Korea. Environ. Int. 23, 205–213.
Banerjee, T., Murari, V., Kumar, M., Raju, M.P., 2015. Source apportionment of airborne particulates through receptor modelling: Indian scenario. Atmos. Res. 164–165, 167–168.
Belis, C.A., Karagulian, F., Larsen, B.R., Hopke, P.K., 2013. Critical review and metaanalysis of ambient particulate matter source apportionment using receptor models in Europe. Atmos. Environ. 69, 94–108.
Belis, C.A., Karagulian, F., Amato, F., Almeida, M., Artaxo, P., Beddows, D.C.S., Bernardoni, V., Bove, M.C., Carbone, S., Cesari, D., Contini, D., Cuccia, E., Diapouli, E., Eleftheriadis, K., Favez, O., El Haddad, I., Harrison, R.M., Hellebust, S., Hovoka, J., Jang, E., Jorquera, H., Kammermeier, T., Karl, M., Lucarelli, F., Mooibroek, D., Nava, S., Nojgaard, J.K., Paatero, P., Pandolfi, M., Perrone, M.G., Petit, J.E., Pietrodangelo, A., Valli, G., Vecchi, R., Vestenius, M., Yubero, E., Hopke, P.K., 2015. A new methodology to assess the performance and uncertainty of source apportionment models II: the results of two European intercomparison exercises. Atmos. Environ. 123, 240–250.
Birch, M.E., Cary, R.A., 1996. Elemental carbon-based method for monitoring occupational exposures to particulate diesel exhaust. Aerosol Sci. Technol. 25, 221–241.
Boreddy, S.K.R., Kawamura, K., Okuzawa, K., Kanaya, Y., Wang, Z., 2017. Temporal and diurnal variations of carbonaceous aerosols and major ions in biomass burning influenced aerosols over Mt. Tai in the North China Plain during MTX2006. Atmos. Environ. 154, 106–117.
Brunekreef, B., Holgate, S.T., 2002. Air pollution and health. Lancet 360, 1233–1242.
Cao, J.J., Wu, F., Chow, J.C., Lee, S.C., Li, Y., Chen, S.W., An, Z.S., Fung, K.K., Watson, J.G., Zhu, C.S., Liu, S.X., 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, J.J., Chow, J.C., Watson, J.G., Wu, F., Han, Y.M., Jin, Z.D., Shen, Z.X., An, Z.S., 2008. Size-differentiated source profiles for fugitive dust in the Chinese Loess Plateau. Atmos. Environ. 42, 2261–2275.
Cao, J.J., Shen, Z.X., Chow, J.C., Watson, J.G., Lee, S.C., Tie, X.X., Ho, K.F., Wang, G.H., Han, Y.M., 2012. Winter and summer $\mathrm{PM}_{2.5}$ chemical compositions in fourteen Chinese cities. J. Air Waste Manage. Assoc. 62, 1214–1226.
Cesari, D., Amato, F., Pandolfi, M., Alastuey, A., Querol, X., Contini, D., 2016a. An intercomparison of $\mathrm{PM_{10}}$ source apportionment using PCA and PMF receptor models in three European sites. Environ. Sci. Pollut. Res. 23, 15133–15148.
Cesari, D., Donateo, A., Conte, M., Contini, D., 2016b. Inter-comparison of source apportionment of $\mathrm{PM_{10}}$ using PMF and CMB in three sites nearby an industrial area in central Italy. Atmos. Res. 182, 282–293.
Chan, C.K., Yao, X., 2008. Air pollution in mega cities in China. Atmos. Environ. 42, 1–42.
Chen, J., Qiu, S., Shang, J., Wilfrid, O.M.F., Liu, X., Tian, H., Boman, J., 2014. Impact of relative humidity and water soluble constituents of $\mathrm{PM}_{2.5}$ on visibility impairment in Beijing, China. Aerosol Air Qual. Res. 14, 260–268.
Chen, P., Wang, T., Hu, X., Xie, M., 2015. Chemical mass balance source apportionment of size-fractionated particulate matter in Nanjing, China. Aerosol Air Qual. Res. 15 (5), 1855–1867.
Chen, P., Wang, T., Lu, X., Yu, Y., Kasoar, M., Xie, M., Zhuang, B., 2017. Source apportionment of size-fractionated particles during the 2013 Asian Youth Games and the 2014 Youth Olympic Games in Nanjing, China. Sci. Total Environ. 579, 860–870.
Cheng, Y., Engling, G., He, K.B., Duan, F.K., Ma, Y.L., Du, Z.Y., Liu, J.M., Zheng, M., Weber, R.J., 2013. Biomass burning contribution to Beijing aerosol. Atmos. Chem. Phys. 13, 7765–7781.
Chuang, M.T., Chou, C.C.K., Sopajaree, K., Lin, N.H., Wang, J.L., Sheu, G.R., Chang, Y.J., Lee, C.T., 2013. Characterization of aerosol chemical properties from near-source biomass burning in the northern Indochina during 7-SEAS/Dongsha experiment. Atmos. Environ. 78, 72–81.
Cohen, D.D., Garton, D., Stelcer, E., Hawa, O., 2004. Accelerator based studies of atmospheric pollution process. Radiat. Phys. Chem. 71, 759–767.
Contini, D., Genga, A., Cesari, D., Siciliano, M., Donateo, A., Bove, M.C., Guascito, M.R., 2010. Characterisation and source apportionment of $\mathrm{PM_{10}}$ in an urban background site in Lecce. Atmos. Res. 95, 40–54.
Cyrys, J., Heinrich, J., Hoek, G., Meliefste, K., Lewne, M., Gehring, U., Bellander, T., Fischer, P., Van Vliet, P., Brauer, M., Wichmann, H.E., Brunekreef, B., 2003. Comparison between different traffic-related particle indicators: elemental carbon (EC), $\mathrm{PM}_{2.5}$ mass, and absorbance. J. Expo. Anal. Environ. Epidemiol. 13, 134–143.
Dai, W., Gao, J., Cao, G., Ouyang, F., 2013. Chemical composition and source identification of $\mathrm{PM}_{2.5}$ in the suburb of Shenzhen, China. Atmos. Res. 122, 391–400.
Deng, J., Zhang, Y., Hong, Y., Xu, L., Chen, Y., Du, W., Chen, J., 2016. Optical properties of and the impacts of chemical compositions in the coastal city Xiamen in China. Sci. Total Environ. 557–558, 665–675.
Ding, A.J., Fu, C.B., Yang, X.Q., Sun, J.N., Zheng, L.F., Xie, Y.N., Herrmann, E., Nie, W., Petäjä, T., Kerminen, V.-M., Kulmala, M., 2013. Ozone and fine particle in the western Yangtze River Delta: an overview of 1 yr data at the SORPES station. Atmos. Chem. Phys. 13, 5813–5830.
Duan, F.K., He, K.B., Ma, Y.L., Yang, F.M., Yu, X.C., Cadle, S.H., Chan, T., Mulawa, P.A., 2006. Concentration and chemical characteristics of $\mathrm{PM}_{2.5}$ in Beijing, China: 2001–2002. Sci. Total Environ. 355, 264.
Feng, J., Hu, J., Xu, B., Hu, X., Sun, P., Han, W., Gu, Z., Yu, X., Wu, M., 2015. Characteristics and seasonal variation of organic matter in $\mathrm{PM}_{2.5}$ at a regional background site of the Yangtze River Delta region, China. Atmos. Environ. 123, 288–297.
Finlayson-Pitts, B.J., Hemminger, J.C., 2000. Physical chemistry of airborne sea salt particles and their components. J. Phys. Chem. A 104 (49), 11463–11477.
Flament, P., Mattielli, N., Aimoz, L., Choel, M., Deboudt, K., Jong, J., Rimetz-Planchon, J., Weis, D., 2008. Iron isotopic fractionation in industrial emissions and urban aerosols. Atmos. Environ. 73 (11), 1793–1798.
Gray, H.A., Cass, G.R., 1998. Source contributions to atmospheric fine carbon particle concentrations. Atmos. Environ. 32 (22), 3805–3825.
Han, F., Kota, S.H., Wang, Y., Zhang, H., 2017. Source apportionment of $\mathrm{PM}_{2.5}$ in Baton Rouge, Louisiana during 2009–2014. Sci. Total Environ. 586, 115–126.
He, K.B., Huo, H., Zhang, Q., 2002. Urban air pollution in China: current status, characterizes and progress. Annu. Rev. Energy Environ. 27 (1), 397–431.
He, L.Y., Hu, M., Zhang, Y.H., Huang, X.F., Yao, T.T., 2008. Fine particle emissions from on-road vehicles in the Zhujiang Tunnel, China. Environ. Sci. Technol. 42, 4461–4466.
Henry, R.C., 1987. Current factor analysis models are ill-posed. Atmos. Environ. 21, 1815–1820.
Henry, R.C., 2003. Multivariate receptor modeling by N-dimensional edge detection. Chemom. Intell. Lab. Syst. 65, 179–189.
Henry, R.C., Lewis, C.W., Hopke, P.K., Williamson, H.J., 1984. Review of receptor model fundamentals. Atmos. Environ. 18, 1507–1515.
Heo, J.B., Hopke, P.K., Yi, S.M., 2008. Source apportionment of $\mathrm{PM}_{2.5}$ in Seoul, Korea. Atmos. Chem. Phys. 9, 4957–4971.
Hopke, P.K., 2003. Recent developments in receptor modeling. J. Chemom. 17, 255–265.
Hopke, P.K., 2010. The application of receptor modeling to air quality data. Pollut. Atmos. 91–109.
Hopke, P.K., Ito, K., Mar, T., Christensen, W.F., Eatough, D.J., Henry, R.C., Kim, E., Laden, F., Lall, R., Larson, T.V., Liu, H., Neas, L., Pinto, J., Stölzel, M., Suh, H., Paatero, P., Thurston, G.D., 2006. PM source apportionment and health effects: 1. Intercomparison of source apportionment results. J. Expo. Sci. Environ. Epidemiol. 16, 275–286.
Huggins, F.E., Huffman, G.P., 2004. Quantifying hazardous species in particulate matter derived from fossil-fuel combustion. Environ. Sci. Technol. 38 (6), 1836–1842.
Huo, H., Zhang, Q., Guan, D., Su, X., Zhao, H., He, K., 2014. Examining air pollution in China using production- and consumption-based emissions accounting approaches. Environ. Sci. Technol. 48 (24), 14139–14147.
Larsen, R.K., Baker, J.E., 2003. Source apportionment of polycyclic aromatic hydrocarbons in the urban atmosphere: a comparison of three methods. Environ. Sci. Technol. 37, 1873–1881.
Laskin, A., Iedema, J.M., Cowin, P.J., 2002. Quantitative time-resolved monitoring of nitrate formation in sea salt particles using a CCSEM/EDX single particle analysis. Environ. Sci. Technol. 36 (23), 4948–4955.
Lewis, C.W., Norris, G.A., Conner, T.L., Henry, R.C., 2003. Source apportionment of Phoenix $\mathrm{PM}_{2.5}$ aerosol with the Unmix receptor model. J. Air Waste Manage. Assoc. 53, 325–338.
Li, L., Chen, C.H., Fu, J.S., Huang, C., Streets, D.G., Huang, H.Y., Zhang, G.F., Wang, Y.J., Jang, C.J., Wang, H.L., Chen, Y.R., Fu, J.M., 2011. Air quality and emissions in the Yangtze River Delta, China. Atmos. Chem. Phys. 11, 1621–1639.
Liu, D., Li, J., Zhang, Y., Xu, Y., Liu, X., Ding, P., Shen, $\mathrm{C}_{\ast}$ Chen, Y., Tian, C., Zhang, G., 2013. The use of Levoglucosan and radiocarbon for source apportionment of $\mathrm{PM}_{2.5}$ carbonaceous aerosols at a background site in East China. Environ. Sci. Technol. 47 (18), 10454–10461.
Liu, B., Wu, J., Zhang, J., Wang, L., Yang, J., Liang, D., Dai, Q., Bi, X., Feng, Y., Zhang, Y., Zhang, Q., 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, 10–22.
Lough, G.C., Schauer, J.J., Park, J.S., Shafer, M.M., Deminter, J.T., Weinstein, J.P., 2005. Emissions of metals associated with motor vehicle roadways. Environ. Sci. Technol. 39 (3), 826–836.
Luo, K.L., Zhang, X.M., Chen, C.H., Lu, Y.L., 2004. Estimate of arsenic emission amount from the coal power stations in China. Chin. Sci. Bull. 49, 2183–2189.
Masiol, M., Hopke, P.K., Felton, H.D., Frank, B.P., Rattigan, O.V., Wurth, M.J., LaDuke, G.H., 2017. Source apportionment of $\mathrm{PM}_{2.5}$ chemically speciated mass and particle number concentrations in New York City. Atmos. Environ. 148, 215–229.
McCulloch, A., Aucott, M.L., Benkovitz, C.M., Graedel, T.E., Kleiman, G., Midgley, P.M., Li, Y.F., 1999. Global emissions of hydrogen chloride and chloromethane from coal combustion, incineration and industrial activities: Reactive Chlorine Emissions Inventory. J. Geophys. Res. 104 (D7), 8391–8403.
Moffet, R.C., Desyaterik, Y., Hopkins, R.J., Tivanski, A.V., Gilles, M.K., Wang, Y., Shutthanandan, V., Molina, L.T., Abraham, R.G., Johnson, K.S., Mugica, V., Molina, M.J., Laskin, A., Prather, K.A., 2008. Characterization of aerosols containing Zn, Pb, and Cl from an industrial region of Mexico City. Environ. Sci. Technol. 42 (19), 7091–7097.
Morawska, L., Zhang, J., 2002. Combustion sources of particles. 1. Health relevance and source signatures. Atmos. Environ. 1045–1058.
Niu, X., Cao, J., Shen, Z., Ho, S.S.H., Tie, X., Zhao, S., Xu, H., Zhang, T., Huang, R., 2016. $\mathrm{PM}_{2.5}$ from the Guanzhong Plain: chemical composition and implications for emission reductions. Atmos. Environ. 147, 458–469.
Norris, G.N., Duval, R., Brown, S., Bai, S., 2014. EPA Positive Matrix Factorization (PMF) 5.0 Fundamentals and User Guide. USEPA Office of Research and Development, Washington, DC 20460.
Ogundele, L.T., Owoade, O.K., Olise, F.S., Hople, P.K., 2016. Source identification and apportionment of $\mathrm{PM}_{2.5}$ and $\mathrm{PM}_{2.5-10}$ in iron and steel scrap smelting factory environment using PMF, PCFA and UNMIX receptor models. Environ. Monit. Assess. 188, 574.
Paatero, P., 1997. Least squares formulation of robust non-negative factor analysis. Chemom. Intell. Lab. Syst. 37 (1), 23–35.
Paatero, P., Tapper, U., 1994. Positive matrix factorization: a non-negative factor model with optimal utilization of error estimates of data values. Environmetrics 5 (2), 111–126.
Paatero, P., Hopke, P.K., Begum, B.A., Biswas, S.K., 2005. A graphical diagnostic method for assessing the rotation in factor analytical models of atmospheric pollution. Atmos. Environ. 39, 193–201.
Paatero, P., Eberly, S., Brown, S.G., Norris, G.A., 2013. Methods for estimating uncertainty in factor analytic solutions. Atmos. Meas. Tech. 6, 7593–7631.
Pekney, N.J., Davidson, C.I., Robinson, A., Zhou, L., Hopke, P., Eatough, D., Rogge, W.F., 2006. Major source categories for $\mathrm{PM}_{2.5}$ in Pittsburgh using PMF and UNMIX. Aerosol Sci. Technol. 40, 910–924.
Poeschl, U., 2005. Atmospheric aerosols: composition, transformation, climate and health effects. Angew. Chem. Int. Ed. 44, 7520–7540.
Poirot, R.L., Wishinski, P.R., Hopke, P.K., Polissar, A.V., 2001. Comparative application of multiple receptor methods to identify aerosol sources in northern Vermont. Environ. Sci. Technol. 35, 4622–4636.
Shen, G.F., Yuan, S.Y., Xie, Y.N., Xia, S.J., Li, L., Yao, Y.K., Qiao, Y.Z., Zhang, J., Zhao, Q.Y., Ding, A.J., Wu, H.S., 2014. Ambient levels and temporal variations of $\mathrm{PM}_{2.5}$ and $\mathrm{PM_{10}}$ at a residential site in the mega-city, Nanjing, in the western Yangtze River Delta, China. J. Environ. Sci. Health, Part A: Tox. Hazard. Subst. Environ. Eng. 49, 171–178.
Shi, G.L., Liu, G.R., Peng, X., Wang, Y.N., Tian, Y.Z., Wang, W., Feng, Y., 2014. A comparison of multiple combined models for source apportionment, including the PCA/ MLR-CMB, Unmix-CMB and PMF-CMB models. Aerosol Air Qual. Res. 14, 2040–2050.
Song, Y., Xie, S., Zhang, Y., Zeng, L., 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. Sci. Total Environ. 372, 278–286.
Stechmann, H., Dannecker, W., 1990. Characterization and source analysis of vehiclegenerated aerosols. J. Aerosol Sci. 21 (S1), 287–290.
Sun, Y., Zhuang, G., Wang, Y., Han, L., Guo, J., Dan, M., Zhang, W., Wang, Z., Hao, Z., 2004. The air-borne particulate pollution in Beijing—concentration, composition, distribution and sources. Atmos. Environ. 38, 5991–6004.
Tao, J., Zhang, L.M., Zhang, R.J., Wu, Y.F., Zhang, Z.S., Zhang, X.L., Tang, Y.X., Cao, J.J., Zhang, Y.H., 2016. Uncertainty assessment of source attribution of $\mathrm{PM}_{2.5}$ and its water-soluble organic carbon content using different biomass burning tracers in positive matrix factorization analysis—a case study in Beijing, China. Sci. Total Environ. 543, 326–335.
Tao, J., Zhang, L.M., Cao, J.J., Zhong, L.J., Chen, D.S., Yang, Y.H., Chen, D.H., Chen, L.G., Zhang, Z.S., Wu, Y.F., 2017. Source apportionment of $\mathrm{PM}_{2.5}$ at urban and suburban areas of the Pearl River Delta region, south China - with emphasis on ship emissions. Sci. Total Environ. 574, 1559–1570.
Thurston, G.D., Spengler, J.D., 1985. A quantitative assessment of source contributions to inhalable particulate matter pollution in metropolitan Boston. Atmos. Environ. 19 (1), 9–25.
Tie, X.X., Brasseur, G.P., Zhao, C., Granier, C., Massie, S., Qin, Y., Wang, P., Wang, G., Yang, P., Richter, A., 2006. Chemical characterization of air pollution in eastern China and the eastern United States. Atmos. Environ. 40, 2607–2625.
Tullio, A.D., Reale, S., Ciammola, M., Arrizza, L., Picozzi, P., De Angelis, F., 2008. Characterization of atmospheric particulate: relationship between chemical composition, size and emission source. ChemSusChem 1, 110–117.
Viana, M., Kuhlbusch, T.A.J., Querol, X., Alastuey, A., Harrison, R.M., Hopke, P.K., Winiwarter, W., Vallius, M., Szidat, S., Prévôt, A.S.H., 2008. Source apportionment of particulate matter in Europe: a review of methods and results. J. Aerosol Sci. 39, 827–849.
Wang, H., Shooter, D., 2001. Water soluble ions of atmospheric aerosols in three New Zealand cities: seasonal changes and sources. Atmos. Environ. 35, 6031–6040.
Wang, T., Wong, C.H., Cheung, T.F., Blake, D.R., Arimoto, R., Baumann, K., Tang, J., Ding, G.A., Yu, X.M., Li, S.Y., Streets, D.G., Simpson, I.J., 2004. Relationships of trace gases and aerosols and the emission characteristics at Lin'an, a rural site in eastern China, during spring 2001. J. Geophys. Res. 109 (D19). http://dx.doi.org/10.1029/ 2003JD004119.
Wang, Z., Wu, T., Shi, G., Fu, X., Tian, Y., Feng, Y., Wu, X., Wu, G., Bai, Z., Zhang, W., 2012. Potential source analysis for $\mathrm{PM_{10}}$ and $\mathrm{PM}_{2.5}$ in autumn in a northern city in China. Aerosol Air Qual. Res. 12, 39–48.
Wang, Y., Yao, L., Wang, L., Liu, Z., Ji, D., Tang, G., Zhang, J., Sun, Y., Hu, B., Xin, J., 2014. Mechanism for the formation of the January 2013 heavy haze pollution episode over central and eastern China. Sci. China Earth Sci. 57, 14–25.
Wang, Y.Q., Zhang, X.Y., Sun, J.Y., Zhang, X.C., Che, H.Z., Li, Y., 2015. Spatial and temporal variations of the concentrations of $\mathrm{PM}_{10},$ $\mathrm{PM}_{2.5}$ and $\mathrm{{PM_{1}}}$ in China. Atmos. Chem. Phys. 15, 13585–13598.
Wang, H.L., An, J.L., Cheng, M.T., Shen, L.J., Zhu, B., Li, Y., Wang, Y.S., Duan, Q., Sullivan, A., Xia, L., 2016. One year online measurements of water-soluble ions at the industrially polluted town of Nanjing, China: sources, seasonal and diurnal variations. Chemosphere 148, 526–536.
Watson, J.G., Chow, J.C., 2005. Receptor models. In: Zanetti, P. (Ed.), Air Quality Modeling – Theories, Methodologies, Computational Techniques, and Available Databases and Software. Vol. II – Advanced Topics. Air and Waste Management Association and the EnviroComp Institute, Pittsburgh, PA, pp. 455–501.
Watson, J.G., Chow, J.C., Houck, J.E., 2001. $\mathrm{PM}_{2.5}$ chemical source profiles for vehicle exhaust, vegetative burning, geological material, and coal burning in northwestern Colorado during 1995. Chemosphere 43 (8), 1141–1151.
Yao, X., Fang, M., Chan, C.K., 2003. The size dependence of chloride depletion in fine and coarse sea-salt particles. Atmos. Environ. 37 (6), 743–751.
Yatkin, S., Bayram, A., 2008. Source apportionment of $\mathrm{PM_{10}}$ and $\mathrm{PM}_{2.5}$ using positive matrix factorization and chemical mass balance in Izmir, Turkey. Sci. Total Environ. 390, 109–123.
Zhang, R., Jing, J., Tao, J., Hsu, S.C., Wang, G., Cao, J., Lee, S.C.L., Zhu, L., Chen, Z., Zhao, Y., Shen, Z., 2013. Chemical characterization and source apportionment of $\mathrm{PM}_{2.5}$ in Beijing: seasonal perspective. Atmos. Chem. Phys. 13, 7053–7074.
Zhang, H., Hu, J., Kleeman, M., Ying, Q., 2014. Source apportionment of sulfate and nitrate particulate matter in the eastern United States and effectiveness of emission control programs. Sci. Total Environ. 490, 171–181.
Zhang, Y., Zhang, H., Deng, J., Du, W., Hong, Y., Xu, L., Qiu, Y., Hong, Z., Wu, X., Ma, Q., Yao, J., Chen, J., 2017. Source regions and transport pathways of $\mathrm{PM}_{2.5}$ at a regional background site in East China. Atmos. Environ. 167, 202–211.
Zheng, M., Salmon, L.G., Schauer, J.J., Zeng, L., Kiang, C.S., Zhang, Y., Cass, G.R., 2005. Seasonal trends in $\mathrm{PM}_{2.5}$ source contributions in Beijing, China. Atmos. Environ. 39, 3967–3976.
Zhu, C.S., Cao, J.J., Tsai, C.J., Zhang, Z.S., Tao, J., 2017. Biomass burning tracers in rural and urban ultrafine particles in Xi'an, China. Atmos. Pollut. Res. 8 (4), 614–618.
Zong, Z., Wang, X., Tian, C., Chen, Y., Qu, L., Ji, L., Zhi, G., Li, J., Zhang, G., 2016. Source apportionment of $\mathrm{PM}_{2.5}$ at a regional background site in North China using PMF linked with radiocarbon analysis: insight into the contribution of biomass burning. Atmos. Chem. Phys. 16, 11249–11265.
|
atmosphere
|
0012
|
10.1016/j.atmosenv.2010.11.057
|
image
|
Fig. 1. Sampling site in PRD. Green pentacle is regional background site, Wangqingsha (WQS) which is in the central of PRD and surrounded by city clusters.
|
|
atmosphere
|
0012
|
10.1016/j.atmosenv.2010.11.057
|
image
|
Fig. 2. Total ion chromatogram for silylated sample. 1. succinic acid; 2. 2-methylglyceric acid; 3. 3-methyl-2,3,4-trihydroxy-1-butene; 4. glutaric acid; 5. adipic acid; 6. 2-methylthreitol; 7. 2-methylerythritol; 8. 3-hydroxyglutaric acid; 9. pimelic acid; 10. 3-hydroxy- $^{4,4}$ -dimethylglutaric acid; 11. phthalic acid- $\cdot\mathrm{D}_{4}$ (IS); 12. phthalic acid; 13. levoglucosan; 14. levoglucosan- $^{13}C_{6}$ (IS); 15. terephthalic acid; 16. azelaic acid; 17. $\upbeta$ -caryophyllinic acid; 18. hexadecanoic acid- $\cdot\mathrm{D}_{31}$ (IS); 19. hexadecanoic acid; 20. octadecanoic acid. Besides SOA tracers, other polar compounds were also quantified but not involved in the current paper.
|
|
atmosphere
|
0012
|
10.1016/j.atmosenv.2010.11.057
|
image
|
Fig. 3. Total ion chromatogram for methylated sample.1. cis-pinonic acid; 2. pinic acid; 3. 3-methyl-1,2,3-butanetricarboxylic acid; 4. phthalic acid- $\cdot\mathrm{D}_{4}$ (IS); 5. hexadecanoic acid- $\cdot\mathrm{D}_{31}$ (IS).
|
|
atmosphere
|
0012
|
10.1016/j.atmosenv.2010.11.057
|
table
|
Table 1 Summary of SOA tracers $\left(\mathfrak{n g m}^{-3}\right)$ and other parameters.
|
|
atmosphere
|
0012
|
10.1016/j.atmosenv.2010.11.057
|
image
|
Fig. 4. Daily variations of major components (a) and SOA tracers (b).
|
|
atmosphere
|
0012
|
10.1016/j.atmosenv.2010.11.057
|
image
|
Fig. 5. Daily variation of isoprene (a) and $^\circ{}$ -pinene (b) SOA tracers.
|
|
atmosphere
|
0012
|
10.1016/j.atmosenv.2010.11.057
|
image
|
Fig. 6. Five-day back trajectories of air mass $100\;\mathrm{m}$ above ground on day Oct. 27 (green), Nov. 1 (blue) and Nov. 19 (red) using HYSPLIT 4.9 software from the Air Resources Laboratory (ARL) in NOAA (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.).
|
|
atmosphere
|
0012
|
10.1016/j.atmosenv.2010.11.057
|
image
|
Fig. 7. Correlations between isoprene SOA tracers and temperature. $k$ is the slope of correlation.
|
|
atmosphere
|
0012
|
10.1016/j.atmosenv.2010.11.057
|
image
|
Fig. 8. Correlations of aerosol acidity with isoprene (a) and $^\circ{}$ -pinene (b) SOA tracers. k is the slope of correlation.
|
|
atmosphere
|
0012
|
10.1016/j.atmosenv.2010.11.057
|
table
|
Table 2 Pearson coefficients for SOA tracers.a
|
|
atmosphere
|
0012
|
10.1016/j.atmosenv.2010.11.057
|
image
|
Fig. 9. Five-day back trajectory of air mass on the day when $\mathrm{P/M}$ ratio was the highest (Nov. 06) and the lowest (Oct. 28), respectively.
|
|
atmosphere
|
0012
|
10.1016/j.atmosenv.2010.11.057
|
text
| Not supported with pagination yet |
The influence of temperature and aerosol acidity on biogenic secondary organic aerosol tracers: Observations at a rural site in the central Pearl River Delta region, South China
Xiang Ding a,b, Xin-Ming Wang a,\*, Mei Zheng
a State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, 511 Kehua Rd, Tianhe, Guangzhou 510640, China b Pearl River Delta Research Center for Environmental Pollution Control, Chinese Academy of Sciences, Guangzhou 510640, China c School of Earth and Atmospheric Sciences, Georgia Institute of Technology, Atlanta 30332, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 30 August 2010 Received in revised form 20 November 2010 Accepted 30 November 2010
Keywords:
SOA
Tracers
Temperature
Aerosol acidity
Pearl River Delta (PRD)
At a rural site in the central Pearl River Delta (PRD) region in south China, fine particle $\left(\mathsf{P M}_{2.5}\right)$ samples were collected during fall-winter 2007 to measure biogenic secondary organic aerosol (SOA) tracers, including isoprene SOA tracers (3-methyl-2,3,4-trihydroxy-1-butene, 2-methylglyceric acid, 2-methylthreitol and 2-methylerythritol), $\pmb{\mathrm{\Omega}}$ -pinene SOA tracers (cis-pinonic acid, pinic acid, 3-methyl-1,2,3-butanetricarboxylic acid, 3-hydroxyglutaric acid and 3-hydroxy-4,4-dimethylglutaric acid) and a sesquiterpene SOA tracer $\Bumpeq$ caryophyllinic acid). The isoprene-, $\mathfrak{x}$ -pinene- and sesquiterpene-SOA tracers averaged $30.8\pm15.9$ $6.61\pm4.39$ , and $0.54\pm0.56\mathrm{\,ng\,m}^{-3}$ , respectively; and 2-methyltetrols (sum of 2-methylthreitol and 2- methylerythritol, $27.6\pm15.1\mathrm{~ng~m}^{-3}$ ) and cis-pinonic acid $(3.60\pm3.76\;\mathrm{ng\,m}^{-3})$ ) were the dominant isoprene- and $^\circ$ -pinene-SOA tracers, respectively. 2-Methyltetrols exhibited significantly positive correlations $(p<0.05)$ with ambient temperature, probably resulting from the enhanced isoprene emission strength and tracer formation rate under higher temperature. The significantly positive correlation $(p<0.05)$ between 2-methyltetrols and the estimated aerosol acidity with a slope of $59.4\pm13.4\;\mathrm{ng}\,\mathrm{m}^{-3}$ per mmo $1\,[\mathsf{H}^{+}]\,\mathsf{m}^{-3}$ reflected the enhancement of isoprene SOA formation by aerosol acidity, and acid-catalyzed heterogeneous reaction was probably the major formation pathway for 2-methyltetrols in the PRD region. 2-Methylglyceric acid showed poor correlations with both temperature and aerosol acidity. The $\mathfrak{a}$ -pinene SOA tracers showed poor correlations with temperature, probably due to the counteraction between temperature effects on the precursor emission/tracer formation and gas/particle partitioning. Among the $\mathfrak{x}$ -pinene SOA tracers, only cis-pinonic acid and pinic acid exhibited significant correlations with aerosol acidity with slopes of $-11.7\pm3.7$ and $-2.2\pm0.8\;\mathrm{ng}\,\mathrm{m}^{-3}$ per $\sharp\mathrm{mol}$ $[\mathsf{H}^{+}]\;\mathsf{m}^{-3}$ , respectively. The negative correlations observed for $\pmb{\mathrm{\Omega}}$ -pinene SOA tracers might result from their transfer from particle to gas phase with the increase of aerosol acidity. The ratio of cis-pinonic acid plus pinic acid to 3-methyl-1,2,3-butanetricarboxylic acid (MBTCA) ranged from 0.28 to 28.9 with a mean of 7.19, indicating the relatively fresh $\pmb{\mathrm{\Omega}}$ pinene SOA tracers during our campaign.
$\circledcirc$ 2010 Elsevier Ltd. All rights reserved.
1. Introduction
Secondary organic aerosols (SOA) are produced by homogenous reactions of biogenic and anthropogenic volatile organic compounds (VOCs) with ozone $\left(0_{3}\right)$ , OH and $\mathsf{N O}_{3}$ radicals, followed by nucleation reactions and condensation onto pre-existing particles, or by direct heterogeneous reactions of VOCs on particle surfaces (Jang et al., 2002). SOA can affect the earth’s radiative balance (Hoyle et al., 2009) directly by altering the scattering properties of the atmosphere and indirectly by changing cloud properties. On the global scale, the emissions of biogenic VOCs (BVOCs) dominate over those of anthropogenic VOCs (Piccot et al., 1992; Guenther et al., 1995). The global production of SOA is estimated to reach $6.2\,\mathrm{TgC}\,\mathrm{yr}^{-1}$ from isoprene (Henze and Seinfeld, 2006) and $18.5\,\mathrm{TgC}\,\mathrm{yr}^{-1}$ from other biogenic precursors (Griffin et al., 1999). However, biogenic contribution to SOA is expected to be relatively lower in urban areas, due to more anthropogenic sources of SOA precursors. Based on the identified SOA tracers, Stone et al. (2009) found that toluene alone had greater contribution to secondary organic carbon (SOC) at the urban sites in United States (US), compared to biogenic precursors including isoprene, $\pmb{\alpha}$ -pinene and $\upbeta.$ -caryophyllene. With chemical mass balance model and carbon isotope results, Ding et al. (2008a) estimated that the biogenic contributions were less than $50\%$ in SOC at an urban site in the southeastern US. More complex is that urban emissions can accelerate oxidation of BVOCs (Carlton et al., 2010; Weber et al., 2007; Zhang et al., 2009).
The Pearl River Delta (PRD) region in south China covering ${\sim}41700\,\mathrm{km}^{2}$ is one of the most industrialized and densely populated regions in China with city clusters, e.g. Hong Kong, Guangzhou, Shenzhen, Foshan and Dongguan. The rapid growth in economy has resulted in fast increase in anthropogenic air pollutant emissions in this region (Chan and Yao, 2008). Located in the tropical/subtropical area, the PRD region has annual mean temperature of ${\sim}25\,^{\circ}\mathrm{C}$ and higher biogenic emissions are expected here compared to northern parts of China (Zheng et al., 2010). Therefore both anthropogenic and biogenic SOA precursors are comparatively abundant in the PRD region. On the other hand, the elevated atmospheric oxidative capacity in the PRD region was observed, such as the successive increasing trend of $0_{3}$ during 1994e2007 in Hong Kong, where is located downwind of the PRD region during fall-winter season and the air quality is deeply impacted by the PRD region (Wang et al., 2009). Additionally, the abundant amounts of sulfate and nitrate in the air make the aerosol very acidic in the PRD region. In Hong Kong, the particles were reported to be highly acidic with the pH values ranging from $-0.62$ to 2.35 (Pathak et al., 2004). The high aerosol acidity in the PRD region will further favor SOA formation, since acid-catalyzed reactions can significantly enhance SOA yields (Jang et al., 2002; Tanner et al., 2009; Zhang et al., 2007). Either from precursors or from atmospheric oxidative capacity and the acidic aerosols, SOA should be an important component of particles in the PRD region. In fact, the estimated SOC using the ratio of organic carbon (OC) to element carbon (EC) shared about $50\%$ of ambient OC in the PRD region (Cao et al., 2003, 2004; Duan et al., 2007). A study in Hong Kong during summer 2006 observed that SOA levels enhanced about one order of magnitude on days under the PRD regional impact when compared to days under local influence, among which biogenic precursors exhibited dominant contributions to the SOC (Hu et al., 2008). Nevertheless, to understand chemical composition and origins of OC in the heavily polluted PRD region, more field works are needed to characterize biogenic SOA. Up to now in mainland China, however, there is only one report about the biogenic SOA tracers in forest areas during summer time (Wang et al., 2008) and one model study about the spatial distribution of SOC over China (Han et al., 2008).
Previous studies all suggested that air quality during fall-winter season was the worst in a year in the PRD region, based on the monitoring of $\mathsf{P M}_{2.5}$ , $0_{3}$ , visibility and other criteria pollutants (Wang et al., $2003\mathbf{a}$ ; Wu et al., 2005; Xu et al., 2008). As for BVOC emissions in the PRD region, although they are the highest in summer, relatively high emissions are still expected during fall (September to November) in this tropical/subtropical region (Zheng et al., 2010). Indeed, emissions of BVOC are even more significant when compared to those in the northern cities. For example, BVOC emissions $(2.55\times10^{9}\,\mathrm{g}\,\mathrm{C})$ during fall in Hong Kong contributed $26\%$ of the annual emissions (Leung et al., 2010), while BVOC emissions in Beijing were estimated to be $12.4\times10^{9}\mathrm{~g~C~}$ in summer (June to August) (Wang et al., 2003b). Since the land area of Hong Kong $(\sim\!1100\,\mathrm{km}^{2})$ is less than $1/10$ of Beijing $(\sim\!16,\!800\,\mathrm{km}^{2})$ , the emission strength of BVOCs on unit area in Hong Kong during fall was in fact much higher than that in Beijing during summer. Thus, with relatively high BVOC emissions during fall in the PRD region, it is essential to study the behavior of biogenic SOA in this heavily polluted season, and to investigate the formation mechanism of biogenic SOA under the complex air pollution situation. In the present study, 24-h $\mathsf{P M}_{2.5}$ samples were collected consecutively at a regional background site in the central PRD during OctobereNovember, 2007; and biogenic SOA tracers formed from isoprene, $\pmb{\alpha}$ -pinene and sesquiterpene were measured. The purposes are 1) to provide the levels of biogenic SOA tracers in the heavily-polluted season in the PRD region; 2) to check the influence of temperature and aerosol acidity on biogenic SOA tracers in the real atmosphere.
2. Experimental section
2.1. Field sampling
24-Hour $\mathsf{P M}_{2.5}$ samples were collected using a high volume sampler (Tisch Environmental Inc.) at a rate of $1.1\;\mathrm{m}^{3}\,\mathrm{min}^{-1}$ from October 23 to November 24, 2007 at a rural site, Wangqingsha (WQS, $22^{\circ}42^{\prime}\mathsf{N}$ , $113^{\circ}32^{\prime}\mathrm{E}$ in the PRD region. As showed in Fig. 1, the sampling site is located in the central PRD, with the Pearl River estuary in the south and surrounding city clusters approximately $60\,\mathrm{km}$ away. The sampler was put on the rooftop, about $30\;\mathrm{m}$ above ground, of a seven-floor building in a high school. Since the surrounding terrain is flat with large farmland nearby and rare traffic, this site can serve as an ideal location to monitor the regional background level of air pollution in the PRD region. Prefired $8\times10$ inch quartz filters were covered with aluminum foil and stored in a zipped bag containing silica gel at $4\,^{\circ}\!C$ before and $-20\,^{\circ}\mathrm{C}$ after collection. A total of 32 air samples were collected with two field blanks.
2.2. Chemical analysis
A punch $1.5\times1.0\,\mathrm{cm})$ of each filter was taken for the measurements of OC and EC using the thermo-optical transmittance (TOT) method (NIOSH,1999) by an OC/EC Analyzer (Sunset Laboratory Inc.). An additional punch was taken from each filter and extracted in $20\,\mathrm{mL}$ of 18-MOhm milliQ water and sonicated for $30\,\mathrm{min}$ in an ice-water bath. After filtered, the extract was analyzed for sulfate $(\mathsf{S}0_{4}^{2-})$ , nitrate $\left(\mathsf{N O}_{3}^{-}\right)$ and ammonium $\left(\mathsf{N H}\ddagger\right)$ with an ion chromatography (Metrohm 761, Switzerland). All the data were corrected using field blank. Atmospheric visibility was recorded during the sampling by a visibility sensor (Belfort Model 6000).
For the analysis of biogenic SOA tracers, $1/8$ of each filter was extracted three times by sonication with $40\;\mathrm{mL}$ of dichloride methane (DCM)/methanol $(1{:}1,\mathsf{v}/\mathsf{v})$ for each time. Prior to solvent extraction, hexadecanoic acid- $\cdot\mathrm{D}_{31}$ phthalic acid- $\cdot\mathrm{D}_{4}$ and levoglucosan- $^{13}C_{6}$ were spiked into the samples as internal standards. The extracts of each sample were combined, filtered and concentrated to $\sim\!2\,\mathrm{mL}$ . Then each sample was separated in two parts. One was blown to dryness under a gentle stream of nitrogen, and kept at room temperature for one hour to derivatize acids to methyl esters after adding $200~\upmu\mathrm{L}$ of DCM, $10\,\upmu\mathrm{L}$ of methanol and $300\,\upmu\mathrm{L}$ of fresh prepared diazomethane. The methylated extract was blown to $200\,\upmu\mathrm{L}$ and analyzed for $\mathfrak{x}$ -pinene SOA tracers (cis-pinonic acid, pinic acid and 3-methyl-1,2,3-butanetricarboxylic acid). The other part was blown to dryness for silylation with $100\,\upmu\mathrm{L}$ of pyridine and $200\,\upmu\mathrm{L}$ of N,O-bis-(trimethylsilyl)-trifluoroacetamide (BSTFA) plus $1\%$ trimethylchlorosilane (TMCS) in an oven at $70\,^{\circ}\mathrm{C}$ for one hour. The silylated extract was analyzed for $\mathfrak{a}$ -pinene SOA tracers (3-hydroxyglutaric acid and 3-hydroxy-4,4-dimethylglutaric acid), isoprene SOA tracers (3-methyl-2,3,4-trihydroxy-1-butene, 2-methylglyceric acid, 2-methylthreitol and 2-methylerythritol) and sesquiterpene SOA tracer ( $\upbeta$ -caryophyllinic acid). The typical total ion chromatograms for silylated and methylated samples are presented in Figs. 2 and 3.
Samples were analyzed by an Agilent 5973N gas chromatography/mass spectrometer detector (GC/MSD) in the scan mode with a $30\;\mathrm{m}$ HP-5 MS capillary column ( $\mathrm{i.d.}0.25\:\mathrm{mm}$ , $0.25\,\upmu\mathrm{m}$ film thickness). Splitless injection of a $1\ \upmu\mathrm{L}$ sample was performed. The GC temperature was initiated at $80\,^{\circ}C$ (held for $2\;\mathrm{min}$ ) and increased to $290\,^{\circ}\mathrm{C}$ at $5\,^{\circ}\mathrm{C}\,\mathrm{min}^{-1}$ then held for $20\;\mathrm{min}$ . cis-Pinonic acid and pinic acid were quantified by authentic standards. Due to lack of standards, isoprene SOA tracers were quantified using erythritol (Ding et al., 2008b); $\pmb{\mathrm{\Omega}}$ -pinene SOA tracers (3-methyl1,2,3-butanetricarboxylic acid, 3-hydroxyglutaric acid and 3-hydroxy-4,4-dimethylglutaric acid) were quantified using pinic acid. For $\upbeta$ -caryophyllinic acid, octadecanoic acid was used for quantification due to their adjacent retention times (Fig. 2). These
SOA tracers were identified by the comparison of mass spectra with literature data (Claeys et al., 2004, 2007; Jaoui et al., 2007; Kleindienst et al., 2007; Szmigielski et al., 2007; Wang et al., 2005) and their retention times relative to other known compounds in the GC chromatograms. The method detection limits (MDLs) for cis-pinonic acid, pinic acid, erythritol and octadecanoic acid were 0.04, 0.07, 0.09 and $0.02\:\mathrm{ng}\:\mathrm{m}^{-3}$ , respectively under the average volume of $1575\,\mathrm{m}^{3}$ . It should be noted that 2-methyltetrols can form sulfate and nitrate esters (Surratt et al., 2007a, 2008), which will be measured as free 2-methyltetrols with the derivatization GC-MS method (Wang et al., 2008). Table 1 summarizes the average and range of all measured chemical species during this campaign.
2.3. Quality assurance/quality control (QA/QC)
The field and laboratory blanks were extracted and analyzed in the same way as the field samples. The target compounds were not detected in the field and laboratory blanks. Recoveries of the target compounds in six spiked samples (authentic standards spiked into solvent with pre-baked quartz filter) were $104\pm2\%$ for cis-pinonic acid, $68\pm13\%$ for pinic acid, $62\pm14\%$ for erythritol, and $78\pm10\%$ for octadecanoic acid. The relative differences for the target compounds in duplicate samples $\because n=6!$ ) were all $<\!15\%$ .
3. Results and discussion
3.1. Ambient concentrations of biogenic SOA tracers
During our campaign, relative humidity (RH) and temperature were in the range of $40{-}83\,\%$ and $17.6{-}26.6\,^{\circ}\mathrm{C}$ (Table 1), respectively. $\mathsf{P M}_{2.5}$ averaged $113\pm23.6~\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , which was about 3 times higher than USEPA 24-hour standard of $35\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ (USEPA, 2006). The highest level of $\mathsf{P M}_{2.5}$ reached $171\ \upmu\mathrm{g}\,\mathrm{m}^{-3}$ and the worst visibility dropped to $5.5\;\mathrm{km}$ during the sampling period. Sulfate, OC and nitrate were the dominant components in $\mathsf{P M}_{2.5}$ (Table 1). As Fig. 4a showed, $\mathsf{P M}_{2.5}$ , sulfate, nitrate and ammonium exhibited similar daily variation during this period. Among the SOA tracers (Fig. 4b), the isoprene SOA tracers exhibited the highest levels, $30.8\,\pm$ $15.9\,\mathrm{ng\,m}^{-3}$ , followed by the $\pmb{\alpha}$ -pinene SOA tracers $.6.61\pm$ $4.39\,\mathrm{ng}\,\mathrm{m}^{-3})$ , and $\upbeta$ -caryophyllinic acid $(0.54\pm0.56\;\mathrm{ng\,m}^{-3})$ .
Among the four isoprene SOA tracers, 2-methylerythritol had the highest levels (Fig. 5a). 2-Methyltetrols (sum of 2-methylthreitol and 2-methylerythritol) averaged $27.6\pm15.1$ (5.00 to 64.4) $\mathsf{n g}\,\mathsf{m}^{-3}$ , which is comparable to the results reported during the same season in US (Ding et al., $2008\boldsymbol{\mathrm{b}}$ ; Xia and Hopke, 2006) and during summer in China (Hu et al., 2008; Wang et al., 2008). 3- Methyl-2,3,4-trihydroxy-1-butene (MTHB) and 2-methylglyceric acid were $1.16\pm0.83$ and $2.04\pm1.90\mathrm{\,ng\,m}^{-3}$ , respectively. The summed isoprene SOA tracers ranged from 5.77 to $68.3\;\mathrm{ng}\,\mathrm{m}^{\bar{-}3}$ . The lowest occurred on Nov. 1, 2007 (Fig. 5a) when a cold air mass from North China intruded (Fig. 6) with the ambient air temperature dropped to the lowest $(17.7\,^{\circ}\mathrm{C})$ during the campaign. Another low value observed on Nov. 19 also coincided with cold air mass transport (Fig. 6). The levels of isoprene SOA tracers were expected to be lower when the cold front intruding, due to the dilution of strong winds. As showed in Fig. 4a, the levels of primary (EC) and secondary species (sulfate, nitrate and ammonium), as well as those of $\mathsf{P M}_{2.5}$ and OC, all decreased on Nov. 1 and Nov. 19. The highest level of isoprene SOA tracers occurred around Oct. 27, 2007 with the mean air temperature of $25.1\,^{\circ}\mathrm{C}$ and air mass from the eastern seashore (Fig. 6). Ambient temperature appeared to be an important factor influencing the levels of isoprene SOA tracers. As Fig. 7 showed, the higher levels of isoprene SOA tracers always tracked higher temperature (except 2-methylglyceric acid) during our campaign, probably resulting from the enhanced isoprene emission strength and tracer formation rate under higher temperature (Rinne et al., 2002; Ion et al., 2005; Ding et al., 2008b).
The carbon contribution of 2-methyltetrols to OC was $0.07\pm0.04\%$ with the maximum of $0.19\%$ during our campaign. The OC fraction of 2-methyltetrols was reported as $0.21\substack{-0.85\%}$ in southeastern US (Ding et al., 2008b), $0.74{-}2\%$ in the Amazonian rain forest (Claeys et al., 2004), and $0.02{-}0.75\%$ in the forest of Eastern China (Wang et al., 2008). Cao et al. (2003) reported average OC level of $14.7\;\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in the PRD mega-cities during winter 2001. Andreae et al. (2008) reported OC of $22.4\,\upmu\mathrm{g}\,\mathrm{m}^{-\bar{3}}$ during fall 2004 in urban Guangzhou. In the present study, OC averaged as high as $19.3\;\upmu\mathrm{g}\,\mathrm{m}^{-3}$ . Consequently, although 2-methyltetrols in the present study exhibited comparable levels to those reported in the forest (Claeys et al., 2004; Ding et al., 2008b; Wang et al., 2008), their shares in aerosol OC were much lower due to much higher OC in the PRD region than those in forest areas.
The sum of five $\mathfrak{a}$ -pinene SOA tracers ranged from 0.91 to $18.3\;\mathrm{ng}\,\mathrm{m}^{-3}$ with an average of $6.61\pm4.39\mathrm{\,ng\,m^{-3}}$ (Table 1, Fig. 5b). Among the $\mathfrak{x}$ -pinene SOA tracers, cis-pinonic acid had the highest level $(3.60\pm3.76\;\mathrm{ng\,m}^{-3})$ , while the concentrations of pinic acid $\left(1.25\pm0.79\:\mathrm{ng\,m}^{-3}\right)$ ) and 3-methyl-1,2,3-butanetricarboxylic acid (MBTCA, $1.16\pm0.99\mathrm{\,ng\,m}^{-3};$ were about one third of that of cispinonic acid. Only trace amounts of 3-hydroxyglutaric acid and 3-hydroxy-4,4-dimethylglutaric acid (HDMGA) were detected. The levels of these tracers during our campaign were also comparable to those in the same season in US (Ding et al., 2008b; Lewandowski et al., 2008; Sheesley et al., 2004). Unlike the isoprene SOA tracers, the $\mathfrak{x}$ -pinene SOA tracers showed poor correlations with temperature $(p>0.05$ , all tracers). Although high temperature could enhance both $\pmb{\alpha}$ -pinene emission (Guenther et al., 1995) and tracer formation, increasing temperature would favor the evaporation of these tracers from the particle phase into the gas phase (Saathoff et al., 2009). Therefore, the poor correlations observed between $\pmb{\alpha}$ -pinene SOA tracers and temperature might be the result of counteraction of temperature effects on the precursor emission/ tracer formation and gas/particle partitioning. Similarly, the sesquiterpene SOA tracer, $\upbeta$ -caryophyllinic acid, also showed poor correlation with temperature $(p>0.05)$ .
3.2. Correlations between SOA tracers and aerosol acidity
Previous chamber studies have revealed that acid-catalyzed heterogeneous reactions can significantly enhance SOA yields (Jang et al., 2002; Surratt et al., 2010). In this study, aerosol acidity $([\mathsf{H}^{+}]$ $\mathsf{\mu m o l}\,\mathsf{m}^{-3},$ was estimated using a charge balance of $\mathrm{SO}_{4}^{2-}$ , $\mathtt{N O}_{3}^{-}$ and $\mathsf{N H}_{4}^{+}$ and calculated as below (Pathak et al., 2009):
$$
\left[\mathrm{H}^{+}\right]\,=\,2\times\left[50_{4}^{2-}\right]+\left[\mathrm{NO}_{3}^{-}\right]-\left[\mathrm{NH}_{4}^{+}\right]
$$
where [X] represents the molar concentration of the ion $X.$ . In this study, the estimated $[\mathsf{H}^{+}]$ in the particulate ranged from 0.15 to $0.77\,\upmu\mathrm{mol}\,\mathrm{m}^{-3}$ , which was in the same range in the major cities of China (Pathak et al., 2004, 2009).
Significant correlations were observed between the estimated $[\mathsf{H}^{+}]$ and the isoprene SOA tracers (Fig. 8a), reflecting the enhancement of isoprene SOA formation in the acidic aerosols in the PRD. The slopes for MTHB, 2-methylthreitol and 2-methylerythritol were $3.6\pm0.7,$ $22.9\pm4.4$ and $\bar{3}6.5\pm9.2\;\mathrm{ng}\,\mathrm{m}^{-3}$ per $\sharp\mathrm{mol}$ $[\mathrm{H}^{+}]\,\mathrm{~m}^{-3}$ , respectively (Fig. 8a). Similar to chamber simulation (Surratt et al., 2007b), 2-methylglyceric acid had poor correlation with aerosol acidity in our ambient study (Fig. 8a). The slope of 2-methyltetrols vs. acidity was $59.4\pm13.4\;\mathrm{ng}\,\mathrm{m}^{-3}$ per $\sharp\mathrm{mol}$ $[\mathsf{H}^{+}]$ $\mathfrak{m}^{-3}$ or $26.2\pm6.9\mathrm{\,ngC\,m}^{-3}$ per $\sharp\mathrm{mol}$ $1~[\mathsf{H}^{+}]~\mathsf{m}^{-3}$ , which was more than 2 orders of magnitude lower than that $(0.00925\mathrm{~}\upmu\mathrm{g}\,\mathrm{C}\,\mathrm{m}^{-3}$ per nmol $[\mathsf{H}^{+}]\,\mathsf{m}^{-3}$ or $9.{\bar{2}}5\times10^{3}\,\mathrm{ng}\,\mathrm{C}\,\mathrm{m}^{-3}$ per $\mathsf{\mu m o l\,[H^{+}]\,m}^{-3\cdot}$ obtained in chamber studies (Surratt et al., 2007b; Offenberg et al., 2009).
There might be several reasons for the huge gap observed between chamber simulation and our ambient measurement in 2-methyltetrols’ level responding to aerosol acidity. In consideration of the analytical aspects, the calibration standard used to determine 2-methyltetrols was erythritol in this study instead of cis-ketopinic acid (KPA) in the chamber studies (Surratt et al., $2007\mathbf{b}$ ; Offenberg et al., 2009). However, as reported by Hu et al. (2008), the uncertainty caused by using compounds other than KPA was estimated to be within a factor of 3, which is minor compared to 2 orders of magnitude for the observed difference in slopes. Also it is worth noting that the charge balance method was used to estimate aerosol acidity in the present study, while aerosol acidity was determined by measuring pH in the chamber studies (Surratt et al., 2007b; Offenberg et al., 2009). It should be stressed that the charge balance method may have uncertainties, since that only three water soluble ions were considered to estimate aerosol acidity, and that fine particles were not homogeneously mixed, i.e., some particles were not acidic while others were much more acidic. Thus, the two methods would result in difference in measured $[\mathsf{H}^{+}]$ values, but such a difference was unlikely as huge as 2 orders of magnitude. Therefore, despite of the difference in above analytical methods, the major reason for the observed difference in slopes of 2-methyltetrols to $[\mathsf{H}^{+}]$ should be that the ambient conditions were much more complicated than those in chamber simulation.
Firstly, up to now the effects of acidity on the formation of 2-methyltetrols in chamber studies (Surratt et al., $2007\mathbf{b}$ ; Offenberg et al., 2009) are limited to low RH $(30\%)$ . During our field campaign, however, the RH values were much higher $(40{-}83\%)$ . Higher RH was more likely to induce a relatively lower acidity in the aerosol droplets due to more water uptake; the enhancement of 2-methyltetrols formation by acidity was thus reduced. In this aspect, to give a sound explanation to the field results, further chamber studies are needed about the influence of RH on 2-methyltetrols formation, particularly of high RH as occurring in the PRD region. Secondly, ambient isoprene levels were quite different from those in chambers. In the chamber studies (Surratt et al., 2007b; Offenberg et al., 2009), an excess of isoprene was maintained to decouple the influence of isoprene level on the simulation result; while in the ambient, mixing ratios of isoprene were often less than 1 ppbv and its emission from plants varied greatly during a day, typically reaching maximum at noon but nearly zero at night. Thirdly, the compositions of airborne species in both gas and particulate phases were much more complicated with even more complicated interactions than those in chambers. It may be true in the highly manipulated chamber studies that the acidity of the aerosol solely causes the increase of isoprene SOA tracers, but not so in the ambient. In the real atmosphere, isoprene SOA formation depends on not only aerosol acidity but also other factors, such as emission strength, reaction rate, and oxidant levels. Although it is not easy to figure out an explanation for the difference at present, the huge gap observed for the dependence of 2-methyltetrols on aerosol acidity between chamber simulation and our ambient measurement suggested that further chamber studies are needed especially under the situations as occurring in the PRD region to narrow the gap between chamber simulation and the real atmosphere.
Two $\pmb{\alpha}$ -pinene SOA tracers, cis-pinonic acid and pinic acid, exhibited negative correlations with the estimated $[\mathsf{H}^{+}]$ with the slopes of $-11.7\pm3.7$ and $-2.23\pm0.81\mathrm{~ng~m}^{-3}$ per $\sharp\mathrm{mol}$ $\dot{[\mathrm{H}^{+}]}\,\mathrm{m}^{-3}$ , respectively (Fig. 8b). Other three $\mathfrak{x}$ -pinene SOA tracers did not show significant correlations with aerosol acidity $\mathrm{\Delta}p>0.05$ , Fig. 8b). The possible explanation for the negative correlation is the influence of aerosol acidity on the gas/particle partitioning. High acidity could result in more cis-pinonic acid and pinic acid in the molecular form in the aqueous solutions of the particle phase. These two acids were semi-volatile and typically dominant in gas phase (Yu et al., 1999). More cis-pinonic acid and pinic acid in the molecular form in particles under more acidic environment would favor their release from the particle phase into the gas phase, therefore the decreasing trend of these tracers in the particle phase would be expected with increasing aerosol acidity. The sesquiterpene-SOA tracer, $\upbeta{\mathrm{.}}$ -caryophyllinic acid did not show significant correlation with aerosol acidity $(p>0.05)$ .
Since cis-pinonic acid and pinic acid were the dominant compounds among the five $\mathfrak{a}$ -pinene SOA tracers, the sum of $\pmb{\alpha}$ -pinene SOA tracers also exhibited the acidity dependence $r\!=\!-0.41$ , $p=0.02$ , $n=32$ ) with a slope of $-11.3\pm4.6\;\mathrm{ng}\,\mathrm{m}^{-3}$ per mmol $[\mathsf{H}^{+}]\;\mathsf{m}^{-3}$ or $-7.8\pm2.8\mathrm{\ng\,C\,m}^{-3}$ per m $\mathrm{nol~}[\mathrm{H^{+}}]\mathrm{~m~}^{-3}$ . In the chamber study, the summed concentration of nine $\mathfrak{x}$ -pinene tracers decreased linearly by $0.00380\ensuremath{~\upmu\mathrm{g}\,\mathrm{C}\,\mathrm{m}^{-3}}$ per nmol $[\mathsf{H}^{+}]\,\mathsf{~m}^{-3}$ or $3.8\times10^{3}\mathrm{\,ng\,C\,m^{-\bar{3}}}$ per mmol $[\mathsf{H}^{+}]\,\mathsf{m}^{-3}$ (Offenberg et al., 2009). Since the remaining four tracers were not detected during our campaign (Fig. 2), the slope $(-7.8\pm2.8\mathrm{{\ng}{C}\,\mathrm{{m}^{-3}}}$ per $\mathrm{\underline{{\{umol\;\;}}}[H^{+}]\;\;m^{-3}})$ observed in the current study could represent that of the whole nine tracers. This slope was also more than 2 orders of magnitude lower than that from above chamber study (Offenberg et al., 2009). Again, a huge gap was observed for the acidic dependence of $\pmb{\alpha}$ -pinene SOA tracers between chamber simulation and our ambient measurement, suggesting that further chamber studies are needed especially under the situations as occurring in the PRD region to narrow the gap between chamber simulation and the real atmosphere.
3.3. Implication of formation pathway for SOA tracers
As showed in Table 2a, significant correlations were observed between isoprene tracers except for 2-methylglyceric acid. As Surratt et al. (2010) proposed, 2-methyltetrols can be formed by further acid-catalyzed heterogeneous reactions of gaseous isoprene oxidation products, such as epoxydiols of isoprene (IEPOX) under low- $\cdot\mathrm{NO}_{\tt x}$ conditions. The particle-phase ring-opening reaction of IEPOX played an important role in the enhancement of 2-methyletetrols from 0.1 to $\bar{5}.1\ \upmu\mathrm{g}\,\mathrm{m}^{-3}$ when aerosol acidity increased from the neutral to the highly acidic (Surratt et al., 2010). During our sampling, $\mathsf{N O}_{\mathtt{X}}$ level was $31\pm2$ ppbv (Guo et al., 2009). Obviously, the rural site is a low- $\cdot\mathrm{NO}_{\tt X}$ case. Moreover, 2-methyletetrols exhibited significantly positive correlations with aerosol acidity in this study (Fig. 8a); and there was one order of magnitude enhancement of 2-methyltetrols (5.59 to $64.83\mathrm{\;ng\,m^{-3}}$ ) with aerosol acidity increasing from 0.17 to $0.74\,\upmu\mathrm{mol}$ $[\mathsf{H}^{+}]\:\:\mathsf{m}^{-3}$ . Furthermore, since the $C_{5}$ -alkene triol, MTHB also shares the same low- $\cdot\mathrm{NO}_{\tt x}$ formation pathway as 2-methyltetrols (Surratt et al., 2010), the good correlations between MTHB and 2-methyltetrols are expected, which were observed in the current study (Table 2a). All these indicated that acid-catalyzed heterogeneous reaction was the major formation pathway for 2-methyltetrols in the PRD region. The slope of 2-methylthreitol to 2-methylerythritol was $0.52\pm0.03$ , quite near those reported in US (Ding et al., 2008b; Xia and Hopke, 2006; Yan et al., 2009), implying the ratio of the production rates for these two tracers might be relatively constant from place to place. On the other hand, 2-methylglyceric acid was not correlated with other tracers (Table 2a), indicating it had different formation mechanism from other isoprene SOA tracers. In fact, Surratt et al. (2010) proposed that 2-methylgriteric acid was produced by decomposition of $C_{4}$ -hydroxynitrate-PAN in the aerosol phase under high- $\cdot\mathrm{NO}_{\tt x}$ conditions, which is quite different from the pathway of 2-methyltetrols under low- $\cdot\mathrm{NO}_{\tt x}$ conditions. It worth noting that 2-methylglyceric acid is an acid while other three isoprene SOA tracers are polyhydric alcohols. 2-Methylglyceric acid would, as discussed above, probably be more influenced by aerosol acidity in its gas/particle partitioning.
For $\mathfrak{a}$ -pinene SOA tracers, interestingly, their correlations exhibited distinct patterns (Table 2b). cis-Pinonic acid and pinic acid were correlated well, while MBTCA, 3-hydroxyglutaric acid and HDMGA were significantly correlated with each other. However, the former two tracers showed poor correlations with the latter three. Although MBTCA has other precursors apart from $\pmb{\alpha}$ -pinene (Jaoui et al., 2005; Szmigielski et al., 2007), the significant correlations among MBTCA, 3-hydroxyglutaric acid and HDMGA suggested $\pmb{\alpha}$ -pinene was probably its major precursor. Previous studies proposed that cis-pinonic acid and pinic acid could further photo-degrade to MBTCA (Claeys et al., 2007; Szmigielski et al., 2007), which partly explained the relatively low or undetectable levels of cis-pinonic acid and pinic acid in the air (Hu et al., 2008; Wang et al., 2008). If such a photo-degradation process does exist in the real atmosphere, the ratio of cis-pinonic acid plus pinic acid to MBTCA $(\mathsf{P}/\mathsf{M})$ can be employed to check the aging of $\mathfrak{x}$ -pinene SOA. Higher $\mathsf{P/M}$ ratio indicates less transformation of cis-pinonic acid and pinic acid to MBTCA and thus $\pmb{\alpha}$ -pinene SOA is relatively fresh; while lower $\mathsf{P/M}$ ratio points to relatively aged $\mathfrak{x}$ -pinene SOA.
In the current study, the lowest $\mathsf{P/M}$ ratio (0.28) was observed when the air mass came from the eastern seashore; while the highest $\mathsf{P/M}$ (28.9) coupled with the air mass from the continent (Fig. 9). Since there should have fresh input of $\mathfrak{a}$ -pinene in the land but not so over the sea, $\mathfrak{a}$ -pinene SOA from the eastern seashore would be more aged compared to that from the continent. The $\mathsf{P/M}$ ratios averaged $7.19\pm7.48$ , indicating that overall the $\pmb{\alpha}$ -pinene SOA tracers were relatively fresh during our campaign.
Acknowledgements
This research was supported by NSFC-Guangdong Joint Funds (U0833003), the National Science Foundation of China $(40821003/$ 40928005/41003045) and Guangzhou Institute of Geochemistry (GIGCX-08-07). This is contribution no. IS-1276 from GIGCAS.
Appendix. Supplementary data
Supplementary data associated with this article can be found in the online version, at doi:10.1016/j.atmosenv.2010.11.057.
References
Andreae, M.O., Schmid, O., Yang, H., Chand, D., Zhen Yu, J., Zeng, L.M., Zhang, Y.H., 2008. Optical properties and chemical composition of the atmospheric aerosol in urban Guangzhou, China. Atmospheric Environment 42, 6335e6350.
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. Atmospheric Environment 37, 1451e1460.
Cao, J.J., Lee, S.C., Ho, K.F., Zou, S.C., Fung, K., Li, Y., Watson, J.G., Chow, J.C., 2004. Spatial and seasonal variations of atmospheric organic carbon and elemental carbon in Pearl River Delta Region, China. Atmospheric Environment 38, 4447e4456.
Carlton, A.G., Pinder, R.W., Bhave, P.V., Pouliot, G.A., 2010. To what extent can biogenic SOA be controlled? Environmental Science & Technology 44, 3376e3380.
Chan, C.K., Yao, X., 2008. Air pollution in mega cities in China. Atmospheric Environment 42, 1e42.
Claeys, M., Graham, B., Vas, G., Wang, W., Vermeylen, R., Pashynska, V., Cafmeyer, J., Guyon, P., Andreae, M.O., Artaxo, P., Maenhaut, W., 2004. Formation of secondary organic aerosols through photooxidation of isoprene. Science 303, 1173e1176.
Claeys, M., Szmigielski, R., Kourtchev, I., VanderVeken, P., Vermeylen, R., Maenhaut, W., Jaoui, M., Kleindienst, T.E., Lewandowski, M., Offenberg, J.H., Edney, E.O., 2007. Hydroxydicarboxylic acids: markers for secondary organic aerosol from the photooxidation of $\mathfrak{a}$ -pinene. Environmental Science & Technology 41, 1628e1634.
Ding, X., Zheng, M., Edgerton, E.S., Jansen, J.J., Wang, X., 2008a. Contemporary or fossil origin: split of estimated secondary organic carbon in the southeastern United States. Environmental Science & Technology 42, 9122e9128.
Ding, X., Zheng, M., Yu, L., Zhang, X., Weber, R.J., Yan, B., Russell, A.G., Edgerton, E.S., Wang, X., 2008b. Spatial and seasonal trends in biogenic secondary organic aerosol tracers and water-soluble organic carbon in the southeastern United States. Environmental Science & Technology 42, 5171e5176.
Duan, J., Tan, J., Cheng, D., Bi, X., Deng, W., Sheng, G., Fu, J., Wong, M.H., 2007. Sources and characteristics of carbonaceous aerosol in two largest cities in Pearl River Delta Region, China. Atmospheric Environment 41, 2895e2903.
Griffin, R.J., Cocker III, D.R., Seinfeld, J.H., Dabdub, D., 1999. Estimate of global atmospheric organic aerosol from oxidation of biogenic hydrocarbons. Geophysical Research Letters 26, 2721e2724.
Guenther, A., Hewitt, C.N., Erickson, D., Fall, R., Geron, C., Graedel, T., Harley, P., Klinger, L., Lerdau, M., McKay, W.A., Pierce, T., Scholes, B., Steinbrecher, R., Tallamraju, R., Taylor, J., Zimmerman, P., 1995. A global model of natural volatile organic compound emissions. Journal of Geophysical Research 100, 8873e8892.
Guo, H., Jiang, F., Cheng, H.R., Simpson, I.J., Wang, X.M., Ding, A.J., Wang, T.J., Saunders, S.M., Wang, T., Lam, S.H.M., Blake, D.R., Zhang, Y.L., Xie, M., 2009. Concurrent observations of air pollutants at two sites in the Pearl River Delta and the implication of regional transport. Atmospheric Chemistry and Physics 9, 7343e7360.
Han, Z.W., Zhang, R.J., Wang, Q.G., Wang, W., Cao, J.J., Xu, J., 2008. Regional modeling of organic aerosols over China in summertime. Journal of Geophysical Research 113. doi:10.1029/2007JD009436.
Henze, D.K., Seinfeld, J.H., 2006. Global secondary organic aerosol from isoprene oxidation. Geophysical Research Letters 33, L09812. doi:10.1029/2006GL025976.
Hoyle, C.R., Myhre, G., Berntsen, T.K., Isaksen, I.S.A., 2009. Anthropogenic influence on SOA and the resulting radiative forcing. Atmospheric Chemistry and Physics 9, 2715e2728.
Hu, D., Bian, Q., Li, T.W.Y., Lau, A.K.H., Yu, J.Z., 2008. Contributions of isoprene, monoterpenes, $\upbeta$ -caryophyllene, and toluene to secondary organic aerosols in Hong Kong during the summer of 2006. Journal of Geophysical Research 113, D22206. doi:10.1029/2008JD010437.
Ion, A.C., Vermeylen, R., Kourtchev, I., Cafmeyer, J., Chi, X., Gelencsér, A., Maenhaut, W., Claeys, M., 2005. Polar organic compounds in rural PM2.5 aerosols from K-puszta, Hungary, during a 2003 summer field campaign: sources and diurnal variations. Atmospheric Chemistry and Physics 5, 1805e1814.
Jang, M., Czoschke, N.M., Lee, S., Kamens, R.M., 2002. Heterogeneous atmospheric aerosol production by acid-catalyzed particle-phase reactions. Science 298, 814e817.
Jaoui, M., Kleindienst, T.E., Lewandowski, M., Offenberg, J.H., Edney, E.O., 2005. Identification and quantification of aerosol polar oxygenated compounds bearing carboxylic or hydroxyl groups. 2. organic tracer compounds from monoterpenes. Environmental Science & Technology 39, 5661e5673.
Jaoui, M., Lewandowski, M., Kleindienst, T.E., Offenberg, J.H., Edney, E.O., 2007. b-Caryophyllinic acid: an atmospheric tracer for $\upbeta$ -caryophyllene secondary organic aerosol. Geophysical Research Letters 34, L05816. doi:10.1029/2006GL028827.
Kleindienst, T.E., Jaoui, M., Lewandowski, M., Offenberg, J.H., Lewis, C.W., Bhave, P.V., Edney, E.O., 2007. Estimates of the contributions of biogenic and anthropogenic hydrocarbons to secondary organic aerosol at a southeastern US location. Atmospheric Environment 41, 8288e8300.
Leung, D.Y.C., Wong, P., Cheung, B.K.H., Guenther, A., 2010. Improved land cover and emission factors for modeling biogenic volatile organic compounds emissions from Hong Kong. Atmospheric Environment 44, 1456e1468.
Lewandowski, M., Jaoui, M., Offenberg, J.H., Kleindienst, T.E., Edney, E.O., Sheesley, R.J., Schauer, J.J., 2008. Primary and secondary contributions to ambient PM2.5 in the midwestern United States. Environmental Science & Technology 42, 3303e3309.
NIOSH, 1999. Method 5040 Issue 3 (Interim): elemental carbon (diesel exhaust). In: NIOSH Manual of Analytical Methods. National Institute of Occupational Safety and Health, Cincinnati, OH.
Offenberg, J.H., Lewandowski, M., Edney, E.O., Kleindienst, T.E., Jaoui, M., 2009. Influence of aerosol acidity on the formation of secondary organic aerosol from biogenic precursor hydrocarbons. Environmental Science $\&$ Technology 43, 7742e7747.
Pathak, R.K., Louie, P.K.K., Chan, C.K., 2004. Characteristics of aerosol acidity in Hong Kong. Atmospheric Environment 38, 2965e2974.
Pathak, R.K., Wu, W.S., Wang, T., 2009. Summertime $\mathrm{PM}_{2.5}$ ionic species in four major cities of China: nitrate formation in an ammonia-deficient atmosphere. Atmospheric Chemistry and Physics 9, 1711e1722.
Piccot, S., Watson, J., Jones, J., 1992. A global inventory of volatile organic compound emissions from anthropogenic sources. Journal of Geophysical Research 97 (D9), 9897e9912.
Rinne, H.J.I., Guenther, A.B., Greenberg, J.P., Harley, P.C., 2002. Isoprene and monoterpene fluxes measured above Amazonian rainforest and their dependence on light and temperature. Atmospheric Environment 36, 2421e2426.
Saathoff, H., Naumann, K.H., Möhler, O., JonssonÅ, M., Hallquist, M., KiendlerScharr, A., Mentel, T.F., Tillmann, R., Schurath, U., 2009. Temperature dependence of yields of secondary organic aerosols from the ozonolysis of $\mathfrak{u}$ -pinene and limonene. Atmospheric Chemistry and Physics 9, 1551e1577.
Sheesley, R.J., Schauer, J.J., Bean, E., Kenski, D., 2004. Trends in secondary organic aerosol at a remote site in Michigan’s Upper Peninsula. Environmental Science & Technology 38, 6491e6500.
Stone, E.A., Zhou, J., Snyder, D.C., Rutter, A.P., Mieritz, M., Schauer, J.J., 2009. A comparison of summertime secondary organic aerosol source contributions at contrasting urban locations. Environmental Science & Technology 43, 3448e3454.
Surratt, J.D., Kroll, J.H., Kleindienst, T.E., Edney, E.O., Claeys, M., Sorooshian, A., Ng, N.L., Offenberg, J.H., Lewandowski, M., Jaoui, M., Flagan, R.C., Seinfeld, J.H., 2007a. Evidence for organosulfates in secondary organic aerosol. Environmental Science & Technology 41, 517e527.
Surratt, J.D., Lewandowski, M., Offenberg, J.H., Jaoui, M., Kleindienst, T.E., Edney, E.O., Seinfeld, J.H., 2007b. Effect of acidity on secondary organic aerosol formation from isoprene. Environmental Science & Technology 41, 5363e5369.
Surratt, J.D., Gmez-Gonzlez, Y., Chan, A.W.H., Vermeylen, R., Shahgholi, M., Kleindienst, T.E., Edney, E.O., Offenberg, J.H., Lewandowski, M., Jaoui, M., Maenhaut, W., Claeys, M., Flagan, R.C., Seinfeld, J.H., 2008. Organosulfate formation in biogenic secondary organic aerosol. The Journal of Physical Chemistry A 112, 8345e8378.
Surratt, J.D., Chan, A.W.H., Eddingsaas, N.C., Chan, M., Loza, C.L., Kwan, A.J., Hersey, S.P., Flagan, R.C., Wennberg, P.O., Seinfeld, J.H., 2010. Reactive intermediates revealed in secondary organic aerosol formation from isoprene. Proceedings of the National Academy of Sciences of the United States of America 107, 6640e6645.
Szmigielski, R., Surratt, J.D., Gómez-González, Y., Veken, P.V.D., Kourtchev, I., Vermeylen, R., Blockhuys, F., Jaoui, M., Kleindienst, T.E., Lewandowski, M., Offenberg, J.H., Edney, E.O., Seinfeld, J.H., Maenhaut, W., Claeys, M., 2007. 3- Methyl-1,2,3-butanetricarboxylic acid: an atmospheric tracer for terpene secondary organic aerosol. Geophysical Research Letters 34, L24811. doi:10.1029/2007GL031338.
Tanner, R.L., Olszyna, K.J., Edgerton, E.S., Knipping, E., Shaw, S.L., 2009. Searching for evidence of acid-catalyzed enhancement of secondary organic aerosol formation using ambient aerosol data. Atmospheric Environment 43, 3440e3444.
USEPA, 2006. National Ambient Air Quality Standards (NAAQS). Available from: http://www.epa.gov/air/criteria.html.
Wang, T., Poon, C.N., Kwok, Y.H., Li, Y.S., 2003a. Characterizing the temporal variability and emission patterns of pollution plumes in the Pearl River Delta of China. Atmospheric Environment 37, 3539e3550.
Wang, T., Wei, X.L., Ding, A.J., Poon, C.N., Lam, K.S., Li, Y.S., Chan, L.Y., Anson, M., 2009. Increasing surface ozone concentrations in the background atmosphere of Southern China, 1994e2007. Atmospheric Chemistry and Physics 9, 6217e6227.
Wang, W., Kourtchev, I., Graham, B., Cafmeyer, J., Maenhaut, i., Claeys, M., 2005. Characterization of oxygenated derivatives of isoprene related to 2-methyltetrols in Amazonian aerosols using trimethylsilylation and gas chromatography/ ion trap mass spectrometry. Rapid Communications in Mass Spectrometry 19, 1343e1351.
Wang, W., Wu, M.H., Li, L., Zhang, T., Li, H.J., Wang, Y.J., Liu, X.D., Sheng, G.Y., Claeys, M., Fu, J.M., 2008. Polar organic tracers in PM2.5 aerosols from forests in eastern China. Atmospheric Chemistry and Physics 8, 7507e7518.
Wang, Z., Bai, Y., Zhang, S., 2003b. A biogenic volatile organic compounds emission inventory for Beijing. Atmospheric Environment 37, 3771e3782.
Weber, R.J., Sullivan, A.P., Peltier, R.E., Russell, A., Yan, B., Zheng, M., de Gouw, J.A., Warneke, C., Brock, C., Holloway, J.S., Atlas, E.L., Edgerton, E., 2007. A study of secondary organic aerosol formation in the anthropogenic-influenced southeastern United States. Journal of Geophysical Research 112, D13302. doi:10.1029/2007JD008408.
Wu, D., Tie, X., Li, C., Ying, Z., Kai-Hon Lau, A., Huang, J., Deng, X., Bi, X., 2005. An extremely low visibility event over the Guangzhou region: a case study. Atmospheric Environment 39, 6568e6577.
Xia, X., Hopke, P.K., 2006. Seasonal variation of 2-methyltetrols in ambient air samples. Environmental Science & Technology 40, 6934e6937.
Xu, H., Wang, X., Pöesch, U., Feng, S., Wu, D., Yang, L., Li, S., Song, W., Sheng, G., Fu, J., 2008. Genotoxicity of total and fractionated extractable organic matter in fine air particulate matter from urban Guangzhou: comparison between haze and nonhaze episodes. Environmental Toxicology & Chemistry 27, 206e212.
Yan, B., Zheng, M., Hu, Y., Ding, X., Sullivan, A.P., Weber, R.J., Baek, J., Edgerton, E.S., Russell, A.G., 2009. Roadside, urban, and rural comparison of primary and secondary organic molecular markers in ambient $\mathrm{PM}_{2.5}$ Environmental Science & Technology 43, 4287e4293.
Yu, J., Griffin, R.J., Cocker III, D.R., Flagan, R.C., Seinfeld, J.H., Blanchard, P., 1999. Observation of gaseous and particulate products of monoterpene oxidation in forest atmospheres. Geophysical Research Letters 26, 1145e1148.
Zhang, Q., Jimenez, J.L., Worsnop, D.R., Canagaratna, M., 2007. A case study of urban particle acidity and its influence on secondary organic aerosol. Environmental Science & Technology 41, 3213e3219.
Zhang, R., Wang, L., Khalizov, A.F., Zhao, J., Zheng, J., McGraw, R.L., Molina, L.T., 2009. Formation of nanoparticles of blue haze enhanced by anthropogenic pollution. Proceedings of the National Academy of Sciences of the United States of America 106, 17650e17654.
Zheng, J., Zheng, Z., Yu, Y., Zhong, L., 2010. Temporal, spatial characteristics and uncertainty of biogenic VOC emissions in the Pearl River Delta region, China. Atmospheric Environment 44, 1960e1969.
|
atmosphere
|
0013
|
10.1016/j.scitotenv.2005.03.001
|
image
|
Fig. 1. Weekly concentrations of $\mathrm{PM}_{2.5}$ , $\mathrm{PM}_{10}$ and gaseous pollutants at Chegongzhuang site from August 2001 to September 2002.
|
|
atmosphere
|
0013
|
10.1016/j.scitotenv.2005.03.001
|
table
|
Table 1 Seasonal average concentrations of gas pollutants, $\mathrm{PM}_{10}$ , $\mathrm{PM}_{2.5}$ and the chemical species at CGZ site (unit: $\upmu\mathrm{g}\textrm{m}^{-3}$ )
|
|
atmosphere
|
0013
|
10.1016/j.scitotenv.2005.03.001
|
image
|
Fig. 2. Weekly averaged meteorological conditions data in Beijing from August 2001 to September 2002.
|
|
atmosphere
|
0013
|
10.1016/j.scitotenv.2005.03.001
|
table
|
Table 2 Annual average concentrations of major pollutants in Beijing from 1998 to 2002 (unit: $\mathrm{mg~m}^{-3}$ )
|
|
atmosphere
|
0013
|
10.1016/j.scitotenv.2005.03.001
|
table
|
Table 3 Enrichment factors of elements in $\mathrm{PM}_{2.5}$
|
|
atmosphere
|
0013
|
10.1016/j.scitotenv.2005.03.001
|
image
|
Fig. 3. Comparison of $\mathrm{SO}_{4}^{2-}$ and $\mathrm{SO}_{2}$ concentrations at Chegongzhuang site.
|
|
atmosphere
|
0013
|
10.1016/j.scitotenv.2005.03.001
|
table
|
Table 4 Matrix of correlation coefficient among the $\operatorname{PM}_{2.5,}$ , water-soluble ions and carbonaceous species at CGZ site
|
|
atmosphere
|
0013
|
10.1016/j.scitotenv.2005.03.001
|
image
|
Fig. 4. Equivalent concentrations comparison of (a) $\mathrm{NO}_{3}^{-}\!+\!\mathrm{SO}_{4}^{2-}$ vs. $\mathrm{NH_{4}^{+}}$ , (b) total cations vs. total anions.
|
|
atmosphere
|
0013
|
10.1016/j.scitotenv.2005.03.001
|
image
|
Fig. 5. Weekly concentrations of organic carbon and elemental carbon at Chegongzhuang site and Tsinghua site.
|
|
atmosphere
|
0013
|
10.1016/j.scitotenv.2005.03.001
|
image
|
Fig. 6. Contributions of chemical species to $\mathrm{PM}_{2.5}$ mass concentrations at CGZ site and THU site. (a) Annual average contributions. (b) Seasonal average contributions.
|
|
atmosphere
|
0013
|
10.1016/j.scitotenv.2005.03.001
|
text
| Not supported with pagination yet |
Concentration and chemical characteristics of $\mathrm{PM}_{2.5}$ in Beijing, China: 2001–2002
F.K. Duana, K.B. Hea,T, Y.L. Maa, F.M. Yanga, X.C. Yua, S.H. Cadleb, T. Chanb, P.A. Mulawab
aDepartment of Environmental Science and Engineering, Tsinghua University, Beijing 100084, P.R. China bGM R&D Center, Chemical and Environmental Sciences Laboratory, MC 480-106-269, Warren, MI 48090, USA
Received 23 July 2004; accepted 1 March 2005 Available online 23 September 2005
Abstract
Weekly $\mathrm{PM}_{2.5}$ samples were simultaneously collected at a semi-residential (Tsinghua University) and a downtown (Chegongzhuang) site in Beijing from August 2001 through September 2002. The ambient mass concentration and chemical composition of $\mathrm{PM}_{2.5}$ were determined. Analyses including elemental composition, water-soluble ions, and organic and elemental carbon were performed. The annual average concentrations of $\mathrm{PM}_{2.5}$ were $96.5~\upmu\mathrm{g}\,\,\mathrm{m}^{-3}$ and $106.9~\upmu\mathrm{g}~\mathrm{m}^{-3}$ at CGZ and HU site, respectively. More than $80\%$ of the $\mathrm{PM}_{2.5}$ mass concentrations were explained by carbonaceous species, secondary particles, crustal matters and trace elements at the two sites. Carbonaceous species were the most abundant components, constituting about $45\%$ and $48\%$ of the total $\mathrm{PM}_{2.5}$ mass concentrations at CGZ and THU site, respectively. $\mathrm{SO}_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ and $\mathrm{NH_{4}^{+}}$ were three major ions, accounting for $37\%$ , $23\%$ and $20\%$ , respectively, of the total mass of inorganic water-soluble ions. $\copyright$ 2005 Published by Elsevier B.V.
Keywords: $\mathrm{PM}_{2.5}$ ; Water-soluble ions; Carbonaceous species; Crustal matter; $\mathrm{PM}_{2.5}$ mass balance
1. Introduction
Fine particulate matter (with aerodynamic diameter less than $2.5\ \upmu\mathrm{m}$ , $\operatorname{PM}_{2.5}$ ) has drawn worldwide attentions for its adverse impact on human health (Dockery and Pope, 1994; Lipfert and Wyzga, 1995;
Schwartz, 1993; Pagano et al., 1998), association with visibility degradation (Appel et al., 1985; Baik et al., 1996; Chan et al., 1999) and global climate change (Charlson et al., 1991; Novakov and Penner, 1993; Wexler and Ge, 1998). In China, a combination of rapid industrialization and high population density inevitably deteriorates the air pollution problems, of which PM has been frequently observed the principal pollutant in most urban areas. In the past decades in China, studies of PM mainly focused on the total suspended particulate (TSP), and few were related to $\mathrm{PM}_{2.5}$ (Winchester et al., 1981; Winchester and Bi, 1984; Chen et al., 1994; Wei et al., 1999). Some latest studies such as Hu et al. (2002), Cao et al. (2003) and Wang et al. (2002a,b) mainly focused on the chemical components of $\mathrm{PM}_{2.5}$ covering a short session at limited areas. He et al. (2001) and Ye et al. (2003) carried out the more comprehensive studies on $\operatorname{PM}_{2.5}$ in Beijing and Shanghai, respectively, of which intensive and continuous sampling were involved in a long period of 1999–2000. It was reported that in Beijing, the average concentration of $\mathrm{PM}_{2.5}$ exceeded $100~\mathrm{\sf~\sf~}{\sf u g}$ $\mathfrak{m}^{-3}$ and the highest value appeared in winter. Organic carbon (OC) was the most abundant species accounting for about $30\%$ of $\mathrm{PM}_{2.5}$ concentration. In Shanghai, the $\operatorname{PM}_{2.5}$ average concentration was about $60~\upmu\mathrm{g}\,\textrm{m}^{-3}$ , which was relatively lower than that in Beijing.
As the capital city of China, Beijing has faced with the serious problems of air quality raised by the emission of increased energy consumption and vehicle quantity. In 2002, the total consumed energy has increased about $17\%$ compared with that in 1999, and about 192, 000 tons of $\mathrm{SO}_{2}$ was emitted. More than 1,500,000 vehicles were on operation. In order to improve the air quality in Beijing and host the Olympic Games successfully in 2008, a series of control measures have been implemented since the end of 1998. The energy source structure has been gradually changed by generalizing clean fuels and low-sulfur coal. To reduce the local dust emission, the construction activities are supervised by the government and vegetation coverage of naked ground has increased. Emission control measures for vehicle exhausts are also adopted such as implementing new emission standards and converting diesel buses to compressed natural gas ones. In 2002, constructions for 2008 Olympic Games have started and a 4-year project for environmental pollution control has been brought forward in Beijing. Under these circumstances, it is necessary and of great significance to carry out $\operatorname{PM}_{2.5}$ studies so as to monitor the air quality change in Beijing. This paper will report the $\mathrm{PM}_{2.5}$ concentrations level from August 2001 to September 2002 and investigate the characteristics of chemical compositions including water-soluble ions, carbonaceous species and inorganic elements.
2. Methods and materials
Low-flow rate samplers (LFS, Aerosol Dynamic Inc., Berkeley, CA) were deployed to collect $\operatorname{PM}_{2.5}$ samples at a downtown site Chegongzhuang (CGZ) and a semi-residential site Tsinghua University (THU) in Beijing from August 2001 to September 2002. The sampler configuration and sites location have been described in detail elsewhere (He et al., 2001). Briefly, LFS at each site includes three parallel sampling inlets for aerosol collection. $\operatorname{PM}_{2.5}$ samples from the Teflonmembrane filter (Gelman $47\ \mathrm{mm}$ diameter and $2~{\upmu\mathrm{m}}$ pore size) on one inlet were for mass concentrations and elemental measurements. $\mathrm{PM}_{2.5}$ samples from a Teflon-membrane filter and a Nylon filter (Gelman 47 mm diameter, $1\upmu\mathrm{m}$ pore size) on the second inlets were used to analyze water-soluble ions. $\mathrm{PM}_{2.5}$ samples from two $47\,\textrm{m m}$ quartz-fiber filters on the third inlets were used to OC and EC (elemental carbon) analysis. The.5 quartz filters were preheated at $900\ ^{\circ}\mathrm{C}$ for at least 3-h to remove the adsorbed organics before sampling. The sampling flow rate is $0.{\bar{4}}\ \mathrm{~l~}\ \mathrm{min}^{-1}$ with duration of 7 days. Elemental analysis by X-ray fluorescence (XRF) and EC/OC analysis by CHN elemental analyzer were performed at National Research Center for Environmental Analysis and Measurement (NRCEAM) that is affiliation institution of Chinese EPA. Inorganic water-soluble ions analysis by ion chromatography (IC) was performed at Beijing Normal University.
The process of filter extraction and ion measurements was similar to the report of Yao et al. (2002). The XRF analysis was performed referring to the standard methods of Chow and Watson (1994). To evaluate the analysis accuracy of $\operatorname{PM}_{2.5}$ samples, data from IC were compared with those from XRF for some elements such as sulfur, magnesium and potassium. The correlation efficient and slope were 0.94 and 0.77 for sulfur, 0.74 and 0.40 for magnesium, and 0.70 and 0.76 for potassium, respectively. The correlations of chloride and calcium were weak with $R^{2}$ less than 0.6, and the $\mathrm{Cl}^{-}/\mathrm{Cl}$ and $\mathrm{Ca}^{2+}/\mathrm{Ca}$ ratio of some samples were larger than 1. It was likely due to the contamination during filter extraction for IC analysis. After discarding those data, the correlation efficient increased, 0.76 and 0.82 for chloride and calcium, respectively. Aluminum by XRF was not acceptable because the concentration was about 10 times as high as reported previously (Chen et al., 1994). It was deduced that the aluminum alloyed sample saucer in the XRF instrument might bring contamination to the samples. Therefore, the Al data will not discussed in following sections.
The CHN analytical method in this study is the same as reported by Liu et al. (2002) and Duan et al. (2004). Briefly, OC and EC were determined by using a laboratory two-step thermal procedure. Carbon contents were obtained by means of a CHN elemental analyzer (MT-5 YANACO New Science Corporation, Japan). In this protocol, the carrier gas is always helium with $8\%$ oxygen. TC (total carbon) and OC are determined directly when the combustion furnace temperatures are set as $950\ \ ^{\circ}\mathrm{C}$ and $450\ \ ^{\circ}\mathrm{C}$ , respectively. EC content can be obtained by subtracting OC from TC. The low detection limit was $5~\upmu\mathrm{g}~\mathrm{C}$ . To evaluate the accuracy of this method, interlaboratory comparison was performed by measuring the same urban aerosol samples. Our TC data were in good agreement with those obtained from the twostage thermal method of Cachier et al. (1989) as reported by Duan et al. (2004). Our OC and EC data showed systematic discrepancy with higher OC $(1.40\pm0.13)$ and lower EC values $(0.70\pm0.12)$ on the average since the OC/EC split is laboratory operational. Other interlaboratory comparisons clearly point out this problem (Schmid et al., 2001) and underline that the method to reach the OC/EC absolute split point is not yet solved (http://ocs. fortlewis.edu/aerosols/ocec/presentations.htm).
3. Results and discussion
3.1. Ambient concentrations of $P M_{2.5}$ , $P M_{I O}$ and gas pollutants
The weekly variations of $\mathrm{PM}_{2.5}$ concentration are shown in Fig. 1, ranging from 44.5 to $192.6\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ at CGZ site and from 22.8 to $199.4~\upmu\mathrm{g}~\mathrm{m}^{-3}$ at THU site, respectively. The annual average $\mathrm{PM}_{2.5}$ concentrations were 96.5 and $106.9~\upmu\mathrm{g}~\mathrm{m}^{-3}$ for CGZ and THU site, respectively. During heating period (from November to March), $\mathrm{PM}_{2.5}$ concentrations were 122 and $138~\upmu\mathrm{g}$ $\mathfrak{m}^{-3}$ , respectively at CGZ and THU site, which were about $50\%$ and $46\%$ higher than those of non-heating period. Auto-monitoring data of gaseous pollutants $\mathrm{SO}_{2}$ , $\mathrm{NO}_{2}$ , and CO) and $\mathrm{PM}_{10}$ for the same sampling period in this work are available at CGZ site (Beijing EPA, http://www.bjepa.gov.cn), and their seasonal variation are also shown in Fig. 1 and Table 1.
Averaged $\mathrm{SO}_{2}$ , $\mathrm{NO}_{2}$ , CO and $\mathrm{PM}_{10}$ were 70.7, 78.1, 2166.9 and $154.5~\ \upmu\mathrm{g}\ \ \mathrm{m}^{-3}$ , respectively. Gaseous pollutants presented the similar seasonal variation to $\mathrm{PM}_{2.5}$ with high values in wintertime and low values in spring and summer. A combination of low level, persistent temperature inversions and increases in emissions related to heating in wintertime usually leads to the increases in ambient pollutant concentrations. It is interesting to note that $\mathrm{SO}_{2}$ in wintertime $(180~\upmu\mathrm{g}\,\\\\,\mathrm{m}^{-3})$ ) increased by 5 times of those in nonheating period, while $\mathrm{NO}_{2}$ and CO increased by $40\%$ and $60\%$ , respectively. It suggests that the much greater increase in ambient $\mathrm{SO}_{2}$ and $\mathrm{PM}_{2.5}$ compared to $\mathrm{NO}_{2}$ and CO may be ascribed to increased coal consumption for heating. The statistics data show that in 2001, the total consumed coal was about 26,750,000 tons that accounted for about two third of the total energy consumption, and approximately
$10\%$ was used to heating supply (Beijing EPA, http:// www.bjepa.gov.cn). $\mathrm{NO}_{2}$ and CO is generally considered to mobile source emissions related. Mobile source emissions are expected to increase somewhat in the winter due to cold start emissions.
It is notable that $\mathrm{PM_{10}}$ concentration shows different seasonal variation from that of $\operatorname{PM}_{2.5}$ , of which the maximum value $188.5~\upmu\mathrm{g}~\mathrm{m}^{-3})$ ) appears in spring. Spring in north China is generally of low relative humidity (RH) and strong wind (Fig. 2) resulting in frequently occurred dust weather. Dust storms originate from northwest area with vast arid and uncovered soil then transport to its downstream areas such as Beijing. Those dust particles contain crustal matters that exist as coarse mode (Winchester et al., 1981). In this work the low $\mathrm{PM}_{2.5}/\mathrm{PM}_{10}$ ratio (0.46) in spring indicates that coarse particle is the predominant form in Beijing. The averaged $\mathrm{PM}_{2.5}/$ $\mathrm{PM_{10}}$ ratio is 0.62, which is comparable to those of our previous work (He et al., 2001) and other report of China (Wei et al., 1999). However, the $\mathrm{PM}_{2.5}/$ $\mathrm{PM}_{10}$ ratio in winter is as high as 0.79 suggesting that the combination of increased emission for heating supply and the favorable meteorological conditions would result in the fine particles formation in winter.
Compared with the data of July 1999–September 2000 in Beijing (He et al., 2001), $\operatorname{PM}_{2.5}$ concentrations in this study obviously decreased though the sampling period is not matched entirely. As illustrated in Table 2, gaseous pollutants and $\mathrm{PM}_{10}$ also decreased compared with those in 1999, by $16\%$ , $14\%$ and $8\%$ for $\mathrm{SO}_{2}$ , CO and $\mathrm{PM}_{10}$ , respectively. It may be due to the air quality control measures adopted by the government since the end of 1998 as noted previously. TSP has no obvious variation compared with that in 1998, however, higher than that in 1999. It indicates that the evaluation and control of non-anthropogenic source such as dust storms should be taken into account by the government.
3.2. Chemical compositions
3.2.1. Inorganic elements
Average concentrations of 15 elements at CGZ and THU site are listed in Table 1, most of which are comparable to those related studies of Beijing (He et al., 2001; Chen et al., 1994). Elemental concentrations presented obvious seasonal variations. For Mg, Si, K, Ca, Ti and Fe, which are generally related to the crustal sources, the high concentrations appearing in spring are likely ascribed to the dust storm contribution. The non-crustal elements such as Cl, Ni, Cu, Zn, As and Pb, however, their high concentrations appeared in winter. S presented different seasonal variation from above elements with the high concentration in summer, which is mainly due to the secondary sulfate particles formation from the precursor gas $\mathrm{SO}_{2}$ by homogeneous or heterogeneous reaction in the atmosphere (Seinfeld and Pandis, 1998). There is strong correlation between concentrations of IC measured water-soluble sulfate and XRF measured sulfur with a correlation coefficient of 0.94 and a slope of 0.77. About $60\%$ at CGZ site and $55\%$ at THU site of the total sulfur are in the form of watersoluble sulfate.
In order to evaluate the contributions of crustal and non-crustal sources, enrichment factor (EF) analysis is performed at the both sites by the definition of
$$
E F={\frac{\left(C_{i}/C_{n}\right)_{\mathrm{Environment}}}{\left(C_{i}/C_{n}\right)_{\mathrm{Background}}}}
$$
where $C_{i}$ is the concentration of element $i$ , $C_{n}$ is the concentration of reference element. $\left(C_{i}/C_{n}\right)$ )Environment and $(C_{i}/C_{n})_{\mathrm{background}}$ is the concentration ratio of element $i$ to element $n$ in the aerosol sample and in the crust or soil, respectively. EF lower than 10 indicates that crustal matters are likely the predominant sources of element $i$ , while EF larger than 10 demonstrates that element $i$ is obviously enriched in aerosol and probably emitted from non-crustal source (Torfs and Van Grieken, 1997). The reference element $n$ should have the characteristics of abundance in crustal matters, with chemical stability and minor contribution of anthropogenic pollutions. Those elements such as Al, Si, Fe and Ti are usually adopted in previous studies (Winchester et al., 1981; Winchester and Bi, 1984; Gao et al., 2002; Chen et al., 1994; Wei et al., 1999). In this work, Ti is selected as the reference element because of its inert character. The enrichment factor analysis results are listed in Table 3. At CGZ site, the EF values of Na, $\mathrm{Mg}$ , Si, K, Ca, Ti and Fe are less than 10, indicating these elements are contributed dominantly by crustal sources. Ning et al. (1996) reported that crustal elements such as Si, K, Ca, Ti and Fe were mainly concentrated in coarse mode with the range of $4\mathrm{-}8~\upmu\mathrm{m}$ . On the other hand, the EF values of S, Cl, Cu, Zn, As, Se, Br and Pb are much larger than 10, indicating that the predominant contributions of non-crustal sources. These elements generally presented bimodal distribution and mainly existed in fine mode in Beijing (Ning et al., 1996). Studies showed that fossil fuel was the important source of those noncrustal elements in north china (Zhang et al., 1998; Shi et al., 2002). As mentioned above, in Beijing, fossil fuel remains wide consumption in industrial area because of the increased population and energy demand. Additionally, the semi-rural areas lacking of uniform heating supply are usually the habitat of nonlocal populations. Therefore, considerable amounts of coal or biomass fuel are consumed in the residential small boilers for cooking and heating. In 2001, about $11\%$ of the total coal was used in residential way. Studies showed that coal fly ash from residential boilers contained abundant S $89950~\upmu\mathrm{g}~\textrm{g}^{-1}\textrm{,}$ ), Pb $(6364~\upmu\mathrm{g}~\mathrm{g}^{-1})$ and $Z\mathrm{n}~(8559~\upmu\mathrm{g}~\mathrm{g}^{-1})$ because of the inefficient combustion and absence of dust removing equipments (Zhang et al., 1998). Mobile emission is expected another important source of non-crustal elements. The amount of vehicle has exceeded 1,500, 000 in 2002 in Beijing. It was reported that the $\mathrm{Pb}$ and $Z\mathfrak{n}$ contents in vehicle exhaust were 3125 $\upmu\mathrm{g}\,\mathrm{\boldmath~g~}^{-1}$ and $1480~\upmu\mathrm{g}~\mathrm{g}^{-1}$ , respectively (Zhang et al., 1998). Although non-leaded gasoline has been used since 1998 in Beijing, the $\mathrm{Pb}$ content in gasoline $(\le\!0.013\,\textrm{g}\,|^{-1})$ ) remains higher than those in other countries.
3.2.2. Inorganic water-soluble ions
The sum concentration of total inorganic watersoluble ions $\mathrm{(Cl^{-}}$ , $\mathrm{NO}_{3}^{-}$ , $\mathrm{SO}_{4}^{2-}$ , $\mathrm{Na}^{+}$ , $\mathrm{NH}_{4}^{+}$ , $\mathbf{K}^{+}$ , $\mathrm{Mg}^{2+}$ and $\mathrm{{Ca}}^{2+}$ ) accounted for about $28\%$ and $25\%$ of $\operatorname{PM}_{2.5}$ mass concentration at CGZ and THU site, respectively. $\mathrm{SO}_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ and $\mathrm{NH_{4}^{+}}$ are observed the predominant ions in Beijing (Yao et al., 2002; Duan et al., 2003). The annual averages of $\mathrm{SO}_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ and $\mathrm{NH_{4}^{+}}$ were 9.9, 6.9 and $5.8~\upmu\mathrm{g}\:\textrm{m}^{-3}$ , respectively at CGZ site, and accounting for about $37\%$ , $23\%$ and $20\%$ of the total mass of inorganic water-soluble ions. The sum of $\mathrm{SO}_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ and $\mathrm{NH}_{4}^{+}$ accounts for about $23\%$ and $21\%$ of $\mathrm{PM}_{2.5}$ at CGZ and THU site, respectively.
$\mathrm{\bar{SO}}_{4}^{2-}$ concentrations in fall and summer were higher than that in winter by $45\%$ . As illustrated in Fig. 3, there is poor correlation between $\mathrm{SO}_{4}^{2-}$ and $\mathrm{SO}_{2}$ mass concentration, which is consistent with previous report (Hu et al., 2002). However, it is obvious that the data can be divided into two groups. Data of one group are close to the $\mathrm{SO}_{4}^{2-}$ axis and with low $\mathrm{SO}_{2}$ concentration $(<50\;\upmu\mathrm{g}\;\mathrm{m}^{-3})$ ; data of another group are close to the $\mathrm{SO}_{2}$ axis, including most of the winter samples. Relationship between $\mathrm{SO}_{4}^{2-}$ and $\mathrm{SO}_{2}$ can be used to investigate the sulfate formation or source. A large $\mathrm{SO}_{4}^{2-}/\mathrm{SO}_{2}$ mole ratio probably suggests the aqueous phase oxidation such as incloud processes of sulfate formation; contrarily, a lower ratio implies that the sulfate might be generated by heterogeneous phase oxidation (Yao et al., 2003). In this study, $\mathrm{SO}_{4}^{2-}/(\mathrm{SO}_{4}^{2-}\!+\!\mathrm{SO}_{2})$ mass ratio presented significant seasonal variation with 0.35, 0.27, 0.16 and 0.05 for summer, fall, spring and winter, respectively. It indicates that low temperature plays an important role and might be not favorable for secondary sulfate formation. Therefore, gas-particle conversion and furthermore condensation or adsorption on the particle surface is likely the major source of secondary sulfate particles in Beijing (Yao et al., 2002; Wang et al., 2002a,b). $\mathrm{NO}_{3}^{-}$ concentration in winter was higher than in summer, which is consistent with what observed in Qingdao city, Japan and Korea (Hu et al., 2002; Wakamatsu et al., 1996). Sampling artifact is not expected the major factor since a denuder and a back nylon filter were used in this work. The gas-particle partition of ambient $\mathrm{NO}_{3}^{-}$ strongly depends on the meteorological conditions such as temperature and RH. As shown in Fig. 2, the temperature and RH decrease from above $30~^{\circ}\mathrm{C}$ and $80\%$ , respectively in summer to approximately $^{-10}$ $^{\circ}\mathrm{C}$ and $25\%$ , respectively in winter. A combination of seasonal variations of those meteorology is expected to results in a large variation of $\mathrm{NO}_{3}^{-}$ in $\mathrm{PM}_{2.5}$ .
Average concentrations of $\mathrm{Na}^{+}$ , $\mathrm{Mg}^{2+}$ and $\mathrm{{Ca}}^{2+}$ at both sites were similar, 0.6, 0.15 and $1.1\;\;\upmu\mathrm{g}\;\;\mathrm{m}^{-3}$ , respectively. As illustrated in Table 4, ${\mathrm{Mg}}^{2+}$ is correlated well with $\mathrm{Na}^{+}$ and $\mathrm{{Ca}}^{2+}$ , with $R^{2}$ of 0.96 and 0.78, respectively. It indicates that those three ions have the same source. Sea salt emission is usually regarded as the major contribution to $\mathrm{Na}^{+}$ , $\mathrm{Mg}^{2+}$ and $\bar{\mathrm{Ca}^{2^{+}}}$ at coastal area. However, there is a distance of about $200\,\mathrm{km}$ . between the sampling sites in this work and the sea, the contribution of sea salt is considered inappreciable. On the other hand, soil and dust particles in north China are generally observed to contribute to those ions (Wang et al., 2002a,b). $\mathsf{K}^{+}$ mass concentration presented modest seasonal variation at both sites; however, peak concentration was observed in June, $2.4~\upmu\mathrm{g}\:\textrm{m}^{-3}$ that was about $42\%$ higher than the average concentration $1.7\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ . It is probably due to the agriculture field vegetation burning during harvest season in north China (Duan et al., 2004).
As illustrated in Table 4, there is strong linear correlation between $\mathrm{NO}_{3}^{-}$ , $\mathrm{SO}_{4}^{2-}$ and $\mathrm{NH_{4}^{+}}$ in equivalent concentrations, with correlation coefficient $R^{2}$ of 0.85 and 0.86. The average equivalent ratios of $(\mathrm{NO}_{3}^{-}\!+\!\mathrm{SO}_{4}^{2-})/\mathrm{NH}_{4}^{+}$ at both sites were close to1, i.e. 1.06 and 1.18 at CGZ (Fig. 4a) and THU site, respectively. Considering the probably sampling artifacts of $\mathrm{NH_{4}^{+}}$ because of ammonium evaporation especially in summer, it is considered that ammonia can neutralize particulate nitrate and sulfate, therefore, $\mathrm{NH}_{4}\mathrm{NO}_{3}$ and $(\mathrm{NH}_{4})_{2}\mathrm{SO}_{4}$ are the major compositions in $\operatorname{PM}_{2.5}$ . From the comparison of equivalent values between total inorganic anions $(\Sigma\mathrm{Cl}^{-}\!+\!\mathrm{NO}_{3}^{-}\!+\!\mathrm{SO}_{4}^{2-})$ and total inorganic cations $(\mathrm{{\Sigma}}\mathrm{{Na}^{+}}\mathrm{{+NH_{4}^{+}}\mathrm{{+K}^{+}}\mathrm{{+Mg}^{2+}}\mathrm{{+Ca}^{2+}})}$ (Fig. 4b), it is found that for most of the samples, cation-to-anion ratios are higher than unity at both sites, 1.34 at CGZ and 1.38 at THU site. The anion deficiency is probably due to the undetected carbonate ion $\mathrm{(HCO}_{3}^{-}$ , $\mathrm{CO}_{3}^{-}$ ) and organic water-soluble ions (Yao et al., 2002).
3.2.3. Carbonaceous particles
Carbonaceous particles are observed the most abundant components of $\mathrm{PM}_{2.5}$ in this work. OC and EC concentrations averaged $22.9\,\upmu\mathrm{g}\,\mathrm{C}\,\mathrm{m}^{-3}$ and $10.2\,\upmu\mathrm{g}$ $\mathrm{~C~m~}^{-3}$ , respectively at CGZ site, and $28.8~\upmu\mathrm{g}\mathrm{~C~}\mathrm{m}^{-3}$ and $9.6~\upmu\mathrm{g}\mathrm{~C~m}^{-3}$ , respectively, at THU site. The OC concentrations at the both sites were comparable to those measured during 1999–2000 (He et al., 2001), however higher than those of $\mathrm{PM}_{2.5}$ in other areas, for instance, $15.4~\upmu\mathrm{g}\,\textrm{C}\,\mathrm{m}^{-3}$ and $14.3~\upmu\mathrm{g}\,\mathrm{~C~}\,\mathrm{m}^{-3}$ in
Shanghai (Ye et al., 2003), $10.4~\upmu\mathrm{g}\,\mathrm{~C~}\,\mathrm{m}^{-3}$ in Kaohsiung, Taiwan (Lin and Tai, 2001), $10.0~\upmu\mathrm{g}\,\mathrm{~C~}$ $\mathrm{m}^{-3}$ in Seoul (Kim et al., 1999) and $9.8~\upmu\mathrm{g}\mathrm{~C~m}^{-3}$ in Sihwa, Korea (Park et al., 2001), $3.0\ \upmu\mathrm{g}\mathrm{~C~m}^{-3}$ in Helsinki, Finland (Viidanoja et al., 2002). OC concentration presented strong seasonal variation as shown in Fig. 5 and Table 1. At CGZ site, the OC concentration in winter was the maximum $(32.2\ \upmu\mathrm{g}\ C\ \mathrm{m}^{-3})$ , then decreased through fall $(23.2\ \upmu\mathrm{g}\textrm{C m}^{-3})$ ) and spring $(19.6\,\mathrm{\upmug}\,\mathrm{\textrm{C}}\,\mathrm{\mathrm{m}}^{-\bar{3}})$ ), and dropped to the minimum in summer $(14.9\,\upmu\mathrm{g}\,\mathrm{C}\,\mathrm{m}^{-3})$ ). This trend was also observed in our previous work in Beijing (He et al., 2001). The data at THU site showed the similar variation, while OC concentrations in winter and fall were higher (38.7 and $29.3~\upmu\mathrm{g}\mathrm{\,C\,m}^{-3}$ , respectively) compared to those of CGZ site. EC concentration in winter was higher than those of other three seasons; however, the difference was not apparent. Average OC/EC ratios presented similar variations at both sites, higher in winter and fall than in spring and summer. These results were probably contributed to the combination of meteorological conditions and increases of local emission for heating. It was reported that residential coal burning could emit considerable carbonaceous particles (Bond et al., 2002). As discussed above, coal remains the major energy source in Beijing and is consumed in many fields such as power plants, industrial and residential boiler. Many pollutants are likely related to coal burning since their concentrations increased during heating period. For the period from September to early November in Beijing, the meteorological conditions are generally stable with strong sunlight and less precipitation. It is favorable to the secondary organic aerosols (SOA) formation. On the other hand, with the temperature decreased from late fall to winter, the semivolatile organic compounds (SVOCs) apt to coagulate in the particulate phase and will be collected during filters sampling. The strong wind in spring is not favorable for the SOA formation and high temperature in summer usually leads to the gas-particle equilibrium shift of SVOCs, and accordingly resulting in low OC concentrations in these seasons.
3.3. $P M_{2.5}$ mass balance
Fig. 6 illustrates the $\operatorname{PM}_{2.5}$ mass balance for the averaged carbonaceous species, three major watersoluble ions, crustal matters and trace species at the both sites. The method to calculate the mass balance has been described in detail elsewhere (He et al., 2001). In briefly, organics is obtained by multiplying OC by a factor of 1.4, which is commonly used to estimate the hydrogen and oxygen in organic compounds. Crustal matters are the sum of sodium, manganese, silicon, potassium, calcium, titanium and iron oxides. Except above elements and S, Cl, sum of all other elements is defined as trace species, including Ni, Cu, Zn, As, Se, Br, Sr, Ag, I, Ba and Pb.
Those elements with concentrations below the detection limits of XRF method are not considered.
Based on the annual average concentrations, the $\operatorname{PM}_{2.5}$ chemical composition at the two sites is very similar (Fig. 6a). Approximately $83.1\%$ and $84.3\%$ of the $\mathrm{PM}_{2.5}$ mass concentrations were explained by seven components at CGZ and THU site, respectively. Of all the species, organic matters are the most predominant fraction accounting for about $33.6\%$ and $37.8\%$ , respectively at the two sites. The higher OC contribution at THU site indicated the impact of residential cooking and heating on $\operatorname{PM}_{2.5}$ . EC accounted for about $11.6\%$ and $9.8\%$ of $\operatorname{PM}_{2.5}$ at CGZ and THU site, respectively. Three major secondary ionic species, i.e., nitrate, sulfate and ammonium contributed $6.5\%$ , $10.4\%$ and $5.5\%$ , respectively at CGZ site, and $7.0\%$ , $10.2\%$ and $5.1\%$ , respectively at THU site. Crustal matter fractions were notable at sites, $14.5\%$ and $13.5\%$ , respectively. A comparison is performed to the mass balance results in previous work in Beijing (He et al., 2001). The average contributions of carbonaceous species increased obviously, which were likely ascribed to both the decreasing of $\operatorname{PM}_{2.5}$ mass concentrations at both sites and the different analytical method (EC especially). The increased contributions of sulfate and nitrate were probably ascribed to the decrease of gaseous pollutants emission from anthropogenic activities as discussed above. The frequent dust storm was expected to explain the high crustal matters contribution in spring.
The $\mathrm{PM}_{2.5}$ mass balance results show strong seasonal variation as can be observed in Fig. 6b. In fall and spring, about $95\%$ at CGZ site and $90\%$ at THU site of $\mathrm{PM}_{2.5}$ mass concentrations were explained by the seven chemical compositions described above. However, in winter and summer the unidentified fractions were considerable, about $25\%$ and $20\%$ at CGZ and THU site, respectively. The high temperature in summer is expected to be responsible for the large unidentified percentage since sampling artifacts are inevitable. For the volatile inorganic species such as nitrate and chloride, a backup Nylon filter was used to absorb the evaporated $\mathrm{HNO}_{3}$ and HCl decomposed from the particles on Teflon filter (Yao et al., 2002). However, for ammonium and volatile organic species, there was no such method to avoid the sampling artifacts. Additionally, the long sampling period (1 week) might somewhat lead to the negative effect on the $\mathrm{PM}_{2.5}$ mass concentrations. It seems that temperature is not the major factor to the unidentified $\mathrm{PM}_{2.5}$ fraction in winter since it is generally below zero. As noted above, the elements Cl and S were related to coal consumption, and approximately $64\%$ of elements S existed as water-soluble sulfate. That was, about $36\%$ of S existed as non-sulfate. However, Cl and the nonsulfate S were not included by the seven chemical compositions when calculating $\mathrm{PM}_{2.5}$ mass balance. We noted that the sum of Cl and non-sulfate S was predominant in winter and averagely accounted for about $5\%$ of $\mathrm{PM}_{2.5}$ mass concentration at the two sites.
If this fraction were added, the explained $\operatorname{PM}_{2.5}$ in winter would be improved. However, it needs further work to investigate the other unidentified fraction and to explain the difference between the two sites as well.
Acknowledgements
This study was funded by General Motors. We would like to thank Beijing Environmental Protection Bureau (http://www.bjepa.gov.cn) for providing us the gaseous pollutants information.
References
Appel BR, Tokiwa Y, Hsu J, Kothny EL, Hahn E. Visibility as related to atmospheric aerosol constituents. Atmos Environ 1985;9:1525–9.
Baik NJ, Kim YP, Moon KC. Visibility study in Seoul, 1993. Atmos Environ 1996;30:2319–28.
Bond T, Quinn P, Bates T. Particulate emissions from residential combustion of coal from the US and China. The Sixth International Aerosol Conference, Taipei, Taiwan on September 9–13; 2002.
Cachier H, Bremond MP, Buat-Menard P. Determination of atmospheric soot carbon with a simple thermal method. Tellus 1989;41B:379–90.
Cao JJ, Lee SC, Ho KF, Zhang XY, Zou SC, Fung K, et al. Characteristics of carbonaceous aerosol in Pearl River Delta Region, China during 2001 winter period. Atmos Environ 2003; 37:1451–60.
Chan YC, Simpson RW, Mctainsh GH, Vowles PD, Cohen DD, Bailey GM. Source apportionment of visibility degradation problems in Brisbane (Australia) using the multiple linear regression techniques. Atmos Environ 1999;33: 3237–50.
Charlson RJ, Langner J, Rodhe H, Leovy CB, Warren SG. Pertubation of the northern hemisphere radiative balance by backscattering from anthropogenic sulfate aerosols. Tellus 1991; 43AB:152–63.
Chen Z, Ge S, Zhang J. Measurement and analysis for atmospheric aerosol particulates in Beijing. Res Environ Sci 1994;7(3):1–9 [in Chinese, with abstract in English].
Chow JC, Watson JG. Guidelines for $\mathrm{PM_{10}}$ sampling and analysis applicable to receptor modeling. Washington, DC7 US Environmental Protection Agency, Office of Air Quality Planning and Standards; 1994. EPA-452/R-94-009.
Dockery DW, Pope CA. Acute respiratory effects of particulate air pollution. Annu Rev Public Health 1994;15:107–32.
Duan FK, Liu XD, He KB, Lu YQ, Wang L. Atmospheric aerosol concentration level and chemical characteristic of its watersoluble ionic species in wintertime in Beijing, China. J Environ Monit 2003;5:569–73.
Duan FK, Liu XD, Dong SP, Lu YQ, Yu T, Cachier H. Identification and estimate of biomass burning contribution to the urban aerosol organic carbon concentration in Beijing. Atmos Environ 2004;38:1275–82.
Gao Y, Nelson ED, Field MP, Ding Q, Li H, Sherrell RM, et al. Characterization of atmospheric trace elements on $\mathrm{PM}_{25}$ particulate matter over the New York–New Jersey harbor estuary. Atmos Environ 2002;36:1077–86.
He K, Yang F, Ma Y, Zhang Q, Yao X, Chan CK, et al. The characteristics of $\mathrm{PM}_{25}$ in Beijing, China. Atmos Environ 2001; 35:4959–70.
Hu M, He LY, Zhang YH, Wang M, Kim YP, Moon KC. Seasonal variation of ionic species in fine particles at Qingdao, China. Atmos Environ 2002;36:5853–9.
Kim YP, Moon KC, Lee J, Baik NJ. Concentrations of carbonaceous species in particles at Seoul and Cheju in Korea. Atmos Environ 1999;33:2751–8.
Lin JJ, Tai HS. Concentrations and distributions of carbonaceous species in ambient particles in Kaohsiung City, Taiwan. Atmos Environ 2001;35:2627–36.
Lipfert FW, Wyzga RE. Air pollution and mortality: issues and uncertainties. J Air Waste Manag Assoc 1995;45:949–66.
Liu XD, Chi XG, Duan FK, Dong SP, Yu T. Determination of organic carbon and elemental carbon in Chinese urban aerosols by using CHN elemental analyzer. J Aerosol Sci 2002; 31(Suppl. 1):s240–1.
Ning DT, Zhong LX, Chung YS. Aerosol size distribution and elemental composition in urban areas of northern China. Atmos Environ 1996;30:2355–62.
Novakov T, Penner JE. Large contribution of organic aerosols to cloud-condensation nuclei concentrations. Nature 1993;365 323–65.
Pagano P, de Zaiacomo T, Scarcella E, Bruni S, Calamosca M. Mutagenic activity of total and particle-sized fraction of urban particulate matter. Environ Sci Technol 1998;30:3512–6.
Park SS, Kim YJ, Fung K. Characteristics of $\mathrm{PM}_{25}$ carbonaceous aerosol in the Sihwa industrial area, South Korea. Atmos Environ 2001;35:657–65.
Schmid H, Laskus L, J.urgen AH, Baltensperger U, Lavanchy V, Bizjak M, et al. Results of the bcarbon conferenceQ international aerosol carbon round robin test stage I. Atmos Environ 2001;35 2111–21.
Schwartz J. Particulate air pollution and chronic respiratory disease. Environ Res 1993;62:7–13.
Seinfeld JH, Pandis SN. Atmospheric chemistry and physics: from air pollution to climate change. John Wiley & Sons, 1998, New York.
Shi ZB, Shao LY, Li H. Characterization of the $\mathrm{PM}_{10}$ in ambient air of Northwestern Beijing urban area during heating period. Environ Sci 2002;23:30–4 [in Chinese with abstract in English].
Torfs K, Van Grieken R. Chemical relations between atmospheric aerosols, deposition and stone decay layers on historic buildings at the Mediterranean coast. Atmos Environ 1997; 31:179–2192.
Viidanoja J, Sillanpaa M, Laakia J, Kerminen V, Hillamo R, Aarnio P, et al. Organic and black carbon in $\mathrm{PM}_{25}$ and $\mathrm{PM}_{10}$ : 1 year of data from an urban site in Helsinki, Finland. Atmos Environ 2002;36:3183–93.
Wakamatsu S, Utsunnomiya A, Han JS. Seasonal variation in atmospheric aerosols concentration covering northern Kyushu, Japan and Seoul, Korea. Atmos Environ 1996;30:2343–54.
Wang G, Huang L, Gao S, Gao S, Wang L. Characterization of water-soluble species of $\mathrm{PM_{10}}$ and $\mathrm{PM}_{25}$ aerosols in urban area in Nanjing, China. Atmos Environ 2002;36:1299–307.
Wang W, Yue X, Liu HJ, Pan Z, Tang DG, Wang Y, et al. Study on pollution characteristics of aerosol during sand dust storm weather in Beijing. Acta Sci Circumstantiae 2002;22:494–8 [in Chinese with abstract in English].
Wei F, Teng E, Wu G, Hu W, Wilson WE, Chanpman RS, et al. Ambient concentrations and elemental compositions of $\mathrm{PM}_{10}$ and $\mathrm{PM}_{25}$ in four Chinese cities. Environ Sci Technol 1999; 33:4188–93.
Wexler AS, Ge Z. Hydrophobic particles can activate at lower relative humidity than slightly hygroscopic ones: a Kohler theory incorporating surface fixed charge. J Geophy Res 1998; 103:6083–8.
Winchester JW, Bi M. Fine and course aerosol composition in an urban setting: a case study in Beijing, China. Atmos Environ 1984;18:1399–409.
Winchester JW, Lu W, Ren L, Wang M, Maenhaut W. Fine and coarse aerosol composition from a rural area in north China. Atmos Environ 1981;15:933–7.
Yao XH, Chan CK, Fang M, Cadle SC, Chan TL, Mulawa PA, et al. The water-soluble ionic composition of $\mathrm{PM}_{25}$ in Shanghai and Beijing, China. Atmos Environ 2002;36:4223–34.
Yao XH, Lau APS, Fang M, Chan CK, Hu M. Size distributions and formation of ionic species in atmospheric particulate pollutants in Beijing, China: 1-inorganic ions. Atmos Environ 2003;37: 2991–3000.
Ye B, Ji X, Yang H, Yao X, Chan CK, Cadle S, et al. Concentration and chemical composition of $\mathrm{PM}_{25}$ in Shanghai for a 1 yr period. Atmos Environ 2003;37:499–510.
Zhang J, Chen ZL, Wang W. Source apportionment on fine particulates in atmosphere in Beijing. Acta Sci Circumstantiae 1998;18:62–7 [in Chinese, with abstract in English].
|
atmosphere
|
0014
|
10.1016/j.atmosres.2017.01.021
|
image
|
Fig. 1. Locations of ambient sampling sites in Haikou
|
|
atmosphere
|
0014
|
10.1016/j.atmosres.2017.01.021
|
table
|
Table 1 Describes of ambient $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ sampling sites
|
|
atmosphere
|
0014
|
10.1016/j.atmosres.2017.01.021
|
image
|
Fig. 3. Chemical compositions of ambient $\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5}$ in Haikou
|
|
atmosphere
|
0014
|
10.1016/j.atmosres.2017.01.021
|
image
|
Fig. 4. Linear regression of EC-OC data in $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ samples in Haikou
|
|
atmosphere
|
0014
|
10.1016/j.atmosres.2017.01.021
|
image
|
Fig. 5. Profiles of six source categories for particulate matter in Haikou. ammonium sulfate; f: ammonium nitrate; g: sea salt)
|
|
atmosphere
|
0014
|
10.1016/j.atmosres.2017.01.021
|
table
|
Table 2 Average amount of SOC deducting in $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$
|
|
atmosphere
|
0014
|
10.1016/j.atmosres.2017.01.021
|
image
|
Fig. 6. Source contribution to ambient $\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5}$ in urban areas of Haikou
|
|
atmosphere
|
0014
|
10.1016/j.atmosres.2017.01.021
|
image
|
Fig. 7. 72-h air parcel backward trajectories (every $^{6\,\mathrm{h}}$ in sampling days) for winter (a) and spring (b) in Haikou. The percentage of trajectories belonging to a particular cluster is shown in parenthesis
|
|
atmosphere
|
0014
|
10.1016/j.atmosres.2017.01.021
|
text
| Not supported with pagination yet |
Accepted Manuscript
Source apportionment of ambient PM10 and PM2.5 in Haikou, China
Xiaozhen Fang, Xiaohui Bi, Hong Xu, Jianhui Wu, Yufen Zhang, Yinchang Feng
PII: S0169-8095(16)30293-9
DOI: doi: 10.1016/j.atmosres.2017.01.021
Reference: ATMOS 3878
To appear in: Atmospheric Research
Received date: 30 August 2016
Revised date: 12 January 2017
Accepted date: 18 January 2017
Please cite this article as: Xiaozhen Fang, Xiaohui Bi, Hong Xu, Jianhui Wu, Yufen Zhang, Yinchang Feng , Source apportionment of ambient PM10 and PM2.5 in Haikou, China. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Atmos(2016), doi: 10.1016/j.atmosres.2017.01.021
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Source apportionment of ambient $\mathbf{PM_{10}}$ and $\mathbf{PM}_{2.5}$ in Haikou, China
Xiaozhen Fang , Xiaohui Bi , Hong $\mathrm{{Xu}^{a}}$ , Jianhui $\mathrm{Wu}^{\mathrm{~a~}}$ , Yufen Zhang, Yinchang
Feng
State Environmental Protection Key Laboratory of Urban Ambient Air Particulate Matter Pollution Prevention and Control, College of Environmental Science and Engineering, Nankai University, No.38 Tongyan Road, Tianjin 300350, China
Highlights
$\bullet$ The characteristics of ambient PM in Haikou were investigated.
$\bullet$ CMB model was used for source apportionment.
$\bullet$ Particulate matter was more affected by regional sources in winter than in spring through back trajectory analysis.
Abstract In order to identify the sources of $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ in Haikou, 60 ambient air samples were collected in winter and spring, respectively. Fifteen elements (Na, Mg, Al, Si, K, Ca, Ti, V, Cr, Mn, Fe, Ni, Cu, Zn and Pb), water-soluble ions $\mathrm{{SO}}_{4}^{2-}$ and $\mathrm{NO}_{3}^{\mathrm{~-}}$ , and organic carbon (OC) and elemental carbon (EC) were analyzed. It was clear that the concentration of particulate matter was higher in winter than in spring. The value of $\mathrm{PM}_{2.5}/\mathrm{PM}_{10}$ was greater than 0.6. Moreover, the proportions of TC, ions, Na, Al, Si and Ca were more high in $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ . The SOC concentration was estimated by the minimum OC/EC ratio method, and deducted from particulate matter compositions when running CMB model. According to the results of CMB model, the resuspended dust $(17.5–35.0\%)$ , vehicle exhaust $\left.14.9{-}23.6\%\right)$ and secondary particulates $(20.4–28.8\%)$ were the major source categories of ambient particulate matter. Additionally, sea salt also had partial contribution $(3\!-\!8\%)$ . And back trajectory analysis results showed that particulate matter was greatly affected by regional sources in winter, while less affected in spring. So particulate matter was not only affected by local sources, but also affected by sea salt and regional sources in coastal cities. Further research could focuses on establishing the actual secondary particles profiles and identifying the local and regional sources of PM at once by one model or analysis method.
Keywords $\mathrm{PM}_{10}$ and $\mathbf{PM}_{2.5};$ chemical composition; source apportionment; back trajectory analysis.
1 Introduction
With the frequent occurrence of haze, more and more people have paid close attention to atmospheric particulate matter (PM). Furthermore, PM was regarded as one of the most important environmental factors affecting the heart and lung disease and lung cancer (Dockery et al., 1993; Pope et al., 1995; Pope, 2000; Forastiere, 2004; Zhang et al., 2014; WHO, 2009; US.EPA, 2004; Pope et al., 2006; Cohen et al., 2005; Kesavachandran et al., 2012). And it has significant influence on global climate and visibility (Appel et al., 1985; Baik et al., 1996; Law et al., 2007). It has been well recognized that atmospheric particulate matter (PM) is the primary pollutant of air pollution as a result of morphological, physical, chemical, thermodynamic and toxicity properties in China. When the annual average concentration of $\mathrm{PM}_{2.5}$ increased $10\;\upmu\mathrm{g}/\mathrm{m}^{3}$ , the cardiovascular diseases and lung cancer mortality increased $6\%$ and $8\%$ , respectively (Pope et al., 1995). Likewise, when the annual average concentration of $\mathrm{{PM}_{10}}$ increased $10~\upmu\mathrm{g/m}^{3}$ , the total risk of death, death from cardiovascular and respiratory diseases of residents increased $0.35\%$ , $0.44\%$ and $0.56\%$ , respectively (Chen et al., 2012). Besides, long-term exposure to $\mathrm{PM}_{10}$ could increase cardiovascular diseases mortality (Zhang et al., 2014).
Nowadays, the characteristics of particulate matter have been widely studied, such as American (Chow et al., 1996), Spain (Querol et al., 2001; Querol et al., 2004; Rodrı́guez et al., 2004), Switzerland (Hueglin et al., 2005; Gehrig et al., 2007), Italy (Ariola et al., 2006; Marcazzan et al., 2001; Marcazzan et al., 2003; Squizzato et al., 2016), India (Guttikunda et al., 2013; Yadav et al., 2013; Kesavachandran et al., 2012; Tiwari et al., 2016) and China (He et al., 2001; Wang et al., 2006; Wei et al., 1999; Streets et al., 2007; Yang et al., 2002; Gao et al., 2016). And many scholars have study source apportionment of PM (Watson et al., 1979; Harrison et al., 1997; Song et al., 2006; Watson et al., 2008; Guo et al., 2009; Wang et al., 2014; Hu et al., 2015; Hsu et al., 2016). Over the years, Chemical Mass Balance (CMB) model had developed rapidly. And CMB (Cooper and Watson, 1980) was the most widely used to identify the source category and estimate the contributions (Watson et al., 1990; Feng et al., 2005; Bi et al., 2007; Wu et al., $2009\mathrm{a}$ ; Han et al., 2011), which needs sources and receptor profiles. But more attentions were paid on continental cities in China, such as Beijing (Wang et al., 2005), Kaifeng (Wu et al., 2009a) and Wuxi (Han et al., 2011). And the local sources were often considered in source apportionment of PM, while regional sources were relatively considered less. However, little work has been done on the characteristics and origin of coast cities in China.
This study aims to focus on coastal cities– Haikou city. The chemical characteristics and source apportionment of $\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5}$ were discussed in Haikou. Due to Haikou is adjacent to Qiongzhou Strait, which contributes a certain amount of sea salt particles every day, so sea salt particles should be assessed in the study. Secondary particles also are the one of main source categories in Haikou, and its influence shouldn’t be overlook. Therefore, the contributions of six source categories were considered to identify the origin of $\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5}$ . The six source categories included resuspended dust, coal combustion dust, construction and cement dust, vehicle exhaust, secondary particles and sea salt particles. S
2 Material and methods
2.1 Study area
Haikou $\mathrm{(110^{\circ}10^{\prime}{-}110^{\circ}41^{\prime}E}$ , $19^{\circ}32^{\prime}–20^{\circ}05^{\prime}\mathrm{N})$ is situated in the north of Hainan Island (Fig.1) with over 2.05 million people distributed $2304~\mathrm{km}^{2}$ . It is located in the northern margin of the low-latitude tropics. It is a tropical monsoon climate zone. It has a maritime climate and obvious seasonal variations. According to Haikou meteorological observation historical data (from 2006 to 2010 year), the annual average rainfall was $1664\;\mathrm{mm}$ , the annual average humidity was $85\%$ , and the average annual sunshine hours were over than 2000 h. SSE and NE were dominant winds direction in spring and winter, respectively. The average annual mean wind speed was $3.8~\mathrm{m/s}$ . And the annual average temperature was $23.8^{\circ}\mathrm{C}$ , the highest was about $28^{\circ}\mathbf{C}$ , the lowest was about $18{}^{\circ}\mathbf{C}$ .
According to a census of pollution sources in 2007 in Haikou, there were no large-scale industrial sources except Huaneng Haikou Power Plant in Haikou, which was $38~\mathrm{km}$ from the city and not in the dominant wind direction. Besides, emission from various sources did not show significant seasonal differences. Due to the development of clean energy over the past decade, the energy structure of Haikou had changed. It was mainly with petroleum products and electricity in the energy structure during the “Eleventh Five-Year”, followed by natural gas and other fuels. And it had no heating season. Coal consumption was the lowest with a downward trend. According to local statistical yearbooks for 2010 (HPBS, 2011), the urban gasification rate reached $98\%$ , and the clean energy utilization rate increased to $99.2\%$ . However, there were more than 0.41 million registered vehicles in the city as of March, 2012, which indicated that vehicle exhaust may be one of the main sources of particulate matter.
2.2 Air sampling
Haikou has two typical atmospheric matter pollution seasons (winter and spring). In order to identify the sources of particulate matter and estimate their contributions to the ambient PM concentrations, 60 samples were collected at the State Air Quality Monitoring Sites per season. The sampling time in winter was from December 26, 2011 to January 3, 2012, while in spring it was April 17 to 26, 2012. It covered all days of the week to give a fair representation of ambient $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ samples for each season (Table 1).
Ambient samples were collected simultaneously at a flow rate of $100\;\mathrm{L/min}$ in a $22{\sim}23\mathrm{~h~}$ period per day using two $\mathrm{{PM}_{10}}$ samplers and two $\mathrm{PM}_{2.5}$ samplers (Wuhan Tianhong Intelligence Instrumentation Facility, TH-150C Medium Volume Sampler). The sampling height was $5{\sim}20\mathrm{~m~}$ above the ground at the stations. Due to funding and labor constrains, both $\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5}$ samples were collected at Hainan University (HU) and Hainan Normal University (HN), while only $\mathrm{PM}_{10}$ samples were collected in Long Hua (LH) and Xiu Ying (XY). Two types of membrane were used in parallel to collect airborne particles. One was quartz-fiber membrane $90\;\mathrm{mm}$ in diameter, type 2500QAT-UP, Pall Life Sciences, USA) for subsequent water-soluble ion and carbon component analysis, the other was polypropylene-fiber membrane ( $90\;\mathrm{mm}$ in diameter, Beijing Synthetic Fiber Research Institute, China) for element species analysis. The quartz-fiber membrane and polypropylene-fiber membrane were preheated at $450\ ^{\circ}\mathrm{C}$ for $^\mathrm{4h}$ and $65\ ^{\circ}\mathrm{C}$ for $2~\mathrm{h}$ respectively before use. Meanwhile, the membranes were put in a refrigerator to save no matter with before and after loaded. Further details have been previously described (Feng et al., 2005; Bi et al., 2007; Zhao et al., 2013).
Source sampling was collected on basis of the investigation of $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ and air pollution emission inventory reported in the official reports. In this study, only resuspended dust (RD) and coal combustion dust were collected.
Resuspended dust refers to an open source that is continually raised into the surrounding atmosphere by natural forces or human activity after being initially deposited in urban areas. It is a complex mixture. Resuspended dust was collected from east, south, west and upwind direction in urban areas by grab sampling, and other features $5{-}15\,\mathrm{~m~}$ above the ground without any clear nearby PM source.
Coal combustion dust samples were collected from an electric power plant using a dilution sampled system. The preprocessing and analysis of these samples were identical with previous studies (Bi et al., 2007; Han et al., 2011).
2.4 Sample analysis
In this study, all source samples and air samples were weighed by a sensitive microbalance (Metler M5, Switzerland). Gravimetric values of $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ were determined using quartz-fiber membrane. The two types of membrane were placed in a desiccator for $48\ \mathrm{h}$ before gravimetric analysis. Ion chromatography (DX-120, Dionex Ltd., USA) was applied to determine water-soluble ions $\mathrm{{(SO_{4}}^{2-}}$ and $\mathrm{NO}_{3}^{\mathrm{~-~}}$ ) after sample extraction with deionized water. The total carbon (TC, sum of organic carbon and elemental carbon), organic carbon (OC) and elemental carbon (EC) were determined using a carbon analyzer (DRI2001A, Desert Research Institute, USA). 15 elemental species (Na, Mg, Al, Si, K, Ca, Ti, V, Cr, Mn, Fe, Ni, Cu, Zn and $\mathbf{P}\mathbf{b}$ ) were analyzed by inductively coupled plasma analysis (ICP 9000 $(\boldsymbol{\mathbf{N}}+\boldsymbol{\mathbf{M}})$ , Thermo, USA). These experimental procedures were similar to those used in previous studies (Tyler et al., 1992; Chow et al., 1993; Baldwin et al., 1994; Chow et al., 2007; Cao et al., 2008; Han et al., 2011).
2.5 Quality assurance and quality control
All the instruments had been frequently and properly calibrated and maintained. The samplers were placed side by side and loaded with membranes. Then ambient $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ samples were collected simultaneously and weighed in the same conditions. The results were within $3.33\%{-7.49\%}$ . At each site, four field blank samples (two quartz-fiber membrane and two polypropylene-fiber membrane) were used to check the possible contamination from sample collection, transportation and storage of samples before each sampling period. Five laboratory blanks were also used for the same purpose. The laboratory blanks and field blanks were tested with no significant contamination found for any target objects. Parallel samples and duplicate measurements of samples were analyzed to test the precision of the sampling and analytical techniques, respectively. The mean relative standard deviations (RSD) for all the target analyses were less than $10\%$ .
2.6 Methodology
CMB model is well established form of receptor-oriented air quality model (Miller et al., 1972; Watson et al., 1990; Watson et al., 1991; Watson et al., 1994). It assumes that signatory molecules do not undergo chemical transformation from source to receptors (Banerjee et al., 2015). CMB model is composed of a set of linear equations. The total PM concentration of ambient samples can be described as the sum of the contributions of every source categories (Feng et al., 2002). On the basis of its principle, Feng (2002) established nested chemical mass balance (NCMB) —an improved CMB analysis, and had been applied in earlier studies (Bi et al., 2007; Wu et al., 2009a). And it was applied in this study for source apportionment of ambient PM. According to the NCMB method, the chemical profiles from local source emissions and ambient PM concentrations were needed. The details of NCMB approach were as discussed in other researches (Feng et al., 2002; Bi et al., 2007).
The HYSPLIT (Hybrid Single-Particle Lagrangian Integrated Trajectory) model is a complete system for computing simple air parcel trajectories to complex dispersion and deposition simulations (Draxler et al., 1998). Backward trajectories indicate where it comes from. And it has been widely applied in many short-term and case studies (Abdalmogith and Harrison, 2005; Hondula et al., 2010). In an attempt to capture the potential importance of different source regions on aerosol composition at the sampling sites, four 72-h back trajectory at $10\textrm{m}$ above the starting point was computed at 00:00, 06:00, 12:00 and 18:00 UTC for each day during sampling month (Kalnay et al., 1996; Salvador et al., 2008; Kong et al., 2010). The input reanalysis meteorological data were from the National Centers for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR). A K-means Clustering algorithm was used to maximize the homogeneity of the trajectories within the clusters and to maximize the heterogeneity among the clusters (NOAA’s ARL, 2009; Abdalmogith and Harrison, 2005; Makra et al., 2011).
3 Results and discussion
3.1 Ambient PM concentration
Daily $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ mass concentrations ranged from 12.64 to $130.53\;\upmu\mathrm{g}/\mathrm{m}^{3}$ and 8.40 to 91.70 $\upmu\mathrm{g}/\mathrm{m}^{3}$ , respectively (Fig. 2). The daily mean concentration of $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ at each site in spring attained National Ambient Air Quality Standard of China (GB 3095-2012) Grade ǀ and USA EPA NAAQS, while it reached to National Ambient Air Quality Standard of China (GB 3095-2012) Grade ǁ when in winter. But the PM pollution was more serious in winter when compared with spring. The concentration of PM in winter $\mathrm{\bfPM}_{10}$ : $77.23~\upmu\mathrm{g}/\mathrm{m}^{3}$ ; $\mathrm{PM}_{2.5}$ : $48.14~\upmu\mathrm{g}/\mathrm{m}^{3})$ was about two times of the concentration in spring ( $\mathrm{\bfPM}_{10}$ : $35.14~\upmu\mathrm{g}/\mathrm{m}^{3}$ ; $\mathrm{PM}_{2.5}$ : $20.73~\upmu\mathrm{g}/\mathrm{m}^{3})$ . For $\mathrm{{PM}_{10}}$ , the concentration at HN was the lowest among these sites in spring, while it was higher than other three sites in winter. For $\mathrm{PM}_{2.5}$ , the concentration at HU was higher than HN in spring, while it was at HU was lower than HN in winter. Compared with other researches, the concentration of $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ in this study was much lower than those in other continental cities in China (Wang et al., 2005; Bi et al., 2007; Gu et al., 2014; Zhang et al., 2014; Li et al., 2015; Zhang et al., 2015; Xu et al., 2016). It was nearly the same with Shenzhen, Zhuhai and Hong Kong (Lai et al., 2007), while was lower than other coastal sites (Bi et al.,, 2007; Wang et al., 2007; Wu et al., 2009; Xiao et al., 2012; Wu et al., 2013; Zhang et al., 2014; Liu et al., 2015). It indicated that the air quality of Haikou was relatively better in China.
In this study, the ratio of $\mathrm{PM}_{2.5}\;\mathrm{to}\;\mathrm{PM}_{10}$ was higher than $60\;\%$ both in winter and spring. The ratio was slightly lower than the Pearl River Delta (Cao et al., 2004) and Hong Kong (Ho et al., 2003). It indicated that fine particles had become a major component of atmospheric particulate matter pollution in Haikou. Generally, high ratios (e.g., larger than 0.6) were ascribed to secondary particulate formation of species such as $\mathrm{SO}_{4}^{\ 2-},\mathrm{NO}_{3}^{\ -}$ , and $\mathrm{NH_{4}}^{+}$ and organics (Chan et al., 2005). Although the consumption of coal in the energy structure was minimized and had a declining trend (HPBS, 2011), the average annual sunshine hours are over than $2000\ \mathrm{h}$ so that it is great possibility of photochemical reaction occurring in Haikou, which would produce secondary particulate matter.
Fig. 2. The daily average concentration at each site of $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ . (HU: Hainan University; HN: Haikou Normal University; LH: Long Hua; XY: Xiu Ying)
3.2 Chemical composition of PM
Carbonaceous species, ions $\mathrm{NO}_{3}^{\,\ensuremath{-}}$ and $\mathrm{SO}_{4}^{\ 2-}$ ) and elements (Na, Al, Si and Ca) were abundant components in ambient $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ samples as shown in Fig. 3. The following were Mg, K, Fe and Zn, and other elements (Ti, V, Cr, Mn, Ni, Pb) were the lowest. The overall percentage of these chemical species (Carbonaceous species, $\mathrm{NO}_{3}{}^{-}$ , $\mathrm{SO}_{4}{}^{2-}$ , Na, Al, Si and Ca) was in the range of $50.34\!-\!64.54\;\%$ , which was consistent with the conclusion found by Han Bo (2011).
There were still differences in winter and spring. The concentration of each component in PM was higher in winter than in spring. The content of TC and $\mathrm{NO}_{3}^{\mathrm{~-~}}$ were higher in spring than in winter, while $\mathrm{SO}_{4}^{\ 2-}$ was higher in winter than in spring. For example, the average value of $\mathrm{SO}_{4}^{\ 2-}$ was $10.76\;\upmu\mathrm{g}/\mathrm{m}^{3}$ and $7.06\ \upmu\mathrm{g/m}^{3}$ in $\mathrm{{PM}_{10}}$ and $\mathbf{PM}_{2.5}$ samples in winter, respectively, while the average value was 3.60 $\upmu\mathrm{g}/\mathrm{m}^{3}$ and $3.07\ \upmu\mathrm{g}/\mathrm{m}^{3}$ in spring. In addition, the concentration of each component was in the order as in spring, $\mathrm{TC>SO_{4}}^{2-}\mathrm{>NO_{3}^{\prime}>S i>N a>C a>A l>K>M g}$ , whereas in winter, $\mathrm{TC}{>}\mathrm{SO}_{4}^{\;2-}{>}\;\mathrm{NO}_{3}^{\;-}{>}\;\mathrm{Na}{>}\;\mathrm{Si}{>}$ $\mathrm{Al{>}C a{>}M g>}\mathrm{K}$ .
TC has a significant contribution to observed $\mathrm{PM_{10}}$ and $\mathrm{PM}_{2.5}$ mass at sampling sites during the study period, accounting for $17.19\%\:\sim\:24.47\%$ and $25.03\%\:\sim\:32.32\%$ in $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ samples, respectively. On average, OC accounted for $75.05\%$ of the total carbon mass in $\mathrm{{PM}_{10}}$ in Haikou, while accounted for $70.26\%$ of the total carbon mass in $\mathrm{PM}_{2.5}$ mass. As shown in Fig. 4, there was OC–EC correlation $\mathbf{R}^{2}{=}\,0.59)$ in $\mathrm{{PM}_{10}}$ samples, while it has low correlation $\mathbf{\mathcal{R}}^{2}{=}\,0.17)$ in $\mathrm{PM}_{2.5}$ . It suggested the origin of OC and EC was different. There may be other sources of fine particulate matter. Moreover, the OC/EC ratio was 2.8 and $2.3\:\mathrm{~in~}\,\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5}$ , respectively. It indicated that there existed secondary organic carbon (SOC) (Cao et al., 2003).
Besides, TC was a tracer of vehicle exhaust in previous studies (Guo et al., 2009; Kam et al., 2012; Keuken et al., 2012; Pant and Harrison, 2013). $\mathrm{NO}_{3}{}^{-}$ and $\mathrm{SO}_{4}{}^{2-}$ were tracers of secondary particles. This suggested that vehicle exhaust and secondary particles photochemical reactions may play an important role in atmospheric particles. Furthermore, high $\mathrm{Na}$ and Si abundances were observed in ambient samples. Na, mainly originating from crust and sea salt (Kong et al., 2010), indicates that sea salt may be an important contributor to ambient PM and should not be ignored (Wang et al., 2014; Hu et al., 2015; Hsu et al., 2016).
3.3 Source profiles
The main source ingredients spectrum were presented in Fig. 5, including resuspended dust, coal combustion dust, construction and cement dust, vehicle exhaust, secondary particles and sea salt sources. Mass of these compositions except OC ranged between $45.47\%$ and $98.51\%$ in the six source categories.
According to the profile of resuspended dust, the main chemical components which contents were high were crustal elements such as Na, Mg, Al, Si, Ca and Fe. And Si was the highest abundance in Haikou by more than a quarter, which is the same as in some other cities (Vega et al., 2001; Bi et al., 2007; Cao et al., 2008). The content of Ca was lower than inland cities such as Taiyuan and Jinan (Bi et al., 2007), while it was higher than coastal cities such as Tianjin (Bi et al., 2007). Si was generally considered as an identification element of resuspended dust. This was identical with previous studies. The levels of TC, sulfate and nitrate were relatively lower when compared with coal combustion dust and vehicle exhaust.
The major species of coal combustion dust were Al, Si, Ca, $\mathrm{SO}_{4}{}^{2-}$ and TC. In this study, Si was the most abundant species, accounting for $12.00\%$ , other major species were TC, Al, $\mathrm{SO}_{4}{}^{2-}$ , Fe and Na with abundances between $1.64\%$ and $10.10\%$ .
Construction and cement dust is defined as the dust that is emitted from the cement manufacturing industries and construction projects. It exhibited the greatest amount of $\mathrm{Ca}\left(39.21\%\right)$ among all sources examined in this study. Other important elements were Si $(8.51\%)$ and $\mathrm{Al.}(8.18\%)$ . The Si was similar to other cities as reported but the level of Al was much high, while TC and OC were much lower (Chow et al., 1996b; Castro et al., 1999; Vega et al., 2001).
Vehicle exhaust component spectrum obtained from U.S. EPA Speciate 4.2 (Hsu et al., 2009). Carbonaceous species were the primary component of the source profile for vehicle exhaust obtained in this study. TC accounted for $89.87\%$ of the mass determined in vehicle exhaust samples, and other major species were $\mathrm{SO}_{4}^{\ 2-}$ and Fe, accounting for $5.05\%$ .
Secondary sulfate and nitrate in particulate matter were assumed to be the secondary aerosol component produced from the oxidation and conversion of sulfur dioxide and nitrogen dioxide (Khoder, 2002). In this study, pure ammonium bisulfate $((\mathrm{NH}_{4})_{2}\mathrm{SO}_{4})$ and ammonium nitrate $\left(\mathrm{NH}_{4}\mathrm{NO}_{3}\right)$ ) were introduced to give the secondary sulfate and nitrate profiles with $73\%$ $\mathrm{SO}_{4}^{\ 2-}$ and $77\%\ \mathrm{NO}_{3}^{\texttt{-}}$ (Bi et al., 2007).
Sea salt particle component spectrum also obtained from U.S. EPA Speciate 4.2 (Hsu et al., 2009). The principal components of sea salt were Na, K, Ca and $\mathrm{SO}_{4}^{\ 2-}$ , other major species were lower. The proportion of Na was up to $40.00\%$ , which was significantly higher than that is in case in other source categories. This suggested Na was a tracer of sea salt particles.
(a: resuspended dust; b: coal combustion dust; c: construction and cement dust; d: vehicle exhaust; e:
3.4 Source apportionment of $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$
Although the CMB receptor model is well dealt with sources apportionment of primary aerosols that are emitted directly, it could not well do with secondary aerosols, because the source and ambient PM profiles cannot be input to the model directly. And the secondary organic aerosol (SOA) in particulate matter accounted for more and more in recent years, especially in fine particles (Cao et al., 2004; Cao et al., 2007; Guo et al., 2009; Li and Bai, 2009). There may existed secondary organic reactions (SOC) based on section 3.2. So the ambient particulate matter could not be input to chemical mass balance (CMB) receptor model directly in this work.
At present, there was no direct analytical determination method of SOC, so several indirect methodologies had been applied in the evaluation of SOC (Turpin et al., 1991; Cao et al., 2004; Gray et al., 1986; Chan et al., 2005; Cao et al., 2003; Cao et al., 2005; Cao et al., 2007; Duan et al., 2005). In this study, the concentration of SOC was calculated using minimum OC/EC ratio method (Turpin et al.1995; Castro et al., 1999; Wu et al., 2009a).
$$
\mathrm{OC_{\mathrm{sec}}=O C_{\mathrm{tot}}-E C\times(O C/E C)_{\mathrm{min}}}
$$
Where $\mathrm{OC_{sec}}$ is the concentration of secondary OC $(\upmu\mathrm{g}/\mathfrak{m}^{3})$ , $\mathrm{OC_{\mathrm{{tot}}}}$ is the concentration of total OC $(\upmu\mathrm{g}/\mathrm{m}^{3})$ and $(\mathrm{OC/EC)_{min}}$ is the minimum ratio observed.
Due to the differences in source emission and photochemical reaction conditions, SOC concentrations of ambient $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ were calculated for each season separately. Average amount of deducting SOC concentrations of $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ in Haikou were summarized in Table 2. The concentrations of SOC estimated through the minimum OC/EC ratio method were $4.1\ \upmu\mathrm{g/m}^{3}$ and 2.3 $\upmu\mathrm{g}/\mathrm{m}^{3}$ in winter and spring for $\mathrm{\bfPM}_{10}.$ , $3.4~\upmu\mathrm{g}/\mathrm{m}^{3}$ and $1.7~\upmu\mathrm{g}/\mathrm{m}^{3}$ for $\mathrm{PM}_{2.5}$ , respectively. Compared with other studies, the concentration of SOC in Haikou was similar with Shenzhen $(3.0\ \upmu\mathrm{g/m}^{3}$ and $4.4\;\upmu\mathrm{g/m}^{3}$ for $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ , respectively), Zhuhai $(3.4~\upmu\mathrm{g/m}^{3}$ and $3.8~\upmu\mathrm{g/m}^{3}$ for $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ , respectively) and Hong Kong $(2.2~\upmu\mathrm{g/m}^{3}$ and $2.1~\upmu\mathrm{g/m}^{3}$ for $\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5}$ , respectively) (Cao et al., 2004). While compared with Tianjin $(5.82\ \upmu\mathrm{g}/\mathrm{m}^{3}$ for $\mathrm{PM}_{2.5}$ ) (Li et al. 2009; Shi et al., 2012) and Guangzhou (9.3 $\upmu\mathrm{g}/\mathrm{m}^{3}$ and $12.9\ \upmu\mathrm{g/m}^{3}$ for $\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5}$ , respectively) (Cao et al., 2004), the concentration of SOC in Haikou was lower.
The SOC concentrations were deducted from the $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ compositions. The receptor profiles were rebuilt and then introduced to the model with source profiles. In this study, T statistics, chi squared $(\boldsymbol{\chi}^{2}),\;\boldsymbol{\mathrm{R}}^{2}$ of the CMB simulation was 2.0, 0.8, 0.92, respectively. And the source categories accounted for $90\%$ of the measured $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ mass in this study. These values of the above are within a reasonable range (Feng et al., 2002).
The results of CMB receptor model was represented in Fig. 6. It was obvious that the major source categories of particulate matter were resuspended dust, vehicle exhaust and secondary sulfate in Haikou. The contribution of resuspended dust varied from $14.9\%$ to $23.6\%$ , vehicle exhaust ranged from $17.5\%$ to $35\%$ and secondary sulfate changed from $9.5\%$ to $15.7\%$ . In addition, the contribution of sea salt changed from $3\%$ to $8\%$ .
However, there were differences in source apportionment of ambient $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ during different seasons. The contributions of coal combustion dust for $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ in winter were higher than in spring. However, Haikou has no heating seasons, and its primary energy was petroleum products and electricity. Therefore, there may be other sources in winter. The trend of sulfate was identical with coal combustion dust. Moreover, construction and cement dust was founded to have the highest contribution to $\mathrm{PM}_{10}$ in spring, with the contribution of $12.8\%$ . This might be caused by anthropologic activities such as construction activities and weather conditions. And the contributions of resuspended dust and construction and cement dust to $\mathrm{PM}_{10}$ were higher than $\mathrm{PM}_{2.5}$ , while the contributions of vehicle exhaust, sulfate, nitrate and SOC to $\mathrm{{PM}_{10}}$ were lower than $\mathrm{PM}_{2.5}$ .
3.5 Back trajectory analysis
There might be other pollution sources in winter based on section 3.4. However, it had limited local emission of PM in Haikou, and it had no significant seasonal differences. The concentration of particulate matter was higher with the dominant wind of Haikou was ENE or NE in winter. On the contrary, the concentration of particulate matter was lower in spring owing to the mainly wind was SSE or ESE. The air mass transport plays an important pole in the distribution of species. As a result, it could indicate some potential sources to some extent through showing a high concentration of trace elements and a significant contribution of anthropogenic sources (Kong et al., 2010; Wang et al., 2013). Thus there maybe existed pollutants from inland areas which moved to Haikou city through atmospheric dispersion. Therefore, the back trajectory analysis was done by HYSPLIT in this study.
The clustered 72-h backward trajectories were presented in Fig. 7. In winter, when northeasterly winds begin to prevail, the trajectories were categorized into 3 sectors. Sector 1 accounted for $46\%$ , while sector 2 accounting for $19\%$ and sectors 3 accounting for $36\%$ . The three trajectories represented air mass from the northeast of China, passing the Pearl River Delta. It suggested that the concentration of aerosols in winter was affected by regional sources.
In spring, the air masses were mainly originating from southeast of the South China Sea (sector 1 accounting for $78\%$ ). Only a partly accounting for $22\%$ came from inland cities (sector 2). Generally, marine air masses could bring an amount of marine species (Na, K, Mg), but not pollutants. Furthermore, marine air masses may dilute the pollutants. This might be explained why the air quality of Haikou in spring better than in winter. In summary, Haikou might act as a receptor of the huge anthropogenic emissions from the northest of China by long- and medium-range transport under the
influence of the monsoon system in winter period.
4 Conclusions
In this study, ambient PM were collected in winter and spring in Haikou, China. The daily concentrations of $\mathrm{{PM}_{10}}$ were $35.14\;\upmu\mathrm{g}/\mathrm{m}^{3}$ and $77.23\;\upmu\mathrm{g}/\mathrm{m}^{3}$ in spring and winter, respectively. And the daily concentrations of $\mathrm{PM}_{2.5}$ were $20.73~\upmu\mathrm{g}/\mathrm{m}^{3}$ and $48.14~\upmu\mathrm{g/m}^{3}$ in spring and winter, respectively. And the proportion of $\mathrm{PM}_{2.5}$ in $\mathrm{{PM}_{10}}$ was large $_{(>0.6)}$ . The concentration of each chemical component was higher in winter than in spring. And carbonaceous species and ions ( $\mathrm{NO}_{3}$ ⁻and $\mathrm{SO}_{4}{}^{2-}$ ) were abundant constituents in ambient $\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5}$ samples, followed by crustal materials (Al, Si, Na, Mg, K and $\mathrm{Ca}_{.}$ ). The percentage of these chemical species was found in the range of $67.3–79.13\%$ .
Owing to the presence of SOC, the SOC concentration would be subtracted from PM. An indirect method of “OC/EC minimum ratio” was applied to estimate the concentration of SOC. In this work, the percentages of SOC were $5.95\%$ and $7.25\%$ for ambient $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ , respectively. Six major source categories of ambient $\mathrm{PM_{10}}$ and $\mathrm{PM}_{2.5}$ were identified was estimated by CMB model. As a result, the resuspended dust, vehicle exhaust and sulfate were the major contributors. The sea salt had contributions, too. In winter, long- and medium-range transport from the northeast of China had played an important role on the air quality of Haikou under the influence of the monsoon system through back trajectories analysis. Moreover, the SOC and sea salt would be consider in other source apportionment of coast cities by CMB. And the regional sources also had an impact on particulate matter, especially in winter.
Acknowledgement
Authors are grateful to Haikou Environmental Monitoring Center Station to help collect ambient samples and source samples.
References
Abdalmogith, S.S., Harrison, R.M., 2005. The use of trajectory cluster analysis to examine the long-range transport of secondary inorganic aerosol in the UK. Atmos. Environ. 39(35), 6686-6695.
Appel, B.R., Tokiwa, Y., Hsu, J., Kothny, E.L., Hahn, E., 1985. Visibility as related to atmospheric aerosol constituents. Atmos. Environ. 19(9), 1525-1534.
Ariola, V., D’Alessandro, A., Lucarelli, F., Marcazzan, G., Mazzei, F., Nava, S., Garcia-Orellana, I., Prati, P., Valli, G., Vecchi, R., Zucchiatti, A., 2006. Elemental characterization of $\mathrm{{PM}_{10}}$ , $\mathrm{PM}_{2.5}$ and $\mathrm{{PM}_{1}}$ in the town of Genoa (Italy). Chemosphere, 62(2), 226-232.
Baik, N.J., Kim, Y.P., Moon, K.C., 1996. Visibility study in Seoul, 1993. Atmos. Environ. 30(13), 2319-2328.
Baldwin, D.P., Zamzow, D.S., D'Silva, A.P., 1994. Aerosol mass measurement and solution standard additions for quantitation in laser ablation-inductively coupled plasma atomic emission spectrometry. Anal. Chem. 66(11), 1911-1917.
Banerjee, T., Murari, V., Kumar, M., Raju, M.P, 2015. Source apportionment of airborne particulates through receptor modeling: Indian scenario. Atmospheric Research, 164–165, 167-187.
Bi, X.H., Feng, Y.C., Wu, J.H., Wang, Y.Q., Zhu, T., 2007. Source apportionment of $\mathrm{{PM}_{10}}$ in six cities of northern China. Atmos. Environ. 41(5), 903-912.
Blifford, I.H., Meeker, G.O., 1967. A factor analysis model of large scale pollution. Atmos. Environ. 1(2), 147-157.
Cao, J.J., Chow, J.C., Watson, J.G., Wu, F., Han, Y.M., Jin, Z.D., Shen, Z.X., An, Z.S., 2008. Size-differentiated source profiles for fugitive dust in the Chinese Loess Plateau. Atmos. Environ. 42(10), 2261-2275.
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(11), 1451-1460.
Cao, J.J., Lee, S.C., Ho, K.F., Zou, S.C., Fung, K., Li, Y., Watson, J.G., Chow, J.C., 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, J.J., Wu, F., Chow, J.C., Lee, S.C., Li, Y., Chen, S.W., An, Z.S., Fung, K.K., Watson, J.G., Zhu, C.S., Liu, S.X., 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(11), 3127-3137.
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. D112(D22), 22-11.
Castro, L.M., Pio, C.A., Harrison, R.M., Smith, D.J.T., 1999. Carbonaceous aerosol in urban and rural European atmospheres: estimation of secondary organic carbon concentrations. Atmos. Environ. 33(17), 2771-2781.
Chan, C.Y., Xu, X.D., Li, Y.S., Wong, K.H., Ding, G.A., Chan, L.Y., Cheng, X.H., 2005. Characteristics of vertical profiles and sources of $\mathrm{PM}_{2.5}$ , $\mathrm{{PM}_{10}}$ and carbonaceous species in Beijing. Atmos. Environ. 39(28), 5113-5124.
Chen, R., Kan, H., Chen, B., Huang, W., Bai, Z., Song, G., Pan, G., 2012. Association of particulate air pollution with daily mortality: the china air pollution and health effects study. American Journal of Epidemiology, 175(11), 1173-81.
Chow, J.C., Watson, J.G., Chen, L.W.A., Chang, M.O., 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 Manage. Assoc. 57(9), 1014-1023.
Chow, J.C., Watson, J.G., Lowenthal, D.H., Countess, R.J., 1996. Sources and chemistry of $\mathrm{PM}_{10}$ aerosol in Santa Barbara County, CA. Atmos. Environ. 30(9), 1489-1499.
Chow, J.C., Watson, J.G., Lu, Z.Q., Lowenthal, D.H., Frazier, C.A., Solomon, P.A., Thuillier, R.H., Magliano, K., 1996. Descriptive analysis of $\mathbf{PM}_{2.5}$ and $\mathrm{PM}_{10}$ at regionally representative locations during SJVAQS/AUSPEX. Atmos. Environ. 30(12), 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. Atmos. Environ. Part A. General Topics, 27(8), 1185-1201.
Cohen, A.J., Ross Anderson, H., Ostro, B., Pandey, K.D., Krzyzanowski, M., Künzli, N., Gutschmid, K., Pope, A., Romieu, I., Samet, J.M., Smith, K., 2005. The global burden of disease due to outdoor air pollution. J. Toxicol. Environ. Heal. Part A, 68(13-14), 1301-1307.
Cooper, J.A., Watson Jr. J.G., 1980. Receptor oriented methods of air particulate source apportionment. J. Air Pollut. Control Assoc. 30(10), 1116-1125.
Dockery, D.W., Pope, C.A, Xu, X., Spengler, J.D., Ware, J.H., Fay, M.E., Ferris, B.G., Speizer, F.E., 1993. An association between air pollution and mortality in six U.S. cities. New England Journal of Medicine, 329(24), 1753-1759.
Draxler, R.R., Hess, G.D., Draxler, R.R., 1998. An overview of the HYSPLIT_4 modelling system for trajectories. Australian Meteorological Magazine, 47(4), 295-308.
Duan, F.K., He, K.B., Ma, Y.L., Jia, Y.T., Yang, F.M., Lei, Y., Tanaka, S., Okuta, T., 2005. Characteristics of carbonaceous aerosols in Beijing, China. Chemosphere, 60(3), 355-364.
Feng, Y.C., Bai, Z.P., Zhu, T., 2002. The principle and application of improved-source-apportionment technique of atmospheric particulate matter. J. Environ. Sci. 23(Suppl.1), 106-108 (in Chinese).
Feng, Y.C., Peng, L., Wu, J.H., Zhu, T., Lv, A.H., Zhang, K.T., 2005. Analytic studies on source of TSP and $\mathrm{PM}_{10}$ in environmental air of Urumqi city. China Environ. Sci. 25 (Suppl.1), 30-33 (in Chinese).
Forastiere, F., 2004. Fine particles and lung cancer. Occupational & Environmental Medicine, 61(10), 797-8.
Gao, Y., Guo, X.Y., Ji, H.B., Li, C., Ding, H.J., Briki, M., Tang, L., Zhang, Y., 2016. Potential threat of heavy metals and PAHs in PM2.5 in different urban functional areas of Beijing. Atmos. Res.178, 6-16.
Gehrig, R., Hill, M., Lienemann, P., Zwicky, C.N., Bukowiecki, N., Weingartner, E., Baltensperger, U., Buchmann, B., 2007. Contribution of railway traffic to local $\mathrm{{PM}_{10}}$ concentrations in Switzerland. Atmos. Environ. 41(5), 923-933.
Gray, H.A., Cass, G.R., Huntzicker, J.J., Heyerdahl, E.K., Rau, J.A., 1986. Characteristics of atmospheric organic and elemental carbon particle concentrations in Los Angeles. Environ. Sci. Technol. 20(6), 580-589.
Gu J.X., Du S.Y., Han D.W., Hou, L.J., Yi, J., Xu, J., Liu, G.H., Han, B., Yang, G.W., Bai, Z.P., 2014. Major chemical compositions, possible sources, and mass closure analysis of $\mathbf{PM}_{2.5}$ in Jinan, China. Air Qual. Atmos. Health. 7, 251-262.
Guo, H., Ding, A.J., So, K.L., Ayoko, G., Li, Y.S., Hung, W.T., 2009. Receptor modeling of source apportionment of Hong Kong aerosols and the implication of urban and regional contribution. Atmos. Environ. 43(6), 1159-1169.
Guttikunda, S.K., Kopakka, R.V., Dasari, P., Gertler, A.W., 2013. Receptor model-based source apportionment of particulate pollution in Hyderabad, India. Environ. Monit. Assess. 185(7), 5585-5593.
Hainan Provincial Bureau of Statistics., 2011. Statistical yearbook of Hainan city in 2010. China Statistical Publishing House, Beijing.
Han, B., Bi, X.H., Xue, Y.H., Wu, J.H., Zhu, T., Zhang, B.G., Ding, J.Q., Du, Y.X., 2011. Source apportionment of ambient $\mathrm{{PM}_{10}}$ in urban areas of Wuxi, China. Front. Environ. Sci. Eng. China, 5(4), 552-563.
Harrison, R.M., Smith, D.J.T., Piou, C.A., Castro, L.M., 1997. Comparative receptor modelling study of airborne particulate pollutants in Birmingham (United Kingdom), Coimbra (Portugal) and Lahore (Pakistan). Atmos. Environ. 31(20), 3309-3321.
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(29), 4959-4970.
Henry, R.C., Lewis, C.W., Hopke, P.K., Williamson, H.J., 1984. Review of receptor model fundamentals. Atmos. Environ. 18(8), 1507-1515.
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(1), 31-39.
Hondula, D.M., Sitka, L., Davis, R.E., Knight, D.B., Gawtry, S.D., Deaton, M.L., Lee, T.R., Normile, C.P., Stenger, P.J., 2010. A back‐trajectory and air mass climatology for the Northern Shenandoah Valley, USA. Int. J. Climatol. 30(4), 569-581.
Hopke, P.K., 2003. Recent developments in receptor modeling. Journal of Chemometrics. 17(5), 255-265.
Hsu, C.Y., Chiang, H.C., Lin, S.L., Chen, M.J., Lin, T.Y., Chen, Y.C., 2016. Elemental characterization and source apportionment of $\mathrm{PM}_{10}$ , and $\mathbf{PM}_{2.5}.$ , in the western coastal area of central Taiwan. Science of the Total Environment, 541, 1139-1150.
Hsu, Y.K., Divita, $\operatorname{F}.,~2009$ . SPECIATE 4.2: Speciation Database Development Documentation. National Risk Management Research Laboratory, Office of Research and Development, U.S. Environmental Protection Agency.
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. Atmos. Environ. 39(4), 637-651.
Hu, J., Wu, L., Zheng, B., Zhang, Q., He, K., Chang, Q., Li, X., Yang, F., Ying, Q., Zhang, H., 2015. Source contributions and regional transport of primary particulate matter in china. Environmental Pollution, 207, 31-42.
Kalnay, E., Kanamitsu, M., Kistler, R., Collins, W., Deaven, D., Gandin, L., Iredell, M., Saha, S., White, G., Woollen, J., Zhu, Y., Chellian, M., Ebisuzakl, W., Higgins, W., Janowiak, J., Mo, K.C., Ropelewskl, C., Wang, J., Leetmaa, A., Reynolds, R., Jenne, R., Joseph, D., 1996. The NCEP/NCAR 40-year reanalysis project. B. Am. Meteorol. Soc. 77(3), 437-472.
Kam, W., Liacos, J.W., Schauer, J.J., Delfino, R.J., Sioutas, C., 2012. Size-segregatedcomposition of particulate matter (PM) in major roadways and surface streets. Atmos. Environ. 55, 90-97.
Kesavachandran, C., Pangtey, B.S., Bihari, V., Fareed, M., Pathak, M.K., Srivastava, A.K., Mathur, N., 2012. Particulate matter concentration in ambient air and its effects on lung functions among residents in the national capital region, India. Environ. Monit. Assess. 185(2), 1265-1272.
Keuken, M.P., Henzing, J.S., Zandveld, P., van den Elshout, S., Karl, M., 2012. Dispersion of particle numbers and elemental carbon from road traffic, a harbour and an airstrip in the Netherlands. Atmos. Environ. 54, 320-327.
Khoder, M.I., 2002. Atmospheric conversion of sulfur dioxide to particulate sulfate and nitrogen dioxide to particulate nitrate and gaseous nitric acid in an urban area. Chemosphere, 49(6), 675-684.
Kong, S.F., Han, B., Bai, Z.P., Chen, L., Shi, J.W., Xu, Z., 2010. Receptor modeling of $\mathrm{PM}_{2.5}$ , $\mathrm{{PM}_{10}}$ and TSP in different seasons and long-range transport analysis at a coastal site of Tianjin, China. Sci. Total Environ. 408(20), 4681-4694.
Lai S.C., Zou S.C., Cao J.J., Lee, S.C., Ho, K.F., 2007. Characterizing ionic species in $\mathrm{PM}_{2.5}$ and $\mathrm{PM_{10}}$ in four Pearl River Delta cities, South China. J. Environ. Sci. 19, 939-947.
Law, K.S., Stohl, A., 2007. Arctic air pollution: Origins and impacts. Science, 315(5818), 1537-1540.
Li, M.H., Fan, L.C., Mao, B., Yang, J.W., Choi, A.M.K., Cao, W.J., Xu, J.F., 2016. Short-term exposure to ambient fine particulate matter increases hospitalizations and mortality in COPD: a systematic review and meta-analysis. Chest, 149(2), 447-458.
Li W.F., Bai Z.P., 2009. Characteristics of organic and elemental carbon in atmospheric fine particles in Tianjin, China. Particuology, 06, 432–437.
Li Y., Chen Q.L., Zhao H.J., Wang, L., Tao, $\mathrm{R}.,2015$ . Variations in $\mathrm{\bfPM}_{10}$ $\mathrm{PM}_{2.5}$ and $\mathrm{{PM}_{1.0}}$ in an Urban Area of the Sichuan Basin and Their Relation to Meteorological Factors. Atmosphere, 6(1), 150-163.
Liu J.J., Man R., Ma S.X., Li, J.S., Wu, Q., Peng, J.Y., 2015. Atmospheric levels and health risk of polycyclic aromatic hydrocarbons (PAHs) bound to $\mathbf{PM}_{2.5}$ in Guangzhou, China. Mar. Pollut. Bull. 100, 134–143.
Makra, L., Matyasovszky, I., Guba, Z., Karatzas, K., Anttila, P., 2011. Monitoring the long-range transport effects on urban $\mathrm{PM}_{10}$ levels using 3D clusters of backward trajectories. Atmos. Environ. 45(16), 2630-2641.
Marcazzan, G.M., Ceriani, M., Valli, G., Vecchi, R., 2003. Source apportionment of $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ in Milan (Italy) using receptor modeling. Sci. Total Environ. 317(1), 137-147.
Marcazzan, G.M., Vaccaro, S., Valli, G., Vecchi, R., 2001. Characterisation of $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ particulate matter in the ambient air of Milan (Italy). Atmos. Environ. 35(27), 4639-4650.
Miller, M.S., Friedlander, S.K., Hidy, G.M., 1972. A chemical element balance for the Pasadena aerosol. J. Colloid Interf. Sci. 39(1), 165-176.
NOAA ’s Air Resources Laboratory, January 2009. Hysplit4 User’s Guide, Version 4.9. revision. http://www.arl.noaa.gov/HYSPLIT_pubs.php.
Pant P., Harrison R.M., 2013. Estimation of the contribution of road traffic emissions to particulate matter concentrations from field measurements: A review. Atmos. Environ. 77, 78-97.
Pope III, C.A., Dockery, D.W., 2006. Health effects of fine particulate air pollution: lines that connect. J. Air Waste Manage. Assoc. 56(6), 709-742.
Pope, C.A., 2000. Invited commentary: particulate matter-mortality exposure-response relations and threshold. American Journal of Epidemiology, 152(5), 407-12.
Pope, CA. 3rd., Thun, MJ., Namboodiri, MM., Dockery, DW., Evans, JS., Speizer, FE., Heath, CW. Jr., 1995. Particulate air pollution as a predictor of mortality in a prospective study of U.S. adults. American Journal of Respiratory & Critical Care Medicine, 151(3 Pt 1), 669-74.
Querol, X., Alastuey, A., Rodriguez, S., Plana, F., Ruiz, C.R., Cots, N., Massague, G., Puig, O., 2001. $\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5}$ source apportionment in the Barcelona Metropolitan area, Catalonia, Spain. Atmos. Environ. 35(36), 6407-6419.
Querol, X., Alastuey, A., Viana, M.M., Rodriguez, S., Artiñano, B., Salvador, P., Garcia do Santos, S., Fernandez Patier, R., Ruiz, C.R., de la Rosa, J., Sanchez de la Campa, A., Menendez, M., Gil, J.I., 2004. Speciation and origin of $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ in Spain. J. Aerosol Sci. 35(9), 1151-1172.
Rodrı́guez, S., Querol, X., Alastuey, A., Viana, M. M., Alarcón, M., Mantilla, E., Ruiz, C. R., 2004. Comparative $\mathrm{PM}_{10}\mathrm{-PM}_{2.5}$ source contribution study at rural, urban and industrial sites during PM episodes in Eastern Spain. Sci. Total Environ. 328(1), 95-113.
Salvador, P., Artíñano, B., Querol, X., Alastuey, A., 2008. A combined analysis of backward trajectories and aerosol chemistry to characterise long-range transport episodes of particulate matter: the Madrid air basin, a case study. Sci. Total Environ. 390(2), 495-506.
Shi, G.L., Tian, Y.Z., Han, S.Q., Zhang, Y.F., Li, X., Feng, Y.C., Wu, J.H., Zhu, Tan., 2012. Vertical characteristics of carbonaceous species and their source contributions in a Chinese mega city. Atmos. Environ. 60, 358-365.
Song, Y., Zhang, Y.H., Xie, S.D., Zeng, L.M., Zheng, M., Salmon, L.G., Shao, M., Slanina, S., 2006. Source apportionment of $\mathrm{PM}_{2.5}$ in Beijing by positive matrix factorization. Atmos. Environ. 40 (8), 1526-1537.
Squizzato, S., Masiol, M., Agostini, C., Visin, F., Formenton, G., Harrison, R.M., Rampazzo, G., 2016. Factors, origin and sources affecting PM1, concentrations and composition at an urban background site. Atmos. Res. 180, 262-273.
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. Atmos. Environ. 41(3), 480-492.
Tiwari, S., Tunved, P., Hopke, P.K., Srivastava, A.K., Bisht, D.S., Pandey, A.K., 2016. Observations of ambient trace gas and PM10, concentrations at Patna, Central Ganga Basin during 2013–2014: the influence of meteorological variables on atmospheric pollutants. Atmos. Res. 180, 138-149.
Turpin, B.J., Huntzicker, J.J., 1995. Identification of secondary organic aerosol episodes and quantitation of primary and secondary organic aerosol concentrations during SCAQS. Atmos. Environ. 29(23), 3527-3544.
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. Atmos. Environ. Part A. General Topics, 25(2), 207-215.
Tyler G., 1992. ICP vs. AA: how do you choose. Environ Lab, 31– 34.
U.S. EPA., 2004. Air quality criteria for particulate matter, Vol. 1, 2 and 3. Research Triangle Park, NC: National Center for Environmental Assessment-RTP Office.
Vega, E., Mugica, V., Reyes, E., Sanchez, G., Chow, J.C., Watson, J.G., 2001. Chemical composition of fugitive dust emitters in Mexico City. Atmos. Environ. 35(23), 4033-4039.
Wang, L., Qi, J.H., Shi, J.H., Chen, X.J., Gao, H.W., 2013. Source apportionment of particulate pollutants in the atmosphere over the Northern Yellow Sea. Atmos. Environ. 70, 425-434.
Wang, X.H., Bi, X.H., Sheng, G.Y., Fu, J.M., 2006. Chemical composition and sources of $\mathrm{{PM}_{10}}$ and $\mathbf{PM}_{2.5}$ aerosols in Guangzhou, China. Environ. Monit. Assess., 119(1-3), 425-439.
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. Atmos. Environ. 39(21), 3771-3784.
Wang, Y., Li, L., Chen, C., Huang, C., Huang, H., Feng, J., Wang, S., Wang, H., Zhang, G., Zhou, M., Cheng, P., Wu, M., Sheng, G., Fu, J., Hu, Y., Russell, A.G., Wumaer, A., 2014. Source apportionment of fine particulate matter during autumn haze episodes in shanghai, china. Journal of Geophysical Research Atmospheres, 119(4), 1903–1914.
Watson, J.G., Chow, J.C., Pace, T.G., 1991. Chemical mass balance. In: Hopke P K, (Ed.), Receptor Modeling For Air Quality Management (pp. 83-116). New York: Elsevier.
Watson, J.G., Chen, L.W., Chow, J.C., Doraiswamy, P., Lowenthal, D.H., 2008. Source Apportionment: Findings from the U.S. Supersites Program. J. Air Waste Manage. Assoc. 58(2), 265–288.
Watson, J.G., Chow, J.C., Lu, Z.Q., Fujita, E.M., Lowenthal, D.H., Lawson, D.R., Ashbaugh, L.L., 1994. Chemical mass balance source apportionment of $\mathrm{{PM}_{10}}$ during the Southern California Air Quality Study. Aerosol Sci. Tech. 21(1), 1-36.
Watson, J.G., Robinson, N.F., Chow, J.C., Henry, R.C., Kim, B.M., Pace, T.G., Meyer, E.L., Nguyen, Q., 1990. The USEPA/DRI chemical mass balance receptor model, CMB 7.0. Environ. Software, 5(1), 38-49.
Watson, J.G., 1979. Chemical element balance receptor model methodology for assessing the sources of fine and total suspended particulate matter in Portland, Oregon.
Watson, J.G., 1984. Overview of receptor model principles. J. Air Pollut. Control Assoc. 34 (6), 619-623.
Wei, F., Teng, E., Wu, G., Hu, W., Wilson, W.E., Chapman, R.S., Pau, J.C., Zhang, J., 1999. Ambient concentrations and elemental compositions of $\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5}$ in four Chinese cities. Environ. Sci. Technol. 33(23), 4188-4193.
WHO., 2009. Global health risks: mortality and burden of disease attributable to selected major risks. Geneva, Switzerland: World Health Organization.
Wu, H., Zhang, C.Y., Wang, J., Xuan, Z.F., Chu, C.J., Feng, Y.C., Xu, H., 2013. Comparative study on pollution characteristics and source apportionment of $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ in Qingdao. Res. Environ. Sci. 26(6), 583-589 (in Chinese).
Wu, L., Feng, Y.C., Dai, L., Han, S.Q., Zhu, T., 2009b. Characteristics of $\mathrm{{PM}_{10}}$ , $\mathrm{PM}_{2.5}$ and their carbonaceous species in Tianjin City. China Environmental Science, 29(11), 1134-1139 (in Chinese).
Wu, L., Feng Y.C., Wu, J.H., Zhu, T., Bi, X.H., Han, B., Yang, W.H., Yang, Z.Q., 2009a. Secondary organic carbon quantification and source apportionment of $\mathrm{{PM}_{10}}$ in Kaifeng, China. J. Environ. Sci. 21, 1353–1362.
Xiao, Z.M., Bi, X.H., Feng, Y.C., Wang, Y.Q., Zhou, J., Fu, X.Q., Weng, Y.B., 2012. Source apportionment of ambient $\mathrm{{PM}_{10}}$ and $\mathrm{PM}_{2.5}$ in Urban Area of Ningbo City. Res. Environ. Sci. 25(5), 549 -555 (in Chinese).
Xu, H.M., Cao, J.J., Chow, J.C., Huang, R.J., Shen, Z.X., Chen, L.W., Ho, K.F., Watson, J.G., 2016. Inter-annual variability of wintertime $\mathrm{PM}_{2.5}$ chemical composition in Xi'an, China: Evidences of changing source emissions. Sci. Total Environ. 546–555.
Yadav, S., Satsangi, P.G., 2013. Characterization of particulate matter and its related metal toxicity in an urban location in south west India. Environ. Monit. Assess. 185(9), 7365-7379.
Yang, K.L., 2002. Spatial and seasonal variation of $\mathrm{{PM}_{10}}$ mass concentrations in Taiwan. Atmos. Environ. 36(21), 3403-3411.
Zhang, G.X., Yan, Y.L., Guo, L.L., He, Q.S., Chen, L.G., 2015. Composition and Variation Characteristics of Atmospheric Carbonaceous Species in $\mathrm{PM}_{2.5}$ in Taiyuan, China. Environ. Sci. 36(3), 780-786 (in Chinese).
Zhang, L.L., Gao, Y.X., Dao, X., Wang, C., Teng, E.J., 2014. Composition and Distribution of Elements in Air Particulate Matters during Heating Season of Beijing-Tianjin-Hebei Megacities, China. Environ. Monitor. in China, 30(6), 53-61 (in Chinese).
Zhang, L.W., Chen, X., Xue, X.D., Sun, M., Han, B., Li, C.P., Ma, J., Yu, H., Sun, Z.R., Zhao, L.J., Zhao, B.X., Liu, Y.M., Chen, J., Wang, P.P., Bai, Z.P., Tang, N.J., 2014. Long-term exposure to high particulate matter pollution and cardiovascular mortality: a 12-year cohort study in four cities in northern china. Environment International, 62(62), 41-47.
Zhao, P.S., Dong, F., He, D., Zhao, X.J., Zhang, W.Z., Yao, Q., Liu, H.Y., 2013. Characteristics of concentrations and chemical compositions for $\operatorname{PM}_{2.5}\operatorname{in}$ the region of Beijing, Tianjin, and Hebei, China. Atmos. Chem. Phys. 13, 4631–4644.
|
atmosphere
|
0015
|
10.1007/s10653-016-9874-5
|
image
|
Fig. 1 Location of the sampling site (the red dot, SS). MOB meteorological observatory base of Xinxiang
|
|
atmosphere
|
0015
|
10.1007/s10653-016-9874-5
|
image
|
Fig. 2 Wind speed and wind direction during the sampling period in Xinxiang $(\mathrm{m~s^{-1}}.$ )
|
|
atmosphere
|
0015
|
10.1007/s10653-016-9874-5
|
image
|
Fig. 3 Daily variations of $\mathrm{PM}_{2.5}$ mass concentration during the sampling period. The blue line shows that the standard of China National Ambient Air Quality is $75~{\upmu\mathrm{g}}~\mathrm{m}^{-3}\left(24~\mathrm{h}\right)$ , and the green line shows that the standard of WHO Air Quality is $25~{\upmu\mathrm{g}}~\mathrm{m}^{-3}\left(24~\mathrm{h}\right)$
|
|
atmosphere
|
0015
|
10.1007/s10653-016-9874-5
|
table
|
Table 1 Comparison of $\mathrm{PM}_{2.5}$ concentrations observed during spring festival across China $(\upmu\mathrm{g}\textrm{m}^{-3})$
|
|
atmosphere
|
0015
|
10.1007/s10653-016-9874-5
|
table
|
Table 2 Mass concentrations of water-soluble ions, heavy metals and carbonaceous species in $\mathrm{PM}_{2.5}$ $\mathrm{(\upmug\m^{-3})}$ )
|
|
atmosphere
|
0015
|
10.1007/s10653-016-9874-5
|
image
|
Fig. 4 Enrichment factors (EF) for trace elements
|
|
atmosphere
|
0015
|
10.1007/s10653-016-9874-5
|
image
|
Fig. 5 Concentrations and ratios of OC and EC $\mathrm{(\upmug~m}^{-3}.$ )
|
|
atmosphere
|
0015
|
10.1007/s10653-016-9874-5
|
image
|
Fig. 6 Cluster analyses for backward trajectories of air masses during the sampling period. The different colors indicate different clusters of the backward trajectories Cluster means - Standard 20 backward trajectories GDAS Meteorological Data
|
|
atmosphere
|
0015
|
10.1007/s10653-016-9874-5
|
table
|
Table 3 Daily exposure values $(\mathrm{mg~kg^{-1}~}\dot{\mathrm{day}}^{-1})$ and health risks of heavy metals
|
|
atmosphere
|
0015
|
10.1007/s10653-016-9874-5
|
text
| Not supported with pagination yet |
$\mathbf{PM}_{2.5}$ levels, chemical composition and health risk assessment in Xinxiang, a seriously air-polluted city in North China
Jinglan Feng $\bullet$ Hao Yu $\bullet$ Shuhui Liu $\bullet$ Xianfa Su $\bullet$ Yi Li $\bullet$ Yuepeng Pan $\bullet$ Jianhui Sun
Received: 28 April 2016 / Accepted: 6 September 2016
$\copyright$ Springer Science $^+$ Business Media Dordrecht 2016
Abstract Seventeen $\mathrm{PM}_{2.5}$ samples were collected at Xinxiang during winter in 2014. Nine water-soluble ions, 19 trace elements and eight fractions of carbonaceous species in $\operatorname{PM}_{2.5}$ were analyzed. $\mathrm{PM}_{2.5}$ concentrations and elements species during different periods with different pollution situations were compared. The threat of heavy metals in $\operatorname{PM}_{2.5}$ was assessed using incremental lifetime cancer risk. During the whole period, serious regional haze pollution persisted, and the averaged concentration of $\operatorname{PM}_{2.5}$ was $168.5~\upmu\mathrm{g}~\mathrm{m}^{-3}$ , with $88.2~\%$ of the daily samples exhibiting higher $\operatorname{PM}_{2.5}$ concentrations than the national air quality standard II. The high $\mathrm{NO}_{3}{}^{-}/$ $\mathrm{SO}_{4}{}^{2-}$ ratio suggested that vehicular exhaust made an important contribution to atmospheric pollution. All of organic carbon and elemental carbon ratios in this study were above 2.0 for $\operatorname{PM}_{2.5}$ , which might reflect the combined contributions from coal combustion, motor vehicle exhaust and biomass burning. Mean 96-h backward trajectory clusters indicated that more serious air pollution occurred when air masses transported from the Hebei, Shanxi and Zhengzhou. The concentrations of the water-soluble ions and trace elements on haze days were 2 and 1.8 times of those on clear days. The heavy metals in $\operatorname{PM}_{2.5}$ might not cause non-cancerous health issues by exposure through the human respiratory system. However, lifetime cancer risks of heavy metals obviously exceeded the threshold $(10^{-6})$ and might have a cancer risk for residents in Xinxiang. This study provided detailed composition data and comprehensive analysis of $\operatorname{PM}_{2.5}$ during the serious haze pollution period and their potential impact on human health in Xinxiang.
J. Sun e-mail: sunjhhj $@$ 163.com
Keywords $\operatorname{PM}_{2.5}$ Haze $\cdot$ Air trajectory $\cdot$ Health risk $\cdot$ Xinxiang
Introduction
Fine particulate matter $(\mathrm{PM}_{2.5}$ , particles with aerodynamic diameter equal to or less than $2.5~\upmu\mathrm{m})$ in the atmosphere can damage the balance of energy on surface ground by the absorption of solar radiation and scattering effect and affect the earth’s climate system. In addition, it can reduce the atmospheric visibility. Therefore, $\operatorname{PM}_{2.5}$ is an important pollutant deteriorating the quality of atmospheric environment (Huang et al. 2014; Jung et al. 2009; Sokolik and Toon 1996; Wu et al. 2005). Haze, mainly related to elevated $\operatorname{PM}_{2.5}$ concentrations and the stable meteorological conditions, is defined as a weather phenomenon leading to atmospheric visual range less than $10\;\mathrm{km}$ at a relative humidity lower than $90~\%$ (Tian et al. 2014; Yuan et al. 2015). In recent years, China is suffering from severe haze pollution due to the intensive emission of air pollutants, in which Jing-Jin-Tang Region, Yangtze River Delta, Pearl River Delta and Sichuan Basin are four regions with the heaviest haze pollution (Cheng et al. 2013; Tao et al. 2013, 2014; Tan et al. 2016). The chemical composition of $\operatorname{PM}_{2.5}$ is complex, which can produce different effects to human health (Qiao et al. 2014; Watson 2002). Researches on the chemical characteristics of $\operatorname{PM}_{2.5}$ is to explain the source, generation, migration change rule and its effect on biological and to provide a scientific basis (Zhang et al. 2013; Cao et al. 2007; Tang et al. 2014; Schwarz et al. 2012; Borge et al. 2007).
Xinxiang (35.18N, 113.52E), located in the North China Plain, is about $111~\mathrm{km}$ south of Jing-Jin-Tang region and $80~\mathrm{km}$ north of Zhengzhou—the provincial capital of Henan Province. The characteristics of monsoon in Xinxiang are significant with the prevailing northeast winds. Xinxiang is one of the fastest industrializing and urbanizing regions in Henan Province; with the rapid economic development, population expansion and urbanization, Xinxiang has experienced ever-increasing energy consumption and the sharp increase in motor vehicles (in 2014, the total energy consumption reached $7.02\,\times\,1010\mathrm{~kW~h~}$ h in Xinxiang, GDP growth rate in Xinxiang was $9.3~\%$ . In addition, the total number of motor vehicles had reached 571,000, the urbanization rate was $47.6\ \%$ , and the total population reached 6.04 million with the natural population growth rate of $5.38\ \%$ in Xinxiang); the fine particulate matter pollution has become an important problem of atmospheric pollution. In addition to local emissions, input of $\operatorname{PM}_{2.5}$ may accompany the northeast monsoon via long-range transport from the heaviest polluted region (Jing-Jin-Tang). As a result, haze has been observed with increasing days in Xinxiang, especially during winter episode. Therefore, the investigation of atmospheric $\operatorname{PM}_{2.5}$ in winter haze period is necessary and urgent for the protection of human health.
To sum up, this study aims to: (1) obtain the $\operatorname{PM}_{2.5}$ mass concentrations at Xinxiang during heavy haze period in 2014 and compare the differences between haze and clear days; (2) characterize in detail the chemical compositions of atmospheric $\operatorname{PM}_{2.5}$ ; (3) identify possible sources of $\operatorname{PM}_{2.5}$ ; (4) calculate the cancer risks of $\operatorname{PM}_{2.5}$ during the whole period.
Methods
Study area and sampling
The sampling site is located in Henan Normal University, which is in a suburban area of northeastern Xinxiang $(35^{\circ}19^{\prime}29^{\prime\prime}\mathrm{N},$ , $113^{\circ}54^{\prime}27^{\prime\prime}\mathrm{E})$ (Fig. 1). The sampling site is about $30\;\mathrm{m}$ height from the ground. The campus is surrounded by residential communities and business district. In addition, about $150\ \mathrm{m}$ at the south direction of sampling site is a heavy traffic of Construction Road. Xinxiang meteorological observation station is about $850\;\mathrm{m}$ in the southeast of sampling site. In addition, no industrial air pollution sources exist around $2~\mathrm{km}$ range. The $\operatorname{PM}_{2.5}$ samples were collected from 26 January 2014 to 15 February 2014. Each sample was conducted from 9:00 in the morning to 9:00 in the next morning using a high-volume $\operatorname{PM}_{2.5}$ sampler (Multistage Versatile Air Pollutant Sampler, $\mathrm{PM}_{2.5}.$ - PUF-300, China) with a flow rate of $300~\mathrm{L~min}^{-1}$ . A total of 17 samples in 2014 were collected on quartz fliters (Tissu quartz 2500 qat-up, Pallflex membrane fliters, USA), which was pre-combusted in a muffle furnace at $450~^{\circ}\mathrm{C}$ for $4~\mathrm{h}$ to remove any contaminants on the filters before sampling. Two blank samples were collected in each sampling period, which were used to correct for any positive artifacts. All of the samples were properly stored in a freezer at $-20~^{\circ}\mathrm{C}$ to prevent any loss of volatiles. The hourly online $\operatorname{PM}_{2.5}$ mass concentrations for Xinxiang were obtained from Xinxiang Air Quality Release System (http://218.28.71.220: 9875/#).
$\operatorname{PM}_{2.5}$ mass
For $\operatorname{PM}_{2.5}$ mass determinations, the fliters were weighed before and after the samples were collected under controlled environment with temperature of $20{-}23\ ^{\circ}\mathrm{C}$ and relative humidity of $40\pm1\;\%$ . The filters were weighed with a Sartorius BS-124S electronic microbalance $(\pm1~\upmu\mathrm{g}$ sensitivity; Sartorius, G¨ottingen, Germany). The difference among the repeated weightings was less than $15~\upmu\mathrm{g}$ . After weighing, the filters were kept in aluminum foil and stored in the refrigerator at $-18~^{\circ}\mathrm{C}$ until chemical analysis.
Chemical analysis
Water-soluble ions
An area of $8.06~\mathrm{cm}^{2}$ of each quartz filter sample was cut to analyze water-soluble ions. The cut piece was put into $5\mathrm{-mL}$ polypropylene pipe and extracted three times with $5~\mathrm{mL}$ ultrapure water ( $^{18.2}\mathrm{~M}\Omega\;\mathrm{cm}$ under sonication for $30\ \mathrm{min}$ in an ultrasonic ice water bath. The total extract solution was filtered through a PTFE syringe filter (pore size $0.45~{\upmu\mathrm{m}})$ and then diluted with ultrapure water to $50~\mathrm{mL}$ , which were stored in a lowdensity polyethylene bottle until analysis. Four anions $\left(\mathrm{F}^{-}$ , $\mathrm{Cl}^{-}$ , $\mathrm{NO}_{3}^{\mathrm{~-~}}$ and $\mathrm{SO}_{4}^{\ 2-},$ ) in the aqueous extract were measured with a Dionex ICS-1100 (Thermo Fisher Scientific, USA), and five cations $(\mathrm{Na}^{+}$ , $\mathbf{K}^{+}$ , $\mathrm{NH_{4}}^{+}$ , $\mathrm{{Ca}}^{2+}$ and $\mathrm{Mg}^{2+}$ ) were analyzed using Dionex DX-600 (Thermo Fisher Scientific, USA).
Trace elements
An area of $8.06~\mathrm{cm}^{2}$ of each filter loading $\operatorname{PM}_{2.5}$ was placed in Teflon digestion vessel for acid treatment. Each sample was treated with $20{-}50~\mathrm{mL}$ acidic mixture of $\mathrm{HNO}_{3}/\mathrm{HF/HClO}_{4}$ (volumetric ratio of 4:4:1) at $200~^{\circ}\mathbf{C}$ for $24~\mathrm{h}$ to decompose the sample. The digested solution after drying was diluted to $100~\mathrm{mL}$ with $1\ \%$ $\mathrm{HNO}_{3}$ solution. The aliquot was analyzed using inductively coupled plasma mass spectrometer (ICP-MS) (DRC-e PerkinElmer, USA). Nineteen elements, i.e., Pb, Fe, Cr, Co, Ni, Cu, Zn, As, Cd, Li, Na, Mg, Al, K, Ca, Ti, V, Mn and Sr, were measured.
Organic components (EC/OC) analysis
An area of $0.51~\mathrm{cm}^{2}$ punch area of each quartz filter was taken to detect the content of elemental carbon (EC) and organic carbon (OC) by an organic carbon/ elemental carbon analyzer (DRI Model 2001A Atmoslytic Inc, USA). Each sample was heated to produce four OC fractions: OC1, OC2, OC3 and OC4 at temperatures of 120, 250, 450 and $550~^{\circ}\mathrm{C}$ in a nonoxidizing He atmosphere; three EC fractions: EC1, EC2 and EC3 at 550, 700 and $800~^{\circ}\mathrm{C}$ in an oxidizing atmosphere of $2~\%$ $\mathrm{O}_{2}/98~\%$ He; and optically detected pyrolyzed carbon (OPC). OC is defined as $\mathrm{OC1}+\mathrm{OC2}+\mathrm{OC3}+\mathrm{OC4}+\mathrm{OPC}$ , and EC is calculated by $\mathrm{EC1}+\mathrm{EC2}+\mathrm{EC3}{\cdot}\mathrm{OPC}$ .
Quality assurance (QA) and quality control (QC)
All standard solutions of water-soluble ions and trace elements were obtained from J&K Scientific Ltd. (Beijing, China). Before targeted samples analysis, standard solution and blank test were performed and the correlation coefficients of standard samples were above 0.999. The method detection limits of $\mathrm{F}^{-}$ , $\mathrm{Cl^{-}}$ , $\mathrm{NO}_{3}^{\mathrm{~-~}}$ , $\mathrm{SO}_{4}{}^{2-}$ , $\mathrm{Na}^{+}$ , $\mathbf{K}^{+}$ , $\mathrm{NH_{4}}^{+}$ , $\mathrm{{Ca}}^{2+}$ and ${\mathrm{Mg}}^{2+}$ were 0.026, 0.058, 0.013, 0.010, 0.013, 0.087, 0.048, 0.084 and $0.067~\mathrm{mg\,L^{-1}}$ , respectively. The relative standard deviations between real values of standard materials and analyzing results were in the range of $2{-}15\ \%$ , and the detection limits ranged from 0.00001 to $0.0005~\upmu\mathrm{g}$ ${\boldsymbol{\mathrm{L}}}^{-1}$ for trace elements. For carbonaceous species, one in every ten samples was detected twice and the precision was less than $1\ \%$ . Standard concentrations of $\mathrm{CH}_{4}/\mathrm{CO}_{2}$ mixed gases were used for calibrating the analyzer in each day before and after sample analysis. All the reported data of water-soluble ions, trace elements and carbonaceous species were corrected by the filter blanks.
Meteorological parameters
The meteorological parameters including relative humidity (RH), wind speed (WS), wind direction (WD), visibility (Vs), temperature (T) and precipitation were recorded by the meteorological observatory base of Xinxiang. The WS and WD during sampling period are shown in Fig. 2.
Back-trajectory calculation
Although 72-h air mass back-trajectories were widely used in many published studies, $96\mathrm{-h}$ or more than 96-h air mass back-trajectories could provide more detailed information about air mass trajectories than
72-h air mass back-trajectories (Dimitriou and Kassomenos 2014; Schwarz et al. 2016; Zhou et al. 2016). According to these published researches (Dimitriou and Kassomenos 2014; Schwarz et al. 2016; Zhou et al. 2016), 96-h air mass back-trajectories were calculated using NOAA Air Resource Lab HYSPLIT 4 model, where the initial height of $100\mathrm{~m~}$ above ground level was used (Huang et al. 2012). Cluster analysis was applied which could objectively result in subsets of backward trajectories computed 00:00, 06:00, 12:00 and 18:00 UTC in each day. In addition, a time series of mixing layer height (MLH) was calculated using the chosen meteorological data by the NOAA’s READY Archived Meteorology online calculating program (http://ready.arl.noaa.gov/ READYamet.php).
Health risk assessment of $\operatorname{PM}_{2.5}$
Heavy metals in $\operatorname{PM}_{2.5}$ can raise risks to human health. To reveal extra health threats from $\operatorname{PM}_{2.5}$ , the average amount of heavy metals exposure by inhalation $(D_{\mathrm{inh}})$ for adults and children based on individual’s body weight during a given time span was calculated using Eq. (1) (Kong et al. 2012; Granero and Domingo 2002):
$$
D_{\mathrm{inh}}={\frac{C\times\mathrm{InhR}\times\mathrm{EF}\times\mathrm{ED}}{\mathrm{BW}\times\mathrm{AT}}}
$$
The lifetime average daily dose (LADD) of heavy metals expo sure through inhalation was used fo r assessing health risks as follows [Eq. (2)]:
$$
\begin{array}{r l}{\mathrm{LADD}\displaystyle=\frac{C\times\mathrm{EF}}{\mathrm{AT}}}&{}\\ {\times\left(\frac{\mathrm{InhR}_{\mathrm{child}}\times\mathrm{ED}_{\mathrm{child}}}{\mathrm{BW}_{\mathrm{child}}}\!+\!\frac{\mathrm{InhR}_{\mathrm{adult}}\times\mathrm{ED}_{\mathrm{adult}}}{\mathrm{BW}_{\mathrm{adult}}}\right)}\end{array}
$$
where $D_{\mathrm{inh}}$ is the exposure by respiratory inhalation, $\mathrm{mg~kg^{-1}~d a y^{-1}}$ ; InhR is inhalation rate, 7.6 and $20~\mathrm{m}^{3}$ $\mathrm{day^{-1}}$ for children and adults, respectively; EF is exposure frequency, day year ; ED is exposure duration, 6 year for children and 24 year for adults, respectively; BW is average body weight, $15~\mathrm{kg}$ for children and $70~\mathrm{kg}$ for adults; AT is the averaging time, for non-cancer toxic risks, AT (days) $=\mathrm{ED~}\times$ 365; for cancer risk, AT $\mathrm{(days)}=70\,\times\,365$ ; and $C$ is exposure point concentration, which is calculated by upper limit of the $95~\%$ confidence interval for the mean, mg $\mathrm{m}^{-3}$ (Kong et al. 2015).
After $D_{\mathrm{inh}}$ was calculated, Hazard Quotient (HQ) was calculated to represent the non-cancer toxic risk using EXqs. (3, 4):
$$
\mathrm{HQ}={\frac{D_{\mathrm{inh}}}{R_{\mathrm{fD}}}}
$$
$$
\mathrm{HI}=\sum{\mathrm{HQ}_{i}}
$$
where $R_{\mathrm{fD}}$ is the reference dose, $\mathrm{mg\kg^{-1}\,d a y^{-1}}$ ; $\mathrm{HI}$ is hazard index, which can be obtained by summing up the individual HQ to estimate the total risks of all heavy metals. $R_{\mathrm{fD}}$ values are $3.52\,\times\,10^{-3}$ , $2.86~\times$ $10^{-5}$ , $5.71\,\times\,10^{-6}$ , $2.06\,\times\,10^{-2}$ , $3.01\,\times\,10^{-1}$ , $3.01\,\times\,10^{-4}$ , $1\,\times\,10^{-3}$ , $7\,\times\,10^{-3}$ and $1.4\,\times\,10^{-5}$ for Pb, Cr, Co, Ni, Zn, As, Cd, $\mathrm{v}$ and Mn (Ministry of Environmental Protection of the People’s Republic of China 2014). If $\mathrm{HI}\leq1$ , there are no adverse health effects; if $\mathrm{HI}>1$ , possibly adverse health effects exist.
For cancer risk, it was calculated using Eqs. (5, 6):
$$
\begin{array}{l}{R=\mathrm{LADD\timesSF}_{a}}\\ {\displaystyle R_{t}=\sum R}\end{array}
$$
where $\mathrm{SF}_{a}$ is slope factor, $\mathrm{mg~kg^{-1}~d a y^{-1}}$ . The $\mathrm{SF}_{a}$ values of Cr, Co, Ni, As and Cd are 42, 9.8, 0.84, 15.1 and 6.4. An internationally accepted precautionary or threshold value for cancer risk is $10^{-6}$ , above which the risk is unacceptable (Feng et al. 2012; Kulshrestha et al. 2004; Wang et al. 2007).
Results and discussion
Mass concentration of $\operatorname{PM}_{2.5}$
Figure 3 illustrates the temporal variation of daily average $\operatorname{PM}_{2.5}$ concentrations in Xinxiang from 26 January to 15 February in 2014. As shown in Fig. 3, 24-h average $\mathrm{PM}_{2.5}$ mass concentrations range from 35.9 (February 4, 2014) to $345~\upmu\mathrm{g}~\mathrm{m}^{-3}$ (January 27, 2014) with a mean value of $168.5~\upmu\mathrm{g}~\mathrm{m}^{-3}$ . According to Ambient Air Quality Standards (GB3095-2012) issued in 2012 by the Chinese Ministry of Environmental Protection, $88.2~\%$ of the observed $\mathrm{PM}_{2.5}$ mass concentrations exceed grade II $(75~\upmu\mathrm{g}\textrm{m}^{-3})$ , let alone the air quality guideline recommended by the World Health Organization $(25~\upmu\mathrm{g}\textrm{m}^{-3}$ for 24-h average of $\mathsf{P M}_{2.5},$ ) (WHO 2006). Compared with other Chinese urban centers, this level is higher than those in Shanghai, Nanjing, Jinan and Tianjin, lower than those in Yellow River Delta and close to those in Zhengzhou (Table 1). The foregoing shed light on the severity of $\operatorname{PM}_{2.5}$ pollution in Xinxiang and the urgency of macro-control by the government.
During the observation period, $\operatorname{PM}_{2.5}$ concentrations peaked on 27 January $(345~\upmu\mathrm{g}\textrm{m}^{-3})$ . The relatively low wind speed $(2.5\mathrm{~m~s~}^{-1})$ ) facilitated the formation of stagnant atmosphere and low mixed boundary layer $\left(308\mathrm{~m}\right)$ , which were unfavorable for the dispersion of pollutants and resulted in the accumulation of $\operatorname{PM}_{2.5}$ in the atmosphere. On February 4 and February 9, 2014, $\operatorname{PM}_{2.5}$ concentrations were 35.9 and $62.6~\upmu\mathrm{g}~\mathrm{m}^{-3}$ , lower than the secondary standard of China National Ambient Air Quality. During these 2 days, the wind speed was 5.4 and $4.4~\mathrm{m~s}^{-1}$ and the mixing layer height were 736 and $415\mathrm{~m~}$ . Due to the relatively higher boundary layer and high wind speed, the diffusion rates of $\operatorname{PM}_{2.5}$ were faster which cause relatively low levels of $\operatorname{PM}_{2.5}$ . In addition, a light snow occurred on February 4, 2014, which could further explain the lower $\operatorname{PM}_{2.5}$ concentrations since precipitation scavenging was an efficient way of removing $\operatorname{PM}_{2.5}$ from the atmosphere. Published researches indicated positive correlation between $\operatorname{PM}_{2.5}$ concentration and meteorological factors (Li et al. 2009; Sun et al. 2013; Dawson et al. 2007; Wang and Ogawa 2015). However, in this study, no correlations were found between $\operatorname{PM}_{2.5}$ concentrations and meteorological parameters. Wu et al. (2014a, b) revealed that for samples collected in a single period, the influence of meteorological parameters was less important than other factors such as emission and dispersion.
Characterization of chemical species
Water-soluble ions characteristics
Average concentrations of water-soluble ions, heavy metals, OC and EC detected in $\operatorname{PM}_{2.5}$ of Xinxiang during 2014 winter are listed in Table 2. Apparently, $\mathrm{SO}_{4}{}^{2-}$ , $\mathrm{NO}_{3}^{\mathrm{~-~}}$ , $\mathrm{NH_{4}}^{+}$ and $\mathrm{Cl}^{-}$ were the dominant ions in $\operatorname{PM}_{2.5}$ with their average concentrations being 39.78, 32.25, 13.17 and $11.66~\upmu\mathrm{g}\,\,\mathrm{m}^{-3}$ , respectively.
The average concentrations for the other five ions ranged from 0.38 to $4.42~\upmu\mathrm{g}\,\,\mathrm{m}^{-3}$ . The concentrations of the water-soluble ions on haze days were two times of those on clear days. $\mathrm{SO}_{4}{}^{2-}$ presented in $\operatorname{PM}_{2.5}$ of Xinxiang was associated with coal combustion, which was still one of the major energy sources in Xinxiang, both in industry and home heating. The abundance of $\mathrm{NO}_{3}^{\mathrm{~-~}}$ could likely be due to the greater impact of traffic emissions of Xinxiang, because NOx, precursors of $\mathrm{NO}{_3}^{-}$ , was mainly emitted from mobile sources. $\mathrm{NH_{4}}^{+}$ was formed largely through reactions of $\mathrm{NH}_{3}$ with $\mathrm{HNO}_{3}$ , $\mathrm{H}_{2}\mathrm{SO}_{4}$ and their precursors. $\mathrm{NH}_{3}$ was mainly emitted from anthropogenic sources, especially agricultural activities and traffic (Asman et al. 1998; Ianniello et al. 2010; Meng et al. 2011). Due to the shrinking of agricultural activity in winter, the higher concentrations of $\mathrm{NH_{4}}^{+}$ in winter were associated with traffic. In addition, ratios of $\mathrm{NO}_{3}^{\mathrm{~-}}/$ $\mathrm{SO}_{4}{}^{2-}$ can be reasonably used to evaluate the contribution of mobile and stationary sources to sulfur and nitrogen in the atmosphere in China $\mathrm{\DeltaXu}$ et al. 2012). The mass ratios of $\mathrm{NO}_{3}^{\mathrm{~-}}/\mathrm{SO}_{4}^{\mathrm{~2-}}$ during the sampling periods were in the range of 0.56–3.42, with an average value of $0.96\pm\:0.69$ . The ratio of $\mathrm{NO}_{3}{}^{-}/$ $\mathrm{SO}_{4}{}^{2-}$ on haze days (1.01) was about two times than those on clear days (0.56), which suggested that the impact of vehicle exhaust should not be neglected under the rapid increase in motors in the urban area, despite the coal burning making a major contribution to particulate pollutants in China.
$\mathrm{F}^{-}$ , $\mathrm{Cl^{-}}$ , $\mathrm{NO}_{3}^{\mathrm{~-~}}$ and $\mathrm{SO}_{4}^{\ 2-}$ are favorable for the acidity of aerosol, and $\mathrm{NH_{4}}^{+}$ , $\ K^{+}$ , $\mathrm{Na}^{+}$ and ${\mathrm{Mg}}^{2+}$ can increase the $\mathrm{pH}$ of air particles ( $\mathrm{Xu}$ et al. 2012). The calculated anion equivalents (AE) and cation equivalents (CE) with the following Eqs. (7, 8) to determine the ion balance are good indicator to study the acidity of aerosol:
$$
\begin{array}{l}{{\mathrm{AE}={\frac{[{\bf S O}_{4}^{-}{\bf\bar{\Xi}}]}{48}}+{\frac{[{\bf N O}_{3}^{-}{\bf\bar{\Xi}}]}{62}}+{\frac{[{\bf C}1^{-}{\bf\Xi}]}{35.5}}+{\frac{[{\bf F}^{-}{\bf\Xi}]}{19}}}}\\ {{\mathrm{CE}={\frac{[{\bf N a}^{+}{\bf\Xi}]}{23}}+{\frac{[{\bf N H}_{4}^{+}{\bf\Xi}]}{18}}+{\frac{[{\bf K}^{+}{\bf\Xi}]}{39}}+{\frac{[{\bf M g}^{2+}{\bf\Xi}]}{12}}+{\frac{[{\bf C a}^{2+}{\bf\Xi}]}{20}}}}\end{array}
$$
The ratios calculated from all the measured ionic species ranged from 0.88 to 1.73. The mean ratio was 1.43 (close to 1), indicating that almost all of the components had been quantified (Wang et al. 2005; Xu et al. 2012). The anion equivalents were plotted against the cation equivalents, and the slope of the regression line was 0.73 (below 1, $r=0.73)$ ). The result suggested that the concentrations of $\mathrm{H^{+}}$ are not counted in calculation due to parts of $\mathrm{NH_{4}}^{+}$ that was probably vaporized into the gas phase. The ratios of AE/CE was high, indicating that aerosol in Xinxiang was strongly acidic.
Trace elements characteristics
In this study, 19 elements were detected. The total concentrations of the crustal elements (Al, Ca, Fe, $\mathbf{M}\mathrm{g}$ , K and $\mathrm{Na},$ varied from 6.67 to $30.22~\upmu\mathrm{g}\textrm{m}^{-3}$ accounting for $53{-}88\ \%$ of total elements. Additionally, the other thirteen trace elements $Z\mathfrak{n}$ , As, Pb, V, Ti, Cr, Mn, Ni, Sr, Cu, Li, Cd and $\mathrm{Co}$ ) accounted for a small part $(12{-}47\ \%)$ ) of the total elements in $\operatorname{PM}_{2.5}$ of Xinxiang. The concentrations of trace elements in $\operatorname{PM}_{2.5}$ occurred in the descending order: Zn, Ti, Pb, V, Cr, As, Mn, Ni, Sr, Cu, Cd, Li and Co. In addition, trace elements concentrations on haze days were 1.8 times of those on clear days.
The enrichment factors (EF) could identify origins and eva luat e the degree of anthropogenic influence. The EF equation is as follows (Hsu et al. 2010):
$$
\mathrm{EF_{i}}=\frac{\left(\frac{C_{\mathrm{i}}}{C_{\mathrm{r}}}\right)_{\mathrm{aerosol}}}{\left(\frac{C_{\mathrm{i}}}{C_{\mathrm{r}}}\right)_{\mathrm{crust}}}
$$
where $C_{\mathrm{i}}$ is the concentration of heavy metal i and $C_{\mathrm{r}}$ is the reference concentration of heavy metal. Al was selected as the reference metal in this study (Hsu et al. 2010; Hernandez et al. 2003; Rubio et al. 2000), and the reference concentrations of crust were obtained from Wei et al. (1990). During the whole period, the mean EF for metals were in the range of 0.0003–5.47 (Fig. 4); most of metals such as Pb, Cr, Co, Ni, Cu, Zn, As, Cd, Sr, Li, K, V and Mn were lower than 1, which suggested that they were identified to be exclusive of crustal origin. Other metals such as Fe, Na, Mg, Ca and Ti were in the range of 1–10, which indicated that these metals were emitted from crustal source and anthropogenic emission (Hsu et al. 2010).
EC and OC characteristics
OC during haze days was 2.8 times of those on clear days, while EC during haze days was 2.3 times of those on clear days. OC and EC were higher on haze days, characterized by extremely high traffic flows (Huang et al. 2012). Decreasing OC and EC on clear days were related to the weakened sources, like vehicle emission and industrial activities dropped. Daily OC and EC are plotted in Fig. 5, which displayed that temporal variations of OC and EC coincided with each other ( ${\textsl{r}}=0.87$ , $p<0.001]$ . In addition, daily variations of OC/EC ratios in $\operatorname{PM}_{2.5}$ show relative low ratios and variability on clear days and relative high ratios and variability on haze day. OC was significantly correlated with $\operatorname{PM}_{2.5}$ mass $(r=0.856,p<0.001)$ , and, to a lesser extent, EC was correlated with $\operatorname{PM}_{2.5}$ mass as well $(r=0.798$ , $p<0.001)$ . The maximum OC concentration was $44.8~\upmu\mathrm{g}\textrm{m}^{-3}$ , which occurred on January 28, 2014, and the maximum EC ( $16.4~\upmu\mathrm{g}~\mathrm{m}^{-3})$ was recorded on the same day. The lowest concentration of OC $(7.37\ \upmu\mathrm{g}\ \mathrm{m}^{-3})$ was observed on 4 February, whereas the minimum EC $(2.11~\upmu\mathrm{g}\textrm{m}^{-3})$ was on 5 February. The maximum-to-minimum ratios were 6.07 for OC and 7.76 for EC, which indicated that the variability of EC was greater than that of OC.
The ratio of OC to EC could give some indication of the origins of carbonaceous $\mathrm{PM}_{2.5}$ (Gray et al. 1986; Turpin and Huntzicker 1991; Chow et al. 1996). The underlying hypothesis was that EC originated directly from primary anthropogenic sources, while OC might be emitted directly from sources as primary particles, but also could be formed from low vapor pressure products via atmospheric chemical reactions. Daily
OC/EC ratios were plotted in Fig. 4, which ranged from 2.43 to 4.53, with the highest ratio recorded on 5 February. In addition, daily variations of $\mathrm{PM}_{2.5}$ OC/ EC ratios showed lower ratios and variability on clear days and higher ratios and variability on haze day. Based on the published studies, OC/EC ratios ranged from 0.3 to 7.6 from coal combustion, 0.7–2.4 from vehicle emission and 4.1–14.5 from biomass burning (Watson 2002). All of OC/EC ratios in this study were above 2.0 for $\operatorname{PM}_{2.5}$ , which might reflect the combined contributions from coal combustion, motor vehicle exhaust and biomass burning.
Air mass backward trajectory analysis
Mean 96-h backward trajectory clusters at Xinxiang in the sampling period are shown in Fig. 6. During these periods, $45\ \%$ (cluster 1 and cluster 2) was from Mongolia and transported across Beijing and Hebei Province; $30~\%$ (cluster 3) was from Tibet and transported across Shanxi Province; $25~\%$ (cluster 4) was from Central China and transported across Zhengzhou. Air masses backward trajectories for cluster 4 were short, indicating air masses move slowly and pollutants were readily to accumulate due to intensive anthropogenic sources for air pollutants in the Central China. More serious air pollution occurred when air masses transported from Shanxi, Hebei and Zhengzhou, where hold abundant anthropogenic sources in winter, such as increasing coal combustion for heating. The above results further revealed that the major pollutants parcel was mainly from northeast monsoon. Noteworthy, the highest $\operatorname{PM}_{2.5}$ concentrations occurred in the whole period because of the influence of monsoon.
Health risk assessment of heavy metals during the period
The results of average daily dose of heavy metals are shown in Table 3, which include the values of HQ and HI of non-carcinogenic metals in $\operatorname{PM}_{2.5}$ of Xinxiang. As indicated in Table 3, the risk levels of the noncarcinogenic heavy metals for exposure through the respiratory system ranged from $6.18\,\times\,10^{-5}$ to 0.19 for children and from $3.48\,\times\,10^{-5}$ to 0.11 for adults, both were lower than the acceptance risk of 1. The risk levels for the non-carcinogenic heavy metals occurred in the following order: $\mathrm{Zn}>\mathrm{Pb}>\mathrm{Cr}>\mathrm{V}>\mathrm{As}$ $>\mathrm{Mn}>\mathrm{Ni}>\mathrm{Cd}>\mathrm{Co}$ . The sum of the risk levels $\mathrm{(HI)}$ posed by the nine heavy metals was 0.38 for children and 0.21 for adults, respectively. It was readily known that the HI was above 0.1 but below 1, which suggested that non-cancerous effects of heavy metals in $\operatorname{PM}_{2.5}$ of Xinxiang were unlikely happened. In addition, non-carcinogenic metals could cause harm to children more easily than to adult, which indicated that children were more sensitive to non-cancer effects than adults and should be kept from possible exposure to them (Yang et al. 2014).
The LADD and the risk caused by As, Cd, Cr, Ni and Co via inhalation are shown in Table 3. It could be seen that the risk levels for the carcinogenic heavy metals occurred in the following order: $\mathrm{Cr}>\mathrm{As}>\mathrm{Cd}>\mathrm{Ni}>\mathrm{Co}.$ . The carcinogenic risks of Cr, As and Cd surpassed $10^{-6}$ , while those of $\mathrm{Ni}$ $(4.47\,\times\,10^{-7})$ and Co $(2.36\,\times\,10^{-7})$ were lower than $10^{-6}$ . Any cancer risk less than the threshold value of $10^{-6}$ was considered negligible by the US EPA. From these results, the lifetime cancer risks of $\mathrm{Cr}$ , As and Cd obviously exceeded the threshold and might have a cancer risk for residents in Xinxiang.
Conclusions
A total of 19 elements, nine water-soluble ions and eight fractions of carbonaceous species were measured to fully characterize the chemical compositions and to apportion potential sources of atmospheric $\operatorname{PM}_{2.5}$ in Xinxiang during winter in 2014. Serious regional haze pollution occurred and lasted for almost whole sampling period, in which $\operatorname{PM}_{2.5}$ concentrations $\left(168.5~\upmu\mathrm{g}\mathrm{~m}^{-3}\right.$ on average) exceeding the China’s Ambient Air Quality Standard (AAQS) (BG3095-12) grade $\mathrm{II}$ were $88.2~\%$ , respectively. The concentrations of water-soluble ions and trace elements on haze days were 2 and 1.8 times of those on clear days. Vehicular exhaust was an important contributor to the atmospheric pollution due to the relatively high $\mathrm{NO}_{3}^{\mathrm{~-}}/$ $\mathrm{SO}_{4}{}^{2-}$ ratios. Aerosol in Xinxiang was heavily acidic with the AE/CE ratio over 1 in the sampling periods. Fe, Na, Mg, Ca and Ti were in the range of 1–10, which suggested that these heavy metals were emitted from crustal source and anthropogenic emission. Health risk assessment revealed that non-cancerous effects of heavy metals in $\mathrm{PM}_{2.5}$ of Xinxiang were unlikely occurred, while lifetime cancer risks of heavy metals obviously exceeded the threshold and might have a cancer risk for residents in Xinxiang. This study provides an intuitional understanding of $\operatorname{PM}_{2.5}$ in Xinxiang and gives us clues to mitigate air pollution in this area. Enlarging degree of pollution prevention and treatment in Hebei and Zhengzhou is helpful to improve air quality in Xinxiang.
Acknowledgments This study was supported by the National Scientific Foundation of China (Grant No. 41103071), Program for Science and Technology Innovation talents in universities of Henan Province (14HASTIT049) and Foundation for University Key Teacher by Henan Province (2013GGJS-059), Key Project of Science and Technology in Henan Province (152102310316) and Program for Science and Technology Development in Xinxiang (15SF02).
References
Asman, W. A. H., Sutton, M. A., & Schjorring, J. K. (1998). Ammonia: Emission, atmospheric transport and deposition. New Phytologist, 139(1), 27–48.
Borge, R., Lumbreras, J., Vardoulakis, S., Kassomenos, P., & Rodriguez, E. (2007). Analysis of long-range transport influences on urban PM10 using two-stage atmospheric trajectory clusters. Atmospheric Environment, 41(21), 4434–4450.
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. doi:10.1029/2006JD008205.
Cheng, Z., Wang, S., Jiang, J., Fu, Q., Chen, C., Xu, B., et al. (2013). Long-term trend of haze pollution and impact of particulate matter in the Yangtze river delta, China. Environmental Pollution, 182, 101–110.
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(12), 2079–2112.
Dawson, J. P., Adams, P. J., & Pandis, S. N. (2007). Sensitivity of $\mathrm{PM}_{2.5}$ to climate in the eastern US: A modeling case study. Atmospheric Chemistry and Physics, 7(16), 4295–4309.
Dimitriou, K., & Kassomenos, P. (2014). Indicators reflecting local and transboundary sources of $\mathrm{PM}_{2.5}$ and PMCOARSE in Rome-impacts in air quality. Atmospheric Environment, 96, 154–162.
Feng, J., Sun, P., Hu, X., Zhao, W., Wu, M., & Fu, J. (2012). The chemical composition and sources of $\operatorname{PM}_{2.5}$ during the 2009 Chinese new year’s holiday in Shanghai. Atmospheric Research, 118, 435–444.
Granero, S., & Domingo, J. L. (2002). Levels of metals in soils of Alcala´ de Henares, Spain. Environment International, 28(3), 159–164.
Gray, H. A., Cass, G. R., Huntzicker, J. J., Heyerdahl, E. K., & Rau, J. A. (1986). Characteristics of atmospheric organic and elemental carbon particle concentrations in Los Angeles. Environmental Science and Technology, 20(6), 580–589.
Hernandez, L., Probst, A., Probst, J. L., & Ulrich, E. (2003). Heavy metal distribution in some french forest soils: Evidence for atmospheric contamination. Science of the Total Environment, 312(1–3), 195–219.
Hsu, S. C., Liu, S. C., Tsai, F., Engling, G., Lin, I. I., Chou, C. K. C., et al. (2010). High wintertime particulate matter pollution over an offshore island (Kinmen) off Southeastern China: An overview. Journal of Geophysical Research, 115(D17), 1383–1392.
Huang, R.-J., Zhang, Y., Bozzetti, C., Ho, K.-F., Cao, J.-J., Han, Y., et al. (2014). High secondary aerosol contribution to particulate pollution during haze events in China. Nature, 514(7521), 218–222.
Huang, K., Zhuang, G., Lin, Y., Wang, Q., Fu, J. S., Zhang, R., et al. (2012). Impact of anthropogenic emission on air quality over a megacity—Revealed from an intensive atmospheric campaign during the Chinese spring festival. Atmospheric Chemistry and Physics, 12(23), 11631–11645.
Ianniello, A., Spataro, F., Esposito, G., Allegrini, I., Rantica, E., Ancora, M. P., et al. (2010). Occurrence of gas phase ammonia in the area of Beijing (China). Atmospheric Chemistry and Physics, 10(19), 9487–9503.
Jung, J., Lee, H., Kim, Y. J., Liu, X., Zhang, Y., Gu, J., et al. (2009). Aerosol chemistry and the effect of aerosol water content on visibility impairment and radiative forcing in Guangzhou during the 2006 Pearl River delta campaign. Journal of Environmental Management, 90(11), 3231–3244.
Kong, S. F., Li, L., Li, X. X., Yin, Y., Chen, K., Liu, D. T., et al. (2015). The impacts of firework burning at the Chinese spring festival on air quality: Insights of tracers, source evolution and aging processes. Atmospheric Chemistry and Physics, 15(4), 2167–2184.
Kong, S., Lu, B., Ji, Y., Zhao, X., Bai, Z., Xu, Y., et al. (2012). Risk assessment of heavy metals in road and soil dusts within $\mathrm{PM}_{2.5}$ , $\mathrm{{PM}_{10}}$ and $\mathrm{{PM}_{100}}$ fractions in Dongying city, Shandong province, china. Journal of Environmental Monitoring, 14(3), 791.
Kulshrestha, U. C., Nageswara Rao, T., Azhaguvel, S., & Kulshrestha, M. J. (2004). Emissions and accumulation of metals in the atmosphere due to crackers and sparkles during Diwali festival in India. Atmospheric Environment, 38(27), 4421–4425.
Li, X., Guo, X., Liu, X., Liu, C., Zhang, S., & Wang, Y. (2009). Distribution and sources of solvent extractable organic compounds in PM2.5 during, 2007 Chinese spring festival in Beijing. Journal of Environmental Sciences, 21(2), 142–149.
Li, W., Shi, Z., Yan, C., Yang, L., Dong, C., & Wang, W. (2013). Individual metal-bearing particles in a regional haze caused by firecracker and firework emissions. Science of the Total Environment, 443, 464–469.
Liu, J., Man, Y., & Liu, Y. (2014). Temporal variability of PM10 and PM2.5 inside and outside a residential home during, 2014 Chinese spring festival in Zhengzhou, China. Natural Hazards, 73(3), 2149–2154.
Meng, Z. Y., Lin, W. L., Jiang, X. M., Yan, P., Wang, Y., Zhang, Y. M., et al. (2011). Characteristics of atmospheric ammonia over Beijing, China. Atmospheric Chemistry and Physics, 11(12), 6139–6151.
Ministry of Environmental Protection of the People’s Republic of China. (2014). The Chinese environmental protection standard: Technical guidelines for risk assessment of contaminated sites. Beijing: Chinese Environment Science Press.
Qiao, L., Cai, J., Wang, H., Wang, W., Zhou, M., Lou, S., et al. (2014). PM $2.5$ constituents and hospital emergency-room visits in Shanghai, China. Environmental Science and Technology, 48(17), 10406–10414.
Rubio, B., Nombela, M., & Vilas, F. (2000). Geochemistry of major and trace elements in sediments of the Ria de Vigo (NW Spain): An assessment of metal pollution. Marine Pollution Bulletin, 40(11), 968–980.
Schwarz, J., Cusack, M., Karban, J., Chalupnı´cˇkova´, E., Havra´nek, V., Smolı´k, J., et al. (2016). PM2.5 chemical composition at a rural background site in central Europe, including correlation and air mass back trajectory analysis. Atmospheric Research, 176–177, 108–120.
Schwarz, J., Sˇtefancova´, L., Maenhaut, W., Smolı´k, J., & Zˇdı´- mal, V. (2012). Mass and chemically speciated size distribution of Prague aerosol using an aerosol dryer—The influence of air mass origin. Science of the Total Environment, 437, 348–362.
Sokolik, I. N., & Toon, O. B. (1996). Direct radiative forcing by airborne mineral aerosols. Journal of Aerosol Science, 27, S11–S12.
Sun, Z., Mu, Y., Liu, Y., & Shao, L. (2013). A comparison study on airborne particles during haze days and non-haze days in Beijing. Science of the Total Environment, 456–457, 1–8.
Tan, J., Duan, J., Zhen, N., He, K., & Hao, J. (2016). Chemical characteristics and source of size-fractionated atmospheric particle in haze episode in Beijing. Atmospheric Research, 167, 24–33.
Tang, L., Haeger-Eugensson, M., Sjo¨berg, K., Wichmann, J., Molna´r, P., & Sallsten, G. (2014). Estimation of the longrange transport contribution from secondary inorganic components to urban background $\mathrm{{PM}_{10}}$ concentrations in south-western Sweden during 1986–2010. Atmospheric Environment, 89, 93–101.
Tao, J., Zhang, L., Engling, G., Zhang, R., Yang, Y., Cao, J., et al. (2013). Chemical composition of $\mathrm{PM}_{2.5}$ in an urban environment in Chengdu, China: Importance of springtime dust storms and biomass burning. Atmospheric Research, 122, 270–283.
Tao, J., Zhang, L., Ho, K., Zhang, R., Lin, Z., Zhang, Z., et al. (2014). Impact of $\operatorname{PM}_{2.5}$ chemical compositions on aerosol light scattering in Guangzhou—The largest megacity in south China. Atmospheric Research, 135–136, 48–58.
Tian, Y. Z., Wang, J., Peng, X., Shi, G. L., & Feng, Y. C. (2014). Estimation of the direct and indirect impacts of fireworks on the physicochemical characteristics of atmospheric $\mathrm{{PM}_{10}}$ and $\operatorname{PM}_{2.5}$ . Atmospheric Chemistry and Physics, 14(18), 9469–9479.
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. Part A. General Topics, 25(2), 207–215.
Wang, J., & Ogawa, S. (2015). Effects of meteorological conditions on $\mathrm{PM}_{2.5}$ concentrations in Nagasaki, Japan. International Journal of Environmental Research and Public Health, 12(8), 9089–9101.
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., Xu, C., & An, Z. (2007). The air pollution caused by the burning of fireworks during the lantern festival in Beijing. Atmospheric Environment, 41(2), 417–431.
Watson, J. G. (2002). Visibility: Science and regulation. Journal of the Air and Waste Management Association, 52, 628–713.
Wei, F. S., Chen, J. S., Wu, Y. Y., & Zheng, C. J. (1990). Introduction to background value of soil environment. Beijing: China Environmental Science Press.
WHO. (2006). World Health Organization. Geneva: Switzerland.
Wu, D., Tie, X., Li, C., Ying, Z., Kai-Hon Lau, A., Huang, J., et al. (2005). An extremely low visibility event over the Guangzhou region: A case study. Atmospheric Environment, 39(35), 6568–6577.
Wu, D., Wang, Z., Chen, J., Kong, S., Fu, X., Deng, H., et al. (2014a). Polycyclic aromatic hydrocarbons (PAHs) in atmospheric $\mathrm{PM}_{2.5}$ and $\mathrm{{PM}_{10}}$ at a coal-based industrial city: Implication for PAH control at industrial agglomeration regions, China. Atmospheric Research, 149, 217–229.
Wu, Y., Yang, L., Zheng, X., Zhang, S., Song, S., Li, J., et al. (2014b). Characterization and source apportionment of particulate PAHs in the roadside environment in Beijing. Science of the Total Environment, 470–471, 76–83.
Xu, L., Chen, X., Chen, J., Zhang, F., He, C., Zhao, J., et al. (2012). Seasonal variations and chemical compositions of $\mathrm{PM}_{2.5}$ aerosol in the urban area of Fuzhou, China. Atmospheric Research, 104–105, 264–272.
Yang, L., Gao, X., Wang, X., Nie, W., Wang, J., Gao, R., et al. (2014). Impacts of firecracker burning on aerosol chemical characteristics and human health risk levels during the Chinese new year celebration in Jinan, China. Science of the Total Environment, 476–477, 57–64.
Yuan, Q., Li, W., Zhou, S., Yang, L., Chi, J., Sui, X., et al. (2015). Integrated evaluation of aerosols during haze-fog episodes at one regional background site in north China Plain. Atmospheric Research, 156, 102–110.
Zhang, R., Jing, J., Tao, J., Hsu, S.-C., Wang, G., Cao, J., et al. (2013). Chemical characterization and source apportionment of $\operatorname{PM}_{2.5}$ in Beijing: Seasonal perspective. Atmospheric Chemistry and Physics, 13(14), 7053–7074.
Zhou, M., Qiao, L., Zhu, S., Li, L., Lou, S., Wang, H., et al. (2016). Chemical characteristics of particulate matters and trajectory influence on air quality in Shanghai during the heavy haze episode in December, 2013. Environmental Science (in Chinese), 37, 1179–1187.
|
atmosphere
|
0016
|
10.1016/j.atmosenv.2015.08.019
|
image
|
Fig. 1. Location of the sampling site (solid star) and the typical backward trajectories for air parcels arriving at the sampling site (percentages in brackets are the frequencies of the trajectories).
|
|
atmosphere
|
0016
|
10.1016/j.atmosenv.2015.08.019
|
table
|
Table 1 Seasonal concentration of the organic compounds and some important indices of $\mathrm{PM}_{2.5}$ at Lin'an.
|
|
atmosphere
|
0016
|
10.1016/j.atmosenv.2015.08.019
|
table
|
Table 1 (continued )
|
|
atmosphere
|
0016
|
10.1016/j.atmosenv.2015.08.019
|
image
|
Fig. 2. Variation of the LMW/HMW ratio with ambient temperature for PAHs in $\mathrm{PM}_{2.5}$ at LA.
|
|
atmosphere
|
0016
|
10.1016/j.atmosenv.2015.08.019
|
image
|
Fig. 3. Ratios of Flu $/(\mathrm{Flu}+\mathrm{Pyr})$ and $\mathrm{IcP}/(\mathrm{IcP}+\mathrm{BgP})$ in $\mathrm{PM}_{2.5}$ at LA.
|
|
atmosphere
|
0016
|
10.1016/j.atmosenv.2015.08.019
|
table
|
Table 2 Characteristics of $\mathrm{PM}_{2.5}$ associated with air masses from different directions.
|
|
atmosphere
|
0016
|
10.1016/j.atmosenv.2015.08.019
|
image
|
Fig. 4. Spatial distributions of PSCF values of (a) OC, (b) EC, (c) $n$ -alkanes, (d) PAHs, (e) hopanes, (f) levoglucosan in $\mathrm{PM}_{2.5}$ at LA.
|
|
atmosphere
|
0016
|
10.1016/j.atmosenv.2015.08.019
|
text
| Not supported with pagination yet |
Characteristics and seasonal variation of organic matter in PM2.5 at a regional background site of the Yangtze River Delta region, China
Jialiang Feng a, \*, Junchao Hu a, Binhua Xu a, Xiaoling Hu a, Peng Sun a, Wenliang Han b, Zeping Gu a, Xiangming Yu c, Minghong Wu a
a School of Environmental and Chemical Engineering, Shanghai University, Shanghai 200444, China
b Institute of Environmental and Resources Technology, Department of Environmental Science and Engineering, Huaqiao University, Xiamen 361021, China
c Lin'an Background Air Monitoring Station, Lin'an, Zhejiang 311307, China
h i g h l i g h t s
Characters and sources of organics in $\mathrm{PM}_{2.5}$ at LA, a regional background site in YRD were investigated.
High pollution level was found at LA due to the regional anthropogenic activities.
Distinct seasonal variations of concentration and composition of SEOC were observed.
Vehicle emissions should not be the main source of PAHs in $\mathrm{PM}_{2.5}$ in YRD.
a r t i c l e i n f o
Article history:
Received 7 January 2015 Received in revised form 10 July 2015
Accepted 5 August 2015 Available online xxx
Keywords:
PM2.5
Carbonaceous aerosol
Organic tracer
Lin'an
Yangtze River Delta
PSCF
a b s t r a c t
One hundred and ten seasonal $\mathrm{PM}_{2.5}$ samples were collected at Lin'an (LA), a regional background site in the Yangtze River Delta (YRD) region of China, to study the chemical composition, seasonal variation and sources of carbonaceous aerosols in the YRD region. Concentrations of organic carbon (OC), elemental carbon (EC), water-soluble organic carbon (WSOC) and composition of solvent-extractable organic compounds (SEOC) including $n$ -alkanes, hopanes, PAHs, $n$ -fatty acids and levoglucosan were analyzed. Back trajectory and potential source contribution function (PSCF) analysis were conducted to identify the possible source areas of $\mathrm{PM}_{2.5}$ at LA. $\mathrm{PM}_{2.5}$ concentration at LA ranged from $15.7~\upmu\mathrm{g}\;\mathrm{m}^{-3}$ to $182.1~\upmu\mathrm{g}~\mathrm{m}^{-3}$ , with an annual average of $59.7~\upmu\mathrm{g}\;\mathrm{m}^{-3}$ . A large part of the total carbon $\mathrm{^{\prime}O C+E C}$ , $11.8\;\upmu\mathrm{g}\;\mathrm{m}^{-3}.$ was watersoluble ( $42\%$ in winter and spring, $56\%$ in summer and $49\%$ in autumn). SEOC concentration showed distinct seasonal variation of higher in winter and lower in summer. The average concentrations of $n$ - alkanes and PAHs in winter $105~\mathrm{ng}~\mathrm{m}^{-3}$ and $55~\mathrm{ng}~\mathrm{m}^{-3}$ respectively) were 5 and 8 times of that in summer. Diagnostic ratios of PAHs and PSCF analysis suggested that vehicle emission was not the main source of PAHs at LA. Based on the levoglucosan concentration (annual average of $136~\mathrm{ng~m}^{-3}$ , more than $15\%$ of the OC at LA was from biomass burning. This data set provides useful information for the understanding of the characteristics of $\mathrm{PM}_{2.5}$ in the YRD region.
$\circledcirc$ 2015 Elsevier Ltd. All rights reserved.
1. Introduction
With the rapid economic growth and urbanization during the last several decades, air pollution has become a pressing environmental problem in China. Due to the adverse health implication (Dockery et al., 1993; Pope and Dockery, 2006) and the possible impact on global or regional climate (Menon et al., 2002), $\mathsf{P M}_{2.5}$ pollution and the haze events in China are drawing more and more concerns in recent years. The occurrence of large-scale haze events (Xu et al., 2013; Wang et al., 2015a), over 1.5 million $\mathrm{km}^{2}$ sometimes, and the monitoring results in the industrialized and developed areas such as Yangtze River Delta (YRD), Pearl River Delta and the Beijing-eTianjineHebei city clusters revealed the regional character of air pollution in China. Studies of aerosol characteristics at the regional background sites in comparison to those at urban areas can provide invaluable information on the impacts of anthropogenic activities, because the rapid urbanization not only affects the air quality in cities but also the quality at the rural or background areas (Kim et al., 2009; Yin et al., 2010). Studies carried out at several regional background sites and remote areas in China, such as Mountain Tai (Li et al., 2010), Mountain Hua (Wang et al., 2012), Mountain Lu (Li et al., 2014), Jinsha (Zhang et al., 2014) and Shangdianzi (Yan et al., 2012) showed that compositions of particles at the background sites could reflect the characteristics of the regional air pollution.
The YRD is one of the regions with the rapidest economic and social development in China, and the consequent degradation of air quality has been recognized as the bottleneck for the sustainable eco-social development. Lin'an background air monitoring station (LA), which is also one of the Global Atmospheric Watch background stations of the World Meteorological Organization in China, is a site with negligible local anthropogenic emission but influenced by regional transported pollutants. Characteristics of fine particles at LA would thus be more conducive in providing information on the pollution status in the whole YRD region, because information from urban sites might reflect just the local situation and not enough to understand the environmental effect of air pollution in a regional scale. Several studies have been conducted at LA and confirmed that LA station is representative of the regional background in YRD (Xu et al., 2002; Wang et al., 2004; Zheng et al., 2005; Yan et al., 2006, 2012; Tang et al., 2007; Meng et al., 2014; Pu et al., 2014). High loading of $\mathsf{P M}_{2.5}$ , carbonaceous aerosols, O3, VOCs and other gas pollutants were found at LA, indicating the seriousness of regional air pollution in YRD. However, previous studies have mostly focused on gaseous species in the atmosphere, and OC, EC and the inorganic water-soluble ions in particles, and most of the studies were based on short-period monitoring. To our knowledge, the chemical composition of organic matter in the regional background fine particles in the YRD region of China has not been reported.
Organic matter is an important part of the fine particles (Ye et al., 2003; Yan et al., 2012). Organic matter, especially the solvent-extractable organic compounds (SEOC), contains useful molecular markers which have been successfully used for source identification and source apportionment (Simoneit et al., 1991; Schauer et al., 1996; Zheng et al., 2005). For example, hopanes could be markers of vehicle emission (Schauer et al., 1999), levoglucosan is the marker of biomass burning (Simoneit, 2002) and nalkanes with odd carbon number of $\mathsf{C}_{27}-\mathsf{C}_{31}$ are markers of plant wax contribution (Rogge et al., 1993; Alves et al., 2012). The relatively low reactivity of $n$ -alkanes make them interesting tracers for both atmospheric transport and particle origin (Omar et al., 2007). Meanwhile, SEOC has been found to be toxic and can cause DNA mutation even at non-lethal dosages (Hsiao et al., 2000), due to the existence of toxic organic compounds such as polycyclic aromatic hydrocarbons (PAHs) (Ramdahl, 1983; Mirante et al., 2013). Chemical speciation of the organic matter in fine particles is thus very important for the understanding of the characters and sources of the regional background aerosols.
In this study, seasonal $\mathsf{P M}_{2.5}$ samples were collected at the LA regional background station of YRD, and the abundance and characteristics of SEOC were analyzed. Back-trajectories and potential source contribution function (PSCF) analysis were conducted to identify the possible source areas.
2. Material and methods
2.1. Sampling site
$\mathsf{P M}_{2.5}$ samples were collected at the Lin'an Background Air Monitoring Station (LA, $30^{\circ}18^{\prime}\mathrm{N}$ , $119^{\circ}44^{\prime}\mathrm{E}$ , on the top of a mountain of $138\;\mathrm{m}$ -high above sea level), which is in Zhejiang province and is a regional background site for the YRD region of China (Fig. 1). LA is about 11 km north of the downtown of Lin'an county, $50\,\mathrm{km}$ west of HangZhou, the capital city of Zhejiang province, and about $150\;\mathrm{km}$ southwest of Shanghai, the mega-city in the YRD region. NingboShaoxin region, a developed and industrialized area in Zhejiang province, is in the southeast direction of LA. About $100\,\mathrm{km}$ north of the LA station is the SuzhoueWuxieChangzhou region, which is the most developed/industrialized region in Jiangsu province. To the west of LA is the mountainous area of Zhejiang and Anhui provinces. LA station is located in a hilly agricultural/forested area with only several small villages scattered around the sampling site.
2.2. Sampling
$\mathsf{P M}_{2.5}$ samples were collected on quartz fiber filters (Whatman, QM-A $20.3\times25.4~\mathrm{cm}^{2}$ , baked at $450\,^{\circ}\mathrm{C}$ for $^{5\,\mathrm{h}}$ before use) with a high-volume $\mathsf{P M}_{2.5}$ sampler (Graseby, GMWT 2200) in four seasons (4 Aprile2 May 2008, spring; 1 Julye30 July 2008, summer; 4 Octobere2 November 2008, autumn; 19 December 2008e13 January 2009, winter). The flow rate was $1.13\ \mathrm{m}^{3}\ \mathrm{min}^{-1}$ and the sampling time was nominally $24~\mathrm{h}$ (6:00 PM to 6:00 PM) for each sample. After sampling, the filters were wrapped with annealed aluminum foil and stored at $-20~^{\circ}\mathrm{C}$ till analysis. The mass concentrations of $\mathsf{P M}_{2.5}$ were determined by weighing the filters before and after sampling under constant temperature $(20~^{\circ}\mathrm{C})$ and humidity $(45\%)$ . Meteorological parameters including ambient temperature, wind speed, wind direction and rainfall during the sampling periods were recorded by the monitoring station collocated with the sampling site.
2.3. Sample analysis
2.3.1. OC, EC and WSOC measurement
Organic carbon (OC) and elemental carbon (EC) concentrations of the samples were analyzed with a thermal/optical carbon analyzer (DRI 2001A, Atmoslytic Inc., USA) with the IMPROVE temperature program. A punch of quartz filter was heated to $580\,^{\circ}\mathrm{C}$ in four steps under helium to volatilize the OC, and then heated to $870~^{\circ}\mathrm{C}$ in three steps in a $\mathrm{He}{\mathrel{\mathrm{:}}}0_{2}$ environment. A HeeNe laser of $650~\mathrm{nm}$ was used to monitor the charring of OC.
A punch of $6~\mathrm{{cm}}^{2}$ from each high-volume filter was extracted with $10\,\mathrm{~mL~}$ nano-pure water and the concentration of watersoluble organic carbon (WSOC) was measured with a total organic carbon analyzer (Multi $_{\mathrm{N/C}}$ 2100, Analytikjena, Germany).
2.3.2. SEOC analysis
The procedure for SEOC analysis has been described in our previous publications (Feng et al., 2007; Gu et al., 2010) and is briefly described here. The filters were spiked with a deuterated standard mixture including tetracosane- $\cdot{\bf d}_{50},$ , anthracene- $\cdot{\bf d}_{10},$ chrysene- $\mathbf{\cdotd}_{12}\,$ , perylene- $\cdot\mathbf{d}_{12}$ and heptadecanoic acid- $\cdot\mathrm{{d}}_{33}$ , and then ultrasonically extracted with three $40\,\textrm{m L}$ aliquots of dichloromethane/methanol $(2{:}1\,\mathsf{V}/\mathsf{V})$ for $20\;\mathrm{min}$ each at room temperature. The combined extract was filtered, and then reacted with freshly prepared diazomethane to esterify the free organic acids. The esterified extract was then subjected to GC-MS (Agilent 5975 MSD interfaced to a Agilent 6890 GC) analysis. Part of the extract was dried and reacted with N,O-bis-(trimethylsilyl)trifluoroacetamide (BSTFA) for the analysis of levoglucosan. The MS is operated in EI mode and the scan range is 50e550 amu. The GC is equipped with a HP-5MS capillary column $(30\,\mathrm{~m~}\times\,0.25\,\mathrm{~mm~}$ i.d., film thickness $0.25\ \upmu\mathrm{m}\right)$ , with helium as carrier gas. Hexamethylbenzene was added to the extraction before GC-MS analysis to check the recovery of the target compounds. Recoveries of better than $80\%$ were found for all the deuterated standards. Field blanks, filter blanks and solvent blanks were examined. Phthalate esters, small amount of palmitic acid and stearic acid, and trace amount of alkanes $\left(C_{17}{-}C_{25}\right)$ were found in field blanks. The concentrations of the target compounds in the blanks were less than $20\%$ of the real samples, therefore the results were not corrected for blanks.
2.4. Statistical analysis
Statistical analysis of the data, including KolmogoroveSmirnov test, correlation, linear regression and two-tailed student's $t\mathrm{.}$ -test, were done with SPSS 16.
2.5. Back trajectory analysis
Forty-eight hours backward trajectories of the air masses reaching the sampling site at the height of $200\ \mathrm{m}$ above ground during the sampling periods were analyzed with the HYSPLIT model provided by NOAA (http://ready.arl.noaa.gov/HYSPLIT.php) using the NCEP/NCAR (National Centers for Environmental Prediction/National Center for Atmospheric Research) Global Reanalysis Data. One trajectory was calculated every six hours for each day.
2.6. Potential source contribution function (PSCF)
PSCF was used to assess the possible source areas contributing the high concentrations of OC, EC, PAHs, hopanes and other organic components. PSCF is the probability that an air mass arriving at the sampling site after passing through a specific geographical region (or cell) has a pollutant concentration above an arbitrarily set criterion. The ijth component of a PSCF field is defined as
$$
\mathrm{PSCF_{ij}=\frac{m_{i j}}{n_{i j}}}
$$
where $\mathfrak{n}_{\mathrm{ij}}$ is the total number of endpoints of the trajectories that fall in the ijth cell during transport to the receptor site, and $\mathrm{m_{ij}}$ for a given pollutant is a subset of $\mathfrak{n}_{\mathrm{ij}}$ whose back trajectory endpoints are associated with concentrations higher than a pre-defined threshold (Lehrstuhl.,1998).
To reduce the uncertainties caused by small $\mathfrak{n}_{\mathrm{ij}}$ values, the PSCF values were multiplied by an arbitrary weight function $\mathrm{W_{ij}}$ (Wang et al., 2006; Choi et al., 2011). The weight function reduced the PSCF values when $\mathfrak{n}_{\mathrm{ij}}$ was less than three times the average endpoints in each cell. $\mathsf{W}_{\mathrm{ij}}$ was defined as follow:
$$
\mathrm{W_{ij}}=\left\{\begin{array}{l l}{1.0\qquad\quad\mathrm{n_{ij}}\rangle3\mathrm{n_{ave}}}&{}\\ {0.7\quad1.5\mathrm{n_{ave}}\langle\mathrm{n_{ij}}\leq3\mathrm{n_{ave}}}&{}\\ {0.4\quad\mathrm{n_{ave}}\langle\mathrm{n_{ij}}\leq1.5\mathrm{n_{ave}}}&{}\\ {0.2\qquad\quad\mathrm{n_{ij}}\leq\mathrm{n_{ave}}}&{}\end{array}\right.
$$
In this study, the highest $30\%$ of the target pollutants was used to define $\mathbf{m}_{\mathrm{ij}}$ . The geophysical region covered by the trajectories was divided into an array of $0.5^{\circ}\times0.5^{\circ}$ grid cells.
3. Results and discussion
3.1. Concentrations of $P M_{2.5}$ , OC and EC
The concentrations of the quantified components were listed in Table 1, and statistical analysis showed that they all followed normal distribution (KolmogoroveSmirnov test). Distinct seasonal variations of $\mathsf{P M}_{2.5}$ mass were found with spring $(86.6\pm31.8~\upmu\mathrm{g}~\mathrm{m}^{-3})$ ) having the highest concentrations and summer $(36.0\pm11.5\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ the lowest $(p<0.05$ in two-tailed t-test for comparisons between different seasons). The annual concentration of $\mathsf{P M}_{2.5}$ at LA was $59.7~\upmu\mathrm{g}\textrm{m}^{-3}$ , comparable with that in Shanghai (Ye et al., 2003) and obviously higher than the new annual $\mathsf{P M}_{2.5}$ standard of China $(35\ \upmu\mathrm{g}\ \mathrm{m}^{-3})$ . Previous studies at LA showed that the $\mathsf{P M}_{2.5}$ concentration was $90~\upmu\mathrm{g}~\mathrm{m}^{-3}$ in November of 1999 (Xu et al., 2002); $40.1~\upmu\mathrm{g}\;\mathrm{m}^{-3}$ in summer and $71.7~\upmu\mathrm{g}\textrm{m}^{-3}$ in winter for the $\mathsf{P M}_{2.1}$ mass during 2004e2005 (Yan et al., 2012). Our results were similar with the previous studies and indicated that the air pollution in the YRD region remained at high level during the last decade. $\mathsf{P M}_{2.5}$ level at LA were lower than the Shangdianzi background station in the Beijing-Tianjin-Hebei region $(72.4~\upmu\mathrm{g}\,\\,\mathrm{m}^{-3})$ (Yan et al., 2012), and a little bit higher than the JSH background station in Central China $(48.7\pm26.9\mathrm{~}\upmu\mathrm{g~m}^{-3}$ , Zhang et al., 2014), confirming that high level air pollution is a wide spread problem in China.
The annual average concentrations of OC and EC in $\mathsf{P M}_{2.5}$ at LA were $10.3\pm5.44~\upmu\mathrm{g}~\mathrm{m}^{-3}$ and $1.54\pm0.79~\upmu\mathrm{g}~\mathrm{m}^{-3}$ . Organic matter $(1.6\times0.7)$ accounted for $28\%$ of the $\mathsf{P M}_{2.5}$ mass, and EC for $3\%$ . The annual average of total carbon (TC, $\mathrm{OC}+\mathrm{EC}$ , $11.8\pm6.18\mathrm{\;\upmug\;m}^{-3},$ was lower than that in urban Shanghai in 2000 (Ye et al., 2003), but higher than that at Changdao ${p=0.045})$ , a rural island in Bohai sea of northern China (Feng et al., 2007). Higher concentrations of OC and EC were found in spring and winter, while the lowest in summer $\mathcal{p}<0.05$ in $t^{\th}$ -test, except that between spring and winter).
A large part of the organic carbon in $\mathsf{P M}_{2.5}$ at LA was found to be water-soluble. The annual WSOC concentration was $5.31~\upmu\mathrm{g}~\mathrm{m}^{-3}$ accounting for $55\%$ of the OC. The WSOC/TC ratio was the highest in summer (0.56), followed by autumn, then spring and winter (Table 1), indicating the high contribution of secondary organic carbon (SOC) in summer (Yang et al., 2005). The WSOC/TC ratio at LA was higher than that in Shanghai (Feng et al., 2006a, 2013), while comparable with that at Changdao (Feng et al., 2007), suggesting the organic aerosols at LA were transported and aged.
3.2. Composition of SEOC
Quantified SEOC includes $n$ -alkanes, hopanes, PAHs $n$ -fatty acids, and levoglucosan (Table 1). The total concentration of SEOC was $384\,\mathrm{ng}\,\mathrm{m}^{-{\bar{3}}}$ , $138\,\mathrm{ng}\,\mathrm{m}^{-3}$ , $352\;\mathrm{ng}\;\mathrm{m}^{-3}$ and $600\,\mathrm{ng}\,\mathrm{m}^{-3}$ in spring, summer, autumn and winter respectively. Among these homologues, levoglucosan and $n$ -fatty acids had the highest concentrations (with the annual average of $136~\mathrm{{ng}~m^{-3}}$ and $135~\mathrm{ng}~\mathrm{m}^{-3}$ respectively), followed by $n$ -alkanes $(60\,\mathrm{ng}\,\mathrm{m}^{-3})$ , PAHs $(25\mathrm{\;ng\;m}^{-3})$ ) and hopanes $(2.9\;\mathrm{ng}\;\mathrm{m}^{-3}$ ).
3.2.1. n-Alkanes $(C_{I6}-C_{36})$
Distinct seasonal variation of $n$ -alkanes concentration was found $\langle p<0.01\rangle$ , with the highest in winter, followed by spring and autumn, and the lowest in summer (Table 1). Obvious seasonal variation was also found in the composition of $n$ -alkanes, samples in winter and spring had more low-molecular-weight individuals than that in summer and autumn. The sum of $n$ -alkanes with carbon number less than 26 accounted for $47\%$ of the total $n$ -alkanes in winter, while $25\%$ , $19\%$ and $18\%$ in spring, summer and autumn. Alkanes of $<\!C_{26}$ are semi-volatile and exist in both gas and particle phase, change in ambient temperature will shift the equilibrium of gas-particle partitioning (Cincinelli et al., 2007). More semi-volatile alkanes will be in particle phase in winter and increase the contribution of low-molecular-weight alkanes. Similar seasonal variations in $n$ -alkanes composition were also found in other places (Feng et al., 2006a,b, 2007).
Fossil fuel residue (such as gasoline, diesel, Schauer et al., 1999) and plant wax emission (Rogge et al., 1993; Gonçalves et al., 2010) are the two main sources of $n$ -alkanes. $n$ -Alkanes from fossil fuel show no significant carbon number predominance, while which from plant waxes are comprised mainly of $\mathsf C_{25}\mathrm{-}\mathsf C_{33}$ individuals with odd carbon numbers. The highest contributions of plant wax to alkanes occurred in summer $(30\%)$ , followed by spring $(28\%)$ , autumn $(26\%)$ and winter $(15\%)$ (Table 1), indicating that plant wax was the important contributor of $n$ -alkanes in $\mathsf{P M}_{2.5}$ at LA though fossil fuel usage was the main source. Consequently, high carbon predominance index (CPI) was found in summer, spring and autumn (Table 1).
High contributions of plant wax $(>\!50\%)$ were found in several days in spring, such as April 9 and 26. Together with the evidences of high concentrations of ${\mathsf{C a}}^{2+}$ (not presented in this paper) and high wind speeds during these periods, the high contribution of plant wax indicated the influence of dust episodes (Fang et al., 1999), and dust episodes should be an important reason for the high $\mathsf{P M}_{2.5}$ concentration in spring at LA.
3.2.2. Hopanes
Hopanes (cyclic triterpanes) are widely used as organic markers for fossil fuel residue (Schauer et al., 1999). In particular, they are known to be emitted from gasoline and diesel vehicles through the use of engine lubricating oil, and are widely used as markers of vehicular emissions in urban environment (Fine et al., 2004). Hopanes from $C_{27}{\mathrm{-}}C_{32}$ without $C_{28}$ were detected in nearly all samples at LA, and the most abundant compound was $17\alpha(\mathrm{H}),21\beta(\mathrm{H}).$ -hopane $(C_{30}\alpha\beta$ -hopane). The yield of hopanes had very similar seasonal trend with that of OC and $n$ -alkanes, higher in spring and winter than autumn and summer (Table 1), due to the seasonal changes in meteorological conditions. But the composition of hopanes was nearly the same in different seasons and was very similar with what found in particles from tunnels. As the compositions of hopanes from different sources were different (Feng et al., 2005), the similarity in composition suggested that the sources of hopanes did not have obvious seasonal change and vehicular emission be the main contributor of hopanes at LA. Though the concentration of hopanes at LA was lower than the reported value for urban Shanghai (Feng et al., 2006a), as expected, the detection of hopanes in all samples indicated that vehicular emissions had an important impact on the air quality at LA, and YRD.
3.2.3. PAHs
The seasonal average concentrations of the 20 quantified PAHs in $\mathsf{P M}_{2.5}$ at LA were listed in Table 1. The highest PAHs concentration $(55~\mathrm{ng}~\mathrm{m}^{-3})$ ) was found in winter, followed by spring and autumn, and summer had the lowest concentration $(6.\bar{7}\,\mathrm{ng}\,\mathrm{m}^{-3}$ ) $\boldsymbol{p}$ values of t-tests were $_{<0.01}$ except that between spring and autumn). The seasonal trend of PAHs concentration at LA was similar with what found in other places in China (Guo et al., 2003; Feng et al., 2005, 2006a,b; Wang et al., 2015b), and should be the result of the seasonal variation of weather conditions. Benzo(a)pyrene (BaP) had a concentration range of $0.05\ \mathrm{ng\m^{-3}{-5.6}\ n g\ m^{-3}}$ , with an annual average of $1.6~\mathrm{ng}~\mathrm{m}^{-3}$ , which is higher than the new air quality standard of China $\langle1\;\mathrm{ng}\;\mathrm{m}^{-3}\rangle$ .
Benzo $(\mathbf{b}+\mathbf{k})$ fluoranthene was the most abundant individual compound in each season, followed by indeno(1,2,3-cd)pyrene (IcP), benzo(ghi)perylene (BgP) and benzo(e)pyrene (BeP). The 5e6 ring PAHs accounted for $62\%{-}74\%$ of the total PAHs. Generally speaking, the compositions of PAHs in different seasons were quite similar, except the relative contribution of low-molecular-weight species (LMW, molecular weight $^{<228}$ ). The LMW/HMW ratio in spring, summer, autumn and winter was 0.35, 0.29, 0.33 and 0.71 respectively (Table 1). It is well recognized that PAHs with molecular weight less than 252 were semi-volatile, and the seasonal change of the LMW/HMW ratio was in agreement with the gasparticle partitioning shift of PAHs under different ambient temperature. From the variation of LMW/HMW ratio with temperature (Fig. 2), negative linear correlation was found between the LMW/ HMW ratio and temperature when temperature was lower than $15~^{\circ}C$ $\mathbb{R}^{2}=0.69$ , $p<0.01$ ), while the LMW/HMW ratio remained stable when temperature was higher than $15\;^{\circ}\mathbf{C}.$ . The stable LMW/ HMW ratio suggested that the LMW species in $\mathsf{P M}_{2.5}$ at LA under higher ambient temperature (say, $>\!15\ \mathrm{\Sigma^{\circ}C})$ were absorbed inreversibly in the elemental carbon or organic matter.
Ratios of $\mathrm{Flu}/(\mathrm{Flu}+\mathrm{Pyr})$ and $\mathrm{IcP}/(\mathrm{IcP}+\mathrm{BgP})$ have been used to distinguish the emission sources of PAHs (Grimmer et al., 1983; Yunker et al., 2002). Fl $\mathrm{\u/(Flu\,+\,Pyr)}$ ratio of $0.40–0.50$ suggests the combustion of liquid fossil fuels (vehicle and crude oil), whereas larger than 0.50 is characteristic for grass, wood or coal burning (Yunker et al., 2002). $\mathrm{IcP/(IcP\,+\,BgP)}$ of 0.18e0.40 was found in fine particles from vehicle emissions, while ${>}0.5$ for coal/ biomass combustion (Grimmer et al.,1983; Yunker et al., 2002). $\mathrm{Flu}/$ $(\mathrm{Flu}\,+\,\mathsf{P y r})$ ratio at LA ranged from 0.38 to 0.56, with an annual average of 0.48, and IcdP $^{\circ}/(\mathrm{IcP}+\tt B g P)$ ranged from 0.45 to 0.58, with an annual average of 0.51, indicating that PAHs at LA were from mixed emissions of coal/biomass burning and traffic emissions. It can be seen from Table 1 and Fig. 3 that the ratios of $\mathrm{IcP}/(\mathrm{IcP}+\mathrm{BgP})$ and $\mathrm{Flu}/(\mathrm{Flu}\,+\,\mathrm{Pyr})$ were higher in winter and lower in summer, suggesting higher contributions of traffic emissions in summer while higher contributions of coal/biomass burning in winter. Statistical analysis showed that $\mathrm{IcP}/(\mathrm{IcP}+\mathrm{BgP})$ and $\mathrm{Flu}/(\mathrm{Flu}+\mathrm{Pyr})$ had positive correlation $\mathrm{~\boldmath~R~}^{2}=0.2$ , $p=0.01$ ), indicating they were in agreement in reflecting the emission sources. Further, significant negative linear correlation was found between $\mathrm{Flu}/(\mathrm{Flu}+\mathrm{Pyr})$ ratio and ambient temperature $\mathbf{\nabla}\cdot\mathbf{R}^{2}~=~0.55$ and slope $\mathrm{~\ensuremath~{~=~}~}-0.0026$ $p<0.01\$ ), while the correlation between $\mathrm{IcP}/(\mathrm{IcP}+\mathrm{BgP})$ and temperature was weaker $\mathrm{R}^{2}=0.30$ and slope $=-0.0014$ , $p<0.01$ ). IcP and BgP are non-volatile and the temperature dependence of the $\mathrm{IcP/(IcP+BgP)}$ ratio should be the result of seasonal change in emission sources. While the different physical/chemical properties of Flu and Pyr should be an important cause of the stronger correlation and the steeper slope for the $\mathrm{Flu}/(\mathrm{Flu}\,+\,\mathrm{Pyr})$ ratio. The boiling point of Flu $(375\ ^{\circ}\mathrm{C})$ is lower than Pyr $(393\ ^{\circ}C)$ , more Flu than Pyr would partition to gas phase with the increase of temperature or during the atmospheric transport, thus lower the $\mathrm{Flu}/$ $(\mathrm{Flu}+\mathsf{P y r})$ ratio especially for the regional background sites where transported pollutants dominate. So the effect of gas-particle partitioning on the ratio value should be taken into consideration when using $\mathrm{Flu}/(\mathrm{Flu}+\mathrm{Pyr})$ ratio as a tool for source identification of particles.
3.2.4. $n_{\mathrm{~\rightmoon~}}$ -fatty acids $(C_{I2}\sim C_{32})$
The yield of $n$ -fatty acids at LA ranged from $19\,\mathrm{\Omegang\,\m}^{-3}$ to $294\;\mathrm{ng}\;\mathrm{m}^{-3}$ , with an annual average of $133\;\mathrm{ng}\;\mathrm{m}^{-3}$ , lower than the reported concentrations for the urban areas of China, such as Qingdao, Beijing, Shanghai and Guangzhou (Guo et al., 2003; Feng et al., 2005, 2006a,b), but comparable with the rural Changdao Island in northern China (Feng et al., 2007), indicating that the anthropogenic activities in urban areas was the main source of fatty acids. The seasonal trend of $n$ -fatty acids (Table 1) was similar with that of $n$ -alkanes and PAHs. It had been found that the $<\!C_{20}$ homologues of fatty acids have multiple sources such as kitchen emission, biomass burning and vehicle emission, while the $>\!C_{20}$ homologues are mostly from plant wax (Simoneit and Mazurek, 1982; Rogge et al., 1993). The contribution of plant wax to fatty acids in spring, autumn and winter at LA ${\sqrt{\sqrt{2}0\%}}$ Wax% in Table 1) was higher than that in the Chinese cities such as Beijing and Shanghai (Feng et al., 2005, 2006a) but comparable with Changdao Island (Feng et al., 2007) and Jeju Island of Korea (Simoneit et al., 2004), in accordance with the rural properties of the LA station.
Alkenoic acids of oleic acid $\left(\mathsf{C}_{18:1}\right)$ and linoleic acid $\left(C_{18:2}\right)$ were detected in most of the samples, but the concentrations were low in all seasons (Table 1). As alkenoic acids are easily degraded in the air, $\mathsf{C}_{18:1}/\mathsf{C}_{18}$ ratio is often used as an indicator of the aging of aerosols (Kawamura and Gagosian, 1987). Low $\mathsf{C}_{18:1}/\mathsf{C}_{18}$ ratio (0.21) in summer was expectable due to the high ambient temperature and strong solar radiation. The low $\mathsf{C}_{18:1}/\mathsf{C}_{18}$ ratio $(\sim\!0.20)$ in winter and spring at LA, comparing with that in urban Beijing (\~0.7, Feng et al., 2005), indicated that the aerosols at LA were aged and transported.
3.2.5. Levoglucosan and biomass burning contribution
Levoglucosan, a pyrolysis product of cellulose, is widely used as a marker of biomass burning (Simoneit, 2002; Hays et al., 2011). Higher level of levoglucosan was found in winter at LA $(229~\mathrm{~ng~}\,\mathrm{~m}^{-3})$ ), followed by autumn $(158\ \ \mathrm{ng}\ \ \mathrm{m}^{-3})$ , spring $(126~\upmu\mathrm{g}\textrm{m}^{-3})$ and summer $(4\bar{5}\ \mathrm{ng\m}^{-3})$ . The seasonal variation of levoglucosan, significant at confidence level of 0.05, could be attributed to the increased crop residue burning in autumn and the increased residential biomass burning in winter besides the seasonal change of meteorological conditions. In addition, longrange transport of biomass burning emissions could also influence the levoglucosan level at the background sites (Aggarwal and Kawamura, 2009).
Simulated combustion using crop residues in China showed that levoglucosan accounted for $8.2\%$ and $4.5\%$ of the OC and particle mass in $\mathsf{P M}_{2.5}$ from biomass burning (Zhang et al., 2007). Using these ratios as emission factors, contribution of biomass burning to OC in $\mathsf{P M}_{2.5}$ at LA was estimated to be $14\%$ in spring, $9\%$ in summer, $21\%$ in autumn and $23\%$ in winter, with an annual average of $16\%$ . For the $\mathsf{P M}_{2.5}$ mass at LA, $4\%$ in spring, $3\%$ in summer, $7\%$ in autumn and $8\%$ in winter was from biomass burning, with an annual average of $5\%$ . As levoglucosan would undergo photochemical degradation during atmospheric transport (Hennigan et al., 2010; Mochida et al., 2010), the contribution of biomass burning to $\mathsf{P M}_{2.5}$ at LA would be underestimated, especially in summer. So biomass burning is an important contributor to $\mathsf{P M}_{2.5}$ in the YRD region and deserves more attention.
3.3. Characters of $P M_{2.5}$ associated with different trajectories
The back trajectories of air masses reaching the sampling site were categorized into five groups as shown in Fig. 1. Group 1 was for air masses passing over Shanghai and Hangzhou, group 2 for air masses passing over Jiangsu province, group 3 for Anhui province, group 4 for Fujian province and the southern part of Zhejiang province, and group 5 for the eastern part of Zhejiang province. The frequency of each group in the whole sampling periods can be found in Fig. 1, and the characteristics of the $\mathsf{P M}_{2.5}$ associated with each group of trajectories were listed in Table 2. The dominant air trajectory in spring was group 5 and group 2, it was group 4 in summer, and group 2 in autumn and winter.
From Table 2, $\mathsf{P M}_{2.5}$ concentration was higher under trajectories of group 1 and group 5, in accordance with the fact that these trajectories passing over the most developed and industrialized areas in the YRD region. Generally speaking, high concentrations of $\mathsf{P M}_{2.5}$ were found at LA in all trajectory categories, indicating the widespread of pollution sources in the whole region. Our previous studies found higher $\mathsf{P M}_{2.5}$ concentrations at the suburban areas than the urban areas (Feng et al., 2005, 2006a), results of this study suggested that high pollution level in suburban and rural areas could be common in YRD and even in other regions in China. The concentrations of organic matter in $\mathsf{P M}_{2.5}$ associated with different trajectories (Table 2) followed similar trend as that for $\mathsf{P M}_{2.5}$ mass.
Some differences were found in the Alkanes/PAHs and Hopanes/EC ratios in different trajectory categories, suggesting that the emission sources in different areas were different. As discussed previously, variation in ambient temperature would shift the gas-particle partitioning of the semi-volatile organics, and the speed of photochemical degradation of the less stable organics would be different at different temperature. To avoid the impact of ambient temperature on the chemical composition of organics, the Alkanes/PAHs ratios in groups with similar ambient temperature (group 3, 4 and 5, Table 2) were compared. It was found that the ratio in group 5 was significantly higher than that in Group 3 and 4 $(p<0.05)$ , indicating that the sources in eastern Zhejiang, the Ningbo-Shaoxin region, emitted less PAHs relative to alkanes than the sources in Anhui and southern Zhejiang. Alkanes/ PAHs ratio in trajectories of group 1, Shanghai-Hangzhou region, was also higher than that in group 2, the industrialized region in
Jiangsu Province. The higher Hopanes/EC ratio in group 1 and 5 than that in group 2, 3 and 4 (Table 2) also suggested the higher contribution of vehicle emissions in the Shanghai-Hangzhou region and eastern Zhejiang than other areas. While the $\mathrm{Ic}\mathrm{P}/$ $(\mathrm{IcP}\,+\,\mathsf{B g P})$ ratio showed no difference between the 5 groups, suggesting the sources of PAHs associated with different trajectories were nearly the same. So the results of this study indicated that vehicle emissions should not be the main source of the PAHs in $\mathsf{P M}_{2.5}$ in YRD.
3.4. Potential source areas analysis
Potential source areas of the main components/species in PM2.5 at LA identified by the PSCF modeling were listed in Fig. 4. The potential source areas included Shanghai, eastern Zhejiang, Jiangsu, Jiangxi and Shandong provinces, indicating that the air quality at LA was influenced by the transported particles from all the industrialized areas in YRD. Though the trajectory analysis showed that only $14\%$ of the air parcels traveled over the Shanghai-Hangzhou region during the sampling periods, the high pollution events were closely related with this area, suggesting the importance of the emissions in this area for the air quality in YRD. The Yellow Sea and East China Sea were also identified as possible sources (Fig. 4b, c, e); however, these areas should not be actual source areas and might appear due to the trailing effect (Lee et al., 2014).
Fig. 4 also showed that the potential source areas for different components in $\mathsf{P M}_{2.5}$ at LA were different. High concentrations of EC and hopanes were mainly caused by the emissions from the Shanghai-Hangzhou region and eastern Zhejiang, in accordance with the fact that these two areas had the highest density of vehicles. While the high concentration events of PAHs were mainly due to the transported particles from Jiangsu and Shangdong provinces, and it approved to some extent the conclusion of Section 3.3 that vehicle emissions were not the main source of PAHs in the YRD region.
4. Conclusions
Based on the analysis of 110 seasonal $\mathsf{P M}_{2.5}$ samples, the annual average concentration of $\mathsf{P M}_{2.5}$ , OC and EC at LA was $59.7~\upmu\mathrm{g}~\mathrm{m}^{-3}$ $10.3~\upmu\mathrm{g}\textrm{m}^{-3}$ and $1.54\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ , respectively. The results of this study were comparable with that for year 2004 and earlier at the same location, indicating that the air pollution in YRD remained at high level during the last decade. Among the quantified SEOC, levoglucosan and $n$ -fatty acids had the highest concentrations (with the annual average of $136~\mathrm{ng~m}^{-3}$ and $135~\mathrm{ng~m}^{-3}$ respectively), followed by $n$ -alkanes $60\ \mathrm{ng}\ \mathrm{m}^{-3}.$ , PAHs $(25~\mathrm{ng}~\mathrm{m}^{-3})$ ) and hopanes $(2.9\mathrm{~ng~m}^{-3})$ ). More than $70\%$ of the $n$ -alkanes in $\mathsf{P M}_{2.5}$ at LA were found from fossil fuel residue, while ${\sim}40\%$ of the fatty acids were from plant wax. Hopanes were detected in all samples and the composition of hopanes was very similar with that from vehicle emissions, indicating that vehicle emissions had an important impact on the air quality in YRD. The $\mathrm{IcP/(IcP+\,BgP)}$ and $\mathrm{Flu}/$ $(\mathrm{Flu}\,+\,\mathsf{P y r})$ ratios suggested that PAHs in $\mathsf{P M}_{2.5}$ at LA was from mixed emissions of coal/biomass combustion and vehicle emission. Back trajectory and PSCF analysis showed that the sources of EC and hopanes were different with that of PAHs, suggesting that vehicle emissions was not the main source of the particulate PAHs in YRD. Based on the concentration of levoglucosan, more than $15\%$ of the OC in $\mathsf{P M}_{2.5}$ at LA was from biomass burning.
Acknowledgments
The study was financially supported by the National Natural Science Foundation of China (20877052, 41173097), the Program for Changjiang Scholars and Innovative Research Team in University (IRT13078), and the Natural Science Foundation of Fujian Province (2011J05112) for which the authors are grateful. Thanks are giving to the NOAA Air Resources Laboratory (ARL) for the HYSPLIT model used in this work.
References
Aggarwal, S.G., Kawamura, K., 2009. Carbonaceous and inorganic composition in long-range transported aerosols over northern Japan: implication for aging of
water-soluble organic fraction. Atmos. Environ. 43, 2532e2540.
Alves, C., Vicente, A., Pio, C., Kiss, G., Hoffer, A., Decesari, S., Prevo^t, A.S.H., Minguill on, M.C., Querol, X., Hillamo, R., Spindler, G., Swietlicki, E., 2012. Organic compounds in aerosols from selected European sites e biogenic versus anthropogenic sources. Atmos. Environ. 59, 243e255.
Choi, E., Choi, K., Yi, S.M., 2011. Non-methane hydrocarbons in the atmosphere of a metropolitan city and a background site in South Korea: sources and health risk potentials. Atmos. Environ. 45, 7563e7573.
Cincinelli, A., Del Bubba, M., Martellini, T., Gambaro, A., Lepri, L., 2007. Gas-particle concentration and distribution of n-alkanes and polycyclic aromatic hydrocarbons in the atmosphere of Prato (Italy). Chemosphere 68, 472e478.
Dockery, D.W., Pope, C.A., Xu, X.P., Spengler, J.D., Ware, J.H., Fay, M.E., Ferris, B.G., Speizer, F.E., 1993. An association between air pollution and mortality in six U.S. cities. N. Engl. J. Med. 329 (24), 1753e1759.
Fang, M., Zheng, M., Wang, F., Chim, K.S., Kot, S.C., 1999. The long-range transport of aerosols from northern China to Hong Kong e a multi-technique study. Atmos. Environ. 33, 1803e1817.
Feng, J., Chan, C., Fang, M., Hu, M., He, L., Tang, X., 2005. Impact of meteorology and energy structure on solvent extractable organic compounds of PM2.5 in Beijing, China. Chemosphere 61, 623e632.
Feng, J., Chan, C., Fang, M., Hu, M., He, L., Tang, X., 2006a. Characteristics of organic matter in $\mathrm{PM}_{2.5}$ in Shanghai. Chemosphere 64, 1393e1400.
Feng, J., Guo, Z., Chan, C., Fang, M., 2007. Properties of organic matter in $\mathrm{PM}_{2.5}$ at Changdao Island, Chinada rural site in the transport path of the Asian continental outflow. Atmos. Environ. 41, 1924e1935.
Feng, J., Hu, M., Chan, C., Lau, P.S., Fang, M., He, L., Tang, X., 2006b. A comparative study of the organic matter in PM2.5 from three Chinese megacities in three different climatic zones. Atmos. Environ. 40, 3983e3994.
Feng, J., Li, M., Zhang, P., Gong, S., Zhong, M., Wu, M., Zheng, M., Chen, C., Wang, H., Lou, S., 2013. Investigation of the sources and seasonal variations of secondary organic aerosols in $\mathrm{PM}_{2.5}$ in Shanghai with organic tracers. Atmos. Environ. 79, 614e622.
Fine, M.P., Chakrabarti, B., Krudysz, M., Schauer, J.J., Constantions, S., 2004. Diurnal variations of individual organic compound constituents of ultrafine and accumulation mode particulate matter in the Los Angeles Basin. Environ. Sci. Technol. 38 (5), 1296e1304.
Gonçalves, C., Alves, C., Evtyugina, M., Mirante, F., Pio, C., Caseiro, A., Schmidl, C., Bauer, H., Carvalho, F., 2010. Characterisation of PM10 emissions from wood stove combustion of common woods grown in Portugal. Atmos. Environ. 44 (35), 4474e4480.
Grimmer, G., Brune, H., Deutsch-Wenzel, R., Naujack, K.W., Misfeld, J., Timm, J., 1983. On the contribution of polycyclic aromatic hydrocarbons to the carcinogenic impact of automobile exhaust condensate evaluated by local application onto mouse skin. Cancer Lett. 21 (1), 105e113.
Gu, Z., Feng, J., Han, W., Wu, M., Fu, J., Sheng, G., 2010. Characteristics of organic matter in $\mathrm{PM}_{2.5}$ from an e-waste dismantling area in Taizhou, China. Chemosphere 80, 800e806.
Guo, Z., Sheng, L., Feng, J., Fang, M., 2003. Seasonal variation of solvent extractable organic compounds in the aerosols in Qingdao, China. Atmos. Environ. 37, 1825e1834.
Hays, M.D., Gullett, B., King, C., Robinson, J., Preston, W., Touati, A., 2011. Characterization of carbonaceous aerosols emitted from outdoor wood boilers. Energy Fuels 25 (12), 5632e5638.
Hsiao, W.L.W., Mo, Z., Fang, M., Shi, X., Wang, F., 2000. Cytotoxicity of $\mathrm{PM}_{2.5}$ and $\mathrm{PM}_{2.5-10}$ ambient air pollutants assessed by the MTT and the comet assays. Mutat. Res. 471, 45e55.
Hennigan, C.J., Sullivan, A.P., Collett, J.L., Robinson, A.L., 2010. Levoglucosan stability in biomass burning particles exposed to hydroxyl radicals. Geophys. Res. Lett. 37, L09806. http://dx.doi.org/10.1029/2010GL043088.
Kawamura, K., Gagosian, R.B., 1987. Implications of u-oxocarboxylic acids in the remote marine atmosphere for photo-oxidation of unsaturated fatty acids. Nature 325, 330e332.
Kim, Y.J., Woo, J.H., Ma, Y.I., Kim, S., Nam, J.S., Sung, H., Choi, K.C., Seo, J., Kim, J.S., Kang, C.H., Lee, G., Ro, C.U., Chang, D., Young, S., 2009. Chemical characteristics of long-range transport aerosol at background sites in Korea. Atmos. Environ. 43, 5556e5566.
Lee, G., Kim, P., Han, Y., Holsen, T., Lee, S., 2014. Tracing sources of total gaseous mercury to Yongheung Island off the coast of Korea. Atmosphere 5, 273e291.
Li, P., Wang, Y., Li, Y., Wang, Z., Zhang, H., Xu, P., Wang, W., 2010. Characterization of polycyclic aromatic hydrocarbons deposition in $\mathrm{PM}_{2.5}$ and cloud/fog water at Mount Taishan (China). Atmos. Environ. 44, 1996e2003.
Li, W., Chi, J., Shi, Z., Wang, X., Chen, B., Wang, Y., Li, T., Chen, J., Zhang, D., Wang, Z., Shi, C., Liu, L., Wang, W., 2014. Composition and hygroscopicity of aerosol particles at Mt. Lu in South China: implications for acid precipitation. Atmos. Environ. 94, 626e636.
Meng, Z., Zhang, R., Lin, W., Jia, X., Yu, X., Yu, X., Wang, G., 2014. Seasonal variation of ammonia and ammonium aerosol at a background station in the Yangtze River Delta region, China. Aerosol Air Qual. Res. 14, 756e766.
Menon, S., Hansen, J., Nazarenko, L., Luo, Y.F., 2002. Climate effects of black carbon aerosols in China and India. Science 27 (297), 2250e2253.
Mirante, F., Alves, C., Pio, C., Pindado, O., Perez, R., Revuelta, M.A., Arti\~nano, B., 2013. Organic composition of size segregated atmospheric particulate matter, during summer and winter sampling campaigns at representative sites in Madrid, Spain. Atmos. Res. 132e133, 345e361.
Mochida, M., Kawamura, K., Fu, P., Takemura, T., 2010. Seasonal variation of levoglucosan in aerosols over the western North Pacific and its assessment as a biomass-burning tracer. Atmos. Environ. 44, 3511e3518.
Omar, N.Y.M.J., Abas, M.R.B., Rahman, N.A., Tahir, N.M., Rushdi, A.I., Simoneit, B.R.T., 2007. Levels and distributions of organic source tracers in air and roadside dust particles of Kuala Lumpur, Malaysia. Environ. Geol. 52, 1485e1500.
Pope, C.A., Dockery, D.W., 2006. Health effects of fine particulate air pollution: lines that Connect. J. Air Waste Manag. Assoc. 56, 709e742.
Pu, J., Xu, H., He, J., Fang, S., Zhou, L., 2014. Estimation of regional background concentration of $C0_{2}$ at Lin'an Station in Yangtze River Delta, China. Atmos. Environ. 94, 402e408.
Ramdahl, T., 1983. Polycyclic aromatic ketones in environmental samples. Environ. Sci. Technol. 17, 666e670.
Rogge, W.F., Hildemann, L.M., Mazurek, M.A., Cass, G.R., Simoneit, B.R.T., 1993. Sources of fine organic aerosol 4. Particulate abrasion products from leaf surfaces of urban plants. Environ. Sci. Technol. 27, 2700e2711.
Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 1999. Measurement of emissions from air pollution sources. 2. C1 through C30 organic compounds from medium duty diesel trucks. Environ. Sci. Technol. 33, 1578e1587.
Schauer, J.J., Rogge, W.F., Hildemann, L.M., Mazurek, M.A., Cass, G.R., 1996. Source apportionment of airborne particulate matter using origanic compounds as tracers. Atmos. Environ. 30 (22), 3837e3855.
Simoneit, B.R.T., 2002. Biomass burning e a review of organic tracers for smoke from incomplete combustion. Appl. Geochem. 17, 129e162.
Simoneit, B.R.T., Kobayashi, M., Mochida, M., Kawamura, K., Lee, M., Lim, H.J., Turpin, B.J., Komazaki, Y., 2004. Composition and major sources of organic compounds of aerosol particulate matter sampled during the ACE-Asia campaign. J. Geophys. Res. 109.
Simoneit, B.R.T., Mazurek, M.A., 1982. Organic matter of the troposphere e II. Natural background of biogenic lipid matter in aerosols over the rural western United States. Atmos. Environ. 16, 2139e2159.
Simoneit, B.R.T., Sheng, G., Chen, X., Fu, J., Zhang, J., Xu, Y., 1991. Molecular marker study of extractable organic matter in aerosols from urban areas of China. Atmos. Environ. 25A, 2111e2129.
Tang, J., Wang, M., Cheng, H., Ding, G., Yu, X., Zhou, H., Liu, G., 2007. Variation characteristics of ambient NMHCs at Shangdianzi and Lin'an regional GAW sites. Acta Meteorol. Sin. 21, 334e341.
Wang, G., Li, J., Cheng, C., Zhou, B., Xie, M., Hu, S., Meng, J., Sun, T., Ren, Y., Cao, J., Liu, S., Zhang, T., Zhao, Z., 2012. Observation of atmospheric aerosols at Mt. Hua and Mt. Tai in central and east China during spring 2009-part 2: impact of dust storm on organic aerosol composition and size distribution. Atmos. Chem. Phys. 12, 4065e4080.
Wang, H., Zhu, B., Zhang, Z., An, J., Shen, L., 2015a. Mixing state of individual carbonaceous particles during a severe haze episode in January 2013, Nanjing, China. Particuology 20, 16e23.
Wang, J., Li, X., Jiang, N., Zhang, W., Zhang, R., Tang, X., 2015b. Long term observations of PM2.5-associated PAHs: comparisons between normal and episode days. Atmos. Environ. 104, 228e236.
Wang, T., Wong, C.H., Cheung, T.F., Blake, D.R., Arimoto, R., Baumann, K., Tang, J., Ding, G.A., Yu, X.M., Li, Y.S., Streets, D.G., Simpson, I.J., 2004. Relationships of trace gases and aerosols and the emission characteristics at Lin'an, a rural site in eastern China, during spring 2001. J. Geophys. Res. 109, D19S05. http:// dx.doi.org/10.1029/2003JD004119.
Wang, Y., Zhang, X., Arimoto, R., 2006. The contribution from distant dust sources to the atmospheric particulate matter loadings at Xi'An, China during spring. Sci. Total Environ. 368, 875e883.
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. Atmos. Environ. 36, 161e173.
Xu, P., Chen, Y., Ye, X., 2013. Haze, air pollution, and health in China. Lancet 382 (9910), 2067.
Yan, P., Zhang, R., Huan, N., Zhou, X., Zhang, Y., Zhou, H., Zhang, L., 2012. Characteristics of aerosols and mass closure study at two WMO GAW regional background stations in eastern China. Atmos. Environ. 60, 121e131.
Yan, P., Zhang, Y., Yang, D., Tang, J., Zhou, X., 2006. Characteristics of aerosol ionic compositions in summer 2003 at Lin'an of Yangtze Delta Region. Acta Meteorol. Sci. 20 (3), 374e382.
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 (20), 3735e3749.
Ye, B.M., Ji, X.L., Yang, H.Z., Yao, X.H., Chan, C.K., Cadle, S.H., Chan, T., Mulawa, P.A., 2003. Concentration and chemical composition of $\mathrm{PM}_{2.5}$ in Shanghai for a 1- year period. Atmos. Environ. 37, 499e510.
Yin, J., Harrison, R.M., Chen, Q., Rutter, A., Schauer, J.J., 2010. Source apportionment of fine particles at urban background and rural sites in the UK atmosphere. Atmos. Environ. 44, 841e851.
Yunker, M.B., Macdonald, R.W., Vingarzan, R., Mitchell, R.H., Goyette, Darcy, Sylvestre, Stephanie, 2002. PAHs in the Fraser River basin: a critical appraisal of PAH ratios as indicators of PAH source and composition. Org. Geochem. 33, 489e515.
Zhang, F., Cheng, H., Wang, Z., Lv, X., Zhu, Z., Zhang, G., Wang, X., 2014. Fine particles $\left(\mathrm{PM}_{2.5}\right)$ at a CAWNET background site in Central China: chemical compositions, seasonal variations and regional pollution events. Atmos. Environ. 86, 193e202.
Zhang, Y., Shao, M., Zhang, Y., Zeng, L., He, L., Zhu, B., Wei, Y., Zhu, X., 2007. Source profiles of particulate organic matters emitted from cereal straw burnings. J. Environ. Sci. 19, 167e175.
Zheng, X., Chan, C., Cui, H., Qin, Y., Chan, L., Zheng, Y., Lee, Y., 2005. Characteristics of vertical ozone distribution in the lower troposphere in the Yangtze River Delta at Lin'an in the spring of 2001. Science in China Series D. Earth Sci. 48 (9), 1519e1528.
|
atmosphere
|
0017
|
10.1016/j.atmosres.2009.01.003
|
image
|
Fig. 1. Location of the sampling sites in Shanghai.
|
|
atmosphere
|
0017
|
10.1016/j.atmosres.2009.01.003
|
image
|
Fig. 2. Comparison of monthly average (MA) with seasonal average (SA) of $\mathsf{P M}_{10}$ $S0_{2}$ , and $\mathrm{NO}_{2}$ concentrations in Shanghai, China. (Note: The averaged values in the figure were calculated from the daily measurements by the Shanghai Environmental Monitoring Center (http://www.envir.gov.cn/airnews/)).
|
|
atmosphere
|
0017
|
10.1016/j.atmosres.2009.01.003
|
table
|
Table 1 Average concentrations of $\mathrm{PM}_{2.5}$ , OC and EC and their ratios in Shanghai.
|
|
atmosphere
|
0017
|
10.1016/j.atmosres.2009.01.003
|
table
|
Table 2 Comparison of $\mathrm{PM}_{2.5}\,0C$ and EC concentrations measured by TOT method in Shanghai with other cities in China and in the world.
|
|
atmosphere
|
0017
|
10.1016/j.atmosres.2009.01.003
|
image
|
Fig. 3. Variations of $\mathrm{PM}_{2.5}$ mass, OC, EC, OC/EC ratio, and fraction of TCA in $\mathrm{PM}_{2.5}$ at ZB and JD Sites in Shanghai.
|
|
atmosphere
|
0017
|
10.1016/j.atmosres.2009.01.003
|
image
|
Fig. 4. Seasonal correlations of OC and EC in $\mathrm{PM}_{2.5}$ in Shanghai.
|
|
atmosphere
|
0017
|
10.1016/j.atmosres.2009.01.003
|
table
|
Table 3 Levels of SOC and SOA in Shanghai estimated from minimum OC/EC ratios.
|
|
atmosphere
|
0017
|
10.1016/j.atmosres.2009.01.003
|
text
| Not supported with pagination yet |
Characteristics of organic and elemental carbon in $\mathsf{P M}_{2.5}$ samples in Shanghai, China
Yanli Feng a, Yingjun Chen b,c,⁎, Hui Guo d, Guorui Zhi c, Shengchun Xiong a, Jun Li c, Guoying Sheng a,c, Jiamo Fu a,c
a Institute of Environmental Pollution and Health, School of Environmental and Chemical Engineering, Shanghai University, Shanghai 200444, China b Yantai Institute of Coastal Zone Research for Sustainable Development, Chinese Academy of Sciences, Yantai, Shandong Province 264003, China c State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou 510640, China d Mud Logging Company, Great Wall Drilling Project Co. Ltd., Panjin, Liaoning 124010, China
a r t i c l e i n f o
Article history:
Received 22 April 2008
Received in revised form 4 November 2008
Accepted 5 January 2009 Keywords:
Organic carbon
Elemental carbon
Fine particle $(\mathrm{PM}_{2.5})$ Seasonal variation Shanghai
a b s t r a c t
Shanghai is the largest industrial and commercial city in China, and its air quality has been deteriorating for several decades. However, there are scarce researches on the level and seasonal variation of fine particle $(\mathsf{P M}_{2.5})$ as well as the carbonaceous fractions when compared with other cities in China and around the world. In the present paper, abundance and seasonal characteristics of $\mathrm{PM}_{2.5}$ organic carbon (OC) and elemental carbon (EC) were studied at urban and suburban sites in Shanghai during four season-representative months in 2005–2006 year. $\mathrm{PM}_{2.5}$ samples were collected with high-vol samplers and analyzed for OC and EC using thermal-optical transmittance (TOT) protocol. Results showed that the annual average $\mathrm{PM}_{2.5}$ concentrations were $90.3–95.5\,\upmu\mathrm{g}/\mathrm{m}^{3}$ at both sites, while OC and EC were $14.7{-}17.4\,\upmu\mathrm{g}/\mathrm{m}^{3}$ and $2.8{-}3.0\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , respectively, with the OC/EC ratios of 5.0–5.6. The carbonaceous levels ranked by the order of Beijing $>$ Guangzhou $>$ Shanghai $>$ Hong Kong. The carbonaceous aerosol accounted for $\sim\!30\%$ of the $\mathrm{PM}_{2.5}$ mass. On seasonal average, the highest OC and EC levels occurred during fall, and they were higher than the values in summer by a factor of 2. Strong correlations $(r\!=\!0.79\!-\!0.93$ ) between OC and EC were found in the four seasons. Average level of secondary organic carbon (SOC) was $5.7{-}7.2\;\upmu\mathrm{g}/\mathrm{m}^{3}$ , accounting for $\sim\!30\%$ of the total OC. Strong seasonal variation was observed for SOC with the highest value during fall, which was about two times the annual average.
$\circledcirc$ 2009 Elsevier B.V. All rights reserved.
1. Introduction
Carbonaceous aerosol constitutes a significant fraction of fineparticles $\mathrm{PM}_{2.5})$ , and it could account for up to $40\%$ of $\mathsf{P M}_{2.5}$ mass in urban atmosphere (Seinfeld and Pandis, 1998). Carbonaceous species are usually classified into elemental carbon (EC) and organic carbon (OC). EC (sometimes called black carbon) derived from incomplete combustion of carboncontained materials, while OC can be either released directly into the atmosphere (primary OC, POC) or produced from gasto-particle reactions (secondary OC, SOC) (Pandis et al., 1992; Turpin and Huntzicker,1995). EC has strong absorption of solar radiation and is one of the important drivers of global warming (Hansen et al., 2005). OC represents a mixture of hundreds of organic compounds, some of which are mutagenic and/or carcinogenic, such as polycyclic aromatic hydrocarbons (PAHs) and polychlorinated dibenzo- $\cdot p$ -dioxins and dibenzofurans (PCDD/Fs) (Feng et al., 2006; Li et al., 2008).
Carbonaceous aerosol in China has drawn special attention in recent years due to its adverse effects on environment and human health and potential influence on climate change (Hamilton and Mansfield,1991; Qiu and Yang, 2000; Jacobson, 2002). It has been estimated that China contributes roughly one-fifth of the global carbonaceous emissions (Bond et al., 2004). Jacobson (2002) suggested that emission reduction of fossil-fuel carbonaceous particles was possibly the most effective method of slowing global warming. The increased EC aerosols may be responsible for the significant variation of precipitation in eastern China over the past decades (Menon et al., 2002). EC also contributed to the marked degradation of optical depths and visibility in northern China (Qiu and Yang, 2000), and lowered the crop yields by reducing solar radiation that reaches the earth (Chameides et al., 1999).
There are several studies focusing on the field measurements of carbonaceous abundance in $\mathsf{P M}_{2.5}$ in China's industrialized areas, such as Guangzhou and Hong Kong in Pearl River Delta Region (Ho et al., 2002, 2003, 2006; Cao et al., 2003, 2004, 2005; Chow et al., 2005; Duan et al., 2007), Beijing (He et al., 2001, 2004a; Dan et al., 2004; Duan et al., 2006), and other cities (Guo et al., 2004; Cao et al., 2005; Yang et al., 2005b). However, there are only limited studies in Shanghai (Ye et al., 2003; Yang et al., 2005a; Feng et al., 2006), which is the largest commercial and industrial city in China and also one of the world's largest seaports. World Expositions will be held here in 2010. Furthermore, one-year carbonaceous measurements by Ye et al. (2003) and Yang et al. (2005a) were finished during 1999–2000 year, and since then the air pollution in Shanghai is ameliorating observably. For example, annual $\mathsf{P M}_{10}$ concentration in Shanghai was $100.5\,\upmu\mathrm{g}/\mathrm{m}^{3}$ in 2001 versus $86.2\,\upmu\mathrm{g}/\mathfrak{m}^{3}$ in 2006 (http://www. envir.gov.cn/airnews/).
Ambient air quality in Shanghai began to deteriorate in the 1960s due to large consumption of low-quality coal for rapid industrial development. When coal smoke was somewhat controlled in the 1990s, vehicular population in Shanghai soared up to $\sim\!2$ million, and the type of air pollution in Shanghai evolved into the combination of coal smoke with vehicular exhaust (Chen, 2003). Great concern on the visibility reduction and public health has been drawn by the heavy PM pollution in Shanghai (Ye et al., 2000), and the primary carbonaceous emission in 2000 was about $16.3~\mathrm{Gg}$ (gigagrams or ktons) for EC and $34.0~\mathrm{Gg}$ for OC (Cao et al., 2006).
The purpose of this paper is to present updated knowledge on abundance and seasonal characteristics of $\mathsf{P M}_{2.5}$ -associated EC and OC in Shanghai, which is helpful to investigate of their sources and PM control strategies.
2. Methodology
2.1. Sampling sites description
Shanghai is located at the east end of the Yangtze River Delta Region and faces the East China Sea (Fig. 1), and possesses a population of over 15 million and a land area of about $6340~\mathrm{km}^{2}$ . Shanghai belongs to the northern subtropical monsoon climate, and northwest wind prevails in wintertime whereas southeast wind in summertime. Annual average of temperature in Shanghai is $15.8\;^{\circ}\mathrm{C},$ with the lowest in January $(3.6\,^{\circ}\mathrm{C})$ and the highest in July $(27.8\ ^{\circ}\mathrm{C})$ .
In this study, two sites were selected for $\mathsf{P M}_{2.5}$ sample collection in Shanghai: one is in Zha-bei district (ZB) of downtown area, the other is in Jia-ding district (JD) which belongs to suburban area in northwest (Fig. 1). Brief descriptions for the sampling sites are given as follows.
ZB Site: It is located in the Yan-chang campus of Shanghai University in Zha-bei district. The sampler was placed on the rooftop of nine-storeyed office building (about $25\textrm{m}$ above ground level). This site is surrounded by low-rise office and residential buildings. A road with moderate traffic passes by $300\,\mathrm{m}$ away from this site in the north, and another road with heavy traffic $1.4~\mathrm{km}$ away in the west. This site represents a mixed residential, traffic, and commercial environments of urban area.
JD Site: It is situated at the Environmental Monitoring Station (EMS) of Jia-ding district, which is a suburban area with quick urbanization and industrial development in recent years. The sampler was mounted on the top of four-storeyed office building (about $10\;\mathrm{m}$ above ground) and was adjacent to other samplers of the EMS. Many two- or three-storeyed residential buildings are distributed around this site. A road with slight traffic passes by $50\,\mathrm{m}$ away in the south, and some small chimneys (burning low-quality coal) from a glass factory are observed $\sim\!100~\mathrm{m}$ away across the road. This site represents a residential and slightly industrialized environment of suburban area.
2.2. Sample collection
Samples were collected almost simultaneously at both ZB and JD Sites in four months: October in 2005, January, April, and July in 2006, which were selected to represent the four seasons in Shanghai respectively, i.e., fall, winter, spring, and summer. The representativeness of each selected month for the entire season was confirmed by the levels of $\mathsf{P M}_{10}.$ $S0_{2}$ , and $\Nu0_{2}$ in Shanghai during the same calendar year (http://www.envir.gov. cn/airnews/). As illustrated in Fig. 2, the averaged concentrations of $\mathsf{P M}_{10}$ , $S0_{2}$ , and $\mathsf{N O}_{2}$ in the selected four months were very similar to the four seasonal averages, respectively. Furthermore, the annually averaged concentrations of $\mathsf{P M}_{10}$ $S0_{2}$ and $\mathrm{NO}_{2}$ were very close to the mean levels of the selected four months, respectively. For example, the annual average of $\mathrm{PM}_{10}\,\mathrm{Was}\,89.5\,\upmu\mathrm{g}/\mathrm{m}^{3}$ while the average value of thefour months was $87.3\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , and the relative deviation was less than $3\%$ .
During the four weeks of each sampling month, two 24-h $\mathsf{P M}_{2.5}$ samples (from morning to morning) were collected in each week to include both weekday (Wednesday) and weekend (Sunday). Sampler applied was high-volume Andersen model SA235 equipped with $2.5~\upmu\mathrm{m}$ inlet at flow rate of $1.13~\mathrm{m}^{3}/\mathrm{min}$ (Thermal Electric Inc., USA) and quartzfiber filters (QFFs) of $20.3\!\times\!25.4\,\mathrm{cm}$ (Whatman, England). The meteorological parameters, including ambient temperature, relative humidity, wind speed/direction, and precipitation of each sampling day were recorded through a Hand-held Weather Monitoring (Kestrel-1000, USA). A total of 62 valid $\mathsf{P M}_{2.5}$ samples were collected at two sites, excluding one sample during fall at ZB Site and one in winter at JD due to incidental failure of electrical supply.
All QFFs were pre-baked at $450\,^{\circ}\mathrm{C}$ for $^{5\,\mathrm{h}}$ before sampling to remove residual carbon. Before and after exposure, all filters were weighed using an electronic balance (Satouris, $0.01\,\mathrm{\mg},$ Germany) under constant temperature $(25\,^{\circ}\mathrm{C})$ and relative humidity $(50\%)$ in a climate chamber (Binder KBF-115, Germany) for $24\,\mathrm{h}$ All samples were stored in refrigerator at $-20\,^{\circ}\mathrm{C}$ for late analysis. All procedures during handling of filters were strictly qualitycontrolled to avoid any possible contamination.
2.3. Carbonaceous analysis and quality control
The samples were analyzed for EC and OC using a thermaloptical transmittance analyzer (TOT, Sunset Laboratory Inc., USA) in the State Key Laboratory of Organic Geochemistry in Guangzhou, China. This instrument has a temperature- and atmosphere-controlled oven and a laser of $680\,\mathrm{\Omeganm}$ wavelength to generate an operational EC/OC split (Birch and Cary, 1996). Temperature procedure performed in the present study was similar to NIOSH method (Birch, 1998): punch aliquot $\phantom{-}1.5\;\mathrm{cm}^{2}.$ of a QFF sample was heated stepwise in the oven at $250\,^{\circ}\mathrm{C}$ (60 s), $500\,^{\circ}\mathrm{C}$ (60 s), $650\,^{\circ}\mathrm{C}$ (60 s), and $850\,^{\circ}\mathrm{C}$ (90 s) in pure helium atmosphere for OC volatilization, and $550\,^{\circ}\mathrm{C}$ (45 s), $650\,^{\circ}\mathrm{C}$ (60 s), $750\,^{\circ}\mathrm{C}$ (60 s), and $850\,^{\circ}\mathrm{C}$ (80 s) in $2\%$ oxygen-contained helium atmosphere for EC oxidation. Some charred OC during inert period was corrected by the laser transmittance. At the end of sample evolvement, a known volume of methane was analyzed as internal standard for calculation of OC and EC concentrations.
For quality control, the analyzer was calibrated using filter blank (pre-heated QFF punch) and standard sucrose solutions every day. Duplicate punches from each sample were analyzed to eliminate nonuniformity of depositions on the filter. Replicate analyses were performed with $10\%$ of total samples, and the differences indicated by replicate analyses was within $5\%$ for OC and EC. Four field blanks for each site were collected by sampling for $5~\mathrm{\min}$ and analyzed to examine operational contamination of the field samples in four months. Generally, the concentrations of $\mathsf{P M}_{2.5}$ , OC, and EC on the field blanks were less than $1\%$ of the sample batches, and were not subtracted from the samples.
3. Results and discussion
3.1. Levels of $P M_{2.5},$ , OC and EC in Shanghai
The statistics for $\mathsf{P M}_{2.5}$ mass, OC and EC at both sampling sites in Shanghai are presented in Table 1. Firstly, it can be seen that the levels of $\mathsf{P M}_{2.5}$ and carbonaceous fractions at suburban site (JD) $(95.5\pm41.8\,\upmu\mathrm{g}/\mathrm{m}^{3})$ are a little higher than that at urban site (ZB) $(90.3\pm54.9\,\upmu\mathrm{g}/\mathrm{m}^{3})$ , suggesting that fine particle pollution occurred not only in the urban, but also in the suburban area of Shanghai. Similar situation was observed by Feng et al. (2006) that there was no clear difference of carbonaceous concentrations between urban and rural areas in Shanghai. In spite of the difference of sampling heights at two sites, it may be inferred that the rapid urbanization and relocation of industrial plants in the past two decades have blurred the lines of ambient air pollution between urban and rural areas in Shanghai.
Secondly, ambient air pollution of $\mathsf{P M}_{2.5}$ in Shanghai is a serious matter of concern as per the U.S. National Ambient Air Quality Standard (NAAQS) annual average of $15\,\upmu\mathrm{g}/\mathfrak{m}^{3}$ (http:// www.epa.gov/air/criteria.html). Ye et al. (2003) also reported ${\sim}60.0~\upmu\mathrm{g/m}^{3}$ of annual $\mathsf{P M}_{2.5}$ concentration in Shanghai urban areas. However, it was reported that theairqualityin Shanghai in recent years came up to the Class II Level of Chinese national $\mathsf{P M}_{10}$ standards (GB3095-1996, the upper limit for annual average is $100\,\upmu\mathrm{g}/\mathrm{m}^{3}$ for Class II areas, such as urban residential, commercial and traffic, industrial, and rural areas, etc.) (http:// www.envir.gov.cn/airnews/). This situation also occurs in Guangzhou, which is the dirtiest city in Pearl River Delta Region (PRDR) and its $\mathsf{P M}_{2.5}$ level $(102.7–129.9\;\upmu\mathrm{g}/\mathrm{m}^{3})$ (Duan et al., 2007)iscomparabletoShanghaiandBeijing $\left(115.0{-}127.0\,\upmu\mathrm{g}/\mathrm{m}^{3}\right)$ (He et al., 2001), but its annual $\mathrm{PM}_{10}$ concentrations also reached the Class II Level (http://www.gdepb.gov.cn/). The high percentage of $\mathsf{P M}_{2.5}$ in $\mathsf{P M}_{10},$ , e.g., $70\%$ in PRDR (Cao et al., 2003, 2004), may interpret the severe $\mathsf{P M}_{2.5}$ pollutions in these areas. In a word, recent air qualities in these cities reached the Chinese $\mathsf{P M}_{10}$ standard, while were severely polluted as per U.S. $\mathsf{P M}_{2.5}$ standard. This inconsistency implies an urgent demand for $\mathsf{P M}_{2.5}$ standard in China.
The annual average concentrations of OC and EC in Shanghai are $14.7\pm10.1$ and $2.8\pm1.3\,\upmu\mathrm{g}/\mathrm{m}^{3}$ at ZB Site, and $17.5\pm9.8$ and $3.0\,{\pm}\,1.2\,\upmu\mathrm{g}/\mathrm{m}^{3}$ at JD, respectively. The inverted condition that carbonaceous pollution in the suburban area was higher than in the urban one was consistent with the $\mathsf{P M}_{2.5}$ levels discussed above. These carbonaceous abundances in Shanghai were comparable to the reported values in November 2002 $(\sim\!16.0~\upmu\mathrm{g/m}^{3}$ for OC and ${\sim}4.0~\upmu\mathrm{g}/\mathfrak{m}^{3}$ for EC) by Feng et al. (2006). Ye et al. (2003) also presented similar annual concentrations of total carbon (TC, sum of OC and EC) in Shanghai $(\sim\!21.0\,\upmu\mathrm{g}/\mathrm{m}^{3})$ , although the EC fractions were a little higher $(6.2{-}6.8~\upmu\mathrm{g/m^{3}})$ due to the employment of different carbon analysis protocol of thermal-optical reflectance (TOR) (Chow et al., 2001). Source apportionment results suggested that the carbonaceous pollution in Shanghai mainly derived from vehicular exhaust $(\sim\!50\%)$ , coal smoke $(\sim\!15\%)$ , and kitchen emissions $(\sim\!10\%)$ (Feng et al., 2006), and was also significantly affected by biomass burning (Yang et al., 2005a).
Fang et al. (2008) recently reviewed carbonaceous pollution in Asian cities. $\mathsf{P M}_{2.5}$ -associated OC and EC concentrations in mainland China ranged in 1.4–21.2 and $3.3{-}20.2\;\upmu\mathrm{g}/\mathrm{m}^{3}$ , respectively. As a matter of comparison, the values were 17.0 and $10.4{-}11.6\,\upmu\mathrm{g}/\mathrm{m}^{3}$ in Taiwan, 3.3–18.0 and $0.2–8.4~\upmu\mathrm{g}/\mathfrak{m}^{3}$ in Korea, and 1.1–5.2 and $2.3–4.5~\upmu\mathrm{g}/\mathfrak{m}^{3}$ in Japan, respectively (Fang et al., 2008 and references therein). It should be noted that various techniques were used for OCEC measurements other than TOT in these studies, such as TOR, TMO (thermal manganese dioxide oxidation), EA $\mathsf{\Gamma}(\mathsf{C}/\mathsf{H}/\mathsf{N}$ elemental analyzer), combustion method etc. Although a good agreement for TC values was generally obtained from these methods, EC concentration usually differs by a factor of 2 or more (Watson et al., 2005). Thus Table 2 only summarizes some measurements by TOT method to compare the carbonaceous level in Shanghai with other cities in China and around the world. $\mathsf{P M}_{2.5}$ OC and EC concentrations in China ranked in the following order: Beijing $(21.4{-}25.6~\upmu\mathrm{g}/\mathrm{m}^{3}$ and $5.6{-}5.7~\upmu\mathrm{g}/\mathfrak{m}^{3}$ for OC and EC, respectively) $>$ Guangzhou (18.4–22.6 and $4.8\mathrm{-}6.4\,\upmu\mathrm{g/m^{3}})>$ NShanghai (9.9–17.4 and 2.8– $3.1\ \upmu\mathrm{g}/\mathfrak{m}^{3}$ ) $\approx$ Nanjing (13.2–14.2 and $2.9{-}3.7~\upmu\mathrm{g/m}^{3})>\mathrm{Hong}$ Kong (5.8–12.0 and $1.1\!-\!3.4\;\upmu\mathrm{g}/\mathrm{m}^{3}.$ ), and the range of concentrations was obviously higher than those in urban sites of Europe and North America.
3.2. Seasonal characteristics of $P M_{2.5},$ OC and EC
Strong seasonal variations of $\mathsf{P M}_{2.5}$ , OC and EC in Shanghai have been shown in Table 1 and Fig. 3. The seasonally averaged $\mathsf{P M}_{2.5}$ concentrations were highest in spring $(113.5\,\upmu\mathrm{g}/\mathrm{m}^{3}$ at ZB) or fall $(113.3\,\upmu\mathrm{g}/\mathrm{m}^{3}$ at JD) whereas lowest in summer $(50.2\,\upmu\mathrm{g}/\mathrm{m}^{3}$ at ZB and $66.9\,\upmu\mathrm{g}/\mathrm{m}^{3}$ at JD), and they differed approximately by a factor of 2; $\mathsf{P M}_{2.5}$ levels in winter $(93.4~\upmu\mathrm{g}/\mathrm{m}^{3}$ at ZB and $87.9~\upmu\mathrm{g}/\mathrm{m}^{3}$ at JD) were close to the annual average values (Table 1). This pattern of $\mathsf{P M}_{2.5}$ was very similar to the seasonal average $\mathsf{P M}_{10}$ concentrations during 2005–2006 year in Shanghai, i.e., 95.6, 83.7, 115.7 and $59.6~\upmu\mathrm{g}/\mathrm{m}^{3}$ from fall to summer, respectively (http://www. envir.gov.cn/airnews/). Ye et al. (2003) reported similar trends of $\mathsf{P M}_{2.5}$ concentrations in two urban sites during 1999– 2000 yr except the maxima appeared in winter, which may be attributed to the highest $\mathsf{P M}_{2.5}$ concentration period of late fall and early winter (November and December) occurred in their sampling campaign. Fig. 3 illustrates the subtle variations of $\mathsf{P M}_{2.5}$ levels in both the sites. For example, the maximum $\mathsf{P M}_{2.5}$ concentration of $255.9\;\upmu\mathrm{g}/\mathrm{m}^{3}$ occurred on April 19, 2006 at ZB and $173.5\,\upmu\mathrm{g}/\mathfrak{m}^{3}$ on October 30, 2005 at JD, which was 10 and 6 times the minimum values, respectively.
The seasonal characteristics of $\mathsf{P M}_{2.5}$ concentrations in Shanghai can be explained as the combined impact of climatic conditions and local emissions. During monsoon seasons of Shanghai, i.e., usually from May to September, clean winds coming from the East China Sea together with abundant precipitation can relieve ambient air pollution to a great extent. However, during cold seasons which last from late fall till early spring, winds mainly blow from mainland China (north, northwest, and west to Shanghai) where air is polluted. Large power plants distribute intensively in northern part of Shanghai such as those from Baosteel Group Corporation Limited. Local emissions also increase rapidly due to heating in cold seasons. Annually Shanghai consumes about 45 million tons of coal for industrial and residential purposes. Low level temperature inversion makes the air pollution more serious (haze, mostly happens in fall) and resulted in great fluctuation of PM concentrations during these seasons together with other added factors such as dust storms in springtime (Fig. 3).
Carbonaceous species had similar seasonal patterns to $\mathsf{P M}_{2.5}$ mass at both sites, although their highest concentrations distinctly occurred in fall (Table 1). For example, OC and EC concentrations at ZB Site from fall to summer were 21.8, 16.7, 14.1, 7.2, and 3.9, 2.3, 3.1, $1.9~\upmu\mathrm{g}/\mathrm{m}^{3}$ , respectively. The levels of OC and EC in fall were higher than in summer by 2–3 times, and the values in winter and spring were almost equivalent. However, there was asynchronous variation between carbonaceous species and $\mathsf{P M}_{2.5}$ mass, and it was attributable to the seasonality of percentages of total carbonaceous aerosol (TCA) in $\mathsf{P M}_{2.5}$ mass (Table 1 and Fig. 3). In this paper, TCA was calculated by the sum of EC and organic matter (OM) which was estimated by multiplying the amount of OC by 1.6 (Turpin and Lim, 2001). As shown in Table 1, TCA accounted for an annually averaged $28.9\%$ (ZB) and $32.0\%$ (JD) of $\mathsf{P M}_{2.5}$ mass, with the seasonal rank of fall $>$ winter $>$ summer $>$ spring. The obviously lowest contribution of TCA to $\mathsf{P M}_{2.5}$ mass in spring may be affected by the contributions of Asian dust storms which frequently occur during the dry spring seasons (particularly during April– May). A typical example was that a strong dust storm appeared on April 19, 2006 which affected a broad area in China and was recorded by the samples in this study. As Fig. 3 demonstrates, TCA/ $\mathsf{P M}_{2.5}$ ratio in the sample of that particular day at ZB Site has the minimum value of all the samples $(16.9\%)$ while the maximum $\mathsf{P M}_{2.5}$ concentration was $255.9~\upmu\mathrm{g}/\mathrm{m}^{3}$ ; similarly in the corresponding sample at JD (collected on April 20, 2006), the TCA/ $\mathrm{PM}_{2.5}$ ratio and $\mathsf{P M}_{2.5}$ were $22.2\%$ and $165.7\;\upmu\mathrm{g}/\mathrm{m}^{3}$ respectively, thereby indicating the obvious contribution of inorganic material.
3.3. The relationship of OC and EC
OC–EC relationship gives some indication of the origin of carbonaceous particles. Fig. 3 shows the regression between OC and EC concentrations for all $\mathsf{P M}_{2.5}$ samples from both sites in Shanghai except for the two samples affected by dust storm, as mentioned above. Strong correlations $(r)$ of 0.82, 0.93, 0.79, and 0.86 were observed for the four seasons from fall to summer. This indicated that carbonaceous particles in Shanghai derived from common emission sources such as vehicular exhaust and/ or coal combustion, underwent a similar atmospheric dispersion process. The variations of regression slopes (6.46–8.29, see Fig. 4) might have been resulted from the seasonal variability of emission sources and SOA contributions.
Since carbonaceous aerosol represents a mixture of various emission sources (EC and primary OC) and secondary OC formed by atmospheric reaction processes, the ratio of OC to EC concentrations (OC/EC) can be used to study the emission and transformation characteristics. Typical emission sources include diesel- and gasoline-powered vehicular exhaust $(0C/\mathrm{EC}\,{=}\,1.0{-}4.2)$ (Schauer et al., 1999, 2002), wood combustion (16.8–40.0) (Schauer et al., 2001), residential coal smoke (2.5–10.5) (Chen et al., 2006), kitchen emissions (32.9– 81.6) (He et al., 2004b), and biomass burning (7.7) (Zhang et al., 2007), etc. It should be noted that the OC/EC ratios presented above were all measured by TOT method, and were comparatively higher than the values by TOR (Watson et al., 2001).
As Table 1 shows, the seasonal average OC/EC ratios in $\mathsf{P M}_{2.5}$ at ZB Site varied in the range of 3.4–6.8 with annual average of 5.0; while at JD, the range was 4.0–6.9 and the average was 5.6. It can be seen that the OC/EC ratios at JD Site were slightly higher than that at ZB for all seasons. This can be explained by the surroundings of both sites: ZB Site was affected by more traffic emissions, while JD by more emission sources with high OC/EC ratios, such as cooking exhaust and coal smoke (small chimneys in a glass factory), etc. Study by Feng et al. (2006) suggested kitchen emissions was the third important source for carbonaceous aerosol in Shanghai after vehicular exhaust and residential coal combustion and cannot be negligible.
For inter-seasonal comparison, the highest OC/EC ratio occurred in wintertime versus the lowest in summertime, and they differed by a factor of 2 (Table 1). Similar phenomenon was reported in PRDR (Duan et al., 2007), and the reasons included: (1) more semi-volatile organic compounds condensed into aerosol in lower temperature; (2) stagnant and dry meteorological conditions resulted in more SOA formation in wintertime; (3) more residential combustion of coal and wood for space heating occurred in winter; (4) polluted air with higher OC/EC ratio from mainland China in northwest and west deteriorated the ambient air quality in Shanghai during cold seasons (Duan et al., 2007).
3.4. Abundances of SOC and SOA in Shanghai
The importance of SOA has been recognized for decades on account of its relation to haze, visibility, climate, and human health. However, difficulties still exist to directly separate secondary OC (SOC) from primary OC (POC), although there are many methods for quantification of total OC (TOC). An indirect method for estimation of SOC has been usually employed using EC as the tracer for POC, since EC is essentially emitted from combustion sources together with primary organic components (Turpin and Huntzicker, 1995), and the equation is as follows:
$$
{\mathrm{SOC}}={\mathrm{TOC}}-{\mathrm{EC}}\times({\mathrm{OC}}/{\mathrm{EC}})_{\mathrm{primary}},
$$
Where $(\mathrm{OC/EC)_{primary}}$ is the ratio for primary sources contributing to the sample. A OC/EC value of 2.0 has been used to estimate SOC (Chow et al., 1996). However, the primary ratio of OC/EC is usually not available because it is affected by many factors such as the type of emission source as well as its variation in temporal and spatial scales, ambient temperature, and carbon determination method, etc. In many case, $(\mathrm{OC/EC)_{primary}}$ was represented by the observed minimum ratio $(\mathrm{(OC/EC)_{min}})$ , and assumptions regarding the use of this procedure as were discussed in detail by Castro et al. (1999).
As presented in Table 3, the minimum OC/EC ratio at ZB and JD Sites in Shanghai ranged in 2.7–4.1 (average 3.3) and 2.2–5.3 (3.5) in different seasons, respectively, and the highest ratios were observed during wintertime. All these lowest ratios occurred in the samples are affected by the air mass coming from the East China Sea, according to the backward trajectory analysis performed with HYSPLIT from the NOAA ARL website (www.arl.noaa.gov/ready/). These values are comparable with those of Guangzhou (2.3–4.5) (Duan et al., 2007), and California, USA (3.7) (Na et al., 2004), but higher than that of Budapest, Hungary (1.1–2.9) (Salma et al., 2004), and Helsinki, Finland (1.1) (Viidanoja et al., 2002).
The annual average concentrations of estimated SOC in Shanghai $\mathsf{P M}_{2.5}$ samples was $5.7\,\upmu\mathrm{g}/\mathrm{m}^{3}$ at ZB Site and $7.2\,\upmu\mathrm{g}/\mathrm{m}^{3}$ at JD, accounting for $27.0\%$ and $33.2\%$ of OC, respectively. This implied that SOC was an important component of OC mass in Shanghai. Compared to the modeling results by chemical mass balance (CMB) (Feng et al., 2006), i.e., $18–19\%$ of SOC-to-OC percentage for urban site and $23-25\%$ for rural site in Shanghai, our data were slightly higher. The seasonally averaged values showed that SOC concentration was distinctly higher during fall than in other seasons at both the sites, and was about two times the annual average (Table 3). This seasonal pattern of SOC was similar to that of OC and EC concentrations in $\mathsf{P M}_{2.5}.$ but differed with that of the OC/EC ratio in which highest values occurred in winter, as stated before. This was attributable to the contributions of various sources in different seasons while meteorological conditions during fall (such as temperature inversion) availed the formation of SOC.
SOA concentration was calculated here by multiplying SOC by a factor of 1.6, and ranged in $3.2{-}18.5\ \ \upmu\mathrm{g}/\mathrm{m}^{3}$ (averaged $9.1\ \ \upmu\mathrm{g}/\mathrm{m}^{3})$ at ZB Site while $6.8{-}22.0\ \ \upmu\mathrm{g/m^{3}}$ (average $11.6\,\upmu\mathrm{g}/\mathrm{m}^{3})$ at JD (Table 3). The average percentages of $7.8\mathrm{-}10.4\%$ at both sites indicated that SOA contributed a minor fraction of $\mathrm{PM}_{2.5}$ mass in Shanghai, although sometimes it could account for up to $20\%$ in fall.
4. Conclusions
Abundance and seasonal characteristics of $\mathsf{P M}_{2.5}$ and carbonaceous species were investigated at two sites at Shanghai, China. On annual average, $\mathsf{P M}_{2.5}$ concentration was $90.3\,\upmu\mathrm{g}/\mathrm{m}^{3}$ at ZB Site and $95.5\,\upmu\mathrm{g}/\mathrm{m}^{3}$ at JD, indicating the serious situation of fine particle pollution in Shanghai. The concentrations for OC and EC were 14.7 and $2.8\ \upmu\mathrm{g}/\mathrm{m}^{3}$ at ZB whereas 17.4 and $3.0\,\upmu\mathrm{g}/\mathrm{m}^{3}$ at JD, respectively, following the order of Beijing $>$ Guangzhou $>$ Shanghai $>$ Hong Kong. Carbonaceous aerosol (TCA) accounted for $\sim\!30\%$ of $\mathsf{P M}_{2.5}$ mass at both sites. On seasonal average, the highest levels of $\mathsf{P M}_{2.5}$ and carbonaceous fractions were generally observed during fall and were higher than summer by a factor of 2. Strong correlations $(r\!=\!0.79\!-\!0.93)$ ) between OC and EC suggested the contributions of common sources and similar atmospheric process during each season in Shanghai. The averaged OC/EC ratios were 5.0 at ZB and 5.6 at JD with the highest values observed during winter, implying the significant contributions of sources with elevated OC/EC ratios such as residential coal smoke and kitchen emissions. SOC concentrations were estimated by minimum OC/EC ratio, and were 5.7 and $7.2\,\upmu\mathrm{g}/\mathrm{m}^{3}$ at ZB and JD, respectively, accounting for $\sim\!30\%$ of the total OC. The average ratios of SOA/ $\mathsf{P M}_{2.5}$ of $7.8\mathrm{-}10.4\%$ indicated that SOA was a minor fraction in fine particles of Shanghai, although it could constitute up to $20\%$ of $\mathsf{P M}_{2.5}$ mass during fall.
Acknowledgements
This project was financially supported by the Department of Science and Technology of Shandong Province (2006GG2205033, 2007GG2QT06018), National Natural Scientific Foundation of China (40605033, 40503012), and Shanghai Leading Academic Disciplines (S30109). The authors would like to thank Jia-ding Environmental Monitoring Station in Shanghai for the assistant of sampling at the station, Dr. Xiyong Hou from YIC, CAS and Ms. Paromita Chakraborty from GIG, CAS for their helps in this manuscript. The authors also thank all three anonymous reviewers for their helpful comments and suggestions.
References
Birch, M.E., 1998. Analysis of carbonaceous aerosols: interlaboratory comparison. The Analyst 123, 851–857.
Birch, M.E., Cary, R.A., 1996. Elemental carbon-based method for monitoring occupational exposures to particulate diesel exhaust: methodology and exposure issues. The Analyst 121, 1183–1190.
Bond, T.C., Streets, D.G., Yarber, K.F., Nelson, S.M., Woo, J.H., Klimont, Z., 2004. A technology-based global inventory of black and organic carbon emissions from combustion. Journal of Geophysical Research-Atmospheres 109. doi:10.1029/2003JD003697.
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. Atmospheric Environment 37,1451–1460.
Cao, J.J., Lee, S.C., Ho, K.F., Zou, S.C., Fung, K., Li, Y., Watson, J.G., Chow, J.C., 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., An, Z.S., Fung, K.K., Watson, J. G., Zhu, C.S., Liu, S.X., 2005. Characterization and source apportionment of atmospheric organic and elemental carbon during fall and winterof 2003 in Xi'an, China. Atmospheric Chemistry and Physics 5, 3127–3137.
Cao, G., Zhang, X., Zheng, F., 2006. Inventory of black carbon and organic carbon emissions from China. Atmospheric Environment 40, 6516–6527.
Castro, L.M., Pio, C.A., Harrison, R.M., Smith, D.J.T.,1999. Carbonaceous aerosol in urban and rural European atmospheres: estimation of secondary organic carbon concentrations. Atmospheric Environment 33, 2771–2781.
Chameides, W.L., Yu, H., Liu, S.C., Bergin, M., Zhou, X., Mearns, L., Wang, G., Kiang, C.S., Saylor, R.D., Luo, C., 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 of the National Academy of Sciences 96, 13626–13633.
Chen, G., 2003. Orientation of the types of Shanghai air pollution. Shanghai Environmental Sciences 22 (4) (in Chinese, with English abstract).
Chen, Y., Zhi, G., Feng, Y., Fu, J., Feng, J., Sheng, G., Simoneit, B.R.T., 2006. Measurements of emission factors for primary carbonaceous particles from residential raw-coal combustion in China. Geophysical Research Letters 33. doi:10.1029/2006GL026966.
Chow, J.C., Watson, J.G., Lu, Z., Lowenthal, D.H., Frazier, C.A., Solomon, P.A., Thuillier, R.H., Magliano, K.,1996. Descriptive analysis of PM2.5 and PM10 at regionally representative locations during SJVAQS/AUSPEX. Atmospheric Environment 30, 2079–2112.
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, 23–34.
Chow, J.C., Watson, J.G., Louie, P.K.K., Chen, L.W.A., Sin, D., 2005. Comparison of PM2.5 carbon measurement methods in Hong Kong, China. Environmental Pollution 137, 334–344.
Dan, M., Zhuang, G., Li, X., Tao, H., Zhuang, Y., 2004. The characteristics of carbonaceous species and their sources in PM2.5 in Beijing. Atmospheric Environment 38, 3443–3452.
Duan, F.K., He, K.B., Ma, Y.L., Yang, F.M., Yu, X.C., Cadle, S.H., Chan, T., Mulawa, P.A., 2006. Concentration and chemical characteristics of PM2.5 in Beijing, China: 2001–2002. Science of the Total Environment 355, 264–275.
Duan, J.C., Tan, J.H., Cheng, D.X., Bi, X.H., Deng, W.J., Sheng, G.Y., Fu, J.M., Wong, M.H., 2007. Sources and characteristics of carbonaceous aerosol in two largest cities in Pearl River Delta Region, China. Atmospheric Environment 41, 2895–2903.
Fan, X., Brook, J.R., Mabury, S.A., 2003. Sampling atmospheric carbonaceous aerosols using an integrated organic gas and particle sampler. Environmental Science and Technology 37, 3145–3151.
Fan, X., Brook, J.R., Mabury, S.A., 2004. Measurement of organic and elemental carbon associated with PM2.5 during Pacific 2001 study using an integrated organic gas and particle sampler. Atmospheric Environment 38, 5801–5810.
Fang, G.-C., Wu, Y.-S., Chou, T.-Y., Lee, C.-Z., 2008. Organic carbon and elemental carbon in Asia: a review from 1996 to 2006. Journal of Hazardous Materials 150, 231–237.
Feng, J., Chan, C.K., Fang, M., Hu, M., He, L., Tang, X., 2006. Characteristics of organic matter in PM2.5 in Shanghai. Chemosphere 64, 1393–1400.
Guo, Z.G., Feng, J.L., Fang, M., Chen, H.Y., Lau, K.H., 2004. The elemental and organic characteristics of PM2.5 in Asian dust episodes in Qingdao, China, 2002. Atmospheric Environment 38, 909–919.
Hamilton, R.S., Mansfield, T.A., 1991. Airborne particulate elemental carbon: Its sources, transport and contribution to dark smoke and soiling. Atmospheric Environment 25, 715–723.
Hansen, J., Sato, M., Ruedy, R., Nazarenko, L., Lacis, A., Schmidt, G.A., Russell, G., Aleinov, I., Bauer, M., Bauer, S., 2005. Efficacy of climate forcings. Journal of Geophysical Research-Atmospheres 110, D18104.
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 PM2.5 in Beijing, China. Atmospheric Environment 35, 4959–4970.
He, Z., Kim, Y.J., Ogunjobi, K.O., Kim, J.E., Ryu, S.Y., 2004a. Carbonaceous aerosol characteristics of PM2.5 particles in Northeastern Asia in summer 2002. Atmospheric Environment 38, 1795–1800.
He, L.-Y., Hu, M., Huang, X.-F., Yu, B.-D., Zhang, Y.-H., Liu, D.-Q., 2004b. Measurement of emissions of fine particulate organic matter from Chinese cooking. Atmospheric Environment 38, 6557–6564.
Ho, K.F., Lee, S.C., Yu, J.C., Zou, S.C., Fung, K., 2002. Carbonaceous characteristics of atmospheric particulate matter in Hong Kong. The Science of The Total Environment 300, 59–67.
Ho, K.F., Lee, S.C., Chan, C.K., Yu, J.C., Chow, J.C., Yao, X.H., 2003. Characterization ofchemical species in PM2.5 and PM10 aerosolsin Hong Kong.Atmospheric Environment 37, 31–39.
Ho, K.F., Lee, S.C., Cao, J.J., Li, Y.S., Chow, J.C., Watson, J.G., Fung, K., 2006. Variability of organic and elemental carbon, water soluble organic carbon, and isotopes in Hong Kong. Atmospheric Chemistry and Physics 6, 4569–4576.
Jacobson, M.Z., 2002. Control of fossil-fuel particulate black carbon and organic matter, possibly the most effective method of slowing global warming. Journal of Geophysical Research-Atmospheres 107, 4410.
Lewtas, J., Pang, Y., Booth, D., Reimer, S., Eatough, D.J., Gundel, L.A., 2001. Comparison of sampling methods for semi-volatile organic carbon associated with PM2.5. Aerosol Science and Technology 34, 9–22.
Li, H., Feng, J., Sheng, G., Lü, S., Fu, J., Peng, P., Ren, M., 2008. The PCDD/F and PBDD/F pollution in the ambient atmosphere of Shanghai, China. Chemosphere 70, 576–583.
Lonati, G., Ozgen, S., Giugliano, M., 2007. Primary and secondary carbonaceous species in PM2. 5 samples in Milan (Italy). Atmospheric Environment 41, 4599–4610.
Menon, S., Hansen, J., Nazarenko, L., Luo, Y.F., 2002. Climate effects of black carbon aerosols in China and India. Science 297, 2250–2253.
Na, K., Sawant, A.A., Song, C., Cocker, D.R., 2004. Primary and secondary carbonaceous species in the atmosphere of Western Riverside County, California. Atmospheric Environment 38, 1345–1355.
Pandis, S.N., Harley, R.A., Cass, G.R., Seinfeld, J.H., 1992. Secondary organic aerosol formation and transport. Atmospheric Environment (Part A) 26, 2269–2282.
Qiu, J., Yang, L., 2000. Variation characteristics of atmospheric aerosol optical depths and visibility in North China during 1980–1994. Atmospheric Environment 34, 603–609.
Salma, I., Chi, X., Maenhaut, W., 2004. Elemental and organic carbon in urban canyon and background environments in Budapest, Hungary. Atmospheric Environment 38, 27–36.
Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 1999. Measurement of emissions from air pollution sources. 2. C1 through C30 organic compounds from medium duty diesel trucks. Environmental Science and Technology 33, 1578–1587.
Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 2001. Measurement of emissions from air pollution sources. 3. C1–C29 organic compounds from fireplace combustion of wood. Environmental Science and Technology 35, 1716–1728.
Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 2002. Measurement of emissions from air pollution sources. 5. C1–C32 organic compounds from gasoline-powered motor vehicles. Environmental Science and Technology 36, 1169–1180.
Seinfeld, J.H., Pandis, S.N.,1998. Atmospheric Chemistry and Physics: from Air Pollution to Climate Change. Wiley, New York.
Turpin, B.J., Huntzicker, J.J., 1995. Identification of secondary organic aerosol episodes and quantitation of primary and secondary organic aerosol concentrations during SCAQS. Atmospheric Environment 29, 3527–3544.
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, 602–610.
Viana, M., Maenhaut, W., ten Brink, H.M., Chi, X., Weijers, E., Querol, X., Alastuey, A., Mikuska, P., Vecera, Z., 2007. Comparative analysis of organic and elemental carbon concentrations in carbonaceous aerosols in three European cities. Atmospheric Environment 41, 5972–5983.
Viidanoja, J., Sillanpaa, M., Laakia, J., Kerminen, V.-M., Hillamo, R., Aarnio, P., Koskentalo, T., 2002. Organic and black carbon in PM2.5 and PM10: 1 year of data from an urban site in Helsinki, Finland. Atmospheric Environment 36, 3183–3193.
Watson, J.G., Chow, J.C., Houck, J.E., 2001. PM2.5 chemical source profiles for vehicleexhaust, vegetative burning,geological material,and coalburning in Northwestern Colorado during 1995. Chemosphere 43, 1141–1151.
Watson, J.G., Chow, J.C., Chen, L.W.A., 2005. Summary of organic and elemental carbon/black carbon analysis methods and intercomparisons. Aerosol and Air Quality Research 5, 65–102.
Yang, F., He, K., Ye, B., Chen, X., Cha, L., Cadle, S.H., Chan, T., Mulawa, P.A., 2005a. One-year record of organic and elemental carbon in fine particles in downtown Beijing and Shanghai. Atmospheric Chemistry and Physics 5, 1449–1457.
Yang, H., Yu, J.Z., Ho, S.S.H., Xu, J., Wu, W.-S., Wan, C.H., Wang, X., Wang, X., Wang, L., 2005b. The chemical composition of inorganic and carbonaceous materials in PM2.5 in Nanjing, China. Atmospheric Environment 39, 3735–3749.
Ye, S.-H., Zhou, W., Song, J., Peng, B.-C., Yuan, D., Lu, Y.-M., Qi, P.-P., 2000. Toxicity and health effects of vehicle emissions in Shanghai. Atmospheric Environment 34, 419–429.
Ye, B., Ji, X., Yang, H., Yao, X., Chan, C.K., Cadle, S.H., Chan, T., Mulawa, P.A., 2003. Concentration and chemical composition of PM2.5 in Shanghai for a 1-year period. Atmospheric Environment 37, 499–510.
Zhang, Y., Shao, M., Zhang, Y., Zeng, L., He, L., Zhu, B., Wei, Y., Zhu, X., 2007. Source profiles of particulate organic matters emitted from cereal straw burnings. Journal of Environmental Sciences 19, 167–175.
|
atmosphere
|
0018
|
10.1016/j.scitotenv.2014.08.008
|
image
|
Fig. 1. Location of the sampling site Wanqingsha (WQS) and its surrounding environments.
|
|
atmosphere
|
0018
|
10.1016/j.scitotenv.2014.08.008
|
image
|
Fig. 2. Charge balance between cations and anions in all $\mathrm{PM}_{2.5}$ samples.
|
|
atmosphere
|
0018
|
10.1016/j.scitotenv.2014.08.008
|
table
|
Table 1 Concentration of $\mathrm{PM}_{2.5}$ mass, carbonaceous and ionic species in fall and winter from 2007 to 2011 (average ± 95% confidence interval) (unit: μg $\mathfrak{m}^{-3}$ ).
|
|
atmosphere
|
0018
|
10.1016/j.scitotenv.2014.08.008
|
image
|
Fig. 3. Annual variation of $\mathrm{PM}_{2.5}$ mass concentration in fall and winter from 2007 to 2011.
|
|
atmosphere
|
0018
|
10.1016/j.scitotenv.2014.08.008
|
image
|
Fig. 4. The cumulative percentage of $\mathrm{PM}_{2.5}$ OM, $S0_{4}^{2-}$ and $\mathrm{NO}_{3}^{-}$ mass concentrations in fall and winter from 2007 to 2011. The red lines are the different $\mathrm{PM}_{2.5}$ mass concentration standards: WHO 24-h guideline $(25\,\upmu\mathrm{g}\,\mathfrak{m}^{-3}$ ), USEPA 24-h standard $(35\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ and China's new national ambient air quality daily standard guideline $(75\,\upmu\mathrm{g}\,\mathrm{m}^{-3})$ .
|
|
atmosphere
|
0018
|
10.1016/j.scitotenv.2014.08.008
|
image
|
Fig. 5. $\mathrm{PM}_{2.5}$ components in fall and winter from 2007 to 2011.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.