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
|
0101
|
10.1016/j.partic.2016.07.009
|
image
|
Fig. 9. Plot of ${}^{208}\mathrm{Pb}/{}^{207}$ Pb vs. ${}^{207}\mathrm{Pb}/{}^{206}\mathrm{Pb}$ in $\mathrm{PM}_{2.5}$ and source-specific samples.
|
|
atmosphere
|
0101
|
10.1016/j.partic.2016.07.009
|
image
|
Fig. 10. Plot of $^{208}\mathrm{Pb}/^{206}\mathrm{Pb}$ vs. $^{208}\mathrm{Pb}/^{207}\mathrm{Pb}$ in $\mathrm{PM}_{2.5}$ and source-specific samples.
|
|
atmosphere
|
0101
|
10.1016/j.partic.2016.07.009
|
table
|
Table 5 Contributions of different potential sources to Pb in $\mathrm{PM}_{2.5}$ samples (%).
|
|
atmosphere
|
0101
|
10.1016/j.partic.2016.07.009
|
text
| Not supported with pagination yet |
Chemical characteristics and Pb isotopic compositions of $\mathsf{P M}_{2.5}$ in Nanchang, China
Yang Zhao a,b, Ruilian $\mathrm{Yu^{a,*}}$ , Gongren $\mathrm{Hu}^{\,\mathrm{a},\ast}$ , Xiaohui Lin a, Xianrong Liu a
a College of Chemical Engineering, Huaqiao University, Xiamen 361021, China b Environmental Protection Agency of Nanchang City, Nanchang 330038, China
article info
abstract
Article history:
Received 13 April 2016
Received in revised form 17 July 2016
Accepted 22 July 2016
Available online xxx
Keywords:
$\mathrm{PM}_{2.5}$
Water-soluble ions
Trace elements
Pb isotopic composition
Nanchang
In mid-September 2013, $\mathrm{PM}_{2.5}$ samples were collected at six sites in Nanchang, Jiangxi Province, China, to quantify nine water-soluble ions $(C\mathbf{a}^{2+}$ , ${\bf M}{\bf g}^{2+}$ , $\mathsf{K}^{+}$ , $\mathrm{N}\mathbf{a}^{+}$ , $\mathsf{N H_{4}}^{+}$ , $S0_{4}{}^{2-}$ , $C1^{-}$ , $\mathrm{F^{-}}$ , $\mathsf{N O}_{3}^{\mathrm{~-~}}$ , 29 trace elements (Ba, Zn, Pb, Ni, Mo, Cr, Cu, Sr, Sb, Rb, Cd, Bi, Zr, V, Ga, Li, Y, Nb, W, Cs, Tl, Sc, Co, U, Hf, In, Re, Be, and Ta), and to characterize Pb isotopic ratios $(^{207}\mathrm{Pb}/^{206}\mathrm{Pb}$ , $^{208}\mathrm{Pb}/^{206}\mathrm{Pb}$ , and $^{207}\mathrm{Pb}/^{204}\mathrm{Pb})$ for identifying the main source(s) of Pb. The results showed that the average daily $\mathsf{P M}_{2.5}$ concentration $(53.16\!\pm\!24.17)\,\upmu\mathrm{g}/\mathrm{m}^{3}$ was within the secondary level of the Chinese ambient air quality standard. The combined concentrations of $S0_{4}{}^{2-}$ , $\mathsf{N H}_{4}^{\;\;+}$ , and $\mathsf{N O}_{3}^{\mathrm{~-~}}$ to total measured water-soluble ion concentrations in $\mathsf{P M}_{2.5}$ ranged from $79.40\%$ to $95.18\%$ , indicating that anthropogenic sources were significant. Coal combustion and vehicle emissions were both contributors to $\mathsf{P M}_{2.5}$ based on the $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{\mathrm{~2-}}$ ratios. Wushu School experienced the lowest concentrations of $\mathrm{PM}_{2.5}$ and most trace elements among the six sampling sites. Enrichment factor results showed that Tl, Cr, In, Cu, Zn, Pb, Bi, Ni, Sb, and Cd in $\mathrm{PM}_{2.5}$ were affected by anthropogenic activities. Cluster analysis suggested that Cd, Sb, Pb, Re, Zn, Bi, Cs, Tl, Ga, and In were possibly related to coal combustion and vehicle exhaust, while Ni, Nb, Cr, and Mo may have originated from metal smelting. Pb isotopic tracing showed that coal dust, cement dust, road dust and construction dust were the major $\mathsf{P b}$ sources in $\mathrm{PM}_{2.5}$ in Nanchang. Combined, these sources contributed an average of $72.51\%$ of the $\mathrm{Pb}$ measured, while vehicle exhaust accounted for $27.49\%$ of $\mathrm{Pb}$ based on results from a binary Pb isotope mixed model.
$\circledcirc$ 2017 Chinese Society of Particuology and Institute of Process Engineering, Chinese Academy of Sciences. Published by Elsevier B.V. All rights reserved.
Introduction
It is well known that fine atmospheric particles, especially those with diameters less than $2.5\,\upmu\mathrm{m}$ $(\mathsf{P M}_{2.5})$ , can deteriorate air quality and be harmful to human health. Fine particles have relatively long residence times and can be transported over large distances. Because of this, they can also reduce atmospheric visibility through extinction. Fine particles can also adsorb harmful pollutants such as heavy metals, organic matter, bacteria and viruses because of their large specific surface areas. The small particles are easily deposited in the lungs, and can even enter the bloodstream (Chow et al., 2006). Recently, source analyses and health risk assessments of fine particles have become a worldwide focus of study (Hu, Wang, Tao, & Chen, 2013; Liu et al., 2014).
Pb isotopic tracing techniques have been widely used in geological research and have gradually started to be used for studying Pb in the environment. Pb has four naturally stable isotopes ( $^{\cdot204}\mathrm{Pb}$ $^{206}\mathrm{Pb}$ , $^{207}\mathrm{Pb}$ , and $^{208}\mathrm{{Pb}}^{\cdot}$ ) and little isotopic fractionation occurs during industrial and environmental processes (Cheng & Hu, 2010; Zhao et al., 2015). The $\mathsf{P b}$ isotopic composition of an ore or anthropogenic source cannot change during the transition to a secondary form unless there is mixing with secondary $\mathsf{P b}$ sources (Cheng & Hu, 2010). As such, the $\mathsf{P b}$ isotopic compositions in the samples reflect the Pb sources or results of mixing if multiple Pb sources exist (Cheng & Hu, 2010). Therefore, the identification of Pb sources can be performed by matching $\mathsf{P b}$ isotopic compositions at the sampling site with those from potential sources.
Nanchang $\,^{115^{\circ}27^{\prime}\mathrm{E}-116^{\circ}35^{\prime}\mathrm{E}}$ , $28^{\circ}10^{\prime}\mathrm{N}{-}29^{\circ}11^{\prime}\mathrm{N}\rangle$ , the capital of Jiangxi Province, is one of the largest cities in the middle and lower reaches of the Yangtze River in China. Because of rapid development of secondary industries and imperfect environmental management, Nanchang is facing a serious air quality issue because of various anthropogenic sources, such as the coal combustion-based power industry, increased vehicle numbers, and widespread real estate construction. Real-time air quality monitoring data in 2013 showed that Nanchang was suffering from severe air pollution, which was highlighted by several incidences of continuous haze (Qin, Liu, Liang, Cheng, & Li, 2014). Several studies of $\mathsf{P M}_{2.5}$ in Nanchang, including on air quality and environmental protection concerns (Zhang, Yuan et al., 2014), $\mathsf{P M}_{2.5}$ concentration analyses (He, Yuan, & Xiao, 2010), and water-soluble ion distributions (Huang et al., 2012), have been conducted. However, analysis of $\mathsf{P M}_{2.5}$ sources in Nanchang has not been reported. In this study, we measured the concentrations of nine water-soluble ions $(C\mathsf{a}^{2+}$ , ${\mathrm{Mg}}^{2+}$ , $K^{+}$ , $\mathsf{N a}^{+}$ , $\mathrm{NH_{4}}^{+}$ , $S0_{4}{}^{2-}$ , $C1^{-}$ , $\mathsf{F}^{-}$ , $\mathsf{N O}_{3}^{\mathrm{~-~}}$ ) and 29 trace elements (Ba, Zn, Pb, Ni, Mo, Cr, Cu, Sr, Sb, Rb, Cd, Bi, Zr, V, Ga, Li, Y, Nb, W, Cs, Tl, Sc, Co, U, Hf, In, Re, Be, and Ta) in $\mathsf{P M}_{2.5}$ samples collected at six sampling sites in Nanchang. We also identified $\mathsf{P b}$ isotopic ratios $(^{207}\bar{\mathrm{Pb}}/^{206}\mathrm{Pb},^{208}\mathrm{Pb}/^{206}\mathrm{Pb}$ , and $^{207}\mathrm{Pb}/^{204}\mathrm{Pb})$ in the $\mathsf{P M}_{2.5}$ samples collected and some source-specific samples. Using these data, we aimed to identify the sources of $\mathsf{P M}_{2.5}$ so that effective control measures can be put in place to support better air quality.
Methods
Sampling and preparation
In mid-September 2013, $\mathsf{P M}_{2.5}$ samples were collected using medium-volume samplers (TH 150-III, Wuhan, China) at six national monitoring stations in Nanchang, including Wushu School (WS), Forestry Research Institute (FRI), Construction Engineering School (CES), Provincial Foreign Affairs Office (PFAO), Jingdong Town Government (JDTG), and Petrochemical (PC) (Fig. 1). Quartz fiber fliters $\!\!\!\Phi\!=\!90\,\mathrm{{mm})}$ ) were used for quantifying the trace elements and water-soluble ions. Before sampling, the quartz fiber fliters were pre-heated in a muffle furnace at $240\,^{\circ}\mathrm{C}$ for $4\,\mathrm{h}$ to remove the volatile components. Before and after sampling the fliters were stored in desiccators for more than 3 days, after which they were weighed using a precision balance to obtain the mass of $\mathsf{P M}_{2.5}$ collected. Each fliter was weighed at least three times to ensure that the differences between replicate weights were less than $15\,\upmu g$ . Considering the daily activities occurring in Nanchang, 61 source-specific samples were collected using the specific methods. The sampling sites are described briefly in Table 1.
Analysis of water-soluble ions
One-quarter of each loaded fliter was extracted using $10\,\mathrm{mL}$ of distilled deionized water (with a resistivity ${>}18\,\mathrm{M}\Omega$ ) in an ultrasonic bath for $10\,\mathrm{min}$ . After standing the supernatant was used to determine the concentrations of water-soluble ions. Four anions $(\mathsf{S}0_{4}{}^{2-}$ , $\mathrm{NO}_{3}{}^{-}$ , $C1^{-}$ , and $\mathsf{F}^{-}$ ) and five cations $(\mathrm{N}\mathbf{a}^{+},\mathrm{N}\mathsf{H}_{4}^{+},\mathrm{K}^{+}$ ${\mathrm{Mg}}^{2+}$ , and ${\bf C}{\bf a}^{2+}$ ) were quantified using ion chromatography (IC, Dionex DX-100, Dionex Corp., Sunnyvale, CA, USA). Cations were separated using a CS12A column (Dionex Corp.) with $20\,\mathrm{mmol/L}$ methanesulfonic acid (MSA) eluent. Anions were separated using an AS11-HC column (Dionex Corp.) and $0.35\,\mathrm{mmol}/\mathrm{L}\,\mathrm{Na}_{2}\mathrm{CO}_{3}$ and $0.1\,\mathrm{mmol/L\NaHCO_{3}}$ as the eluents. A calibration was performed for each analytical sequence. The detection limits (DLs) were: $\mathsf{N a}^{+}$ $(2.30\,\upmu\mathrm{g/L)}$ , $\mathrm{NH}_{4}{}^{+}\left(1.14\,\upmu\mathrm{g}/\mathrm{L}\right)$ , $\mathrm{K}^{+}\left(8.59\,\upmu\mathrm{g}/\mathrm{L}\right)$ , $\mathrm{Mg}^{2+}\,(7.50\,\upmu\mathrm{g}/\mathrm{L}),$ ${\bf C}{\bf a}^{2+}$ $(10\,\upmu\mathrm{g/L)}$ , $S0_{4}{}^{2-}$ $\left(0.903\,\upmu\mathrm{g/L}\right)$ $,N0_{3}^{\,-}\,(0.844\,\upmu\mathrm{g/L)},\mathrm{Cl^{-}\,(0.242\,\upmu\mathrm{g/L)}},$ $\mathrm{F}^{-}\,(0.635\,\upmu\mathrm{g}/\mathrm{L})$ . Procedural blank values were subtracted from sample concentrations.
Analysis of trace elements
To quantify the concentrations of 29 trace elements in the $\mathsf{P M}_{2.5}$ samples, one-quarter of each loaded fliter was cut into pieces and soaked in $1.6\,\mathrm{mL}$ of concentrated $\mathrm{HNO}_{3}$ and $0.2\,\mathrm{mL}$ of concentrated HF in a Teflon vessel and then digested hermetically at $130\,^{\circ}\mathbf{C}$ for $^{5\,\mathrm{h}}$ and then $180\,^{\circ}\mathrm{C}$ for $^{12\,\mathrm{h}}$ . After cooling down, the vessel was placed on an electric hot plate and heated at $140\,^{\circ}\mathrm{C}$ until dry. A $50\,\upmu\mathrm{g}/\mathrm{L}$ Rh solution $\mathrm{[0.4\,mL)}$ was then added to be used as the internal standard, followed by $0.6\,\mathrm{mL}$ of $\mathrm{HNO}_{3}$ and $0.2\,\mathrm{mL}$ deionized water. The mixture was hermetically heated at $140\,^{\circ}\mathrm{C}$ for $^{5\,\mathrm{h}}$ . After cooling down, the solution was diluted with deionized water to $4\,\mathrm{mL}$ for trace element detection using inductively-coupled plasma mass spectrometry (ICP-MS, ELEMENT XR, USA).
For the analysis of trace element concentrations in each sourcespecific sample, $0.0500\,\mathrm{g}$ of each sample was weighed into a Teflon vessel containing $1\,\mathrm{mL}$ of concentrated $\mathrm{HNO}_{3}$ and $1\,\mathrm{mL}$ of concentrated HF, followed by hermetic heating at $180\,^{\circ}\mathrm{C}$ for $24\,\mathrm{h}$ . After cooling down, the vessel was placed on an electric hot plate and heated at $140\,^{\circ}\mathrm{C}$ until dry. Then, $1\,\mathrm{mL}$ of $50\,\upmu\mathrm{g}/\mathrm{L}$ Rh solution (internal standard) was added, followed by $1.5\,\mathrm{mL}$ of concentrated $\mathrm{HNO}_{3}$ and $0.5\,\mathrm{mL}$ of deionized water. The mixture was then hermetically heated at $140\,^{\circ}\mathrm{C}$ for $^{5\,\mathrm{h}}$ . After cooling down, the solution was diluted with deionized water to $50\,\mathrm{mL}$ for trace element detection using ICP-MS.
Blank values were routinely monitored by simultaneously analyzing blank fliters using the same process as for field samples. The method detection limits ranged from 0.002 to $0.006\,\upmu\mathrm{g/g}$ . The relative standard deviations were less than $5.0\%$ . The standard recoveries ranged between $80\%$ and $120\%$ The final concentration reported for each element was the blank-subtracted value.
Analysis of $P b$ isotopic ratios
Analyses of Pb isotopic compositions were conducted at the Institute of Urban Environment, Chinese Academy of Sciences. The digestion solutions for metal analyses were diluted to approximately $25\,\upmu\mathrm{g}/\mathrm{L}$ Pb using $5\%\ \mathsf{H N O}_{3}$ (metal-oxide-semiconductor pure reagents with very little impurities) based on the measured Pb concentrations. Subsequently, $\boldsymbol{\mathrm{Pb}}$ isotopic ratios were measured using ICP-MS (Agilent $7500c\mathbf{x}$ ). Throughout the process a certified reference material (NBS981, USA) was used for checking the measured accuracies with the certified $^{208}\mathrm{Pb}/^{206}\mathrm{Pb}$ , ${}^{207}\mathrm{Pb}/{}^{206}\mathrm{Pb}$ , and $^{207}\mathrm{Pb}/^{204}\mathrm{Pb}$ values of $2.1681\pm0.0033$ , $0.9146\!\pm\!0.0021$ , and $15.4910\pm0.0005$ , respectively. In this study, the $\boldsymbol{\mathrm{Pb}}$ isotopic ratios of $60\,\mathrm{PM}_{2.5}$ samples and 61 source-specific samples (Table 1) were measured with relative standard deviations of less than $0.5\%$ .
Results and discussion
$P M_{2.5}$ concentrations
The $\mathsf{P M}_{2.5}$ concentrations at six sampling sites in Nanchang during the sampling period are presented in Fig. 2. The mean $\mathsf{P M}_{2.5}$ concentrations $\mathrm{(}\upmu\mathrm{g}/\mathrm{m}^{3})$ at the sampling sites increased from WS $(37.96\pm16.97)\$ <FRI $(47.02\pm26.23)$ <PC $(52.54\pm27.15)<\mathrm{CES}$ $54.51\pm25.83)$ <JDTG $.62.64\pm38.42\rangle$ <PFAO $(64.26\pm28.37)$ . Although the mean daily $\mathsf{P M}_{2.5}$ concentrations at the sampling sites were all lower than the Chinese ambient air quality standard (MEP, 2012) second-grade limit $(75\,\upmu\mathrm{g}/\mathrm{m}^{3})$ , there were still $10\%$ , $40\%$ , $40\%$ , $40\%$ , and $40\%$ of samples at FRI, CES, PFAO, JDTG, and PC, respectively, that exceeded the limit. A comparison of $\mathsf{P M}_{2.5}$ concentrations in Nanchang with those from other cities in China and abroad is presented in Table 2. $\mathsf{P M}_{2.5}$ concentrations in Nanchang were lower than those in Guangzhou (Feng, Ming, Liu, Zhang, & Zheng, 2015) and Xi’an (Gan, Cao, Wang, & Shen, 2011), comparable with those in Nanjing (Shen, Wang, Chen, & Jiang, 2015), and much higher than those in Kowloon Tong in Hong Kong (Jiang, Yang, Chan, & Ning, 2014), and Seoul, Korea (Ahmed, Kim, Shon, & Song, 2015). The mean $\mathsf{P M}_{2.5}$ concentration in this study was lower than that $(63.65\,\upmu\mathrm{g}/\mathrm{m}^{3})$ found from July to August 2007 in Nanchang (He et al., 2010).
Water-soluble ions and pollution sources in $P M_{2.5}$
Fine particles in the atmosphere can promote the formation of cloud condensation nuclei because of the hydrophilicity of water-soluble ions and can have a significant impact on climate and visibility. The measured concentrations of water-soluble ions in $\mathsf{P M}_{2.5}$ collected at the six sampling sites are presented in Fig. 3. The mean concentrations of water-soluble ions decreased in the order of $\mathrm{SO_{4}}^{2-}>\mathrm{NO_{3}}^{-}>\mathrm{NH_{4}}^{+}>\mathrm{Ca}^{2^{+}}>\mathrm{K}^{+}>\mathrm{Na}^{+}>\mathrm{Cl}^{-}>\mathrm{Mg}^{2^{+}}>\mathrm{F}^{-}$ . Concentrations of $\mathsf{N O}_{3}^{\mathrm{~-~}}$ were higher at CES, PFAO, and JDTG, while $S0_{4}{}^{2-}$ concentrations were higher at FRI, PFAO, and JDTG. Concentrations of the other water-soluble ions also showed differences among the sampling sites, indicating that sources differed among the sites.
The percentage contributions of the measured water-soluble ions to total $\mathsf{P M}_{2.5}$ mass concentrations at different sites are presented in Fig. 4. The mean percentage contributions of the total measured water-soluble ions to $\mathsf{P M}_{2.5}$ concentrations at the different sampling sites decreased in the order of PFAO $(39.08\%)>\!\mathrm{PC}$ $(34.45\%)>\mathtt{F R I}\,(33.87\%)>\mathtt{C E S}\,(30.35\%)>\mathtt{W S}\,(29.31\%)>\mathtt{J D T G}\,(28.57\%),$ indicating that water-soluble ions were significant contributors to $\mathsf{P M}_{2.5}$ mass in Nanchang. The overall mean percentages of different water-soluble ions to $\mathsf{P M}_{2.5}$ mass concentrations decreased in the order of $\mathrm{SO_{4}}^{2-}\left(13.63\%\right)\!>\!\mathrm{NO_{3}}^{-}\left(8.89\%\right)\!>\!\mathrm{NH_{4}}^{+}\left(6.06\%\right)\!>\!\mathrm{Ca}^{2+}\left(1.22\%\right)$ $>\mathrm{K}^{+}\;(1.03\%)>\mathrm{Na}^{+}\;(0.88\%)>\mathrm{Cl}^{-}\;(0.59\%)>\mathrm{N}$ $\cdot8\%>\!\mathrm{Cl}^{-}\;(0.59\%)>\!\mathrm{Mg}^{2+}\;(0.17\%)>\!\mathrm{F}^{-}\;(0.13\%).$ The secondary compounds $S0_{4}{}^{2-}$ , $\mathsf{N H}_{4}^{\,\,+}$ , and $\mathrm{NO}_{3}{}^{-}$ were the main water-soluble ions, contributing $79.40\%{-95.18\%}$ of the total measured water-soluble ions, highlighting that secondary pollution is a significant source of $\mathsf{P M}_{2.5}$ . $S0_{4}{}^{2-}$ contributed $13.95\%{-}69.86\%$ ( $\mathrm{mean}\!=\!39.24\%$ to the total measured water-soluble ions, suggesting that coal combustion was an important source of $\mathsf{P M}_{2.5}$ pollution.
The relationship between the anions and the cations, expressed as the sum of the respective equivalent concentration $\left(\upmu\mathrm{mol}/\upmu\mathrm{n}^{3}\right)$ of the measured water-soluble ions, is presented in Fig. 5. The cation equivalent (CE) and anion equivalent (AE) concentrations were calculated using Eqs. (1) and (2), respectively.
$$
\begin{array}{r l}&{C E(\mathrm{Cation\;equivalent})=\cfrac{{\mathrm{N}}\mathsf{a}^{+}}{23.0}+\cfrac{\mathrm{K}^{+}}{39.1}+\cfrac{{\mathrm{NH}}_{4}^{+}}{18.0}+\cfrac{{\mathrm{M}}\mathsf{g}^{2+}}{12.2}+\cfrac{\mathrm{C}\mathsf{a}^{2+}}{20.0}\,,}\\ &{A E(\mathrm{Anion\;equivalent})=\cfrac{\mathrm{F}^{-}}{19.0}+\cfrac{\mathrm{C}\mathsf{I}^{-}}{35.5}+\cfrac{{\mathrm{N}}\mathsf{O}_{3}^{-}}{62.0}+\cfrac{50_{4}^{2-}}{48.0}\,.}\end{array}
$$
The AE showed a significant correlation $'R^{2}\,{=}\,0.6082$ , Fig. 5) with the $C E$ and the slope $(A E/C E)$ of the linear regression was 1.0772, indicating no apparent absence of ions. However, $A E\!>\!C E$ indicated that the particles were acidic, which is similar to the ionic components of $\mathsf{P M}_{2.5}$ during winter in the Beijing-Tianjin-Hebei area (Dao et al., 2015).
Ratios of $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{\mathrm{~2-}}$ have been used as an indicator of the relative contribution of coal combustion (stationary source) compared to fuel burning (mobile source) in the atmosphere (Xiu et al., 2004 ; Yao et al., 2002; Zhao et al., 2011). The stationary source should be considered dominant if $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{\mathrm{~2-}}$ is less than 1, otherwise vehicular sources should be considered dominant (Yao et al., 2002). In this study, the mean value of $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{\mathrm{~2-}}$ in $\mathsf{P M}_{2.5}$ was 0.979, which is comparable with the value of 0.996 in Shanghai from 2011 to
2012 (Ma, 2014), 0.77 in Nanjing from 2010 to 2011 (Zhang, Zhu, & Gong, 2014), and 0.56–1.85 in the Beijing-Tianjin-Hebei area in 2013 (Dao et al., 2015), but much higher than data previous result (0.10) from Nanchang in 2009 (Huang et al., 2012) and Guiyang (0.13) in 2003 (Xiao & Liu, 2004). Our results suggest that the contribution of vehicle exhaust to $\mathsf{P M}_{2.5}$ has significantly increased and this is likely because of a rapid increase in the number of motor vehicles. This work suggests that vehicle exhaust had almost the same contribution as coal combustion to measured $\mathsf{P M}_{2.5}$ concentrations in Nanchang.
Characteristics of trace elements in $P M_{2.5}$
Concentrations of 29 trace elements in $\mathsf{P M}_{2.5}$ samples collected at the six sampling sites are presented in Table 3. Among the measured elements, Ba, Zn, Pb, Ni, Mo, Cr, and Cu were the most abundant. Pb concentrations in the $\mathsf{P M}_{2.5}$ samples from Nanchang were all within the limit $\mathrm{[1000\,ng/m^{3}]}$ of the National Ambient Air Quality Standard of China (GB3095-2012) (MEP, 2012). The concentrations of most measured elements at WS were lower than those at other sampling sites, likely because WS was a suburban location in Wanli district near the Meiling scenic area that experienced fewer anthropogenic activities. The CES, PFAO, and PC sites experienced the highest concentrations of most of the measured elements. These sites were located near areas where substantial infrastructure improvements were being carried out during the sampling period, such as an ongoing subway construction, numerous demolitions of old buildings and construction of new buildings (Zhang, Yuan et al., 2014).
Enrichment factors (EFs) relative to soil background values have been used to evaluate the influence of anthropogenic activities on the natural environment (Ji, Zhu, Feng, Bai, & Liu, 2006). Considering its low variation coefficient (0.437) and high correlations with other elements, Rb was chosen as the reference element to calculate the enrichment factors in this study. Enrichment factors were calculated using Eq. (3):
$$
E F=\left(C_{i}/C_{\mathrm{Rb}}\right)_{\mathrm{PM}_{2.5}}/\left(C_{i}/C_{\mathrm{Rb}}\right)_{\mathrm{soil}},
$$
where Ci/CRb PM and $\left(C_{i}/C_{\mathrm{Rb}}\right)_{\mathrm{soil}}$ are the ratios of element i to the reference element, Rb, in $\mathsf{P M}_{2.5}$ and the background value in Alayer soil from the Jiangxi Province (MEP, 1990), respectively. The background values of Mo, Re, and Nb were not available. As shown in Fig. 6, the EF values of Ta, Zr, Be, Y, Li, Co, V, Hf, Cs, Ga, Sc, U, W, and Sr were lower than 10, indicating little enrichment relative to the Earth’s crust (Dai, 2006). However, the EF values of Ba, Tl, Cr, In, Cu, Zn, Pb, Bi, Ni, Sb, and Cd were higher than 10, indicating anthropogenic impacts. The $E F$ values for Cd ranged from 371.9 to 3741, revealing significant impacts related to coal burning activities (Duan & Tan, 2013).
Hierarchical cluster analysis was performed according to Ward’s method using the concentration data from all of the measured trace elements in the $\mathsf{P M}_{2.5}$ samples to obtain information on potential pollution sources. The results of this analysis are presented in Fig. 7. TheelementsU,Hf,Y,Zr,Sc,Sr,Ba,Cu,Li,Be,Co,Rb,Ta,V,and W were from rock weathering or soil and had low EFs. The links among Cd, Sb, Pb, Re, Zn, Bi, Cs, Tl, Ga, and In indicated possible coal combustion and motor vehicle exhaust sources, particularly because Cd is typically found in fly ash from coal burning and from fuel burning (Tian et al., 2010), and Pb, Zn and Cd originate from vehicular exhaust emission, lubricating oil, tire and brake wearing (Duan & Tan, 2013). This result was similar to that reported by Lin, Zhao, Fan, Hu, and Yu (2016). However, Ni, Nb, Cr, and Mo could also originate from metal smelting (Duan & Tan, 2013).
Pb isotopic compositions in $P M_{2.5}$
The main sources of $\mathsf{P b}$ in $\mathsf{P M}_{2.5}$ include vehicles, metal refining plants, coal combustion, cement factories, and soil erosion (Widory, Liu, & Dong, 2010). However, there are only a few studies on Pb isotopic compositions of local Chinese coal combustion and vehicle exhaust samples. Pb concentrations and isotopic ratios in the $\mathsf{P M}_{2.5}$ and source-specific samples from Nanchang showed considerable differences (Table 4). Pb concentrations in the $\mathsf{P M}_{2.5}$ samples were far higher than those in the source-specific samples, showing considerable Pb enrichment. Pb isotopic compositions in the $\mathsf{P M}_{2.5}$ samples were within the ranges of the source-specific samples, indicating effects related to different sources, but the $\boldsymbol{\mathrm{Pb}}$ sources contributing to $\mathsf{P M}_{2.5}$ samples at different sampling sites could not be distinguished clearly through comparisons of the Pb isotopes.
A plot of $^{206}\mathrm{Pb}/^{207}\mathrm{Pb}\,\mathrm{vs}.1/[\mathrm{Pb}]$ can be used to identify Pb sources (Álvarez-Iglesias, Rubio, & Millos, 2012; N’Guessan, Probst, Bur, & Probst, 2009). We found that $^{206}\mathrm{Pb}/^{207}\mathrm{Pb}$ of our samples was not related to $1/[\mathsf{P b}]$ $R^{2}\!=\!0.03493$ , Fig. 8), suggesting that $\mathsf{P b}$ variations were from more than two sources with different $\boldsymbol{\mathrm{Pb}}$ isotopic compositions.
A plot of $^{208}\mathrm{Pb}/^{207}\mathrm{Pb}$ vs. $^{207}\mathrm{Pb}/^{206}\mathrm{Pb}$ for the $\mathsf{P M}_{2.5}$ and sourcespecific samples is presented in Fig. 9. Pb isotopic compositions of most $\mathsf{P M}_{2.5}$ samples (except one sample collected at PFAO) fell outside the range of motor vehicle exhaust dust, indicating that Pb concentrations in $\mathsf{P M}_{2.5}$ samples from Nanchang were less influenced by motor vehicle exhaust than in other locations. This result is consistent with the banning of leaded gasoline in China since 2000. Also, Pb isotopic ratios from most soil dust samples were not close to those of the collected $\mathsf{P M}_{2.5}$ samples, indicating that soil dust was also not a significant contributor to $\mathsf{P M}_{2.5}$ in Nanchang. Most $\mathsf{P M}_{2.5}$ samples were connected with cement dust, road dust, coal combustion, and construction dust. Smelting dust occasionally affected the $\mathsf{P b}$ isotopic composition in $\mathsf{P M}_{2.5}$ samples from JDTG, which was located near the Taihao high-tech industrial zone, which featured several smelting-related activities.
With the development of $\boldsymbol{\mathrm{Pb}}$ isotope tracing technology, major sources of $\mathsf{P b}$ have been analyzed using binary or ternary mixed models (Yu, Hu, Yuan, & Zhao, 2009). In this study, $^{208}\bar{\mathrm{Pb}}/^{206}\mathrm{Pb}$ and $^{208}\mathrm{Pb}/^{207}\mathrm{Pb}$ were chosen to calculate the contributions of Pb sources because of their strong correlation. The plot of $^{208}\mathrm{Pb}/^{206}\mathrm{Pb}$ vs. $^{208}\mathrm{Pb}/^{207}\mathrm{Pb}$ (Fig. 10) shows that the $\mathsf{P b}$ isotopic compositions of most $\mathsf{P M}_{2.5}$ samples (except one sample collected at PFAO) ranged between motor vehicle exhaust and other mixed anthropogenic sources, and were closest to construction dust, road dust, smelting dust and coal combustion.
The result is consistent with $\mathsf{208\,P b/^{207}P b\,v s.^{207}P b/^{206}P b},$ , and fulflils the requirement of the binary mixed model (Monna, Lancelot, Croudace, Cundy, & Lewis, 1997). As such, a binary mixed model was used to calculate the contribution of two pollution groups using Eqs. (4) and (5):
$$
f_{1}=\frac{R_{S}-R_{2}}{R_{1}-R_{2}},
$$
$$
f_{2}=1-f,\kappa
$$
where $R_{s}$ is the $^{208}\mathrm{Pb}/^{207}\mathrm{Pb}$ of $\mathsf{P M}_{2.5}$ samples, $R_{1}$ represents the average (2.4292) of $^{208}\mathrm{Pb}/^{207}\mathrm{Pb}$ in motor vehicle exhaust, $R_{2}$ represents the average (2.1079) of $^{208}\mathrm{Pb}/^{207}\mathrm{Pb}$ for the mixed sources (construction dust, road dust, smelting dust, and coal combustion), and $f_{1}$ and $f_{2}$ are the relative contributions. As shown in Table 5, the average contributions of mixed sources and motor vehicle exhaust source were $72.51\%$ and $27.49\%$ , respectively. The PC site was most affected by vehicle exhaust because gas stations were located nearby.
Conclusions
Concentrations of $\mathsf{P M}_{2.5}$ in Nanchang during autumn 2013 were lower than the Chinese second-grade limit value $(75\,\upmu\mathrm{g}/\mathrm{m}^{3})$ ). $S0_{4}{}^{2-}$ , $\mathsf{N H}_{4}^{+}$ , and $\mathsf{N O}_{3}^{\ensuremath{-}}$ were the main water-soluble ions in the $\mathsf{P M}_{2.5}$ samples, and there was no absence of cations and/or anions evident.
Coal combustion and vehicle emissions were both contributors to $\mathsf{P M}_{2.5}$ in Nanchang according to the $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{\mathrm{~2-}}$ ratio. Tl, Cr, In, Cu, Zn, Pb, Bi, Ni, Sb, and Cd showed enrichment relative to the earth crust, reflecting significant anthropogenic impacts. Cluster analysis results showed that the trace elements U, Hf, Y, Zr, Sc, Sr, Ba, Cu, Li, Be, Co, Rb, Ta, V, and W were mainly derived from rock weathering or Earth’s crust, and Cd, Sb, Pb, Re, Zn, Bi, Cs, Tl, Ga, and In were likely from coal combustion and motor vehicle exhaust, while Ni, Nb, Cr, and Mo could be from metal smelting. Pb isotopic compositions showed that coal combustion, cement dust, road dust, and construction dust were the main anthropogenic sources of Pb in $\mathsf{P M}_{2.5}$ . More specifically, Pb in Nanchang mainly originated from motor vehicle exhaust $(27.49\%)$ and mixed sources $(72.51\%)$ according to the combined results of a Pb isotopic analysis diagram and a binary mixed model, where the PC site was most affected by vehicle exhaust emissions.
Acknowledgements
This work was supported by the National Natural Science Foundation of China (21377042, 21477042) and the Natural Science Foundation of Fujian Province (2016 J01065). The authors express their heartfelt thanks to the colleagues who participated in the sampling work. Mr. James Ing, a native English speaker from the UK is gratefully acknowledged for reviewing this paper prior to resubmission. We thank the anonymous reviewers for their constructive comments.
References
Ahmed, E., Kim, K. H., Shon, Z. H., & Song, S. K. (2015). Long-term trend of airborne particulate matter in Seoul, Korea from 2004 to 2013. Atmospheric Environment, 101, 125–133.
Álvarez-Iglesias, P., Rubio, B., & Millos, J. (2012). Isotopic identification of natural vs. anthropogenic lead sources in marine sediments from the inner Ría de Vigo (NW Spain). Science of the Total Environment, 437, 22–35.
Cheng, H. F., & Hu, Y. N. (2010). Lead (Pb) isotopic fingerprinting and its applications in lead pollution studies in China: A review. Environmental Pollution, 158(5), 1134–1146.
Chow, J. C., Watson, J. G., Mauderly, J. L., Costa, D. L., Wyzga, R. E., Vedal, S., et al. (2006). Health effects of fine particulate air pollution: Lines that connect. Journal of the Air & Waste Management Association, 56(10), 1368–1380.
Dai, S. G. (2006). Environmental chemistry. Beijing: Higher Education Press (in Chinese).
Dao, X., Zhang, L. L., Wang, C., Chen, Y., Lyu, Y. B., & Teng, E. J. (2015). Characteristics of mass and ionic compounds of atmospheric particles in winter and summer of Beijing-Tianjin-Hebei area, China. Environmental Chemistry, 34(1), 60–69 (in Chinese).
Duan, J. C., & Tan, J. H. (2013). Atmospheric heavy metals and arsenic in China: Situation, sources and control policies. Atmospheric Environment, 74, 93–101.
Feng, X. D., Ming, C. B., Liu, H., Zhang, Y. H., & Zheng, M. (2015). Microscopic morphology and size distribution of $\mathrm{PM}_{2.5}$ in Guangzhou urban area in fall 2011. China Environmental Science, 35(4), 1013–1018 (in Chinese).
Gan, X. F., Cao, J. J., Wang, Q. Y., & Shen, Z. X. (2011). Concentration characteristics and sources of chemical elements in atmospheric fine particles $\left(\mathrm{PM}_{2.5}\right)$ ) in autumn in Xi’an city. Meteorological and Environmental Research, 2(4), 5–8 (in Chinese).
He, Z. J., Yuan, S. L., & Xiao, M. (2010). Pollution levels of airborne particulate matter $\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5}$ in summer in Nanchang city. Journal of Anhui Agricultural Sciences, 38(3), 1336–1338, 1384. (in Chinese).
Hu, Z. M., Wang, J., Tao, Z. K., & Chen, Z. L. (2013). Pollution level and health risk assessment of heavy metals in $\mathrm{PM}_{2.5}$ Shanghai. Acta Scientiae Circumstantiae, 33(12), 3399–3406 (in Chinese).
Huang, H., Zou, C. W., Cao, J. J., Tsang, P. K., Zhu, F. X., Yu, C. L., et al. (2012). Watersoluble ions in $\mathrm{PM}_{2.5}$ on the Qianhu campus of Nanchang University, Nanchang City: Indoor-outdoor distribution and source implications. Aerosol and Air Quality Research, 12(3), 435–443.
Ji, Y. Q., Zhu, T., Feng, Y. C., Bai, Z. P., & Liu, H. X. (2006). Application of the enrichment factor to analyze the pollution of elements in soil dust in China. Acta Scientiarum Naturalium Universitatis Nankaiensis, 29(2), 94–99.
Jiang, S. Y. N., Yang, F. H., Chan, K. L., & Ning, Z. (2014). Water solubility of metals in coarse PM and $\mathrm{PM}_{2.5}$ in typical urban environment in Hong Kong. Atmospheric Pollution Research, 5(2), 236–244.
Liu, F. L., Lu, X., Wu, M. L., Liu, J., Ren, Y. R., & Guo, Z. B. (2014). Pollution characters and source apportionment of heavy metals in $\mathrm{PM}_{2.1}$ in Nanjing. Chinese Journal of Environmental Engineering, $8(2)$ , 652–658 (in Chinese).
Lin, X. H., Zhao, Y., Fan, X. J., Hu, G. R., & Yu, R. L. (2016). Enrichment characteristics and source analysis of metal elements in $\mathrm{PM}_{2.5}$ in autumn in Nanchang City. Environmental Science, 37(1), 35–40 (in Chinese).
Ma, J. L. (2014). Characteristics and sources analysis of $\mathrm{PM}_{2.5}$ in Baoshan District of Shanghai. Environmental Science and Management, 39(4), 120–125 (in Chinese).
MEP (Ministry of Environmental Protection of the People’s Republic of China). (2012). Ambient air quality standard (GB 3095-2012).. Retrieved from Ministry of Environmental Protection of the People’s Republic of China website: http:// hbj.new.cqcs.gov.cn/upflies/2013-3/2013327153015207.pdf. (in Chinese)
MEP (Ministry of Environmental Protection of the People’s Republic of China). (1990). Chinese soil element background value. pp. 330–483. Beijing: China Environmental Press (in Chinese).
Monna, F., Lancelot, J., Croudace, I. W., Cundy, A. B., & Lewis, J. T. (1997). Pb isotopic composition of airborne particulate material from France and the Southern United Kingdom: Implications for Pb pollution sources in urban areas. Environmental Science and Technology, 31(8), 2277–2286.
N’Guessan, Y. M., Probst, J. L., Bur, T., & Probst, A. (2009). Trace elements in stream bed sediments from agricultural catchments (Gascogne region, S–W France): Where do they come from? Science of the Total Environment, 407(8), 2939–2952.
Qin, W., Liu, X. Z., Liang, Y., Cheng, D. Z., & Li, L. J. (2014). Research on the pollution of $\mathrm{PM}_{2.5}/\mathrm{PM}_{10}$ in Nanchang city in 2013. Science Mosaic, (9), 31–34 (in Chinese).
Shen, T. D., Wang, T. J., Chen, P. L., & Jiang, Z. Q. (2015). Relationship between atmospheric visibility and chemical composition of $\mathrm{PM}_{2.5}$ in the summer and autumn of Nanjing. China Environmental Science, 35(3), 652–658 (in Chinese).
Tian, H. Z., Wang, Y., Xue, Z. G., Cheng, K., Qu, Y. P., Chai, F. H., et al. (2010). Trend and characteristics of atmospheric emissions of Hg, As, and Se from coal combustion in China, 1980–2007. Atmospheric Chemistry $\delta\sigma$ Physics, 10(23), 11905–11919.
Widory, D., Liu, X. D., & Dong, S. P. (2010). Isotopes as tracers of sources of lead and strontium in aerosols (TSP & $\mathrm{PM}_{2.5}$ ) in Beijing. Atmospheric Environment, 44(30), 3679–3687.
Xiao, H. Y., & Liu, C. Q. (2004). Chemical characteristics of water-soluble components in TSP over Guiyang, SW China, 2003. Atmospheric Environment, 38(37), 6297–6306.
Xiu, G. L., Zhang, D. N., Chen, J. Z., Huang, X. J., Chen, Z. X., Guo, H. L., et al. (2004). Characterization of major water-soluble inorganic ions in size-fractionated particulate matters in Shanghai campus ambient air. Atmospheric Environment, 38(2), 227–236.
Yao, X. H., Chan, C. K., Fang, M., Cadle, S., Chan, T., Mulawa, P., et al. (2002). The watersoluble ionic composition of $\mathrm{PM}_{2.5}$ in Shanghai and Beijing, China. Atmospheric Environment, 36(26), 4223–4234.
Yu, R. L., Hu, G. R., Yuan, X., & Zhao, Y. H. (2009). Development in research on pollution source of heavy metals from atmospheric dust-recognition and analysis. Earth and Environment, 37(1), 73–79 (in Chinese).
Zhang, L. J., Yuan, Z. K., Maddock, J. E., Zhang, P., Jiang, Z. Q., Lee, T., et al. (2014). Air quality and environmental protection concerns among residents in Nanchang, China. Air Quality, Atmosphere & Health, 7(4), 441–448.
Zhang, Q. C., Zhu, B., & Gong, D. L. (2014). Characterization of water-soluble ion species of aerosol in Nanjing, China. China Environmental Science, 34(2), 311–316 (in Chinese).
Zhao, P. S., Zhang, X. L., Meng, W., Yang, B. Y., Fan, W. Y., & Liu, H. Y. (2011). Characteristics of inorganic water-soluble ions from aerosols in Beijing-Tianjin-Hebei area. Environmental Science, 32(6), 1546–1549.
Zhao, Z. Q., Zhang, W., Li, X. D., Yang, Z., Zheng, H. Y., Ding, H., et al. (2015). Atmospheric lead in urban Guiyang, Southwest China: Isotopic source signatures. Atmospheric Environment, 115, 163–169.
|
atmosphere
|
0102
|
10.1002/jgrd.50599
|
image
|
Figure 1. Geographical location of the sampling site. Adapted from www.earthobservatory.nasa.gov.
|
|
atmosphere
|
0102
|
10.1002/jgrd.50599
|
table
|
Table 1. Seasonal Meteorological Characteristics at Lulang
|
|
atmosphere
|
0102
|
10.1002/jgrd.50599
|
image
|
Figure 2. Temporal variations of analyzed chemical species in total suspended particle (TSP) samples from Lulang (July 2008 to July 2009) (yellow: episode type I; gray: episode type II).
|
|
atmosphere
|
0102
|
10.1002/jgrd.50599
|
image
|
Figure 3. Three day air backward-in-time air mass trajectory analysis for high-aerosol loading episodes at Lulang.
|
|
atmosphere
|
0102
|
10.1002/jgrd.50599
|
table
|
Table 2. Arithmetic Averages $\pm$ Standard Deviations $(\upmu\mathrm{g}\:\mathrm{m}^{-3})$ for Mass and Chemical Components During Four Seasons
|
|
atmosphere
|
0102
|
10.1002/jgrd.50599
|
image
|
Figure 4. Relationship between the concentrations of organic carbon (OC) and elemental carbon (EC) in four seasons.
|
|
atmosphere
|
0102
|
10.1002/jgrd.50599
|
table
|
Table 3. Comparison of Major Chemical Components From This Study With Measurements at Other Background Sites
|
|
atmosphere
|
0102
|
10.1002/jgrd.50599
|
image
|
Figure 5. Seasonal and episodic variations of mass concentrations and fractions of organic matter, elemental carbon, nitrate, sulfate, sea salt, and crustal material normalized to aerosol mass concentration.
|
|
atmosphere
|
0102
|
10.1002/jgrd.50599
|
image
|
Figure 6. Enhancement factors for selected species in total suspended particles at Lulang during eight high-aerosol loading events. The enhancement factor for species $i\:(\mathrm{EnF}_{i})$ is the ratio of that species’ TSP mass fraction on the designated date to the ratio of its annual average concentration to the annual average of TSP (i.e., $\begin{array}{r}{\mathrm{EF}_{i k}=\mathrm{F}_{i k}{\tilde{/}}\big(\!\sum_{n=1}^{K}\!\mathrm{F}_{i k}\mathrm{TSP}_{n}/\!\sum_{n=1}^{K}\!\mathrm{TSP}_{n}\big)}\end{array}$ , where $\mathrm{EnF}_{i k}\!=$ enhancement factor for species $i$ on sample $k$ , $\mathrm{F}_{i k}\!=$ ratio of species $i$ concentration to TSP concentration on sample $k,K$ total number of samples, and $\mathrm{TSP}_{n}\,{=}\,\mathrm{TSP}$ concentration on sample n).
|
|
atmosphere
|
0102
|
10.1002/jgrd.50599
|
image
|
Figure 7. NCEP/NCAR wind trajectory analysis at $500\ \mathrm{hPa}$ and satellite (MODIS) integrated aerosol optical depth (AOD) for typical high-aerosol episode days: (a, b, and c) NCEP/NCAR wind trajectories for 17 to 20 March and (d) average AOD for 17 to 20 March. (The yellow circle: sampling site).
|
|
atmosphere
|
0102
|
10.1002/jgrd.50599
|
image
|
Figure 8. NCEP/NCAR wind trajectory integrated UV-AI (OMI) for typical episode days: (a, b, and c) NCEP/NCAR wind trajectories for 18 to 21 May and (d) UV-AI from OMI for 17 to $^{20\,\mathrm{M}}$ arch.
|
|
atmosphere
|
0102
|
10.1002/jgrd.50599
|
image
|
Figure 9. Source apportionments by positive matrix factorization (PMF) receptor model. (a) Source profiles for total suspended particles from PMF, (b) factor contributions from PMF in relation to sampling time, and (c) factor loading for TSP mass (yellow arrow: episode type I; gray arrow: episode type II).
|
|
atmosphere
|
0102
|
10.1002/jgrd.50599
|
image
|
Figure 9. (continued)
|
|
atmosphere
|
0102
|
10.1002/jgrd.50599
|
image
|
Figure 10. Detailed wind patterns and the corresponding ultraviolet aerosol index (AI) distributions from the Ozone Monitoring Instrument: (a, b, and c) NCEP/NCAR wind trajectories for 11 to 14 March and (d) UV-AI for 11 to 14 March.
|
|
atmosphere
|
0102
|
10.1002/jgrd.50599
|
text
| Not supported with pagination yet |
Aerosol particles at a high-altitude site on the Southeast Tibetan Plateau, China: Implications for pollution transport from South Asia
Zhuzi Zhao,1 Junji Cao,1 Zhenxing Shen,2 Baiqing Xu,3 Chongshu Zhu,1 L.-W. Antony Chen,4 Xiaoli Su,1 Suixin Liu,1 Yongming Han,1 Gehui Wang,1 and Kinfai Ho 1,5
Received 30 October 2012; revised 16 May 2013; accepted 23 June 2013; published 4 October 2013.
[1] Bulk aerosol samples were collected from 16 July 2008 to 26 July 2009 at Lulang, a high-altitude $\mathord{\left/{\vphantom{\left(-333000\right)}}\right.\kern-\nulldelimiterspace}>33000\mathrm{m}$ above sea level) site on the southeast Tibetan Plateau (TP); objectives were to determine chemical characteristics of the aerosol and identify its major sources. We report aerosol (total suspended particulate, TSP) mass levels and the concentrations of selected elements, carbonaceous species, and water-soluble inorganic ions. Significant buildup of aerosol mass and chemical species (organic carbon, element carbon, nitrate, and sulfate) occurred during the premonsoon, while lower concentrations were observed during the monsoon. Seasonal variations in aerosol and chemical species were driven by precipitation scavenging and atmospheric circulation. Two kinds of high-aerosol episodes were observed: one was enriched with dust indicators (Fe and $\mathrm{Ca}^{2+}$ ), and the other was enhanced with organic and elemental carbon (OC and EC), $\mathrm{SO}_{4}{}^{2-}$ , $\mathrm{NO}_{3}\,^{-}$ , and Fe. The TSP loadings during the latter were 3 to 6 times those on normal days. The greatest aerosol optical depths (National Centers for Environmental Protection/National Center for Atmospheric Research reanalysis) occurred upwind, in eastern India and Bangladesh, and trajectory analysis indicates that air pollutants were transported from the southwest. Northwesterly winds brought high levels of natural emissions (Fe, $\mathrm{Ca}^{2+}$ ) and low levels of pollutants $(\mathrm{SO}_{4}^{\;\;\bar{2}-},\mathrm{NO}_{3}^{\;\;-},\mathrm{K}^{+};$ , and EC); this was consistent with high aerosol optical depths over the western deserts and Gobi. Our work provides evidence that both geological and pollution aerosols from surrounding regions impact the aerosol population of the TP.
Citation: Zhao, Z., et al. (2013), Aerosol particles at a high-altitude site on the Southeast Tibetan Plateau, China: Implications for pollution transport from South Asia, J. Geophys. Res. Atmos., 118, 11,360–11,375, doi:10.1002/jgrd.50599.
1. Introduction
[2] The Tibetan Plateau (TP), the world’s highest and largest plateau, exerts a major influence on the monsoonal circulation in Asia [An et al., 2001]. Moreover, the TP not only affects large-scale atmospheric circulation, but it also impacts the hydrological cycle of the entire Asian continent. These effects are mediated through its dynamical and thermal influences on the atmosphere as well as through snow and glacier melt [Lau et al., 2010; Wu et al., 2007].
[3] The TP is at the juncture of several important natural and anthropogenic aerosol sources, including the Taklimakan and Gobi Desert to the north and northwest, deserts in southwestern Asia and the Middle East to the west and southwest, densely populated areas in the Indo-Gangetic Plain, and areas in South Asia where biomass is burned extensively [Xia et al., 2011]. Transport of pollutants from the densely populated countries (such as India, Pakistan, China, and Nepal) to the Himalayas may not only result in substantial radiative forcing in South Asia with potential effects on the monsoon circulation but also on regional climate and hydrological cycles [Venzac et al., 2008]. Black carbon (BC) that is deposited on the TP absorbs solar radiation, and it is one of the most important substances that cause the melting of ice and snow. Recent studies show that over the past decade, Tibetan glaciers have been melting at an accelerating rate, raising the threat of future shortages in the water supply in neighboring countries [Xu et al., 2009]. [4] High-elevation sites are well suited for sampling the free troposphere, and studies at those sites are particularly useful for characterizing typical background conditions and investigating the influences of human activity on its composition. In the past decade, numerous scientific studies have been conducted in the Indo-Asia-Pacific region, and an extensive observational system has been developed within the Indian Ocean Experiment project [Ramanathan and Crutzen, 2003]. Studies at these sites have served to better understand the “brown cloud” which can extend from the Indian Ocean to the Himalayan range, attain a vertical thickness of about $3~\mathrm{km}$ [Ramanathan et al., 2007] during the dry season (especially from November to March), and affect some of the most populous Asian regions, currently home of more than 2 billion persons.
[5] Atmospheric aerosols build up over the southern slope of the TP during the premonsoon season, and they are lifted by the Himalayan topography [Xia et al., 2011]. Aerosol mass and BC concentrations increased significantly over the southeastern TP during a dry period when numerous fires burned, and there was transport of pollution from nearby regions in South Asia and the northern part of the Indian Peninsula [Engling et al., 2011]. Some previous investigators have already suggested that the southern Himalayas are affected by significant amounts of pollution that is either uplifted by the typical valley circulation or advected by regional and long-range transport events [Hindman and Upadhyay, 2002; Bonasoni et al., 2008]. Observations at the Nepal Climate Observatory-Pyramid provide convincing evidence that the southern side of the high Himalayan valleys represents a “direct channel” through which brown cloud pollutants can be transported up to $5\,\textrm{k m}$ above sea level (asl) especially during the premonsoon period; this is a case where normally clean atmospheric conditions could be strongly compromised [Bonasoni et al., 2010]. However, the effects of the Tibetan aerosol on regional climate variability remain largely unknown, and this is due, at least in part, to the limited number of observations over the plateau.
[6] Although the monitoring of atmospheric composition at high altitudes is recognized as important for understanding climate change, experimental work in high-altitude regions has been sparse [Carrico et al., 2003; Hindman and Upadhyay, 2002]. To this end, we conducted a 1 year study of bulk aerosol (total suspended particulate, TSP) at Lulang, which is on the southeastern flank of the TP. The study focused on selected aerosol chemical components, including organic (OC), elemental carbon (EC), water-soluble inorganic ions, and dust-derived trace elements, and it included an assessment of their seasonal variability, especially in relation to the monsoonal circulation. We show how anthropogenic activities dramatically perturb the background aerosol levels and discuss how the aerosol chemical composition changes in such cases (episodes). Potential sources for the aerosol particles and their transport pathways during high-aerosol loading episodes are also evaluated and discussed.
2. Data and Methodology
2.1. Aerosol Sampling
[7] The Lulang sampling site (Figure 1) is situated in Linzhi Prefecture, which is on the southeastern margin of the TP $(94.44^{\circ}\mathrm{E}$ , $29.46^{\circ}\mathrm{N})$ . To the west of the sampling site is the Yarlung Tsangpo River Valley, which has an altitude of $3300\,\mathrm{m}$ , and the Himalaya Mountains rise south of the site. Lulang is approximately 30 to $50~\mathrm{km}$ west of several small villages [Cao et al., 2010]. There are extensive forests in Linzhi, and the sampling site is located in a remote area with no major anthropogenic sources nearby. Hence, long-range transport from upwind regions to the plateau is the presumptive cause for elevated pollutant concentrations. During the summer monsoon, low-pressure systems over the plateau induce a flow of moist warm air from the Indian and Pacific Oceans to the Indian subcontinent and TP. In winter, on the other hand, high-pressure systems drive cold, dry air out of the plateau [Byson, 1986; Tang, 1998].
[8] Sampling was normally conducted once every 6 days starting at 10:30 Beijing Time (BJT), and the sampling intervals were typically 3 days. Aerosol (TSP) samples were collected at a flow rate of $40\,\mathrm{\L/min}$ with a KC-120H QingDao Laoshan sampler (Laoshan Electronic Instrument Factory Co., LTD., QingDao, China). The flow rate of the sampler was lowered from $100\;\mathrm{L/min}$ to $40\;\mathrm{L/min}$ to adjust for the atmospheric pressure in this high-altitude site. The aerosol particles were collected on $90\,\mathrm{mm}$ Whatman quartz fiber filters (QM-A™, Whatman, Clifton, NJ, USA) from 16 July 2008 to 26 July 2009. The filters were preconditioned by heating at $900^{\circ}\mathrm{C}$ for $3\,\mathrm{~h~}$ to remove residual carbon. After sampling, the filters were transported in a portable cooler to the aerosol laboratory at the Institute of Earth Environment, Chinese Academy of Sciences, to avoid the loss of volatile compounds. Field blank filters were also collected periodically by exposing filters in the sampler without drawing air through them; these were used to account for any artifacts introduced during the sample handling process.
[9] An automatic weather station was installed at the sampling site, and meteorological data, including wind speed and direction, air temperature, precipitation, air pressure, and relative humidity, were collected routinely. Abrupt variations in relative humidity (RH), wind direction (WD), and precipitation were used to identify the onset and conclusion of the summer monsoon and winter seasons. Based on the seasonality of atmospheric circulation, the annual cycle was defined as follows: monsoon (16 July 2008 to 3 October 2008; 28 April 2009 to 26 July 2009), postmonsoon (4 October 2008 to 9 November 2008), winter season (10 November 2008 to 17 February 2009), and premonsoon (18 February 2009 to 30 April 2009). Table 1 summarizes the meteorological conditions during the study. A total of 61 samples were collected; 27 of these were collected during the monsoon, 6 during the postmonsoon, 16 during the winter season, and 12 during the premonsoon.
2.2. Experimental
2.2.1. Gravimetric Analysis
[10] After ${\sim}24\,\mathrm{h}$ equilibration at a temperature between 20 and $23^{\circ}\mathbf{C}$ and RH between 35 and $45\%$ , the quartz fiber filters were analyzed gravimetrically for mass concentrations with the use of a Sartorius MC5 electronic microbalance that had a sensitivity of $\pm1\,\upmu\mathrm{g}$ (Sartorius, Göttingen, Germany). Each filter was weighed at least three times before and after sampling, and the net mass accumulation was obtained by subtracting the difference between the averaged presampling and postsampling weights.
2.2.2. Water-Soluble Inorganic Ions Analysis
[11] An aliquot of a sample filter $({\sim}4.33\ \ \mathrm{cm}^{2})$ was extracted with $10~\mathrm{mL}$ of ultrapure water for the inorganic ion studies. Eight inorganic ion concentrations $\bar{(\mathrm{Na}^{+}}$ , $\mathrm{NH_{4}}^{+}$ , $\mathbf{K}^{+}$ , $\mathrm{Mg}^{2+}$ , $\mathrm{{{Ca}}}^{2+}$ , $\bar{\mathrm{Cl}^{-}}$ , $\mathrm{NO}_{3}{}^{-}$ , and $\mathrm{SO}_{4}{}^{2-}$ ) were determined with the use of a DX600 ion chromatography system (Dionex Inc., Sunnyvale, CA, USA). A CS12 column $\left\langle150\times4\,\mathrm{mm}\right\rangle$ and an AS14 column $150\times4\,\mathrm{mm})$ ) were used for cation and anion analysis, respectively. Standard reference materials produced by the National Research Center for Certified Reference Materials (Beijing, China) were analyzed for quality control and assurance purposes [Zhang et al., 2011]. Field blank levels were averaged and subtracted from the samples. Ten percent of the samples were submitted for replicate analyses, yielding coefficients of variance of $\pm0.25\%$ for $\mathrm{Na}^{+}$ , $\pm22.8\%$ for $\bar{\mathrm{NH}_{4}}^{+}$ , $\pm1.82\%$ for $\mathsf{K}^{+}$ , $\pm2.90\%$ for $\mathrm{Mg}^{2+}$ , $\pm3.75\%$ for $C\mathbf{a}^{2+}$ , $\pm1.59\%$ for $\mathrm{Cl}^{-}$ , $\pm2.34\%$ for $\mathrm{NO}_{3}^{\mathrm{~-~}}$ , and $\pm1.39\%$ for $\mathrm{SO}_{4}{}^{2-}$ .
2.2.3. OC and EC Analysis
[12] A $0.5\,\mathrm{cm}^{2}$ punch from each quartz filter was analyzed for OC and EC by the IMPROVE_A thermal/optical reflectance protocol [Chow et al., 2007] and the use of a Desert Research Institute (Reno, NV, USA) Model 2001 thermal/ optical carbon analyzer (Atmoslytic Inc., Calabasas, CA, USA). Details of the method, including quality assurance/ quality control (QA/QC) procedures, are described in Cao et al. [2003].
2.2.4. Elemental Analysis
[13] All samples were analyzed by energy dispersive X-ray fluorescence spectrometry with the use of an Epsilon 5 XRF analyzer (PANalytical, Almelo, Netherlands). The X-ray source was a side window $\mathrm{\DeltaX}$ -ray tube with a gadolinium anode; the instrument operated at an acceleration voltage between 25 and $100\;\mathrm{\kV}$ and a current of 0.5 to $24\,\mathrm{mA}$ (maximum power: $600~\mathrm{W}$ ). The characteristic X-ray radiation was detected with the use of a PAN 32 germanium detector.
[14] Concerns over the validity of the elemental data for the quartz fiber filters led to an intercomparison of elemental concentrations for samples collected on two different types of filters. The intercomparison study was based on XRF measurements made on nine collocated Teflon-membrane and quartz fiber filters from Xi’an. The concentrations were found to be comparable, with correlations $(r)$ ranging from 0.982 for Fe and $Z\mathfrak{n}$ (slopes of 1.054 and 1.062, respectively) to 0.915 for As (slope of 1.204) for these elements. Measurement precision was calculated as the standard deviation of several analyses of the same sample; the precisions were $\pm7.6\%$ for Fe, $\pm8.6\%$ for Ti, $\pm12.5\%$ for Mn, $\pm7.6\%$ for Zn, $\pm23.5\%$ for As, $\pm33.3\%$ for Br, and $\pm7.9\%$ for $\mathrm{Pb}$ at typical concentration levels [Cao et al., 2012]. The $\mathrm{QA/QC}$ of the analysis is described by Xu [Xu et al., 2012].
[15] In this study, Fe and K were used to estimate the loadings of crustal matter and to evaluate the influence of biomass burning, respectively. Prior studies have shown that Fe accounts for $4\%$ of Asian dust and Chinese loess [Zhang et al., 2003]; hence, soil dust concentrations were estimated from the following equation:
$$
\mathrm{C_{\mathrm{{soil}\ d u s t}}=C_{F e}/4\%}
$$
where $\mathrm{{C_{soil\dust}}}$ is soil dust concentration and $\mathrm{C_{Fe}}$ is the elemental Fe concentration. And $\mathrm{K}^{+}/\mathrm{K}$ ratio was used to evaluate the extent of biomass burning or dust in different episodes [Shen et al., 2007]. Moreover, S, K, Fe, Ca, and Ti were used for source apportionments by positive matrix factorization (PMF).
2.3. Satellite Data
[16] Space-based sensors provide near-real-time information on atmospheric aerosols, and the data obtained with them are invaluable because the coverage is global and spatial resolution kilometer scale [King et al., 1999; Kaufman et al., 2002]. Satellite aerosol products considered in this study consist of both level-2.0 aerosol optical depth (AOD) from the Moderate Resolution Imaging Spectrometer (MODIS) which is deployed on the Terra satellite [Remer et al., 2005] and the level-2.0G UV aerosol index (AI) derived from the Ozone Monitoring Instrument (OMI) carried by the Aura spacecraft [Torres et al., 2007]. MODIS AOD has been shown to be suitable for the detection of aerosol pollution [Chu et al., 2003; Li et al., 2007], while the OMI AI has been used for long-term remote sensing of UV absorbing aerosols, such as desert dust [Moulin and Chiapello, 2004; Huang et al., 2007]. Accordingly, we used both MODIS AOD and OMI AI for the case studies described in section 3.4.
2.4. Air Mass Back Trajectory Analysis
[17] Air mass back trajectory analysis is another useful tool for identifying the probable sources and transport pathways for air pollutants. In this study, $72\,\mathrm{h}$ air mass back trajectories starting $3000\;\mathrm{m}$ above ground level at 0:00 UTC were calculated using the NOAA HYSPLIT4 trajectory model. The trajectories were used to study the origins of the aerosol pollution, and to this end, the high-aerosol loading episodes were classified according to the direction of air mass transport.
2.5. Source Apportionment Method
[18] Receptor modeling by positive matrix factorization (PMF 3.0, developed by the U.S. Environmental Protection Agency) was used to apportion the measured ambient aerosol concentrations among potential sources. PMF can use the information contained in aerosol concentration and composition data to identify presumptive sources. This approach has been widely used for receptor modeling, and it is typically applied when the source profiles are unknown [Chen et al., 2007, 2010]. Here PMF was applied as a way of evaluating the origins of the aerosols over the southeastern TP.
3. Results and Discussion
3.1. Variations of Aerosol Mass Concentrations
[19] The arithmetic mean annual aerosol mass concentration was $23.5\!\pm\!20.3\ensuremath{\,\upmu\mathrm{g}\,\mathrm{m}^{-3}}$ , and it ranged from 8.2 to $139.2\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ . Figure 2 shows the seasonal variations of TSP loadings and major chemical species levels at the Lulang site. The TSP mass exhibits a well-defined seasonal cycle, with the highest mass concentration $(43.7\pm37.2\,\upmu\mathrm{g\dot{\m}}^{-3})$ ) observed during the premonsoon and the lowest value during the monsoon $(\bar{1}4.8\pm6.6\,\upmu\mathrm{g}\;\;\mathrm{m}^{-3})$ . The aerosol levels in the premonsoon period were larger by about a factor of 3 compared with mean level during the monsoon; however, the standard deviation of aerosol mass during the premonsoon was much larger than during the monsoon period. This seasonal variation is similar to the aerosol mass and black carbon concentration variations found at the Nepal Climate Observatory at Pyramid (NCO-P) site in the southern Himalayas [Marinoni et al., 2010].
[20] High concentrations of aerosol mass and large variability at high elevations, particularly prior to the onset of the monsoon, have been attributed to long-range pollution transport [Shrestha et al., 2000]. As shown in Figure 2, major peaks in the TSP mass concentrations always coincided with high loadings of OC, EC, major ions, and Fe. This implies that the same physical processes, including those during both production and transport, caused these species’ concentrations to covary. The aerosol mass seasonality we observed at Lulang is similar to the variations observed at other sites in Nepal and India; that is, lower loadings during the monsoon season and higher concentrations in the winter and premonsoon periods [Ganguly et al., 2006; Marinoni et al., 2010; Mouli et al., 2006]. Wet scavenging is one likely cause for the lower aerosol concentrations during the monsoon while in winter, higher aerosol concentrations correspond with diminished precipitation in air brought by strong westerly flow.
[21] Eight high-aerosol loading events, operationally defined by aerosol mass loadings $>35\,\upmu\mathrm{g}\,\:\stackrel{.}{\mathrm{~m}}^{-3}$ , were observed during the year of measurements. Two types of high-aerosol episodes were identified on the basis of differences in transport pathways as depicted by back trajectory analysis (Figure 3). Type I episodes (marked in yellow in Figures 2 and 3) were classified as a dust events owing to the abundance of Fe. Three type I events occurred from the premonsoon to monsoon season; these were on 11 to 14 March, 10 to 13 April, and 18 to 21 May. The corresponding aerosol mass loadings for these three events were $\dot{1}39.2\,\upmu\mathrm{{g}\ m}^{-3}$ , $40.0\,\upmu\mathrm{g}\mathrm{\;m}^{-3}$ , and $37.3\,\upmu\mathrm{g}\mathrm{\;m}^{-3}$ , giving an arithmetic mean of $72.\bar{2}\,\upmu\mathrm{g}\,\,\mathrm{m}^{-3}$ , which is over 3 times the yearly mean value.
[22] Type II episodes (marked in gray as shown in Figures 2 and 3) were defined as pollution events because they were characterized by high concentrations of OC, EC, and three major ions $(\mathrm{K}^{+},\,\mathrm{SO}_{4}^{\phantom{\,+}2-}$ and $\mathrm{NO}_{3}^{\mathrm{~-}}$ ). Five type II events occurred over a total of 15 days; one event in winter (9 to 12 February) and four events during the premonsoon period (15 to 18 February, 5 to 8 March, 17 to 20 March, and 23 to 27 April). The mean aerosol mass for type II episodes was $53.8\,\upmu\mathrm{g}\:\:\mathrm{m}^{-3}$ , which is over 2 times the annual value. Chemical profiles for type II episodes were characterized by high $\bar{\bf K}^{+}/\bar{\bf K}$ ratios, indicating important contributions from biomass burning upwind. Xia et al. [2011] monitored a pollution episode at the Namco site in the central TP from 14 to 19 March 2009, and their results, consistent with ours, indicated that during that time, air pollution was transported to the TP from upwind sources.
[23] The high concentrations of carbonaceous species and major ions during type II episodes can be linked to air flow driven by the SW monsoon; the transport pathways often pass over densely populated areas of the Indo-Gangetic Plain. This region is characterized by heavy anthropogenic emissions, and during the premonsoon season, it is usually dry there [Marinoni et al., 2010]. In contrast, for type I episodes, the back trajectories show a simple flow pattern in which air masses, mainly from western China, bring crustal matter from desert dust source regions to Lulang.
3.2. Chemical Composition
3.2.1. Carbonaceous Aerosol
[24] The abundances of TSP, OC, EC, water-soluble ionic species (WSIS), and OC/EC ratios in the aerosol samples are presented in Table 2 where they are grouped according the type of high-aerosol episode. The overall average concentrations of OC and EC were $4.28\pm2.05\,\upmu\overline{{\mathrm{g}}}\,\,\,\mathrm{~m}^{-3}$ and $0.52\pm0.35\,\upmu\mathrm{g}\,\,\mathrm{m}^{-3}$ , respectively. Both OC and EC showed distinct seasonal variations with remarkably higher concentrations in late winter into the early premonsoon compared with the other seasons (Figure 2). The EC seasonal trends are in good agreement with the record of EC in ice cores taken from Tibetan glaciers where peaks occur during November to March [Xu et al., 2009]. For OC, the concentrations decreased in the following order: premonsoon $(5.15\,\upmu\mathrm{g}$ $\mathrm{m}^{-3})\,>$ monsoon $(4.27\,\upmu\mathrm{g}\;\mathrm{m}^{-\bar{3}})$ , and postmonsoon $(4.30\,\upmu\mathrm{g}$ $\mathrm{m}^{-3})\,>$ winter $(3.65\,\upmu\mathrm{g}\,\\,\mathrm{m}^{-3})$ , while EC followed a decreasing order of premonsoon $(0.84\,\upmu\mathrm{g}\ \ m^{-3})>$ postmonsoon $(0.67\,\upmu\mathrm{g}\,\,\mathrm{m}^{-3})\,\Bar{>}$ winter $(0.58\,\upmu\mathrm{g}\,\,\mathrm{m}^{-3})>$ monsoon $(0.31\,\upmu\mathrm{g}$ $\mathrm{m}^{-3})$ . The most abundant precipitation occurred during the monsoon (Table 1), and the removal of aerosol particles from suspension by precipitation scavenging was the most likely reason for the lower EC concentrations at that time of year.
[25] The ratios of OC to EC ranged from 1.7 to 8.6, 7.8 to 58.4, 1.9 to 8.4, and 1.7 to 9.6 in premonsoon, monsoon, postmonsoon and winter, respectively; and the corresponding averages were 6.3, 17.7, 6.7, and 6.5. These seasonally averaged OC/EC ratios are comparable with those reported for other high-altitude sites such as Mt. Abu (range: 4.8 to 27.2) and Manora Peak (range: 3.0 to 11.5) [Ram et al., 2008]. Comparisons with other sites on the TP show that the OC concentrations at Lulang (Table 3) were larger than those at the Namco site in the central TP [Cong et al., 2007; Ming et al., 2010]. EC was higher than the value measured at Waliguan in northeast TP [Wen et al., 2001] or Muztagh Ata Mountain, in the western TP [Cao et al., 2009]. These comparisons suggest that Lulang site is more strongly influenced by pollution sources upwind than some other parts of the TP.
[26] The data for the high-aerosol episodes were excluded to characterize the chemical composition of the aerosol on more normal days. OC was found to be more abundant during the monsoon and postmonsoon than in the other two seasons. These trends can be explained by the emission of plant spores and pollen as well as the formation of greater quantities of secondary organic carbon (SOC) in the periods with the higher OC loadings. The average OC $(3.88\pm1.21\,\upmu\mathrm{g}$ $\mathrm{m}^{-3}.$ ) and EC $(0.44\!\pm\!0.22\,\mathrm{\bar{\mu}g\,m}^{-3})$ concentrations on normal days were ${\sim}9$ to $14\%$ lower than the annual mean values, and neither OC nor EC showed significant differences on normal days in the four periods (the average OC in four seasons were 2.89, 4.21, 4.30, and $3.5\dot{2}\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ , while EC were 0.50, 0.31,
0.67 and $0.57\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ , respectively). During type II episodes, the average OC and EC concentrations reached 7.93 (4.86 to 16.58) and 1.30 (0.74 to 2.23) $\upmu\mathrm{g}\:\mathfrak{m}^{-3}$ , these were much higher than those during type I episodes or on normal days. The higher OC and EC levels in the type II samples imply that compared with the others, combustion sources were stronger contributors during the type II events.
[27] The OC and EC concentrations were generally well correlated, with a least squares linear regression for the full data set yielding the following equation: ${\mathrm{OC}}\,{=}\,4.1\times\mathrm{EC}\,{+}\,2.1$ (with a correlation coefficient $r\!=\!0.70$ and probability for chance occurrence $p\,{<}\,0.0001$ , $n\!=\!61$ ). Relationships between the concentrations of OC and EC in the monsoon and other periods are shown in Figure 4 to illustrate the seasonal differences in aerosol characteristics. However, the correlations for the four separate seasons are relatively lower ${\bf\nabla}r\!=\!0.3$ to 0.6) than the entire data set; this suggests the existence of multiple sources of OC and EC that were not always linked. 3.2.2. Water-Soluble Ionic Species (WSIS)
[28] The sums of the concentrations of the eight ions analyzed were 3.81, 1.92, 1.60, and $1.67\,\upmu\mathrm{g}\,\textrm{m}^{-\widecheck{3}}$ in the premonsoon, monsoon, postmonsoon, and winter samples, respectively, and these sums amounted to $16.24\%$ , $8.17\%$ , $6.81\%$ , and $7.10\%$ of the total TSP mass. The dominant compounds were $\mathrm{SO}_{4}{}^{2-}$ , $C\mathbf{a}^{2+}$ , and $\mathrm{NO}{_3}^{-}$ , whose combined concentrations accounted for more than $75\%$ of the total water-soluble ion mass. Ion mass balance calculations were used to evaluate the acid-base balance of aerosol and to determine whether significant quantities of undetected ions were present in the samples [Jain et al., 2000]. A strong correlation $(r\!=\!0.87)$ ) between cation and anion equivalents for all samples implies that the ions measured in our study were in fact the most abundant ions on the filters. The ratio of anion to cation equivalents was ${\sim}0.86$ , which was higher than the ratios determined at Zhuzhang and Lhasa, Tibet [Qu et al., 2008; Tang et al., 2005], and this suggests that the Lulang TSP is more alkaline than the aerosol from the other two sites.
[29] Interestingly, most of the $\mathrm{NH_{4}}^{+}$ values were below the detection limit (data not shown in Table 1)—the exceptions were samples collected during three heavy pollution events (15 to 18 February, 5 to 8, and 17 to 20 March). Indeed, the aerosol during type I episodes was depleted in $\mathrm{NH_{4}}^{+}$ compared with the samples collected during type II episodes. Prior studies have similarly shown depletions in $\bar{\mathrm{NH}_{4}}^{+}$ during dust storm events compared with pollution events [Shen et al., 2009]. The pattern in $\mathrm{NH_{4}}^{+}$ concentrations at Lulang is similar to what was observed in previous studies at the NCO-P site on the southern Himalayas. There the ammonium levels also were below the detection limit except during the premonsoon [Decesari et al., 2010]. The low concentrations of ammonium at these high-latitude sites highlight the somewhat usual ion profiles of particular matter (PM) from the TP; one implication of the low $\mathrm{NH_{4}}^{+}$ loadings is that other cations, such as $C\mathbf{a}^{2+}$ and $\mathrm{Mg}^{2+}$ , should play relatively more important roles in ion balance.
[30] In the data sets with the high-aerosol events excluded, none of the ions showed significant variations except for $\mathrm{Na^{+}}$ and $\mathrm{Cl}^{-}$ ; this implies that there were no substantial changes in the pollution sources under clean background conditions. The rank order for the most abundant to least abundant ionic species followed the sequence $\mathrm{SO_{4}}^{2-}\mathrm{>}\mathrm{Na}^{+}\mathrm{>}\mathrm{~Ca}^{2+}\mathrm{>NO_{3}}^{-}\mathrm{>Cl}^{-}$ ; $>\mathrm{Mg}^{2+}\mathrm{>K^{+}\mathrm{>NH}_{4}}^{+}$ . In the monsoon season, when winds are predominantly from the Arabian Sea, the concentrations of $\bar{\mathbf{N}}\mathbf{a}^{+}$ and $\mathrm{Cl^{-}}$ were relatively high (average: 0.42 and $0.16\,\upmu\mathrm{g}$ $\mathfrak{m}^{-3}$ , respectively), in fact, twice the loadings in the other periods. This pattern is similar to the seasonality in sea salt observed at Mt Abu, India [Rastogi and Sarin, 2005].
[31] Previous studies have shown that biomass burning is a major source of aerosol particles in Southern Asia [Chan et al., 2003; Duncan et al., 2003]. The concentrations of $\mathsf{K}^{+}$ , a tracer of biomass burning [Andreae, 1983; Shen et al., 2009], in aerosol samples during the type II episodes $(0.38\,\upmu\mathrm{g}\,\,\mathrm{m}^{-\bar{3}})$ ) were much higher than those in the premonsoon $(0.04\,\upmu\mathrm{g}$ $\mathfrak{m}^{-3},$ ), monsoon $(0.04\,\upmu\mathrm{g}\,\textrm{m}^{-3})$ , postmonsoon $\left(0.03\,\upmu\mathrm{g}\,\textrm{m}^{-3}\right)$ ), winter seasons $(0.03\,\upmu\mathrm{g}\:\textrm{m}^{-3})$ , or type I episodes $(0.04\,\upmu\mathrm{g}$ $\mathfrak{m}^{-3},$ ). This can be explained by strong emissions from biomass burning sources upwind of the measurement site during the type II episodes. The concentrations of $\mathrm{SO}_{4}{}^{2-}$ and $\mathrm{NO}_{3}^{\mathrm{~-~}}$ during these episodes also were enhanced by factors of 4 to 5 over those on normal days. The elevated loadings of $\mathsf{K}^{+}$ , along with $\mathrm{SO}_{4}{}^{2-}$ , $\mathrm{NO}_{3}^{\mathrm{~-~}}$ , $\mathrm{OC}$ , and EC in the type II samples, suggest that emissions from a combination of biomass burning and fossil fuel combustion sources were brought to Lulang during those episodes, mostly likely as a result of long-range transport.
3.3. Material Balance
[32] The relative contributions of major chemical species to the aerosol mass for different seasons (excluding the pollution events) as well as during two type episodes are shown in Figure 5. Organic matter (OM) concentrations were calculated as 1.8 times the OC content as suggested by Turpin and Lim [2001]. Sea salt was estimated from $\mathrm{Cl}^{\dot{-}}\,{+}\,(1.4486\times\mathrm{Na}^{+})$ where 1.4486 is the ratio of the sum concentration of all elements except $\mathrm{Cl}^{-}$ in sea water to the concentration of $\mathrm{Na}^{+}$ [Li et al., 2010]. Crustal matter (geological material) was calculated by $\mathrm{Fe/4\%}$ [Zhang et al., 2003].
[33] The sums of the measured and estimated components accounted for 68, 91, 79, and $79\%$ of the aerosol mass in the premonsoon, monsoon, postmonsoon, and winter samples, respectively. During the premonsoon, crustal matter was the most abundant species, followed by OM, sulfate, EC, sea salt, and nitrate. In contrast, during the monsoon and postmonsoon, OM dominated, accounting for $57\%$ and $42\%$ of the mass, respectively, and crustal matter decreased when compared with the premonsoon period. Sea salt accounted for only $1.7\%$ of the aerosol mass in the premonsoon and winter season, but the percentage did increase by a factor of $\sim\!\!2.2$ during the monsoon. For the winter samples, the contributions in order of importance were geological material $>\mathrm{OM}>$ sulfate $>$ elemental carbon $>\mathbf{S}\mathbf{e}\mathbf{a}$ salt $>$ nitrate. Approximately 9 to $32\%$ of the measured mass was not quantified by the chemical analysis, and this can be ascribed to residual water and other unmeasured species, underestimations of the weighting factors for OM, sea salt, and geological material. Possible errors in the gravimetric analyses, which would mainly be due to uncertainties in the equilibration of the filters [Kajino et al., 2006], also must be acknowledged.
[34] During type I events, crustal matter was the most abundant aerosol constituent, contributing $53\%$ to the mass, and it was followed by OM $(19\%)$ . These results can be compared with data for the dust storm observed at Zhenbeitai; in that study, mineral dust contributed $51\%$ to the $\operatorname{PM}_{2.5}$ mass, followed by carbonaceous matter $(12\%)$ $[X u$ et al., 2004]. For type II episodes, the percentage of the aerosol mass contributed by crustal matter decreased to $40\%$ , while the percentages of sulfate, nitrate, and EC increased; these are relatively similar to their contributions in the premonsoon and winter. The high crustal material loadings and high sulfate, nitrate, and EC, in terms of both concentration and percentage, show that both geological and anthropogenic sources provided significant quantities of the aerosol that was transported to the southeastern TP in type II episodes.
3.4. Species Enhancements During HighAerosol Episodes
[35] Enhancement factors (EnF, plural EnFs) are defined here as the ratio of the mass fraction for a species during the eight high-aerosol episodes normalized to the annual average mass fraction (see text for description of enhancement factors in the caption of Figure 6) [Watson et al., 2002]. Here we calculate the EnFs to further investigate the chemical composition of PM during the high-aerosol loading events (Figure 6).
[36] For all events, the EnFs of OC and $\mathrm{Na^{+}}$ were less than unity, showing that these species did not contribute significantly to the high-aerosol loads. In addition, as Figure 6 shows, the EnFs of most combustion- and industry-related species $\propto\!^{+}$ , $\mathrm{SO}_{4}{}^{2-}$ , EC) showed mass fractions less than the annual average $\mathrm{EnFs}\,<\,1\mathrm{)}$ for the 21 May, 13 April, and 14 March 2009 episodes, and the moderate enhancement of crustal K for these days is consistent with regional transport from the northwest. For the type $\mathrm{II}$ episodes (typical haze events), the EnFs for $\mathsf{K}^{+}$ , $\mathrm{\dot{SO}}_{4}{}^{2-}$ were relatively high, while the EnFs for $\mathrm{NO}_{3}^{-}$ , S, and K were lower. Compared with type I episodes, the EnFs for the type II events seem to be more variable, and it is worth noting that the EnFs for each of these pollution events were different from one other.
[37] During one severe episode (15 to 18 February), several species associated with combustion products $\left(\mathrm{NO}_{3}\right)^{-}$ , $\mathrm{SO}_{4}{}^{2-}$ , $\mathrm{Pb}$ , and $\mathbf{K}^{+}$ ) showed significant enhancements (EnFs: 2 to 10), and this is an indication that the high-aerosol load during this episode was due, at least in part, to emissions produced from fossil fuel combustion. In contrast, the EnFs for crustal matter (Fe) and $\mathrm{{Ca}}^{2+}$ were less than unity during the event, indicating a relatively small contribution from dust. This haze episode evidently was influenced by a combination of strong anthropogenic emissions on Indian subcontinent coupled with dry meteorological conditions during regional transport; this conclusion also is supported by back trajectory analysis (as shown in Figure 3).
[38] During another episode (17 to 20 March), $\mathsf{K}^{+}$ and K were the most strongly enhanced components of all measured species (EnFs of 4.0 and 1.5, respectively), and the EnFs for $\mathrm{NO}_{3}^{\mathrm{~-~}}$ , EC, $\mathrm{SO}_{4}{}^{2-}$ , and S also were relatively high. This can be explained by the transport of polluted boundary layer air enriched with products from biomass burning and fossil fuel combustion. The EnF analysis indicated that several kinds of pollutants were enhanced several fold as a result of transport from upwind regions, and this is important because anthropogenic substances can exert several types of influences on the environment and climate of the TP; these effects include alterations of the radiative balance as well as melting the glaciers on the TP.
3.5. Further Evidence for Long-Range Aerosol Transport
[39] The seasonally averaged levels of TSP, EC, OC, and ions are compared with other measurements from the TP, Nepal, and the Indian Peninsula in Table 3. The concentrations of OC, EC, and ions are comparable with those at two high-altitude sites—the Atmospheric Brown Cloud-Pyramid Observatory [Bonasoni et al., 2008] and NCO-P [Decesari et al., 2010]—but they are much lower than those on the Indian subcontinent [Rastogi and Sarin, 2005, 2009; Rengarajan et al., 2007; Venkataraman et al., 2002], especially urban sites (Ahmedabad and Mumbai). Such comparisons along with the air flow analyses (see below) suggest that the high-aerosol emissions over the Indian Peninsula were major influences on the TP air.
[40] To further investigate the transport of anthropogenic aerosol to the southeastern TP, we analyzed National
Centers for Environmental Protection/National Center for Atmospheric Research (NCEP/NCAR) winds at $500\ \mathrm{hPa}$ , the corresponding MODIS AOD at $550\,\mathrm{nm}$ , and the ultraviolet aerosol index values (UV-AI) from the Ozone Monitoring Instrument for two typical episodes (Figures 7 and 8). These two figures are comprehensive maps of the spatial distributions of AOD/AI over the study region. Daily NCEP/NCAR reanalysis wind trajectories were used to determine the daily trends in the wind directions during the episodes.
[41] As shown in Figure 7, the 17 to 20 March high-aerosol episode had the highest concentrations of $\mathrm{SO}_{4}{}^{2-}$ , $\bar{\bf N}{\bf O}_{3}{}^{-}$ , $\mathsf{K}^{+}$ , OC, and EC (Figure 2) of all the events. For this case, the wind direction in the vicinity of the measurement site was from the west or southwest, with a mean speed of $10\,\textrm{m}$ $\mathrm{s}^{-1}$ . The air mass passed over the Thar Desert, located in northwestern India, and this was likely responsible for the high-dust load during this period. Results from MODIS showed that Lulang was shrouded in high-aerosol loads at this time, with the AOD value exceeding 0.9 (Figure 7d). The remarkably high AOD area stretched along the southern slope of the Himalayas (including southeastern India and areas east of Nepal and northeast of Myanmar) during this episode, and the information on wind fields and AOD in Figure 7 clearly indicates a likely pathway for pollutants as they were transported to Lulang. Another recent study similarly showed that high BC concentrations at Lulang were associated with the transport from the southwest [Cao et al., 2010].
[42] Figure 8 shows that the high mass and high Fe (dust) concentrations observed from 18 to 21 May were associated with northwesterly winds with speeds of ${<}10\,\mathrm{m~s}^{-1}$ . The OMI AI in the northern TP (regions adjacent to the Taklimakan Desert and Qaidam Desert) and in northwestern India (parts of the Thar-Cholistan Desert) were high, that is, ${>}2$ (Figure 8d). The NCEP/NCAR wind trajectories showed that winds from the northwest prevailed during the period, and the patterns in OMI AI suggest that the winds carried large quantities of geological material to the site. In contrast, the concentrations of combustion products $(\mathrm{SO}_{4}{}^{2-}$ , $\mathrm{NO}_{3}{}^{-}$ , $\mathsf{K}^{+}$ , and EC) during this period were relatively low. Other studies conducted at this sampling site have shown that transport from northwest was associated with low BC levels but high-dust loadings [Cao et al., 2010; Xia et al., 2008].
3.6. Source Apportionment With the Use of a PMF Model
[43] All of the aerosol data, that is, the concentrations of OC, EC, S, K, Ca, Ti, Fe, $\mathrm{Na}^{+}$ , $\mathbf{K}^{+}$ , $\mathrm{Mg}^{2+}$ , $\mathrm{{Ca}}^{2+}$ , $\mathrm{Cl}^{-}$ , $\mathrm{NO}_{3}{}^{-}$ , and $\mathrm{SO}_{4}^{\ 2-}$ , were included in PMF analyses, which followed the procedures described in Reff et al. (2007) and Green et al. (2012). The frequency distribution of scaled residuals was taken into account in the model; most of these were between $^{-2}$ and $+2$ , and this attests to a good agreement between the PMF model results and the input data. The profiles for each factor and the factor loadings are shown in Figure 9.
[44] Factor 1 was dominated by OC, EC, K, $\ K^{+}$ , Ti, and Fe (Figure 9a), and it is best interpreted as a mixture of dust and combustion aerosol. It is noteworthy that the Factor 1 contribution to TSP mass was high in premonsoon season (Figure 9b), especially from 15 to 18 February, 5 to 8 March, 17 to 20 March, and 23 to 27 April. All of these intervals were characterized as type II high-aerosol events. This result is also supported by air mass back trajectories and MODIS AODs, which provide evidence for the long-range transport of mixed aerosols from the IndoGangetic Plain along the valley of the Yarlung Tsangpo River to the southeastern TP. Nearly one third of the TSP mass was accounted for by this factor (Figure 9c). [45] The second PMF factor had high loadings for $\mathrm{Na^{+}}$ , $\mathring{\mathrm{K}}^{+}$ , $\bar{\mathrm{Mg}^{2\bar{+}}}$ , $C\mathbf{a}^{2+}$ , $\mathrm{Cl}^{-}$ , and OC, and this factor is most readily explained as a combination of SOC and sea salt. The Factor
2 contribution was higher in summer, and this factor accounted for $22\%$ of annual TSP mass.
[46] The $\mathrm{NO}_{3}^{\mathrm{~-~}}$ , $\mathrm{SO}_{4}{}^{2-}$ , and sulfur concentrations in the profiles of the third factor were very high, and this can be attributed to the formation of secondary aerosols. The highest contribution from Factor 3 occurred from 9 to 12 February and 17 to 20 March (marked by gray arrows in Figure 9b). The strong loadings of Factor 3, combined with air mass trajectories for these dates, indicated that aged air that had passed over the Indo-Gangetic Plain carried high loadings of secondary aerosols to Lulang. We note that the sample collected from 11 to 14 March (yellow arrow) had a large contribution from both Factor 3 (secondary aerosol) and Factor 4 (dust storms). This can be explained by the back trajectory analysis (Figure 3) which also is substantiated by the wind directions determined from the NCEP/NCAR reanalysis (Figure 10). That is, from 13 to 14 March, the wind direction at Lulang changed from northwest to southwest, and therefore, one can see how this flow could bring a mixture of crustal matter from the northwest and secondary pollutants from the southwest to our sampling site. The back trajectory for this period (Figure 3) shows how the air flow changed course and passed over the northern edge of India before arriving at the southeastern TP. This factor accounted for $17\%$ of the annual average TSP mass, and high contributions occurred in the winter and premonsoon season.
[47] Factor 4 was enriched in crustal elements (e.g., Ca, Fe, Ti, K, ${\mathrm{Mg}}^{2+}$ , $\mathrm{{Ca}}^{2+}$ , and $\mathrm{Cl^{-}}$ ), and this factor clearly represents mineral aerosol and fugitive dust. In the same way as Factor 1, major contributions occurred during 11 to 14 March, 10 to 13 April, and 18 to 21 May (marked by yellow arrows in Figure 9b); these were all characterized as type I episodes. This factor accounted for $28\%$ of TSP mass.
4. Summary and Conclusions
[48] One year of continuous observations of aerosol composition (from July 2008 to July 2009) were conducted at Lulang ( $3300\,\mathrm{m}$ above sea level) on the southeastern TP. The data set provides insights into the chemical composition and the sources of aerosols over a remote, high-altitude site on the TP. Major results were summarized as below.
[49] 1. During the premonsoon, the highest seasonal values of TSP, OC, EC, $\mathbf{K}^{+}$ , $\mathrm{NO}_{3}^{\mathrm{~-~}}$ , and $\mathrm{SO}_{4}{}^{2-}$ were observed, while the lowest concentrations of EC and TSP occurred during the monsoon, and OC was lowest in winter. The clear seasonal variations of these compounds at Lulang can be explained by seasonal differences in atmospheric conditions.
[50] 2. Source apportionment by PMF produced four factors of the TSP mass; there were (1) mixed dust and combustion aerosols, (2) SOC and sea salt, (3) secondary aerosols, and (4) mineral aerosol and fugitive dust. The mixed dust and pollution factor was the largest contributor, accounting for $33.8\%$ of TSP mass, followed by mineral aerosol and fugitive dust $(28.0\%)$ . The highest contributions for these two factors were coincident with the high-aerosol episodes, and the interpretations were supported by air mass back trajectory and MODIS AOD analyses.
[51] 3. The concentrations of major chemical species increased several fold during pollution episodes compared with normal days. Evidence obtained from back trajectories and NCEP/NCAR reanalysis highlighted the impact of pollution from northern India on the southeastern TP.
[52] Further studies are needed to better understand the sources and transport of anthropogenic particles from South Asia and dust particles from the surrounding deserts; additional studies also are needed to reveal their potential climatic and environmental impacts.
[53] Acknowledgments. This study was supported by the National Natural Science Foundation of China (41230641, 40925009) and projects from the “Strategic Priority Research Program” of the Chinese Academy of Science (grant XDA05100401). The authors are grateful to Haiqing Xu and Yongjie Wang for their assistance in the aerosol sampling and analyses. Special thanks should go to Darrel Baumgardner for his kind help with grammar revising.
References
An, Z. S., J. E. Kutzbach, W. L. Prell, and S. C. Porter (2001), Evolution of Asian monsoons and phased uplift of the Himalaya-Tibetan plateau since Late Miocene times, Nature, 411(6833), 62–66. Andreae, M. O. (1983), Soot carbon and excess fine potassium: Long-range transport of combustion-derived aerosols, Science, 220(4602), 1,148–1,151. Bonasoni, P., et al. (2008), The ABC-Pyramid Atmospheric Research Observatory in Himalaya for aerosol, ozone and halocarbon measurements, Sci. Total Environ., 391(2–3), 252–261.
Bonasoni, P., et al. (2010), Atmospheric brown clouds in the Himalayas: First two years of continuous observations at the Nepal Climate ObservatoryPyramid $(5079\,\mathrm{m})$ , Atmos. Chem. Phys., 10(15), 7515–7531.
Byson, R. A. (1986), Airstream Climatology of Asia, American Meteorological Society, Boston, pp. 604–617.
Cao, J. J., B. Q. Xu, J. Q. He, X. Q. Liu, Y. M. Han, G. H. Wang, and C. S. Zhu (2009), Concentrations, seasonal variations, and transport of carbonaceous aerosols at a remote mountainous region in western China, Atmos. Environ., 43(29), 4444–4452.
Cao, J. J., S. C. Lee, K. F. Ho, X. Y. Zhang, S. C. Zou, K. Fung, J. C. Chow, and J. G. Watson (2003), Characteristics of carbonaceous aerosol in Pearl River Delta Region, China during 2001 winter period, Atmos. Environ., 37(11), 1451–1460.
Cao, J. J., X. X. Tie, B. Q. Xu, Z. Z. Zhao, C. S. Zhu, G. H. Li, and S. X. Liu (2010), Measuring and modeling black carbon (BC) contamination in the SE Tibetan Plateau, J. Atmos. Chem., 67(1), 45–60.
Cao, J. J., Z. X. Shen, J. C. Chow, J. G. Watson, S. C. Lee, X. X. Tie, K. F. Ho, G. H. Wang, and Y. M. Han (2012), Winter and summer PM2.5 chemical compositions in fourteen Chinese cities, J. Air Waste Manage., 62(10), 1214–1226, doi:10.1080/10962247.2012.701193.
Carrico, C. M., M. H. Bergin, A. B. Shrestha, J. E. Dibb, L. Gomes, and J. M. Harris (2003), The importance of carbon and mineral dust to seasonal aerosol properties in the Nepal Himalaya, Atmos. Environ., 37(20), 2811–2824.
Chan, C. Y., L. Y. Chan, J. M. Harris, S. J. Oltmans, D. R. Blake, Y. Qin, Y. G. Zheng, and X. D. Zheng (2003), Characteristics of biomass burning emission sources, transport, and chemical speciation in enhanced springtime tropospheric ozone profile over Hong Kong, J. Geophys. Res., 108(D1), 4015, doi:10.1029/2001JD001555.
Chen, L.-W. A., J. G. Watson, J. C. Chow, and K. L. Magliano (2007), Quantifying $\mathrm{PM}_{2.5}$ source contributions for the San Joaquin Valley with multivariate receptor models, Environ. Sci. Technol., 41(8), 2818–2826.
Chen, L. -W. A., D. H. Lowenthal, J. G. Watson, D. Koracin, N. Kumar, E. M. Knipping, N. Wheeler, K. Craig, and S. Reid (2010), Toward effective source apportionment using positive matrix factorization: Experiments with simulated $\mathrm{PM}_{2.5}$ Data, J. Air Waste Manage. Assoc., 60, 43–54.
Chow, J. C., J. G. Watson, L. W. A. Chen, M. C. O. Chang, N. F. Robinson, D. Trimble, and S. Kohl (2007), The IMPROVE: A temperature protocol for thermal/optical carbon analysis: Maintaining consistency with a longterm database, J. Air Waste Manage. Assoc., 57(9), 1014–1023.
Chu, D. A., Y. J. Kaufman, G. Zibordi, J. D. Chern, J. Mao, C. Li, and B. N. Holben (2003), Global monitoring of air pollution over land from the Earth Observing System-Terra Moderate Resolution Imaging Spectroradiometer (MODIS), J. Geophys. Res., 108(D21), 4661, doi:10.1029/2002JD003179.
Cong, Z. Y., S. C. Kang, X. D. Liu, and G. F. Wang (2007), Elemental composition of aerosol in the Nam Co region, Tibetan Plateau, during summer monsoon season, Atmos. Environ., 41(6), 1180–1187.
Decesari, S., et al. (2010), Chemical composition of $\mathrm{PM}_{10}$ and $\mathrm{PM}_{1}$ at the high-altitude Himalayan station Nepal Climate Observatory-Pyramid (NCO-P) (5079m a.s.l.), Atmos. Chem. Phys., 10, 4583–4596.
Duncan, B. N., R. V. Martin, A. C. Staudt, R. Yevich, and J. A. Logan (2003), Interannual and seasonal variability of biomass burning emissions constrained by satellite observations, J. Geophys. Res., 108(D2), 4100, doi:10.1029/2002JD002378.
Engling, G., Y.-N. Zhang, C.-Y. Chan, X.-F. Sang, M. Lin, K.-F. Ho, Y.-S. Li, and J. J. Lee (2011), Characterization and sources of aerosol particles over the southeastern Tibetan Plateau during the Southeast Asia biomass-burning season, Tellus B, 63(1), 117–128.
Ganguly, D., A. Jayaraman, T. A. Rajesh, and H. Gadhavi (2006), Wintertime aerosol properties during foggy and nonfoggy days over urban center Delhi and their implications for shortwave radiative forcing, J. Geophys. Res., 111, D15217, doi:10.1029/2005JD007029.
Green, M. C., L.-W. A. Chen, D. W. DuBois, and J. V. Molenar (2012), Fine particulate matter and visibility in the Lake Tahoe Basin: Chemical characterization, trends, and source apportionment, J. Air Waste Manage. Assoc., 62(8), 953–965.
Hindman, E. E., and B. P. Upadhyay (2002), Air pollution transport in the Himalayas of Nepal and Tibet during the 1995–1996 dry season, Atmos. Environ., 36(4), 727–739.
Huang, J., P. Minnis, Y. Yi, Q. Tang, X. Wang, Y. Hu, Z. Liu, K. Ayers, C. Trepte and D. Winker (2007), Summer dust aerosols detected from CALIPSO over the Tibetan Plateau, Geophys. Res. Lett., 34, L18805, doi:10.1029/2007GL029938.
Jain, M., U. C. Kulshrestha, A. K. Sarkar, and D. C. Parashar (2000), Influence of crustal aerosols on wet deposition at urban and rural sites in India, Atmos. Environ., 34(29–30), 5129–5137.
Kajino, M., W. Winiwarter, and H. Ueda (2006), Modeling retained water content in measured aerosol mass, Atmos. Environ., 40(27), 5202–5213.
Kaufman, Y. J., D. Tanré, and O. Boucher (2002), A satellite view of aerosols in the climate system, Nature, 419, 215–223, doi:10.1038/ nature01091.
King, D. M., Y. J. Kaufman, D. Tanré, and T. Nakajima (1999), Remote sensing of tropospheric aerosols from space: Past, present, and future, Bull. Am. Meterol. Soc., 80(11), 2,229–2,259, doi:10.1175/1520-0477 (1999)080.
Lau, W. K. M., M.-K. Kim, K.-M. Kim, and W.-S. Lee (2010), Enhanced surface warming and accelerated snow melt in the Himalayas and Tibetan Plateau induced by absorbing aerosols, Environ. Res. Lett., 5(2), doi:10.1088/1748-9326/5/2/025204.
Li, L., W. Wang, J. L. Feng, D. P. Zhang, H. J. Li, Z. P. Gu, B. J. Wang, G. Y. Sheng, and J. M. Fu (2010), Composition, source, mass closure of $\mathrm{PM}_{2.5}$ aerosols for four forests in eastern China, J. Environ. Sci., 22(3), 405–412.
Li, L. J., Y. Wang, Q. Zhang, T. Yu, Y. Zhao, and J. Jin (2007), Spatial distribution of aerosol pollution based on MODIS data over Beijing, China, J. Environ. Sci., 19(8), 955–960.
Marinoni, A., et al. (2010), Aerosol mass and black carbon concentrations, a two year record at NCO-P $5079\,\mathrm{m}$ , Southern Himalayas), Atmos. Chem. Phys., 10, 8551–8562.
Ming, J., C. D. Xiao, J. Y. Sun, S. C. Kang, and P. Bonasoni (2010), Carbonaceous particles in the atmosphere and precipitation of the Nam Co region, central Tibet, J. Environ. Sci., 22(11), 1748–1756.
Mouli, P., S. Mohan, and S. Reddy (2006), Chemical composition of atmospheric aerosol $(\mathrm{PM}_{10})$ at a semi-arid urban site: Influence of terrestrial sources, Environ. Monit. Assess., 117(1), 291–305.
Moulin, C., and I. Chiapello (2004), Evidence of the control of summer atmospheric transport of African dust over the Atlantic by Sahel sources from TOMS satellites (1979–2000), Geophys. Res. Lett., 31, L02107, doi:10.1029/2003GL018931.
Qu, W. J., X. Y. Zhang, R. Arimoto, D. Wang, Y. Q. Wang, L. W. Yan, and Y. Li (2008), Chemical composition of the background aerosol at two sites in southwestern and northwestern China: Potential influences of regional transport, Tellus B, 60(4), 657–673.
Ram, K., M. M. Sarin, and P. Hegde (2008), Atmospheric abundances of primary and secondary carbonaceous species at two high-altitude sites in India: Sources and temporal variability, Atmos. Environ., 42(28), 6785–6796.
Ramanathan, V., and P. J. Crutzen (2003), New directions: Atmospheric
brown “clouds”, Atmos. Environ., 37(28), 4033–4035.
Ramanathan, V., et al. (2007), Atmospheric brown clouds: Hemispherical and regional variations in long-range transport, absorption, and radiative forcing, J. Geophys. Res., 112, D22S21, doi:10.1029/2006JD008124.
Rastogi, N., and M. M. Sarin (2005), 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(30), 5541–5554.
Rastogi, N., and M. M. Sarin (2009), Quantitative chemical composition and characteristics of aerosols over western India: One-year record of temporal variability, Atmos. Environ., 43(22–23), 3481–3488.
Reff, A., S. I. Eberly, and P. V. Bhave (2007), Receptor modeling of ambient particulate matter data using positive matrix factorization: Review of existing methods, J. Air Waste Manage. Assoc., 57(2), 146–154.
Remer, L. A., et al. (2005), The MODIS aerosol algorithm, products and validation, J. Atmos. Sci., 62, 947–973.
Rengarajan, R., M. M. Sarin, and A. K. Sudheer (2007), Carbonaceous and inorganic species in atmospheric aerosols during wintertime over urban and high-altitude sites in North India, J. Geophys. Res., 112, D21307, doi:10.1029/2006JD008150.
Shen, Z. X., J. J. Cao, R. Arimoto, R. J. Zhang, D. M. Jie, S. X. Liu, and C. S. Zhu (2007), Chemical composition and source characterization of spring aerosol over Horqin sand land in northeastern China, J. Geophys. Res., 112, D14315, doi:10.1029/2006JD007991.
Shen, Z. X., J. J. Cao, R. Arimoto, Z. W. Han, R. J. Zhang, Y. M. Han, S. X. Liu, T. Okuda, S. Nakao, and S. Tanaka (2009), Ionic composition of TSP and $\mathrm{PM}_{2.5}$ during dust storms and air pollution episodes at Xi’an, China, Atmos. Environ., 43(18), 2911–2918.
Shrestha, A. B., C. P. Wake, J. E. Dibb, P. A. Mayewski, S. I. Whitlow, G. R. Carmichael, and M. Ferm (2000), Seasonal variations in aerosol concentrations and compositions in the Nepal Himalaya, Atmos. Environ., 34(20), 3349–3363.
Tang, J., X. L. Yu and H. S. Xue (2005), Observation of the water-soluble species in aerosols in summer in Lhasa area: Contribution of biomass burning emission, Paper presented at eighth symposium on aerosol research and aerosol technology in China and cross the Taiwan Strait, pp. 355, Nanjing, China. (in Chinese).
Tang, M. (1998), Formation, Evolution and Variability Characteristics of Qinghai-Tibetan Plateau Monsoon, Guangdong Sciences and Technology Press, Guangzhou, pp. 161–182.
Torres, O., A. Tanskanen, B. Veihelmann, C. Ahn, R. Braak, P. K. Bhartia, P. Veefkind, and P. Levelt (2007), Aerosols and surface UV products from ozone monitoring instrument observations: An overview, J. Geophys. Res., 112, D24S47, doi:10.1029/2007JD008809.
Turpin, B. J., and H.-J. Lim (2001), Species contributions to $\mathrm{PM}_{2.5}$ mass concentrations: Revisiting common assumptions for estimating organic mass, Aerosol Sci. Tech., 35(1), 602–610.
Venkataraman, C., C. K. Reddy, S. Josson, and M. S. Reddy (2002), Aerosol size and chemical characteristics at Mumbai, India, during the INDOEXIFP (1999), Atmos. Environ., 36(12), 1979–1991.
Venzac, H., et al. (2008), High frequency new particle formation in the Himalayas, Proc. Natl. Acad. Sci. U. S. A., 105(41), 15,666–15,671.
Watson, J. G., T. Zhu, J. C. Chow, J. P. Engelbrecht, E. M. Fujita, W. E. Wilson (2002), Receptor modeling application framework for particle source apportionment, Chemosphere, 49(9), 1093–1136.
Wen, Y. P., X. B. Xu, J. Tang, X. C. Zhang, and Y. C. Zhang (2001), Enrichment characteristics and origin of atmospheric aerosol elemental at Mt. Waliguan, J. Appl. Meteorol. Sci., 12(4), 400–408 (in Chinese).
Wu, G. X., Y. M. Liu, T. M. Wang, R. J. Wan, X. Liu, W. P. Li, Z. Z. Wang, Q. Zhang, A. M. Duan, and X. Y. Liang (2007), The influence of mechanical and thermal forcing by the Tibetan plateau on Asian climate, J. Hydrometeorol., 8(4), 770–789.
Xia, X. A., X. M. Zong, Z. Y. Cong, H. B. Chen, S. C. Kang, and P. C. Wang (2011), Baseline continental aerosol over the central Tibetan plateau and a case study of aerosol transport from South Asia, Atmos. Environ., 45(39), 7370–7378.
Xia X. A., P. C. Wang, Y. S. Wang, Z. Q. Li, J. Y. Xin, J. Liu and H. B. Chen (2008), Aerosol optical depth over the Tibetan plateau and its relation to aerosols over the Taklimakan Desert, Geophys. Res. Lett., 35, L16804, doi:10.1029/2008GL034981.
Xu, B. Q., et al. (2009), Black soot and the survival of Tibetan glaciers, Proc. Natl. Acad. Sci. U. S. A., 106(52), 22,114–22,118.
Xu, H. M., et al. (2012), Lead concentrations in fine particulate matter after the phasing out of leaded gasoline in Xi’an, China, Atmos. Environ., 46, 217–224.
Xu, J., M. H. Bergin, R. Greenwald, J. J. Schauer, M. M. Shafer, J. L. Jaffrezo, and G. Aymoz (2004), Aerosol chemical, physical, and radiative characteristics near a desert source region of northwest China during ACE-Asia, J. Geophys. Res., 109, D19S03, doi:10.1029/2003JD004239.
Zhang, T., et al. (2011), Water-soluble ions in atmospheric aerosols measured in Xi’an, China: Seasonal variations and sources, Atmos. Res., 102(1–2), 110–119.
Zhang, X. Y., S. L. Gong, R. Arimoto, Z. X. Shen, F. M. Mei, D. Wang, and Y. Cheng (2003), Characterization and temporal variation of Asian dust aerosol from a site in the Northern Chinese deserts, J. Atmos. Chem., 44(3), 241–257.
|
atmosphere
|
0103
|
10.1016/j.chemosphere.2016.06.032
|
table
|
Table 1 Description of sampling sites and duration. Please note that the words in parenthesis against each site represent the short name of each site, and subscripts u, r, c, b and s indicate urban, regional, coastal and cruise site types, respectively. I and II refer to the first and second ship cruise measurement.
|
|
atmosphere
|
0103
|
10.1016/j.chemosphere.2016.06.032
|
image
|
Fig. 1. Box-Whisker plot for $\mathrm{PM}_{2.5}$ mass concentration of each filed campaign. The box denotes the 25, 75 percentiles. The whiskers denote the 5th and 95th percentiles. The points represent max/min value. Crosses and horizontal lines represent the mean value and median value of $\mathrm{PM}_{2.5}$ concentration for entire study period, respectively.
|
|
atmosphere
|
0103
|
10.1016/j.chemosphere.2016.06.032
|
image
|
Fig. 2. The mass concentration of $\mathrm{PM}_{2.5}$ and its major chemical speciation at different locations in China. (unit: $\upmu\mathrm{g}\:\mathsf{m}^{-3}$ ).
|
|
atmosphere
|
0103
|
10.1016/j.chemosphere.2016.06.032
|
image
|
Fig. 3. Comparison of mass concentration of $\mathrm{PM}_{2.5}$ and major chemical species for urban and regional sites in each region. Each color corresponds to a chemical species. The size of each cycle is proportional to the correlation values between the temporal trends of each species at urban and regional site.
|
|
atmosphere
|
0103
|
10.1016/j.chemosphere.2016.06.032
|
image
|
Fig. 4. Relative contribution of each species (OM, $S0_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ , $\mathrm{NH_{4}^{+}}$ , EC, and others) to $\mathrm{PM}_{2.5}$ is shown as a function of total mass concentration of $\mathrm{PM}_{2.5}$ The grey line denotes the trend of fraction of SNA as a function of $\mathrm{PM}_{2.5}$ level.
|
|
atmosphere
|
0103
|
10.1016/j.chemosphere.2016.06.032
|
text
| Not supported with pagination yet |
Spatial distributions and chemical properties of PM2.5 based on 21 field campaigns at 17 sites in China
Jing Zheng a, Min Hu a, \*, Jianfei Peng a, Zhijun Wu a, Prashant Kumar b, c, Mengren Li a, Yujue Wang a, Song Guo
a State Key Joint Laboratory of Environmental Simulation and Pollution Control, College of Environmental Sciences and Engineering, Peking University,
Beijing, 100871, China
b Department of Civil and Environmental Engineering, Faculty of Engineering and Physical Science (FEPS), University of Surrey, Guildford GU2 7XH, Surrey,
United Kingdom
c Environmental Flow (EnFlo) Research Centre, FEPS, University of Surrey, Guildford GU2 7XH, Surrey, United Kingdom
h i g h l i g h t s
Chemical characteristics of $\mathrm{PM}_{2.5}$ are evaluated at 17 diversified sites in China.
Regional air pollution in major areas of China are characterized.
The dominant contributors to $\mathsf{P M}_{2.5}$ are identified for major Northern and Southern China.
a r t i c l e i n f o
Article history:
Received 16 January 2016
Received in revised form
7 June 2016
Accepted 8 June 2016
Available online 21 June 2016
Handling Editor: R. Ebinghaus
Keywords:
Air pollution
China
PM2.5
Chemical properties
Secondary inorganic ions
a b s t r a c t
Severe air pollution and its associated health impacts have become one of the major concerns in China. A detailed analysis of $\mathrm{PM}_{2.5}$ chemical compositions is critical for optimizing pollution control measures. In this study, daily 24-h bulk filter samples were collected and analyzed for totally 21 field campaigns at 17 sites in China between 2008 and 2013. The 17 sites were classified into four groups including six urban sites, seven regional sites, two coastal sites in four fast developing regions of China (i.e. Beijing-TianjinHebei region, Yangtze River Delta, Pearl River Delta and Sichuan Basin), and two ship cruise measurements covered the East China Sea and Yellow Sea of China. The high average concentrations of $\mathrm{PM}_{2.5}$ and the occurrences of extreme cases at most sites imply the widespread air pollution in China. Fine particles were largely composed of organic matter and secondary inorganic species at most sites. High correlation between the temporal trends of $\mathrm{PM}_{2.5}$ and secondary species of urban and regional sites highlights the uniformly distributed air pollutants within one region. Secondary inorganic species were the dominant contributors to the high $\mathrm{PM}_{2.5}$ concentration in Northern China. However in Southern China, the relative contributions of different chemical species kept constant as $\mathsf{P M}_{2.5}$ increased. This study provides us a better understanding of the current state of air pollution in diversified Chinese cities. Analysis of chemical signatures of $\mathrm{PM}_{2.5}$ could be a strong support for model validation and emission control strategy.
$\circledcirc$ 2016 Elsevier Ltd. All rights reserved.
1. Introduction
Air pollution is one of the most prominent environmental concerns in China due to its association with adverse effects on human health (Heal et al., 2012). High concentrations of gaseous pollutants such as sulfur dioxide $\left({\mathrm{SO}}_{2}\right)$ , ozone $\left(0_{3}\right)$ and particulate matter (PM) from diverse sources coexist in the atmosphere, which far exceed the atmospheric self-purification capacity, and the complicate interactions among them lead to the formation of a complex air pollution that is difficult to disentangle (Shao et al., 2006). Air pollution comes from a complex mixture of sources such as traditional coal combustion, vehicular emissions, and secondary pollution (Kumar et al., 2011; Huang et al., 2014). In 2014, the annual average concentration of $\mathsf{P M}_{2.5}$ (PM with aerodynamic equivalent diameters less than $2.5\ \ \upmu\mathrm{m}\mathrm{)}$ ) in 161 cities reached $62\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , while only $11.2\%$ of these cities met the Grade II of China National Ambient Air Quality Standard (Report on the state of the Environment in China, 2015). In recent years, severe and longlasting haze affected China on many occasions. The pollution episodes which affected 1.3 million $\mathrm{km}^{2}$ and 800 million people gained worldwide attention during the first quarter of 2013 (Huang et al., 2014).
Among all the air pollutants, ambient particles are of great interest to the scientific community and policy makers because of their ability to carry health risks, as well as influence air quality and global climate (Dusek et al., 2006; Kumar et al., 2010). The researches on chemical characteristics of particles remain key priorities since the mass and ratios of different components imply the various sources and formation mechanisms. It is crucial to develop detailed chemical databases of particles to precisely conduct source apportionment and evaluate environmental impacts. Significant efforts have been dedicated to understanding the characteristics and formation of fine particle pollution in megacities, especially for city clusters and megacities with high population density such as the Beijing, Tianjin and Hebei (BTH) region (He et al., 2001; Guo et al., 2010), Yangtze River Delta (YRD) region (Wang et al., 2006; Du et al., 2011), Pearl River Delta (PRD) region (Hu et al., 2008, Hu et al., 2012), and Sichuan Basin (Tao et al., 2013; Chen and Xie, 2014). However, most of the studies so far have focused on urban areas, ignoring the rural regions. Yang et al. (2011b) compiled chemical composition of $\mathsf{P M}_{2.5}$ for about 13 urban sites, 2 rural sites, and one mountain forest site. However, the different sampling and analytical techniques deployed by individual measurements may introduce a bias in comparative studies on air pollutants. Therefore, harmonized and systematic measurements are a key premise for characterizing the general mapping of air quality in China. In addition, pair studies of urban, regional sites of particle pollution need to be conducted to explore the influence of regional air pollution.
Here we present the mass concentrations and chemical compositions of fine particles based on the results from 21 field campaigns conducted at urban, regional, coastal sites and during two different ship cruise measurements. Chemical properties of $\mathsf{P M}_{2.5}$ at the particular sites and seasons are elucidated. Characteristics of regional air pollutions are also analyzed by urban-regional pair studies. The formation of heavy pollution episodes is also discussed. This article, therefore, provides fundamental information for future studies that aim to evaluate the effects of air pollution on human health and climate change as well as to optimize the emission control strategies for the different regions in China and similar environmental conditions elsewhere.
2. Methodology
2.1. Sites description
Twenty one individual field campaigns were conducted at 17 different sites between 2008 and 2013. Each campaign lasted for around 30 days. According to the population density, geographic location (i.e. the distance between observation site and large pollution source), and energy structure, they were broadly classified into four different types (Table 1). These included 6 urban sites, 7 regional sites, 2 coastal sites and 2 ship cruise measurements (see Supporting Information, SI, Section S1).
To highlight regional similarities and differences in particle characteristics, these sites were furtherly divided into 4 large regions according to their geographical location: BTH region $\mathbf{(BJ_{u},Y F_{r}}$ $\mathsf{W Q}_{\mathrm{r}}$ and $\mathbf{CD}_{\mathsf{C}}$ ) in Northern China, YRD Region $\mathrm{(WX_{u},J H_{u},\,H Z L_{r}}$ and $\mathsf{W L}_{\mathsf{C}/\mathrm{r}})$ , PRD Region $\mathrm{GZ_{u},S Z_{u},J M_{r},\,H S_{r}}$ and $\mathsf{C H}_{\mathrm{r}}$ ), and Sichuan Basin $(Z\Upsilon_{\Gamma})$ in Southern China. These four groups represented hazeaffected regions with rapid industrialization and high population density, causing severe air pollution episodes. Among the 17 sites, three urban-regional site pairs were included: $\mathsf{B}\mathsf{J}_{\mathsf{U}}{-}\mathsf{W}\mathsf{Q}_{\mathrm{r}}$ in BTH region, $\mathrm{JH}_{\mathrm{U}}{-}\mathsf{W L}_{\mathrm{c}/\mathrm{r}}$ in YRD region and $\mathsf{G Z_{u}-C H_{r}}$ in PRD region. The regional sites are more than $50~\mathrm{km}$ away from urban centers and have less local emissions compared to urban sites.
2.2. Sampling method
During each individual field campaign, a four-channel $\mathsf{P M}_{2.5}$ sampler (TH-16A, Tianhong, China) was deployed to collect samples on Teflon and Quartz filters. The time resolution was 12-h for the $\mathrm{BJ_{u}},\,\mathrm{GZ_{u}}\,(2008),\mathrm{YF_{r}},\,\mathrm{WQ_{r}},\,\mathrm{JM_{r}},\,\mathrm{ZY_{r}}$ and $\mathsf{C H}_{\mathrm{r}}$ sites. The samples were collected during daytime and nighttime separately. For the remaining campaigns, the time resolution was 24-h. The sampling flow rate was $16.7\,\bar{\mathrm{L}}\,\mathrm{min}^{-1}$ . Each sample set consisted of one or two Teflon filters for the total mass and individual water-soluble ions measurement. Analyses of carbonaceous species were conducted with quartz filters. The quartz filters were pre-treated by heating at $550~^{\circ}\mathrm{C}$ for $6~\mathrm{{h}}$ before each use. After sampling, the filters were stored in the refrigerator at $-20\,^{\circ}\mathrm{C}$ until they were analyzed.
$\mathsf{P M}_{2.5}$ mass concentration was obtained with an analytical balance by the gravimetric method (MettlerToledo AG285) (Yang et al., 2011b). As described in Guo et al. (2010, 2012), seven major watersoluble inorganic compounds (Kþ, $\mathrm{Mg}^{2+}$ , ${\mathsf{C a}}^{2+}$ , $\boldsymbol{\mathrm{NH}}_{4}^{+}$ , $\mathtt{N O}_{3}^{-}$ , $\mathrm{SO}_{4}^{2-}$ and $\mathsf{C l}^{-}.$ ) were analyzed by ion-chromatograph (DIONEX, ${\mathrm{ICS}}{-}2500/$ 2000). Organic carbon (OC) and elemental carbon (EC) of samples from all the sites except for $S Z_{\mathrm{u}}$ were analyzed via the thermaloptical transmission (TOT) method using a Sunset Laboratorybased instrument with NIOSH method (He et al., 2006). The OC and EC concentrations in samples from $\mathsf{S Z}_{\mathrm{u}}$ were obtained by a DRI carbon analyzer, following the IMPROVE thermal optical reflectance (TOR) protocol (Cao et al., 2007). Comparison of EC concentrations in 27 samples showed EC results determined by TOR correction were about $35\%$ higher than those by TOTcorrection (Fig. S1), which was close to the results obtained by Chow et al. (2004).
3. Results and discussion
3.1. Spatial distribution of chemical species in $P M_{2.5}$
3.1.1. $P M_{2.5}$ concentration
Total 1140 sets of $\mathsf{P M}_{2.5}$ samples were collected at the 17 sites between 2008 and 2013. Descriptive statistics for all the valid observations are summarized in Table S2. In general, the mass concentrations of fine particles at different types of sites were 2e5 times higher than those of corresponding sites in European and American cities (Hidy, 2009; Putaud et al., 2010). As shown in Fig. 1, high levels of $\mathsf{P M}_{2.5}$ concentration (exceeding $100~{\upmu\mathrm{g}}~\mathsf{m}^{-3}.$ were observed at most of sites and covered from the urban to regional scale, indicating pollution episodes did not occur at or dependent on one particular season or site, but spread all over China. The average $\mathsf{P M}_{2.5}$ concentrations were nearly identical at urban and regional pair sites of BTH, PRD and YRD regions, indicating the regional feature in the whole city cluster. The mean concentrations at two coastal sites $\mathrm{{[CD_{c}}}$ and $\mathsf{W L}_{\mathsf{c}/\mathrm{r}}$ ) lie well above the median values due to the strong increases during injection of continental air mass with large amounts of air pollutants, which lead to skewed distributions of concentrations. The $\mathsf{P M}_{2.5}$ concentrations measured during ship cruise measurements were relatively higher than that of Indian seas (Quinn and Bates, 2005), indicating that the East China sea and Yellow sea were strongly influenced by the inland anthropogenic activity.
3.1.2. Chemical speciation
Relative contributions of each species to $\mathsf{P M}_{2.5}$ at different sites are plotted in Fig. 2. The daily averaged chemical data sets for each individual field campaign are also presented in Table S2. Here we report organic matter (OM) concentration by scaling OC by a constant factor of 1.8 $(0\mathsf{M}=1.8\,\times\,0\mathsf{C})$ for all sites (Turpin and Lim, 2001). However it is possible that the actual scaling factor is inconstant and it might be higher at rural sites for organic aerosols due to a higher degree of aging (Zhang et al., 2011). The most abundant species in $\mathsf{P M}_{2.5}$ in all campaigns were OM, sulfate, nitrate, ammonium, and EC. As shown in Fig. 2, despite the dominant components were same in China, the fractions of each compound varied between different sites and sampling periods. The chemical speciations measured in China are similar to the European studies in the sense, that OM accounted for the highest fraction of $\mathsf{P M}_{2.5}$ at urban sites and that sulfate contributed more at regional and background sites (Putaud et al., 2004).
3.1.2.1. Carbonaceous species. Carbonaceous aerosols in China are generated from the widespread use of coal, biomass burning, and petroleum products. Residential coal contributes significantly to black carbon in China (Bond et al., 2013). Carbonaceous species represented a large fraction of $\mathsf{P M}_{2.5}.$ , accounting for around $30\%$ . The variations of OC/EC ratio may be used as an index reflecting emission sources and secondary organic aerosol (SOA) formation (Zeng and Wang, 2011). In this study, OC/EC ratios ranged from 2.74 to 7.46, within the ranges reported by the literatures of China (Table S3). Comparatively higher OC/EC ratios were observed during ship cruises measurements, due to the SOA formation during long-range transportation from inland to the ocean and biogenically driven OC’s contribution to marine aerosol (O’Dowd et al., 2004).
For urban sites, the results were comparable to the concentrations of carbonaceous aerosols at other Chinese sites and some Asia urban sites (Table S3). The heating season with a greater consumption of coals in northern China is a plausible explanation for the high OC concentration in $\mathtt{B J_{u}}$ . The OC concentration in the wintertime in $Z\Upsilon_{\Gamma}$ in Sichuan Basin ranked second after $\mathtt{B J u}$ wintertime. The stagnant dispersion conditions caused by mountain-basin topography in this region enhanced the accumulation of pollutants, leading to a high concentration of OC. In addition, biomass burning in this region also contributed to the high concentration of carbonaceous aerosol (Chen and Xie, 2014).
Relatively high concentrations of OC and EC were found at the regional sites $\left(\mathsf{W L}_{\mathsf{c}/\mathrm{r}}\right)$ compared to those of regional/background sites in other part of world (Table S3). This may be due to the strong anthropogenic influences from YRD region, respectively. The measurement at $\mathbf{CD}_{\mathsf{c}}$ was comparable to the previous observation during the spring of 2003 (Feng et al., 2007). Asian continental flow carried pollutants to this downstream site.
3.1.2.2. Inorganic species. The anthropogenic emissions of precursors ( $S0_{2}$ , NOx, and $\mathsf{N H}_{3}$ ) strongly influence the concentrations and compositions of SNA aerosol (sum of the mass concentration of $\mathrm{{NH}\ddagger}$ , $\mathrm{SO}_{4}^{2-}$ and $\mathrm{NO}_{3}^{-}$ ) (Wang et al., 2013). The high concentration of precursor gaseous pollutants can partially explain the high SNA concentrations in China (Table S1). In this study, the contributions of sulfate were almost equal to those of OM in urban field campaigns, and higher contributions of sulfate were observed in regional sites. The concentrations of sulfate were $5{-}10$ times the measured concentrations in Europe and United States (Putaud et al., 2004; Hidy, 2009).
Nitric acid is mainly formed by the reaction of $\mathsf{N O}_{2}$ with OH radical during daytime or through hydrolysis of $\mathsf{N}_{2}0_{5}$ during nighttime. The former pathway dominates $\mathrm{HNO}_{3}$ formation (Alexander et al., 2009). Although the PRD region has higher NOx emission intensity (Wang et al., 2013), the concentrations and the relative contributions of nitrate in this region were lower than those of other regions (below $10\%$ of $\mathsf{P M}_{2.5}$ ). The reason is partly attributed to the difficulties in precisely measuring particulate nitrate due to its volatilization under high temperatures.
In general, SNA contributed to more than $90\%$ of the total inorganic ionic species mass and accounted for $\sim\!50\%$ of $\mathsf{P M}_{2.5}$ during all field campaigns. Regional sites presented a comparatively higher percentage of SNA than urban sites, indicating relative higher contribution of secondary aerosol formation. It is consistent with European results that the regional background aerosol of Spain exhibited similar secondary characteristics (Salvador et al., 2012).
For the sites strongly influenced by biomass burning (i.e. ZYr and $\mathrm{HS}_{\mathrm{r}}.$ ), the concentrations of $K^{+}$ could be more than $1~{\upmu\mathrm{g}}~\mathrm{m}^{-3}$ , accounting for $\sim\!\!2\%$ in $\mathsf{P M}_{2.5}$ . Higher chloride concentration at $\mathbf{CD}_{\mathsf{c}}$ was mainly attributed to sea salt aerosols, for the coastal site was influenced by marine air masses (Wu et al., 2006). Relative higher concentrations of $C a^{2+}$ were observed at $\mathrm{UR}_{\mathrm{u}}$ and $\mathsf{W}\mathsf{X}_{\mathsf{u}}$ . It has been reported that in spring, soil dust was transported from Jungger Basin and Guebantongute desert in the north of $\mathrm{UR}_{\mathrm{u}}$ , resulting in a higher contribution of ${C a^{2+}}$ (Li et al., 2008). The neutralizing level of $\mathsf{P M}_{2.5}$ was determined by the equivalent charge ratio of major cations to anions. The result shows the acidic aerosol can be neutralized at most of the sites (see Fig. S2).
The contribution of unidentified fraction (others) is estimated by taking the difference of the sum of OM, EC, and inorganic constituents from the $\mathsf{P M}_{2.5},$ , accounting for $2{-}40\%$ among all the field campaigns. The unidentified matter mainly consists of crustal dust and sea salt. Crustal species tend to play a minor role in $\mathsf{P M}_{2.5}$ (around $10\%$ ) (Zhou et al., 2016). But given their importance in source apportionment, detailed analysis of elements will be done in future analysis.
3.2. Chemical characterizations of fine particles
3.2.1. Characteristic of regional air pollution
The role of regional air pollution due to the rapid urbanization has been studied extensively in the past few years (Guo et al., 2010; Yang et al., 2011a; Li et al., 2014). Regional pollution is regarded as the regional scale elevated concentrations of air pollutants, which mainly due to the secondary formation and the horizontal and vertical homogeneity of the meteorological conditions of the whole region (Jia et al., 2008). Previous results revealed that persistent regional stagnation conditions favored the elevation of PM mass or $\bar{\mathrm{S}}0_{4}^{2-}$ on a regional scale from the mid-western to northeastern of United States (Blanchard et al., 2013). Although several studies have been performed at urban/rural pair sites to investigate regional air pollution in BTH region (Guo et al., 2010; Yang et al., 2011a), a thorough analysis of the chemical components of fine particles for pair sites in China has not been conducted.
The comparison of chemical properties of fine particles at a pair sites supports the assessment of regional air pollution. For most chemical species, urban concentrations were similar to their regional counterparts, indicating the uniform distribution of regional pollution. For EC, the urban concentration reached $222\%$ higher than that of regional sites, showing a stronger local contribution than other components (see Fig. 3). The linear correlations between temporal trends of each species (e.g. concentration of $\mathsf{P M}_{2.5},$ OC, $\bar{\mathrm{S}}0_{4}^{\bar{2}-}$ , $\mathrm{NO}_{3}^{-}$ , $\mathsf{N H}_{4}^{+}$ ) of pair sites were analyzed here in terms of the Spearman correlation analysis, for example the correlation of $\mathsf{P M}_{2.5}$ concentrations of $\mathtt{B J u}$ and $\mathrm{{WQ}_{r}}$ to investigate the regional pollutions of fine particles during the observation periods. Table S4 summarizes the results of the correlation coefficient of each species. The temporal trends of $\mathsf{P M}_{2.5}$ at the urban sites typically correlated well with those of corresponding regional sites. The regional wide elevated PM concentrations indicated the occurrence of regional air pollution. The analysis also showed generally high correlation values between the daily data of pair sites for the secondary inorganic species. The $\mathrm{SO}_{4}^{2-}$ correlations for BTH, YRD, and PRD region ranged from 0.74 to 0.78. This is consistent with a regional formation mechanism with $S0_{4}^{2-}$ . For the primary traffic-related components, EC, the correlation coefficient decreased to around 0.5 for the three regions, indicating the influence of local emissions (see Fig. 3). High correlations between the temporal trends of $\mathsf{P M}_{2.5}$ and different chemical components of urban and regional sites highlighted the uniformly distributed air pollutants within one region. The regional influences were greater for secondary species than primary components.
3.2.2. Formation of heavy pollution episodes
Given the coupling interactions between complex meteorological conditions, pollution sources, and atmospheric transformation processes, characterization of formation mechanisms of severe haze with high $\mathsf{P M}_{2.5}$ levels at different sites presents a major challenge (Guo et al., 2014). Fig. 4 shows the relative contribution of OM, EC, $\mathrm{SO}_{4}^{2-}$ , $\mathtt{N O}_{3}^{-}$ , $\mathrm{NH}_{4}^{+}$ and others to $\mathsf{P M}_{2.5}$ as a function of $\mathsf{P M}_{2.5}$ mass concentration. In Northern China (BTH region and $\mathbf{CD}_{\mathsf{C}}\,$ ), the fraction of OM in fine particle decreased with increasing $\mathsf{P M}_{2.5}$ concentration, while the relative contributions of secondary inorganic species increased at the urban, regional, and coastal sites. When $\mathsf{P M}_{2.5}$ concentration reached peak values, the fractions of SNA were around $60\%{-}80\%$ , indicating the importance of secondary formation in the pollution episodes. Previous research with back trajectory analysis revealed that the clean air masses from the north had a larger contribution from primary aerosol emissions, characterized by a higher contribution of OM, while haze pollution occurred when stagnant air masses with secondary regional aerosols dominated (Huang et al., 2010; Sun et al., 2010). A broadly consistent pattern was also observed in Taiyuan, which is a heavily polluted city in Northern China (Meng et al., 2007). However in the Southern China (YRD and PRD region), the relative contributions of different chemical components remained relatively constant (the increase of secondary aerosols concentration was lower than $15\%$ as the $\mathsf{P M}_{2.5}$ concentration increased. No significant increase of SNA fraction in $\mathsf{P M}_{2.5}$ was also observed during a heavy polluted episode in the YRD region, with $42\%$ on moderate days and $52\%$ on the heavily polluted days (Fu et al., 2008).
Although secondary formation was a major cause of particulate pollution in the whole China, the discrepancy of relative contribution from different chemical components to $\mathsf{P M}_{2.5}$ level implies diverse formation mechanisms of heavy air pollution at different sites of China. Comparatively stronger emissions of gas precursors ( $\mathbf{e}.\mathbf{g}\;\mathbf{S}0_{2},\;\mathbf{NO}_{\mathrm{x}})$ in Northern China would have a potential effect on SNA formation (Wang et al., 2013). It may be a plausible explanation for the dominant contribution of SNA. It is of interest to perform detail analyses on the relationship of geographical features with formation mechanisms of heavy pollution at specific sites in China.
4. Summary and conclusions
The paper presents comprehensive assessments of the chemical property of fine aerosol particles at different sites in China. Datasets of chemical characteristics of fine particles from 21 field campaigns at 17 sites are analyzed to understand the sources and transformation of $\mathsf{P M}_{2.5}$ in the atmosphere.
Heavy pollution episodes with $\mathsf{P M}_{2.5}$ higher than $100~\upmu\mathrm{g}~\mathrm{m}^{-3}$ were observed during most of the campaigns, independent of sites and seasons. High $\mathsf{P M}_{2.5}$ mass concentrations of coastal sites and during cruise measurements indicated offshore regions of China were strongly influenced by continental air pollutants. Organics and sulfate were the most abundant species in $\mathsf{P M}_{2.5}$ . SNA accounted for about $50\%$ of the $\mathsf{P M}_{2.5}$ concentration.
Simultaneous measurements conducted at three urban-regional pairs of sites demonstrated the urban and regional sites had similar concentrations of pollutants. The similar variations of temporal trends of $\mathsf{P M}_{2.5}$ and secondary inorganic species of pairs sites indicated a uniform distribution of pollution.
Secondary inorganic species were the dominant contributors to high $\mathsf{P M}_{2.5}$ concentration in Northern China, indicating a high level of the secondary formation during haze days. The relative contributions of each species as a function of $\mathsf{P M}_{2.5}$ mass concentration demonstrated that Northern China and Southern China has different formation mechanisms of heavy pollution episode.
Our study presents a comprehensive picture of $\mathsf{P M}_{2.5}$ chemical compositions at a numerous of different types of sites in China and evaluates air-pollutant compositional similarities and differences. This dataset can be further used for more in-depth research on source apportionment and secondary pollutants formation mechanisms, besides carrying out a thorough scientific assessment of health impacts.
Acknowledgements
This research was supported by the National Basic Research Program (2013CB228503), National Natural Science Foundation of China (91544214, 21190052, 41121004) and China Ministry of Environmental Protection’s Special Funds for Scientific Research on Public Welfare (20130916). Prashant Kumar, Jing Zheng and Min Hu thank the University of Surrey’s International Relations Office for the Santander Postgraduate Mobility Award that helped Jing Zheng to visit University of Surrey, UK, to develop this research article collaboratively, and Misti Levy (Texas A&M University) for revising English writing on the manuscript. We also thank Min Shao, Shaodong Xie, Yuanhang Zhang for their support to different field campaigns.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.chemosphere.2016.06.032.
References
Alexander, B., Hastings, M.G., Allman, D.J., Dachs, J., Thornton, J.A., Kunasek, S.A., 2009. Quantifying atmospheric nitrate formation pathways based on a global model of the oxygen isotopic composition (D17O) of atmospheric nitrate. Atmos. Chem. Phys. 9, 5043e5056.
Blanchard, C.L., Hidy, G.M., Tanenbaum, S., Edgerton, E.S., Hartsell, B.E., 2013. The Southeastern aerosol research and characterization (SEARCH) study: spatial variations and chemical climatology, 1999e2010. J. Air Waste Manag. 63, 260e275.
Bond, T.C., Doherty, S.J., Fahey, D.W., Forster, P.M., Berntsen, T., DeAngelo, B.J., Flanner, M.G., Ghan, S., K€archer, B., Koch, D., Kinne, S., Kondo, Y., Quinn, P.K., Sarofim, M.C., Schultz, M.G., Schulz, M., Venkataraman, C., Zhang, H., Zhang, S., Bellouin, N., Guttikunda, S.K., Hopke, P.K., Jacobson, M.Z., Kaiser, J.W., Klimont, Z., Lohmann, U., Schwarz, J.P., Shindell, D., Storelvmo, T., Warren, S.G., Zender, C.S., 2013. Bounding the role of black carbon in the climate system: a scientific assessment. J. Geophys. Res. 118, 5380e5552.
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. 112.
Chen, Y., Xie, S.-d., 2014. Characteristics and formation mechanism of a heavy air pollution episode caused by biomass burning in Chengdu, Southwest China. Sci. Total Environ. 473e474, 507e517.
Chow, J.C., Watson, J.G., Chen, L.-W.A., Arnott, W.P., Moosmuller, H., 2004. Equivalence of elemental carbon by thermal/optical reflectance and transmittance with different temperature protocols. Environ. Sci. Technol. 38, 4414e4442.
Du, H., Kong, L., Cheng, T., Chen, J., Du, J., Li, L., Xia, X., Leng, C., Huang, G., 2011. Insights into summertime haze pollution events over Shanghai based on online water-soluble ionic composition of aerosols. Atmos. Environ. 45, 5131e5137.
Dusek, U., Frank, G.P., Hildebrandt, L., Curtius, J., Schneider, J., Walter, S., Chand, D., Drewnick, F., Hings, S., Jung, D., Borrmann, S., Andreae, M.O., 2006. Size matters more than chemistry for cloud-nucleating ability of aerosol particles. Science 312, 1375e1378.
Feng, J., Guo, Z., Chan, C.K., 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.
Fu, Q., Zhuang, G., Wang, J., Xu, C., Huang, K., Li, J., Hou, B., Lu, T., Streets, D.G., 2008. Mechanism of formation of the heaviest pollution episode ever recorded in the Yangtze River Delta, China. Atmos. Environ. 42, 2023e2036.
Guo, S., Hu, M., Wang, Z.B., Slanina, J., Zhao, Y.L., 2010. Size-resolved aerosol watersoluble ionic compositions in the summer of Beijing: implication of regional secondary formation. Atmos. Chem. Phys. 10, 947e959.
Guo, S., Hu, M., Guo, Q., Zhang, X., Zheng, M., Zheng, J., Chang, C.C., Schauer, J.J., Zhang, R., 2012. Primary sources and secondary formation of organic aerosols in Beijing, China. Environ. Sci. Technol. 46, 9846e9853.
Guo, S., Hu, M., Zamora, M.L., Peng, J., Shang, D., Zheng, J., Du, Z., Wu, Z., Shao, M., Zeng, L., Molina, M.J., Zhang, R., 2014. Elucidating severe urban haze formation in China. Proc. Natl. Acad. Sci. U. S. A. 111, 17373e17378.
He, K.B., Yang, F.M., Ma, Y.L., Zhang, Q., Yao, X.H., Chan, C.K., Cadle, S., Chan, T., Mulawa, P., 2001. The characteristics of $\mathrm{PM}_{2.5}$ in Beijing, China. Atmos. Environ. 35, 4959e4970.
He, L.Y., Hu, M., Huang, X.F., Zhang, Y.H., Tang, X.Y., 2006. Seasonal pollution characteristics of organic compounds in atmospheric fine particles in Beijing. Sci. Total Environ. 359, 167e176.
Heal, M.R., Kumar, P., Harrison, R.M., 2012. Particles, air quality, policy and health. Chem. Soc. Rev. 41, 6606e6630.
Hidy, G.M., 2009. Surface-level fine particle mass concentrations: from hemispheric distributions to megacity sources. J. Air Waste Manag. 59, 770e789.
Hu, M., Wu, Z., Slanina, J., Lin, P., Liu, S., Zeng, L., 2008. Acidic gases, ammonia and water-soluble ions in $\mathrm{PM}_{2.5}$ at a coastal site in the Pearl River Delta, China. Atmos. Environ. 42, 6310e6320.
Hu, W.W., Hu, M., Deng, Z.Q., Xiao, R., Kondo, Y., Takegawa, N., Zhao, Y.J., Guo, S., Zhang, Y.H., 2012. The characteristics and origins of carbonaceous aerosol at a rural site of PRD in summer of 2006. Atmos. Chem. Phys. 12, 1811e1822.
Huang, R.J., Zhang, Y., Bozzetti, C., Ho, K.F., Cao, J.J., Han, Y., Daellenbach, K.R., Slowik, J.G., Platt, S.M., Canonaco, F., Zotter, P., Wolf, R., Pieber, S.M., Bruns, E.A., Crippa, M., Ciarelli, G., Piazzalunga, A., Schwikowski, M., Abbaszade, G., Schnelle-Kreis, J., Zimmermann, R., An, Z., Szidat, S., Baltensperger, U., El Haddad, I., Prevot, A.S., 2014. High secondary aerosol contribution to particulate pollution during haze events in China. Nature 514, 218e222.
Huang, X.F., He, L.Y., Hu, M., Canagaratna, M.R., Sun, Y., Zhang, Q., Zhu, T., Xue, L., Zeng, L.W., Liu, X.G., Zhang, Y.H., Jayne, J.T., Ng, N.L., Worsnop, D.R., 2010. Highly time-resolved chemical characterization of atmospheric submicron particles during 2008 Beijing olympic games using an aerodyne high-resolution aerosol mass spectrometer. Atmos. Chem. Phys. 10, 8933e8945.
Jia, Y., Rahn, K.A., He, K., Wen, T., Wang, Y., 2008. A novel technique for quantifying the regional component of urban aerosol solely from its sawtooth cycles. J. Geophys. Res. 113.
Kumar, P., Gurjar, B.R., Nagpure, A.S., Harrison, R.M., 2011. Preliminary estimates of nanoparticle number emissions from road vehicles in megacity Delhi and associated health impacts. Environ. Sci. Technol. 45, 5514e5521.
Kumar, P., Robins, A., Vardoulakis, S., Britter, R., 2010. A review of the characteristics of nanoparticles in the urban atmosphere and the prospects for developing regulatory controls. Atmos. Environ. 44, 5035e5052.
Li, J., Zhuang, G., Huang, K., Lin, Y., Xu, C., Yu, S., 2008. Characteristics and sources of air-borne particulate in Urumqi, China, the upstream area of Asia dust. Atmos. Environ. 42, 776e787.
Li, W., Wang, C., Wang, H., Chen, J., Yuan, C., Li, T., Wang, W., Shen, H., Huang, Y., Wang, R., Wang, B., Zhang, Y., Chen, H., Chen, Y., Tang, J., Wang, X., Liu, J., Coveney Jr., R.M., Tao, S., 2014. Distribution of atmospheric particulate matter (PM) in rural field, rural village and urban areas of northern China. Environ. Pollut. 185, 134e140.
Meng, Z.Y., Jiang, X.M., Yan, P., Lin, W.L., Zhang, H.D., Wang, Y., 2007. Characteristics and sources of $\mathrm{PM}_{2.5}$ and carbonaceous species during winter in Taiyuan, China. Atmos. Environ. 41, 6901e6908.
Ministry of Environment Protection, 2015. Report on the State of the Environment in China.
O’Dowd, C.D., Facchini, M.C., Cavalli, F., Ceburnis, D., Mircea, M., Decesari, S., Fuzzi, S., Yoon, Y.J., Putaud, J.-P., 2004. Biogenically driven organic contribution to marine aerosol. Nature 437, 676e680.
Putaud, J.-P., Raes, F., Van Dingenen, R., Brüggemann, E., Facchini, M.C., Decesari, S., Fuzzi, S., Gehrig, R., Hüglin, C., Laj, P., Lorbeer, G., Maenhaut, W., Mihalopoulos, N., Müller, K., Querol, X., Rodriguez, S., Schneider, J., Spindler, G., Brink, H.t., Tørseth, K., Wiedensohler, A., 2004. A European aerosol phenomenologyd2: chemical characteristics of particulate matter at kerbside, urban, rural and background sites in Europe. Atmos. Environ. 38, 2579e2595.
Putaud, J.P., Van Dingenen, R., Alastuey, A., Bauer, H., Birmili, W., Cyrys, J., Flentje, H., Fuzzi, S., Gehrig, R., Hansson, H.C., Harrison, R.M., Herrmann, H., Hitzenberger, R., Hüglin, C., Jones, A.M., Kasper-Giebl, A., Kiss, G., Kousa, A., Kuhlbusch, T.A.J., L€oschau, G., Maenhaut, W., Molnar, A., Moreno, T., Pekkanen, J., Perrino, C., Pitz, M., Puxbaum, H., Querol, X., Rodriguez, S., Salma, I., Schwarz, J., Smolik, J., Schneider, J., Spindler, G., ten Brink, H., Tursic, J., Viana, M., Wiedensohler, A., Raes, F., 2010. A European aerosol phenomenology e 3: physical and chemical characteristics of particulate matter from 60 rural, urban, and kerbside sites across Europe. Atmos. Environ. 44, 1308e1320.
Quinn, P.K., Bates, T.S., 2005. Regional aerosol properties: comparisons of boundary layer measurements from ACE 1, ACE 2, Aerosols99, INDOEX, ACE Asia, TARFOX, and NEAQS. J. Geophys. Res. 110.
Salvador, P., Artí\~nano, B., Viana, M., Alastuey, A., Querol, X., 2012. Evaluation of the changes in the Madrid metropolitan area influencing air quality: analysis of 1999e2008 temporal trend of particulate matter. Atmos. Environ. 57, 175e185.
Shao, M., Tang, X., Zhang, Y., Li, W., 2006. City clusters in China: air and surface water pollution. Front. Ecol. Environ. 4, 353e361.
Sun, J., Zhang, Q., Canagaratna, M.R., Zhang, Y., Ng, N.L., Sun, Y., Jayne, J.T., Zhang, X., Zhang, X., Worsnop, D.R., 2010. Highly time- and size-resolved characterization of submicron aerosol particles in Beijing using an Aerodyne Aerosol Mass Spectrometer. Atmos. Environ. 44, 131e140.
Tao, J., Chen, T., Zhang, R., Cao, J., Zhu, L., Wang, Q., Luo, L., Zhang, L., 2013. Chemical composition of $\mathrm{PM}_{2.5}$ at an urban site of Chengdu in Southwestern China. Adv. Atmos. Sci. 30, 1070e1084.
Turpin, B.J., Lim, H.-J., 2001. Species contributions to $\mathrm{PM}_{2.5}$ mass concentrations: revisiting common assumptions for estimating organic mass. Aerosol Sci. Technol. 35, 602e610.
Wang, Y., Zhang, Q.Q., He, K., Zhang, Q., Chai, L., 2013. Sulfate-nitrate-ammonium aerosols over China: response to 2000e2015 emission changes of sulfur dioxide, nitrogen oxides, and ammonia. Atmos. Chem. Phys. 13, 2635e2652.
Wang, Y., Zhuang, G., Zhang, X., Huang, K., Xu, C., Tang, A., Chen, J., An, Z., 2006. The ion chemistry, seasonal cycle, and sources of $\mathrm{PM}_{2.5}$ and TSP aerosol in Shanghai. Atmos. Environ. 40, 2935e2952.
Wu, D., Tie, X., Deng, X., 2006. Chemical characterizations of soluble aerosols in southern China. Chemosphere 64, 749e757.
Yang, F., Huang, L., Duan, F., Zhang, W., He, K., Ma, Y., Brook, J.R., Tan, J., Zhao, Q., Cheng, Y., 2011a. Carbonaceous species in $\mathrm{PM}_{2.5}$ at a pair of rural/urban sites in Beijing, 2005e2008. Atmos. Chem. Phys. 11, 7893e7903.
Yang, F., Tan, J., Zhao, Q., Du, Z., He, K., Ma, Y., Duan, F., Chen, G., Zhao, Q., 2011b. Characteristics of PM2.5 speciation in representative megacities and across China. Atmos. Chem. Phys. 11, 5207e5219.
Zeng, T., Wang, Y., 2011. Nationwide summer peaks of OC/EC ratios in the contiguous United States. Atmos. Environ. 45, 578e586.
Zhang, Q., Jimenez, J.L., Canagaratna, M.R., Ulbrich, I.M., Ng, N.L., Worsnop, D.R., Sun, Y., 2011. Understanding atmospheric organic aerosols via factor analysis of aerosol mass spectrometry: a review. Anal. Bioanal. Chem. 401, 3045e3067.
Zhou, X., Cao, Z., Ma, Y., Wang, L., Wu, R., Wang, W., 2016. Concentrations, correlations and chemical species of $\mathrm{PM}_{2.5}/\mathrm{PM}_{10}$ based on published data in China: potential implications for the revised particulate standard. Chemosphere 144, 518e526.
|
atmosphere
|
0104
|
10.1016/j.atmosres.2016.03.019
|
image
|
Fig. 1. Location of sampling sites in Baotou. BGH—Baogang Hotel, STU—Inner Mongolia Science and Technology University, QSH-Qingshan Hotel, EMB—Environmental monitoring station of Baotou, EPB—Environmental Protection Agency of Baotou, and DHE—Environmental Protection Agency of Donghe District
|
|
atmosphere
|
0104
|
10.1016/j.atmosres.2016.03.019
|
table
|
Table 1 Descriptive statistics and meteorological conditions during sampling period
|
|
atmosphere
|
0104
|
10.1016/j.atmosres.2016.03.019
|
image
|
Fig. 2. Plots of wind rose for sampling period.
|
|
atmosphere
|
0104
|
10.1016/j.atmosres.2016.03.019
|
table
|
Table 2 The concentrations of PM, OC, EC $\left(\upmu\mathrm{g}\cdot\up m^{-3}\right)$ and OC/EC ratios in Baotou.
|
|
atmosphere
|
0104
|
10.1016/j.atmosres.2016.03.019
|
image
|
Fig. 3. Seasonal variation for OC and EC in (3a) $\mathrm{PM}_{10}$ and (3b) $\mathrm{PM}_{2.5}$ in Baotou.
|
|
atmosphere
|
0104
|
10.1016/j.atmosres.2016.03.019
|
table
|
Table 3 The concentrations of OC, EC, and OC/EC ratios in different cities.
|
|
atmosphere
|
0104
|
10.1016/j.atmosres.2016.03.019
|
image
|
Fig. 5. Values (mean $\pm\,S\mathrm{D}.$ for OC, EC, and OC/EC ratios of the (a) $\mathrm{PM}_{10}$ and (b) $\mathrm{PM}_{2.5}$ at EPB site.
|
|
atmosphere
|
0104
|
10.1016/j.atmosres.2016.03.019
|
image
|
Fig. 6. Meteorological conditions in Baotou.
|
|
atmosphere
|
0104
|
10.1016/j.atmosres.2016.03.019
|
image
|
Fig. 7. Seasonal Scatter plots between OC and EC in (4a) $\mathrm{PM}_{10}$ and (4b) $\mathrm{PM}_{2.5}$
|
|
atmosphere
|
0104
|
10.1016/j.atmosres.2016.03.019
|
image
|
Fig. 8. Mean values of gaseous pollutants and OC, EC in $\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5}$ at EPB site
|
|
atmosphere
|
0104
|
10.1016/j.atmosres.2016.03.019
|
image
|
Fig. 9. Cluster analysis of gaseous pollutants and OC, EC, at EPB site
|
|
atmosphere
|
0104
|
10.1016/j.atmosres.2016.03.019
|
table
|
Table 4 The Pearson correlation coefficients between carbonaceous species, meteorological conditions and gaseous pollutants.
|
|
atmosphere
|
0104
|
10.1016/j.atmosres.2016.03.019
|
text
| Not supported with pagination yet |
The distribution of $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ carbonaceous aerosol in Baotou, China
Haijun Zhou a,b, Jiang He a,c,⁎, Boyi Zhao a, Lijun Zhang b, Qingyun Fan b, Changwei Lü c, Dudagula b, Tao Liu b, Yinghui Yuan a
a College of Life Sciences, Inner Mongolia University, 010021 Hohhot, China b Environmental Monitoring Center of Inner Mongolia, 010011 Hohhot, China c College of Environment and Resources, Inner Mongolia University, 010021 Hohhot, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 11 November 2015
Received in revised form 17 March 2016
Accepted 18 March 2016
Available online xxxx
Keywords:
Organic carbon
Elemental carbon
Secondary organic carbon
$\mathrm{PM}_{10}$
$\mathrm{PM}_{2.5}$
Particulate matter (PM), including $\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5},$ is one of the major impacts on air quality, visibility, climate change, earth radiation balance, and public health. Organic carbon (OC) and elemental carbon (EC) are the major components of PM. 804 samples ( $\mathrm{PM}_{10}$ and $\mathsf{P M}_{2.5.}$ ) were simultaneously collected from six urban sites covering 3 districts in Baotou, in January, April, September, and November 2014. As to a long-term study on the effects of carbonaceous aerosol, data were collected annually at Environmental Protection Agency of Baotou (EPB). The concentrations of $\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5}$ , the spatial distribution and content of OC and EC, the relationship between OC and EC, and the formation of secondary organic carbon (SOC) have been investigated. The findings indicated that the concentrations of these particle matter are higher than that in US or European standards. The average concentrations of OC in $\mathsf{P M}_{10}$ and $\mathrm{PM}_{2.5}$ follow the order: January $>$ November $>$ April $>$ September; and for EC in $\mathrm{PM}_{10}$ and $\mathrm{PM}_{2.5}$ follow the order: January $>$ November $>$ September $>$ April. Affected by metrological factors, it was indicated that high wind speed and low relative humidity were beneficial for removal of OC and EC in January and November. Pearson correlations and cluster analysis on OC and EC concentrations in $\mathsf{P M}_{10}$ and $\mathrm{PM}_{2.5}$ with gaseous pollutants $(\mathsf{S O}_{2},\mathsf{N O}_{2}$ and CO) suggested that OC shared the same emission sources with $S0_{2}$ and CO from combustion, while EC's sources mainly came from vehicles exhaust and combustion which contributed to $\mathsf{N O}_{2}$ as well. The OC concentration is mainly primary in warm months, while it appears secondary in cold months in Baotou. There is a common characteristic among the cities with higher SOC in winter, wherever the coal combustion can lead to the severe pollution. This work is important for the construction of the database of OC and EC concentrations in $\mathrm{PM}_{10}$ and $\mathsf{P M}_{2.5}$ at spatial and time intervals, and it can provide scientific suggestion for similar PM atmospheric pollutant control and air quality improvement in Baotou.
$\circledcirc$ 2016 Elsevier B.V. All rights reserved.
1. Introduction
Particulate matter (PM), including $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ , is one of the major pollutants which worsen the air quality and visibility, accelerate climate change, disturb earth radiation balance, and threaten public health (Habre et al., 2014; Kuo et al., 2013; Menon et al., 2002; Nelin et al., 2012; Ramanathan et al., 2001; Van Ryswyk et al., 2014). Organic carbon (OC) and elemental carbon (EC) are the major components of PM (Daher et al., 2012; Dellinger et al., 2001; Nunes and Pio, 1993; Shafer et al., 2010; Stone et al., 2010; Subhasis et al., 2009; Xia and Nel, 2004). The major sources of OC come from direct emissions called primary organic carbon (POC) and photochemical reaction of volatile hydrocarbon named secondary organic carbon (SOC). It contains polycyclic aromatic hydrocarbons (PAHs), n-alkanes, organic acid, carboxyl compound, polychlorinated biphenyls (PCBs), and etc, most of which are most hazardous, toxic, and even carcinogenic (Mauderly and Chow, 2008; Pope and Dockery, 2006; Vedal, 1997). EC is produced during the incomplete combustion of fossil fuel and biomass, and it can interact with gases ( $S0_{2}$ , $\mathsf{N O}_{\mathtt{X}}$ ) and form toxic species which would further impact human health (Bollasina et al., 2011; Bond et al., 2013; Hansen et al., 2000; Hansen et al., 2005; Lee et al., 2012; Novakov, 1984; Park et al., 2002). China is a major global source of carbonaceous aerosol due to its high usage of coal and biofuels (Cao et al., 2006; Junker and Liousse, 2008). Therefore, it is important to control PM levels and reduce its impacts on the environment, economy, and health by mitigating primary particulate emissions from fossil fuel and biomass combustion. This can further reduce the emissions of secondary aerosol precursors (Ru-Jin et al., 2014).
Baotou is a traditionally industrial city and it is also the largest city in Inner Mongolia. According to the government statistics, coal consumption reached 48.7 million tons in 2014, in which $48\%$ came from electricity generation and heating and the rest came from industrial production (Baotou statistical yearbook, 2015). In recent years, Baotou has suffered several typical air pollution accidents such as visibility degradation, which PM was monitored as the principal pollutant. The systematic study of the spatial distribution, composition, and sources of $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ carbonaceous aerosol in Baotou is rarely reported until recent years. The concentration of $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ , the spatial distribution and content of OC and EC, the relationship between OC and EC, and the formation of SOC have also been studied. The objectives are to construct the database of OC and EC in $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ at spatial and time intervals and to provide scientific suggestion to similar PM atmospheric pollutant control and air quality improvement in Baotou.
2. Material and methods
2.1. Study area and sampling
Baotou $(40^{\circ}15^{\prime}{-}42^{\circ}43^{\prime}\mathrm{N}$ , $109^{\circ}15^{\prime}–110^{\circ}26^{\prime}\mathrm{E})$ is located to $670~\mathrm{km}$ west of Beijing (Fig. 1). It is a typical heavy industrial city in Northern China, with an area of $27.768\ \mathrm{km}^{2}$ and a population of approximately 2.8 million. The average annual precipitation is $240{-}400\ \mathrm{mm}$ which mainly occurs from July to September. Northwest wind prevails in winter and southeast wind prevails in spring and summer. Frequent dust storms occur from March to May. The average wind speed, temperature, and relative humidity is $2.3–3.0\ensuremath{\,\mathrm{m\cdots^{-1}}}$ , $-\,10.3$ to $15.0\,^{\circ}\mathrm{C},$ and $48–68\%$ during the sampling period, respectively (Table 1).
In order to explore the spatial distribution, composition, and source of $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ carbonaceous aerosol in Baotou, 6 sites were selected, including Baogang Hotel (BGH), Inner Mongolia Science and Technology University (STU), Qingshan Hotel (QSH), Environmental Monitoring Station of Baotou (EMB), Environmental Protection Agency of Baotou (EPB), and Environmental Protection Agency of Donghe District (DHE), which all belong to the National Air Quality Monitoring Station (Fig. 1). Besides, these selected sites also cover the 3 districts of Baotou city. BGH and STU were responsible for sampling collection of the Kundulun district and these two sites sit next to the largest pollution emission firm named Baotou Iron and Steel Group. QSH, EMB, and EPB were selected for the pollution control of the Qingshan district, where industrial productions are distributed unevenly, including second power plant of Baotou, first machinery plant and second machinery plant, and etc. For Donghe district, DHE was arranged to control the pollution emission industries, such as the third power plant and aluminum manufacturer of Baotou.
The sampling months were selected not only to investigate the spatial distribution of carbonaceous aerosol but also to investigate the seasonal impacts as well. Therefore, it was marked on the basis of seasonal representatives of heavy pollution in January, frequent dust storm in April, frequent precipitation on September, and winter heating beginning from November. Moreover, to study the long-term seasonal impacts, it has selected EPB site for annual sample collection.
972 samples were collected, in which 804 samples were collected in January, April, September, and November 2014, and 168 samples were collected at EPB site every Tuesday and Saturday in 2014. For the 804 samples, the sampling period were during 4th–21st January, 9th–28th April, 13th–29th September, 11th–25th November, and 18, 18, 16, and 15 samples at each sites for both $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ were collected, respectively. Gaseous pollutants ( $S0_{2}$ , $\mathsf{N O}_{2}$ , $0_{3}$ and CO) concentrations were obtained from National Air Quality Monitoring Stations. Temperature, humidity, wind speed, and wind direction data were provided by the Baotou Meteorological Institute. All the data were synchronous obtained with 972 samples.
The samples were collected with 3 samplers at each site, which were specialized for $\mathsf{P M}_{10},$ $\mathsf{P M}_{2.5}.$ , and field blank, respectively. Samples were simultaneously collected from 10 to 9 a.m. next day at six sites with six operators. The $23\,\mathrm{h}$ samples were continuously collected using quartz filters ( ${\mathcal{O}}90\;\mathrm{mm}$ , Pall Life Sciences, USA) with medium-volume samplers (2050 Intelligent Air/TSP Comprehensive Sampler, manufactured by Qingdao Laoshan Applied Technology Research Institute) at a flow rate of $100\,\mathrm{L}\cdot\mathrm{min}^{-1}$ . The filters were heated at $450\,^{\circ}\mathrm{C}$ for $4\,\mathrm{h}$ to remove the residual OC before sampling (Pipal et al., 2015; Wang et al., 2016; $\mathrm{xu}$ et al., 2015). The filters were weighed by a highly precise balance (Sartorius Corporation, Precision $0.01~\mathrm{mg}$ ), after being stabilized for at least $24\,\mathrm{h}$ under controlled temperature $(21\pm1\mathrm{~}^{\circ}\mathrm{C})$ and relative humidity $(45\pm5\%)$ . After weighed, the filters were frozen at $-20~^{\circ}\mathrm{C}$ until analysis. All procedures were strictly controlled to avoid contaminating the samples. Hourly means of the wind direction frequency and speed are plotted in Fig. 2.
2.2. OC and EC analyses
OC and EC were analyzed by Thermal/Optical Carbon Analyzer (DRI Model 2001, Atmoslytic Inc., USA). A punch of $0.2997~\mathrm{cm}^{2}$ filter was applied for eight carbon fractions analysis following the IMPROVE TOR protocol (Allen et al., 2000; Cao et al., 2003; Cao et al., 2004; Chow and Watson, 2002; Fung et al., 2002). The oven temperature was stepwise heated to 140, 280, 480, and $580\,^{\circ}\mathrm{C}$ , respectively, with helium environment, which produced four OC fractions(OC1, OC2, OC3, and OC4). The oven was further heated to 580, 740, and $840\,^{\circ}\mathrm{C}_{:}$ , respectively, with $2\%$ oxygen/ $\prime98\%$ helium atmosphere, which produced three EC fractions (EC1, EC2, and EC3). During volatilization of OC, part of OC was pyrolyzed to EC called pyrolyzed carbon (OCPyro). OCPyro was determined at the point of oxygen added to the combustion atmosphere, where intensity of transmitted light returned to the initial value. On the other hand, OC was operationally defined as the sum of OC1, OC2, OC3, OC4, and OCPyro, whereas EC was calculated by $\mathsf{E C}1+\mathsf{E C}2+\mathsf{E C}3-$ OCPyro. Methane has been set as internal standard; sucrose was used to establish the calibration curve (Internal Standard Calibration Method). Field blank and replicate analyses were carried out once per 10 samples.
3. Results and discussion
3.1. Distribution of PM
$\mathsf{P M}_{10}$ concentrations ranged at $17.2–681\;|\upmu\mathrm{g}\cdot\mathrm{m}^{-3},$ , with a mean and standard deviation of $176\pm89.0\,\upmu\mathrm{gm}^{-3}$ , which was 4.4 times of that in the Air Quality Standards of the European Union (AQSEU) $(40\,\upmu\mathrm{g}\,\mathfrak{m}^{-3}$ , European Commission department) and 3.5 times of that in National Ambient Air Quality Standards of USA (NAAQSUS)
$(50\,\upmu\mathrm{g}\cdot\upmu^{-3}$ , U.S Environmental Protection Agency). The average concentration of $\mathsf{P M}_{10}$ for each site was 6.3 (DHE), 4.0 (EPB), 3.7 (EMB), 4.0 (QSH), 3.9 (BGH), and 3.8 (STU) times of AQSEU and 5.1 (DHE), 3.2 (EPB), 3.0 (EMB), 3.2 (QSH), 3.1 (BGH), and 3.0 (STU) times of NAAQSUS. $\mathsf{P M}_{2.5}$ concentrations ranged at $8.70{-}457\;\upmu\mathrm{g}\cdot\textrm{m}^{-3}$ with a mean and standard deviation of $89.6\pm55.2\;\upmu\mathrm{g}\cdot\,\mathrm{m}^{-3}$ , that was 6.0 and 3.6 times of that in NAAQSUS ( $15~\upmu\mathrm{g}\cdot\upmu^{-3}$ , U.S Environmental Protection Agency) and AQSEU $(25\,\upmu\mathrm{g}\cdot\mathfrak{m}^{-3}$ , European Commission department), respectively. The average concentration of $\mathsf{P M}_{2.5}$ for each sites were 4.9 (DHE), 3.3 (EPB), 3.0 (EMB), 3.1 (QSH), 3.2 (BGH), and 3.2 (STU) times of AQSEU and 8.1 (DHE), 5.4 (EPB), 5.0 (EMB), 5.2 (QSH), 5.3 (BGH), and 5.4 (STU) times of NAAQSUS (Table 2, Fig. 3). The highest $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ concentrations were at DHE, which may be due to the major resuspension of fugitive dust resulting from the demolition of Beiliang shantytowns (about $13\ \mathrm{km}^{2}\$ ) in the sampling months. In conclusion, no matter at NAAQUS or AQSEU, the $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ concentrations all highly exceeded standards.
The seasonal variations of average concentrations of $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ were counted according to the order of January, April, September, and November, respectively, as $237\pm92.0$ , $176\pm71.2$ , $122\pm67.6$ , and
$151\pm82.2\,\upmu\mathrm{gm}^{-3}$ and $147\pm61.0,80.9\pm27.2,47.3\pm25.7.$ , and $69.2\pm$ $39.4\,\upmu\mathrm{gm}^{-3}$ , respectively. The results showed the mean concentrations of $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ following the order: January $>$ Apri $>$ November $>$ September.
In addition, according to the annual samples results, the $\mathsf{P M}_{2.5}$ accounted for 62, 47, 40, and $45\%$ of $\mathsf{P M}_{10}$ in January, April, September, and November, respectively, while the annual mean $\mathsf{P M}_{2.5}$ mass contributes about $50\%$ of $\mathsf{P M}_{10}$ (Table 2). The ratio of $\mathsf{P M}_{2.5}$ in $\mathsf{P M}_{10}$ indicated that coarse particulate has been the major pollutants for Baotou and it is being replaced by fine particulate gradually.
3.2. Distributions of OC and EC
Generally, the OC and EC concentrations of $\mathsf{P M}_{10}$ at DHE site were 1.8 and 1.6 times of average OC and EC concentration at the other five sites,
1.7 and 1.5 times for $\mathsf{P M}_{2.5}$ , respectively. The concentrations of OC and EC both in $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ at DHE site were much higher than those of other sites, but the mass percentages of them did not show the same feature (Fig. 4), it may be due to high concentration of PM resulting from the demolition of Beiliang. BGH has higher concentrations of OC and EC than EPB, EMB, QSH, and STU sites, which may be due to its close location at the east of Baotou Iron & Steel Group (Fig. 1) with frequent west or southwest wind (Fig. 2) in January and November. Moreover, according to the annual data collected at EPB (Fig. 5), the concentrations of OC and EC were higher in January and November as well, which indicated that the heating season did great contribution to the concentrations of OC and EC, and the concentration for heating seasons is significantly higher $\mathrm{[}\,\mathsf{p}<0.01\,\$ than that in the non-heating season since more coal was combusted for heating in Baotou, even in northern China.
Table 3 compares the carbonaceous levels in Baotou with other cities. It is found that the concentrations of OC and EC in $\mathsf{P M}_{10}$ in Baotou are equal to Guangzhou (summer time), higher than Hong Kong (summer time), Kaohsiung, and Nanjing, while much lower than Santiago. For $\mathsf{P M}_{2.5}$ , the concentrations of OC and EC are similar to Guangzhou (summer time), Nanjing, and Seoul and higher than Hong Kong (summer time), while lower than Beijing, Tianjin, and Agra. Baotou is a heavy industrial city, with 6 months of heavy coal combustion for heating; therefore, OC and EC are higher than most of the cities in compared list, however, lower than some cities in Northern China with high density of motor vehicles.
Concentrations of OC and EC indicated obvious seasonal variation, with higher concentration in January and November, and lower concentration in April and September. The average concentrations of OC in $\mathsf{P M}_{10}$ were $38.5~\pm~21.2$ , $12.1~\pm~4.49$ , $11.5~\pm~5.75$ , and $24.4~\pm$ $13.0\,\upmu\mathrm{g}\cdot\upmu^{-3}$ during sampling months, and in January, the concentration was roughly 3.2, 3.4, and 1.6 times of April, September, and November, respectively. For $\mathsf{P M}_{2.5},$ it was found to be $29.4\pm17.6$ , $7.56\pm2.83$ , $6.92\pm2.85$ , and $18.6\pm10.2\,\upmu\mathrm{g}\cdot\mathrm{m}^{-3}$ , and about 3.9, 4.2, and 1.6 times, respectively. The results showed the mean concentrations of OC in $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ following the order: January $>$ November $>$ April $>$ September. In 2014, the precipitation of Baotou was mainly occurred from July to September, accounting for $68.1\%$ of annual precipitation (Baotou statistical yearbook, 2015). Therefore, the lower contents of OC and EC in September may be due to washout effect, while coal combustion heating and unfavorable meteorological condition including high calm wind frequency and low mixed layer height (Fig. 6) leads to higher concentration in January and November. EC concentrations were found to be $8.63\pm2.88,\,5.04\pm2.14,\,7.02\pm3.09$ , and $10.7~\pm$ $5.49\,\upmu\mathrm{g}\;\mathrm{m}^{-3}$ in $\mathsf{P M}_{10},$ while $6.83\pm2.31,3.51\pm1.18,5.13\pm2.43$ , and $7.94\pm3.74\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ in $\mathsf{P M}_{2.5}$ during sampling months, respectively. However, the average concentrations of EC in $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ in Baotou ranked in the following order: January $>$ November $>$ September $>$ April. Both OC and EC in the heating season are higher than those in nonheating season, which EC1, OC2, OC3, and OC4 were the dominant species with 34.4, 13.0, 19.8, and $18.1\%$ in $\mathsf{P M}_{10}$ , and 28.4, 14.7, 20.2, and $19.9\%$ in $\mathsf{P M}_{2.5}$ , respectively.
It was recommend by Turpin and Lim's study that total carbonaceous aerosol (TCA) is reasonably calculated by using an average organic molecular weight per carbon weight ratio of $1.6\pm0.2$ for urban aerosols (Terzi et al., 2010; Turpin and Lim, 2001); therefore, the amount of organic matter in the city area of Baotou was estimated by multiplying the amount of OC by 1.6 as $\mathrm{TCA}=1.60{\mathsf{C}}+\mathrm{EC}$ (Cao et al., 2003; Gu et al., 2010; Turpin and Huntzicker, 1995). The means of TCA accounted for 29.2, 14.4, 22.5, and $34.3\%$ in $\mathsf{P M}_{10},$ while 35.7, 19.9, 40.3, and $61.1\%$ in $\mathsf{P M}_{2.5}$ during sampling months, respectively, indicating that TCA has enriched in $\mathsf{P M}_{2.5}$ and contributed more in $\mathsf{P M}_{2.5}$ over $\mathsf{P M}_{10}$ (Fig. 4). Furthermore, OC was the major contributor to total carbon (TC), accounting for 81.1, 70.6, 62.3, and $68.9\%$ of TC in $\mathsf{P M}_{10}$ , while those were 80.3, 68.4, 58.0, and $68.9\%$ in $\mathsf{P M}_{2.5}$ during sampling months, respectively.
3.3. Correlations between carbonaceous species and gaseous pollutants
The Scatter plots of OC and EC concentrations and their correlation coefficients $(\mathbb{R}^{2})$ in different sampling periods are depicted in Fig. 7. Significant correlations between OC and EC were observed, with correlation coefficient $(\mathbb{R}^{2})$ of 0.63, 0.51, 0.83, and 0.78 in January, April, September, and November in $\mathsf{P M}_{10}$ , and 0.75, 0.90, 0.86, and 0.92 in $\mathsf{P M}_{2.5}$ , respectively. The different plots by linear regression equations in various sampling periods could be associated with the seasonal variation of emissions and SOC formation. The $\mathbb{R}^{2}$ between OC and EC of $\mathsf{P M}_{10}$ was lower in April $\mathrm{\Delta[R^{2}~=~}0.51$ ) than those in January,
September, and November $\mathit{\Omega}^{\left(0.63<{\mathrm R}^{2}<0.83\right)}$ , indicating that additional sources of OC must be appeared other than EC. OC and EC presented significant correlations in $\mathsf{P M}_{2.5}$ ( $[0.75<\mathsf{R}^{2}<0.92]$ , indicating that they were likely derived from common sources, such as coal combustion and vehicles emissions.
The OC/EC ratios for PM fractions at the six urban sites were collected simultaneously, and their variation can be used to index the origins of the carbonaceous aerosol during the sampling months (Cao et al., 2003; Novakov et al., 2005). The mean and standard deviation of OC/EC ratios in $\mathsf{P M}_{10}$ were $4.36\pm0.87,2.53\pm0.80,1.68\pm0.28,$ and $2.37\pm0.47$ during sampling months, respectively. Similarly, in $\mathsf{P M}_{2.5}$ the mean and standard deviation of OC/EC ratios were $4.14\pm0.87$ , $2.18\pm0.25$ , $1.44\pm0.21$ , and $2.31\pm0.43$ during sampling months, respectively. The results indicated that the SOC appeared in cold months rather than warm months that OC was primary replaced by SOC. The elevated OC/EC ratios in January could be attributed to several reasons.
First, coal consumption for winter heating contributes more OC than EC and also increases the emission of volatile organic precursors (Chen et al., 2014; Li and Bai, 2009). Secondly, low temperature in winter leads to the adsorption and condensation of semi-volatile organic compounds on particles (Zhao et al., 2013). Thirdly, the low mixing layer height in January would enhance the SOC formation (Park et al., 2001; Strader et al., 1999). The same phenomenon of higher winter OC/EC ratio was also observed in Beijing (Dan et al., 2004), Guangzhou, and Hong Kong (Duan et al., 2007).
The Pearson correlation coefficients were applied by using the annual data from EPB. The correlation between carbonaceous species, meteorological conditions, and gaseous pollutants $S0_{2}$ , $\mathsf{N O}_{2}$ , $0_{3}$ , and CO) were summarized in Table 4 and Fig. 8. OC and EC both in $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ were negatively correlated with wind speed, and positively correlated with relative humidity, indicating that most OC and EC were removed by high wind speed and low relative humidity during January and November. The results show a big picture that $S0_{2}$ , $\mathrm{NO}_{2}$ , and CO presented significant positive correlations with both OC and EC either in $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ . Cluster analysis between carbonaceous species and gaseous pollutants indicated that OC shared the same source with $S0_{2}$ and CO, while EC shared the same source with $\mathrm{NO}_{2}$ (Fig. 9). $S0_{2}$ and CO were mainly produced during coal combustion (Mancilla et al., 2015), while $\Nu0_{2}$ were mainly from vehicles exhaust and power plants (John and Seinfeld, 2006; Kim et al., 2011). Therefore, coal combustion is the main source of OC, while vehicles exhaust and coal combustion both contribute to EC. In April, the EC and OC presented poor correlation with all factors in general, only EC in $\mathsf{P M}_{2.5}$ correlated with $\mathsf{N O}_{2}$ at a significant level of $\mathrm{P}=0.01$ . This might be caused by the frequent fugitive dust during the period of time. On the other hand, $0_{3}$ presented significant negative correlations with both OC and EC either in $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ during sampling months. $0_{3}$ is a secondary pollutant and the formation and destruction of $0_{3}$ depend on many factors, such as $\mathrm{NO}_{2}$ , $S0_{2}$ , CO, volatile organic compounds (VOCs), and solar radiation etc (Gao et al., 2015; Gorai et al., 2015; Lou et al., 2015). According to the annual data, the findings indicate that higher concentration of $0_{3}$ accompanied with low gases contents $\phantom{}{\.}\Nu0_{2}$ , $S0_{2}$ , CO) in warmer months (June to September). It may be due to the implication in forming the SOC, and it is formed in the atmosphere by photochemical reactions of volatile organic compounds (VOCs) of biogenic and anthropogenic origin followed by the condensation of reaction products on particles (Chen et al., 2015; Kroll and Seinfeld, 2008; Ling and Guo, 2014; Ou et al., 2015). Furthermore, it may be also caused by the emission of VOCs, strong sunlight, and high temperature in summer (Ghude et al., 2008; Mishra and Goyal, 2016).
3.4. Estimation of secondary organic carbon
OC is emitted from both primary and secondary sources, since EC is essentially emitted from combustion of fossil fuels and biomass. Furthermore, resistant to the chemical reactions, it has often been used to indicate primary anthropogenic pollutants and estimate the SOC concentrations (Turpin and Huntzicker, 1991; Strader et al., 1999; Dan et al., 2004). When compared to the same ratio in PM, the ratio of OC/EC in source emissions will provide an indication of the presence of secondary organic aerosol (SOA) formation. In the EC tracer method (Cabada et al., 2004; Turpin and Huntzicker, 1995), the SOC concentration is estimated by means of the following equation:
$$
S O C=O C_{t o t}-E C\Big(\frac{O C}{E C}\Big)_{\mathrm{{min}}}
$$
where SOC is the secondary organic carbon and $O C_{t o t}$ is the total organic carbon, the primary OC could be calculated from the formula EC (OC/ $\mathrm{EC})_{\mathrm{min}}$ .(Castro et al., 1999). The EC/OC minimum ratio method has a limitation in the fact that they do not account for the variability of primary sources and do not consider non-combustion OC associated with EC, which mainly comes from biogenic origin (Khan et al., 2016). A regression approach has been suggested to overcome this limitation (Turpin and Huntzicker, 1995). The SOC concentration can be also estimated by the following equation:
$$
S O C=O C_{t o t}\!-\!P O C=O C_{t o t}\!-\!(a E C+b)
$$
where $a$ is the primary combustion ratio (Grivas et al., 2012; Khan et al., 2016); $b$ is the primary OC background concentration (Cao et al., 2007; Zhao et al., 2013). Values of $a$ and $b$ are calculated by least-square regression using samples with the lowest $5.20\%$ OC/EC ratios (Cao et al., 2007). In this work, parameters $a$ and $b$ were calculated using samples with the lowest $10\%$ of OC/EC ratios. For the approach, average concentrations of SOC in $\mathsf{P M}_{10}$ samples were 19.6, 3.8, 3.6, and $10.9\,\upmu\mathrm{g}\cdot\mathrm{m}^{-3}$ accounting for 50.8, 32.5, 31.8, and $44.8\%$ of OC, while in $\mathsf{P M}_{2.5}$ , these were found to be 12.7, 2.4, 1.7, and $6.9\,\upmu\mathrm{g}\cdot\upmu^{-3}$ , accounting for 43.2, 33.2, 24.4, and $37.1\%$ of OC during sampling months, respectively.
The variation of SOC and $O C_{t o t}$ presented elevating trends during heating season is due to coal consumption in winter which led to enhance emissions of volatile organic precursors, as well as meteorological conditions that result in SOC precursor stagnation and SOC formation (Cao et al., 2007; Dan et al., 2004; Duan et al., 2005; Park et al., 2001; Strader et al., 1999). The results indicated that the coal combustion could likely be the predominant sources of the organic aerosols in Baotou.
The SOC fraction in $\mathsf{P M}_{2.5}$ was $34\%$ , which was lower than that in previous studies, i.e. $40\%$ by Cao et al (Cao et al., 2007) and $57\%$ by Zhang et al. (Zhang et al., 2008) in urban sites of China. The concentration of OC in Baotou is mainly primary. Numerous studies report higher SOC in summer (Castro et al., 1999; Gu et al., 2010; Khan et al., 2010; Na et al., 2004). However, the results of this work showed down the higher SOC concentration in winter which was also supported by previous studies (Dan et al., 2004; Li and Bai, 2009). There is a common characteristic among the cities with higher SOC in winter, wherever the coal combustion can lead to the severe pollution.
4. Conclusions
This work selected 6 sites covering 3 districts in Baotou, and collected 927 samples in total. 804 samples simultaneously collected on the basis of seasonal impacts in January, April, September, and November. For a long-term study of the effects of carbonaceous aerosol, data were collected annually at EPB site. The concentration of $\mathsf{P M}_{10}$ and $\mathrm{PM}_{2.5},$ the spatial distribution and content of OC and EC, the relationship between OC and EC, and the formation of SOC were investigated. The findings indicated that the average concentrations of OC in $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ following the order: January $>$ November $>$ April $>$ September; For EC in $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ following the order: January $>$ November $>$
September $>$ April. Correlations of OC and EC both in $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ with metrological conditions indicated that high wind speed and low relative humidity were beneficial for the removal of OC and EC in January and November. Pearson correlations and cluster analysis of OC and EC both in $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ with gaseous pollutants ( $S0_{2}$ , $\mathsf{N O}_{2}$ and CO) suggested OC shared the same emission sources with $S0_{2}$ and CO from combustion and EC's sources including vehicles exhaust and combustion both contributed to $\Nu0_{2}$ as well. The OC concentration is mainly primary in warm months, while it appears secondary in cold months in Baotou. There is a common characteristic among the cities with higher SOC in winter, wherever the coal combustion can lead to the severe pollution. This work is important for constructing the database of OC and EC in $\mathsf{P M}_{10}$ and $\mathsf{P M}_{2.5}$ at spatial and time intervals and providing scientific consultation for PM atmospheric pollutant control and air quality improvement in Baotou.
Acknowledgements
This project is supported by the Postgraduate Scientific Research Innovation Foundation of Inner Mongolia (no. 1402020201314).
References
Allen, J.O., et al., 2000. Characterization and evolution of primary and secondary aerosols during PM2. 5 and PM10 episodes in the South Coast Air Basin. Environmental Engineering and Science Department California Institute of Technology, Pasadena, CA. Final report, CRC Project No, A-22.
Baotou Statistical Bureau, 2015. Baotou Statistical Year Book 2015. China Statistics Press, Beijing.
Bollasina, M.A., Ming, Y., Ramaswamy, V., 2011. Anthropogenic aerosols and the weakening of the South Asian summer monsoon. Science 334, 502–505.
Bond, T.C., et al., 2013. Bounding the role of black carbon in the climate system: A scientific assessment. J. Geophys. Res. Atmos. 118, 5380–5552.
Cabada, J.C., et al., 2004. Estimating the secondary organic aerosol contribution to PM2. 5 using the EC tracer method special issue of aerosol science and technology on findings from the fine particulate matter supersites program. Aerosol Sci. Technol. 38, 140–155.
Cao, G., Zhang, X., Zheng, F., 2006. Inventory of black carbon and organic carbon emissions from China. Atmos. Environ. 40, 6516–6527.
Cao, J., et al., 2003. Characteristics of carbonaceous aerosol in Pearl River Delta Region, China during 2001 winter period. Atmos. Environ. 37, 1451–1460.
Cao, J., et al., 2004. Spatial and seasonal variations of atmospheric organic carbon and elemental carbon in Pearl River Delta Region, China. Atmos. Environ. 38, 4447–4456.
Cao, J.J., et al., 2007. Spatial and seasonal distributions of carbonaceous aerosols over China. J. Geophys. Res. Atmos. 112, 88–97.
Castro, L., Pio, C., Harrison, R.M., Smith, D., 1999. Carbonaceous aerosol in urban and rural European atmospheres: estimation of secondary organic carbon concentrations. Atmos. Environ. 33, 2771–2781.
Chen, W., Tang, H., Zhao, H., 2015. Diurnal, weekly, and monthly spatial variations of air pollutants and air quality of Beijing. Atmos. Environ. 119, 21–34.
Chen, Y., Xie, S., Luo, B., Zhai, C., 2014. Characteristics and origins of carbonaceous aerosol in the Sichuan Basin, China. Atmos. Environ. 94, 215–223.
Chow, J.C., Watson, J.G., 2002. PM2. 5 carbonate concentrations at regionally representative Interagency Monitoring of Protected Visual Environment sites. J. Geophys. Res. Atmos. (1984–2012) 107 (pp. ICC 6-1-ICC 6-9).
Daher, N., et al., 2012. Characterization, sources and redox activity of fine and coarse particulate matter in Milan, Italy. Atmos. Environ. 49, 130–141.
Dan, M., Zhuang, G., Li, X., Tao, H., Zhuang, Y., 2004. The characteristics of carbonaceous species and their sources in PM2. 5 in Beijing. Atmos. Environ. 38, 3443–3452.
Dellinger, B., et al., 2001. Role of free radicals in the toxicity of airborne fine particulate matter. Chem. Res. Toxicol. 14, 1371–1377.
Didyk, B.M., Simoneit, B.R., Pezoa, L.A., Riveros, M.L., Flores, A.A., 2000. Urban aerosol particles of Santiago, Chile: organic content and molecular characterization. Atmos. Environ. 34, 1167–1179.
Duan, F., et al., 2005. Characteristics of carbonaceous aerosols in Beijing, China. Chemosphere 60, 355–364.
Duan, J., et al., 2007. Sources and characteristics of carbonaceous aerosol in two largest cities in Pearl River Delta Region, China. Atmos. Environ. 41, 2895–2903.
European Commission department, 2013. Air Quality Standards. http://www.ec.europa. eu/environment/air/quality/standards.htm.
Fung, K., Chow, J.C., Watson, J.G., 2002. Evaluation of OC/EC speciation by thermal manganese dioxide oxidation and the IMPROVE method. J. Air Waste Manage. Assoc. 52, 1333–1341.
Gao, J., et al., 2015. A case study of surface ozone source apportionment during a high concentration episode, under frequent shifting wind conditions over the Yangtze River Delta, China. Sci. Total Environ. 544, 853–863.
Ghude, S.D., et al., 2008. Ozone in ambient air at a tropical megacity, Delhi: characteristics, trends and cumulative ozone exposure indices. J. Atmos. Chem. 60, 237–252.
Gorai, A.K., et al., 2015. An innovative approach for determination of air quality health index. Sci. Total Environ. 533, 495–505.
Grivas, G., Cheristanidis, S., Chaloulakou, A., 2012. Elemental and organic carbon in the urban environment of Athens. Seasonal and diurnal variations and estimates of secondary organic carbon. Sci. Total Environ. 414, 535–545.
Gu, J., et al., 2010. Characterization of atmospheric organic carbon and element carbon of PM2. 5 and PM10 at Tianjin, China. Aerosol Air Qual. Res. 10, 167–176.
Habre, R., et al., 2014. The effects of PM2. 5 and its components from indoor and outdoor sources on cough and wheeze symptoms in asthmatic children. J. Expo. Sci. Environ. Epidemiol. 24, 380–387.
Hansen, J., Sato, M., Ruedy, R., Lacis, A., Oinas, V., 2000. Global warming in the twenty-first century: an alternative scenario. Proc. Natl. Acad. Sci. 97, 9875–9880.
Hansen, J., et al., 2005. Efficacy of climate forcings. J. Geophys. Res. Atmos. (1984–2012) 110.
He, K., et al., 2001. The characteristics of PM 2.5 in Beijing, China. Atmos. Environ. 35, 4959–4970.
John, H., Seinfeld, S.N.P., 2006. Atmospheric Chemistry and Physics: From Air Pollution to Climate Change. second ed. Wiley, New York.
Junker, C., Liousse, C., 2008. A global emission inventory of carbonaceous aerosol from historic records of fossil fuel and biofuel consumption for the period 1860–1997. Atmos. Chem. Phys. 8, 1195–1207.
Khan, M.B., et al., 2016. Carbonaceous PM2.5 and secondary organic aerosol across the Veneto region (NE Italy). Sci. Total Environ. 542 (Part A), 172–181.
Khan, M.F., Shirasuna, Y., Hirano, K., Masunaga, S., 2010. Characterization of PM 2.5, PM 2.5–10 and $\mathrm{PM}>10\$ in ambient air, Yokohama, Japan. Atmos. Res. 96, 159–172.
Kim, K.H., Sekiguchi, K., Kudo, S., Sakamoto, K., 2011. Characteristics of atmospheric elemental carbon(char and soot) in ultrafine and fine particles in a roadside environment, Japan. Aerosol Air Qual. Res. 11, 1–12.
Kroll, J.H., Seinfeld, J.H., 2008. Chemistry of secondary organic aerosol: formation and evolution of low-volatility organics in the atmosphere. Atmos. Environ. 42, 3593–3624.
Kuo, C.-Y., et al., 2013. Analysis of the major factors affecting the visibility degradation in two stations. J. Air Waste Manage. Assoc. 63, 433–441.
Lee, J.Y., Lane, D.A., Heo, J.B., Yi, S.M., Kim, Y.P., 2012. Quantification and seasonal pattern of atmospheric reaction products of gas phase PAHs in PM2.5. Atmos. Environ. 55, 17–25.
Li, B., et al., 2015. Seasonal variation of urban carbonaceous aerosols in a typical city Nanjing in Yangtze River Delta, China. Atmos. Environ. 106, 223–231.
Li, W., Bai, Z., 2009. Characteristics of organic and elemental carbon in atmospheric fine particles in Tianjin, China. Particuology 7, 432–437.
Lin, J.J., Tai, H.-S., 2001. Concentrations and distributions of carbonaceous species in ambient particles in Kaohsiung City, Taiwan. Atmos. Environ. 35, 2627–2636.
Ling, Z.H., Guo, H., 2014. Contribution of VOC sources to photochemical ozone formation and its control policy implication in Hong Kong. Environ. Sci. Pol. 38, 180–191.
Lou, S., Liao, H., Yang, Y., Mu, Q., 2015. Simulation of the interannual variations of tropospheric ozone over China: roles of variations in meteorological parameters and anthropogenic emissions. Atmos. Environ.
Mancilla, Y., Herckes, P., Fraser, M.P., Mendoza, A., 2015. Secondary organic aerosol contributions to PM 2.5 in Monterrey, Mexico: temporal and seasonal variation. Atmos. Res. 153, 348–359.
Mauderly, J.L., Chow, J.C., 2008. Health effects of organic aerosols. Inhal. Toxicol. 20, 257–288.
Menon, S., Hansen, J., Nazarenko, L., Luo, Y., 2002. Climate effects of black carbon aerosols in China and India. Science 297, 2250–2253.
Mishra, D., Goyal, P., 2016. Neuro-Fuzzy approach to forecasting Ozone Episodes over the urban area of Delhi, India. Environ. Technol. Innov. 5, 83–94.
Na, K., Sawant, A.A., Song, C., Cocker, D.R., 2004. Primary and secondary carbonaceous species in the atmosphere of Western Riverside County, California. Atmos. Environ. 38, 1345–1355.
Nelin, T.D., Joseph, A.M., Gorr, M.W., Wold, L.E., 2012. Direct and indirect effects of particulate matter on the cardiovascular system. Toxicol. Lett. 208, 293–299.
Novakov, T., 1984. The role of soot and primary oxidants in atmospheric chemistry ☆. Sci. Total Environ. 36, 1–10.
Novakov, T., Menon, S., Kirchstetter, T., Koch, D., Hansen, J., 2005. Aerosol organic carbon to black carbon ratios: analysis of published data and implications for climate forcing. J. Geophys. Res. Atmos. (1984–2012) 110.
Nunes, T.V., Pio, C.A., 1993. Carbonaceous aerosols in industrial and coastal atmospheres. Atmos. Environ. Part A 27, 1339–1346.
Ou, J., et al., 2015. Speciated OVOC and VOC emission inventories and their implications for reactivity-based ozone control strategy in the Pearl River Delta region, China. Sci. Total Environ. 530, 393–402.
Pachauri, T., Singla, V., Satsangi, A., Lakhani, A., Kumari, K.M., 2013. Characterization of carbonaceous aerosols with special reference to episodic events at Agra, India. Atmos. Res. 128, 98–110.
Park, S.S., Kim, Y.J., Fung, K., 2001. Characteristics of PM 2.5 carbonaceous aerosol in the Sihwa industrial area, Korea. Atmos. Environ. 35, 657–665.
Park, S.S., Kim, Y.J., Fung, K., 2002. PM 2.5 carbon measurements in two urban areas: Seoul and Kwangju, Korea. Atmos. Environ. 36, 1287–1297.
Pipal, A.S., Satsangi, P.G., Satsangi, P.G., 2015. Study of carbonaceous species, morphology and sources of fine (PM2.5) and coarse (PM10) particles along with their climatic nature in India. Atmos. Res. 154, 103–115.
Pope III, C.A., Dockery, D.W., 2006. Health effects of fine particulate air pollution: lines that connect. J. Air Waste Manage. Assoc. 56, 709–742.
Ramanathan, V., et al., 2001. Indian Ocean experiment: an integrated analysis of the climate forcing and effects of the great Indo-Asian haze. J. Geophys. Res. Atmos. (1984–2012) 106, 28371–28398.
Ru-Jin, H., et al., 2014. High secondary aerosol contribution to particulate pollution during haze events in China. Nature 514, 218–222.
Shafer, M.M., et al., 2010. Reactive oxygen species activity and chemical speciation of sizefractionated atmospheric particulate matter from Lahore, Pakistan: an important role for transition metals. J. Environ. Monit. Jem 12, 704–715.
Stone, E.A., et al., 2010. Characterization of emissions from South Asian biofuels and application to source apportionment of carbonaceous aerosol in the Himalayas. J. Geophys. Res. Atmos. 115, 620–631.
Strader, R., Lurmann, F., Pandis, S.N., 1999. Evaluation of secondary organic aerosol formation in winter. Atmos. Environ. 33, 4849–4863.
Subhasis, B., et al., 2009. Oxidative potential of semi-volatile and non volatile particulate matter (PM) from heavy-duty vehicles retrofitted with emission control technologies. Environ. Sci. Technol. 43, 3905–3912.
Terzi, E., et al., 2010. Chemical composition and mass closure of ambient PM10 at urban sites. Atmos. Environ. 44, 2231–2239.
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. A. Gen. Top. 25, 207–215.
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, 3527–3544.
Turpin, B.J., Lim, H.-J., 2001. Species contributions to PM2.5 mass concentrations: revisiting common assumptions for estimating organic mass. Aerosol Sci. Technol. 35, 602–610.
U.S Environmental Protection Agency, 1997. National ambient air quality standards. http://www3.epa.gov/ttn/naaqs/standards/pm/s_pm_history.html.
U.S Environmental Protection Agency, 2012. National ambient air quality standards. http://www3.epa.gov/ttn/naaqs/criteria.html.
Van Ryswyk, K., et al., 2014. Impact of microenvironments and personal activities on personal PM2. 5 exposures among asthmatic children. J. Expo. Sci. Environ. Epidemiol. 24, 260–268.
Vedal, S., 1997. Ambient particles and health: lines that divide. J. Air Waste Manage. Assoc. 47, 551–581.
Wang, F., Guo, Z., Lin, T., Rose, N.L., 2016. Seasonal variation of carbonaceous pollutants in PM2.5 at an urban ‘supersite’ in Shanghai, China. Chemosphere 146, 238–244.
Xia, T., Nel, A., 2004. Quinones and aromatic chemical compounds in particulate matter induce mitochondrial dysfunction: implications for ultrafine particle toxicity. Environ. Health Perspect. 112, 1347–1358.
Xu, Z., Wen, T., Li, X., Wang, J., Wang, Y., 2015. Characteristics of carbonaceous aerosols in Beijing based on two-year observation. Atmos. Pollut. Res. 6, 202–208.
Zhang, X., Wang, Y., Zhang, X., Guo, W., Gong, S., 2008. Carbonaceous aerosol composition over various regions of China during 2006. J. Geophys. Res. Atmos. (1984–2012) 113.
Zhao, P., et al., 2013. Characteristics of carbonaceous aerosol in the region of Beijing, Tianjin, and Hebei, China. Atmos. Environ. 71, 389–398.
|
atmosphere
|
0105
|
10.1016/j.atmosenv.2016.03.054
|
image
|
Fig. 1. Location of four sampling sites in China.
|
|
atmosphere
|
0105
|
10.1016/j.atmosenv.2016.03.054
|
table
|
Table 1 Summary of water-soluble ions, OC, EC in $\mathrm{PM}_{2.5}$ at four cities of China.
|
|
atmosphere
|
0105
|
10.1016/j.atmosenv.2016.03.054
|
image
|
Fig. 2. Seasonal variation of $\mathrm{PM}_{2.5}$ and major constituents at four sits of China.
|
|
atmosphere
|
0105
|
10.1016/j.atmosenv.2016.03.054
|
image
|
Fig. 3. Seasonal variation of OC and EC in $\mathrm{PM}_{2.5}$ at four sits of China. Wuqing (WQ), Haining (HN), Zhongshan(ZS), Deyang(DY).
|
|
atmosphere
|
0105
|
10.1016/j.atmosenv.2016.03.054
|
table
|
Table 2 Comparison of PM2.5, water-soluble ions and carbonaceous species at different sites over the world.
|
|
atmosphere
|
0105
|
10.1016/j.atmosenv.2016.03.054
|
image
|
Fig. 4. Correlations between anion equivalents and cation equivalents at four sampling sites.
|
|
atmosphere
|
0105
|
10.1016/j.atmosenv.2016.03.054
|
image
|
Fig. 5. Relationship between $C1^{-}/K^{+}$ and $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$
|
|
atmosphere
|
0105
|
10.1016/j.atmosenv.2016.03.054
|
image
|
Fig. 6. The 3-day back trajectories of air masses arriving at sampling sites in Wuqing (a), Haining (b), Zhongshan (c), Deyang (d) during summertime.
|
|
atmosphere
|
0105
|
10.1016/j.atmosenv.2016.03.054
|
image
|
Fig. 7. The 3-day back trajectories of air masses arriving at sampling sites in Wuqing (a), Haining (b), Zhongshan (c), Deyang (d) during wintertime.
|
|
atmosphere
|
0105
|
10.1016/j.atmosenv.2016.03.054
|
table
|
Table 3 Factor loading from PCA in $\mathrm{PM}_{2.5}$ at four sites of China.
|
|
atmosphere
|
0105
|
10.1016/j.atmosenv.2016.03.054
|
text
| Not supported with pagination yet |
Characterizing and sourcing ambient PM2.5 over key emission regions in China I: Water-soluble ions and carbonaceous fractions
Jiabin Zhou a, Zhenyu Xing a, Junjun Deng b, Ke Du a,
a Department of Mechanical and Manufacturing Engineering, University of Calgary, Alberta, T2N 1N4, Canada b Center for Excellence in Urban Atmospheric Environment, Institute of Urban Environment, Chinese Academy of Sciences, Xiamen, 361021, China
h i g h l i g h t s
Seasonal variation of ion and OC/EC in PM2.5 over 4 regional sites was characterized.
The sites were affected by local emissions and air masses via long-range transport.
Main sources are from vehicular emission, coal/biomass combustion industry source.
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 17 September 2015
Received in revised form
20 March 2016
Accepted 29 March 2016
Available online 30 March 2016
Keywords:
$\mathrm{PM}_{2.5}$
Air quality
Source apportionment
Aerosol
China
During the past decade, huge research resources have been devoted into studies of air pollution in China, which generated abundant datasets on emissions and pollution characterization. Due to the complex nature of air pollution as well as the limitations of each individual investigating approach, the published results were sometimes perplexing and even contradicting. This research adopted a multi-method approach to investigate region-specific air pollution characteristics and sources in China, results obtained using different analytical and receptor modeling methods were inter-compared for validation and interpretation. A year-round campaign was completed for comprehensive characterization of PM2.5 over four key emission regions: Beijing-Tianjin-Hebei (BTH), Yangzi River Delta (YRD), Pearl River Delta (PRD), and Sichuan Basin (SB). Atmospheric $\mathrm{PM}_{2.5}$ samples were collected from 10/2012 to 08/2013 at four regional sites, located on the diffusion paths of air masses from their corresponding megacities (i.e., Beijing, Shanghai, Guangzhou, and Chengdu). The annual average $\mathsf{P M}_{2.5}$ mass concentrations showed distinct regional difference, with the highest observed at BTH and lowest at PRD site. Nine water-soluble ions together contributed $33\!-\!41\%$ of $\mathsf{P M}_{2.5}$ mass, with three dominant ionic species being $S0_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ , $\mathsf{N H}_{4}^{+}$ , and carbonaceous particulate matter contributed $16{-}23\%$ of $\mathrm{PM}_{2.5}$ mass. This implied that combustion and secondary formation were the main sources for PM2.5 in China. In addition, $S0_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ , $\mathrm{{NH}\ddagger}$ , and carbonaceous components (OC, EC) showed clear seasonal patterns with the highest concentration occurring in winter while the lowest in summer. Principal component analysis performed on aerosol data revealed that vehicular emissions, coal/biomass combustion, industry source, soil dust as well as secondary formation were the main potential sources for the ionic components of $\mathsf{P M}_{2.5}$ . The characteristic chemical species combined with back trajectory analysis indicated that BTH was heavily influenced by air masses originating from Mongolia and North China Plain regions, whereas SB suffered from both local emissions of Sichuan Basin and biomass burning via long-range transport from South Asia. Sourcing conclusions from this study will be compared, validated and interpreted with those obtained using organic molecular marker and carbon isotope analyses to be presented parts II and III of this series.
$\circledcirc$ 2016 Elsevier Ltd. All rights reserved.
1. Introduction
Atmospheric aerosols consist of a mixture of suspended solid and liquid particles ranging over a few orders of magnitudes in sizes from natural or anthropogenic origin (Seinfeld and Pandis,
2006). The chemical composition of aerosol particles depends on multiple factors including primary emission source as well as postformation processes. In general, water-soluble ions and carbonaceous components comprise a significant fraction of the ambient aerosol mass and influence the climate, visibility and human health through aerosol direct or indirect effects (Jacobson et al., 2000; Putaud et al., 2004; Raizenne et al., 1996).
The atmosphere in China has been increasingly contaminated by anthropogenic gases and aerosols due to fast industrialization and urbanization in the past decades. Recently, most of megacities in China suffered haze pollution with severe atmospheric visibility degradation, which was mainly due to light extinction effects of fine aerosols in the atmosphere. Previous studies revealed that Beijing-Tianjin-Hebei area (BTH), Yangtze River Delta area (YRD), Pearl River Delta area (PRD) and Sichuan Basin area (SB) are the four major black carbon (BC) emission areas in China (Cao et al., 2006; Xing et al., 2014). Black carbon contributes to visibility impairment by efficiently absorbing light. In contrast, water-soluble inorganic ions such as $\mathrm{SO}_{4}^{2-}$ and $\mathrm{NO}_{3}^{-}$ were generally thought as major chemical species that contribute to the visibility reduction by light scattering effect. Although a number of studies concerning inorganic ions in aerosol have been conducted in China (He et al., 2001; Hu et al., 2002; Wang et al., 2002, 2006; Shen et al., 2009; Pathak et al., 2009; Tan et al., 2009; Cao et al., 2012), they mainly focused on individual large cities or seasons. Due to the complexity of air pollution in nature and limitations of each individual analytical approach, the conclusions are sometimes perplexing and even contradicting (Zhang et al., 2013). Therefore, characterizing aerosol chemical compositions at large spatial scale in China using multiple analytical techniques are necessary to evaluate aerosols’ effects on atmospheric visibility, human health, and climate change. In this study, we conducted a year-round sampling campaign at four key emission regions and investigated the effects of sources and long range transportation by analyzing the ionic component, OC/EC, organic molecular markers, and carbon isotope. In this paper, we focused on seasonal characterization of PM2.5 ionic and carbonaceous components at four regional sites that have large pollution footprint to investigate the impact of potential sources and long range transportation on regional air quality.
The four sites were located in four satellite cities of their corresponding megacities Beijing, Shanghai, Guangzhou, Chengdu, and approximate $100~\mathrm{km}$ away from the megacities. The sites are within major black carbon (BC) emission areas in China. Daily samples were collected continuously for one week during each season of a year. The meteorological data over the last few years indicated that the chosen sites locate on the dispersion path of air mass from their corresponding megacities. Therefore, $\mathsf{P M}_{2.5}$ collected at these sites could represent overall aged aerosol from corresponding megacities and are free from individual pollution plume. The objectives of this work are to investigate the seasonal variations of water-soluble ions and carbonaceous components in $\mathsf{P M}_{2.5}$ at four sites of China and reveal their potential sources and transport processes. The results presented here would shed light on some major factors determining the distributions of major chemical components in $\mathsf{P M}_{2.5}$ aerosol which is useful for making effective strategies for regional air quality improvement in China.
2. Experiments and methods
2.1. Sample collection
Atmospheric $\mathsf{P M}_{2.5}$ samples were collected at four sites located in Wuqing $39^{\circ}39^{\prime}\mathsf{N}$ , $117^{\circ}39^{\prime}\mathrm{E})$ ), Haining $30^{\circ}51^{\prime}\mathrm{N}$ , $120^{\circ}69^{\prime}\mathrm{E})$ , Zhongshan $22^{\circ}56^{\prime}\mathsf{N}$ , $113^{\circ}33^{\prime}\mathrm{E}^{\prime}$ , Deyang $.31^{\circ}10^{\prime}\mathsf{N}$ , $104^{\circ}39^{\prime}\mathrm{E})$ in China during the period in November 2012, January 2013, April
2013 and July 2013 (Fig. 1). Daily aerosol $\mathsf{P M}_{2.5}$ samples were collected on quartz fiber filters ( $\Phi90\,\mathrm{mm}$ , Millipore) at a flow rate of $100\,\mathrm{L/min}$ with approximate $24\,\mathrm{h}$ on the roofs of the buildings with the height of about $10{-}30\;\mathrm{m}$ using a medium volume air sampler (TH-150C III, Tianhong, Wuhan, China) with a $\mathsf{P M}_{2.5}$ impactor (THPM2.5-100, Tianhong, Wuhan, China). After sampling, the quartz filters were immediately placed in pre-baked aluminum foil and then stored frozen at $-18~^{\circ}\mathsf{C}$ until analyzed. The location of sampling sites is shown in Fig.1. Detailed sampling information of these sites has been described in our previous paper (Xing et al., 2014).
2.2. Measurement of water-soluble ions and carbonaceous components
Blank and sample filters were weighed on an electronic balance (Sartorius-BT125D, Germany) with a reading precision of $0.01~\mathrm{mg}$ to determine the aerosol mass concentration after its having been equilibrated at $25\pm0.5\,^{\circ}\mathrm{C}$ and $40\pm5\%$ relative humidity for $24\,\mathrm{h}$ . The collected aerosol $\mathsf{P M}_{2.5}$ filter was ultrasonically extracted with $20~\mathrm{mL}$ Milli-Q water $^{18.2}\mathrm{~M}\Omega\mathrm{cm})$ for $40\ \mathrm{min}$ . Then the extracted solution was filtered through $0.45~{\upmu\mathrm{m}}$ PTFE syringe filter (Pall Co. Ltd, USA). Analysis of extracted solutions was performed with an ion chromatograph (Metrohm, Switzerland) equipped with a conductivity detector. The concentrations of water-soluble components including five cations $(\mathrm{NH}_{4}^{+},\,\mathrm{Na}^{+},\,\mathrm{K}^{+},\,\mathrm{Ca}^{2+},\,\mathrm{Mg}^{2+})$ and four anions $\mathrm{(F^{-},}$ $C1^{-}$ , $\mathrm{NO}_{3}^{-}$ , $S0_{4}^{2-}$ ) were determined in this study. The concentrations of organic carbon (OC) and elemental carbon (EC) were measured using a Sunset Carbon Aerosol analyzer (Sunset Laboratory Inc., USA), following the NIOSH TOT protocol and assuming carbonate carbon in the sample to be negligible. Typically, a $1.5\,\mathrm{cm}^{2}$ punch of the filter was placed in a quartz boat inside the thermal desorption chamber of the analyzer, and then stepwise heating was applied (Schauer and Cass, 2000). Quality assurance and quality control tests including field blank, laboratory blank, method detection limit, and recovery efficiency were conducted. Multi-point calibrations using standard solutions were carried out and the result of correlation coefficient was higher than 0.999. All the reported ion concentrations have been corrected using field blanks.
2.3. Back trajectory analysis
In order to characterize the origin and transport pathway of the air masses to the sampling sites, back trajectory simulation was performed by using Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model developed by NOAA/ARL (Draxler and Hess, 1998). $^{72\mathrm{~h~}}$ back trajectories were calculated using the HYSPLIT model with an endpoint height of $500\,\mathrm{m},1000\,\mathrm{m}$ and $1500\;\mathrm{m}$ , respectively. The meteorological data used for back trajectory simulation are six-hourly archived data from GDAS (Global Data Assimilation System) of NCEP (National Centers for Environmental Prediction).
3. Results and discussions
3.1. Characterization of $P M_{2.5}$ and major components
Table 1 presented a summary of statistics of water-soluble ions as well as EC, OC in $\mathsf{P M}_{2.5}$ over the sampling period. The annual average $\mathsf{P M}_{2.5}$ mass concentrations at Wuqing, Haining, Zhongshan, Deyang sampling sites were $148.9\pm91.1~\upmu\mathrm{g}/\mathrm{m}^{3}$ , $109.6\pm59.4\,\mathrm{{\bar{\upmu}g/m^{3}}}$ , $60.5\pm46.5~\mathrm{\bar{{\upmu}g/m^{3}}}$ , and $121.5\pm101.1~\upmu\mathrm{g}/\mathrm{m}^{3}$ , respectively. The $\mathsf{P M}_{2.5}$ mass concentration in this campaign is comparable to those measured in other Chinese cities (Wang et al., 2005; Tao et al., 2014; Zhang et al., 2011; Hu et al., 2014) and India site (Verma et al., 2010), however its value is much higher than the new National Ambient Air Quality Standards of China for annual $\mathsf{P M}_{2.5}$ $\langle35\;\upmu\mathrm{g}/\mathrm{m}^{3}\rangle$ . As can be seen from the Table 1, sulfate was the most abundant watersoluble species. The concentrations of individual ions were in the order of $\mathrm{\bar{SO_{4}^{2-}}>N O_{3}^{-}>N H_{4}^{+}>K^{+}>N a^{+}>M g^{2+}}$ respectively. On average, nine water-soluble ions together contributed $33\!-\!41\%$ of $\mathsf{P M}_{2.5}$ mass. Meanwhile, carbonaceous matter was also found as major component in $\mathsf{P M}_{2.5}$ aerosol. The average total carbon $(0C+\mathbb{E}C)$ accounted for $10-15\%$ of $\mathsf{P M}_{2.5}$ mass during the campaign period. Assuming that the ratio of organic matter (OM) to OC was 1.6 (Turpin and Lim, 2001), the total carbonaceous particle matter $(0\mathsf{M}\,+\,\mathsf{E C})$ can be estimated and it approximately comprised $16{-}23\%$ of $\mathsf{P M}_{2.5}$ mass. It is worthy to note that the OM in this study might be underestimated especially during the warmer seasons when the fractions of secondary organic components and of biological material are very high (Hueglin et al., 2005). The $\mathsf{P M}_{2.5}$ aerosol at Wuqing site is characterized by higher water-soluble inorganic ions and lower OM composition. In contrast, fine aerosol at Zhongshan site showed higher relative abundance of OM particles whereas lower water-soluble ions.
3.2. Seasonal variation of water-soluble ions
The seasonal variation of $\mathsf{P M}_{2.5}$ and major constituents at four sites of China are shown in Fig. 2. The annual average $S0_{4}^{2-}$ concentration ranged from 9.8 to $24.2~\upmu\mathrm{g}/\mathrm{m}^{3}$ at four sites, accounting for $38{-}49\%$ of the total mass of water-soluble ions. Specifically, Wuqing site suffered from the highest loading of water-soluble ions, while Zhongshan site recorded the lowest. The sulfate aerosols are mainly produced by chemical reactions of gaseous precursors (i.e. $S0_{2}$ gas from anthropogenic sources, dimethyl sulfide from oceans), which occur either in the gas phase with the OH radical or in cloud droplets with hydrogen peroxide $\left({\mathrm{H}}_{2}0_{2}\right)$ or ozone (Pandis et al., 1990). The seasonal variation of sulfate was obvious and slightly different temporal trend was found among these sites. In general, sulfate showed the highest concentration in winter. Such high level of $\mathrm{SO}_{4}^{2-}$ in winter is likely due to anthropogenic emissions (i.e. coal combustion) coupled with poor dispersion condition during cold season (He et al., 2001). It is noticed that summer samples at Wuqing and Deyang showed an elevated level of sulfate with respect to fall. As secondary ion in fine aerosol, $\mathrm{SO}_{4}^{2-}$ might be caused by enhanced photochemical reaction and aqueous processing at high temperature, intensive solar radiation and high relative humidity.
The annual concentration of $\mathrm{NO}_{3}^{-}$ varied from 6.4 to $19.6\;\upmu\mathrm{g}/\mathrm{m}^{3}$ and it contributed to $23-33\%$ of total mass of water-soluble ions. The seasonal variations of $\mathtt{N O}_{3}^{-}$ are characterized by winter/spring maxima and summer minima at these sites. These findings were consistent with those reported in other Chinese cities (Duan et al., 2006; Hu et al., 2008; Cao et al., 2012). The nitrate aerosol is formed through heterogeneous reactions of nitrogen radicals such as NOx, and $\mathrm{HNO}_{3}$ on aerosol surfaces (Bauer et al., 2007), depending mainly on the thermodynamic state of its precursor and the environmental conditions such as atmospheric temperature and relative humidity (Lin et al., 2010). The highest values of nitrate measured in winter might be ascribed to local pollution sources, such as vehicular traffic (Paraskevopoulou et al., 2015). Meanwhile, the low temperature in winter and spring is beneficial to the formation of nitrate aerosol.
The annual concentration of $\mathsf{N H}_{4}^{+}$ varied from 2.8 to $8.5~\upmu\mathrm{g}/\mathrm{m}^{3}$ , and it showed a strong seasonal cycle with peak values in winter/ spring and relatively low in summer. $\mathsf{N H}_{4}^{+}$ was observed as the most abundant cation in this study and contributed to $13{-}15\%$ of total mass of water-soluble ions. Ammonium $\left(\mathsf{N H}_{4}^{+}\right)$ in aerosol is produced from the reaction between $\mathsf{N H}_{3}$ and acidic species present in either the gas or aerosol phase. In winter, lower temperature and higher acid species such as sulfate and nitrate will favor the gasparticle reactions of $\mathsf{N H}_{3}$ to $\mathsf{N H}_{4}^{+}$ (Pathak et al., 2009; Meng et al., 2014). Compared to the other sites over the world, the $S0_{4}^{2-}$ , $\Nu0_{3}^{-}$ , $\mathrm{{NH}\ddagger}$ concentrations observed in this study were higher than Athens, Greece (Paraskevopoulou et al., 2015) and Seoul, Korea (Shon et al., 2013) except Raipur, India (Verma et al., 2010).
3.3. Seasonal variation of carbonaceous matter
$1.6\ \upmu\mathrm{g}/\mathrm{m}^{3}$ , $1.4~{\upmu\mathrm{g}}/{\mathrm{m}}^{3}$ , $1.2\ \upmu\mathrm{g}/\mathrm{m}^{3}$ and $1.4~{\upmu\mathrm{g}}/{\mathrm{m}}^{3}$ for sampling sites at Wuqing, Haining, Zhongshan and Deyang, respectively (Fig. 3). EC concentration exhibited least seasonal variability at these sites, suggesting a fairly uniform local source, i.e. primary particles from fossil fuel incomplete combustion. The OC concentration varied between 9.0 and $14.1~\upmu\mathrm{g}/\mathrm{m}^{3}$ at these sampling sites. The OC mass concentrations at Wuqing and Deyang sites are around $80\%$ higher than those measured at Haining and Zhongshan sites. Furthermore, OC concentration showed strong seasonal variation with highest in winter, followed by fall and spring, and lowest in summer.
OC/EC ratios during this campaign period were found to be 3.0e27.9 with a mean of 5.8e8.9. These ratios are higher than those reported for the megacities in China (Yao et al., 2002; Cao et al., 2003; Ho et al., 2003). The high OC/EC ratios at these sites suggest that OC is largely produced by photochemical processes during long-range transport. Furthermore, secondary organic carbon (SOC) concentration was estimated with the EC-tracer method based on the following equation (Turpin and Huntzicker, 1995),
1999; Zhou et al., 2012). In this study, we selected data with OC/EC ratio in the lowest $10\%$ to determine the primary OC/EC ratio, assuming that the primary OC/EC ratio can be determined from the OC/EC ratios observed during the periods in which conditions are highly unfavorable for the production of SOC. Primary OC/EC ratio of 2.2 was determined by least-square regression as the threshold value to estimate the concentration of SOC.
It is worthy to note that OC/EC ratios at Haining and Zhongshan sites observed highest in summer are likely due to the local photochemical production of secondary OC during warm season. Actually, the contribution of SOC to the total OC ranged from 42.5 to $74.3\%$ with an average of $62.5\%$ , indicating that more SOA were formed through photochemical reaction during summertime at these sites. Interestingly, the observations of increasing ratios of OC/EC (14.2e15.5) at Deyang and Wuqing sites during wintertime might be attributed to additional input of primary OC from different sources which enhance the OC value on transport pathway to these sites (Aggarwal and Kawamura, 2009). In fact, the high percentage of SOC/OC $(61.2{-}78.9\%)$ observed during wintertime was probably due to the biomass burning emissions because this source has much higher primary OC/EC ratio and can cause overestimation of SOC.
Table 2 listed the concentrations of PM2.5, water-soluble ions and carbonaceous species at different sites over the world. In this study, the concentrations of OC and EC are comparable to some urban sites in China, e.g. Chengdu (Tao et al., 2014), and much higher than Athens, Greece (Paraskevopoulou et al., 2015), however apparently lower than tropical pasture site in Rond^onia, Brazil (Kundu et al., 2010).
3.4. Aerosol chemistry and ions balance
The molar ratios of cation equivalents (CE) and anion equivalents (AE) are frequently employed to infer the acidity of aerosol (Chow et al.,1994; Hennigan et al., 2015). At this study, ions balance was evaluated using the following equations.
$$
\mathrm{SOC}=\mathrm{OC}_{\mathrm{tot}}-\mathrm{EC}\times\mathrm{(OC/EC)}_{\mathrm{pri}}
$$
where $0C_{\mathrm{{tot}}}$ is the total OC, and $(\mathrm{OC/EC)_{pri}}$ is the primary OC/EC ratio.
$$
C E=\frac{[N a^{+}]}{23}+\frac{\Big[N H_{4}^{+}\Big]}{18}+\frac{[K^{+}]}{39}+\frac{[C a^{2+}]}{20}+\frac{[M g^{2+}]}{12}
$$
It should be noted that some uncertainty in the estimation method since primary OC/EC emission ratio vary between sources and it is usually influenced by meteorology condition (Strader et al.,
$$
A E=\frac{\left[S O_{4}^{2-}\right]}{48}+\frac{\left[N O_{3}^{-}\right]}{62}+\frac{\left[C l^{-}\right]}{35.5}+\frac{\left[F^{-}\right]}{19}
$$
As shown in Fig. 4, both anions and cations are strongly correlated and the ratios of AE/CE were slightly higher than one, indicating the $\mathsf{P M}_{2.5}$ aerosols at four sites were characterized by acidic in nature. It means the neutralization of sulfate and nitrate in aerosol was not completely achieved at these sites. It is known that among the ammonium-associated compounds, $(\mathrm{NH}_{4})_{2}\mathrm{SO}_{4}$ is preferentially formed due to its least volatility, while $\mathsf{N H}_{4}\mathsf{N O}_{3}$ is relatively volatile, and $\mathsf{N H}_{4}\mathsf{C l}$ is the most volatile. Furthermore, the non-sea-salt fractions of sulfate $(\mathrm{nss-SO_{4}^{2-}})$ , potassium $(\mathfrak{n s s-K}^{+\cdot}$ and calcium (nss- $\cdot{\bf C}{\bf a}^{2+}.$ ) were estimated with the following equations using measured sodium concentrations and the known ratios of each ion to $\mathtt{N a}^{+}$ in the bulk sea water, assuming the soluble $\mathsf{N a}^{+}$ in aerosol particulate comes only from sea-salts and sea-salts has the same chemical composition as the sea water (Morales et al., 1998; Harrison and van Grieken, 1998).
$$
\mathrm{nss-SO_{4}^{2-}=\left[S O_{4}^{2-}\right]-\left[N a^{+}\right]\times0.2455}
$$
$$
\mathsf{n s s-K^{+}}=\left[\mathsf{K^{+}}\right]-\left[\mathsf{N a^{+}}\right]\times0.0355
$$
$$
\mathsf{n s s-C a^{2+}}=\left[\mathsf{C a}^{2+}\right]-\left[\mathsf{N a}^{+}\right]\times0.0373
$$
between nss- $\cdot S0_{4}^{2-}$ and $\mathrm{NH}_{4}^{+}$ $\mathrm{~\AA~}^{\prime}{\bf R}=0.92{-}0.98)$ at a very significant level $\langle\mathsf{P}<0.001\rangle$ ) revealed the sulfate aerosol primarily existed in the form of $(\mathsf{N H}_{4})_{2}\mathsf{S O}_{4}$ in the atmosphere at these sites.
Previous studies showed that the mass ratio of nitrate/sulfate is generally used to evaluate the relative contribution of mobile and stationary sources in the atmosphere (Wang et al., 2005). The average mass ratios of $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ were 0.43, 0.57, 0.83, 0.93 for Deyang, Zhongshan, Wuqing and Haining, respectively, suggesting that the stationary source emissions were predominant at these sites (Khoder and Hassan, 2008). Comparatively among the sites, Deyang has been influenced by more emissions from coal burning, whereas Haining received more loadings from vehicle emissions. Chloride $(\mathbf{Cl}^{-})$ in coarse aerosol particles is mainly from marine source, and it might also accumulate in fine particulate originating from anthropogenic emissions such as coal combustion, biomass burning. Potassium $(\mathsf{K}^{+})$ in fine mode aerosols is generally served as a diagnostic tracer for biomass burning source (Andreae et al., 1998). The ratios of $C1^{-}/K^{+}$ at this study are relatively higher in summer with large variation, and lower in fall and winter (Fig. 5). Since $K^{+}$ is steady, it remains in particulate phase once emitted from fire source, thus its accumulation in the aerosol phase decrease the ratio of $C1^{-}/K^{+}$ in cold season. The emission of biomass burning in cold season and regional long-range transport in summer are likely most responsible for this observation (Yin et al., 2014). It is worth noting that the highest ratios of $C1^{-}/K^{+}$ were found at Wuqing site during summer might be associated with coal combustion (McCulloch et al., 1999). Similarly, ratios of $\mathbf{Cl}^{-}/\mathbf{K}^{+}$ ranging from 2.8 to 6.5 combined with ratios of $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ ranging between 0.1 and 0.2 confirmed the coal combustion emissions at Haining site during summer period. Moreover, significant increase in concentrations of $\mathsf{P M}_{2.5}$ $1_{2.5}\:(1\bar{1}6.4\substack{-135.4\;\upmu\mathrm{g/m^{3}}})$ coupled with high $\mathsf{C a}^{2+}\,(0.7\mathrm{-}1.2~\upmu\mathrm{g/m}^{3})$ were observed at Wuqing and Deyang sites during spring, implying these sites were affected by local emissions and continental dust through long-range transport.
3.5. Influence of air masses transport
$^{72\mathrm{~h~}}$ back trajectories were calculated by HYSPLIT model to get insight into the origin and transport pathway of the air masses arriving at these sites. As illustrated in Fig. 6, the majority of the air masses during summertime reaching these sites were from western Pacific Ocean passing over Yangtze River Delta (YRD), Pearl River Delta (PRD) regions, whereas at Deyang site the air mass originated from southern mainland coupling with local Sichuan Basin (SB) emissions and long-range transport from South Asia. This is confirmed by the high concentration of $\mathrm{Na}^{+}\,(0.2{-}0.3\;\upmu\mathrm{g/m}^{3})$ in marine air masses, while lower $\mathtt{N a}^{+}$ values observed at Deyang on July 14, 2013.
Fig. 7 showed that the dust originating in the Mongolia moved southeastward and mixed with air masses passing over North China Plain (Hebei province and Shanxi province), and finally arrived at Wuqing site during wintertime. The aerosol chemical data on January 18, 2013 is characterized by high level of ${\mathsf{C}}{\mathsf{a}}^{2+}$ $(1.3\,\upmu\mathrm{g}/\mathrm{m}^{3})$ and $\mathrm{\overline{{Mg^{2+}}}}(0.2\,\upmu\mathrm{g}/\mathrm{m}^{3})$ , together with high concentration of $\mathrm{NO}_{3}^{-}$ $(25.2\,\upmu\mathrm{g}/\mathrm{m}^{3})$ , $50_{4}^{2-}(40.9\,\upmu\mathrm{g}/\mathrm{m}^{3})$ , OC $(27.3\,\upmu\mathrm{g/m}^{3})$ ), $\ K^{+}\left(7.2\,\upmu\mathrm{g}/\right.$ ${\mathfrak{m}}^{3}.$ ), suggesting the polluted air masses on that day was mixed with fossil fuel combustion and continental dust via long-range transport. We noted that high correlation of $\mathsf{K}^{+}$ with $\mathrm{S}0_{4}^{\overline{{2}}-}$ $(\boldsymbol{\mathrm{R}}^{\bar{2}}=0.90)$ during winter, hence $\mathsf{K}^{+}$ may exist as $\mathrm{K}_{2}\mathrm{SO}_{4}$ which is often present in aged smoke especially when the aerosol in this campaign is acidic and ammonium-deficient (Li et al., 2003).
Meanwhile, air masses arrived at Deyang site on January 19, 2013 carrying high level of $\mathrm{SO_{4}^{2-}\,(46.0\;\mu g/m^{3}),N O_{3}^{-}\,(28.4\;\mu g/m^{3}),K^{+}}$ $(7.0\,\upmu\mathrm{g}/\mathrm{m}^{3})$ . It indicated that the aerosol was under the influence of fossil fuel combustion and biomass burning from both the lower part of atmospheric boundary layer (ABL) in Central China and long-range transport from South Asia (Fig. 7). These results together suggest that both long-range transport and local sources have important impact on ions and carbonaceous particles in the aerosol at these sites.
It is worth noting that a haze event in eastern China was recorded in January 2013. The overall higher $\mathrm{SO}_{4}^{2-}/\mathrm{PM}_{2.5}$ ratios for Wuqing (0.18) compared with Haining and Zhongshan site during the winter sampling campaign indicate relatively higher coal combustion emission in NCP region. In contrast, $\mathsf{N O}_{3}^{-}/\mathsf{P M}_{2.5}$ ratios at Haining and Zhongshan sites (0.13e0.15) in winter have been found to be significantly higher than in Wuqing, suggesting a relatively greater contribution from traffic emissions. Interestingly, the ratios of $K^{+}/\mathsf{P M}_{2.5}$ (0.02e0.03) varied little during winter among the four sites, which imply that these haze events were equally affected by biomass combustion in all four regions. The present results are in general agreement with the findings of Andersson et al. during the same study period which revealed that high coal contributions to EC in NCP region and more liquid fossil in YRD and PRD region using dual carbon isotope constrained ( $\Delta^{14}C$ and $\S^{13}\C u$ source apportionment method (Andersson et al., 2015).
3.6. Source identification of ions using principal component analysis
In this study, principal component analysis (PCA) technique was applied with major aerosol data to identify the main contributions of processes and emission sources of water-soluble ions. Table 3 presented PCA results with varimax rotation using $\mathsf{P M}_{2.5}$ chemical component data at four sites respectively. All principal factors with eigenvalues ${>}1$ were extracted.
At Wuqing site, three principal components were extracted and accounted for $90\%$ of the total variance. The principal component 1 had highly positive loading from $\mathsf{N H}_{4}^{+}$ (0.942), $\bar{\mathrm{S}}0_{4}^{2-}$ (0.912), $K^{+}$ (0.934), nss- $\bar{S}0_{4}^{2-}$ (0.911), nss- $\cdot K^{+}$ (0.934), OC (0.920), indicating industry source mixing with biomass burning. The principal component 2 is only characterized by $C1^{-}$ (0.543), which can be related to coal combustion. Additionally, the principal component 3 is linked with high value of $\mathrm{NO}_{3}^{-}$ (0.543). It can be identified as
vehicle emission.
We noted that the first component at Haining site is characterized by $\mathrm{NH4}\,(0.946),\mathrm{K}^{+}\,(0.929),\mathrm{nss-K}^{+}\,(0.929),$ , suggesting likely origins from industry source mixing with biomass burning. The second factor, explaining $14.59\%$ of the variance, had high factor loading for $\mathsf{C a}^{2+}\,\bar{(0.881)}$ and nss- $.C a^{2+}$ (0.896), indicating possible source from soil dust. The third factor is dominated by EC, which implies the component is associated with primary particles from incomplete combustion. The fourth factor with high loading for $C1^{-}$ (0.558) and low loading for $\mathtt{N a}^{+}$ (0.339) indicated the possible source from coal combustion.
Similarly, the principal component analysis performed on aerosol data set for Zhongshan site revealed industry source, soil dust and coal/biomass combustion as the main source for the ions in $\mathsf{P M}_{2.5}$ . At Deyang site, water-soluble ionic components in $\mathsf{P M}_{2.5}$ are possibly affected by industry source and soil dust. It is worth noting that the nss- $\cdot S0_{4}^{2-}$ together with OC had high loading on the first factor, which might be attributed to the secondary photochemical formation at this site.
4. Conclusions
The seasonal variation of aerosol $\mathsf{P M}_{2.5}$ and its ionic constituents as well as carbonaceous matter from four satellite city sites in China over one-year period of 2012e2013 are presented. The annual average $\mathsf{P M}_{2.5}$ mass concentrations ranged from 60.5 to $148.9~\upmu\mathrm{g}/$ $\mathfrak{m}^{3}$ . On average, Wuqing site had the highest concentration of aerosol $\mathsf{P M}_{2.5}$ and chemical species, followed by Deyang and Haining, while Zhongshan site was found being the lowest. The water-soluble ions were dominated by three ionic species $(\mathrm{SO}_{4}^{2-}$ , $\mathsf{N O}_{3}^{-},\mathsf{N H}_{4}^{+})$ accounting for $33{-}41\%$ of $\mathsf{P M}_{2.5}$ mass. In general, watersoluble ions exhibited a clear seasonal pattern with highest concentration in winter while lower in summer. Additionally, the total carbonaceous particle matter approximately comprised $16{-}23\%$ of $\mathsf{P M}_{2.5}$ mass.
Ion balance calculations showed the aerosols at four sites were characterized by acidic in nature. The aerosol chemical composition combined with back trajectory analysis confirmed that both longrange transport and local emissions have important impact on ions and carbonaceous particles in the aerosol at these sites. PCA indicated that major sources of ions are vehicular emissions, coal/ biomass combustion, industry source, as well as soil dust. Although this study provided us with valuable information on spatiotemporal variability of major chemical constituents in $\mathsf{P M}_{2.5}$ , our follow-up investigations into the detailed individual molecular compositions of fine aerosol at these sites are ongoing and therefore will give a better understanding of sources of fine aerosol, and impact of aerosol chemical compositions on atmospheric visibility impairment during haze events in China.
Acknowledgements
The authors thanks to NSERC Discovery grant, Queen Elizabeth II Scholarships, and the Strategic Priority Research Program (B) of the Chinese Academy of Sciences (XDB05060200) for financial support of this research.
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.
Andersson, A., Deng, J., Du, K., Zheng, M., Yan, C., Sk€old, M., Gustafsson, €O., 2015. Regionally-varying combustion sources of the January 2013 severe haze events over China. Environ. Sci. Technol. 49, 2038e2043.
Andreae, M.O., Andreae, T.W., Annegarn, H., Beer, J., Cachier, H., Canut, P., Elbert, W., Maenhaut, W., Salma, I., Wienhold, F.G., Zenker, T., 1998. Airborne studies of aerosol emissions from savanna fires in southern Africa: aerosol chemical
Bauecro, mS.pEo.,s itKioocnh. ,J . DG.,e oUpnhgyesr., RNe.,s . M10e3t z(gDe2r,4 )S,. 3M2.,1 1S9hien3d2e1ll2, 8.D.T., Streets, D.G., 2007. Nitrate aerosols today and in 2030: a global simulation including aerosols and tropospheric ozone. Atmos. Chem. Phys. 7, 5043e5059.
Cao, G., Zhang, X., Zheng, F., 2006. Inventory of black carbon and organic carbon emissions from China. Atmos. Environ. 40, 6516e6527.
Cao, J.J., Lee, S.C., Ho, K.F., Zhang, X.Y., Zou, S.C., Fung, K.K., Chow, J.C., Watson, J.G., 2003. Characteristics of carbonaceous aerosol in Pearl River Delta region, China during 2001 winter period. Atmos. Environ. 37, 1451e1460.
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, 1214e1226.
Chow, J.C., Watson, J.G., Fujita, E.M., Lu, Z., Lawson, D.R., 1994. Temporal and spatial variations of $\mathrm{PM}_{2.5}$ and $\mathrm{PM_{10}}$ aerosol in the Southern California air quality study. Atmos. Environ. 28, 2061e2080.
Draxler, R.R., Hess, G.D., 1998. An overview of the HYSPLIT 4 modelling system for trajectories, dispersion, and deposition. Aust. Meteorol. Mag. 47, 295e308.
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, 264e275.
Harrison, R.M., van Grieken, R.E., 1998. Atmospheric Particles. John Wiley and Sons.
He, K.B., Yang, F.M., Ma, Y.L., Zhang, Q., Yao, X.H., Chan, C.K., Cadle, S.H., Chan, T., Mulawa, P.A., 2001. The characteristics of $\mathrm{PM}_{2.5}$ in Beijing, China. Atmos. Environ. 35, 4959e4970.
Hennigan, C.J., Izumi, J., Sullivan, A.P., Weber, R.J., Nenes, A., 2015. A critical evaluation of proxy methods used to estimate the acidity of atmospheric particles. Atmos. Chem. Phys. 15, 2775e2790.
Ho, K.F., Lee, S.C., Chan, C.K., Yu, J., Chow, J.C., Yao, X.H., 2003. Characterization of chemical species in $\mathrm{PM}_{2.5}$ and $\mathrm{PM_{10}}$ aerosols in Hong Kong. Atmos. Environ. 37, 31e39.
Hu, G., Zhang, Y.M., Sun, J.Y., Zhang, L.M., Shen, X.J., Lin, W.L., Yang, Y., 2014. Variability, formation and acidity of water-soluble ions in $\mathrm{PM}_{2.5}$ in Beijing based on the semi-continuous observations. Atmos. Res. 145e146, 1e11.
Hu, M., Ling, Y.H., 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, 5853e5859.
Hu, M., Wu, Z.J., Slanina, J., Lin, P., Liu, S., Zeng, L.M., 2008. Acidic gases, ammonia and water-soluble ions in $\mathrm{PM}_{2.5}$ at a coastal site in the Pearl River Delta, China. Atmos. Environ. 42, 6310e6320.
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, nearcity and rural sites in Switzerland. Atmo. Environ. 39, 637e651.
Jacobson, M.C., Hansson, H.C., Noone, K.J., Charlson, R.J., 2000. Organic atmospheric aerosols: review and state of the science. Rev. Geophys. 38, 267e294.
Khoder, M.I., Hassan, S.K., 2008. Weekday/weekend differences in ambient aerosol level and chemical characteristics of water-soluble components in the city centre. Atmos. Environ. 42, 7483e7493.
Kundu, S., Kawamura, K., Andreae, T.W., Hoffer, A., Andreae, M.O., 2010. Diurnal variation in the water-soluble inorganic ions, organic carbon and isotopic compositions of total carbon and nitrogen in biomass burning aerosols from the LBA-SMOCC campaign in Rond^onia. Braz. Aerosol Sci. 41, 118e133.
Li, J., Posfai, M., Hobbs, P.V., Buseck, P.R., 2003. Individual aerosol particles from biomass burning in southern Africa: 2 compositions and aging of inorganic particles. J. Geophys. Res. 108 (D13) http://dx.doi.org/10.1029/2002JD002310.
Lin, Y.C., Cheng, M.T., Lin, W.H., Lan, Y.Y., Tsuang, B.J., 2010. Causes of the elevated nitrate aerosol levels during episodic days in Taichung urban area, Taiwan. Atmos. Environ. 44, 1632e1640.
McCulloch, A., Aucott, M.L., Benkovitz, C.M., Graedel, T.E., Kleiman, G., Midgley, P.M., Li, Y., 1999. Global emissions of hydrogen chloride and chloromethane from coal combustion, incineration and industrial activities: reactive chlorine emissions inventory. J. Geophys. Res. 104 (D7), 8391e8403.
Meng, Z.Y., Zhang, R.J., Lin, W.L., Jia, X.F., Yu, X.M., Yu, X.L., Wang, G.H., 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.
Morales, J.A., Pirela, D., de Nava, M.G., de Borrego, B.S., Vel asquez, H., Dur an, J., 1998. Inorganic water soluble ions in atmospheric particles over Maracaibo Lake Basin in the western region of Venezuela. Atmos. Res. 46, 307e320.
Pandis, S.N., Seinfeld, J.H., Pilinis, C., 1990. The Smog-Fog-Smog cycle and acid deposition. J. Geophys. Res. 95 (D11), 18489e18500.
Paraskevopoulou, D., Liakakou, E., Gerasopoulos, E., Mihalopoulos, N., 2015. Sources of atmospheric aerosol from long-term measurements (5 years) of chemical composition in Athens, Greece. Sci. Total Environ. 527e528, 165e178.
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. Atmos. Chem. Phys. 9, 1711e1722.
Putaud, J.P., Dingenen, R.V., Dell'Acqua, A., Raes, F., Matta, E., Decesari, S., Facchini, M.C., Fuzzi, S., 2004. Size-segregated aerosol mass closure and chemical composition in Monte Cimone (I) during MINATROC. Atmos. Chem. Phys. 4, 889e902.
Raizenne, M., Neas, L.M., Damokosh, A.L., Dockery, D.W., Spen-gler, J.D., Koutrakis, P., Ware, J.H., Speizer, F.E., 1996. Health effects of acid aerosols on North American children: pulmonary function. Environ. Health Perspect. 104, 506e514.
Schauer, J.J., Cass, G.R., 2000. Source apportionment of wintertime gas-phase and particle-phase air pollutants using organic compounds as tracers. Environ. Sci. Technol. 34, 1821e1832.
Seinfeld, J.H., Pandis, S.N., 2006. Atmospheric Chemistry and Physics from Air Pollution to Climate Change. John Wiley, Sons, Inc., New York.
Shen, Z.X., Cao, J.J., Arimoto, R., Han, Z.W., Zhang, R.J., Han, Y.M., Liu, S.X., Okuda, T., Nakao, S., Tanaka, S., 2009. Ionic composition of TSP and PM2.5 during dust storms and air pollution episodes at Xi'an, China. Atmos. Environ. 43, 2911e2918.
Shon, Z.-H., Ghosh, S., Kim, K.-H., Song, S.-K., Jung, K., Kim, N.-J., 2013. Analysis of water-soluble ions and their precursor gases over diurnal cycle. Atmos. Res. 132e133, 309e321.
Strader, R., Lurmann, F., Pandis, S., 1999. Evaluation of secondary organic aerosol formation in winter. Atmos. Environ. 33, 4849e4863.
Tan, J.H., Duan, J.C., Chen, D.H., Wang, X.H., Guo, S.J., Bi, X.H., Sheng, G.Y., He, K.B., Fu, J.M., 2009. Chemical characteristics of haze during summer and winter in Guangzhou. Atmos. Res. 94, 238e245.
Tao, J., Gao, J., Zhang, L., Zhang, R., Che, H., Zhang, Z., Lin, Z., Jing, J., Cao, J., Hsu, S.-C., 2014. PM2.5 pollution in a megacity of southwest China: source apportionment and implication. Atmos. Chem. Phys. 14, 8679e8699.
Turpin, B., Huntzicker, J., 1995. Identification of secondary organic aerosol episodes and quantitation of primary and secondary organic aerosol concentrations during SCAQS. Atmos. Environ. 29, 3527e3544.
Turpin, B.J., Lim, H.J., 2001. Species contributions to $\mathrm{PM}_{2.5}$ mass concentrations: revisiting common assumptions for estimating organic mass. Aerosol Sci. Tech. 35, 602e610.
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, 65e76.
Wang, G.H., Huang, L.M., Gao, S.X., Gao, S.T., Wang, L.C., 2002. Characterization of water- soluble species of PM10 and PM2.5 aerosols in urban area in Nanjing, China. Atmos. Environ. 36, 1299e1307.
Wang, Y., Zhuang, G., Tang, A., Yuan, H., Sun, Y., Chen, S., Zheng, A., 2005. The ion chemistry and the source of $\mathrm{PM}_{2.5}$ aerosol in Beijing. Atmos. Environ. 39, 3771e3784.
Wang, Y., Zhuang, G.S., Zhang, X.Y., Huang, K., Xu, C., Tang, A., Chen, J.M., An, Z.S., 2006. The ion chemistry, seasonal cycle and sources of $\mathrm{PM}_{2.5}$ and TSP aerosol in Shanghai. Atmos. Environ. 40, 2935e2952.
Xing, Z.Y., Deng, J.J., Mu, C., Wang, Y., Du, K., 2014. Seasonal variation of mass absorption efficiency of elemental carbon in the four major emission areas in China. Aerosol Air Qual. Res. 14, 1897e1905.
Yao, X.H., Chak, K.C., Fang, M., Cadle, S., Chan, T., Mulawa, P., He, K.B., Ye, B.M., 2002. The water-soluble ionic composition of $\mathrm{PM}_{2.5}$ in Shanghai and Beijing. China. Atmos. Environ. 36, 4223e4234.
Yin, L.Q., Niu, Z.C., Chen, X.Q., Chen, J.S., Zhang, F.W., Xu, L.L., 2014. Characteristics of water-soluble inorganic ions in $\mathrm{PM}_{2.5}$ and $\mathrm{PM}_{2.5-10}$ in the coastal urban agglomeration along the Western Taiwan Strait Region, China. Environ. Sci. Pollut. Res. 21, 5141e5156.
Zhang, R., Jing, J., Tao, J., Hsu, S.-C., Wang, G., Cao, J., Lee, C.S.L., Zhu, L., Chen, Z., Zhao, Y., Shen, Z., 2013. Chemical characterizations and source apportionment of PM2.5 in Beijing: seasonal perspective. Atmos. Chem. Phys. 13, 7053e7074.
Zhang, T., Cao, J.J., Tie, X.X., Shen, Z.X., Liu, S.X., Ding, H., Han, Y.M., Wang, G.H., Ho, K.F., Qiang, J., 2011. Water-soluble ions in atmospheric aerosols measured in Xi'an, China: seasonal variations and sources. Atmos. Res. 102, 110e119.
Zhou, S.Z., Wang, Z., Gao, R., Xue, L.K., Yuan, C., Wang, T., Gao, X.M., Wang, X.F., Nie, W., Xu, Z., Zhang, Q.Z., Wang, W.X., 2012. Formation of secondary organic carbon and long-range transport of carbonaceous aerosols at Mount Heng in South China. Atmos. Environ. 63, 203e212.
|
atmosphere
|
0106
|
10.1007/s11356-018-2238-1
|
image
|
Fig. 1 Location of the sampling site in Sanya, China
|
|
atmosphere
|
0106
|
10.1007/s11356-018-2238-1
|
table
|
Table 1 Limit of detection and relative error of analysis methods for carbonaceous aerosol, watersoluble ions, and selected elemental components
|
|
atmosphere
|
0106
|
10.1007/s11356-018-2238-1
|
image
|
Fig. 2 Time series of meteorological conditions and $\operatorname{PM}_{2.5}$ concentrations
|
|
atmosphere
|
0106
|
10.1007/s11356-018-2238-1
|
image
|
unaccounted-for mass after subtracting the sum of measured components from the $\mathrm{PM}_{2.5}$ mass. Unaccounted-for mass may be composed of unmeasured geological material (e.g., calcium carbonate), OM with a higher than assumed fraction of oxygen, residual water, and other unmeasured species. Uncertainties in gravimetric analyses also could introduce errors
|
|
atmosphere
|
0106
|
10.1007/s11356-018-2238-1
|
image
|
Fig. 4 Relationship between neutralization ratio and free acidity
|
|
atmosphere
|
0106
|
10.1007/s11356-018-2238-1
|
image
|
Fig. 5 Three-day air-mass trajectories calculated backwards in time reaching Sanya at $500\;\mathrm{m}$ above ground every hour during the campaign period and their grouping into three trajectory clusters
|
|
atmosphere
|
0106
|
10.1007/s11356-018-2238-1
|
table
|
Table 3 Characteristics of three trajectory clusters and $\operatorname{PM}_{2.5}$ , species concentrations, and parameters grouped by cluster
|
|
atmosphere
|
0106
|
10.1007/s11356-018-2238-1
|
image
|
Fig. 6 Horizontal winds $200\,\textrm{m}$ above ground level and a EC the EC model concentrations assuming b no emissions from Hainan and c concentrations obtained with the Weather Research and Forecasting- no emissions from South China for 30 January to 1 February 2012 Elemental or Black Carbon model at Hainan and surrounding areas and
|
|
atmosphere
|
0106
|
10.1007/s11356-018-2238-1
|
text
| Not supported with pagination yet |
Chemical characterization of $\mathsf{P M}_{2.5}$ from a southern coastal city of China: applications of modeling and chemical tracers in demonstration of regional transport
Jiamao Zhou $^{1,2}\mathbb{\oplus}$ & Steven Sai Hang Ho1,3 & Junji Cao1 & Zhuzi Zhao1 & Shuyu Zhao1 & Chongshu Zhu1 & Qiyuan Wang1 & Suixin Liu1 & Ting Zhang1 $\cdot$ Youzhi Zhao4 & Ping Wang4 & Xuexi Tie1
Received: 9 November 2017 /Accepted: 4 May 2018
$\copyright$ Springer-Verlag GmbH Germany, part of Springer Nature 2018
Abstract
An intensive sampling campaign of airborne fine particles $(\mathrm{PM}_{2.5})$ was conducted at Sanya, a coastal city in Southern China, from January to February 2012. Chemical analyses and mass reconstruction were used identify potential pollution sources and investigate atmospheric reaction mechanisms. A thermodynamic model indicated that low ammonia and high relative humidity caused the aerosols be acidic and that drove heterogeneous reactions which led to the formation of secondary inorganic aerosol. Relationships among neutralization ratios, free acidity, and air-mass trajectories suggest that the atmosphere at Sanya was impacted by both local and regional emissions. Three major transport pathways were identified, and flow from the northeast (from South China) typically brought the most polluted air to Sanya. A case study confirmed strong impact from South China (e.g., Pearl River Delta region) (contributed $76.8\%$ to EC, and then this result can be extended to primary pollutants) when the northeast winds were dominant. The Weather Research Forecasting Black carbon model and trace organic markers were used to apportion local pollution versus regional contributions. Results of the study offer new insights into the atmospheric conditions and air pollution at this coastal city.
Keywords $\mathrm{PM}_{2.5}$ . Chemical composition $\cdot$ Aerosol acidity $\cdot$ Regional transport $\cdot$ Sanya $\cdot$ Organic markers
Introduction
There have been growing concerns over urban air quality across China in recent decades due to the extremely high aerosol loadings that have accompanied the country’s rapid economic development (Zhang et al. 2013; Guan et al. 2014). Atmospheric pollution by particulate matter (PM) with aerodynamic equivalent diameters of $\mathrm{Dp}<2.5\ \upmu\mathrm{m}\ (\mathrm{PM}_{2.5})$ has become a critical issue owing to its effects on climate (IPCC 2013), visibility (Watson 2002), cloud condensation nuclei (CCN) (Seinfeld and Pandis 2006), and cloud, rain, and fog water (Hong et al. 2002). Numerous studies have been conducted in China to investigate $\operatorname{PM}_{2.5}$ composition, aerosol physical characteristics, haze formation mechanisms, regional pollution, and environmental effects in large cities, coastal regions, and background sites (Liu et al. 2013, Guo et al. 2014, Pu et al. 2015, Feng et al. 2015, Guo et al. 2017, Zhao et al. 2015a, 2013).
Here, we present the results of a study conducted at Sanya, which is located on the southern end of Hainan Island in the South China Sea and a popular tourist destination. A few studies have been conducted in Sanya to investigate ecological environments, sediment distributions, and meteorological conditions (Li et al. 2013; Qiao et al. 2015), but there has been a lack of comprehensive monitoring of aerosol particles in Sanya or Hainan Island. Recently, (Wang et al. 2015) reported on carbonaceous $\operatorname{PM}_{2.5}$ at Sanya, and these authors concluded that there were impacts from a variety of pollution sources in winter while emissions from the burning of fossil fuels were most important in summer. Renowned for its tropical climate, the air quality in Sanya is nearly always better than in most of the Chinese megacities, but occasional pollution episodes do occur there. These affect both air quality and visibility, and they occur most often in wintertime.
The geography of Sanya and its maritime atmosphere make it subject to effects land-sea breeze circulation. The first objective of the work described here was to determine the chemical composition of airborne $\mathrm{PM}_{2.5}$ at Sanya during a 1-month intensive monitoring study in winter. We recorded the atmospheric conditions, measured selected chemical components, and characterized transport pathways that can affect the air quality in the area. The Weather Research Forecasting Black Carbon (WRF-BC) model and an organic tracer method were then used to evaluate the contributions of regional emissions to $\operatorname{PM}_{2.5}$ at Sanya. Studies of particle formation mechanisms undertaken for the study have led to a better understanding of what controls the aerosol populations, and the results also contribute to the developing national chemical database on $\mathrm{PM}_{2.5}$ in China.
Sampling and methodology
Sampling site and methods
Sampling was conducted at the Hainan Tropical Ocean University in the northeastern part of Sanya City, Hainan Province, China $\left\langle{18^{\circ}\ 18^{\prime}\ N}\right.$ , $109^{\circ}\,\,31^{\prime}\mathrm{~E~}$ ; Fig. 1) from January 8 to February 8, 2012. Sanya is in the tropical marine monsoon climate zone, and the sampling site itself was is in a suburban area, $\sim10~\mathrm{km}$ from the South China Sea (Wang et al. 2015).
The sampling equipment was installed on the roof of an academic office building on the campus $\sim20\,\textrm{m}$ above ground level. Twenty-four hour integrated daily samples (from $10{\cdot}00\ \mathrm{a.m}$ . to $10{\cdot}00\ \mathrm{a.m}$ . in the next day) were collected on $47{-}\mathrm{mm}$ -diameter quartz-fiber filters $(\mathrm{QM}/\mathrm{A}\mathbb{B})$ , Whatman Inc., UK), and these were used for measurements of carbonaceous aerosols and water-soluble ions. Another set of samples was collected using Teflon $\ensuremath{\mathcal{R}}$ -membrane filters ( $^{47}\mathrm{~mm}$ , $\mathrm{QM/A\textcircled{\mathrm{B}}}$ , Whatman Inc., UK), and they were used for elemental analyses. Both sets of samples were collected using Airmetrics $\mathrm{PM}_{2.5}$ Mini-Volume samplers (Springfield, OR, USA), which operated at flow rates of $5\;\mathrm{L}\;\mathrm{min}^{-1}$ . Two blank samples were collected and analyzed to account for any positive artifacts. The sampler was regularly calibrated using a Defender 510 Volumetric Primary Flow Standard (Bios International Corporation, Butler, NJ, USA). The variance in flow was approximately $\pm\,2\%$ . All of the samples were properly stored in a freezer at $4\,^{\circ}\mathrm{C}$ until analysis to prevent any loss of volatile matter. Meteorological data were provided by the Sanya Meteorology Administration.
Chemical analyses
Mass measurements
For the mass measurements, the Teflon $\textsuperscript{\textregistered}$ filters were equilibrated at a temperature between 20 and $23\ ^{\circ}\mathrm{C}$ and relative humidity (RH) of $35{-}45\%$ for at least $24~\mathrm{h}$ , and the masses of the filters were determined be weighing the filters before and after the samples were collected. A Sartorius MC5 electronic microbalance with $\pm\,1~\upmu\mathrm{g}$ sensitivity (Sartorius, Gottingen, Germany) was used for the mass determinations. Each filter was weighed at least three times, and the net mass was obtained by subtracting the difference between the averaged pre- and post-sampling weights.
Carbonaceous aerosol analysis
Prior to sampling, all of the quartz-fiber filters were preheated in a furnace at $800\,^{\circ}\mathrm{C}$ for $^{3\,\mathrm{h}}$ to remove any contaminants, and both before and after sampling, the filters were stored in the freezer at $<\!4\,^{\circ}\mathrm{C}$ . The carbon analyses were carried out using a Desert Research Institute (DRI) Model 2001 carbon analyzer (Atmoslytic Inc., Calabasas, CA, USA) following the IMPROVE_A thermal/optical reflectance (TOR) protocol (Chow et al. 1993; Chow et al. 2004). A punch aliquot of each quartz filter sample was heated in a stepwise manner to obtain data for four organic carbon (OC) fractions (OC1, OC2, OC3, and OC4 in a helium atmosphere at 140, 280, 480, and $580\ ^{\circ}\mathrm{C})$ , and three elemental carbon (EC) fractions (EC1, EC2, and EC3 in a $2\%$ oxygen $\rho8\%$ helium atmosphere at 580, 740, and $840\ ^{\circ}\mathrm{C})$ . Pyrolyzed carbon (OP) was produced at $<580~^{\circ}\mathrm{C}$ in the inert atmosphere, and the decrease in reflected light caused by OP was used to correct for charred OC. Total OC was defined as the sum of OP plus the four OC fractions, and total EC was the sum of the three EC fractions minus OP.
Elemental analysis
The concentrations of 40 elements (Na, Mg, Al, Si, P, S, Cl, K, Ca, Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Ga, As, Se, Br, Rb, Sr, Y, Mo, Pd, Ag, Cd, Sn, Sb, Cs, Ba, La, W, Au, Hg, Pb, and Tl) on the Teflon $\textsuperscript{\textregistered}$ filters were measured by energy dispersive xray fluorescence (ED-XRF) spectrometry (Epsilon 5 ED-XRF, PANalytical B.V., the Netherlands). The spectrometer used a three-dimensional polarizing geometry with 11 secondary targets (i.e., $\mathrm{CeO}_{2}$ , CsI, Ag, Mo, Zr, KBr, Ge, Zn, Fe, Ti, and Al) and one target $\left(\mathrm{Al}_{2}\mathrm{O}_{3}\right)$ ) that resulted in high signal-to-noise ratios and low detection limits (Watson et al. 1999). The X-ray source was a side-window X-ray tube with a gadolinium anode that operated at a current of 0.5 to $24~\mathrm{mA}$ and an accelerating voltage of 25 to $100\;\mathrm{kV}$ (maximum power $600~\mathrm{W}$ ). The characteristic X-ray radiation was detected by a germanium detector (PAN 32). Each filter sample was counted for $30\;\mathrm{min}$ .
Ion analysis
One quarter of each quartz-fiber sample filter was cut using and extracted in $10~\mathrm{ml}$ of high-purity water. Ten major ions $\left(\mathrm{Na}^{+}$ , $\mathrm{NH_{4}}^{+}$ , $\mathrm{K}^{+}$ , $\mathrm{Mg}^{2+}$ , $C\mathrm{a}^{2+}$ , $\mathrm{NO}_{3}^{\phantom{\mathrm{-}}}$ , $\mathrm{SO}_{4}{}^{2-}$ , $\mathrm{Cl}^{-}$ , $\mathrm{F}^{-}.$ , and Br−) were measured with a DX600 ion chromatograph (Dionex Inc., Sunnyvale, CA, USA) (Chow and Watson 1999). A
CS12 column $(150\times4~\mathrm{mm})$ and an AS14 column $150\times$ $4\;\mathrm{mm})$ were used for chromatographic separations of cations and anions analyses, respectively. The mothd details can be found in (Zhang et al. 2011).
Organic speciation
Non-polar organic compounds, including polycyclic aromatic hydrocarbon (PAHs), alkanes, and hopanes, were measured on the quartz-fiber filter using in-injection port thermal desorption-gas chromatography/mass spectrometry (TD-GC/ MS). The sensitivity of TD-GC/MS method is much better than the traditional solvent extraction approach, and the thermal desorption method also requires less labor and uses less chemicals (Chow et al. 2007). The experimental procedures we used have been described in detail by Ho and Yu (2004)), Ho et al. (2008)), and Ho et al. (2011)). In brief, a filter aliquot $(0.5–1.0\ \mathrm{cm}^{2})$ was cut and thermally extracted inside the GC injector port (GC 6890, Agilent Technology, Inc., Santa Clara, CA, USA) at $275~^{\circ}\mathrm{C}$ . The desorbed organic compounds were separated with a capillary column and detected by a MS detector $(m/z~50{-}550)$ (MS 5975, Agilent Technology, Inc.).
Quality assurance and quality control (QA/QC)
The limit of detections (LODs) and relative errors for carbonaceous aerosol, water-soluble ions, and elemental components are shown in Table 1. The LOD is defined as the amount of an analyte that generates an analytical signal equal to the sum of the mean blank signal plus three times the standard deviation of the blank signals (Meier and Zünd 2005).
QA/QC procedures have been described in detail for the carbon analysis (Cao et al. 2003), elemental analyses (Xu et al. 2012, 2016), and ion chromatographic measurements (Shen et al. 2009a, b; Zhang et al. 2011). The QA/QC procedures for the organic species analyzed by TD-GC/MS also have been described elsewhere (Ho et al. 2008, 2011). Replicate analyses were done for each group of ten samples. Blank filters were also analyzed, and the sample results were corrected for the average of the blank concentrations.
Non-sea salt sulfate calculation
Sea salt sulfate $\mathrm{(ssSO_{4}}^{2-})$ and non-sea salt sulfate $\mathrm{(nssSO}_{4}{}^{2-})$ were calculated by assuming that $\mathrm{Na^{+}}$ was only from by sea salt and multiplying the measured $\mathrm{Na^{+}}$ by the seawater $\mathrm{SO}_{4}{}^{2}$ $^{-}/\mathrm{Na}^{+}$ mass ratio of 0.25 as follows:
$$
\mathrm{ssSO}_{4}^{2-}=0.25\times\mathrm{Na}^{+}
$$
$$
\mathrm{nssSO}_{4}^{2-}=\mathrm{SO}_{4}^{2-}\mathrm{-ssSO}_{4}^{2-}
$$
Thermodynamic model
Acidity is one of the most important factors that determine the aerosols’ physical and chemical properties. Investigations of in situ aerosol properties and how they vary in relation to acidity and water content are fundamental to assessing their involvement in heterogeneous reactions, especially nitrate formation (Pathak et al. 2009). Many researchers (Zhang et al. 2000; Pathak et al. 2004; Zhang et al. 2007; Pathak et al. 2009; Engelhart et al. 2011; Pathak et al. 2011; Squizzato et al. 2013) have applied the Extended Aerosol Inorganic Model (E-AIM, http://www.aim.env.uea.ac.uk/aim/aim.php) to simulate the in situ aerosol acidity $(\mathrm{[H^{+}]_{F r e e}})$ , the activities of ionic species in aqueous aerosols, and the solid- and liquid-phase composition of the aerosol (Clegg et al. 1998). In this study, a recent version of the E-AIM model IV (E-AIM4, (Friese and Ebel 2010) was used to simulate the acidity and thermodynamic properties of the $\mathrm{H^{+}}\mathrm{-NH_{4}^{+}\mathrm{-Na^{+}\mathrm{-SO_{4}^{2-}\mathrm{-NO_{3}^{-}\mathrm{-Cl^{-}\mathrm{-}H_{2}O}}}}}$ mixture in $\operatorname{PM}_{2.5}$ . The average ambient temperature, relative humidity (RH), and molar concentrations of total aerosol acidity $\left(\left[\mathrm{H}^{+}\right]\right.$ Total) were input into E-AIM4 to obtain $[\mathrm{H}^{+}]_{\mathrm{Free}}$ and the number of moles of selected chemical species in the aqueous phase. In this study, $[\mathrm{H}^{+}]_{\mathrm{Total}}$ was estimated using the ionic balance of the dominant inorganic ionic species (Lippmann et al. 2000; Pathak et al. 2009); these include sulfate, nitrate, chloride, ammonium, and sodium.
$$
[H^{+}]_{T o t a l}=2S O_{4}^{2-}+N O_{3}^{-}+C l^{-}-\left(N H_{4}^{+}+N a^{+}\right)
$$
Results and discussion
Time series of meteorological conditions and $\mathsf{P M}_{2.5}$
Meteorological data provided by Sanya Meteorology Administration showed that the prevailing winds during study were northeasterly, and a time series plot shows that the wind speed was relatively constant over the course of the study (Fig. 2). The average wind speed, temperature, and relative humidity (RH) were $0.6\pm0.2\,\mathrm{ms}^{-1}$ , $22.5\pm1.4\,^{\circ}\mathrm{C},$ , and $90.9\pm$ $5.6\%$ , respectively. The daily concentrations of $\operatorname{PM}_{2.5}$ ranged from 8.6 to $35.3~\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , and the arithmetic mean $\mathrm{PM}_{2.5}$ mass was $20.4~\upmu\mathrm{g}~\mathrm{m}^{-3}$ . Four typical periods, that is, two episodes with high $\mathrm{PM}_{2.5}$ (identified with red rectangles in Fig. 2) and two episodes with low $\operatorname{PM}_{2.5}$ (shown in blue rectangles), were selected for discussion. We found that higher $\operatorname{PM}_{2.5}$ mass concentrations occurred when the winds were from the northnortheast wind and the RH was low. Lower $\operatorname{PM}_{2.5}$ loadings occurred when the winds were south or southwesterly and the RH was high. These results show that fine particles at Sanya are affected by wind direction, which in turn implies that transport affects the aerosol. More discussion of the daily changes in $\operatorname{PM}_{2.5}$ is presented below.
$\mathsf{P M}_{2.5}$ chemical composition
The $\mathrm{PM}_{2.5}$ mass concentrations and those of chemicals that mostly concerned are shown in Table 2. The $\operatorname{PM}_{2.5}$ mass and chemical species concentrations at Sanya were obviously lower than those in other Chinese coastal cities including Hong Kong, Qingdao, Shanghai, Xiamen, and Guangzhou where the same analytical methods were used (Cao et al. 2012). The concentrations of $\operatorname{PM}_{2.5}$ , carbonaceous aerosol, watersoluble aerosol, and elemental components also were lower compared with a group of coastal cities (13 cities) along the Western Taiwan Straits region. The $\mathrm{PM}_{2.5}$ mass concentrations were below the $24\mathrm{-h}$ limits of $75~\upmu\mathrm{g}~\mathrm{m}^{-3}$ promulgated by the Chinese National Environmental Protection Agency, and they also were lower than the recommended levels established by other countries and international health organizations. The good air quality at Sanya can be explained by the relatively small population of $\sim500{,}000$ and limited impacts from anthropogenic pollution sources, such as heavy industries. Tourism is a pillar industry in Sanya and that has small to environmental impacts other than limited emissions from local transport and vehicles used for tourism. In addition, the geographical location of Sanya is favorable: the land breezesea breeze circulation prevails (Fig. 1), and meteorological conditions are generally conducive to pollutant dispersal.
The geography and climate at Sanya have a degree of similarity with Hong Kong where a comprehensive study on atmospheric fine particle was conducted (Cao et al. 2012). In Hong Kong, seasonal variations in $\operatorname{PM}_{2.5}$ were apparent, and high loadings of various chemical components were found in winter. However, the concentration differences of most species (e.g., means of OC and EC) between summer and winter were vanishingly small in Sanya (Wang et al. 2015). The lack of seasonality can be ascribed to the consistency of local pollution inputs. In addition, the concentrations of several chemicals, including Ti and Fe, were very low in Sanya, much closer to the levels in Hong Kong in summer than in the more polluted winter. These results simply indicate that there were different pollution sources for the two cities even though both are located on coastlines.
Aerosol mass balance
A material balance for the aerosol was calculated to rank the mass contributions of important marker species (Fig. 3). The order of importance was organic matter $\mathrm{(OM)}>$ nss-sulfate $\mathrm{(nss{-}S O_{4}}^{2-})$ $>$ geological material $(\mathrm{GM})\!>\!\mathrm{EC}\!>$ nitrate $\left(\mathrm{NO}_{3}\right)$ $>$ sea salts $>$ ammonium $\mathrm{(NH_{4}}^{+})$ . Due to the low concentrations, several chemical components in some samples were not detected, including $\mathrm{Cl}^{-}$ , $\mathrm{NO}_{2}^{\phantom{\mathrm{~-~}}}$ , $\mathrm{Na^{+}}$ , $\mathrm{NH_{4}}^{+}$ , and $\mathrm{K}^{+}$ which had limits of detection of 0.0087, 0.0050, 0.0005, 0.0010, and $0.0011~\upmu\mathrm{g}\;\mathrm{m}\Gamma^{-1}$ , respectively. The average concentrations for those analytes were calculated by substituting half of the LOD for the undetected samples to avoid overestimation of the concentrations.
Approximately $21.0\%$ of the $\operatorname{PM}_{2.5}$ mass was not identified. This value is higher than the results shown by (Cao et al. 2012) and that may have been due to underestimations of the weighting factors for certain components or uncertainties in the gravimetric measurements. For example, the OM would have underestimated if the non-quantified mass ratio of 1.6 that was used in the calculations is not representative of the OM at Sanya. In fact, the lush vegetation at Sanya and surrounding areas may provide precursors for secondary aerosol formation, including OM, and the photochemical reactions at the site would be favored by the high temperatures and strong solar radiation (Wang et al. 2015). Andrews et al. (2011) also found that $28–42\%$ of the $\operatorname{PM}_{2.5}$ collected in national park could not be identified, attributed to the formations of multifunctional organic compounds. GM accounted for $17.8\%$ of the $\mathrm{PM}_{2.5}$ mass in Sanya compared with 12 to $34\%$ at 14 Chinese megacities (Cao et al. 2012). This type of material is typically due to soil and dust re-suspension, construction projects, and agricultural activities. A relatively high contribution of GM at Sanya can be explained by the fact that an active construction site was next to the sampling location during the study. The relative abundances of other fractions were consistent with the results for other cities, including Beijing, Shanghai, Guangzhou, and Chongqin (Cao et al. 2012; Zhao et al. 2013), where OM and nss- $\cdot\mathrm{{SO}}_{4}{}^{2-}$ were two major PM components. At Sanya, OM and nss- $\cdot\mathrm{SO}_{4}{}^{2-}$ accounted for 24.5 and $20.5\%$ , respectively, of the total PM mass.
Oxidation-reduction reactions involving $\mathrm{H}_{2}\mathrm{O}_{2}$ contribute to sulfate formation in the aqueous phase, particularly when the $\mathrm{pH}$ is below 5 (Chandler et al. 1988; Das and Aneja 1994;
Fig. 3 Chemical composition of $\operatorname{PM}_{2.5}$ from Sanya. Organic matter was estimated as $1.6\times\mathrm{OC}$ (El-Zanan et al. 2005; Chen and $\mathrm{Yu}~2007$ ; ElZanan et al. 2012) to account for the unmeasured hydrogen and oxygen. Geological material was estimated as $\mathrm{Al}/0.07$ (Zhang et al. 2003). Sea salts were estimated as $\mathrm{C}1^{-}+(1.4486\times\mathrm{Na}^{+})$ where 1.4486 is the ratio of the concentration of all elements in sea water except $\mathrm{Cl}^{-}$ to the concentration of $\mathrm{Na^{+}}$ (Li et al. 2010).BOthers^ was the remaining
Fung et al. 1991; He et al. 2010; Meagher et al. 1990), and the results indicate that high RH greatly promoted the oxidation of sulfur dioxide $(\mathrm{SO}_{2})$ ) to $\mathrm{SO}_{4}^{\ 2-}$ . Sea salt was estimated to be $3.67\%$ of the $\operatorname{PM}_{2.5}$ mass, and therefore, sea salt $\mathrm{SO}_{4}^{\ 2-}$ was almost negligible, accounting for only $0.34\%$ of $\mathrm{PM}_{2.5}$ mass. The relatively low value for sea salt is close to the contributions of 4.2 and $5.0\%$ reported at the urban sites of Borgerhout (Bencs et al. 2008). However, the small sea salt proportion in the atmosphere of Sanya can be explained, at least in part, by the seasonality of the prevailing winds; that is, northeasterly winds prevail in winter, and therefore, the aerosol is mainly influenced by air masses from the mainland. The phenomenon of Bchlorine loss^ (Ohta and Okita 1990) and multivariate analysis (Thurston and Spengler 1985) of the data for water-soluble ions showed that sea salts were significantly impacted by the sea breeze while the secondary salts, like sulfate, were mainly contributed by land breeze. It is also supported by the mean value $[\mathrm{Cl}^{-}]/[\mathrm{Na}^{+}]$ (0.42 in this study) which is lower than the value in the sea water (1.16) (Keene et al. 1986).
Ammonium ion had an average concentration of $0.4\pm$ $0.6\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ , and it accounted for only $1.8\%$ of the $\operatorname{PM}_{2.5}$ mass and was undetectable in 32 samples. Animal manure and fertilizer applications are well-known sources for ammonia $\left(\mathrm{NH}_{3}\right)$ (Cui et al. 2013; Gu et al. 2014), but the results indicate there was relatively little ammonium from these sources around Sanya. Other anthropogenic, non-agricultural sources for $\mathrm{NH}_{3}$ , such as motor vehicles, fuel combustion, and landfills also exist, but their impact apparently was small (Pierson and Brachaczek 1983; Sutton et al. 2000; Battye et al. 2003; Wilson et al. 2004; Sutton et al. 2008). As noted above, tourism is the main industry in Sanya, and emissions from the industrial and agricultural sectors are relatively small.
Wu et al. (2016)) used annual girded emissions data to construct and annual Chinese $\mathrm{NH}_{3}$ emission map from 2001 to 2008.The annual $\mathrm{NH}_{3}$ emission for the Hainan Province was only $0.5–2.1\;\mathrm{t}\,\mathrm{km}^{-2}\,\mathrm{mol}\,\mathrm{km}^{-3}$ , which was consistent with our results showing low ammonium levels. Due to its short atmospheric lifetime, $\mathrm{NH}_{3}$ typically deposits on surfaces near the source regions (Walker et al. 2004). As the primary gaseous base in the atmosphere, $\mathrm{NH}_{3}$ influences the acidity of solid- and aqueous-phase aerosol species, cloud water, and precipitation. Ammonia may be either wet- or dry-deposited as a gas or react with sulfuric $\mathrm{(H_{2}S O4)}$ , nitric $\mathrm{(HNO_{3})}$ , nitrous acid (HONO), and hydrochloric acid (HCl) to form ammonium sulfate $[(\mathrm{NH}_{4})_{2}\mathrm{SO}_{4}]$ , bisulfate $\mathrm{(NH_{4}H S O_{4})}$ ), nitrate $\left(\mathrm{NH}_{4}\mathrm{NO}_{3}\right)$ , nitrite $\mathrm{(NH}_{4}\mathrm{NO}_{2})$ , and chloride $\left(\mathrm{NH_{4}C l}\right)$ aerosols. The Pearson correlation coefficient calculated for $\mathrm{NH_{4}}^{+}$ and $\mathrm{SO}_{4}^{\ 2-}$ at Sanya was 0.94 followed by 0.69 for and $\mathrm{NO}_{2}^{\mathrm{~-~}}$ and 0.64 for $\mathrm{NH_{4}}^{+}$ and $\mathrm{NO}_{3}^{\phantom{\mathrm{\,-}}}$ . These correlations suggest that $\mathrm{NH_{4}}^{+}$ preferentially neutralizes $\mathrm{H}_{2}\mathrm{SO}_{4}$ , which has a relatively low vapor pressure. However, when ammonium concentrations are low and neutralization ratios are small, therefore, other anions can neutralize the residual acids by forming sodium, potassium, calcium, and magnesium salts.
Aerosol acidity
Two important parameters commonly used to represent the aerosol acidity. The first is total acidity $\mathrm{([H^{+}]_{T o t a l})}$ , which is defined as the total amount of the deliquesced aerosol acid measured as strong acids, including sulfuric, nitric, and chloric acid (Li et al. 2014) and calculated by the method described in the BThermodynamic model^ section. The second parameter, in situ acidity $\mathrm{[H^{+}]_{F r e e}}.$ , is the actual acidity of aerosol, which is influenced by the chemical and photochemical reactions in the atmosphere and included in E-AIM model IV. The average $[\mathrm{H}^{+}]_{\mathrm{Total}}$ and $[\mathrm{H}^{+}]_{\mathrm{Free}}$ calculated by the thermodynamic method were $168.2\pm84.8$ and $108.9\pm50.8\;\mathrm{nmol}\;\mathrm{m}^{-3}$ , respectively. These values are lower than what has been reported for megacities such as Beijing and Shanghai, but they are higher than those at Guangzhou and Lanzhou in summer (Pathak et al. 2009). In addition, the aerosol at Sanya was more acidic than that estimated for Xiamen, another coastal city, in winter $\mathrm{\DeltaWu}$ et al. 2017). It is noteworthy that $[\mathrm{H^{+}}]_{\mathrm{Free}}$ accounted for $64.7\%$ of $[\mathrm{H}^{+}]_{\mathrm{Total}}$ . The proportion is almost double the averages measured in the five cities referenced above. Aerosol acidities can vary over a wide range for different source materials, either anthropogenic or natural, and they can be affected by regional emissions and weather conditions. Higher water contents in aerosol particles can promote the uptake of sulfur dioxide $(\mathrm{SO}_{2})$ , $\mathrm{H}_{2}\mathrm{SO}_{4}$ , and $\mathrm{HNO}_{3}$ ; this can accelerate $\mathrm{SO}_{2}$ oxidation and increase the formation of $\mathrm{H}_{2}\mathrm{SO}_{4}$ in the liquid phase. To summarize, the warm and humid climate at Sanya likely supported heterogeneous reactions that led to the formation of sulfate, and the high acidity can be explained by the high aerosol $\mathrm{SO}_{4}{}^{2-}$ combined with the limited local $\mathrm{NH}_{3}$ emissions and high RHs (on average $91.9\%$ ).
The E-AIM model IV (E-AIM4) only considers the major water-soluble ions in the atmosphere; these include $\mathrm{NH_{4}}^{+}$ , $\mathrm{Na^{+}}$ , $\mathrm{SO}_{4}{}^{2-}$ , $\mathrm{NO}_{3}^{\phantom{\,}-}$ , and $\mathrm{Cl}^{-}$ but not $\mathrm{{Ca}}^{2+}$ , $\ K^{+}$ , or $\mathrm{Mg}^{2+}$ . However, these latter ions cannot be ignored at the Sanya site, and therefore, the acidity estimated by E-AIM4 may not be accurate. A linear equation between cation and anion is given as
$$
[\mathrm{Cation}]=0.68\times[\mathrm{Anion}]+0.001
$$
by using the formula
$$
\begin{array}{c}{{\mathrm{Cation\equivalent}=\displaystyle\frac{\mathrm{Na}^{+}}{23}+\frac{\mathrm{NH}_{4}^{+}}{18}+\frac{\mathrm{K}^{+}}{39}+\frac{\mathrm{Mg}^{2+}}{12}\ }}\\ {{+\displaystyle\frac{\mathrm{Ca}^{2+}}{20}}}\\ {{\mathrm{Anion\equivalent}=\displaystyle\frac{\mathrm{SO}_{4}^{2-}}{47}+\frac{\mathrm{NO}_{3}^{-}}{62}+\frac{\mathrm{Cl}^{-}}{35.5}+\frac{\mathrm{Br}^{-}}{80}+\frac{\mathrm{F}^{-}}{19}}}\end{array}
$$
where [cation] and [anion] are expressed in normality. Calculation of the cation and anion balance was used to support for the findings regarding aerosol acidity in Sanya City.
Moreover, the acidity ratio (Engelhart et al. 2011) or neutralization ratio (NR) (Bencs et al. 2008) can be defined as
$$
\mathrm{NR}=\left(\left[\mathrm{NH}_{4}^{+}\right]\right)/\left[\mathrm{SO}_{4}^{2-}\right]+\left[\mathrm{NO}_{3}^{-}\right]
$$
which is a way of expressing the degree of neutralization of sulfate and nitrate by ammonium (expressed in equivalents). That is, NR indicates the aerosol acidity characteristics by accounting for the possible neutralization of only the two major inorganic acids (nitric and sulfuric) with ammonium. In $\mathrm{PM}_{2.5}$ , $\mathrm{NR}<1$ typically indicates the partial neutralization of acidic aerosols. The NR in Sanya was much less than unity, indicating limited neutralization, and it also lower than the values calculated for other megacities (Pathak et al. 2009; Fu et al. 2015; Wu et al. 2017). Various atmospheric chemical and physical processes affect the acidity of the aerosol. For example, at Beijing and Shanghai, $\mathrm{NH}_{3}$ neutralized $40-$ $50\%$ of $\mathrm{SO}_{4}{}^{2-}$ and $\mathrm{NO}_{3}^{\phantom{\,+}}$ , and a complete neutralization was seen in Lanzhou and Guangzhou (Pathak et al. 2009). Such neutralization evidently did not occur at Sanya because NR was only 0.18 due to an insufficient supply of ammonia.
The NR ratio was highly correlated with the free acidity $(R^{2}=0.77)$ as shown in Fig. 4, and the degree of neutralization increased with free acidity. In contrast, Fu et al. (2015) and $\mathrm{Wu}$ et al. (2017) found that NR values decreased when free acidities increased at Guangzhou and Xiamen in winter, and furthermore, the relationship for those sites was logarithmic. Sanya, Guangzhou, and Xiamen are all near the sea, but the local pollution emissions at the latter two sites are much higher than those in Sanya, and this is a likely reason the aerosol acidity differed among the sites. Positive correlations between the degree of neutralization and free acidity were likely the result of the conversion of $\mathrm{SO}_{2}$ to $\mathrm{SO}_{4}{}^{2-}$ through aqueous reactions, but as discussed below, there it is also likely that the correlations were driven by regional emission and transported to the sampling site.
Regional scale transport
Clusters of air-mass trajectories
Even though local pollution sources, such as motor vehicle emissions, normally are the main sources for the aerosol at Sanya (Wang et al. 2015), potential impacts from regional emissions are worth evaluating. Pollutant transport to the site is mainly driven by synoptic winds, local sea-land breeze circulation, and orographic effects. Three-day air-mass trajectories, calculated backwards in time and reaching at Sanya at $500\,\textrm{m}$ above ground level were calculated for every hour during the campaign by using the National Oceanic and Atmospheric Administration (NOAA) Air Resource Lab (ARL) Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model and gridded meteorological data (Global Data Assimilation System, GDAS1) (www.arl. noaa.gov/HYSPLIT_pubs.php) (Stein et al. 2015).
Most of the air masses came from east and northeast during the sampling period (Fig. 5), and this is consistent with the wintertime climatology. The underlying premise for evaluating transport pathways is that air masses pass over potential source areas and accumulate aerosols and their precursors before they arrive at the receptor site. The air masses were grouped into three clusters according to their origin (Fig. 5). The average concentrations of chemical species and parameters for the three clusters are summarized in Table 3.
As shown in Fig. 5, 16 trajectories were grouped into cluster 1: this group primarily originated over South Asia and South China, and it had the highest average $\operatorname{PM}_{2.5}$ concentration $(26.5\ \upmu\mathrm{g}\ \mathrm{m}^{-3})$ ) of the three clusters. There were ten trajectories associated with cluster 2 which primarily originated from eastern coastal regions and the East China Sea, and these had the lowest average $\operatorname{PM}_{2.5}$ concentration of the clusters ( $\cdot13.9\ \upmu\mathrm{g}\ \mathrm{m}^{-3})$ ). The remaining six trajectories were grouped into cluster 3, and that group, which had an intermediate $\mathrm{PM}_{2.5}$ loading of $15.2~\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , originated from the South China Sea or the ocean and then passed over the South China Sea.
The trajectories were classified as Bpolluted^ if the $\operatorname{PM}_{2.5}$ concentration for the sample matched to the trajectory (you might briefly describe how were the samples matched to the trajectories) was higher than the average value of $20.4\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ . Otherwise, if the $\operatorname{PM}_{2.5}$ loading for a sample was below the average the trajectory was identified as Bnormal.^ Based on this approach, 13 days with trajectories in cluster 1 were classified as Bpolluted.^ No polluted day were found for clusters 2 or 3, demonstrating that the air quality was better when the flow was from eastern coastal regions or the South China Sea. This supports our conclusion that the air masses that passed over southern Asia and South China were the main regional pollution sources for Sanya.
In addition to the $\operatorname{PM}_{2.5}$ mass loadings, the concentrations of various chemical species showed different rankings among the three clusters. Cluster 1, which had all of the Bpolluted^ trajectories, had the highest concentrations of OC and secondary inorganic aerosol (SIA) compounds such as $\mathrm{SO}_{4}{}^{2-}$ , $\mathrm{NO}_{3}^{\mathrm{~-~}}$ , and $\mathrm{NH_{4}}^{+}$ , and the highest values for $[\mathrm{H}^{+}]_{\mathrm{Total}}.$ $[\mathrm{H}^{+}]_{\mathrm{Free}}$ and NR. The high concentrations of these species show that the transport pathways represented by this cluster were associated with the worst air quality at the receptor site, and those effects can be traced to pollutant emissions from southern Asia. Meanwhile, cluster 3 had the highest concentration of $\mathrm{Na^{+}}$ and ${\mathrm{Cl}}^{-}$ , which are major components of sea salt, and therefore, this cluster appears the most representative of marine air. Dimethyl sulfide (DMS), produced by marine phytoplankton, is one of the major natural sources of atmospheric sulfate, and methanesulfonic acid (MSA) and non-sea salt of $\mathrm{SO}_{4}^{\ 2-}$ , which are oxidation products of DMS, are two major contributors to natural aerosol acidity (Andreae and Barnard 1984; Nguyen et al. 1992; Yang et al. 2011). The mixing of marine air with the local aerosols likely led to the comparatively high concentrations of $\mathrm{SO}_{4}^{\ 2-}$ , $[\mathrm{H}^{+}]_{\mathrm{Total}}.$ and $[\mathrm{H}^{+}]_{\mathrm{Free}}$ in cluster 3, but their sources and ways in which they formed were different from those in cluster 1.
The lengths of the trajectories are indicative of transport speed; that is, a longer pathway equates to a faster transport. The trajectories grouped into cluster 2 passed over the Eastern China and the East China Sea, and they had the longer lengths compared with those in the other two clusters. The trajectories in cluster 2 also were associated with cold air and strong winds, which would be conducive to convection and promote the dispersal of pollutants, leading to good air quality in Sanya. Cluster 2 would be subject to atmospheric removal process like dry and wet depositions during long-range transport. Overall, the $\operatorname{PM}_{2.5}$ aerosol in Sanya was incompletely neutralized and thus acidic, but the samples for clusters 2 and 3 had lower neutralization ratios than those in cluster 1, indicating relatively less $\mathrm{NH}_{3}$ and $\mathrm{NH_{4}}^{+}$ for those pathways.
A case study of high $\mathsf{P M}_{2.5}$
To investigate the sources for $\operatorname{PM}_{2.5}$ , we focused on an episode at Sanya on 31 January 2012 and the day before the episode and day after it. The filter samples for the 31 January episode were assigned to cluster 1, and the concentrations of pollutants in that sample were significantly higher compared with the averages for the other sampling days. The increases over average were as follows: $\mathrm{NH_{4}}^{+}\,(404\%)\!>\!\mathrm{SO_{4}}^{2-}$ $(167\%)>\mathrm{OC}$ $(140\%)>\mathrm{K}^{+}$ $(108\%)>\mathrm{NO}_{3}^{\;\,-}$ $(102\%)>\mathrm{EC}$ $(85.9\%)>\mathrm{PM}_{2.5}$ mass $(72.7\%)$ . Three-day back trajectories for the episode showed the air masses originated from the middle of China, passed along with Pearl River Delta (PRD) region, and finally arrived at Sanya on the day of the episode. To apportion the potential pollution sources for this episode, a simplified Weather Research and Forecasting chemistry (WRF-Chem) model was run in conjunction with the (WRF) model. WRF is a numerical weather prediction system and widely used in both operational forecasting and atmospheric research. Detailed descriptions of the WRF model have been reported elsewhere (www.wrf-model.org/index. php). In addition to the standard features, EC has been added as a chemical tracer in the WRF-EC model (Zhao et al. 2015b), and it was used to track the long-range transport of EC from source regions to the sample area.
EC is useful as a tracer for atmospheric transport because it is directly emitted from combustion sources and has relatively low chemical reactivity (Zhao et al. 2015b; Cao et al. 2013). The complex chemical schemes for gases and aerosols in the standard WRF-Chem model rely on modules that perform online calculations of dynamic inputs (e.g., winds, temperature, boundary layer); adjective, convective, and diffusive transport; dry deposition; tracer particles; and surface emissions. The newly added calculations for EC use an emission inventory obtained from (Streets et al. 2003). The horizontal resolution of the model was set at $3\times3~\mathrm{km}$ in a $450\times300~\mathrm{km}$ domain centered on Sanya for our study.
The EC simulated with the WRF-EC model was generally consistent with the measured EC, suggesting the WRF-EC model was able to capture the sources and processes responsible for the variability in the concentrations of this material. Nonetheless, the simulated values underestimated the measured concentrations by $\sim\!20\%$ , and that was probably due to the effects of local meteorological conditions and uncertainties in the EC emission database for Hainan. The spatial distributions of horizontal winds at $200~\mathrm{m}$ above the ground for the episode and the days before and after are shown in Fig. 6a. The prevailing winds over South China and the South China Sea were northeasterly on 30 January (first day) and 31 January (episode day). The simulated EC concentration on the first day was lower than the episode day in upstream areas and the PRD region was downwind. On 1 February, the day after the event, the simulated EC concentration in Sanya was the highest, but the winds to Sanya on that day came from east and passed over the South China Sea, which would have added little to the EC aerosol. As a result, higher concentrations of cluster 1 pollutants were measured in Sanya on the second day of the case study compared to those on the third day. More important, the modelbased analysis indicates that emissions from Southern China, especially from the PRD region, were as the main source for the EC and other pollutants during the episode.
The contributions of local and regional emission to the pollution aerosol at Sanya also were estimated with the use of the WRF-BC model. Figure 6b, c depicts the simulated EC concentrations without emissions from Hainan and South China, respectively, during the pollution episode. The contribution of local primary emissions was only $18.1\%$ of the total EC whereas emissions from areas to northeast of Sanya accounted for $76.8\%$ of EC during the episode totally explained ca. $95\%$ of the at least aerosol pollution origin indicating that the main source during this episode is the primary emission pollutants from Hainan and South China.
Trace organic marker species, such as PAHs, can be used to evaluate regional impacts on air quality; indeed, vehicular emissions have been recognized as a common contributor to atmospheric pollution at both Sanya and the PRD region (Gao et al. 2012; Wang et al. 2015). Of the 3 days in the case study, the highest concentrations of benzo[a]pyrene (BaP), benzo[e]pyrene (BbP), and dibenzo[a,h]anthracene (DahA) were found on the day of the pollution episode. The PAHs most likely originated from industrial sources in the Guangdong Province in the PRD region (Wang et al. 2016), and benzo[k]fluoranthene (BfF), a marker for coal combustion (Ravindra et al. 2008), exhibited its highest concentration on 31 January $(0.69\,\mathrm{ng}\,\mathrm{m}^{-3})$ ) compared with 0.21 and $0.42\,\mathrm{ng}\,\mathrm{m}^{-3}$ on the other 2 days of the case study. In addition, the diagnostic ratio of indeno [1,2,3-cd]pyrene (IcdP)/ $\mathrm{(IcdP\+}$ benzo[ghi]perylene (BghiP)) was 0.52 on the episode day, and that was higher than the values $<0.50$ on the first and third days of the case study. A combination of pollution emissions, such as those from wood fires, grass fires, and coal burning, is indicated for $\mathrm{IcdP/(IcdP+BghiP)}$ ratios $>0.5$ . Interestingly, the highest total quantified concentrations of PAHs and $n$ -alkanes were found on the first day of the case study, not the episode day itself, and this demonstrates that the individual or diagnostic ratios of organic markers for specific source are useful for identification and tracking. The analysis of the case study provides further proof that transport from the PRD region can significantly impact the air quality at Sanya when the prevailing winds are from the northeast.
Conclusions
The chemical composition of $\operatorname{PM}_{2.5}$ was measured and a mass balance of the $\operatorname{PM}_{2.5}$ from Sanya was calculated to provide an overall assessment of the air quality during winter at the southernmost city on Hainan Island. The concentrations of typical pollutants were lower than in most Chinese cities due to geographical and climate conditions at Sanya and less industrial development. The low proportion of $\mathrm{NH_{4}}^{+}$ in $\operatorname{PM}_{2.5}$ suggested that the pollution sources were different from those in other areas and the main chemical reactions leading to SIA formation also were unlike those in some large cities. A thermodynamic model indicated that the aerosol acidities were relatively high and that acidity promoted heterogeneous reactions that led to aerosol $\mathrm{SO}_{4}^{\ 2-}$ . Three major transport pathways were evaluated based on clusters of 3-day air-mass back trajectories. The air parcels that arrived at the site were predominately from the northeast, and the characteristics of the aerosol and pollution impacts varied among transport pathways. Transport from the South China, which was likely affected by emissions from the PRD region, brought polluted air to the site, and both regional influences and impacts from local emissions were evident. The WRF-EC model showed that regional emissions and transport was responsible for the high EC loadings, and the trajectories that passed over northeast Hainan province contributed $76.8\%$ to the EC in $\operatorname{PM}_{2.5}$ . The evaluation based on trace organic markers supports the modeling results and demonstrates a close linkage between the pollutant loadings and transport from PRD region. One important implication of our finding is that maintaining or improving the air quality at clean sites such as Sanya will not only require effective controls on local emissions but also controls on distant sources. The results presented here were for a short sampling campaign, and more extensive measurements are needed to determine fully appreciate the relative influences of local emissions and atmospheric transport on air quality: a long-term study on the effects of the Asian monsoon and related large-scale processes would be particularly valuable. A more in-depth evaluation of the atmospheric reaction mechanisms between gases and particulate matter also should be performed with emphasis on their relations to air quality, regional environment, and climate.
Funding information This research is supported by the project from the BStrategic Priority Research Program^ of the Chinese Academy of Science (Grant No. XDB05060500) and the project from Ministry of Science and Technology (2013FY112700). It is also supported by the Natural Science Foundation of Hainan Province, China (Grant No. 417151), The Doctoral Scientific Research Foundation of Hainan Tropical Ocean University (Grant No. RHDXB201613), Educational Reform of Hainan Tropical Ocean University (Grant No. RDJGb2016- 18), and Key Discipline Construction Program of Hainan Provincial Department of Education (Marine Geology-2017).
References
Andreae MO, Barnard WR (1984) The marine chemistry of dimethylsulfide. Mar Chem 14(3):267–279. https://doi.org/10.
1016/0304-4203(84)90047-1 Andrews E, Saxena P, Musarra S, Hildemann LM, Koutrakis P, McMurry PH, Olmez I, White WH (2011) Concentration and composition of atmospheric aerosols from the 1995 SEAVS experiment and a review of the closure between chemical and gravimetric measurements. J Air Waste Manage Assoc 50(5):648–664. https://doi.org/
10.1080/10473289.2000.10464116 Battye W, Anejia VP, Roelle PA (2003) Evaluation and improvement of ammonia emissions inventories. Atmos Environ 37(27):3873–3883. https://doi.org/10.1016/S1352-2310(03)00343-1 Bencs L, Ravindra K, de HJ, Rasoazanany EO, Deutsch F, Bleux N, Berghmans P, Roekens E, Krata A, van Grieken R (2008) Mass and ionic composition of atmospheric fine particles over Belgium and their relation with gaseous air pollutants. J Environ Monit
10(10):1148–1157 Cao JJ, Lee SC, Ho KF, Zhang XY, Zou SC, Fung K, Chow JC, Watson JG (2003) Characteristics of carbonaceous aerosol in Pearl River Delta region, China during 2001 winter period. Atmos Environ 37(11):1451–1460. https://doi.org/10.1016/ S1352-2310(02)01002-6
Cao JJ, Shen ZX, Chow JC, Watson JG, Lee SC, Tie XX, Ho KF, Wang GH, Han YM (2012) Winter and summer $\mathrm{PM}_{2.5}$ chemical compositions in fourteen Chinese cities. J Air Waste Manag Assoc 62(10): 1214–1226. https://doi.org/10.1080/10962247.2012.701193
Cao J-J, Zhu C-S, Tie X-X, Geng F-H, Xu H-M, Ho SSH, Wang G-H, Han Y-M, Ho K-F (2013) Characteristics and sources of carbonaceous aerosols from Shanghai, China. Atmos Chem Phys 13(2): 803–817. https://doi.org/10.5194/acp-13-803-2013
Chandler AS, Choularton TW, Dollard GJ, Eggleton AEJ, Gay MJ, Hill TA, Jones BMR, Tyler BJ, Bandy BJ, Penkett SA (1988) Measurements of $\mathrm{H}_{2}\mathrm{O}_{2}$ and $\mathrm{SO}_{2}$ in clouds and estimates of their reaction rate. Nature 336(6199):562–565. https://doi.org/10.1038/ 336562a0
Chen X, Yu JZ (2007) Measurement of organic mass to organic carbon ratio in ambient aerosol samples using a gravimetric technique in combination with chemical analysis. Atmos Environ 41(39):8857– 8864. https://doi.org/10.1016/j.atmosenv.2007.08.023
Chow JC, Watson JG (1999) Ion chromatography in elemental analysis of airborne particles. In: Landsberger S, Creatchman M (eds) Elemental analysis of airborne particles. Gordon and Breach Science, Amsterdam, vol 1, pp 97–137
Chow JC, Watson JG, Pritchett LC, Pierson WR, Frazier CA, Purcell RG (1993) The dri thermal/optical reflectance carbon analysis system. Description, evaluation and applications in U.S. air quality studies. Atmos Environ Part A Gen Top 27(8):1185–1201. https://doi.org/ 10.1016/0960-1686(93)90245-T
Chow JC, Watson JG, Chen LWA, Arnott WP, Moosmüller H, Fung K (2004) Equivalence of elemental carbon by thermal/optical reflectance and transmittance with different temperature protocols. Environ Sci Technol 38(16):4414–4422
Chow JC, Yu JZ, Watson JG, Ho SSH, Bohannan TL, Hays MD, Fung KK (2007) The application of thermal methods for determining chemical composition of carbonaceous aerosols: a review. J Environ Sci Health A Tox Hazard Subst Environ Eng 42(11): 1521–1541. https://doi.org/10.1080/10934520701513365
Clegg SL, Brimblecombe P, Wexler AS (1998) Thermodynamic model of the system $\mathrm{H^{+}\mathrm{-NH_{4}^{+}\mathrm{-SO_{4}^{2-}\mathrm{-NO_{3}^{-}\mathrm{-H_{2}O}}}}}$ at tropospheric temperatures. J Phys Chem A 102(12):2137–2154. https://doi.org/10.1021/ jp973042r
Cui S, Shi Y, Groffman PM, Schlesinger WH, Zhu Y-G (2013) Centennial-scale analysis of the creation and fate of reactive nitrogen in China (1910-2010). Proc Natl Acad Sci U S A 110(6):2052– 2057. https://doi.org/10.1073/pnas.1221638110
Das M, Aneja VP (1994) Measurements and analysis of concentrations of gaseous hydrogen peroxide and related species in the rural Central Piedmont region of North Carolina. Atmos Environ 28(15):2473– 2483. https://doi.org/10.1016/1352-2310(94)90398-0
El-Zanan HS, Lowenthal DH, Zielinska B, Chow JC, Kumar N (2005) Determination of the organic aerosol mass to organic carbon ratio in IMPROVE samples. Chemosphere 60(4):485–496. https://doi.org/ 10.1016/j.chemosphere.2005.01.005
El-Zanan HS, Zielinska B, Mazzoleni LR, Hansen DA (2012) Analytical determination of the aerosol organic mass-to-organic carbon ratio. J Air Waste Manag Assoc 59(1):58–69. https://doi.org/10.3155/1047- 3289.59.1.58
Engelhart GJ, Hildebrandt L, Kostenidou E, Mihalopoulos N, Donahue NM, Pandis SN (2011) Water content of aged aerosol. Atmos Chem Phys 11(3):911–920. https://doi.org/10.5194/acp-11-911-2011
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 PM2.5 at a regional background site of the Yangtze River Delta region, China. Atmos Environ 123:288–297. https://doi.org/10.1016/j.atmosenv. 2015.08.019
Friese E, Ebel A (2010) Temperature dependent thermodynamic model of the system $\mathrm{H^{+}}\mathrm{-NH_{4}^{+}}\mathrm{-Na^{+}}\mathrm{-SO_{4}^{2-}}\mathrm{-NO_{3}^{-}}$ -Cl-- $\cdot\mathrm{H}_{2}\mathrm{O}$ . J Phys Chem A 114(43):11595–11631. https://doi.org/10.1021/jp101041j
Fu X, Guo H, Wang X, Ding X, He Q, Liu T, Zhang Z (2015) $\operatorname{PM}_{2.5}$ acidity at a background site in the Pearl River Delta region in fallwinter of 2007-2012. J Hazard Mater 286:484–492. https://doi.org/ 10.1016/j.jhazmat.2015.01.022
Fung CS, Misra PK, Bloxam R, Wong S (1991) A numerical experiment on the relative importance of $\mathrm{H}_{2}\mathrm{O}_{2}$ , $\mathrm{O}_{3}$ in aqueous conversion of SO2 to SO42-. Atmos Environ Part A Gen Top 25(2):411–423. https://doi.org/10.1016/0960-1686(91)90312-U
Gao B, Guo H, Wang X-M, Zhao X-Y, Ling Z-H, Zhang Z, Liu T-Y (2012) Polycyclic aromatic hydrocarbons in $\operatorname{PM}_{2.5}$ in Guangzhou, southern China: spatiotemporal patterns and emission sources. J Hazard Mater 239-240:78–87. https://doi.org/10.1016/j.jhazmat. 2012.07.068
Gu B, Sutton MA, Chang SX, Ge Y, Chang J (2014) Agricultural ammonia emissions contribute to China’s urban air pollution. Front Ecol Environ 12(5):265–266. https://doi.org/10.1890/14.WB.007
Guan D, Su X, Zhang Q, Peters GP, Liu Z, Lei Y, He K (2014) The socioeconomic drivers of China’s primary PM 2.5 emissions. Environ Res Lett 9(2):24010. https://doi.org/10.1088/1748-9326/9/ 2/024010
Guo S, Hu M, Zamora ML, Peng J, Shang D, Zheng J, Du Z, Wu Z, Shao M, Zeng L, Molina MJ, Zhang R (2014) Elucidating severe urban haze formation in China. Proc Natl Acad Sci U S A 111(49):17373– 17378. https://doi.org/10.1073/pnas.1419604111
Guo J, Xia F, Zhang Y, Liu H, Li J, Lou M, He J, Yan Y, Wang F, Min M, Zhai P (2017) Impact of diurnal variability and meteorological factors on the PM2.5-AOD relationship. Implications for PM2.5 remote sensing. Environ Pollut 221:94–104. https://doi.org/10.1016/ j.envpol.2016.11.043
He SZ, Chen ZM, Zhang X, Zhao Y, Huang DM, Zhao JN, Zhu T, Hu M, Zeng LM (2010) Measurement of atmospheric hydrogen peroxide and organic peroxides in Beijing before and during the 2008 Olympic Games. Chemical and physical factors influencing their concentrations. J Geophys Res 115(D17):459. https://doi.org/10. 1029/2009JD013544
Ho SSH, Yu JZ (2004) In-injection port thermal desorption and subsequent gas chromatography-mass spectrometric analysis of polycyclic aromatic hydrocarbons and n-alkanes in atmospheric aerosol samples. J Chromatogr A 1059(1–2):121–129
Ho SSH, Yu JZ, Chow JC, Zielinska B, Watson JG, Sit EHL, Schauer JJ (2008) Evaluation of an in-injection port thermal desorption-gas chromatography/mass spectrometry method for analysis of nonpolar organic compounds in ambient aerosol samples. J Chromatogr A 1200(2):217–227. https://doi.org/10.1016/j.chroma. 2008.05.056
Ho SSH, Chow JC, Watson JG, Ting Ng LP, Kwok Y, Ho KF, Cao J (2011) Precautions for in-injection port thermal desorption-gas chromatography/mass spectrometry (TD-GC/MS) as applied to aerosol filter samples. Atmos Environ 45(7):1491–1496. https:// doi.org/10.1016/j.atmosenv.2010.12.038
Hong YM, Lee BK, Park KJ, Kang MH, Jung YR, Lee DS, Kim MG (2002) Atmospheric nitrogen and sulfur containing compounds for three sites of South Korea. Atmos Environ 36(21):3485–3494. https://doi.org/10.1016/S1352-2310(02)00289-3
IPCC (2013) Climate Change 2013: the physical science basis. Contribution of working group I to the fifth assessment report of the intergovernmental panel on climate change. Cambridge University Press, Cambridge. https://doi.org/10.1017/ CBO9781107415324
Keene WC, Pszenny APP, Galloway JN, Hawley ME (1986) Seasalt corrections and interpretation of constituent ratios in marine precipitation. J Geophys Res 91:6647–6658
Li L, Wang W, Feng J, Zhang D, Li H, Gu Z, Wang B, Sheng G, Fu J (2010) Composition, source, mass closure of PM2.5 aerosols for four forests in eastern China. J Environ Sci 22(3):405–412. https:// doi.org/10.1016/S1001-0742(09)60122-4
Li XB, Huang H, Lian JS, Liu S, Huang LM, Yang JH (2013) Spatial and temporal variations in sediment accumulation and their impacts on coral communities in the Sanya Coral Reef Reserve, Hainan, China. Deep Sea Res Pt II 96:88–96. https://doi.org/10.1016/j.dsr2.2013. 04.015
Li L, Yin Y, Kong S, Wen B, Chen K, Yuan L, Li Q (2014) Altitudinal effect to the size distribution of water soluble inorganic ions in PM at Huangshan, China. Atmos Environ 98:242–252. https://doi.org/10. 1016/j.atmosenv.2014.08.077
Lippmann M, Xiong JQ, Li W (2000) Development of a continuous monitoring system for $\mathrm{PM}_{10}$ and components of $\mathrm{PM}_{2.5}$ . Appl Occup Environ Hyg 15(1):57–67. https://doi.org/10.1080/ 104732200301854
Liu D, Li J, Zhang Y, Xu Y, Liu X, Ding P, Shen C, Chen Y, Tian C, Zhang G (2013) The use of levoglucosan and radiocarbon for source apportionment of PM(2.5) carbonaceous aerosols at a background site in East China. Environ Sci Technol 47(18):10454–10461. https://doi.org/10.1021/es401250k
Meagher JF, Olszyna KJ, Weatherford FP, Mohnen VA (1990) The availability of $\mathrm{H}_{2}\mathrm{O}_{2}$ and $\mathrm{O}_{3}$ for aqueous phase oxidation of $\mathrm{SO}_{2}$ . The question of linearity. Atmos Environ Part A Gen Top 24(7):1825– 1829. https://doi.org/10.1016/0960-1686(90)90514-N
Meier PC, Zünd RE (2005) Statistical methods in analytical chemistry, 2nd edn. Chemical analysis, vol 153. Wiley, New York
Nguyen BC, Mihalopoulos N, Putaud JP, Gaudry A, Gallet L, Keene WC, Galloway JN (1992) Covariations in oceanic dimethyl sulfide, its oxidation products and rain acidity at Amsterdam Island in the southern Indian Ocean. J Atmos Chem 15(1):39–53. https://doi. org/10.1007/BF00053608
Niu Z, Zhang F, Chen J, Yin L, Wang S, Xu L (2013) Carbonaceous species in PM2.5 in the coastal urban agglomeration in the Western Taiwan Strait region, China. Atmos Res 122:102–110. https://doi.org/10.1016/j.atmosres.2012.11.002
Ohta S, Okita T (1990) A chemical characterization of atmospheric aerosol in Sapporo. Atmos Environ Part A Gen Top 24(4):815–822. https://doi.org/10.1016/0960-1686(90)90282-R
Pathak RK, Louie PKK, Chan CK (2004) Characteristics of aerosol acidity in Hong Kong. Atmos Environ 38(19):2965–2974. https://doi. org/10.1016/j.atmosenv.2004.02.044
Pathak RK, Wu WS, Wang T (2009) Summertime $\mathrm{PM}_{2.5}$ ionic species in four major cities of China. Nitrate formation in an ammoniadeficient atmosphere. Atmos Chem Phys 9(5):1711–1722. https:// doi.org/10.5194/acp-9-1711-2009
Pathak RK, Wang T, Ho KF, Lee SC (2011) Characteristics of summertime $\mathrm{PM}_{2.5}$ organic and elemental carbon in four major Chinese cities. Implications of high acidity for water-soluble organic carbon (WSOC). Atmos Environ 45(2):318–325. https://doi.org/10.1016/j. atmosenv.2010.10.021
Pierson WR, Brachaczek WW (1983) Emissions of ammonia and amines from vehicles on the road. Environ Sci Technol 17(12):757–760. https://doi.org/10.1021/es00118a013
Pu W, Zhao X, Shi X, Ma Z, Zhang X, Yu B (2015) Impact of long-range transport on aerosol properties at a regional background station in Northern China. Atmos Res 153:489–499. https://doi.org/10.1016/j. atmosres.2014.10.010
Qiao YT, Zhang CH, Jian MQ (2015) Role of the 10–20-Day Oscillation in sustanined rainstorms over Hainan, China in October 2010. Adv Atmos Sci 32(3):363–374
Ravindra K, Sokhi R, Grieken RV (2008) Atmospheric polycyclic aromatic hydrocarbons: source attribution, emission factors and regulation. Atmos Environ 42(13):2895–2921
Seinfeld JH, Pandis SN (2006) Atmospheric Chemistry and Physics: from air pollution to climate change, 2nd edition. J Wiley, New York
Shen Z, Cao J, Arimoto R, Han Z, Zhang R, Han Y, Liu S, Okuda T, Nakao S, Tanaka S (2009a) Ionic composition of TSP and $\operatorname{PM}_{2.5}$ during dust storms and air pollution episodes at Xi’an, China. Atmos Environ 43(18):2911–2918. https://doi.org/10.1016/j.atmosenv. 2009.03.005
Shen ZX, Cao JJ, Tong Z, Liu SX, Reddy LSS, Han Y, Zhang T, Zhou J (2009b) Chemical characteristics of submicron particles in winter in Xi’an. Aerosol Air Qual Res:80–93. https://doi.org/10.4209/aaqr. 2008.10.0050
Squizzato S, Masiol M, Brunelli A, Pistollato S, Tarabotti E, Rampazzo G, Pavoni B (2013) Factors determining the formation of secondary inorganic aerosol. A case study in the Po Valley (Italy). Atmos Chem Phys 13(4):1927–1939. https://doi.org/10.5194/acp-13- 1927-2013
Stein AF, Draxler RR, Rolph GD, Stunder BJB, Cohen MD, Ngan F (2015) NOAA’s HYSPLIT atmospheric transport and dispersion modeling system. Bull Am Meteorol Soc 96(12):2059–2077. https://doi.org/10.1175/BAMS-D-14-00110.1
Streets DG, Bond TC, Carmichael GR, Fernandes SD, Fu Q, He D, Klimont Z, Nelson SM, Tsai NY, Wang MQ, Woo J-H, Yarber KF (2003) An inventory of gaseous and primary aerosol emissions in Asia in the year 2000. J Geophys Res 108(D21):213. https://doi.org/ 10.1029/2002JD003093
Sutton MA, Dragosits U, Tang YS, Fowler D (2000) Ammonia emissions from non-agricultural sources in the UK. Atmos Environ 34(6):855– 869. https://doi.org/10.1016/S1352-2310(99)00362-3
Sutton MA, Erisman JW, Dentener F, Möller D (2008) Ammonia in the environment: from ancient times to the present. Environ Pollut 156(3):583–604. https://doi.org/10.1016/j.envpol.2008.03.013
Thurston GD, Spengler JD (1985) A quantitative assessment of source contributions to inhalable particulate matter pollution in metropolitan Boston. Atmos Environ (1967) 19(1):9–25. https://doi.org/10. 1016/0004-6981(85)90132-5
Walker JT, Whitall DR, Robarge W, Paerl HW (2004) Ambient ammonia and ammonium aerosol across a region of variable ammonia emission density. Atmos Environ 38(9):1235–1246. https://doi.org/10. 1016/j.atmosenv.2003.11.027
Wang JZ, Ho SSH, Cao JJ, Huang RJ, Zhou JM, Zhao YZ, Xu HM, Liu SX, Wang GH, Shen ZX, Han YM (2015) Characteristics and major sources of carbonaceous aerosols in $\operatorname{PM}_{2.5}$ from Sanya, China. Sci Total Environ 530-531:110–119. https://doi.org/10.1016/j.scitotenv. 2015.05.005
Wang J, Ho SSH, Ma S, Cao J, Dai W, Liu S, Shen Z, Huang R, Wang G, Han Y (2016) Characterization of $\mathrm{PM}_{2.5}$ in Guangzhou, China: uses of organic markers for supporting source apportionment. Sci Total Environ 550:961–971. https://doi.org/10.1016/j. scitotenv.2016.01.138
Watson JG (2002) Visibility: science and regulation. J Air Waste Manag Assoc 52(6):628–713
Watson JG, Chow JC, Frazier CA (1999) X-ray fluorescence analysis of ambient air samples. In: Landsberger S, Creatchman M (eds) Elemental analysis of airborne particles. Gordon and Breach Science, Amsterdam, vol 1, pp 67–96
Wilson LJ, Bacon PJ, Bull J, Dragosits U, Blackall TD, Dunn TE, Hamer KC, Sutton MA, Wanless S (2004) Modelling the spatial distribution of ammonia emissions from seabirds in the UK. Environ Pollut 131(2):173–185. https://doi.org/10.1016/j.envpol.2004.02.008
Wu Y, Gu B, Erisman JW, Reis S, Fang Y, Lu X, Zhang X (2016) $\operatorname{PM}_{2.5}$ pollution is substantially affected by ammonia emissions in China. Environ Pollut 218:86–94. https://doi.org/10.1016/j.envpol.2016. 08.027
Wu X, Deng JJ, Chen JS, Hong YW, Xu LL, Yin LQ, Du WJ, Hong ZY, Dai NZ, Yuan C-S (2017) Characteristics of water-soluble inorganic components and acidity of $\operatorname{PM}_{2.5}$ in a coastal city of China. Aerosol Air Qual Res 17(9):2152–2164. https://doi.org/10.4209/aaqr.2016. 11.0513
Xu HM, Cao JJ, Ho KF, Ding H, Han YM, Wang GH, Chow JC, Watson JG, Khol SD, Qiang J, Li WT (2012) Lead concentrations in fine particulate matter after the phasing out of leaded gasoline in Xi’an, China. Atmos Environ 46:217–224. https://doi.org/10.1016/j. atmosenv.2011.09.078
Xu L, Yu Y, Yu J, Chen J, Niu Z, Yin L, Zhang F, Liao X, Chen Y (2013) Spatial distribution and sources identification of elements in PM2.5 among the coastal city group in the Western Taiwan Strait region, China. Sci Total Environ 442:77–85. https://doi.org/10.1016/j. scitotenv.2012.10.045
Xu H, Cao J, Chow JC, Huang R-J, Shen Z, Chen LWA, Ho KF, Watson JG (2016) Inter-annual variability of wintertime $\mathrm{PM}_{2.5}$ chemical composition in Xi’an, China: evidences of changing source emissions. Sci Total Environ 545-546:546–555. https://doi.org/10.1016/ j.scitotenv.2015.12.070
Yang G-P, Zhang H-H, Zhou L-M, Yang J (2011) Temporal and spatial variations of dimethylsulfide (DMS) and dimethylsulfoniopropionate (DMSP) in the East China Sea and the Yellow Sea. Cont Shelf Res 31(13):1325–1335. https://doi. org/10.1016/j.csr.2011.05.001
Yin L, Niu Z, Chen X, Chen J, Zhang F, Xu L (2014) Characteristics of water-soluble inorganic ions in PM2.5 and PM 2.5-10 in the coastal urban agglomeration along the Western Taiwan Strait region, China. Environ Sci Pollut Res Int 21(7):5141–5156. https://doi.org/10. 1007/s11356-013-2134-7
Zhang Y, Seigneur C, Seinfeld JH, Jacobson M, Clegg SL, Binkowski FS (2000) A comparative review of inorganic aerosol thermodynamic equilibrium modules. Similarities, differences, and their likely causes. Atmos Environ 34(1):117–137. https://doi.org/10.1016/ S1352-2310(99)00236-8
Zhang XY, Gong SL, Shen ZX, Mei FM, Xi XX, Liu LC, Zhou ZJ, Wang D, Wang YQ, Cheng Y (2003) Characterization of soil dust aerosol in China and its transport and distribution during 2001 ACE-Asia. 1. Network observations. J Geophys Res 108(D9). https://doi.org/10.
1029/2002JD002632 Zhang Q, Jimenez JL, Worsnop DR, Canagaratna M (2007) A case study of urban particle acidity and its influence on secondary organic aerosol. Environ Sci Technol 41(9):3213–3219. https://doi.org/10.
1021/es061812j Zhang T, Cao JJ, Tie XX, Shen ZX, Liu SX, Ding H, Han YM, Wang GH, Ho KF, Qiang J, Li WT (2011) Water-soluble ions in atmospheric aerosols measured in Xi’an, China. Seasonal variations and sources. Atmos Res 102(1–2):110–119. https://doi.org/10.1016/j.atmosres.
2011.06.014 Zhang X, Wu L, Zhang R, Deng S, Zhang Y, Wu J, Li Y, Lin L, Li L, Wang Y, Wang L (2013) Evaluating the relationships among economic growth, energy consumption, air emissions and air environmental protection investment in China. Renew Sust Energ Rev 18:
259–270. https://doi.org/10.1016/j.rser.2012.10.029 Zhao PS, Dong F, He D, Zhao XJ, Zhang XL, Zhang WZ, Yao Q, Liu HY (2013) Characteristics of concentrations and chemical compositions for $\mathrm{PM}_{2.5}$ in the region of Beijing, Tianjin, and Hebei, China. Atmos Chem Phys 13(9):4631–4644. https://doi. org/10.5194/acp-13-4631-2013 Zhao M, Huang Z, Qiao T, Zhang Y, Xiu G, Yu J (2015a) Chemical characterization, the transport pathways and potential sources of PM2.5 in Shanghai. Seasonal variations. Atmos Res 158-159:66–
78. https://doi.org/10.1016/j.atmosres.2015.02.003 Zhao S, Tie X, Cao J, Li N, Li G, Zhang Q, Zhu C, Long X, Li J, Feng T, Su X (2015b) Seasonal variation and four-year trend of black carbon in the Mid-west China. The analysis of the ambient measurement and WRF-Chem modeling. Atmos Environ 123:430–439. https:// doi.org/10.1016/j.atmosenv.2015.05.008
|
atmosphere
|
0107
|
10.1016/j.atmosres.2017.11.014
|
image
|
Fig. 1. Google map of Bohai Bay and the two sampling sites.
|
|
atmosphere
|
0107
|
10.1016/j.atmosres.2017.11.014
|
table
|
Table 1 Average meteorological parameters at Bohai Sea and Tanggu from Sep. 8th to Oct. 8th in 2013. (T, temperature; RH, relative humidity; WS, wind speed).
|
|
atmosphere
|
0107
|
10.1016/j.atmosres.2017.11.014
|
table
|
Table 2 Concentrations of gaseous pollutants, principal water-soluble ions and acidity in $\mathrm{PM}_{2.5}$ at Bohai Sea and Tanggu from Sep. 8th to Oct. 8th in 2013.
|
|
atmosphere
|
0107
|
10.1016/j.atmosres.2017.11.014
|
image
|
Fig. 2. Relative mass distribution among principal water-soluble ionic species in $\mathrm{PM}_{2.5}$ in the Bohai Sea and Tanggu.
|
|
atmosphere
|
0107
|
10.1016/j.atmosres.2017.11.014
|
image
|
Fig. 4. The relationship between ammonium equivalent concentration and the sum of sulfate and nitrate equivalent concentration in $\mathrm{PM}_{2.5}$ samples.
|
|
atmosphere
|
0107
|
10.1016/j.atmosres.2017.11.014
|
image
|
Fig. 5. Scatter plot and linear regression that compares the molar concentrations of $[\mathrm{H}^{+}]_{\mathrm{Total}}$ to ${\mathrm{so}}_{4}^{\ 2-}$ in $\mathrm{PM}_{2.5}$ from Sep. 8th to Oct. 8th in 2013 at Bohai Sea and Tanggu.
|
|
atmosphere
|
0107
|
10.1016/j.atmosres.2017.11.014
|
image
|
Fig. 6. The relationship between the ratios of $[\mathrm{NO}_{3}^{\mathrm{~-}}]/[\mathrm{SO}_{4}^{\mathrm{~2~-}}]$ and $[\mathrm{NH}_{4}^{~+}]/[\mathrm{SO}_{4}^{~2\,-}]$ under differernt aerosol acidity in Bohai Sea and Tanggu. (LA, less acidic; MA, more acidic; Group A and B are data sets in more and less acidic condition in Tanggu respectively).
|
|
atmosphere
|
0107
|
10.1016/j.atmosres.2017.11.014
|
text
| Not supported with pagination yet |
Spatial and temporal characteristics of $\mathbf{PM}_{2.5}$ acidity during autumn in marine and coastal area of Bohai Sea, China, based on two-site contrast
Ming Zhoua, Yufen Zhanga, Yan Hana, Jianhui Wua, Xiang $\mathrm{{Du}^{a}}$ , Hong $\mathrm{\DeltaXu^{a}}$ , Yinchang Fenga,, Suqin Hanb
aState Environmental Protection Key Laboratory of Urban Ambient Air Particulate Matter Pollution Prevention and Control, College of Environmental Science and
Engineering, Nankai University, Tianjin 300071, China
b Tianjin Institute of Meteorological Science, Tianjin 300074, China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Acidity
PM2.5
Spatial and temporal variation
Marine
Coast
Nitrate formation
In-situ pH of atmospheric particulate, defined as the ${\mathsf{p H}}$ value of aqueous phase in ambient aerosol, has been reported to have significant influence on the formation progress of secondary aerosol, especially through the heterogeneous pathway. In this study, $\mathbf{PM}_{2.5}$ samples were collected in the marine and costal area of Bohai Sea from September 8th to October 8th in 2013, with daytime and nighttime separated. Eight water-soluble ions including ${\mathrm{so}}_{4}^{\ 2-}$ , $\mathrm{NO}_{3}{}^{-}$ , $\mathrm{Cl}^{-}$ , $\mathrm{NH_{4}}^{+}$ , $\kappa^{+}$ , ${\mathrm{Ca}}^{2\,+}$ , $\mathrm{Na}^{+}$ and ${\mathrm{M}}{\mathrm{g}}^{2\,+}$ were analyzed by ion chromatography. The insitu pH of $\mathbf{PM}_{2.5}$ was estimated using Aerosol Inorganics Model II, with meteorological parameters (temperature and relative humidity) and basic chemical composition data (concentrations of water-soluble ions) serving as input. Five indicators were conjunctively applied to describe the spatial and temporal characteristics of $\mathbf{PM}_{2.5}$ acidity over Bohai Sea during autumn. As a result, strong acidity was found in both marine and coastal area. Marine area had a stronger acidity under a more $\mathrm{NH_{4}}^{+}$ -deficiency and humid condition. And the difference of $\mathbf{PM}_{2.5}$ acidity between daytime and nighttime was more obvious in coastal area than that in marine area, with stronger acidity observed during the daytime. Local $S0_{2}$ emission was identified as an important factor influencing the diurnal variation of aerosol acidity. Meanwhile, sulfurous species were identified as a mixture of $\mathrm{NH}_{4}\mathrm{HSO}_{4}$ and $\mathrm{H}_{2}\mathrm{SO}_{4}$ in marine area while a mixture of $\mathrm{NH}_{4}\mathrm{HSO}_{4}$ and $(\mathrm{NH}_{4})_{2}\mathrm{SO}_{4}$ in the coastal area. Analysis in the impact of aerosol acidity on nitrate formation has indicated that heterogeneous pathways were important in nitrate formation in coastal area of Bohai Sea as well as the homogeneous pathways.
Capsule abstract: $\mathrm{PM}_{2.5}$ was highly acidic around Bohai Sea, and the acidity of $\mathbf{PM}_{2.5}$ was stronger in marine area than coastal area during autumn.
1. Introduction
Ambient aerosol appears to be acidic because acidic ions such as $^{\mathrm{SO}_{4}^{\ 2-}}$ and $\mathrm{NO}_{3}^{\mathrm{~-~}}$ cannot be fully neutralized by alkaline ions, such as ${\mathrm{Ca}}^{2\,+}$ and $\mathrm{NH_{4}}^{+}$ (Pathak et al., 2011; Xue et al., 2011; Yao et al., 2011; Rengarajan et al., 2011; Huang et al., 2011). Recently, acidic aerosol has been reported to ubiquitously exist worldwide. Except for the direct inhalation, acidic aerosol can expose human health to a high risk by activating the hazardous components in particulates as well (He et al., 2012; Zhang et al., 2007; Amdur and Chen, 1989; Utell and Looney, 1995). Meanwhile, acidic aerosol can do harm to ecosystem through wet/dry deposition (He et al., 2012; Larssen et al., 2006; Likens et al., 1996). And the larger hygroscopicity of acidic aerosol can result in degradation of atmospheric visibility and disturbance in radiative balance of atmosphere because particulates' ability to scatter light and to nucleate cloud droplets can be enhanced by larger hygroscopicity (Zhang et al., 2007; Watson, 2002; Boucher and Anderson, 1995; Crumeyrolle et al., 2008).
The formation of secondary aerosol (SA) is closely interrelated to aerosol acidity, especially in the heterogeneous reactions. For the secondary inorganic aerosol (SIA), recent research has found that aerosol acidity can tip the scale in the pathway of sulfate formation. The large unidentified source of sulfate during the heavy-haze days in Beijing was explained by the oxidation from $\mathrm{NO}_{2}$ in aerosol water content. Changes occurred in pH played a critical role in the transition to an $\mathrm{NO}_{2}$ - dominated process of aqueous sulfate production under haze days from an $\mathrm{H}_{2}\mathrm{O}_{2}$ -dominated process of aqueous sulfate production in cloud droplets (Cheng et al., 2016). And for the secondary organic aerosol (SOA), it is reported that the catalytic action of acid in heterogeneous reactions might play a significant role in enhancing the SOA production in the atmosphere (Jang et al., 2002; Edney et al., 2005; Gao et al., 2004b; Iinuma et al., 2005). Therefore, aerosol acidity deserves to be lucubrated in the field of atmospheric physics and chemistry.
Generally, the actual aerosol acidity can't be measured directly because of low water content in particles. Thus thermodynamic models have been developed to estimate the in-situ pH in deliquesced aerosol. Thermodynamic equilibrium models such as SCAPE2 (Meng et al., 1995), ISORROPIA (Nenes et al., 1998) and EQUISOLV II (Jacobson, 1999) can extrapolate in-situ aerosol pH through measuring the $\ensuremath{\mathbf{p}}\ensuremath{\mathbf{H}}$ of aerosol extracts and estimating liquid water content based on the information of ionic species, temperature and relative humidity during sampling (Yao et al., 2006). Another thermodynamic equilibrium model named AIM II (Aerosol Inorganics Model II, Clegg et al., 1998) is also widely utilized to estimate the in-situ pH of aerosol using measured strong acidity, concentrations of ionic species, average temperature and relative humidity as the input. And by comparison, Model II is more suitable to estimate in-situ aerosol pH in the atmosphere than SCAPE2 and ISORROPIA (Yao et al., 2006).
Same with other maritime countries, coastal regions own the largest population density and the most developed technology and economy in China. Circum-Bohai Sea Region, together with Yangtze River delta and Pearl River Delta has the fastest economic growth in China. Bohai Sea is semi-enclosed with an area about $78\,\mathrm{km}^{2}$ and about one third of China's water transport vessels and fishing boats were navigated over the Bohai Sea by 2013 (China Statistical Yearbook, 2013; Chinese Fishery Statistical Yearbook, 2013). Bohai Bay, a shallow bay located in the west of Bohai Sea, is adjacent to Beijing-Tianjin-Hebei Region (the political center of China) and surrounded by Shandong province, Hebei province and Tianjin municipality. Located at the downwind area of North China Plain, Bohai Bay is inevitably impacted by regional transport of air pollution and the local emission from ships in recent years (Zhang et al., 2016).
Numbered previous studies about air pollution around Bohai Sea were focused on the characteristics of particle concentration and compositions (Gu et al., 2011; Zhang et al., 2014; Ni et al., 2013; Xu et al., 2015). However, there is rare published literature studying the characteristics of $\mathrm{PM}_{2.5}$ acidity over the Bohai area. In this study, the experiment was designed to explore the spatial and temporal variation of $\mathrm{PM}_{2.5}$ acidity in Bohai area during autumn. Thus, a coastal and a marine sampling site were chosen and sampling time was divided to daytime and nighttime. The $\mathrm{PM}_{2.5}$ pollution characteristics were analyzed and five indicators were applied to discuss the characteristics of $\mathrm{PM}_{2.5}$ acidity. Meanwhile, the impact of aerosol acidity on nitrate formation was discussed as well.
2. Experimental methods
2.1. Sampling sites
Aiming at studying the characteristics of $\mathrm{\bfPM}_{2.5}$ acidity in the region of Bohai Sea, the $\mathrm{PM}_{2.5}$ samples were collected at two chosen sites noted on Fig. 1. One monitoring station was located in Tanggu district of Tianjin municipality which represents the coastal area. And the other station in Bohai Sea was also set up to compare the differences between the coast and sea.
The Tanggu (TG) sampling site $(117^{\circ}43^{\prime}\mathrm{E},39^{\circ}3^{\prime}\mathrm{N})$ was located in Binhai meteorological warning center, approximately $4\,\mathrm{km}$ southwest, $8\,\mathrm{km}$ west and $9\,\mathrm{km}$ northeast from the coastline, which can represent the west coastal region of Bohai Sea. Tanggu district, located in the east of Tianjin, is the central area of Bohai Sea Economic Circle. It has the largest comprehensive trading port of northern China, Tianjin Port. Tianjin Port has an annual throughput of 250 million tons and maintains trading with over 180 countries and regions, which means that this site is inevitably affected by local emissions due to the industrialization as well as pollution transporting from Beijing-TianjinHebei district.
The Bohai Sea (BS) sampling site $(118^{\circ}25^{\prime}{\mathrm{E}},38^{\circ}27^{\prime}{\mathrm{N}})$ was located at an offshore oil drilling platform which lied in Bohai Sea and ${\sim}70\,\mathrm{km}$ away from the northern coast of Bohai Sea Gulf and $\sim\!40\,\mathrm{km}$ from the southern, as it is shown in Fig. 1. Detailed introduction about the oil drilling platform is available in Text S1. Bohai Sea, surrounded by mainland from three sides, is the only inland sea of China. So it seems that atmospheric pollutants generated from adjacent area, such as Liaotung Peninsula in the north and the whole Beijing-Tianjin-Hebei region in the west, could transport to this sampling site and play an important role due to the open marine environment.
The sampler at Tanggu sampling site was approximately $10\;\mathrm{m}$ above the ground and that at Bohai Sea sampling site was $30\,\mathrm{m}$ over the sea level. All of the $\mathrm{PM}_{2.5}$ samples were collected from September 8th to October 8th in 2013 using medium-volume samplers (TH-150, Wuhan Tianhong Intelligence Instrumental Facility, Wuhan, China) with flow rates of $100\,\mathrm{L}\,\mathrm{min}^{\,-\,1}$ . Considering the difference of pollutants emission and meteorological conditions between daytime and nighttime, the sampling time was divided into two 11-hour periods, 8:00–19:00 for the daytime and 20:00–7:00 (the next day) for the nighttime. Field blanks were used to determine any possible contamination. Concentration of gaseous pollutants in Tanggu was downloaded from an official website (http://106.37.208.233:20,035/) while the information of gaseous pollutants was not available in Bohai Sea. Meanwhile, hourly meteorological information including wind speed (WS, $\mathrm{~m~s~}^{-1}$ , temperature (T, $^\circ C)$ and relative humidity (RH, $\%$ ) was obtained from auto-monitoring station set at Tanggu and Bohai Sea.
2.2. Gravimetric and chemical composition analysis
Particles were collected on quartz fiber filters for subsequent gravimetric and chemical composition analysis. Each quartz filter was pre-baked in a muffle furnace for $4\,\mathrm{h}$ at $900\,^{\circ}\mathrm{C}$ to reduce residual carbon level before sampling. All the filters were weighted twice on a Mettler Toledo microbalance (resolution $1\;\upmu g)$ before and after sampling. It is important to emphasize that filters were conditioned for $48\,\mathrm{h}$ in a balance room under controlled relative humidity (RH, $50\ \pm\ 5\%)$ and temperature (T, $20\ \pm\ 1\ ^{\circ}\mathrm{C})$ before weighting. In addition, all the quartz filters were stored in a freezer $(4\,^{\circ}\mathrm{C})$ before subsequent composition analysis to improve the accuracy of experimental results.
The quartz fiber filters were used to analyze water-soluble ions, including $^{\mathrm{SO}_{4}^{\ 2-}}$ , $\mathrm{NO}_{3}^{\mathrm{~-~}}$ , $\mathrm{Cl}^{-}$ , $\mathrm{NH_{4}}^{+}$ , $\mathsf{K}^{+}$ , ${\mathrm{Ca}}^{2\,+}$ , $\mathrm{Na}^{+}$ and ${\mathrm{M}}{{g}^{2}}^{\mathrm{~+~}}$ , by Ion chromatography (DX-120 IC). The method detection limits (MDLs) were within the range of 0.01 to $0.04\,\upmu\mathrm{g}\,\upmu^{-\,3}$ for cations and 0.03 to $0.07\,\upmu\mathrm{g}\,\up m^{\,-\,3}$ for anions (Hsu et al., 2007; Liu et al., 2016b). A quarter of each quartz fiber filter was placed into the glass tube and deionized water was applied to extract. To ensure the water-soluble ions in samples were extracted completely into the solution, the extraction procedure was carried out for at least three times. All the extract solutions were stored under $4\,^{\circ}\mathrm{C}$ until the moment for detection by ion chromatography. Prior to the ions detection, standard solutions were prepared and detected for over three times and low relative standard deviations (RSD) were obtained. Analytical quantification was carried out by using calibration curves made from standard solutions prepared in a concentration range of 1 to $120\mathrm{\,mg\,L^{-1}}$ (vary with each ion). Expect for ionic species, carbonaceous species and elements had been analyzed as well and the analytical methods are introduced in Text S2.
The background contamination was observed regularly by using blank tests, which were applied to validate and correct data. The analysis of blank and duplicate samples was carried out for approximately $10\%$ of all the samples. For quality assurance and quality control (QA and QC), certified reference materials (CRMs, produced by National Research Center for Certified Reference Materials, China) were pretreated and analyzed with the same procedure which had been described in previous studies (Tian et al., 2014; Liu et al., 2016b; Zhao
et al., 2013; Kong et al., 2010; Bi et al., 2007; Wu et al., 2009; Xue et al., 2010).
2.3. Indicators of $P M_{2.5}$ acidity
2.3.1. Ratio of cation/anion
In this study, we use the C/A ratio (equivalent charge ratio of major cations to anions) to indicate the neutralizing level of $\mathrm{PM}_{2.5}$ . The C/A ratio is defined as the following equation (Zhang et al., 2002; Adams et al., 1999):
$$
R_{C/A}=\frac{[N H_{4}^{+}]+2\times[C a^{2+}]}{2\times[S O_{4}^{2-}]+[N O_{3}^{-}]}
$$
We used square bracelets to indicate the molar concentrations of the species inside here and henceforth. Only four ions $\mathrm{NH_{4}}^{+}$ , ${\mathrm{Ca}}^{2\,+}$ , ${\mathrm{so}}_{4}^{\ 2-}$ and $\mathrm{NO}_{3}{}^{-}$ are involved in this equation on account of the small quantity and little influence to aerosol acidity of other ions, as it is shown in Table 2. Theoretically when $\mathrm{R}_{\mathrm{C/A}}\,<\,1$ , it indicates that the aerosol is acidic, while most of acids deem to be neutralized when $\mathtt{R_{C/}}$ $_\mathrm{~A~}\geq1$ (He et al., 2012). However, considering the uncertainty of evaluation method of neutralization level of $\mathrm{PM}_{2.5},\mathrm{we}$ use $\mathrm{R_{C/A}}$ value of 0.9 to divide $\mathrm{PM}_{2.5}$ samples into two groups: more acidic aerosol group $(\mathrm{R_{C/}}$ $_\mathrm{~A~}<\ 0.9)$ and less acidic aerosol group $(\mathrm{R}_{\mathrm{C/A}}\geq0.9)$ (He et al., 2012; Liu et al., 2016a).
2.3.2. [H]Total, $[H]_{I n s}$ and pH
$[H]_{\mathrm{Total}}$ is defined as the total amount $\mathrm{\dot{\m}\ n m o l\ m}^{-\,3}\mathrm{)}$ of $\boldsymbol{\mathrm{H}}^{+}$ referring to the total acid contributed by strong acids in the aqueous extract of collected aerosol. $[H_{\mathrm{{Total}}}$ include $[H]_{\mathrm{Ins}},$ $[\mathrm{HSO}_{4}\mathrm{~}^{-}]$ and any other $[\mathbf{H}^{+}]$ in the solid phase of aerosol at nitrate and/or sulfate equilibrium (Pathak et al., 2009; Zhang et al., 2007). Here we use the ionic balance of relevant inorganic ionic species (Pathak et al., 2009; Lippmann et al., 2000) to estimate the total aerosol acidity of $\mathrm{PM}_{2.5}$ as $[H]_{T o t a l}=2\times[S O_{4}{}^{2}{}^{-}]+[N O_{3}{}^{-}]-[N H_{4}{}^{+}]$ .
$[H]_{\mathrm{Ins}}$ is defined as the molar concentration $\mathrm{(in\nmol\,m^{-\,3})}$ of free $\boldsymbol{\mathrm{H}}^{+}$ in the aqueous phase of ambient aerosol. Free $\boldsymbol{\mathrm{H}}^{+}$ is a portion of $\boldsymbol{\mathrm{H}}^{+}$ that has chemical activity and directly decides the aerosol $\mathsf{p H}$ which plays an extremely important role in atmospheric chemistry, especially in the heterogeneous processes (Yao et al., 2006; Chameides and Stelson, 1992; Sievering et al., 1995; Van Oss et al., 1998; Gao et al., 2004a; Nemitz et al., 2004).
The Extended AIM Thermodynamic Model (E-AIM, Model II, http:// www.aim. env.uea.ac.uk/aim/) was employed to estimate the in-situ acidity and aerosol water content in the deliquesced particles. E-AIM II is an equilibrium thermodynamic model that can simulate liquid and solid phase of ionic compositions accurately in the $\mathrm{H^{+}\mathrm{~-~}N H_{4}^{+}\mathrm{~-~}S O_{4}^{\mathrm{~2~-~}}\mathrm{-NO_{3}^{\mathrm{~-~}}\mathrm{-H_{2}O}}}$ system under certain temperature and relative humidity (Clegg et al., 1998; He et al., 2012). The weekly average temperature $(\mathrm{T},\ {}^{\circ}\mathrm{C})$ , relative humidity (RH, 0.1–1) and molar concentrations of relevant species $\(\mathrm{[NH_{4}}^{+}]$ , $[{\cal S}{\cal O}_{4}{}^{2}\,]$ , $[\mathrm{NO}_{3}{}^{-}]$ and $[H_{\mathrm{{Total}}})$ were input in the E-AIM II model. Of particular importance is the charge balance between the input cations and anions in this system (Zhang et al., 2007; Yao et al., 2006; Pathak et al., 2009).
$\mathrm{Na}^{+}$ and $\mathrm{Cl}^{-}$ were not included in E-AIM II model but the influence was negligible in our research. Compared with ${\mathrm{so}}_{4}^{\ 2-}$ , $\mathrm{NO}_{3}^{\mathrm{~-~}}$ and $\mathrm{NH_{4}}^{+}$ dominant in inorganic ions, $\mathrm{Na}^{+}$ and $\mathrm{Cl}^{-}$ accounted for only a small portion in inorganic ions ( $.1\%$ and $3\%$ on average respectively, 1–2 magnitude lower than those of ${\mathrm{so}}_{4}^{\ 2-}$ , $\mathrm{NO}_{3}^{\mathrm{~-~}}$ and $\mathrm{NH_{4}}^{+}$ ) although the sampling sites located in marine area, as it is shown in Table 2. Previous study (Yao et al., 2006) has found that the omission of $\mathrm{Na}^{+}$ and $\mathrm{Cl}^{-}$ from E-AIM II did not create a significant error in the estimated in-situ pH of $\mathrm{PM}_{2.5}$ when the portion of $\mathrm{Na}^{+}$ and $\mathrm{Cl}^{-}$ was very small in $\mathrm{PM}_{2.5}$ $8\%$ in Yao's work, also a coastal city, Hong Kong).
Up to now, there have been four E-AIM models can be selected. The E-AIM III and E-AIM IV have already included $\mathrm{Na}^{+}$ and $\mathrm{Cl}^{-}$ but they have some restrictions: E-AIM III only operates at $298.3\:\mathrm{K}$ and the input RH cannot below 0.6 in E-AIM IV (while the average RH was below 0.6 during the daytime in Tanggu sampling site). The setting of temperature and relative humidity can greatly affect the result of in-situ pH. As E-AIM III and E-AIM IV were improper to simulate the real situation, EAIM II was selected because of its adaptation of a wide range of temperature and relative humidity. Other models, such as SCAPE and ISORROPIA, cannot be used in current study because they required gaseous $\mathrm{HNO}_{3}$ and $\mathrm{NH}_{3}$ (He et al., 2012; Yao et al., 2006).
In principle, the pH is calculated as $\mathrm{pH}=\mathrm{~-}\log(a_{H^{+}}),$ , $a_{H^{+}}$ is the molality-based activity of $\boldsymbol{\mathrm{H}}^{+}$ (aq). However, this is extremely challenging to measure the $a_{H^{+}}$ in the aqueous phase of aerosol directly. EAIM II was used to solve for the amount of aqueous phase $\boldsymbol{\mathrm{H}}^{+}$ with measured particle constituents. The approach using AIM II model requires a basic assumption of an internally mixed population of particles (Murphy et al., 2017). However, it cannot always be the ideal case and the best accuracy can be provided by models include activity coefficient (γ) that account for non-ideal conditions (Murphy et al., 2017). The insitu pH of aerosol aqueous phase was calculated using the following formula:
$$
\mathrm{In-situ\,pH}=-{\log}(\gamma\times[H^{+}]_{F r a c})
$$
$\upgamma$ is the activity coefficient of aqueous phase $\boldsymbol{\mathrm{H}}^{+}$ and $[H^{+}]_{F r a c}$ is the molar fraction of aqueous phase $\boldsymbol{\mathrm{H}}^{+}$ . $\gamma\times[H^{+}]_{F r a c}$ represents the aqueous phase activity of $\boldsymbol{\mathrm{H}}^{+}$ (Zhang et al., 2007; He et al., 2012; Pathak et al., 2009; Yao et al., 2006). At the same time, in-situ acidity of aerosol and water content can also be simulated directly (Zhang et al., 2007; Xue et al., 2011). However, some defects exist in the E-AIM II model as well. For instance, the influence of calcium, magnesium and organic species is neglected in E-AIM II model (Yang, 2014; He et al., 2012).
Large bias may exist when the concentrations of ${\mathrm{Ca}}^{2\,+}$ and ${\mathrm{M}}{{g}^{2}}^{+}$ of samples are high, for example, in the mineral dust (Ziemba et al., 2007). However, ${\mathrm{Ca}}^{2\,+}$ and ${\mathrm{M}}\mathbf{g}^{2\,+}$ only accounted for $1\%$ and $0.1\%$ (on average) in the total concentration of inorganic species. Thus, the omission of ${\mathrm{Ca}}^{2\,+}$ and ${\mathrm{M}}{{g}^{2}}^{\mathrm{~+~}}$ is expected to have little influence on acidity estimation although this influence cannot be quantitatively estimated (Yao et al., 2006). The influence of ${\mathrm{Ca}}^{2\,+}$ and ${\mathrm{M}}{{g}^{2}}^{+}$ was ignored in many previously published works as well (Zhang et al., 2007; He et al., 2012; Pathak et al., 2009, 2004; Yao et al., 2006; Xue et al., 2011).
The omission of organic acid in E-AIM II model has little influence on estimation of aerosol acidity as well because their low abundance in aerosols. Organic aerosol, especially organic acids, may contribute free $\boldsymbol{\mathrm{H}}^{+}$ in aerosol aqueous phase and affect partitioning/dissociation of inorganic species in acidic particles. However, previous observation founded that oxalic acid, the single most abundant organic aerosol compound identified so far (Huang and Yu, 2007), contributed little to the free acidity of rain water (Huang et al., 2010). This implied that contribution of organic acids to aerosol free $\boldsymbol{\mathrm{H}}^{+}$ is most likely minor. Besides, quantity of free $\boldsymbol{\mathrm{H}}^{+}$ released from organosulfates was proved small compared with that estimated from inorganic anion and cation balance (Chan et al., 2010; Olson et al., 2011). Secondary organic aerosol may also affect aerosol acidity because of the hygroscopic property nevertheless its hygroscopicity is significantly weaker than inorganic species such as NaCl and $(\mathrm{NH}_{4})_{2}S0_{4}$ (Ansari and Pandis, 2000). To sum up, organic species generally have minor influence on insitu $\ensuremath{\mathbf{p}}\ensuremath{\mathbf{H}}$ of aerosol. So, we did not take the organic species into concentration in our work, neither did previous works about the in-situ acidity of aerosol (Zhang et al., 2007; He et al., 2012; Pathak et al., 2009, 2004; Yao et al., 2006; Xue et al., 2011).
2.3.3. $[N H_{4}^{+}J_{m e a s}/L N H_{4}^{+}J_{n e u}$
The ratio of measured $\mathrm{NH_{4}}^{+}$ concentration (i $\mathrm{\Omega}_{1\mathrm{\nmol\,m}}^{-\,3\cdot}$ ) to the concentration (in $\mathrm{nmol}\;\mathrm{m}^{\mathrm{~-~}3\mathrm{\cdot~}}$ ) of $\mathrm{NH_{4}}^{+}$ needed to fully neutralize the anions was expressed as $[\mathrm{NH_{4}}^{+}]_{\mathrm{meas}}/[\mathrm{NH_{4}}^{+}]_{\mathrm{neu}}$ (Zhang et al., 2007). The value of this ratio can be applied to evaluate the degree of stoichiometric neutralization for the ensemble of measured particles.
$$
\frac{[N H_{4}^{+}]_{m e a s}}{[N H_{4}^{+}]_{n e u}}=[N H_{4}^{+}]/(2\times[S O_{4}^{2-}]+[N O_{3}^{-}]+[C l^{-}])
$$
In this research, when $[\mathrm{NH}^{4}\,^{+}]_{\mathrm{meas}}/[\mathrm{NH}^{4}\,^{+}]_{\mathrm{neu}}\approx1$ , particles are deemed to be mostly neutralized (Zhang et al., 2007). Meanwhile, particles become more acidic along with the decline of this value.
3. Results and discussions
3.1. Meteorological parameters
Table 1 presents the average meteorological parameters (T, RH, WS) at Bohai Sea and Tanggu during the sampling time, daytime and nighttime separated. The average temperature, relative humidity and wind speed in the Bohai Sea was $22\ ^{\circ}\mathrm{C};$ $71\%$ and $6\:\mathrm{m}\:\mathrm{s}^{\mathrm{-1}}$ . Compared with Bohai Sea, Tanggu had a lower level in temperature $(21\ ^{\circ}\mathrm{C})$ and relative humidity $(67\%)$ . Besides, wind speed was much lower than Bohai Sea $2\,\mathrm{m}\,\mathrm{s}^{\mathrm{-1}}$ v.s. $6\:\mathrm{m}\:\mathrm{s}^{-\:1\cdot}$ ) as the consequence of land-sea thermal discrepancy. As to the difference between day and night, it was more obvious in Tanggu in temperature and relative humidity, while the difference of wind speed in Bohai Sea was relatively more distinct. After all, higher relative humidity was found on Bohai Sea $(>70\%)$ and it was higher during the nighttime than daytime for both sites. At Bohai Sea sampling site, the wind speed was higher in the nighttime. However, there was not obvious difference of wind speed between daytime and nighttime in coastal area. This indicated that coastal area had more stable atmosphere and the movement of atmospheric disturbance was weaker during the night in Bohai Sea.
3.2. Concentration and composition of $P M_{2.5}$
Concentrations of gaseous pollutants, principal water-soluble ions and $\mathrm{PM}_{2.5}$ acidity in the daytime and nighttime from Sep. 8th to Oct. 8th are averaged for the two sampling sites, Bohai Sea and Tanggu, as it is shown in Table 2. The mass concentrations of $\mathrm{PM}_{2.5}$ were similar in the daytime at the two sites $(\sim\!140\,\upmu\mathrm{g}\,\mathsf{m}^{-\,3})$ . However, it is obvious that Bohai Sea had a higher $\mathrm{\bfPM}_{2.5}$ mass concentration $(\sim\!141\;\upmu\mathrm{g}\;\mathbf{m}^{-\;3})$ during the night than the Tanggu sampling site $(-128\,\upmu\mathrm{g}\,\mathbf{m}^{-\,3})$ . To some extent, high level of $\mathbf{PM}_{2.5}$ mass concentration on the sea indicated a high regional $\mathrm{PM}_{2.5}$ background concentration in Bohai Rim. Interestingly, we have found that $\mathrm{\bfPM}_{2.5}$ concentration was higher in the daytime than nighttime at Tanggu, which might result from the anthropogenic activities in Tanggu economic development zone. As for the total ionic species (sum of ${\mathrm{SO}}_{4}^{\ 2-}$ , $\mathrm{NO}_{3}{}^{-}$ , $\mathrm{NH_{4}}^{+}$ , $\mathrm{Cl}^{-}$ , $\mathrm{Na}^{+}$ , ${\mathrm{Ca}}^{2\,+}$ ${\mathrm{M}}\mathfrak{g}^{2\,+}$ and $\mathrm{K}^{+}$ ) in $\mathrm{{PM}}_{2.5}$ , Tanggu shared a similar mass concentration with Bohai Sea. Nevertheless, the proportion of total ionic species was higher in Tanggu, especially during the night $(\sim\!41\%)$ . This was comparable with those results ( $\sim\!40\%$ in Tianjin, 2014; $39{-}44\%$ in Tai'an, 2014; $27{-}36\%$ in Beijing, 2006) in previous research around Bohai Bay (Wu et al., 2015; Liu et al., 2016b; He et al., 2012). Compared with other components in $\mathrm{PM}_{2.5.}$ , (total elements, organic matter and elemental carbon), inorganic ions accounted for a major part in $\mathrm{PM}_{2.5}$ (Text S2).
For both sampling sites, lower mass concentrations of ${\mathsf{S O}}_{4}^{\ 2-}$ were found during the night than the daytime. The concentrations of $\mathrm{NO}_{3}{}^{-}$ and $\mathrm{NH_{4}}^{+}$ showed no obvious difference in the two sites. Compared with previous research, the mass concentration of $\mathrm{NO}_{3}{}^{-}$ was higher although possible underestimation of nitrate might exist because of the omission of $\mathrm{HNO}_{3}$ . However, the mass concentrations of ${\mathrm{so}}_{4}^{\ 2-}$ and $\mathrm{NH_{4}}^{+}$ were comparable with previous researches (Liu et al., 2016b; He et al., 2012). This might attribute to the proper environmental condition for nitrate generation in the marine area. Mass concentration of ${\mathrm{Ca}}^{2\,+}$ was a little higher in the samples collected in the daytime of Tanggu. The major sources of calcium were soil dust, resuspended dust and construction dust reported by previous researches (Almeida et al., 2013; Liu et al., 2016b; Bi et al., 2007; Tullio et al., 2008; Xue et al., 2010). Tanggu district is a recently developed area in Tianjin, China. Higher calcium concentration during daytime than nighttime indicated that anthropogenic activities relative to construction and road traffic were tenser during daytime in Tanggu area.
Compared with the mass contributions of ${\mathrm{SO}}_{4}^{\ 2-}$ , $\mathrm{NO}_{3}^{\mathrm{~-~}}$ , $\mathrm{NH_{4}}^{+}$ (together accounted for $89{-}92\%$ in total water-solute ions),the other water-solute ions $\left(\mathbf{Cl}^{-}\right)$ , $\mathrm{Na}^{+}$ , ${\mathrm{Ca}}^{2\,+}$ , ${\mathrm{M}}\mathfrak{g}^{2\,+}$ and $\mathsf{K}^{+}$ ) were negligible $(0.07\mathrm{-}5.6\%$ individually in inorganic ions). The relative distributions of ${\mathrm{so}}_{4}^{\ 2-}$ , $\mathrm{NO}_{3}{}^{-}$ , $\mathrm{NH_{4}}^{+}$ , $\mathrm{Cl}^{-}$ , $\mathrm{Na}^{+}$ , ${\mathrm{Ca}}^{2\,+}$ , ${\mathrm{M}}{{g}^{2}}^{+}$ and $\mathrm{K}^{+}$ among watersolute ions are depicted in Fig. 2. Secondary ionic species, including ${\mathrm{so}}_{4}^{\ 2-}$ , $\mathrm{NO}_{3}{}^{-}$ and $\mathrm{NH_{4}}^{+}$ , were dominated in water-soluble ionic species. Sulfate had the largest fraction $(44-52\%)$ ) in water-solute ions in the two sampling sites. Then it was $\mathrm{NO}_{3}{}^{-}$ $(23{-}27\%)$ and $\mathrm{NH_{4}}^{+}$ $(17{-}18\%)$ . Higher distributions of $\mathrm{NO}_{3}^{\mathrm{~-~}}$ was found during the nighttime in the two sampling sites. And the fractions of $\mathrm{Cl}^{-}$ were evidently higher in Tanggu during the night $(\sim\!5\%)$ compared with other samples in the two sites $(\sim\!2\%)$ . This may be relevant to organic halides discharged from local industrial sources situated in Tanggu during the nighttime (Pathak et al., 2009). The contributions of sea salt ions $\left(\mathbf{Na}^{+}$ and ${\mathrm{M}}\mathfrak{g}^{2\,+}$ ) were at a low level $\zeta<2\%$ together) in spite of the marine environment. The distribution of $^{\mathrm{SO}_{4}^{\ 2-}}$ had a slight decrease while that of $\mathrm{NO}_{3}^{\mathrm{~-~}}$ seemed to ascend mildly with the spatial transformation from Bohai Sea to the coastal area. This probably because the transformation from $S O_{2}$ to ${\mathrm{SO}}_{4}^{\ 2-}$ can be accelerated in more acidic and hygroscopic aerosol (He et al., 2012) and the spatial distribution of aerosol acidity in the two sites will be discussed in Section 3.3.1.
3.3. Spatiotemporal variation in $P M_{2.5}$ acidity
3.3.1. Spatial distribution
Acidity-dependent chemical reaction progress in atmospheric aerosol, typically the heterogeneous oxidation of $S O_{2}$ on particulates surfaces and the generation of organic aerosol (OA), can be affected by the in-situ acidity (Chameides and Stelson, 1992; Sievering et al., 1995; Van Oss et al., 1998; Gao et al., 2004a; Nemitz et al., 2004; Surratt et al., 2007; Sudheer and Sarin, 2011).
To discuss the spatial characteristics of $\mathbf{PM}_{2.5}$ acidity during the research period, the values of all the indicators of $\mathrm{PM}_{2.5}$ acidity, including $[\mathrm{H^{+}}]_{\mathrm{Total}},\ [\mathrm{H^{+}}]_{\mathrm{Ins}},\ \mathrm{pH},\mathrm{C/A}$ and $[\mathrm{NH_{4}}^{+}]_{\mathrm{meas}}/[\mathrm{NH_{4}}^{+}]_{\mathrm{neu}},$ are averaged in Fig. 3. The estimating and calculating methods were discussed in Section 2.3. Strong acidity of $\mathrm{PM}_{2.5}$ has been found both in Bohai Sea and Tanggu. From marine to coastal area, there was an obviously downward trend in aerosol acidity $(\mathrm{[H^{+}]_{T o t a l}})$ and in-situ acidity $([\mathrm{H}^{+}]_{\mathrm{Ins}})$ in $\mathrm{PM}_{2.5}$ . Higher aerosol acidity $(282\,\mathrm{nmol}\,\mathrm{m}^{-3})$ ) was found in Bohai Sea samples. And relatively lower level of average aerosol acidity was found in Tanggu $(207\,\mathrm{nmol}\,\mathrm{m}^{-\,3})$ . These results were comparable with the aerosol acidity previously measured in Beijing $({\sim}390\;\mathrm{nmol}\;\mathrm{m}^{-\,3}$ , 2005),Shanghai $({\sim}220\;\mathrm{nmol}\;\mathrm{m}^{-\;3}$ , 2005) (Pathak et al., 2009).
The in-situ acidity $\mathrm{[H^{+}]_{I n s}})$ was $85\,\mathrm{nmol}\,\mathrm{m}^{\,-\,3}$ ( $30\%$ of $[\mathrm{H}^{+}]_{\mathrm{Total}})$ , $35\,\mathrm{nmol}\,\mathrm{m}^{-\,3}$ $.17\%$ of $[\mathrm{H}^{+}]_{\mathrm{Total}})$ in Bohai Sea and Tanggu. The existing difference in $[\mathrm{H}^{+}]_{\mathrm{Ins}}/[\mathrm{H}^{+}]_{\mathrm{Total}}$ ratio was due to the different levels of aerosol water content in $\mathrm{PM}_{2.5}$ samples. Samples in Bohai Sea had a larger water content $(\sim43\mathrm{~}\upmu\mathrm{g}\mathrm{~m~}^{-\mathrm{~}3})$ compared with the average aerosol water content $(\sim\!30\,\upmu\mathrm{g}\,\mathrm{m}^{-\,3})$ in Tanggu. Therefore, considerable fraction $(30\%)$ of aerosol acidity was released as in-situ acidity in Bohai Sea samples. Interestingly, although distinct spatial difference was observed both in aerosol acidity and in-situ acidity, the $\mathsf{p H}$ in the two sites samples were similar, 1.2 and 1.3 in Bohai Sea and Tanggu samples respectively. To sum up, the $\mathrm{PM}_{2.5}$ samples collected in Bohai Sea were a little more acidic than those in Tanggu. And water content played a dual role in the determination of in-situ acidity (Pathak et al., 2004). Obviously, larger aerosol water content would free more reactive $\boldsymbol{\mathrm{H}}^{+}$ from strong and weak acid molecules, such as $\mathrm{H}_{2}\mathrm{SO}_{4}$ and $\mathrm{HSO_{4}}^{-}$ . But, on the other hand, aerosol water content can also serve a dilution function to reactive $\boldsymbol{\mathrm{H}}^{+}$ concentration in the aqueous aerosol. After all, the results of $\mathsf{p H}$ in the two sampling sites are close to the in-situ pH of $\mathrm{PM}_{2.5}$ in previous research conducted in Beijing, where the in-situ pH of $\mathrm{PM}_{2.5}$ fluctuated at a range of 0.5 to 2.4 with a mean value around 1.5 during autumn in 2005 (He et al., 2012).
Conclusions summarized from the variation trend of C/A and $[\mathrm{NH_{4}}^{+}]_{\mathrm{meas}}/[\mathrm{NH_{4}}^{+}]_{\mathrm{neu}}$ are another added proofs that atmospheric aerosol was much less acidic in Tanggu than Bohai Sea. The average $\mathsfit{C}/$ A value in Tanggu was 0.8, higher than that in Bohai Sea (0.7), as it is shown in Table 2. The average values of $[\mathrm{NH_{4}}^{+}]_{\mathrm{meas}}/[\mathrm{NH_{4}}^{+}]_{\mathrm{neu}}$ ratio in Bohai Sea and Tanggu were 0.6 and 0.7. This proclaimed that sufficiency of ammonium was vital in lessening the $\mathrm{\bfPM}_{2.5}$ acidity. In general, the unanimous conclusion summarized from $[\mathrm{H}^{+}]_{\mathrm{Total}},~[\mathrm{H}^{+}]_{\mathrm{Ins}},$ pH, C/A and $[\mathrm{NH_{4}}^{+}]_{\mathrm{meas}}/[\mathrm{NH_{4}}^{+}]_{\mathrm{neu}},$ , provides important evidence that $\mathrm{PM}_{2.5}$ in Bohai Sea was more acidic than that in Tanggu and the high relative humidity together with deficiency of ammonia in marine area may be the most important causes. Besides, our research could prove that the joint utilization of the five indicators can help to describe the characteristics of aerosol acidity more expressly and elaborately.
The relationship between ammonium and the sum of sulfate and nitrate equivalent concentration is also critical to describe the acidic characteristics of aerosol acidity because a large proportion of sulfate and nitrate is neutralized by ammonia in ambient atmosphere. If the ratio of $[\mathrm{NH_{4}}^{+}]/([\mathrm{SO_{4}}^{2}{}^{-}]\,+\,[\mathrm{NO_{3}}^{-}])$ (equivalent concentration, neq $\mathrm{m}^{-3})$ is $<1$ ,we regard that these samples are $\mathrm{NH_{4}}^{+}$ -poor. In other words, sulfate and nitrate can't be fully neutralized by ammonia in aqueous atmospheric aerosol (Pathak et al., 2009). In Bohai Sea and Tanggu samples, the ratio was almost all $<\,1$ no matter it was in the daytime or in the nighttime. This indicated that the samples collected in Bohai Sea and Tanggu were definitely $\mathrm{NH_{4}}^{+}$ -poor. In addition, larger deficiency was found in marine sampling site (Bohai Sea) compared with that in coastal sampling site (Tanggu). This indicated that the atmospheric aerosol was more acidic on the sea and neutralization of alkaline ions, especially ammonia, may be one important explanation for the weaker acidity of $\mathrm{PM}_{2.5}$ in the coastal samples. Meanwhile, the ammonium deficiency became obviously more severe when the sum of sulfate and nitrate equivalent concentration was on a higher level as it is shown in Fig. 4. This is consistent with the conclusion that the sulfurous species in the samples of Bohai Sea and Tanggu were a mixture of $\mathrm{NH}_{4}\mathrm{HSO}_{4}$ and $\mathrm{H}_{2}\mathrm{SO}_{4}$ because of the $\mathrm{NH_{4}}^{+}$ -poor conditions, which we will discuss in the next section.
During the study period, $[\mathrm{H}^{+}]_{\mathrm{Total}}$ correlated well with $[{\bf S O}_{4}^{\mathrm{~\,~2~-~}}]$ in Bohai Sea $(r=0.95$ , Fig. 5) and Tanggu sampling site $(r=0.93$ , Fig. 5). The dotted line (2:1) indicates that ${{S O}_{4}}^{2\mathrm{~-~}}$ in particulates is in the form of $\mathrm{H}_{2}\mathrm{SO}_{4}$ while the dotted line (1:1) indicates that ${\mathsf{S O}}_{4}^{\ 2-}$ exists in the form of $\mathrm{HSO_{4}}^{-}$ . And the horizontal axis means $\mathrm{H}_{2}\mathrm{SO}_{4}$ is fully neutralized by ammonium and ${\mathrm{so}}_{4}^{\ 2-}$ exists in the form of $(\mathrm{NH}_{4})_{2}S0_{4}$ . Fig. 5 illustrated $\mathrm{NH}_{4}\mathrm{HSO}_{4}$ appeared to be the mainly existent form of sulfate at Tanggu as the slope was 1.05. This observation is a little different with previous research in Pittsburgh that the sulfurous species in aerosol was a mixture of $\mathrm{NH}_{4}\mathrm{HSO}_{4}$ and $(\mathrm{NH}_{4})_{2}\mathrm{SO}_{4}$ with very little $\mathrm{H}_{2}\mathrm{SO}_{4}$ in most urbanized area (Zhang et al., 2007; Liu et al., 1996). However, Liu's previous research (Liu et al., 1996) had also found that sulfurous species were primarily $\mathrm{H}_{2}S0_{4}$ and $\mathrm{NH}_{4}\mathrm{HSO}_{4}$ in semi-rural areas. Tanggu district is a geographically-special area which locates on the border region of Tianjin megacity and the Bohai
Sea. That may be the reason why the samples in Tanggu shared a mediate composition of sulfurous species compared with urbanized and semi-rural area studied in Liu's research. Here other sulfurous specious, such as $\mathrm{K}_{2}\mathrm{SO}_{4},$ $\mathrm{CaSO_{4}}$ and $\mathrm{MgSO_{4}}$ , were ignored due to the minor concentration of $\mathsf{K}^{+}$ , ${\mathrm{Ca}}^{2\,+}$ and ${\mathrm{M}}{{g}^{2}}^{+}$ compared with $\mathrm{NH_{4}}^{+}$ (Table 2).
Contrastively, the sulfurous species in $\mathrm{PM}_{2.5}$ collected in Bohai Sea was a mixture of $\mathrm{NH}_{4}\mathrm{HSO}_{4}$ and $\mathrm{H}_{2}\mathrm{SO}_{4}$ (slope was 1.34) as it is illustrated in Fig. 5. Although there were few anthropogenic activities on the marine area, the samples collected in Bohai Sea had the same rule with those in semi-rural areas studied in Liu's previous research (Liu et al., 1996). Taking the high $\mathrm{PM}_{2.5}$ concentration of Bohai Sea into consideration, we can contribute this phenomenon to the pervasivelyexisting regional transportation of $\mathrm{{PM}}_{2.5}$ pollution.
3.3.2. Temporal distribution
Comparison of aerosol acidity between daytime and nighttime may be helpful to explore the factors determining aerosol acidity. Nevertheless, there was not visible difference of in-situ $\mathsf{p H}$ between daytime and nighttime found in Bohai Sea, 1.1 and 1.2 for the daytime and nighttime respectively. Although $\sim\!37\,\mathrm{nmol}\,\mathrm{m}^{-\,3}$ decrease of $[\mathrm{H}^{+}]_{\mathrm{Total}}$ was found during the nighttime in Bohai Sea, the ratio of $[\mathrm{H}^{+}]_{\mathrm{Ins}}/[\mathrm{H}^{+}]_{\mathrm{Total}}$ maintained stable from day to night, $30.3\%$ v.s. $29.6\%$ in Bohai Sea. Consistently, $\mathrm{C/A}$ and $[\mathrm{NH_{4}}^{+}]_{\mathrm{meas}}/[\mathrm{NH_{4}}^{+}]_{\mathrm{neu}}$ ratios in the nighttime had little difference compared with those during the daytime, as it is shown in Fig. 3b. Aerosol water content was also similar in the daytime and nighttime, $45\,\upmu\mathrm{g}\,\mathsf{m}^{\,-\,3}$ v.s. $43\,\upmu\mathrm{g}\,\mathrm{m}^{-\,3}$ .
Interestingly, the samples of $\mathrm{PM}_{2.5}$ collected in Tanggu were obviously more acidic in the daytime than nighttime. The in-situ pH was 1.0 in the daytime and 1.6 in the nighttime. There was a $\sim\!75\,\mathrm{nmol\,m}^{-\,3}$ decrease in $[\mathrm{H}^{+}]_{\mathrm{Total}}$ observed from day to night in Tanggu, two times of that in Bohai Sea. The ratio of $[\mathrm{H}^{+}]_{\mathrm{Ins}}/[\mathrm{H}^{+}]_{\mathrm{Total}}$ was increased from day to night, $14.6\%$ v.s. $19.0\%$ in Tanggu. Other parameters, $[\mathrm{H}^{+}]_{\mathrm{Ins}},\mathrm{C/A}$ and $[\mathrm{NH_{4}}^{+}]_{\mathrm{meas}}/[\mathrm{NH_{4}}^{+}]_{\mathrm{neu}}$ had little diurnal difference. The decrease in aerosol acidity might be attributed to the decreased concentration of ${\mathrm{so}_{4}}^{2\mathrm{~-~}}$ although the water content was at a higher level during the nighttime $(35.9\,\upmu\mathrm{g}\,\mathrm{m}^{-\,3})$ . The mass concentration of ${\mathrm{so}}_{4}^{\ 2-}$ was $27\,\upmu\mathrm{g}\,\mathrm{m}^{\,-\,3}$ and $20\,\upmu\mathrm{g}\,\upmu^{-\,3}$ in the daytime and nighttime respectively. The ratios of $[\mathrm{NH}_{4}^{~+}]/[{\mathrm{SO}_{4}}^{2}-]$ and $[\mathrm{NO}_{3}^{\mathrm{~-}}]/$ $[{\cal S}{\cal O}_{4}^{\ 2-}]$ were both much lower during the daytime in Tanggu due to the high ${\mathrm{SO}}_{4}^{\ 2-}$ level. From Table 2 we can find out that the observed concentrations of $S0_{2}$ and $0_{3}$ were approximately 1.5 and 2.5 times higher in the daytime than nighttime. Therefore, increased concentration of ${\mathrm{SO}}_{4}^{\ 2-}$ could be attributed to $S O_{2}$ emission from coal burning in the local area and the photochemical oxidation from ozone during the daytime.
3.3.3. Impact of aerosol acidity on nitrate formation
During the formation progress of $\mathrm{NO}_{3}{}^{-}$ , ${\mathrm{so}}_{4}^{\ 2-}$ and $\mathrm{NH_{4}}^{+}$ play a determinative effect. As ${\mathrm{so}}_{4}^{\ 2-}$ competes with $\mathrm{NO}_{3}{}^{-}$ for $\mathrm{NH_{4}}^{+}$ , the pathway of $\mathrm{NO}_{3}{}^{-}$ formation is suggested to be different under different $^{\mathrm{SO}_{4}^{\ 2-}}$ levels. Moreover, variational $\mathrm{NH_{4}}^{+}$ conditions may also lead to different formation mechanisms of $\mathrm{NO}_{3}^{\mathrm{~-~}}$ .The relationship between $\mathrm{NO}_{3}^{\mathrm{~-~}}$ and $\mathrm{NH_{4}}^{+}$ under different ${\mathrm{so}}_{4}^{\ 2-}$ levels has been widely expressed as a function relation of $[\mathrm{NO}_{3}^{\mathrm{~-}}]/[\mathrm{SO}_{4}^{\mathrm{~2~-}}]$ and $[\mathrm{NH}_{4}^{+}]/$ $[{\cal S}{\cal O}_{4}^{\ 2-}]$ (Pathak et al., 2004, 2009; He et al., 2012; Liu et al., 2016a). Previous research has suggested that the linear correlation between $[\mathrm{NO}_{3}^{\mathrm{~-}}]/[\mathrm{SO}_{4}^{\mathrm{~2~-}}]$ and $[\mathrm{NH}_{4}^{~+}]/[{\mathrm{SO}_{4}}^{2}-]$ in $\mathrm{NH_{4}}^{+}$ -rich conditions indicate the homogeneous chemical reaction between nitric and ammonia to generate $\mathrm{NO}_{3}^{\mathrm{~-~}}$ (Pathak et al., 2004, 2009):
$$
H N O_{3}(g)+N H_{3}(g)\leftrightarrow N H_{4}N O_{3}(s,a q)
$$
While, in Pathak's study, no relationship was observed between $[\mathrm{NO}_{3}^{\mathrm{~-}}]/[\mathrm{SO}_{4}^{\mathrm{~2~-}}]$ and $[\mathrm{NH}_{4}^{~+}]/[\mathrm{SO}_{4}^{~2~-}]$ in $\mathrm{NH_{4}}^{+}$ -poor conditions and the hydrolysis of $\mathrm{N}_{2}\mathrm{O}_{5}$ on preexisiting aerosol was suggested to be the contributor of the high level of $\mathrm{NO}_{3}{}^{-}$ (Pathak et al., 2009):
$$
N_{2}O_{5}(a q)+H_{2}O(a q)\leftrightarrow2N O_{3}^{-}(a q)+2H^{+}(a q)
$$
In summary, previous research conducted by Pathak et al. (Pathak et al., 2004, 2009) implied that homogeneous reaction between $\mathrm{HNO}_{3}$ (g) and $\mathrm{NH}_{3}$ (g) should be the dominant pathway of nitrate formation under $\mathrm{NH_{4}}^{+}$ -poor condition while heterogeneous reaction between $\mathrm{N}_{2}\mathrm{O}_{5}$ (aq) and $\mathrm{H}_{2}\mathrm{O}$ (aq) should play a dominanting role in $\mathrm{NH_{4}}^{+}$ -rich condition. However, in our research, as it is shown in Fig. 6, significant correlations were observed between $[\mathrm{NO}_{3}{}^{-}]/[\mathrm{SO}_{4}{}^{2}\,{}^{-}]$ and $[\mathrm{NH}_{4}^{\ +}]/}$ $[{\bf S O}_{4}^{\mathrm{~\,~2~-~}}]$ in Bohai Sea $(r=0.80)$ and Tanggu $(r=0.81)$ ), where they were both $\mathrm{NH_{4}}^{+}$ -poor conditions. This is consistent with He′s research (He et al., 2012). Therefore, instead of ammonium level, aerosol acidity was applied to analyze the pathway of nitrate formation. As it was discussed in Section 2.3.1, the ratio of $\mathrm{{C/A}}$ was used to divide the $\mathrm{PM}_{2.5}$ samples into more acidic aerosol group $\left(\mathrm{R_{C/A}}~<~0.9\right)$ and less acidic aerosol group $(\mathbf{R}_{\mathsf{C/A}}\geq0.9)$ . Almost all of $\mathrm{PM}_{2.5}$ samples in Bohai Sea were more acidic while in Tanggu sites, less acidic and more acidic samples were comparable.
In order to illustrate the relationship between nitrate formation and aerosol acidity, two groups of data were selected, as it is shown in Fig. 6. Group A represents a data set under more acidic condition in Tanggu and Group B represents a data set under less acidic condition. Through the comparison between Group A and Group B, it is easy to find that the ratio of $[\mathrm{NO}_{3}^{\mathrm{~-}}]/[\mathrm{SO}_{4}^{\mathrm{~2~-}}]$ tend to be higher in more acidic samples despite the same level of $[\mathrm{NH}_{4}^{~+}]/[{\mathrm{SO}_{4}}^{2}-]$ ratio. This indicated that nitrate was more likely to generate in a more acidic condition and heterogeneous reaction should be dominant in nitrate formation. Because previous laboratory and field researchs have found that the heterogeneous hydrolysis of $\mathrm{N}_{2}\mathrm{O}_{5}$ can be promoted by greater surface area, hygroscopicity and acidity of preexisting particulates. (McLaren et al., 2004; Anttila et al., 2006; Pathak et al., 2009; Hu and Abbatt, 1997; Hallquist et al., 2003; Martinez et al., 2000). Relatively high RH and aerosol acidity will increase the uptake coefficients of $\mathrm{N}_{2}\mathrm{O}_{5}$ on particulate surface (Hu and Abbatt, 1997; Kane et al., 2001; Hallquist et al., 2003). Consequently, the high RH $(\sim\!59\%$ in the daytime and $\sim\!72\%$ in the nighttime) and low in-situ pH $(<2)$ constructed an appropriate condition for nitrate generation from heterogeneous pathway.
Additionally, the good correlation between $[\mathrm{NO}_{3}^{\mathrm{~-}}]/[\mathrm{SO}_{4}^{\mathrm{~2~-}}]$ and $[\mathrm{NH}_{4}^{~+}]/[{\mathrm{SO}_{4}}^{2}-]$ in Bohai Sea $(r=0.80$ , Fig. 6) and Tanggu $(r=0.81$ , Fig. 6) proved that the homogeneous gas-phase reaction between nitric acid and ammonia was available for nitrate formation as well (Pathak et al., 2004, 2009).
4. Conclusions
High mass concentration of $\mathrm{PM}_{2.5}$ $\langle{\sim}140\,\upmu\mathrm{g}\,\mathrm{m}^{-\,3}\rangle$ was observed in the two sampling sites, Bohai Sea and Tanggu, from September 8th to October 8th in 2013. Regional pollution of $\mathrm{PM}_{2.5}$ around Bohai Bay was severe during autumn. Conclusion summarized from the five indicators of $\mathrm{PM}_{2.5}$ acidity indicated that $\mathrm{PM}_{2.5}$ was highly acidic in both two sampling sites (the average in-situ $\ensuremath{\mathbf{p}}\ensuremath{\mathbf{H}}$ was 1.2 in Bohai Sea and 1.3 in Tanggu) and there was a decrease in $\mathrm{PM}_{2.5}$ acidity from the marine area to coastal area. Because of the $\mathrm{NH_{4}}^{+}$ -poor conditions, sulfurous species were primarily a mixture of $\mathrm{H}_{2}\mathrm{SO}_{4}$ and $\mathrm{NH}_{4}\mathrm{HSO}_{4}$ in Bohai Sea and a mixture of $\mathrm{NH}_{4}\mathrm{HSO}_{4}$ and $(\mathrm{NH}_{4})_{2}\mathrm{SO}_{4}$ in Tanggu. Furthermore, there is not visible difference of $\mathrm{PM}_{2.5}$ acidity between daytime and nighttime found in Bohai Sea. However, stronger aerosol acidity was observed during the daytime than the nighttime in Tanggu sampling site which might be attributed to dramatically increasing concentration of ${\mathrm{so}}_{4}^{\ 2-}$ in the daytime. And $S O_{2}$ emission from coal burning in the local area and the photochemical oxidation from ozone were supposed to be the potential reasons for ${\mathrm{so}}_{4}^{\ 2-}$ increasement during the daytime. Aerosol acidity can promote nitrate generation through heterogeneous hydrolysis of $\mathrm{N}_{2}\mathrm{O}_{5}$ . The high mass concentration of nitrate in the marine and coastal aerosol was attributed to both homogeneous and heterogeneous reaction on the surface of particulate matter.
Acknowledgements
This work was also supported by the National Key R&D Program of China (grant numbers 2016YFC0208500, 2016YFC0208505); National Natural Science Foundation of China [grant numbers 21207069, 41205089]; Tianjin Science and Technology Project [grant number 16YFZCSF00260]; Natural Science Foundation of Tianjin (grant number 13JCQNJC08300). We would like to thank Tianjin Institute of Meteorological Science for the support. We also gratefully acknowledge Simon Clegg, Peter Brimblecombe and Anthony Wexler for sharing the E-AIM model.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https:// doi.org/10.1016/j.atmosres.2017.11.014.
References
Adams, P.J., Seinfeld, J.H., Koch, D.M., 1999. Global concentrations of tropospheric sulfate, nitrate, and ammonium aerosol simulated in a general circulation model. J. Geophys. Res.-Atmos. 104, 13791–13823.
Almeida, S.M., Freitas, M.C., Pio, C.A., Pinheiro, M.T., Felix, P., 2013. Fifteen years of nuclear techniques application to suspended particulate matter studies. J. Radioanal. Nucl. Chem. 297, 347–356.
Amdur, M.O., Chen, L.C., 1989. Furnace-generated acid aerosols speciation and pulmonary effects. Environ. Health Perspect. 79, 147–150.
Ansari, A.S., Pandis, S.N., 2000. Water absorption by secondary organic aerosol and its effect an inorganic aerosol behavior. Environ. Sci. Technol. 34, 71–77.
Anttila, T., Kiendlerscharr, A., Tillmann, R., Mentel, T.F., 2006. On the reactive uptake of gaseous compounds by organic-coated aqueous aerosols: theoretical analysis and application to the heterogeneous hydrolysis of $\mathrm{N}_{2}\mathrm{O}_{5}$ . J. Phys. Chem. A 110 (35), 10435 (2006).
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, 903–912.
Boucher, O., Anderson, T.L., 1995. General circulation model assessment of the sensitivity of direct climate forcing by anthropogenic sulfate aerosols to aerosol size and chemistry. J. Geophys. Res. - Atmos. 100, 26117–26134.
Chameides, W.L., Stelson, A.W., 1992. Aqueous-phase chemical processes in deliquescent sea-salt aerosols: a mechanism that couples the atmospheric cycles of S and sea-salt. J. Geophys. Res. 97, 20565–20580.
Chan, M.N., Surratt, J.D., Claeys, M., Edgerton, E.S., Tanner, R.L., Shaw, S.L., Zheng, M., Knipping, E.M., Eddingsaas, N.C., Wennberg, P.O., Seinfeld, J.H., 2010. Characterization and quantification of isoprene-derived epoxydiols in ambient aerosol in the Southeastern United States. Environ. Sci. Technol. 44, 4590–4596.
Cheng, Y.F., Zheng, G.J., Wei, C., Mu, Q., Zheng, B., Wang, Z.B., Gao, M., Zhang, Q., He, K.B., Carmichael, G., Pöschl, U., Su, H., 2016. Reactive nitrogen chemistry in aerosol water as a source of sulfate during haze events in China. Sci. Adv. 2016 (2), e1601530. http://dx.doi.org/10.1126/sciadv.1601530.
China Statistical Yearbook, 2013. Part XVI: Transport. Post and Telecommunication Services.
Chinese Fishery Statistical Yearbook, 2013. Part III: Production Factors.
Clegg, S.L., Brimblecombe, P., Wexler, A.S., 1998. Thermodynamic model of the system $\mathrm{H^{+}-N H_{4}}^{+}-\mathrm{SO_{4}}^{2}-\mathrm{-NO_{3}}^{-}-\mathrm{H_{2}O}$ at tropospheric temperatures. J. Phys. Chem. A 102, 2137–2154.
Crumeyrolle, S., Gomes, L., Tulet, P., Matsuki, A., Schwarzenboeck, A., Crahan, K., 2008. Increase of the aerosol hygroscopicity by cloud processing in a mesoscale convective system: a case study from the AMMA campaign. Atmos. Chem. Phys. 8, 6907–6924. http://dx.doi.org/10.5194/acp-8-6907-2008.
Edney, E.O., Kleindienst, T.E., Jaoui, M., Lewandowski, M., Offenberg, J.H., Wang, W., Claeys, M., 2005. Formation of 2-methyl tetrols and 2-methylglyceric acid in secondary organic aerosol from laboratory irradiated isoprene $\mathrm{NO}_{\mathrm{X}}/\mathrm{SO}_{2},$ air mixtures and their detection in ambient $\mathrm{PM}_{2.5}$ samples collected in the eastern United States. Atmos. Environ. 39, 5281–5289.
Gao, S., Keywood, M., Ng, N.L., Surratt, J., Varutbangkul, V., Bahreini, R., Flagan, R.C., Seinfeld, J.H., 2004a. Low-molecular-weight and oligomeric components in secondary organic aerosol from the ozonolysis of cycloalkenes and alpha-pinene. J. Phys. Chem. A 108, 10147–10164.
Gao, S., Ng, N.L., Keywood, M., Varutbangkul, V., Bahreini, R., Nenes, A., He, J., Yoo, K.Y., Beauchamp, J.L., Hodyss, R.P., Flagan, R.C., Seinfeld, J.H., 2004b. Particle phase acidity and oligomer formation in secondary organic aerosol. Environ. Sci. Technol. 38, 6582–6589.
Gu, J.X., Bai, Z.P., Li, W.F., Wu, L.P., Liu, A.X., Dong, H.Y., Xie, Y.Y., 2011. Chemical composition of $\mathrm{PM}_{2.5}$ during winter in Tianjin, China. Particuology 9 (3), 215–221.
Hallquist, M., Stewart, D.J., Stephenson, S.K., Cox, R.A., 2003. Hydrolysis of $\mathrm{N}_{2}\mathrm{O}_{5}$ on submicron sulfate aerosols. Phys. Chem. Chem. Phys. 5, 3453–3463.
He, K.B., Zhao, Q., Ma, Y.L., Yang, F., Shi, Z., Chen, G., 2012. Spatial and seasonal variability of $\mathrm{PM}_{2.5}$ acidity at two Chinese megacities: insights into the formation of secondary inorganic aerosols. Atmos. Chem. Phys. 12, 1377–1395.
Hsu, S.-C., Liu, S.C., Kao, S.-J., Jeng, W.-L., Huang, Y.-T., Tseng, C.-M., Tsai, F., Tu, J.-Y., Yang, Y., 2007. Water-soluble species in the marine aerosol from the northern South China Sea: high chloride depletion related to air pollution. J. Geophys. Res.-Atmos. 112 (D19304).
Hu, J.H., Abbatt, J.P.D., 1997. Reaction probabilities for $\mathrm{N}_{2}\mathrm{O}_{5}$ hydrolysis on sulfuric acid and ammonium sulfate aerosols at room temperature. J. Phys. Chem. A 101, 871–878.
Huang, X.F., Yu, J.Z., 2007. Is vehicle exhaust a significant primary source of oxalic acid in ambient aerosols? Geophys. Res. Lett. 34, L02808. http://dx.doi.org/10.1029/ 2006 GL028457.
Huang, X.F., He, L.Y., Li, X.A., Feng, N., Hu, M., Niu, Y.W., Zeng, L.W., 2010. 5-Year study of rainwater chemistry in a coastal mega-city in South China. Atmos. Res. 97, 185–193.
Huang, X., Qiu, R., Chan, C.K., Kant, P.R., 2011. Evidence of high $\mathrm{PM}_{2.5}$ strong acidity in ammonia-rich atmosphere of Guangzhou, China: transition in pathways of ambient ammonia to form aerosol ammonium at $[\mathrm{NH_{4}}^{+}]/[\mathrm{SO_{4}}^{2-}]=1.5$ . Atmos. Res. 99 (3–4), 488–495.
Iinuma, Y., Boge, O., Miao, Y., Sierau, B., Gnauk, T., Herrmann, H., 2005. Laboratory studies on secondary organic aerosol formation from terpenes. Faraday Discuss. 130, 279–294.
Jacobson, M.Z., 1999. Studying the effects of calcium and magnesium on size-distributed nitrate and ammonium with EQUISOLV II. Atmos. Environ. 33, 3635–3649.
Jang, M.S., Czoschke, N.M., Sangdon, L., Kamens, R.M., 2002. Heterogeneous atmospheric aerosol production by acidcatalyzed particle-phase reactions. Science 298, 814–817.
Kane, S.M., Caloz, F., Leu, M.-T., 2001. Heterogeneous uptake of gaseous $\mathrm{N}_{2}\mathrm{O}_{5}$ by $(\mathrm{NH}_{4})_{2}\mathrm{SO}_{4},$ $\mathrm{NH}_{4}\mathrm{HSO}_{4},$ and $\mathrm{H}_{2}\mathrm{SO}_{4}$ aerosols. J. Phys. Chem. A 105, 6465–6470.
Kong, S.F., Han, B., Bai, Z.P., Chen, L., Shi, J.W., Xu, J.W., 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, 4681–4694.
Larssen, T., Lydersen, E., Tang, D.G., He, Y., Gao, J.X., Liu, H.Y., Duan, L., Seip, H.M., Vogt, R.D., Mulder, J., Shao, M., Wang, Y.H., Shang, H., Zhang, X.S., Solberg, S., Aas, W., Okland, T., Eilertsen, O., Angell, V., Liu, Q.R., Zhao, D.W., Xiang, R.J., Xiao, J.S., Luo, J.H., 2006. Acid rain in China. Environ. Sci. Technol. 40, 418–425.
Likens, G.E., Driscoll, C.T., Buso, D.C., 1996. Long-term effects of acid rain: response and recovery of a forest ecosystem. Science 272, 244–246.
Lippmann, M., Xiong, J.Q., Li, W., 2000. Development of a continuous monitoring system for $\mathrm{PM_{10}}$ and components of $\mathrm{PM}_{2.5}$ . App. Occup. Environ. Hyg. 15, 57–67.
Liu, L.J.S., Burton, R., Wilson, W.E., Koutrakis, P., 1996. Comparison of aerosol acidity in urban and semi-rural environments. Atmos. Environ. 30, 1237–1245.
Liu, Z.R., Hu, B., Zhang, J.K., Yu, Y.C., Wang, Y.S., 2016a. Characteristics of aerosol size distributions and chemical compositions during wintertime pollution episodes in Beijing. Atmos. Res. 168, 1–12.
Liu, B.S., Song, N.A., Dai, Q.L., Mei, R.B., Sui, B.H., Bi, X.H., Feng, Y.C., 2016b. Chemical composition and source apportionment of ambient $\mathrm{PM}_{2.5}$ during the non-heating period in Taian, China. Atmos. Res. 170, 23–33.
Martinez, M., Perner, D., Hackenthal, E.-M., Kulzer, S., Schutz, L., 2000. $\mathrm{NO}_{3}$ at Helgoland during the NORDEX campaign in October 1996. J. Geophys. Res. 105, 22685–22695.
McLaren, R., Salmon, R.A., Liggio, J., Hayden, K.L., Anlauf, K.G., Leaitch, W.R., 2004. Nighttime chemistry at a rural site in the Lower Fraser Valley. Atmos. Environ. 38, 5837–5848.
Meng, Z., Seinfeld, J.H., Saxena, P., Kim, Y.P., 1995. Atmospheric gas–aerosol equilibrium, IV: thermodynamics of carbonate. Aerosol Sci. Technol. 23, 131–154.
Murphy, J.G., Gregoire, P.K., Tevlin, A.G., Wentworth, G.R., Ellis, R.A., Markovic, M.Z., VandenBoer, T.C., 2017. Observational constraints on particle acidity using measurements and modelling of particles and gases. Faraday Discuss. 2017 (Advance Article). http://dx.doi.org/10.1039/C7FD00086C.
Nemitz, E., Sutton, M.A., Wyers, G.P., Jongejan, P.A.C., 2004. Gas-particle interactions above a Dutch heathland: I surface exchange fluxes of $\mathrm{NH}_{3}$ , $S0_{2}$ HNO3, HCl. Atmos. Chem. Phys. 4, 989–1005.
Nenes, A., Pandis, S.N., Pilinis, C., 1998. ISORROPIA: a new thermodynamic equilibrium model for multiphase multi-component inorganic aerosols. Aquat. Geochem. 4, 123–152.
Ni, T.R., Li, P.H., Han, B., Bai, Z.P., Ding, X., Wang, Q.W., Huo, J., Lu, B., 2013. Spatial and temporal variation of chemical composition and mass closure of ambient $\mathrm{PM_{10}}$ in Tianjin, China. Aeros. Air Qual. Res. 13, 1832–1846.
Olson, C.N., Galloway, M.M., Yu, G., Hedman, C.J., Lockett, M.R., Yoon, T., Stone, E.A., Smith, L.M., Keutsch, F.N., 2011. Hydroxycarboxylic acid-derived organosulfates: synthesis, stability, and quantification in ambient aerosol. Environ. Sci. Technol. 45, 6468–6474.
Pathak, R.K., Liou, P.K.K., Chan, C.K., 2004. Characteristics of aerosol acidity in Hong Kong. Atmos. Environ. 38, 2965–2974.
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. Atmos. Chem. Phys. 9, 1711–1722.
Pathak, R.K., Wanga, T., Wu, W.S., 2011. Nighttime enhancement of $\mathrm{PM}_{2.5}$ nitrate in ammonia-poor atmospheric conditions in Beijing and Shanghai: plausible contributions of heterogeneous hydrolysis of $\mathrm{N}_{2}\mathrm{O}_{5}$ and $\mathrm{HNO}_{3}$ partitioning. Atmos. Environ. 45 (5), 1183–1191.
Rengarajan, R., Sudheer, A.K., Sarin, M.M., 2011. Aerosol acidity and secondary organic aerosol formation during wintertime over urban environment in western India. Atmos. Environ. 45 (11), 1940–1945.
Sievering, H., Gorman, E., Ley, T., Pszenny, A.A.P., Spring Young, M., Boatman, J.F., Kim, Y., Nagamoto, C., Wellman, D.L., 1995. Ozone oxidation of sulfur in sea-salt aerosol particles during the Azores Marine Aerosol and Gas Exchange experiment. J. Geophys. Res. 100, 23075–23081.
Sudheer, R.R.K., Sarin, M.M., 2011. Aerosol acidity and secondary organic aerosol formation during wintertime over urban environment in western India. 2011. Atmos. Environ. 45 (11), 1940–1945.
Surratt, J.D., Lewandowski, M., Offenberg, J.H., Jaoui, M., Kleindienst, T.E., Edward Edney, E.O., Seinfeld, J.H., 2007. Effect of acidity on secondary organic aerosol formation from isoprene. Environ. Sci. Technol. 41, 5363–5369.
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 $\mathrm{PM}_{2.5}$ . Atmos. Chem. Phys. 14, 9469–9479.
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.
Utell, M.J., Looney, R.J., 1995. Environmentally induced asthma. Toxicol. Lett. 82 (83), 47–53.
Van Oss, R., Duyzer, J., Wyers, P., 1998. The influence of gas to particle conversion on measurements of ammonia exchange over forest. Atmos. Environ. 32, 465–471.
Watson, J.G., 2002. Visibility: science and regulation. J. Air Waste Manage. Assoc. 52, 628–713.
Wu, L., Feng, Y.C., Wu, J.H., Zhu, T., Bi, X.H., Han, B., Yang, W.H., Yang, Z.Q., 2009. Secondary organic carbon quantification and source apportionment of $\mathrm{PM_{10}}$ in Kaifeng, China. J. Environ. Sci. 21, 1353–1362.
Wu, H., Zhang, Y.F., Han, S.Q., Wu, J.H., Bi, X.H., Shi, G.L., Wang, J., Yao, Q., Cai, Z.Y., Liu, J.L., Feng, Y.C., 2015. Vertical characteristics of $\mathrm{PM}_{2.5}$ during the heating season in Tianjin, China. Sci. Total Environ. 523, 152–160.
Xu, H., Bi, X.H., Zheng, W.W., Wu, J.H., Feng, Y.C., 2015. Particulate matter mass and chemical component concentrations over four Chinese cities along the western Pacific coast. Environ. Sci. Pollut. Res. 22, 1940–1953.
Xue, Y.H., Wu, J.H., Feng, Y.C., Dai, L., Bi, X.H., Li, X., 2010. Source characterization and apportionment of $\mathrm{PM_{10}}$ in Panzhi-hua, China. Aerosol Air Qual. Res. 10, 367–377.
Xue, J., Lau, A.K.H., Yu, J.Z., 2011. A study of acidity on $\mathrm{PM}_{2.5}$ in Hong Kong using online ionic chemical composition measurements. Atmos. Environ. 45 (38), 7081–7088.
Yang, Y.W., 2014. Study on the Characteristics of $\mathrm{PM}_{2.5}$ Acidity on the High Temporal Resolution Observations [D]. Fudan University, Shanghai (in Chinese).
Yao, X.H., Ling, T.Y., Fang, M., Chan, C.K., 2006. Comparison of thermodynamic predictions for in situ pH in $\mathrm{PM}_{2.5}$ . Atmos. Environ. 40, 2835–2844.
Yao, X.H., Rehbein, P.J.G., Lee, C.J., Evans, G.J., Corbin, J., Jeong, C.-H., 2011. A study on the extent of neutralization of sulphate aerosol through laboratory and field experiments using an ATOFMS and a GPIC. Atmos. Environ. 45 (34), 6251–6256.
Zhang, X.Y., Cao, J.J., Li, L.M., Arimoto, R., Cheng, Y., Huebert, B., Wang, D., 2002. Characterization of atmospheric aerosol over Xi'an in the south margin of the Loess Plateau, China. Atmos. Environ. 36, 4189–4199.
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. Environ. Sci. Technol. 41, 3213–3219.
Zhang, F., Chen, Y., Tian, C., Wang, X., Huang, G., Fang, Y., Zong, Z., 2014. Identification and quantification of shipping emissions in Bohai Rim. Sci. Total Environ. 497, 570–577.
Zhang, J.Q., Chen, J., Xia, X.G., Che, H.H., Fan, X.H., Xie, Y.Y., Han, Z.W., Chen, H.B., Lu, D.R., 2016. Heavy aerosol loading over the Bohai Bay as revealed by ground and satellite remote sensing. Atmos. Environ. 124, 252–261.
Zhao, X.J., Zhao, P.S., Xu, J., Meng, W., Pu, W.W., Dong, F., He, D., Shi, Q.F., 2013. Analysis of a winter regional haze event and its formation mechanism in the North China Plain. Atmos. Chem. Phys. 13, 5685–5696.
Ziemba, L.D., Fischer, E., Griffin, R.J., Talbot, R.W., 2007. Aerosol acidity in rural New England: Temporal trends and source region analysis. J. Geophys. Res.-Atmos. 112 (D10S22). http://dx.doi.org/10.1029/2006jd007605.
|
atmosphere
|
0108
|
10.1016/j.atmosres.2014.08.014
|
image
|
Fig. 1. Location of the sampling site (five-pointed star: Shandong University; triangle: business office) in Jinan, Shandong Province, China.
|
|
atmosphere
|
0108
|
10.1016/j.atmosres.2014.08.014
|
table
|
Table 1 Characteristics of the indoor environment and sampling locations as well as sample numbers.
|
|
atmosphere
|
0108
|
10.1016/j.atmosres.2014.08.014
|
table
|
Table 2 Average concentrations of $\mathrm{PM}_{2.5}\,(\upmu\mathrm{g}\,\,\mathrm{m}^{-3})$ , inorganic ions $\left(\upmu\mathrm{g}\:\mathsf{m}^{-3}\right)$ ) and 15 PAHs $\left(\mathfrak{n g}\;\mathfrak{m}^{-3}\right)$ ) monitored in indoor and outdoor air samples in summer and autumn.
|
|
atmosphere
|
0108
|
10.1016/j.atmosres.2014.08.014
|
image
|
Fig. 2. Trends for the variations in the mass concentrations of indoor/outdoor $\mathrm{PM}_{2.5}$ in summer and autumn.
|
|
atmosphere
|
0108
|
10.1016/j.atmosres.2014.08.014
|
image
|
Fig. 3. Mean contributions of individual PAHs to total PAH concentrations. Indoor samples are shown in white, and outdoor samples are shown in black.
|
|
atmosphere
|
0108
|
10.1016/j.atmosres.2014.08.014
|
image
|
Fig. 4. The indoor-to-outdoor concentration ratios of $\mathrm{PM}_{2.5,}$ water-soluble ions and PAHs during the daytime and nighttime in summer and autumn.
|
|
atmosphere
|
0108
|
10.1016/j.atmosres.2014.08.014
|
image
|
Fig. 5. Diurnal and nocturnal variations in the percentage contributions of 5–6-ring PAHs to total PAHs as well as the temperature fluctuation over the sampling period.
|
|
atmosphere
|
0108
|
10.1016/j.atmosres.2014.08.014
|
table
|
Table 3 Diagnostic ratios of particulate PAHs in the outdoor air at the office.
|
|
atmosphere
|
0108
|
10.1016/j.atmosres.2014.08.014
|
table
|
Table 4 Toxic-equivalent concentrations $\left(\mathrm{ng~m}^{-3}\right)$ of particulate PAHs.
|
|
atmosphere
|
0108
|
10.1016/j.atmosres.2014.08.014
|
text
| Not supported with pagination yet |
Indoor/outdoor relationships and diurnal/nocturnal variations in water-soluble ion and PAH concentrations in the atmospheric $\mathsf{P M}_{2.5}$ of a business office area in Jinan, a heavily polluted city in China
Yanhong Zhu a,b, Lingxiao Yang a,b,⁎, Chuanping Meng a, Qi Yuan a, Chao Yan a, Can Dong a, Xiao Sui a, Lan Yao a, Fei Yang a, Yaling Lu a, Wenxing Wang a,c
a Environment Research Institute, Shandong University, Jinan 250100, China b School of Environmental Science and Engineering, Shandong University, Jinan 250100, China c Chinese Research Academy of Environmental Sciences, Beijing 100012, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 24 April 2014
Received in revised form 27 August 2014
Accepted 27 August 2014
Available online 16 September 2014
Keywords:
$\mathrm{PM}_{2.5}$
Water-soluble ions
PAHs
Indoor-to-outdoor ratio
Diurnal and nocturnal variations
Indoor/outdoor and diurnal/nocturnal variations in $\mathrm{PM}_{2.5}$ and associated water-soluble ions and polycyclic aromatic hydrocarbons (PAHs) were examined in a business office during the summer and autumn of 2010 in Jinan, China. Both indoor and outdoor $\mathrm{PM}_{2.5}$ levels were higher than the value recommended by the WHO, and outdoor sources were found to be the major contributors to indoor $\mathrm{PM}_{2.5}$ $S0_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ and $\mathrm{NH}_{4}^{+}$ were the dominant water-soluble ions in both indoor and outdoor particles. During daytime, $\mathrm{NO}_{3}^{-}$ mainly came from indoor sources, which was related to the temperature difference between the indoor and outdoor air. During daytime, the 15 monitored PAHs were all largely from indoor sources, while during nighttime, the 3–4-ring PAHs were mainly generated indoors and the 5–6-ring PAHs predominantly came from the outdoor air. The diurnal/ nocturnal variations of PAHs suggested that gas/particle partitioning driven by temperature makes a significant contribution to the variation in PAH concentrations. The diagnostic ratios revealed that biomass burning had an important contribution to outdoor PAH concentrations in autumn. The results of a risk assessment of PAH pollution suggested that indoor PAHs present more carcinogenic and mutagenic risks during daytime. Our results indicated that serious indoor air pollution in a business office presents a high health risk for workers.
$\circledcirc$ 2014 Elsevier B.V. All rights reserved.
1. Introduction
Over the last twenty years, considerable efforts around the world have been made to clarify the influence of indoor air quality on human health (Gupta et al., 1996; Hayakawa et al., 1999; Lee and Chang, 2000; Jo and Seo, 2005; Kotzias et al., 2009; Yang et al., 2009; Pegas et al., 2011; Hasheminassab et al., 2014). The results of these studies showed that people are exposed to a multitude of chemical and biological stressors, some of which cause health problems (allergy, asthma, sensory irritation, lung cancer, etc.) (Samet and Spengler, 2003; Bernstein et al., 2008; Rios et al., 2009; Jie et al., 2011; Zhou et al., 2013; Jovanović et al., 2014). However, previous studies of indoor air have mainly been conducted in residential and school buildings (e.g., Blondeau et al., 2005; Fromme et al., 2008; Polidori et al., 2009; Massey et al., 2012; Krugly et al., 2014; Rivas et al., 2014), while, in comparison, few studies have been performed in business offices (Saraga et al., 2010; Sangiorgi et al., 2011, 2013), despite the fact that they are the workplace for most urban office workers. In addition, few studies have focused on the diurnal/nocturnal differences in indoor and outdoor pollution levels (Reisen and Arey, 2005; Wang et al., 2007, 2010; Zimmermann et al., 2012; Souza et al., 2014), though the differences in anthropogenic emissions between daytime and nighttime could affect the contributions from various sources of pollution.
$\mathsf{P M}_{2.5}$ , which can easily move from the outdoors into the indoors due to its effective penetration ability (Massey et al., 2012; Barraza et al., 2014; Hassanvand et al., 2014), is an important pollutant in indoor environments. Water-soluble ions are a major chemical component of $\mathsf{P M}_{2.5}$ and account for one-third or more of the particulate mass in Chinese urban regions (Saliba et al., 2009; Chithra and Shiva Nagendra, 2013; Hassanvand et al., 2014). Moreover, the acidity of $\mathsf{P M}_{2.5}.$ , which depends on the balance between sulphate and nitrate as the acidic compounds and ammonium as a neutralizing species (Seinfeld and Pandis, 2006), has also been associated with negative effects on human health, mainly in terms of lung function. In addition, by acting as surface-active reagents, water-soluble ions can increase the solubility of toxic organic compounds, such as PAHs, and therefore increase their toxicity to humans (Pongpiachan et al., 2013; Krugly et al., 2014; Kamal et al., 2014). PAHs have received increased attention in recent years in indoor air pollution research due to their potential cytotoxicity, mutagenicity and carcinogenicity in humans (WHO, 1998) and their ability to directly or indirectly damage DNA (Baird et al., 2005; Novotna et al., 2007; Li et al., 2014). Therefore, the presence of water-soluble ions and PAHs within $\mathsf{P M}_{2.5}$ strongly increases their potential for adverse health effects.
Shandong Province, which is located in northern China, is China's third-biggest economic power, and anthropogenic emissions there contributed approximately $10.00\%$ of $S0_{2}$ $8.46\%$ of NOx, $9.02\%$ of VOC, $9.11\%$ of $\mathsf{P M}_{2.5}$ $7.33\%$ of BC and $6.66\%$ of OC to China's overall emissions in 2006 (National Bureau of Statistics of China, 2009; Zhang et al., 2009). Jinan, as the capital of Shandong Province, was listed in the group of large cities with the highest pollution concentrations, including $S0_{2}$ , NOx, PAHs and $\mathsf{P M}_{2.5}$ (Baldasano et al., 2003; Gao et al., 2011). In such industrial cities, scientific evidence has shown that indoor air environments can be become seriously polluted (Balasubramanian and Lee, 2007). Moreover, with economic development, the number of large office buildings equipped with air-conditioning is increasing. Advances in information technology have also increased the quantity of and transformed the nature of electronic equipment (such as computers, printers, copier machines and fax machines) used by office workers. These office tools have been found to be a source of ozone, particulate matter and volatile organic compounds and have a serious impact on indoor air quality (Wolkoff et al., 2006; Destaillats et al., 2006). Hence, the indoor air quality in business offices could have a significant influence on human health. The objectives of this study were as follows: (1) to measure and understand the indoor and outdoor $\mathsf{P M}_{2.5}$ mass concentrations and associated water-soluble ion and PAH concentrations in a business office and in Shandong University; (2) to examine the indoor to outdoor ratios $\left(\mathrm{I}/\mathrm{O}\right)$ and the relationships $(\mathbb{R}^{2})$ between the indoor and outdoor concentrations of $\mathsf{P M}_{2.5}$ , water-soluble ions and PAHs and analyse their source implications; (3) to illustrate the diurnal and nocturnal variations in $\mathsf{P M}_{2.5},$ , water-soluble ions and PAHs and assess their potential controlling factors; and (4) to assess the sources and health risks of PAHs in the air of indoor and outdoor environments in a business office.
2. Methodology
2.1. Site description
As shown in Fig. 1, a business office was selected as the indoor sampling site, and it is located in an area characterized by considerable human activity and intense traffic. There were no smoking areas at the indoor site. However, electronic equipment (computers, printers, copier machines, etc.) was operated during daytime for all of the sampling periods. The air supply systems were operated in summer, but natural ventilation was provided in autumn. The outdoor sampling site was located on the roof (approximately $20.36\;\mathrm{m}$ above the ground) of the Information Science and Engineering School, Central Campus, Shandong University. This building is surrounded by commercial and residential areas, and a busy road is located in front of the site.
2.2. Sampling strategy
Two sampling campaigns were conducted (Table 1): summer (17 July–26 July 2010) and autumn (11 October–27 October 2010). The sampling was performed simultaneously at both the indoor and outdoor sites. For daytime and nighttime $\mathsf{P M}_{2.5}$ measurements, indoor and outdoor sampling works were conducted from 8:30 to 18:30 and from 19:30 to 8:00 the next day, respectively. To guarantee the quality of the $\mathsf{P M}_{2.5}$ measurements and chemical analyses, field blank filters were collected during each campaign. $\mathsf{P M}_{2.5}$ samples were not collected on July 20 or during the nighttime of October 18, October 20 and October 21 due to rainfall. Half of each filter was used to determine the ionic constituents, and the other half was used for the PAH analysis.
2.3. $P M_{2.5}$ collection
In this study, two TH-16A Intelligent $\mathsf{P M}_{2.5}$ samplers (Wuhan Tianhong Corporation, China) were used to collect indoor and outdoor $\mathsf{P M}_{2.5}$ samples simultaneously at a flow rate of $100\,\mathrm{L}\,\mathrm{min}^{-1}$ . Before and after sampling, the airflow rates of the two samplers were calibrated. All quartz fibre-filters were pre-treated by baking at $600~^{\circ}\mathrm{C}$ for $4\;\mathrm{h}$ to remove any organic contaminants. After baking, the filters were transferred to a room with a constant temperature and humidity (temperature of $20\pm1~^{\circ}\mathrm{C},$ relative humidity of $50\pm2\%$ for $24\,\mathrm{h}$ and then, they were weighed using a Sartorius analytical balance (detection limit $0.001~\mathrm{~mg})$ . Loaded filters were similarly conditioned for $24\,\mathrm{~h~}$ before weighing. The weighed filters were placed in plastic petri dishes and stored at a $-4~^{\circ}\mathsfit{C}$ in a refrigerator before further analysis of PAHs.
2.4. Analysis of ionic components
Particulates on the quartz filters were dissolved completely in an ultrasonic bath in ultra-pure water of 18.2 MU cm (purified by Millipore Water Purification System) for $60\ \mathrm{min}$ . The water extracts were filtered through a $13{-}\mathrm{mm}$ filter with
$0.2\ \upmu\mathrm{m}$ pores (13JP020AN, ADVANTEC). An ion chromatography system (Dionex ICs-90, Dionex Corporation, USA) was deployed to analyse the water-soluble ions in $\mathsf{P M}_{2.5}$ . The anions, including $\mathsf{F}^{-}$ , $\mathsf{C l}^{-}$ , $\mathrm{NO}_{2}^{-}$ , $\mathrm{NO}_{3}^{-}$ and $S0_{4}^{2-}$ , were analysed using an AS14A Column with an AMMS 300 Suppresser and were eluted with $3.5\,\mathrm{mmol}\,\mathrm{L}^{-1}\,\mathrm{Na}_{2}\mathrm{CO}_{3}/1.0\,\mathrm{mmol}\,\mathrm{L}^{-1}\,\mathrm{NaHCO}_{3}$ The cations, $\mathtt{N a}^{+}$ , $\mathrm{NH}_{4}^{+}$ , $\mathsf{K}^{+}$ , ${\mathrm{Mg}}^{2+}$ and ${\mathsf{C}}{\mathsf{a}}^{2+}$ , were analysed using a CS12A Column with a CSRS Ultra II Suppresser and were eluted with $20.0\ \mathrm{mmol\}\mathrm{L}^{-1}$ methanesulfonic acid. The minimum detection limit (MDL) values for all of the ions ranged from 0.01 to $0.08\,\upmu\mathrm{g}\,\mathfrak{m}^{-3}$ .
2.5. Analysis of PAH components
The samples were extracted with a 4:1 $({\boldsymbol{\mathbf{v}}}/{\boldsymbol{\mathbf{v}}})$ mixture of dichloromethane and n-hexane using an accelerated solvent extraction (ASE 300, Dionex) for $15\,\mathrm{min}$ . Then, the extracts were concentrated in a rotary evaporator under a purified nitrogen stream $(99.99\%,$ purified by an activated carbon column) to a volume of $1~\mathrm{mL}$ . Internal standards, including naphthalene-D8, acenaphthene-D10, phenanthrene-D10, chrysene-D10 and perylene-D12, were added into the samples immediately prior to analysis.
The samples were analysed using a gas chromatograph (GC) system (Agilent 6890N) that was equipped with an Agilent 5973 Network Mass Selective Detector. A $30\;\mathrm{m}\times0.25\;\mathrm{mm}\times$ $0.50\,\upmu\mathrm{m}$ DB5 capillary column was used for the separation of the PAHs. The injections were splitless, and the sample volume was
1 μL. High-purity helium was used as the carrier gas at a constant flow rate of $1\;\mathrm{mL}\,\mathrm{min}^{-1}$ . The chromatographic conditions were $60~^{\circ}C$ (maintained for $1\ \mathrm{min}$ ), ramping from 60 to $290~^{\circ}\mathrm{C}$ at $18~{}^{\circ}\mathrm{C}~\operatorname*{min}^{-1}$ and $290~^{\circ}\mathrm{C}$ (maintained for $15\;\mathrm{min}$ ). The MS detection was conducted by electron impact ionization of $70\,\mathrm{eV}$ and the mass fragment scan was from 50 to 550 amu. The organic compounds in the samples were identified by comparing the sample mass spectra with those in the National Institute of Standards and Technology mass spectral reference library (NIST05a) and were further confirmed by comparison with standards for retention time and mass spectra. Compound quantification was based on the calibration curves of the corresponding standards. The values of the MDLs for the PAHs ranged from $0.01\;\mathrm{ng}\;\mathrm{m}^{-3}$ (Ace) to $0.14\,\mathrm{ng}\;\mathrm{m}^{-3}$ (BghiP).
The PAHs monitored in this study are abbreviated as follows: acenaphthylene (Acy, $\mathrm{m}/\mathrm{z}~152\$ ), acenaphthene (Ace, $\mathrm{m}/\mathrm{Z}\,154\$ , fluorene (Fl, $\boldsymbol{\mathrm{m}}/\boldsymbol{\mathrm{Z}}\,166\$ , phenanthrene (Phe, $\mathrm{m}/\mathrm{z}\,178\$ , anthracene (Ant, $\mathfrak{m}/\mathfrak{z}\,178$ ), fluoranthene (Flu, $\mathrm{m}/\mathrm{z}\,202$ ), pyrene (Pyr, $\scriptstyle\mathrm{m}/\mathrm{Z}\ 202\mathrm{\Omega}$ ), benz[a]anthracene (BaA, $\mathfrak{m}/\mathfrak{z}\ 228$ ), chrysene (Chr, $\mathfrak{m}/\mathbf{Z}\,228$ ), benzo[b]fluoranthene (BbF, $\mathrm{m}/\mathrm{z}\,252$ ), benzo[k] fluoranthene (BkF, $\mathsf{m}/\mathsf{Z}\ 252\$ ), benzo[a]pyrene (BaP, $\mathrm{m}/\mathrm{z}\,252$ ), indeno[1,2,3-cd]pyrene (IcdP, $\scriptstyle\mathrm{m}/\tau$ 276), dibenz[a, h]anthracene (DBA, $\mathrm{m}/z\,278$ ) and benzo[ghi]perylene (BghiP, $\scriptstyle{\mathrm{m}}/\mathrm{Z}\,276\mathrm{)}$ .
2.6. Quality control
The analytical method was based on the USEPA method TO-13. Field blanks were performed to identify the level of background contamination from the sampling sites. Method blanks and spiked filter samples were also analysed to ensure the quality of the pre-treatment, ICs and GC–MS analysis steps. These blanks were extracted and analysed in the same manner as the samples. Neither water-soluble ions nor PAHs were detected in any of the blank samples. In this study, concentrations of $\mathsf{F}^{-}$ and $\Nu0_{2}^{-}$ below the detection limit were excluded from discussion. The mean recoveries for the 15 PAHs ranged from $70.45\%$ to $125.32\%,$ , and the data reported in the study were corrected for their recoveries. Because the recovery of naphthalene-D8 was low, the naphthalene results were not included in this study. The PAHs found in the field blanks were generally below the instrumental detection limits (IDLs) (the amount of analyte that would generate a signal-to-noise ratio of 3:1). The MDLs were assigned to be 3 times the IDLs.
3. Results and discussion
3.1. Occurrence of $P M_{2.5}$
3.1.1. $P M_{2.5}$ mass concentrations
As shown in Table 2, the mean indoor $\mathsf{P M}_{2.5}$ concentration levels were $51.52\,\pm\,28.22$ and $93.52\,\pm\,41.07~\upmu\mathrm{g}\,\mathrm{~m}^{-3}$ in summer and autumn, respectively, which were 2.06 and 3.74 times the WHO-recommended daily $\mathsf{P M}_{2.5}$ standard value of $25\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ (WHO, 2006). Thus, the daily concentration overlimit ratio was $86.72\%{-}100.00\%$ , indicating that the indoor air quality was very poor. In summer and autumn, the mean outdoor $\mathsf{P M}_{2.5}$ concentration levels were $115.92\pm44.84$ and $150.88\pm73.07\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ , which were 1.77 and 2.32 times the outdoor $\mathsf{P M}_{2.5}$ daily standard of $65~\upmu\mathrm{g}~\mathrm{m}^{-3}$ recommended by the National Ambient Air Quality Standard (NAAQS), suggesting that Jinan suffers from serious air pollution.
3.1.2. Indoor–outdoor concentration relationships and diurnal and nocturnal variations in $P M_{2.5}$
The ratio of indoor-to-outdoor concentrations is commonly used to describe the $1/0$ relationships of air pollutants (Wang et al., 2003, 2006; Duan et al., 2007). In general, if I/O is greater than 1.00, the source is mainly from the indoors; when $1/0$ is less than or equal to 1.00, pollutants are mainly from the ambient air. The correlation coefficient $(\mathbb{R}^{2})$ between the indoor and outdoor concentrations was used to indicate whether the PM species measured indoors originated from the outdoors.
In this study, the I/O ratios in Table 2 (0.44 in summer and 0.62 in autumn were all smaller than 1.00) and $\mathrm{R}^{2}$ values in Fig. 2 (0.21 in summer and 0.65 in autumn, suggesting that indoor and outdoor $\mathsf{P M}_{2.5}$ concentrations have good correlation) indicated that the indoor concentrations of these respirable pollutants were dominated by outdoor sources. The lower I/O ratio and $\mathbb{R}^{2}$ value in the summer can be attributed to the extensive use of air conditioning at the office. It has been reported that the use of air conditioning in a room can reduce particle concentrations and the infiltration rates of outdoor particles (Vette et al., 2001; Guo et al., 2008; Lv and Zhu, 2013). However, in autumn, the office relied on natural ventilation, and doors and windows were left open.
As shown in Fig. 4, during daytime and nighttime, the I/O ratios for $\mathsf{P M}_{2.5}$ in summer were 0.51 and 0.43, respectively, while in autumn, these I/O ratios were 0.72 and 0.58, respectively. The daytime ratios were higher than those at nighttime, which could be due to the operation of electronic equipment during daytime.
3.2. Inorganic ions
3.2.1. Inorganic ion concentrations
The mean concentrations of the sum of the 8 measured inorganic ions $\mathrm{[Cl^{-}}$ , $\mathrm{NO}_{3}^{-}$ , $S0_{4}^{2-}$ , $\mathsf{N a}^{+}$ , $\mathrm{NH_{4}^{+}}$ , $\mathsf{K}^{+}$ , ${\mathrm{Mg}}^{2+}$ and $(\mathbf{a}^{2+})$ in the indoor and outdoor air were $40.01\,\pm\,9.10$ and $105.24\pm19.74\ensuremath{\,\upmu\mathrm{g\m}^{-3}}$ in summer, respectively, which were higher than those in autumn $38.13\,\pm\,5.52$ and $71.33\,\pm$ $10.48\,\upmu\mathrm{g}\,\mathrm{m}^{-3}.$ ). In addition, in summer, the total concentrations of the 8 inorganic ions accounted for $77.53\%$ and $90.29\%$ of the indoor and outdoor $\mathsf{P M}_{2.5}$ mass concentrations, respectively, which were also higher than those in autumn $.40.84\%$ and $47.67\%,$ respectively). $S0_{4}^{2-}$ , $\mathtt{N O}_{3}^{-}$ and $\mathrm{NH}_{4}^{+}$ were the three main inorganic ions, accounting for approximately $95.88\%$ and $97.86\%$ of the indoor and outdoor total ion concentrations in summer, respectively, and $80.76\%$ and $89.58\%$ of the indoor and outdoor total ion concentrations in autumn. Compared with other cities, the concentrations of $S0_{4}^{2-}$ , $\mathrm{NH}_{4}^{+}$ and $\mathrm{NO}_{3}^{-}$ were 3.36–10.66 times higher than those of foreign cities, such as Thessaloniki, Greece, in 1999 (Mouratidou and Samara, 2004) and Munich, Germany, in 2005 (Fromme et al., 2008), and 1.46–1.78 times higher than those of Chinese locations, such as Guizhou in 2008 (Wang et al., 2010).
Among the inorganic ions, $S0_{4}^{2-}$ was the largest component, and the $S0_{4}^{2-}$ concentrations exhibited a strong seasonal variation, with a greater contribution observed during the summer. $S0_{4}^{2-}$ is formed by the oxidation of $S0_{2}$ , and in summer, the increase in photochemical reactions promotes the oxidation of the gaseous precursor $S0_{2}$ by hydroxyl radicals (Mazzera et al., 2001; Seguin et al., 2011). Significant reductions in the contributions of nitrate and ammonium were observed during summer when compared with autumn. The concentration of $\mathrm{NO}_{3}^{-}$ is mainly affected by the chemical equilibrium, which depends on temperature and relative humidity (Lin et al., 2010). Compared with summer, the low temperature in autumn is beneficial to the formation of nitrate. The contribution of the $\mathrm{Cl}^{-}$ ion to $\mathsf{P M}_{2.5}$ was higher in autumn, which may be related to a greater amount of coal combustion (Sun et al., 2006). Biomass burning around Shandong (https://firms. modaps.eosdis.nasa.gov/firemap/) was the main factor contributing to the higher contribution of $\mathrm{K}^{+}$ in autumn relative to that in summer.
3.2.2. Indoor-outdoor concentration relationships and diurnal and nocturnal ion variations
As shown in Table 2, the average $1/0$ ratios of $\Nu0_{3}^{-}$ , $S0_{4}^{2-}$ , $\mathsf{N a}^{+}$ , $\mathsf{N H}_{4}^{+}$ and $\mathsf{K}^{+}$ ions were less than 1.00 in summer and autumn, indicating that these ions were mainly from outdoor sources, whereas the $1/0$ ratios of $\mathrm{Cl}^{-}$ , ${\mathrm{Mg}}^{2+}$ and ${\mathsf{C}}{\mathsf{a}}^{2+}$ were greater than 1.00, indicating the presence of a significant indoor source. In the indoor environment, a $\mathrm{NH}_{4}\mathrm{NO}_{3}$ aerosol is rapidly volatilized and adheres to room surfaces, which may be a reason for the lower indoor $\Nu0_{3}^{-}$ and $\mathrm{NH_{4}^{+}}$ concentrations (Fromme et al., 2008; Saliba et al., 2009). The large amount of chlorinated water used for cleaning and the associated evaporation of chlorine into the air during the cleaning process contributed to the indoor $\mathsf{C l}^{-}$ concentrations. Several studies have found that cleaning products that contain bleach will release significant amounts of $\mathrm{Cl}^{-}$ (Loupa et al., 2007). The resuspension of particles in association with the movement of people across the office contributed to the indoor ${\mathsf{C}}{\mathsf{a}}^{2+}$ and ${\mathrm{Mg}}^{2+}$ concentrations.
The correlation coefficients $(\mathbb{R}^{2})$ between indoor and outdoor inorganic ion concentrations were evaluated to determine which ion species measured indoors are influenced by contributions from outdoors. The $\mathrm{NO}_{3}^{-}$ $_3^{-},50_{4}^{2-},\mathrm{Na}^{+},\mathrm{NH}_{4}^{+}$ and $\mathsf{K}^{+}$ ions had strong correlations ( $\mathrm{~R}^{2}$ values ranging from 0.52 to 0.87), which indicate that these ions predominantly originated from the transport of outdoor air. The indoor and outdoor ${\mathrm{Mg}}^{2+}$ and $\mathsf{C}a^{2\bar{+}}$ concentrations were moderately correlated $^{\prime}\mathrm{R}^{2}=0.28\substack{-0.32}^{}$ , whereas the $\mathrm{Cl^{-}}$ correlation was very weak $(\mathsf{R}^{2}=0.07–0.08)$ , indicating the significant contribution of $C^{1^{-}}$ from indoor sources (cleaning activities).
As shown in Fig. 4, in summer, the $1/0$ ratios for $\mathtt{N O}_{3}^{-}$ and $\mathsf{C l}^{-}$ were higher during daytime than at nighttime, which may be related to the use of air conditioner systems in the office during daytime. Additionally, the indoor temperature was lower than the ambient temperature, which would reduce the reactions of $\mathrm{NH}_{4}\mathrm{NO}_{3}$ and $\mathsf{N H}_{4}\mathsf{C l}$ that resulted to their decomposition into gaseous $\mathrm{NH}_{3}$ , $\mathrm{HNO}_{3}$ and HCl. In autumn, the $1/0$ ratios of $\mathrm{NO}_{3}^{-}$ and $\mathrm{Cl}^{-}$ measured during daytime exceeded the ratio observed during nighttime, which may be associated with the large temperature difference between day and night in autumn. Thus, when the outdoor temperature decreased, the decomposition of particulates $\mathrm{NH}_{4}\mathrm{NO}_{3}$ and $\mathsf{N H}_{4}\mathbf{Cl}$ was inhibited. Notably, during daytime, cleaning products were heavily used and there was a great deal of human movement, which resulted in higher $1/0$ ratios of $\mathrm{Cl}^{-}$ and ${\mathsf{C}}{\mathsf{a}}^{2+}$ and ${\mathrm{Mg}}^{2+}$ , compared to nighttime. The $1/0$ ratio of $\mathsf{K}^{+}$ in autumn was higher in daytime and may have been affected by biomass burning.
3.3. PAHs
3.3.1. PAH concentrations
The mean concentrations of individual PAHs and the $\sum$ PAHs in the indoor and outdoor environments are given in Table 2. The $\sum$ PAH concentrations in the indoor and outdoor samples were $111.07\,\pm\,5.95$ and $140.34\,\pm\,7.59\,\mathrm{~ng~}\,\mathrm{m}^{-3}$ respectively, in autumn, which were 3.83 and 4.56 times those in summer. Moreover, in autumn, the indoor PAH concentrations were more than 2.12 times higher than those at an urban sampling site with heating sources in China (Li et al., 2006) and Portuguesa, Venezuela (Castro et al., 2011). In addition, the outdoor PAH concentrations in autumn were much greater than those measured in residential areas located in North America, such as Houston (Fraser et al., 2002), Atlanta and Gulfport (Zheng et al., 2002), and more than 2.41 times greater than those at urban sites in Birmingham, Britain (Zheng et al., 2002) and Guangzhou, China (Li et al., 2006).
The higher $\sum$ PAH concentrations in autumn may be associated with crop straw burning and the lower temperature during this sampling period. It was found that crop straw burning is common and represents a considerable source of PAHs (Zhang et al., 2007; Shen et al., 2013). Low temperatures can increase the transition of PAHs from the vapour phase to the particulate phase and decrease the photochemical decomposition of PAHs (Duan et al., 2007).
Fig. 3 shows the relative contributions of individual PAHs to the total PAH concentrations. In summer and autumn, the contribution profiles of all of the individual PAHs were very similar for both the indoor and outdoor samples. The 3–4-ring PAH compounds provided higher contributions in the indoor air, while the 5–6-ring PAHs were the predominant contributors in the outdoor air. Notably, a relatively low proportion of 5–6-ring PAHs was observed in the indoor air in summer. These lower concentrations may have been related to the use of air conditioner systems in the office. According to Lv and Zhu (2013), the concentrations and percentages of the indoor 5–6- ring PAHs decline when air conditioner systems are operated continuously compared with the use of natural ventilation.
3.3.2. Indoor–outdoor concentration relationships and diurnal and nocturnal variations in PAHs
As shown in Table 2, in summer, the indoor concentrations of the 3–4-ring PAHs were considerably higher than those of the outdoors, resulting in I/O values greater than 1.00; however, for the less volatile 5–6-ring PAHs, the situation was reversed, and the mean I/O ratios were less than 1.00. In autumn, the I/O ratios of the 3-ring PAH concentrations were greater than 1.00, while the ratios for the 4–6-ring PAHs were less than 1.00. Moreover, in summer, the coefficients of determination, $\mathbf{R}^{2}$ , for the 3–4-ring PAHs were lower than 0.50 but were greater than 0.50 for the 5–6-ring PAHs. In autumn, the $\mathbb{R}^{2}$ values for the 3-ring PAHs were less than 0.50 but greater than 0.50 for the 4–6-ring PAHs. These results indicate that the low-ring PAHs mainly came from indoor sources, whereas the high-ring PAHs predominantly originated from the transport of outdoor air.
The diurnal and nocturnal I/O ratios for particulate PAHs are shown in Fig. 4. During the daytime in both summer and autumn, however, the I/O ratios of the 15 monitored PAHs were all greater than 1.00; thus, we believe that the PAHs were mainly from indoor sources. However, during nighttime, the $1/0$ ratios for the 3–4-ring PAHs were greater than 1.00 in both summer and autumn, while those for the 5–6-ring PAHs were less than 1.00. These results indicate that the emission sources of 3–4-ring PAHs mainly derive from indoor air, while the 5–6- ring PAHs were predominantly from outdoor sources.
Some studies implied emission source as a predominant factor controlling the changes of PAHs with the evidence of positive correlation between PAH concentration and $\mathsf{P M}_{2.5}$ contents (Chen et al., 2009; Hu et al., 2012). Interestingly, in this study, the $\mathsf{P M}_{2.5}$ concentrations had no significant correlations with particulate PAH concentrations ( $\mathrm{~\,~r=~0.12~}$ ), which implied that not only emission sources but also gas-particle partitioning processes control the variations of the particulate PAH concentrations (Sitaras et al., 2004; Vasilakos et al., 2007). The results of diurnal and nocturnal I/O ratios for particulate PAHs were strong evidence for the contribution of gas-particle partitioning.
These results could be explained by temperature fluctuations due to the use of air conditioning in summer and the temperature difference between day and night in autumn in Jinan. In summer, air conditioning reduced the indoor temperature in the office, which decreased the transition of PAHs from the particulate phase to the vapour phase. Thus, during daytime, the concentration of particulate PAHs was high and the I/O ratios of the 15 monitored PAHs were all greater than 1.00. In autumn, the higher concentrations of 5–6-ring PAHs in the outdoor air at nighttime may have been associated with the temperature difference between day and night (Liu et al., 2013). Fig. 5 shows that the percentage contribution of 5–6-ring PAHs to the total PAH concentration was higher at nighttime than that during daytime. During nighttime, the low temperatures caused a decrease in the transition of 5–6-ring PAHs from the particulate phase to the vapour phase, which resulted in an increase in the 5–6-ring PAH concentrations. Moreover, significant correlations were observed between the percentage contribution of 5–6-ring PAHs and temperature in summer $(\mathrm{r}=-0.56)$ and autumn $(\mathrm{r}=-0.49)$ . A number of studies have also reported that atmospheric temperature has an important influence on the variations in particulate PAH concentrations (Li et al., 2006; Tan et al., 2006; Vasilakos et al., 2007).
3.3.3. PAH source apportionment by diagnostic ratios
Considering the fact that outdoor particle sources have a strong impact on indoor environments, we used diagnostic ratios to analyse the sources of the outdoor particulate PAHs. Diagnostic ratios are useful indicators of PAH sources because isomer pairs are diluted to a similar extent when mixed with natural particulate matter. They are also distributed similarly among phases because they have comparable thermodynamic partitioning and mass transfer coefficients (Dickhut et al., 2000). Certain PAH isomer-pair ratios, including IcdP/(IcdP $^+$ BghiP), BaA $/(\mathrm{BaA}\,+\,\mathrm{Chr})$ and BaP/BghiP, have been widely used to distinguish among possible sources of PAHs due to their relatively stable features (Yunker et al., 2002). The results of these diagnostic ratios are listed in Table 3.
The ratio of IcdP/(IcdP $^+$ BghiP) was 0.46 in summer and 0.68 in autumn. The result indicated that PAHs mainly originated from fuel combustion outdoors in the summer, and in the autumn, outdoor PAHs primarily originated from coal/biomass combustion sources (Mandalakis et al., 2002; Hu et al., 2012). The value of the BaA/(BaA + Chr) ratio in summer was 0.41, indicating pyrolytic sources, whereas in autumn, the ratio was between 0.20 and 0.35, indicating a strong contribution from combustion (Soclo et al., 2000). The BaP/BghiP ratio in the outdoor air in summer was between 0.30 and 0.40, indicating a strong contribution from traffic (Park et al., 2002). In contrast, the ratio in autumn was greater than 0.41 but lower than 0.90, implying a mixture of contributions from coal/biomass combustion and traffic sources (Liu et al., 2007).
3.3.4. Risk assessment of PAH pollution in the indoor air of the office
The carcinogenic potency of each PAH was assessed in terms of its benzo[a]pyrene-equivalent concentration (BaPeq). The calculation of the BaPeq concentration for each individual PAH species requires the use of its toxic equivalent factor (TEF), which represents the relative cancer potency of a given species (Nisbet and LaGoy, 1992). The concentrations of carcinogenic PAHs were calculated as:
$$
\mathrm{BaPeq}=\mathrm{PAH_{i}*T E F_{i;\,a n d\,T E Q}=\sum(P A H_{i}*T E F_{i})}
$$
where $\mathrm{PAH_{i}}$ is the concentration of the PAH congener i; $\mathrm{TEF_{i}}$ is the toxic equivalent factor for the PAH congener i; and TEQ is the toxic equivalent of the reference compound.
Table 4 shows that the TEQ values in summer were 1.3528 and $3.0471~\mathrm{~ng~}\mathrm{~m}^{-3}$ in the indoor and outdoor samples, respectively, which were lower than the national standard of $10\;\;\mathrm{ng}\;\;\mathrm{m}^{-3}$ but were significantly higher than the WHO standard $1\ \mathrm{ng}\ \mathrm{m}^{-3}$ , WHO, 1987). In autumn, the indoor and outdoor TEQ values were 18.4011 and $29.4012~\mathrm{~ng~}\mathrm{~m}^{-3}$ , respectively, which were much greater than the national standard and also approximately tenfold greater than those in the summer. Notably, although the indoor TEQ values were lower than the outdoor values in both summer and autumn, the contributions of carcinogenic PAHs, such as BaA, BkF, IcdP and DBA, to the inhalation exposure of TEQ in the indoor air were $36.97\%$ in summer and $54.70\%$ in autumn, suggesting a relatively higher human health risk in the indoor air in the office in Jinan compared to the values for the outdoor air in summer $(43.99\%)$ and in autumn $(51.04\%)$ .
In addition, we also used the values of the $\sum$ TEF-PAHs to assess the lifetime lung cancer risk attributed to the indoor and outdoor samples. The WHO proposed a unit of risk of $8.70~\times$ $10^{-5}$ $(\mathrm{ng~m}^{\bar{-}3})^{-1}$ for a lifetime (70 years) of PAH exposure (WHO, 2000). The values of the lifetime lung cancer risks from the exposure to indoor particulate PAHs in the office in Jinan were $1.18\,\times\,10^{-4}$ in summer and $1.60\,\times\,10^{-3}$ in autumn, which exceed the health guideline $(10^{-5})$ (Boström et al., 2002). Therefore, there was high health risk to individuals from exposure to PAHs in the indoor air in Jinan. Moreover, the values of the lifetime lung cancer risks from particulate PAHs in the outdoor are were $2.65\times10^{-4}$ and $2.56\times10^{-3}$ in summer and autumn, respectively, which were 2.57 and 1.60-fold greater than those of the indoor air, suggesting a relatively high human health risk in Jinan.
During daytime, the indoor TEQ value was $0.4980\mathrm{\,ng\,m^{-3}}$ in summer and $15.8476~\mathrm{ng~m}^{-3}$ in autumn, which was 1.99 and 6.20-fold greater than those at nighttime, respectively. Additionally, the values of the lifetime lung cancer risks during the daytime $(4.33\times10^{-5}$ in summer and $1.38\times10^{-3}$ in autumn) were also greater than those for the nighttime $(2.17\times10^{-5}$ in summer and $2.22\times10^{-4}$ in autumn). These results suggest that indoor PAHs have more carcinogenic and mutagenic effects during daytime.
4. Summary
The mean indoor and outdoor $\mathsf{P M}_{2.5}$ mass concentrations were $51.52\pm28.22$ and $115.92\pm44.84\;\upmu\mathrm{g}\;\mathrm{m}^{-3}$ in summer and $93.52\pm41.07$ and $150.88\pm73.07\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ in autumn. The average $1/0$ ratios of the $\mathsf{P M}_{2.5}$ indicated that the indoor $\mathsf{P M}_{2.5}$ was mainly transported from outdoor sources. The $S0_{4}^{2-}$ , $\mathrm{NO}_{3}^{-}$ and $\mathsf{N H}_{4}^{+}$ ions were the three main inorganic ions in the indoor and outdoor air. In summer and autumn, $\mathrm{NO}_{3}^{-}$ , $S0_{4}^{2-}$ , $\mathsf{N a}^{+}$ , $\mathsf{N H}_{4}^{+}$ and $\mathsf{K}^{+}$ ions mainly originated from outdoor sources, whereas $\mathrm{Cl}^{-}$ , ${\mathrm{Mg}}^{2+}$ and ${\mathsf{C}}{\mathsf{a}}^{2+}$ had significant indoor sources. During daytime, the $1/0$ ratio for $\mathrm{NO}_{3}^{-}$ was higher than that at nighttime, which was related to the temperature difference between the indoor and outdoor air. The 3–4-ring PAH compounds provided a higher contribution to PAHs in the indoor air, while the 5–6-ring PAHs were the predominant contributor in the outdoor air. During daytime, the 15 monitored PAHs were all mainly from indoor sources. In contrast, during nighttime, the 3–4-ring PAHs were mainly generated indoors, but the 5–6-ring PAHs were predominantly transported indoors from the outdoor air. According to the correlations of $\mathsf{P M}_{2.5}$ and particulate PAH concentrations, we can conclude that not only emission sources but also gasparticle partitioning processes control the variations in particulate PAH concentrations. Diagnostic ratios indicated that biomass burning had an important role in outdoor PAH concentrations in autumn. The results of the risk assessment of PAH pollution indicated that indoor PAHs have more carcinogenic and mutagenic effects during daytime.
Acknowledgements
This work was supported by the National Basic Research Program (973 Program) of China (2005CB422203), Key Project of Shandong Provincial Environmental Agency (2006045), Promotive Research Fund for Young and Middle-aged Scientists of Shandong Province (BS2010HZ010), Independent Innovation Foundation of Shandong University (2009TS024) and Special Research for Public-Beneficial Environment Protection (201009001-1).
References
Baird, W.M., Hooven, L.A., Mahadevan, B., 2005. Carcinogenic polycyclic aromatic hydrocarbon-DNA adducts and mechanism of action. Environ. Mol. Mutagen. 45, 106–114.
Balasubramanian, R., Lee, S.S., 2007. Characteristics of indoor aerosols in residential homes in urban locations: a case study in Singapore. J. Air Waste Manage. Assoc. 57, 981–990.
Baldasano, J., Valera, E., Jiménez, P., 2003. Air quality data from large cities. Sci. Total Environ. 307, 141–165.
Barraza, F., Jorquera, H., Valdivia, G., Montoya, L.D., 2014. Indoor PM2.5 in Santiago, Chile, spring 2012: source apportionment and outdoor contributions. Atmos. Environ. 94, 692–700.
Bernstein, J.A., Alexis, N., Bacchus, H., Bernstein, I.L., Fritz, P., Horner, E., Oullette, J., 2008. The health effects of nonindustrial indoor air pollution. J. Allergy Clin. Immunol. 121, 585–591.
Blondeau, P., Iordache, V., Poupard, O., Genin, D., Allard, F., 2005. Relationship between outdoor and indoor air quality in eight French schools. Indoor Air 15, 2–12.
Boström, C.E., Gerde, P., Hanberg, A., Jernström, B., Johansson, C., Kyrklund, T., Rannug, A., Törnqvist, M., Victorin, K., Westerholm, R., 2002. Cancer risk assessment, indicators, and guidelines for polycyclic aromatic hydrocarbons in the ambient air. Environ. Health Perspect. 110, 451–488.
Castro, D., Slezakova, K., Delerue-Matos, C., Alvim-Ferraz, M.D.C., Morais, S., Pereira, M.D.C., 2011. Polycyclic aromatic hydrocarbons in gas and particulate phases of indoor environments influenced by tobacco smoke: levels, phase distributions, and health risks. Atmos. Environ. 45, 1799–1808.
Chen, K., Li, H., Wang, H., Wang, W., Lai, C., 2009. Measurement and receptor modeling of atmospheric polycyclic aromatic hydrocarbons in urban Kaohsiung, Taiwan. J. Hazard. Mater. 166, 873–879.
Chithra, V.S., Shiva Nagendra, S.M., 2013. Chemical and morphological characteristics of indoor and outdoor particulate matter in an urban environment. Atmos. Environ. 77, 579–587.
Destaillats, H., Singer, B.C., Lee, S.K., Gundel, L.A., 2006. Effect of ozone on nicotine desorption from model surfaces: evidence for heterogeneous chemistry. Environ. Sci. Technol. 40, 1799–1805.
Dickhut, R.M., Canuel, E.A., Gustafson, K.E., Liu, K., Arzayus, K.M., Walker, S.E., Edgecombe, G., Gaylor, M.O., MacDonald, E.H., 2000. Automotive sources of carcinogenic polycyclic aromatic hydrocarbons associated with particulate matter in the Chesapeake Bay region. Environ. Sci. Technol. 34, 4635–4640.
Duan, J.C., Bi, X.H., Tan, J.H., Sheng, G.Y., Fu, J.M., 2007. Seasonal variation on size distribution and concentration of PAHs in Guangzhou City, China. Chemosphere 67, 614–622.
Fraser, M., Yue, Z., Tropp, R., Kohl, S., Chow, J., 2002. Molecular composition of organic fine particulate matter in Houston, TX. Atmos. Environ. 36, 5751–5758.
Fromme, H., Diemer, J., Dietrich, S., Cyrys, J., Heinrich, J., Lang, W., Twardella, D., 2008. Chemical and morphological properties of particulate matter (PM10, PM2.5) in school classrooms and outdoor air. Atmos. Environ. 42, 6597–6605.
Gao, X.M., Yang, L.X., Cheng, S.H., Gao, R., Zhou, Y., Xue, L.K., Nie, W., 2011. Semicontinuous measurement of water-soluble ions in PM2.5 in Jinan, China: temporal variations and source apportionments. Atmos. Environ. 45, 6048–6056.
Guo, H., Morawska, L., He, C., Gilbert, D., 2008. Impact of ventilation scenario on air exchange rates and on indoor particle number concentrations in an airconditioned classroom. Atmos. Environ. 42, 757–768.
Gupta, P., Harger, W.P., Arey, J., 1996. The contribution of nitro-and methylnitronaphthalenes to the vapor-phase mutagenicity of ambient air samples. Atmos. Environ. 30, 3157–3166.
Hasheminassab, S., Daher, N., Shafer, M.M., Schauer, J.J., Delfino, R.J., Sioutas, C., 2014. Chemical characterization and source apportionment of indoor and outdoor fine particulate matter (PM2.5) in retirement communities of the Los Angeles Basin. Sci. Total Environ. 490, 528–537.
Hassanvand, M.S., Naddafi, K., Faridi, S., Arhami, M., Nabizadeh, R., Sowlat, M.H., Kashani, H., 2014. Indoor/outdoor relationships of PM10, PM2.5, and PM1 mass concentrations and their water-soluble ions in a retirement home and a school dormitory. Atmos. Environ. 82, 375–382.
Hayakawa, K., Kizu, R., Ando, K., 1999. Study on atmospheric behavior and toxicity of carcinogenic nitroarenes by high-performance liquid chromatography using chemiluminescence detection. Chromatogr. J. Sep. Detect. Sci. 20, 37–43.
Hu, J., Liu, C., Zhang, G., Zhang, Y., 2012. Seasonal variation and source apportionment of PAHs in TSP in the atmosphere of Guiyang, Southwest China. Atmos. Res. 118, 271–279.
Jie, Y., Ismail, N.H., Isa, Z.M., 2011. Do indoor environments influence asthma and asthma-related symptoms among adults in homes? A review of the literature. J. Formos. Med. Assoc. 110, 555–563.
Jo, W.K., Seo, Y.J., 2005. Indoor and outdoor bioaerosol levels at recreation facilities, elementary schools, and homes. Chemosphere 61, 1570–1579.
Jovanović, M., Vučićević, B., Turanjanin, V., Živković, M., Spasojević, V., 2014. Investigation of indoor and outdoor air quality of the classrooms at a school in Serbia. Energy 1–7.
Kamal, A., Malik, R.N., Martellini, T., Cincinelli, A., 2014. Cancer risk evaluation of brick kiln workers exposed to dust bound PAHs in Punjab Province (Pakistan). Sci. Total Environ. 493, 562–570.
Kotzias, D., Geiss, O., Tirendi, S., Barrero-Moreno, J., Reina, V., Gotti, A., Sarigiannis, D., 2009. Exposure to multiple air contaminants in public buildings, schools and kindergartens—the european indoor air monitoring and exposure assessment (AIRMEX) study. Fresenius Environ. Bull. 18, 670–681.
Krugly, E., Martuzevicius, D., Sidaraviciute, R., Ciuzas, D., Prasauskas, T., Kauneliene, V., Kliucininkas, L., 2014. Characterization of particulate and vapor phase polycyclic aromatic hydrocarbons in indoor and outdoor air of primary schools. Atmos. Environ. 82, 298–306.
Lee, S., Chang, M., 2000. Indoor and outdoor air quality investigation at schools in Hong Kong. Chemosphere 41, 109–113.
Li, J., Zhang, G., Li, X., Qi, S., Liu, G., Peng, X., 2006. Source seasonality of polycyclic aromatic hydrocarbons (PAHs) in a subtropical city, Guangzhou, South China. Sci. Total Environ. 355, 145–155.
Li, G., Lang, Y., Yang, W., Peng, P., Wang, X., 2014. Source contributions of PAHs and toxicity in reed wetland soils of Liaohe estuary using a CMB–TEQ method. Sci. Total Environ. 490, 199–204.
Lin, Y.C., Cheng, M.T., Lin, W.H., Lan, Y.Y., Tsuang, B.J., 2010. Causes of the elevated nitrate aerosol levels during episodic days in Taichung urban area, Taiwan. Atmos. Environ. 44, 1632–1640.
Liu, M., Cheng, S.B., Ou, D.N., Hou, L.J., Gao, L., Wang, L.L., Xie, Y.S., Yang, Y., Xu, S.Y., 2007. Characterization, identification of road dust PAHs in central Shanghai areas, China. Atmos. Environ. 41, 8785–8795.
Liu, J., Li, J., Lin, T., Liu, D., Xu, Y., Chaemfa, C., Zhang, G., 2013. Diurnal and nocturnal variations of PAHs in the Lhasa atmosphere, Tibetan Plateau: implication for local sources and the impact of atmospheric degradation processing. Atmos. Res. 124, 34–43.
Loupa, G., Kioutsioukis, I., Rapsomanikis, S., 2007. Indoor–outdoor atmospheric particulate matter relationships in naturally ventilated offices. Indoor Built Environ. 16, 63–69.
Lv, J., Zhu, L., 2013. Effect of central ventilation and air conditioner system on the concentration and health risk from airborne polycyclic aromatic hydrocarbons. J. Environ. Sci. 25, 531–536.
Mandalakis, M., Tsapakis, M., Tsoga, A., Stephanou, E.G., 2002. Gas–particle concentrations and distribution of aliphatic hydrocarbons, PAHs, PCBs and PCDD/Fs in the atmosphere of Athens (Greece). Atmos. Environ. 36, 4023–4035.
Massey, D., Kulshrestha, A., Masih, J., Taneja, A., 2012. Seasonal trends of PM10, PM 5.0, PM 2.5, PM1.0 in indoor and outdoor environments of residential homes located in North-Central India. Build. Environ. 47, 223–231.
Mazzera, D.M., Lowenthal, D.H., Chow, J.C., Watson, J.G., 2001. Sources of PM10 and sulfate aerosol at McMurdo station, Antarctica. Chemosphere 45, 347–356.
Mouratidou, T., Samara, C., 2004. PM2.5 and associated ionic component concentrations inside the archaeological museum of Thessaloniki, N. Greece. Atmos. Environ. 38, 4593–4598.
National Bureau of Statistics of China, 2009. China Statistical Yearbook. China Statistics Press, Beijing (in Chinese).
Nisbet, I.C., LaGoy, P.K., 1992. Toxic equivalency factors (TEFs) for polycyclic aromatic hydrocarbons (PAHs). Regul. Toxicol. Pharmacol. 16, 290–300.
Novotna, B., Topinka, J., Solansky, I., Chvatalova, I., Lnenickova, Z., Sram, R.J., 2007. Impact of air pollution and genotype variability on DNA damage in Prague policemen. Toxicol. Lett. 172, 37–47.
Park, S.S., Kim, Y.J., Kang, C.H., 2002. Atmospheric polycyclic aromatic hydrocarbons in Seoul, Korea. Atmos. Environ. 36, 2917–2924.
Pegas, P., Alves, C., Evtyugina, M., Nunes, T., Cerqueira, M., Franchi, M., Freitas, M., 2011. Seasonal evaluation of outdoor/indoor air quality in primary schools in Lisbon. J. Environ. Monit. 13, 657–667.
Polidori, A., Cheung, K.L., Arhami, M., Delfino, R.J., Schauer, J.J., Sioutas, C., 2009. Relationships between size-fractionated indoor and outdoor trace elements at four retirement communities in southern California. Atmos. Chem. Phys. 9, 4521–4536.
Pongpiachan, S., Tipmanee, D., Deelaman, W., Muprasit, J., Feldens, P., Schwarzer, K., 2013. Risk assessment of the presence of polycyclic aromatic hydrocarbons (PAHs) in coastal areas of Thailand affected by the 2004 tsunami. Mar. Pollut. Bull. 76, 370–378.
Reisen, F., Arey, J., 2005. Atmospheric reactions influence seasonal PAH and nitro-PAH concentrations in the Los Angeles Basin. Environ. Sci. Technol. 39, 64–73.
Rios, J.L.D.M., Boechat, J.L., Gioda, A., Santos, C.Y.D., Aquino Neto, F.R.D., Lapa Silva, J.R., 2009. Symptoms prevalence among office workers of a sealed versus a non-sealed building: associations to indoor air quality. Environ. Int. 35, 1136–1141.
Rivas, I., Viana, M., Moreno, T., Pandolfi, M., Amato, F., Reche, C., Sunyer, J., 2014. Child exposure to indoor and outdoor air pollutants in schools in Barcelona, Spain. Environ. Int. 69, 200–212.
Saliba, N., Atallah, M., Al-Kadamany, G., 2009. Levels and indoor–outdoor relationships of PM10 and soluble inorganic ions in Beirut, Lebanon. Atmos. Res. 92, 131–137.
Samet, J.M., Spengler, J.D., 2003. Indoor environments and health: moving into the 21st century. Am. J. Public Health 93, 1489–1493.
Sangiorgi, G., Ferrero, L., Perrone, M., Bolzacchini, E., Duane, M., Larsen, B., 2011. Vertical distribution of hydrocarbons in the low troposphere below and above the mixing height: tethered balloon measurements in Milan, Italy. Environ. Pollut. 159, 3545–3552.
Sangiorgi, G., Ferrero, L., Ferrini, B., Lo Porto, C., Perrone, M., Zangrando, R., Bolzacchini, E., 2013. Indoor airborne particle sources and semi-volatile partitioning effect of outdoor fine PM in offices. Atmos. Environ. 65, 205–214.
Saraga, D.E., Maggos, T.E., Sfetsos, A., Tolis, E.I., Andronopoulos, S., Bartzis, J.G., Vasilakos, C., 2010. PAHs sources contribution to the air quality of an office environment: experimental results and receptor model (PMF) application. Air Qual. Atmos. Health 3, 225–234.
Seguin, A.M., Norman, A.L., Eaton, S., Wadleigh, M., 2011. Seasonality in size segregated biogenic, anthropogenic and sea salt sulfate aerosols over the North Atlantic. Atmos. Environ. 45, 6947–6954.
Seinfeld, J.H., Pandis, S.N., 2006. Atmospheric Chemistry and Physics—From Air Pollution to Climate Change, second ed. John Wiley & Sons.
Shen, H., Huang, Y., Wang, R., Zhu, D., Li, W., Shen, G., Lu, Y., 2013. Global atmospheric emissions of polycyclic aromatic hydrocarbons from 1960 to 2008 and future predictions. Environ. Sci. Technol. 47, 6415–6424.
Sitaras, I.E., Bakeas, E.B., Siskos, P.A., 2004. Gas/particle partitioning of seven volatile polycyclic aromatic hydrocarbons in a heavy traffic urban area. Sci. Total Environ. 327, 249–264.
Soclo, H.H., Garrigues, P., Ewald, M., 2000. Origin of polycyclic aromatic hydrocarbons (PAHs) in coastal marine sediments: case studies in Cotonou (Benin) and Aquitaine (France) areas. Mar. Pollut. Bull. 40, 387–396.
Souza, K.F., Carvalho, L.R., Allen, A.G., Cardoso, A.A., 2014. Diurnal and nocturnal measurements of PAH, nitro-PAH, and oxy-PAH compounds in atmospheric particulate matter of a sugar cane burning region. Atmos. Environ. 83, 193–201.
Sun, Y., Zhuang, G., Tang, A., Wang, Y., An, Z., 2006. Chemical characteristics of PM2.5 and PM10 in haze-fog episodes in Beijing. Environ. Sci. Technol. 40, 3148–3155.
Tan, J.H., Bi, X.H., Duan, J.C., Rahn, K.A., Sheng, G.Y., Fu, J.M., 2006. Seasonal variation of particulate polycyclic aromatic hydrocarbons associated with PM10 in Guangzhou, China. Atmos. Res. 80, 250–262.
Vasilakos, C., Levi, N., Maggos, T., Hatzianestis, J., Michopoulos, J., Helmis, C., 2007. Gas–particle concentration and characterization of sources of PAHs in the atmosphere of a suburban area in Athens, Greece. J. Hazard. Mater. 140, 45–51.
Vette, A.F., Rea, A.W., Lawless, P.A., Rodes, C.E., Evans, G., Highsmith, V.R., Sheldon, L., 2001. Characterization of indoor–outdoor aerosol concentration relationships during the Fresno PM exposure studies. Aerosol Sci. Technol. 34, 118–126.
Wang, G., Wang, H., Yu, Y., Gao, S., Feng, J., Gao, S., Wang, L., 2003. Chemical characterization of water-soluble components of PM10 and PM2.5 atmospheric aerosols in five locations of Nanjing, China. Atmos. Environ. 37, 2893–2902.
Wang, Y., Zhuang, G., Zhang, X., Huang, K., Xu, C., Tang, A., An, Z., 2006. The ion chemistry, seasonal cycle, and sources of PM2.5 and TSP aerosol in Shanghai. Atmos. Environ. 40, 2935–2952.
Wang, G., Kawamura, K., Zhao, X., Li, Q., Dai, Z., Niu, H., 2007. Identification, abundance and seasonal variation of anthropogenic organic aerosols from a mega-city in China. Atmos. Environ. 41, 407–416.
Wang, S., Wei, W., Li, D., Aunan, K., Hao, J., 2010. Air pollutants in rural homes in Guizhou, China—concentrations, speciation, and size distribution. Atmos. Environ. 44, 4575–4581.
WHO, 1987. Polynuclear aromatic hydrocarbons (PAH). Air Quality Guidelines for Europe. World Health Organization Regional Office Europe, Copenhagen, pp. 105–117.
WHO, 1998. Environmental Health Criteria 202: Selected Non-Heterocyclic Polycyclic Aromatic Hydrocarbons. International Programme on Chemical Safety. World Health Organisation, Geneva.
WHO, 2000. Air quality guidelines, second ed. WHO Regional Publications, European Series No. 91, pp. 186–194, (Copenhagen).
WHO, 2006. WHO Air Quality Guidelines for Particulate Matter, Ozone, Nitrogen Dioxide and Sulfur Dioxide. World Health Organization, p. 9.
Wolkoff, P., Wilkins, C., Clausen, P., Nielsen, G., 2006. Organic compounds in office environments—sensory irritation, odor, measurements and the role of reactive chemistry. Indoor Air 16, 7–19.
Yang, W., Sohn, J., Kim, J., Son, B., Park, J., 2009. Indoor air quality investigation according to age of the school buildings in Korea. J. Environ. Manage. 90, 348–354.
Yunker, M.B., Macdonald, R.W., Vingarzan, R., Mitchell, R.H., Goyette, D., Sylvestre, S., 2002. PAHs in the Fraser River basin: a critical appraisal of PAH ratios as indicators of PAH source and composition. Org. Geochem. 33, 489–515.
Zhang, K., Wang, Y.S., Wen, T.W., Meslmani, Y., Murray, F., 2007. Properties of nitrate, sulfate and ammonium in typical polluted atmospheric aerosols (PM10) in Beijing. Atmos. Res. 84, 67–77.
Zhang, Q., Streets, D.G., Carmichael, G.R., He, K., Huo, H., Kannari, A., Fu, J., 2009. Asian emissions in 2006 for the NASA INTEX-B mission. Atmos. Chem. Phys. 9, 5131–5153.
Zheng, M., Cass, G.R., Schauer, J.J., Edgerton, E.S., 2002. Source apportionment of PM2. 5 in the southeastern United States using solvent-extractable organic compounds as tracers. Environ. Sci. Technol. 36, 2361–2371.
Zhou, B., Zhao, B., Guo, X., Chen, R., Kan, H., 2013. Investigating the geographical heterogeneity in PM10-mortality associations in the China air pollution and health effects study (CAPES): a potential role of indoor exposure to PM10 of outdoor origin. Atmos. Environ. 75, 217–223.
Zimmermann, K., Atkinson, R., Arey, J., Kojima, Y., Inazu, K., 2012. Isomer distributions of molecular weight 247 and 273 nitro-PAHs in ambient samples, NIST diesel SRM, and from radical-initiated chamber reactions. Atmos. Environ. 55, 431–439.
|
atmosphere
|
0109
|
10.1016/j.atmosres.2017.12.013
|
image
|
Fig. 1. Location of sampling site on BH Island.
|
|
atmosphere
|
0109
|
10.1016/j.atmosres.2017.12.013
|
image
|
Fig. 2. Scatter plot showing the seasonal correlations between OC and EC in $\mathrm{PM}_{2.5}$ .
|
|
atmosphere
|
0109
|
10.1016/j.atmosres.2017.12.013
|
image
|
Fig. 3. Abrupt variation of OC/EC by M-K test on BH. Two pink dotted lines show the range of significant level at $95\%$ .
|
|
atmosphere
|
0109
|
10.1016/j.atmosres.2017.12.013
|
image
|
Fig. 4. EFs of inorganic elements in $\mathrm{PM}_{2.5}$ on BH Island.
|
|
atmosphere
|
0109
|
10.1016/j.atmosres.2017.12.013
|
table
|
Table 1 Seasonal EFs for Co, Mn, Cr, V, Ni, Cu, As, Zn, Pb and Cd. The bold refers to the highest value.
|
|
atmosphere
|
0109
|
10.1016/j.atmosres.2017.12.013
|
image
|
Fig. 5. The correlation between observed $\mathrm{PM}_{2.5}$ and reconstructed $\mathrm{PM}_{2.5}$ with the information of chemical mass closure on BH Island.
|
|
atmosphere
|
0109
|
10.1016/j.atmosres.2017.12.013
|
image
|
Fig. 6. Seven source profiles (bars) and percentage contributions (dots); time series of daily contribution from each identified source (continuous line).
|
|
atmosphere
|
0109
|
10.1016/j.atmosres.2017.12.013
|
image
|
Fig. 7. The respective contribution of the seven factors to PM2.5 on BH Island
|
|
atmosphere
|
0109
|
10.1016/j.atmosres.2017.12.013
|
image
|
Fig. 8. The PSCF maps of coal combustion $^+$ biomass burning, industry source, ship emission, sea salt, mineral dust, refined chrome industry and vehicle emission.
|
|
atmosphere
|
0109
|
10.1016/j.atmosres.2017.12.013
|
text
| Not supported with pagination yet |
Accepted Manuscript
iPMF and PSCF based source apportionment of PM2.5 at a regional background site in North China
Zheng Zong, Xiaoping Wang, Chongguo Tian, Yingjun Chen, Shanfei Fu, Lin Qu, Ling Ji, Jun Li, Gan Zhang
PII: S0169-8095(17)31162-6
DOI: https://doi.org/10.1016/j.atmosres.2017.12.013
Reference: ATMOS 4149
To appear in: Atmospheric Research
Received date: 6 November 2017
Revised date: 19 December 2017
Accepted date: 28 December 2017
Please cite this article as: Zheng Zong, Xiaoping Wang, Chongguo Tian, Yingjun Chen, Shanfei Fu, Lin Qu, Ling Ji, Jun Li, Gan Zhang , iPMF and PSCF based source apportionment of PM2.5 at a regional background site in North China. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Atmos(2017), https://doi.org/10.1016/j.atmosres.2017.12.013
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.
iPMF and PSCF based source apportionment of PM2.5 at a regional background site in North China
Zheng Zong1,2, Xiaoping Wang2, Chongguo Tian1,\*, Yingjun Chen3,\*, Shanfei $\mathrm{Fu}^{4}$ , Lin ${\mathrm{Qu}}^{5}$ , Ling $\mathrm{Ji}^{5}$ , Jun $\mathrm{Li}^{2}$ , Gan Zhang2
1 Key Laboratory of Coastal Environmental Processes and Ecological Remediation, Yantai Institute of Coastal Zone Research, Chinese Academy of Sciences, Yantai, 264003, China 2 State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou, 510640, China
3 Key Laboratory of Cities' Mitigation and Adaptation to Climate Change in Shanghai (CMA), College of Environmental Science and Engineering, Tongji University, Shanghai, 200092, China 4 School of Environmental and Civil Engineering, Jiangnan University, Wuxi, Jiangsu Province, 214122, China
5 Yantai Oceanic Environmental Monitoring Central Station, SOA, Yantai, 264006, China
\* Corresponding author:
Chongguo Tian, Yantai Institute of Coastal Zone Research, CAS. Phone: +86-535-2109-160; Fax: +86-535-2109-000; e-mail: cgtian $@$ yic.ac.cn
Yingjun Chen, College of Environmental Science and Engineering, Tongji University. Phone: $+86.$ - 535-2109-160; Fax: $+86.$ -535-2109-000; e-mail: yjchentj $@$ tongji.edu.cn
Abstract
To apportion regional $\mathrm{PM}_{2.5}$ (atmospheric particles with aerodynamic diameter less than $2.5\ \upmu\mathrm{m}\phantom{.}$ ) source types and their geographic pattern in North China, 120 daily $\mathrm{PM}_{2.5}$ samples on Beihuangcheng Island (BH, a regional background site in North China) were collected from August $20^{\mathrm{th}}$ , 2014 to September $15^{\mathrm{th}}$ , 2015 showing one-year period. After the chemical analyses on carbonaceous species, water-soluble ions and inorganic elements, various approaches, such as Mann-Kendall test, chemical mass closure, ISORROPIA II model, Positive Matrix Factorization (PMF) linked with Potential Source Contribution Function (PSCF), were used to explore the PM2.5 speciation, sources, and source regions. Consequently, distinct seasonal variations of $\mathrm{PM}_{2.5}$ and its main species were found and could be explained by varying emission source characteristics. Based on PMF model, seven source factors for $\mathrm{PM}_{2.5}$ were identified, which were coal combustion $^+$ biomass burning, vehicle emission, mineral dust, ship emission, sea salt, industry source, refined chrome industry with the contribution of $48.21\%$ , $30.33\%$ , $7.24\%$ , $6.63\%$ , $3.51\%$ , $3.2\%$ , and $0.88\%$ , respectively. In addition, PSCF analysis using the daily contribution of each factor from PMF result suggested that Shandong peninsula and Hebei province were identified as the high potential region for coal combustion $^+$ biomass burning; BeijingTianjin-Hebei (BTH) region was the main source region for industry source; Bohai Sea and East China Sea were found to be of high source potential for ship emission; Geographical region located northwest of BH Island was possessed of high probability for sea salt; Mineral dust presumably came from the region of Mongolia; Refined chrome industry mostly came from Liaoning, Jilin province; The vehicle emission was primarily of BTH region origin, centring on metropolises, such as Beijing and Tianjin. These results provided precious implications for PM2.5 control strategies in North China.
Keywords: Source apportionment, $\mathrm{PM}_{2.5}$ , PMF, PSCF, Geographical origin
1 Introduction
Aerosols are airborne solid or liquid particles, which are emitted directly or result from gas-toparticle conversions from multiple natural and anthropogenic sources (Bressi et al., 2014). In recent years, fine particles less than 2.5 micron in aerodynamic diameter $(\mathbf{PM}_{2.5})$ becomes the primary air pollutant in China due to some combined effects, such as rapid industrialization, high population densities, and meteorological factors that lead to weak dilution and dispersion (Niu et al., 2016). Furthermore, it evolves to a top environmental issue because of its strong adverse effects on human health, visibility, weather and climate (Liu et al., 2014), which has triggered both public alarm and official concern. Based on the epidemiology and resulting guidance of international organization, national legislation in many countries has fixed thresholds, policies for controlling PM2.5 mass concentration (Masiol et al., 2017). In China, the government promulgated the First Grade National Ambient Air Quality Standard of China ( $35~\upmu\mathrm{g}\,\textrm{m}^{-3}$ for PM2.5, GB 3095-2012) and introduced the Action Plan for Air Pollution Prevention and Control (2013-17), which aimed at marked improvements in air quality until 2017. Although the political abatement has alleviated the polluted status, air pollution in China is still far from being controlled, especially in the heavily polluted area, such as the North China. In 2016, the proportion of fine days with PM2.5 concentration lower than $35~\upmu\mathrm{g}~\mathrm{m}^{-3}$ in this region was only $56.8\%$ , $22.0\%$ lower than the national average. Also, nine of the ten cities ranking at the bottom of 74 cities in China with the relative worse air quality were located in the area (http://www.zhb.gov.cn/). Therefore, the improvement of air quality in North China is the crux to achieve the national setting goal by 2017.
Source identification and quantification are the key points to reduce the concentration of PM2.5. For this, various studies have been conducted targeting big cities in this region for establishing an efficient control strategy to improve air quality (Li et al., 2016; Wu et al., 2015). However, such research results seem to be rather insufficient for pollution control, which may be ascribed to the lack of the regional source information of $\mathrm{PM}_{2.5}$ (Kotchenruther, 2016). It is because the diffusion and transport of $\mathrm{PM}_{2.5}$ could pose a huge challenge to the pollution control (Wang et al., 2016b). Thus, regional scale efforts to assess source apportionment of ambient PM2.5 should be addressed. Monitoring data from background area could be used to assess the sources, as they are beyond the influence of pointing source, and therefore, better suited for investigating the regional pollution. As an illustration, biomass burning and biogenic emission were found to be the predominant sources for aerosols in East China in summer and autumn based on an observation at a regional background site in Ningbo (Liu et al., 2013). In addition, the geographical origin of $\mathrm{PM}_{2.5}$ in North China is also important, but poorly documented. To the best of our knowledge, few studies have addressed this issue in this region (Zhang et al., 2013). Thus, regional source apportionment of $\mathrm{PM}_{2.5}$ including both source and geographical origin information is urgently needed. In the present study, year-round $\mathrm{PM}_{2.5}$ samples were collected on Beihuangcheng (BH) Island, a regional background site in North China, for a better understanding of the source and geographical origin of PM2.5. The objects of this study are (1) to qualify the regional scale concentration burden and characteristics of $\mathrm{PM}_{2.5}$ and its species; (2) to assess the sources of $\mathrm{PM}_{2.5}$ based on chemical mass closure method and Positive Matrix Factorization (PMF) model; (3) to distinguish the source region of $\mathrm{PM}_{2.5}$ using Potential Source Contribution Function (PSCF) model linked with PMF model. This effort is vital, as it would help to facilitate effective control strategies to mitigate air pollution in North China.
2 Materials and methods
2.1 Sampling site and sample collection
The sampling campaign was conducted from August $20^{\mathrm{th}}$ , 2014 to September $15^{\mathrm{th}}$ , 2015 at the Environmental Monitoring Station of State Ocean Administration of China on BH Island ( $38^{\circ}24^{\prime}\mathrm{N}$ $120^{\circ}55^{\prime}\mathrm{E})$ ). The island lies at the demarcation line between Bohai Sea and Yellow Sea (Fig. 1), which is about $65\,\mathrm{km}$ north of Shandong peninsula, $185\,\mathrm{km}$ east of Beijing-Tianjin-Hebei (BTH) region, and $43\;\mathrm{km}$ south of Liaodong peninsula. There is no industry found on it, and the majority of islanders live by fishing and seafood farming. Influenced by the Asian monsoon, the arriving air masses covered over most area of North China (Fig. S1), showing prefect regional background features.
During the sampling period, a total of $120\ \mathrm{PM}_{2.5}$ samples were collected every three day by a Tisch high volume sampler at a flow rate of $1.13\,\mathrm{\m}^{3}\,\mathrm{\min}^{-1}$ . Blank samples were also collected to subtract possible contamination occurring during or after sampling. The duration of every sampling was $24\,\mathrm{{h}}$ starting at 06:00 am (local time). The quartz fiber filters used were preheated at $450\;^{\circ}\mathbf{C}$ for 6 h in muffle furnace to remove the impurity. Before and after each sampling, quartz fiber filters were subjected to $24\,\mathrm{~h~}$ equilibration at $25\ \pm\ 1\ \ ^{\circ}\mathrm{C}$ and $50~\pm~2\%$ relative humidity, then analyzed gravimetrically using a Sartorius MC5 electronic microbalance. Each sample was weighted at least three times, and acceptable difference among the repetitions was less than $20~\upmu\mathrm{g}$ for a sampled filter and $10~\upmu\mathrm{g}$ for a blank filter. After weighting, filters were folded, wrapped in aluminium foil, sealed in airtight plastic bags, and then stored in refrigerators $(-20\mathrm{~}^{\circ}\mathrm{C})$ until chemical analysis.
2.2 Chemical analysis
Organic carbon (OC) and element carbon (EC) were analyzed by a Desert Research Institute (DRI) Model 2001 Carbon analyzer (Atmoslytic Inc., Calabasas, CA) based on the Interagency Monitoring of Protected Visual Environment (IMPROVE_A) thermal/optical reflected (TOR) protocol. Detailed information can be referred to our previous study (Zong et al., 2015). Regarding water-soluble ions (Na, $\mathbf{K}^{+}$ , $\mathrm{{Ca}}^{2+}$ , ${\mathrm{Mg}}^{2+}$ , $\mathrm{NH4}^{+}$ , Cl, $\mathrm{NO}{3}^{\ensuremath{-}}$ and $\mathrm{SO}4^{2-}$ ), they were measured by an ion chromatograph (Dionex ICS3000, Dionex Ltd., America) following the analysis method (Shahsavani et al., 2012) with the detection limit of $10~\mathrm{\,ng\,\ml^{-1}}$ (error $<\,5\%$ ). The concentrations of inorganic elements were determined by an inductively coupled plasma coupled with mass spectrometer (ICP-MS of ELAN DRC II type, Perkin Elmer Ltd., Hong Kong) based on previous analytical method (Wang et al., 2006). The detection limit was less than $0.01\;\mathrm{ng}\;\mathrm{ml}^{-1}$ , and error $<5\%$ . Noted, all the analytical species were blank-corrected by subtracting the average field blank value.
2.3 Data analysis method
2.3.1 HYSPLIT model
The hybrid single-particle Lagrangian integrated trajectory (HYSPLIT) model is a complete system for computing simple air parcel trajectories to complex dispersion simulation (Zhang et al., 2014), which is available on the National Oceanic and Atmospheric Administration Air Resource Laboratory website (www.arl.noaa.gov/ready/hysplit4.html). In this study, it was used to generate 72 h backward trajectory with 6 h interval, and 480 backward trajectories in total were calculated at 500 m above ground level (half of mixing height in North China). The calculated trajectories were then bunched into clusters by the clustering function in this model.
2.3.2 PMF model
Positive Matrix Factorization (PMF) recommended by the USEPA was utilized to qualify $\mathrm{PM}_{2.5}$ sources in present research. It is available at the website: www.epa.gov/air-research/positive-matrixfactorizationmodel-environmental-data-analyses. PMF does not require the source profiles prior to analysis and has no limitation on source number, which is an effective source apportionment receptor model and has been widely used in works about the source apportionment of $\mathrm{PM}_{2.5}$ (Zhang et al., 2013). The principles of PMF can be referred to elsewhere in detail (Amil et al., 2016). For species uncertainty, if the concentration is less than or equal to the method detection limit (MDL) used, the uncertainty (Unc) is calculated using the following equation: 人
$$
U\!n c\!=\!\frac{5}{6}\!\times\!M\!D\!L
$$
If the concentration is greater than the MDL used, the uncertainty would be calculated based on the equation:
$$
U n c\!=\!\sqrt{(E r r o r F r a c t i o n\!\times\!c o n c e n t r a t i o n)^{2}+(0.5\!\times\!M\!D L)^{2}}
$$
In total, twenty-eight chemical components were used for the PMF model, which included OC, EC, Cl-, $\mathrm{SO}_{4}{}^{2\cdot}$ -, $\mathrm{NO}{_{3}}^{-}$ , $\mathrm{Na^{+}}$ , $\mathbf{K}^{+}$ , $\mathrm{{Ca}}^{2+}$ , ${\mathrm{Mg}}^{2+}$ , $\mathrm{NH_{4}}^{+}$ , Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Cd, La, Ce, Pr, Nd, Pb, Th, U. To determine the optimal number of source factors, a string of effective test, in which factors number was from four to ten, was carried out. The resulting Q parameters were shown in Fig. S2. Obviously, there was a lowest QRobust value (7338) at seven factors in moving from four to ten factors. Although Qexpected has been decreasing in the process, Q/Qexpected shared similar variation with QRobust showing the lowest value at seven factors. In addition, PMF was run many times with different Fpeak to determine the range within which the objective QRobust remained relatively constant. Finally, an Fpeak value (-0.1) at seven factors demonstrated the most physically reasonable source profiles. Uncertainty of PMF model is usually estimated by bootstrapping (BS), displacement (DISP), and bootstrapping with displacement (BS-DISP). Here, characteristics of factors nearby seven, where QRobust was relative lower, were explored. With six factors, four factors (coal combustion $^+$ biomass burning, ship emission, sea salt, vehicle emission) were mapped $100\%$ of BS, while industry source and mineral dust were mapped $88\%$ and $91\%$ , respectively of runs. There were no swaps with DISP, and $100\%$ of the BS-DISP runs were successfully. At seven factors, results were more stable with all factors but ship emission (mapped $96\%$ of runs) mapped in BS in $100\%$ (Table S1), no swaps occurred with DISP and all BS-DISP runs were successfully. However, the solution became less stable in moving from seven to eight factors. The new vehicle dust factor was only mapped in BS in $76\%$ and ship emission factor was mapped in BS in $90\%$ , other factors were mapped in $100\%$ of runs. No swaps were found in DISP, but $24\%$ of BS-DISP runs were rejected due to factor swaps. Therefore, it suggested that seven factors were the optimal solution in this study.
2.3.3 PSCF model
The Potential Source Contribution Function (PSCF) is an effective method to identify regional source on the basis of HYSPLIT model. In our study, it was adopted to assess the potential source regions of the indicative source factors from PMF result using the respective daily contribution data (Bressi et al., 2014). Briefly, the ijth component of a PSCF field can be resumed as follows:
$$
P S C F_{i j}\!=\!m_{i j}/n_{i j}
$$
Where $\mathbf{n}_{\mathrm{ij}}$ is the total number of end points that fall in the ijth cell, and $\mathbf{m}_{\mathrm{ij}}$ is the number of endpoints of that parcel for which measured values exceed a user-determined threshold criterion ( $(75^{\mathrm{th}}$ percentile chosen in this study). Noted, cells with few endpoints may result in high uncertainty in PSCF method. Thus, to remove these high uncertainties, an arbitrary weight function $\mathbf{W}(\mathbf{n}_{\mathrm{{ij}}})$ was multiplied into the PSCF value:
$$
W(n_{i j}\,)\!=\!\!\left\{\!\!\begin{array}{c c}{{\![.00\!}}&{{\!(n\!\geq\!40)}}\\ {{\!0.7\!}}&{{\!(I0\!\leq\!n\!\leq\!40)}}\\ {{\!0.42\!}}&{{\!(5\!\leq\!n\!\leq\!I0)}}\\ {{\!0.I7\!}}&{{\!(n\!\leq\!5)}}\end{array}\!\!\right.
$$
3 Results and discussion
3.1 General characteristics of PM2.5 and its identified species
Table S2 shows a statistic summary of concentrations of $\mathrm{PM}_{2.5}$ , carbonaceous components, water-soluble ions (WSI) and inorganic elements on BH Island in the entire sampling period (annual and seasons). Generally, the concentration of $\mathrm{PM}_{2.5}$ ranged from $5.28~\upmu\mathrm{g}\;\mathrm{m}^{-3}$ to $267.11~\upmu\mathrm{g}\,\mathrm{m}^{-3}$ with an annual mean of $63.10\pm39.00\;\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , which was about twice of the First Grade National Ambient Air Quality Standard $(35~\upmu\mathrm{g}\,\:\mathrm{m}^{-3})$ of China (GB 3059-2012, www.zhb.gov.cn). Compared with other regions in China (Table S3), it was notably lower than those in typical urban areas, such as Beijing $(135~\upmu\mathrm{g}\,\\\,\mathrm{m}^{-3})$ ), Chengdu $(119~\upmu\mathrm{g}\,\\,\mathrm{m}^{-3})$ ) and Guanzhong $(134.7~\upmu\mathrm{g}\ \mathrm{m}^{-3})$ , while was similar with some background sites. This indicated our measurement primarily reflected a regional pollution pattern, which was further confirmed by its seasonal back trajectory clusters (Fig. S1). A distinct seasonal variation of $\mathrm{PM}_{2.5}$ was observed with the highest concentration $(72.45\mathrm{\;\mug\,m^{-3}})$ in winter and the lowest value ( $45.84\;\;\upmu\mathrm{g}\;\;\mathrm{m}^{-3})$ in summer. The concentration of $\mathrm{PM}_{2.5}$ is usually governed by emission, meteorological condition and deposition process (Tao et al., 2014), thus the higher $\mathrm{PM}_{2.5}$ concentration in winter could be partly attributed to the relative weaker atmospheric horizontal diffusion ability and vertical exchange capacity (Kumar et al., 2017). In addition, the wintertime residential coal combustion for heating in North China may be a great additional source for $\mathrm{PM}_{2.5}$ (Huang et al., 2017). Except for OC, EC and $\mathrm{SO}_{2}$ , it was also the major source of nitrogen oxide $(\mathbf{NO}_{\mathrm{x}})$ in winter, significantly promoting the formation of new particle (Pan et al., 2016; Wang et al., 2016a). By contrast, summer was possessed of favourable conditions for lower concentration, such as less emission and abundant rainfall (Lewandowski et al., 2007). On BH Island, the rainfall averaged $153.52\,\mathrm{~mm}$ in summer accounting for $66.96\%$ of the total year, which accelerated the efficient removal of PM2.5 from the atmosphere.
Carbonaceous species are important components in PM2.5 (Feng et al., 2009). On BH Island, the average concentrations of OC and EC were $4.90\pm3.69$ and $2.28\pm1.69\,\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , respectively. Seasonally, the variations of OC and EC concentrations were both similar with that of $\mathrm{PM}_{2.5}$ following the decreasing order of winter, autumn, spring and summer. The consistent variation of OC and EC may indicate their strong association of emission sources. For further understanding the relationship between OC and EC, the scatter plot was conducted as shown in Fig. 2. Obviously, OC and EC exhibited a good liner relation $(\mathbf{r}>0.52,\,\mathrm{p}<0.01)$ during all the sampling time indicating their strong co-emission, especially in winter $(\mathbf{r}=0.91)$ ). Generally, EC derives primarily from the incomplete combustion of carbon-contained materials, while OC includes primary organic carbon (POC) and secondary organic carbon (SOC) (Hu et al., 2014; Wang et al., 2014). The concentrations of SOC and POC were calculated by the EC tracer method:
SOCOC(OC/EC)primEC
POCOCSOC
Where $(\mathrm{OC/EC})_{\mathrm{prim}}$ is the ratio of OC/EC for primary emission. The minimum OC/EC was adopted as (OC/EC)prim (Du et al., 2014). The estimated concentrations of SOC and POC were 3.08 and $1.81~\upmu\mathrm{g}$ $\mathrm{m}^{-3}$ , respectively, with large seasonal discrepancies. The highest POC/OC (0.72) was found in winter followed by summer (0.66), spring (0.55) and autumn (0.45). It consisted with the order of correlation coefficient of OC and EC, also suggesting that the largest co-emission of OC and EC occurred in winter. Low-temperature burning, such as biomass burning, emits more OC compared with hightemperature burning (e.g. vehicle exhaust) (Gibson et al., 2013). Therefore the ratio of OC/EC could be used to evaluate the relative contribution of high- and low-temperature emission for OC and EC (Zong et al., 2016b). The ratio of OC/EC was 2.59, 1.80, 2.11 and 2.78 in spring, summer, autumn and winter, respectively. It suggested biomass burning contributed clearly to the carbonaceous species in winter on BH Island. Mann-Kendall (M-K) test could be adopted to test a non-liner trend and to determine the timing of each abrupt change for the time variation (Wang et al., 2014). In order to further identify the temporal tend of OC/EC, it was performed as shown in Fig. 3. Apparently, the abrupt point occurred at mid-May, exhibiting significant decrease at a $95\%$ confidence interval level. But this decrease trend started as early as the end of March. It was an interesting phenomenon because the change point was in accordance with the time of stopping heating in North China. Thus coal combustion may be an important source for OC and EC in winter in North China.
For WSI, their concentrations averaged $25.99\pm20.71\;\upmu\mathrm{g}\,\mathrm{m}^{-3}$ , which were $41.20\%$ of $\mathrm{PM}_{2.5}$ mass concentration. Among the ions, $\mathrm{SO}_{4}{}^{2-}$ ranked the highest with a mean value of $11.38\pm10.26\ \upmu\mathrm{g\m}^{-3}$ , followed by $\mathrm{NO}{_3}^{-}$ ( $6.45\pm4.92\mathrm{\;\upmug\;m^{-3}})$ and $\mathrm{NH_{4}}^{+}\,(4.91\pm4.10\mathrm{\,\upmug\,m}^{-3})$ . The three secondary inorganic aerosols (SIA) constituted the majority $(87.46\%)$ of the total WSI, which was relative higher than urban areas agreeing with the regional scale of their precursors in North China (Zhao et al., 2012). It further suggested that our measurement provided a regional signal of $\mathrm{PM}_{2.5}$ pollution in North China. By using ISORROPIA II model (Text S1), it can be seen $\mathrm{SO}_{4}{}^{2-}$ and $\mathrm{NO}{_3}^{-}$ were mostly combined with $\mathrm{NH4}^{+}$ presenting in the form of $\mathrm{NH}4\mathrm{NO}3$ and $\mathrm{(NH4)_{2}S O_{4}}$ , while the remainder $\mathrm{SO}_{4}{}^{2-}$ neutralized the alkalis, and all Cl was exclusively presented in the form of $\mathrm{NH}_{4}\mathrm{Cl}$ (Table S4; Text S2). $\mathbf{K}^{+}$ is usually treated as the marker of biomass burning (Fourtziou et al., 2017). Its concentration was relative higher in winter demonstrating the significance of biomass burning at that time, which consisted with the conclusion indicated by the higher ratio of OC/EC. Some typical ocean emission tracers, such as $\mathrm{Na^{+}}$ and ${\mathrm{Mg}}^{2+}$ , exhibited the highest concentration in spring, which may be due to the mechanical disruption of the ocean surface under the higher wind speed. The ratio of $\mathrm{{Cl}^{-}/\mathrm{{Na}^{+}}}$ was highest in winter with the value of 3.03, followed by autumn (1.47), summer (1.47) and spring (1.26). Compared with seawater (Cl/Na, 1.80), the highest ratio in winter can be ascribed to the additional emission of Cl from coal combustion (Tian et al., 2016), while the lower values in other seasons may be due to the Cl deletion from reaction of nitric and sulfuric acid with NaCl particles and subsequent volatilization of HCl (Genga et al., 2017). In addition, the highest concentration of $\mathrm{{Ca}}^{2+}$ , the typical tracer of dust, was found in spring, corresponding to the prevalence of dust events at that moment (Huang et al., 2010).
Inorganic elements constitute a minor fraction of $\mathrm{PM}_{2.5}$ mass. However, as primary pollutant, they are expected to conserve the finger-print of emission source for $\mathrm{PM}_{2.5}$ (Tan et al., 2014). Total concentration of analyzed inorganic elements was $781.82\pm670.96\;\mathrm{ng\;m^{-3}}$ , contributing $1.24\%$ to the $\mathrm{PM}_{2.5}$ mass. Its seasonal variability was not obvious. The lowest concentration was found in autumn, while values in spring, summer and winter were insignificant different $(95\%)$ . Anthropogenic Zn $\langle214.22\pm196.80\mathrm{\,ng\,m^{-3}}\rangle$ ) ranked the highest, followed by Fe $(212.10\pm267.48\mathrm{{\,ng\,m^{-3}}})$ and $\mathrm{Pb}$ (151.13 $\pm\ 257.06\ \mathrm{ng\,\m^{-3}})$ , implying the great contribution of human activities. In order to evaluate the contaminating degree, enrichment factors (EFs) method (Text S3) was employed as displayed in Fig. 4. Manifestly, the annual EFs of Ti, Yb, Lu, Tm, Er, Y, Ho, Th, Dy, Nd, Pr, Tb, Sm, Gd, Ce, Eu, U and La were lower than 10, indicating their main nature sources, while that of Co and Mn was slightly above 10, which respected a hybrid contribution from crustal and anthropogenic sources. The EFs of Cr, V, Ni, Cu, Zn, As, Cd and Pb were higher than 100, confirming their predominantly originations from human activities (Tao et al., 2014). Compared with other regions of China, such as Chengdu and Foshan, EFs of Ni and V (typical tracers of ship emission) on BH Island were much higher, revealing the ship emission for $\mathrm{PM}_{2.5}$ in Bohai region (Zhang et al., 2014). This was further confirmed by the ratio of V/Ni $[1.21\,\pm\,0.5)$ ), much higher than the threshold value (0.7). For further identifying the contaminating characteristic of these anthropogenic elements (Co, Mn, Cr, V, Ni, Cu, As, Zn, Pb and Cd), the seasonal features of EFs were exhibited in Table 1. The highest EFs of V and Ni were found in summer indicating the biggest shipping contribution at that point. This agreed with previous study demonstrating shipping activities were more frequent in summer (Wang et al., 2013). Apart from V and Ni, Cu and As, Pb and Cd shared homogeneous seasonal fluctuation for EFs, which may suggest their co-emission in Bohai region.
3.2 Source apportionment of PM2.5
3.2.1 Chemical mass closure
PM2.5 mass was reconstructed on a seasonal and annual basis by employing chemical mass closure method (Text S4). On an annual scale, good correlation between observed and reconstructed PM2.5 was found with an r of 0.87 (Fig. 5), suggesting a fine performance of the equation applied. The slope (0.68) was similar with that of Beijing (0.68) (Zhang et al., 2013), but lower than those in Chengdu (0.92) (Tao et al., 2014), Chongqing (0.80) (Yang et al., 2011) and Lanzhou (0.97) (Wang et al., 2016c). Correspondingly, the unidentified portion reached $32.07\%$ . One reason for such high ratio of the unidentified was the lack of water content in this study as various studies have shown that water content was one of the important components in PM2.5 mass. By using a KF system equipped with a controlled heating device, a recent study reported that the PM-bound water could constitute up to $22\%$ of the total PM matter (Perrino et al., 2016). Another possible reason was the varying factors adopted for the transformation from measured species to certain components. For example, OM/OC ratio often ranges from 1.4 to 2.2, which was intended as 1.8 in present study. If we employed the factor of 2.2, the unidentified percentage would fall by 3.11%.
OM, $\mathrm{SO}_{4}{}^{2\cdot}$ , $\mathrm{NO}{_3}^{\ensuremath{-}}$ , $\mathrm{NH_{4}}^{+}$ , EC, sea salt, mineral, TEO, $\mathbf{K}^{+}$ contributed $13.98\%$ , $18.03\%$ , $10.23\%$ , $7.78\%$ , $3.61\%$ , $3.4\%$ , $8.4\%$ , $1.08\%$ and $1.44\%$ of PM2.5 mass, respectively, with apparent seasonal variation. As displayed in Fig. S3, spring was possessed of the highest contribution of sea salt and mineral in the four seasons, which was in good agreement with the seasonal tendency of $\mathrm{Na^{+}}$ and $\mathrm{{Ca}}^{2+}$ caused by the great mechanical disruption of the ocean surface and the prevalence of dust events in spring. The highest contribution of SIA and TEO were found in summer. This can be ascribed to the perfect condition, such as opulent sunshine and oxidant, for the photochemistry and oxidation reaction at that point. Similar phenomenon was also reported in Beijing (Zhang et al., 2013). The proportions of OM, EC and $\mathbf{K}^{+}$ were maximal in winter led by the great contribution of biomass burning and resident coal combustion as discussed above. Overall, $\mathrm{SO}_{4}{}^{2\cdot}$ , $\mathrm{NO}{_3}^{-}$ and $\mathrm{NH_{4}}^{+}$ maintained the major components in $\mathrm{PM}_{2.5}$ throughout the year albeit with seasonal variation.
3.2.2 PMF model
By employing USEPA PMF5.0 model together with the obtained data set (120 samples $\times\ 28$ species), source contribution of $\mathrm{PM}_{2.5}$ was quantitatively explored. After iterative testing from 4 to 10 factors in modelling exercises, the pattern of seven factors were found providing the most physically reasonable source profiles with an Fpeak value of -0.1 and the lowest Q value (7338). The identified factors were coal combustion $^+$ biomass burning, industry source, ship emission, sea salt, mineral dust, refined chrome industry and vehicle emission, respectively. Their source profiles and percentage contributions were shown in Fig. 6. The first factor interpreted high proportions of OC, EC, $\mathrm{SO}_{4}{}^{2-}$ and $\mathbf{K}^{+}$ , which matched a mixing emission profiles including coal combustion and biomass burning (Zhang et al., 2013). Generally, $\mathbf{K}^{+}$ is an excellent tracer of biomass burning (Fourtziou et al., 2017), while coal combustion is often indicated by high OC, EC and $\mathrm{SO}_{4}{}^{2-}$ (Zong et al., 2016b). This factor was the largest contributor to $\mathrm{SO}_{4}{}^{2\cdot}$ , consisting with its emission inventory in North China (Zhao et al., 2013). Besides, the ratio of $\mathrm{NO}_{3}{}^{-}/\mathrm{SO}_{4}{}^{2-}$ was relative lower (0.18) proving the feature of coal combustion (Liu et al., 2014). Fig. 6 also displayed the time series of daily contribution from each factors. It can be seen that higher contribution of this factor occurred in winter, indicating the dominant status of residential coal and biomass burning at coal time in North China (Fourtziou et al., 2017; Zhou et al., 2017). This agreed with the discussed above, such as the higher OC/EC ratio, M-K test result and the relative higher concentration of $\mathbf{K}^{+}$ in winter.
The second factor was industry source, characterized by high Cu, Zn, As, Cd and Pb loadings. The factor profile was in line with that of Qimu Island in our previous study (Zong et al., 2016b). Equally, iron and steel industry may be very important among these industrial processes, because sintering process for iron and steel industry could produce a mass of heavy metals, especially for Cu, Pb and Zn (Wang et al., 2016c). In addition, the scale of steel production in North China was huge. As national statistical data shown, about half the world’s production of crude steel was from China, while BTH region and Shandong province contributed $25.3\%$ and $7.8\%$ , respectively (http: //www.stats.gov.cn/tjsj/ndsj/). Fig. 6 suggested industry source had higher contribution in winter, when the air masses mostly passed through the two regions (Fig. S1). The third factor represented ship emission characterized by high contributions of Ni and V and a high V/Ni ratio. High loadings of Ni and V are typically related with emissions of residual oil derived from ship activities (Pey et al., 2013). Previous study showed a high V/Ni ratio, usually more than 0.7, is always considered as a symbol of aerosols influenced by ship emissions (Zhang et al., 2014). In this profile, the V/Ni ratio was 1.69. Meanwhile, higher contribution of this factor happened in summer (Fig. 6) corresponding with the fact that summer was the most frequent period of shipping in Bohai Sea due to the perfect conditions of weather and sea state (Wang et al., 2013), which was also in accordance with the EFs result.
The fourth factor was characterized by high loadings of Cl, $\mathrm{Na^{+}}$ and ${\mathrm{Mg}}^{2+}$ , which was treated as signals of sea salt (Manousakas et al., 2017). It mostly came from the mechanical disruption of the sea surface. For example, winter and spring exhibited a relative higher contribution of this factor, and its highest contribution occurred at $28^{\mathrm{th}}$ April, 2015, when a strong wind raged in Bohai Sea. The Cl-/Na+ ratio was 1.43 in this profile, which was lower than the corresponding ratio in average seawater (1.80). This can be attributed to the Cl depletion as mentioned above (Masiol et al., 2017). The fifth factor was assigned as mineral dust with some typical crustal elements, such as Ti, Mn, Fe and rare earth metals (Xiong et al., 2017). The contribution of this factor was obvious higher in spring agreeing well with the prevalence of dust event at that time. The sixth factor, refined chrome industry, was characterized by high loading of Cr accompanied with some related metal elements (Mn, Fe, Co, Cu and Zn). Cr is one of the most important alloying elements. In China, Cr industry mainly distributes in Northeast China represented by some big chrome alloy factories, such as Jilin, Liaoyang, and Jinzhou. Correspondingly, summer demonstrated higher contribution for this factor, when part of air masses encountering BH Island derived from Northeast China (Fig. S1). Noticeably, this type of backward trajectory was significant different from other seasons. The seventh factors revealed high proportions of EC, $\mathrm{NO}{_3}^{-}$ and $\mathrm{NH_{4}}^{+}$ , which were all enriched in vehicle emission (Cui et al., 2016). Thereinto, $\mathrm{NH_{4}}^{+}$ was from vehicles equipped with three-way catalytic converters with the rapid increase of the vehicle number in nowadays. The $\mathrm{NO}_{3}^{-}/\mathrm{SO}_{4}^{2-}$ ratio was 1.74 in this profile, and this factor contributed $62.70\%$ of $\mathrm{NO}{_3}^{\ensuremath{-}}$ during the sampling period, suggesting the great vehicle emission. In addition, Fig. 6 indicated this factor’s contribution held no obvious seasonal variation, manifesting vehicle pollution in North China having regional feature.
Fig. 7 describes the respective contribution of the seven factors to PM2.5. Obviously, coal combustion $^+$ biomass burning and vehicle emission were the dominant contributors, accounting for $48.21\%$ and $30.33\%$ , respectively, during the study period. They were followed by mineral dust $(7.24\%)$ , ship emission $(6.63\%)$ , sea salt $(3.51\%)$ , industry source $(3.2\%)$ and refined chrome industry $(0.88\%)$ in the decreasing order. Comparatively, this PMF result was basically equal to that of chemical mass closure, validating the effectiveness of modelling simulation. For example, the contribution of sea salt was $3.51\%$ in PMF result, and that in chemical mass closure was $3.4\%$ . While the contribution of mineral dust was relative lower in PMF result, which may be caused by the irrelevant classification.
3.3 Regional source deduced from PSCF analysis
Based on the described above, we can see that the regional sources and transports of $\mathrm{PM}_{2.5}$ could exert a significant influence on the air quality of BH Island. To get a better understanding of this issue, the PSCF analysis on account of HYSPLIT model was used for identifying the potential source regions of the seven source factors given by PMF result (Bressi et al., 2014). Fig. 8 shows the PSCF map produced using the daily contribution date with the $75^{\mathrm{th}}$ percentile cut of each source factor. The colours represent the contribution level of regions, and dark colour could be associated with the high possibility for the grid cell to be an emission source, while the light colour indicated low possibility. Based on this map, the region surrounding Shandong peninsula and Hebei province were identified as the mediums to high potential region for coal combustion $^+$ biomass burning in North China, which was consistent with that found by previous studies. For example, Zong et al (Zong et al., 2016b) suggested biomass burning from Shandong peninsula contributed a large portion of PM2.5 in North China. In addition, it was noted that high contribution of this mixed source frequently occurred in cool season, when most back trajectories passed through the two regions. There were a large number of grid cells with PSCF values greater than 0.6 from the BTH region for industry source, indicating this region was an important industry seminary for $\mathrm{PM}_{2.5}$ in North China (Yao et al., 2016). It also agreed with mentioned above, illustratively, the contribution of crude steel in BTH region could occupy about $25.3\%$ in China, which contributed half of the world’s production of crude steel. Bohai Sea and East
China Sea were found to be of high source potential for ship emission (Zhang et al., 2014), where a great number of ships were running. As reported, 0.18 million water transport vessels and 1.07 million fishing boats active in Chinese sea by 2013, of which 1/3 lay in Bohai Sea. Besides, ship activities distributing in west coast of Korea may also influence this source (Song and Shon, 2014).
Regarding sea salt, the high probability partly came from geographical regions located northwest of BH Island. The mechanical disruption of sea surface was the dominant source for sea salt, which significantly hung on the high wind speed. Impacted by the East winter monsoon, the strong northwest wind became the source causer for BH Island. In addition, East China Sea affected by summer storms was also important source region. Mineral dust presumably came from the region of Mongolia. In fact, plenty of studies have suggested that Mongolia could be the main source region of dust emission for North China. When comparing mineral dust and refined chrome industry, distinct feature from the map can be found. Refined chrome industry mostly came from Liaoning, Jilin province and a lesser extent from BTH region and Shandong peninsula although contribution from overseas was also relative noticeable. This distribution area was consistent with the locations of some big chrome alloy factories, such as Jilin, Liaoyang, Jinzhou and Shougang as mentioned above. Lastly, vehicle emission was primarily of BTH region origin, centred on metropolises, such as Beijing and Tianjin. Indeed, vehicle emission is becoming one of the most important sources of severe air pollution due to the huge quantity and sharp growth of vehicles in Chinese megacities (Guo et al., 2016). Moreover, Shandong peninsula was also identified as an important source region, as except for the acknowledged high level of vehicle emission in BTH region, there were also many vehicle sources existing in Shandong peninsula. For example, the car ownership in Shandong province has ranked first in China by 2013 (Zong et al., 2016a).
4 Conclusions
The annual average concentration of PM2.5 was $63.10\pm39.00\ \upmu\mathrm{g\m}^{-3}$ on BH Island, a regional background site in North China. Compared with other regions in China, it was notably lower than typical urban areas, while was similar with some background sites, indicating our measurement primarily reflected a regional pollution pattern. A distinct seasonal variation of PM2.5 was observed with the highest concentration in winter and the lowest value in summer. The average concentration of OC and EC was $4.90\pm\:3.69$ and $2.28\pm1.69~\upmu\mathrm{g}~\mathrm{m}^{-3}$ accounting for $7.77\%$ and $3.61\%$ of $\mathrm{PM}_{2.5}$ , respectively. The consistent variation and good relationship between OC and EC suggested their strong co-emission during the sampling period. Biomass burning and residential coal combustion may be important sources for OC and EC in winter based on the higher OC/EC ratio and M-K test result. Among WSI, $\mathrm{SO}_{4}{}^{2-}$ ranked the highest with a mean concentration of $11.38\pm10.26\ \upmu\mathrm{g\m}^{-3}$ , followed by $\mathrm{NO}{_3}^{\prime}$ , $\mathrm{NH_{4}}^{+}$ , Cl, $\mathbf{K}^{+}$ , $\mathrm{Na^{+}}$ , $\mathrm{{Ca}}^{2+}$ and ${\mathrm{Mg}}^{2+}$ . In the thermodynamic equilibrium condition, $\mathrm{SO}_{4}{}^{2-}$ and $\mathrm{NO}{_3}^{-}$ were mostly combined with $\mathrm{NH_{4}}^{+}$ presenting in the form of $\mathrm{NH}_{4}\mathrm{NO}_{3}$ and $(\mathrm{NH}4)_{2}\mathrm{SO}_{4}$ , while the remainder $\mathrm{SO}_{4}{}^{2-}$ neutralized the alkalis, all Cl was exclusively presented in the form of $\mathrm{NH}4C1$ . Total concentration of analyzed inorganic elements was $781.82\pm670.96\,\mathrm{ng\,m^{-3}}$ , which accounted for $1.24\%$ of the $\mathrm{PM}_{2.5}$ mass. The EFs analysis implied Cr, V, Ni, Cu, Zn, As, Cd and Pb originated predominantly from human activities.
Chemical mass closure of $\mathrm{PM}_{2.5}$ was successfully conducted on an annual and seasonal basis. Results showed that OM, $\mathrm{SO}_{4}{}^{2-}$ , $\mathrm{NO}{3}^{\ensuremath{-}}$ , $\mathrm{NH4}^{+}$ , EC, sea salt, mineral, TEO, $\mathbf{K}^{+}$ contributed $13.98\%$ , $18.03\%$ , $10.23\%$ , $7.78\%$ , $3.61\%$ , $3.4\%$ , $8.4\%$ , $1.08\%$ and $1.44\%$ of $\mathrm{PM}_{2.5}$ mass, respectively, with apparent seasonal variation. Based on PMF model, seven source factors were identified; they are coal combustion $^+$ biomass burning $(48.21\%)$ , vehicle emission $(30.33\%)$ , mineral dust $(7.24\%)$ , ship emission $(6.63\%)$ , sea salt $(3.51\%)$ , industry source $(3.2\%)$ and refined chrome industry $(0.88\%)$ , respectively. Comparatively, this PMF result was basically equal to that of chemical mass closure, validating the effectiveness of modelling result. In addition, PSCF analysis using the daily contribution of each factor from PMF result was adopted to identify their potential source regions. Shandong peninsula and Hebei province were identified as high potential regions for coal combustion $^+$ biomass burning; BTH region was the main source region for industry source; Bohai Sea and East China Sea were found to be of high source potential for ship emission; Geographical regions located northwest of BH Island was possessed of high probability for sea salt; Mineral dust presumably came from the region of Mongolia; Refined chrome industry mostly came from Liaoning, Jilin province and a lesser extent from BTH region; The vehicle emission was primarily of BTH region origin, centred on metropolises, such as Beijing and Tianjin. Considering their respective contribution in $\mathrm{PM}_{2.5}$ , coal combustion $^+$ biomass burning in Shandong peninsula and Hebei province, and vehicle emission in BTH should be critically managed for the $\mathrm{PM}_{2.5}$ alleviation in North China.
Acknowledgements
This work was supported by the CAS Strategic Priority Research Program (Nos. XDA11020402, XDB05030303), and the Natural Scientific Foundation of China (Nos. 41471413; 31370464). The authors gratefully acknowledge the National Oceanic and Atmospheric Administration’s Air Resources Laboratory for providing the HYSPLIT transport model and the READY website (http://www.arl.noaa.gov/ready.html). The authors also acknowledge the office-holders of Environmental Monitoring Station of State Ocean Administration of China on Beihuangcheng (BH) Island for the sampling work.
References
Amil N, Latif MT, Khan MF, Mohamad M. Seasonal variability of $\mathrm{PM}_{2.5}$ composition and sources in the Klang Valley urban-industrial environment. Atmospheric Chemistry and Physics 2016; 16: 5357-5381.
Bressi M, Sciare J, Ghersi V, Mihalopoulos N, Petit JE, Nicolas JB, et al. Sources and geographical origins of fine aerosols in Paris (France). Atmospheric Chemistry and Physics 2014; 14: 8813- 8839.
Cui M, Chen Y, Tian C, Zhang F, Yan C, Zheng M. Chemical composition of $\mathrm{PM}_{2.5}$ from two tunnels with different vehicular fleet characteristics. Science of the Total Environment 2016; 550: 123- 132.
Du Z, He K, Cheng Y, Duan F, Ma Y, Liu J, et al. A yearlong study of water-soluble organic carbon in Beijing I: Sources and its primary vs. secondary nature. Atmospheric Environment 2014; 92: 514-521.
Feng Y, Chen Y, Guo H, Zhi G, Xiong S, Li J, et al. Characteristics of organic and elemental carbon in $\mathrm{PM}_{2.5}$ samples in Shanghai, China. Atmospheric Research 2009; 92: 434-442.
Fourtziou L, Liakakou E, Stavroulas I, Theodosi C, Zarmpas P, Psiloglou B, et al. Multi-tracer approach to characterize domestic wood burning in Athens (Greece) during wintertime. Atmospheric Environment 2017; 148: 89-101.
Genga A, Ielpo P, Siciliano T, Siciliano M. Carbonaceous particles and aerosol mass closure in $\mathrm{PM}_{2.5}$ collected in a port city. Atmospheric Research 2017; 183: 245-254.
Gibson MD, Pierce JR, Waugh D, Kuchta JS, Chisholm L, Duck TJ, et al. Identifying the sources driving observed $\mathrm{PM}_{2.5}$ temporal variability over Halifax, Nova Scotia, during BORTAS-B. Atmospheric Chemistry and Physics 2013; 13: 7199-7213.
Guo X, Fu L, Ji M, Lang J, Chen D, Cheng S. Scenario analysis to vehicular emission reduction in Beijing-Tianjin-Hebei (BTH) region, China. Environmental Pollution 2016; 216: 470-479.
Hu G, Zhang Y, Sun J, Zhang L, Shen X, Lin W, et al. Variability, formation and acidity of watersoluble ions in PM2.5 in Beijing based on the semi-continuous observations. Atmospheric Research 2014; 145-146: 1-11.
Huang D, Xiu G, Li M, Hua X, Long Y. Surface components of $\mathrm{PM}_{2.5}$ during clear and hazy days in Shanghai by ToF-SIMS. Atmospheric Environment 2017; 148: 175-181.
Huang K, Zhuang G, Li J, Wang Q, Sun Y, Lin Y, et al. Mixing of Asian dust with pollution aerosol and the transformation of aerosol components during the dust storm over China in spring 2007. Journal of Geophysical Research-Atmospheres 2010; 115.
Kotchenruther RA. Source apportionment of $\mathrm{PM}_{2.5}$ at multiple Northwest US sites: Assessing regional winter wood smoke impacts from residential wood combustion. Atmospheric Environment 2016; 142: 210-219.
Kumar M, Raju MP, Singh RK, Singh AK, Singh RS, Banerjee T. Wintertime characteristics of aerosols over middle Indo-Gangetic Plain: Vertical profile, transport and radiative forcing. Atmospheric Research 2017; 183: 268-282.
Lewandowski M, Jaoui M, Kleindienst TE, Offenberg JH, Edney EO. Composition of PM2.5 during the summer of 2003 in Research Triangle Park, North Carolina. Atmospheric Environment 2007; 41: 4073-4083.
Li H, Wang Qg, Yang M, Li F, Wang J, Sun Y, et al. Chemical characterization and source apportionment of PM2.5 aerosols in a megacity of Southeast China. Atmospheric Research 2016; 181: 288-299.
Liu D, Li J, Zhang Y, Xu Y, Liu X, Ding P, et al. The Use of Levoglucosan and Radiocarbon for Source Apportionment of $\mathrm{PM}_{2.5}$ Carbonaceous Aerosols at a Background Site in East China. Environmental Science & Technology 2013; 47: 10454-10461.
Liu J, Li J, Zhang Y, Liu D, Ding P, Shen C, et al. Source Apportionment Using Radiocarbon and Organic Tracers for $\mathrm{PM}_{2.5}$ Carbonaceous Aerosols in Guangzhou, South China: Contrasting
Local- and Regional-Scale Haze Events. Environmental Science & Technology 2014; 48:
12002-12011.
Manousakas M, Papaefthymiou H, Diapouli E, Migliori A, Karydas AG, Bogdanovic-Radovic I, et al. Assessment of $\mathrm{PM}_{2.5}$ sources and their corresponding level of uncertainty in a coastal urban area using EPA PMF 5.0 enhanced diagnostics. Science of the Total Environment 2017; 574: 155-164. 人
Masiol M, Hopke PK, Felton HD, Frank BP, Rattigan OV, Wurth MJ, et al. Source apportionment of PM2.5 chemically speciated mass and particle number concentrations in New York City. Atmospheric Environment 2017; 148: 215-229.
Niu X, Cao J, Shen Z, Ho SSH, Tie X, Zhao S, et al. PM2.5 from the Guanzhong Plain: Chemical composition and implications for emission reductions. Atmospheric Environment 2016; 147: 458-469.
Pan Y, Wang Y, Zhang J, Liu Z, Wang L, Tian S, et al. Redefining the importance of nitrate during haze pollution to help optimize an emission control strategy. Atmospheric Environment 2016; 141: 197-202.
Perrino C, Catrambone M, Farao C, Canepari S. Assessing the contribution of water to the mass closure of $\mathrm{{PM}_{10}}$ . Atmospheric Environment 2016; 140: 555-564.
Pey J, Perez N, Cortes J, Alastuey A, Querol X. Chemical fingerprint and impact of shipping emissions over a western Mediterranean metropolis: Primary and aged contributions. Science of the Total Environment 2013; 463: 497-507.
Shahsavani A, Naddafi K, Haghighifard NJ, Mesdaghinia A, Yunesian M, Nabizadeh R, et al. Characterization of ionic composition of TSP and $\mathrm{{PM}_{10}}$ during the Middle Eastern Dust (MED) storms in Ahvaz, Iran. Environmental Monitoring and Assessment 2012; 184: 6683-6692.
Song S-K, Shon Z-H. Current and future emission estimates of exhaust gases and particles from shipping at the largest port in Korea. Environmental Science and Pollution Research 2014; 21: 6612-6622.
Tan JH, Duan JC, Ma YL, Yang FM, Cheng Y, He KB, et al. Source of atmospheric heavy metals in winter in Foshan, China. Science of the Total Environment 2014; 493: 262-70.
Tao J, Gao J, Zhang L, Zhang R, Che H, Zhang Z, et al. PM2.5 pollution in a megacity of southwest China: source apportionment and implication. Atmospheric Chemistry and Physics 2014; 14: 8679-8699.
Tian M, Wang H, Chen Y, Yang F, Zhang X, Zou Q, et al. Characteristics of aerosol pollution during heavy haze events in Suzhou, China. Atmospheric Chemistry and Physics 2016; 16: 7357- 7371.
Wang G, Zhang R, Gomez ME, Yang L, Zamora ML, Hu M, et al. Persistent sulfate formation from London Fog to Chinese haze. Proceedings of the National Academy of Sciences of the United States of America 2016a; 113: 13630-13635.
Wang L, Qi JH, Shi JH, Chen XJ, Gao HW. Source apportionment of particulate pollutants in the atmosphere over the Northern Yellow Sea. Atmospheric Environment 2013; 70: 425-434. O
Wang S-H, Hung W-T, Chang S-C, Yen M-C. Transport characteristics of Chinese haze over Northern Taiwan in winter, 2005-2014. Atmospheric Environment 2016b; 126: 76-86.
Wang X, Bi X, Sheng G, Fu H. Hospital indoor $\mathrm{PM_{10}}/\mathrm{PM_{2.5}}$ and associated trace elements in Guangzhou, China. Science of the Total Environment 2006; 366: 124-135.
Wang X, Chen Y, Tian C, Huang G, Fang Y, Zhang F, et al. Impact of agricultural waste burning in the Shandong Peninsula on carbonaceous aerosols in the Bohai Rim, China. Science of the Total Environment 2014; 481: 311-6.
Wang Y, Jia C, Tao J, Zhang L, Liang X, Ma J, et al. Chemical characterization and source apportionment of PM2.5 in a semi-arid and petrochemical-industrialized city, Northwest China. Science of the Total Environment 2016c; 573: 1031-1040.
Wu H, Zhang Y-f, Han S-q, Wu J-h, Bi X-h, Shi G-l, et al. Vertical characteristics of $\mathrm{PM}_{2.5}$ during the heating season in Tianjin, China. Science of the Total Environment 2015; 523: 152-160.
Xiong Y, Zhou J, Schauer JJ, Yu W, Hu Y. Seasonal and spatial differences in source contributions to PM2.5 in Wuhan, China. Science of the Total Environment 2017; 577: 155-165.
Yang F, Tan J, Zhao Q, Du Z, He K, Ma Y, et al. Characteristics of PM2.5 speciation in representative megacities and across China. Atmospheric Chemistry and Physics 2011; 11: 5207-5219.
Yao L, Yang L, Yuan Q, Yan C, Dong C, Meng C, et al. Sources apportionment of $\mathrm{PM}_{2.5}$ in a background site in the North China Plain. Science of the Total Environment 2016; 541: 590- 598.
Zhang F, Chen Y, Tian C, Wang X, Huang G, Fang Y, et al. Identification and quantification of shipping emissions in Bohai Rim, China. Science of the Total Environment 2014; 497-498: 570-7.
Zhang R, Jing J, Tao J, Hsu SC, Wang G, Cao J, et al. Chemical characterization and source apportionment of PM2.5 in Beijing: seasonal perspective. Atmospheric Chemistry and Physics 2013; 13: 7053-7074.
Zhao B, Wang P, Ma JZ, Zhu S, Pozzer A, Li W. A high-resolution emission inventory of primary pollutants for the Huabei region, China. Atmospheric Chemistry and Physics 2012; 12: 481- 501.
Zhao PS, Dong F, He D, Zhao XJ, Zhang XL, Zhang WZ, et al. Characteristics of concentrations and chemical compositions for $\mathrm{PM}_{2.5}$ in the region of Beijing, Tianjin, and Hebei, China. Atmospheric Chemistry and Physics 2013; 13: 4631-4644.
Zhou S, Yang L, Gao R, Wang X, Gao X, Nie W, et al. A comparison study of carbonaceous aerosols in a typical North China Plain urban atmosphere: Seasonal variability, sources and implications to haze formation. Atmospheric Environment 2017; 149: 95-103.
Zong Z, Chen Y, Tian C, Fang Y, Wang X, Huang G, et al. Radiocarbon-based impact assessment of open biomass burning on regional carbonaceous aerosols in North China. Science of the Total Environment 2015; 518-519: 1-7. 人
Zong Z, Wang X, Tian C, Chen Y, Han G, Li J, et al. Source and formation characteristics of watersoluble organic carbon in the anthropogenic-influenced Yellow River Delta, North China. Atmospheric Environment 2016a; 144: 124-132.
Zong Z, Wang X, Tian C, Chen Y, Qu L, Ji L, et al. 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. Atmospheric Chemistry and Physics 2016b; 16: 11249- 11265.
Highlights
► 120 $\mathsf{P M}_{2.5}$ samples were collected at a regional background site in North China
► PMF combined PSCF were adopted to explore the source and origin regions for PM2.5
► Coal combustion $^+$ biomass burning and vehicle emission were the dominant sources
► Shandong and BTH were the main source areas for $\mathsf{P M}_{2.5}$ in North China
|
atmosphere
|
0110
|
10.1016/j.apr.2017.05.001
|
image
|
Fig. 1. Spatial distribution of five sampling sites in Dongguan and their positions relative to two neighboring cities (Guangzhou and Shenzhen).
|
|
atmosphere
|
0110
|
10.1016/j.apr.2017.05.001
|
table
|
Table 1 Meteorological conditions and sample information during the four sampling seasons in 2014 in Dongguan.
|
|
atmosphere
|
0110
|
10.1016/j.apr.2017.05.001
|
image
|
Fig. 2. Typical back trajectories in the sampling seasons in Dongguan.
|
|
atmosphere
|
0110
|
10.1016/j.apr.2017.05.001
|
image
|
Fig. 3. Chemical compositions of $\mathrm{PM}_{2.5}$ in Dongguan: (a) composition structure of annual average $\mathrm{PM}_{2.5};$ (b) seasonal average compositions; (c) spatial distribution of annual average compositions.
|
|
atmosphere
|
0110
|
10.1016/j.apr.2017.05.001
|
table
|
Table 2 Comparison of major chemical compositions of $\mathrm{PM}_{2.5}$ between Dongguan and other cities $(\upmu\mathrm{g}/\upmu^{3})$ .
|
|
atmosphere
|
0110
|
10.1016/j.apr.2017.05.001
|
image
|
B.-B. Zou et al. / Atmospheric Pollution Research xxx (2017) 1e10 Fig. 4. Explained variations of eight factors and factor profiles of $\mathrm{PM}_{2.5}$ obtained by PMF.
|
|
atmosphere
|
0110
|
10.1016/j.apr.2017.05.001
|
image
|
B.-B. Zou et al. / Atmospheric Pollution Research xxx (2017) 1e10 Fig. 5. Seasonal and spatial variations of mass contributions of the eight factors output by PMF and the SOA calculated: (a) seasonal variations; (b) spatial variations.
|
|
atmosphere
|
0110
|
10.1016/j.apr.2017.05.001
|
image
|
Fig. 6. Source structure of $\mathrm{PM}_{2.5}$ in Dongguan in 2014: (a) source structure of annual average $\mathrm{PM}_{2.5}$ at the five sites; (b) the relationship between the measured and modelreconstructed $\mathrm{PM}_{2.5}$ mass.
|
|
atmosphere
|
0110
|
10.1016/j.apr.2017.05.001
|
image
|
Fig. 7. Contributions of sources of $\mathrm{PM}_{2.5}$ on different pollution levels in fall and winter (numbers above the bars indicate the increasing ratio from NSD to ESD).
|
|
atmosphere
|
0110
|
10.1016/j.apr.2017.05.001
|
text
| Not supported with pagination yet |
Source apportionment of PM2.5 pollution in an industrial city in southern China
Bei-Bing Zou, Xiao-Feng Huang, Bin Zhang, Jing Dai, Li-Wu Zeng, Ning Feng, Ling-Yan He\*
Key Laboratory for Urban Habitat Environmental Science and Technology, School of Environment and Energy, Peking University 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 December 2016
Received in revised form
26 April 2017
Accepted 1 May 2017
Available online xxx
Keywords:
PM2.5
Source apportionment
Secondary organic aerosol (SOA)
Positive matrix factorization (PMF)
Pearl River Delta (PRD)
Severe $\mathsf{P M}_{2.5}$ pollution has become a great challenge to atmospheric pollution control in China. Most of previous aerosol source apportionment studies in China focused only on part of $\mathrm{PM}_{2.5}$ (e.g., organic matter in composition or $\mathrm{PM}_{1}$ in size) or lacked source contributions identified with necessary tempospatial variations, which makes the results not convincible enough for policy making. In this study, five various sites were selected for simultaneous $\mathrm{PM}_{2.5}$ observation in an industrial city in the Pearl River Delta (PRD) of South China during all four seasons of 2014. A positive matrix factorization (PMF) model was applied to the datasets of measured chemical species to perform source apportionment with the results as: (1) The annual mean $\mathsf{P M}_{2.5}$ concentration was $53~\upmu\mathrm{g}/\mathrm{m}^{3}$ , with vehicle emissions, secondary sulfate, biomass combustion, and secondary organic aerosol (SOA) identified as the major sources, contributing $21\%$ , $20\%$ , $11\%$ , and $10\%$ to $\mathrm{PM}_{2.5}$ , respectively. Ship emissions, fugitive dust, secondary nitrate, industrial emissions, and coal burning each contributed $5\%{-}8\%$ (2) The tempo-spatial variations of sources reveal that secondary sulfate, biomass combustion, SOA, and ship emissions had obvious regional pollution characteristics; however, vehicle emissions, secondary nitrate, coal burning, fugitive dust, and industrial emissions showed obvious local emission characteristics. (3) The exceeding standard days $(\mathsf{P M}_{2.5}{>}75\ \upmu\mathrm{g}/\mathsf{m}^{3})$ appeared with secondary nitrate, SOA, and biomass burning increasing mostly in concentration, indicating that the relevant primary sources or precursor emissions should be controlled more strictly. This study highlights the importance of SOA in $\mathrm{PM}_{2.5}$ pollution in China, which has been scarcely quantified for bulk $\mathrm{PM}_{2.5}$ in the literature.
$\copyright$ 2017 Turkish National Committee for Air Pollution Research and Control. Production and hosting by Elsevier B.V. All rights reserved.
1. Introduction
Environmental particulate matter plays an important role in air pollution, especially fine particulate matter $(\mathsf{P M}_{2.5})$ , which can scatter and absorb sunlight, reduce atmospheric visibility, increase radiation forcing, and affect the global climate (Charlson et al., 1992). The complex chemical composition of fine particulate matter includes acidic substances, polycyclic aromatic hydrocarbons (PAHs) and heavy metals, which can penetrate the lungs of humans to cause respiratory and cardiovascular diseases (Sarnat et al., 2008). In 2010, more than 3 million people died prematurely owing to severe fine particulate matter pollution, which has become one of the most important risk factors for disease in East Asia (Lim et al., 2012).
China is experiencing severe air pollution because of its rapid economic development and urbanization. According to satellite aerosol optical depth (AOD) data (Donkelaar et al., 2010) and ground $\mathsf{P M}_{2.5}$ monitoring data, $\mathsf{P M}_{2.5}$ pollution is most serious in the BeijingeTianjineHebei region, Sichuan Basin, Yangtze River Delta, and Pearl River Delta (PRD), in which the $\mathsf{P M}_{2.5}$ annual mean concentration is $52{\-}{\-}149~\upmu\mathrm{g}/\mathfrak{m}^{3}$ (Zhang et al., 2015; Zhou et al., 2016). The PRD is one of the most economically developed regions in China and is also a region of severe air pollution. Owing to the early implementation of strict air pollution control measures, the air quality in the PRD region has improved over other regions with severe air pollution; however, the understanding of sources of $\mathsf{P M}_{2.5}$ in the PRD is still lacking. Internationally used mathematical models such as positive matrix factorization (PMF), chemical mass balance (CMB), UNMIX, and principal component analysis (PCA) were applied to determine the source apportionment of $\mathsf{P M}_{2.5}$ (Viana et al., 2008; Watson et al., 2008; Karagulian et al., 2015). In recent studies, the CMB model has been rarely used to analyze the source apportionment of $\mathsf{P M}_{2.5}$ in the PRD (Wang et al., 2016). Instead, the PMF model has been widely applied in regions such as Guangzhou (Gao et al., 2013; Kuang et al., 2015), Shenzhen (Huang et al., 2014), Dongguan (Wang et al., 2015) and Foshan (Tan et al., 2016). However, such studies have either few sampling sites or short-term observation, either of which creates difficulty in accurately identifying the spatial and temporal characteristics of the sources. In terms of source apportionment studies based on on-line measurements, such as using the popular aerosol mass spectrometers, they mainly focused on organic matter in submicron particles rather than bulk $\mathsf{P M}_{2.5}$ (e.g., He et al., 2011; Gong et al., 2012).
Dongguan, with a population of 8.2 million, is one of the fastest-growing cities in the PRD and is one of the world's largest manufacturing bases. Consequently, the air pollution in this region remains severe; therefore, accurate determination of the sources of $\mathsf{P M}_{2.5}$ in Dongguan is urgently needed. In this study, long-term $\mathsf{P M}_{2.5}$ sample collection at multiple sites was conducted in Dongguan. Most of the components of $\mathsf{P M}_{2.5}$ were analyzed; the sources of $\mathsf{P M}_{2.5}$ were apportioned by applying the PMF model, and the source characteristics of $\mathsf{P M}_{2.5}$ on different pollution levels were analyzed. Dongguan is representative of an industrial city experiencing severe air pollution owing to rapid economic growth. Therefore, the source apportionment of $\mathsf{P M}_{2.5}$ in this study provides useful references for air pollution control strategies in this region.
2. Instruments and methods
2.1. Sampling and analysis
Dongguan is located on the east coast of Pearl River estuary, bordering Guangzhou to the northwest, Shenzhen to the south, and Huizhou to the east, and its climate is subtropical monsoon. The five sampling sites are distributed along the approximately eastewest axis and include Qingxi (QX), Songshan Lake (SSL), Changan (CA), Nancheng (NC), and Machong (MC), respectively (Fig. 1). MC, CA, and SSL are located in northwest, southwest, and middle suburban areas, where industry is dominant. NC represents the downtown area. QX is regarded as a background site, since it is located on the eastern edge of Dongguan and thus little influenced by the other areas of Dongguan. In addition, the QX site is surrounded by mountains and has negligible local emissions, implying its air pollution is highly controlled by regional transport, especially from the north in winter as discussed later. At each site, a sampler was set up on the roof of a 3e5 floor building, where no significant pollutant emission or vehicle traffic exists at least within $200~\mathrm{m}$ Thermo 2300 four-channel samplers (Thermo Fisher Scientific Inc., Waltham, Massachusetts, USA, with a flowrate of $16.7\,\mathrm{L/min}$ for two channels and a flowrate of $10.0~\mathrm{L/min}$ for the other two channels) were set at NC and SSL, and TH-16A four-channel samplers (Tianhong Corp., Wu Han, China, with a flow rate of $16.7~\mathrm{L/min}$ for all channels) were set at MC, CA, and QX. The parallel sampling by the two types of samples showed a relative deviation of less than $5\%$ Teflon filters were applied in two channels of each sampler, and quartz filters were applied in the remaining two channels. December 11, 2013eJanuary 10, 2014; April 3eMay 6, 2014; July 5eAugust 4, 2014; and October 6eNovember 11, 2014, were carefully selected to represent winter, spring, summer, and fall seasons in 2014, respectively. Sampling was conducted intermittently for $^{24\mathrm{~h~}}$ during the period of observation. From the five sampling sites, 301 samples were collected.
The weight difference of Teflon filters before and after sampling was used to calculate the daily average mass concentration of $\mathsf{P M}_{2.5}$ . The Teflon filter samples of one channel were analyzed for 15 metallic elements by using an inductively coupled plasma mass spectrometer (ICP-MS, 7500c; Agilent Technologies, Santa Clara, California, USA); those for the second channel were analyzed for water-soluble inorganic ions by using an ion chromatography system (ICS-2500, Dionex; Sunnyvale, California, USA). The quartz filter samples were analyzed for organic carbon (OC) and elemental carbon (EC) by using an OC/EC analyzer (Desert Research Institute, Reno, Nevada, USA). Organic matter (OM) mass was estimated by $1.8\,\times\,00$ (He et al., 2011). We used the same laboratory and instruments as detailed in He et al. (2008).
Table 1 presents the general meteorological conditions during the sampling seasons in Dongguan, indicating that the city had a hot and humid summer and a cool and dry winter, while spring and summer were two transition seasons. Fig. 2 further presents the back trajectory clusters of each season during the period of observation, which were determined by using the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) Model. The 48-h back trajectories beginning at 12:00 UTC were calculated every day during the observation period, and the starting altitude was $100\;\mathrm{m}$ in Dongguan. The results of the HYSPLIT Model showed that the air mass originated from the northern mainland in winter, from northeast coastal areas in spring, from the South China Sea in summer, and from northern mainland and northeast coastal areas in the fall. Therefore, the pollution characteristics of different seasons could be determined by observation during the representative months of the four seasons because the sources of the air mass varied among the seasons.
2.2. Source apportionment using PMF receptor model
A receptor model is a mathematical model used to analyze the chemical composition and physical properties of ambient atmospheric particulates through quantitative identification of the contribution of pollution sources based on source profiles or source fingerprint characteristics (Norris and Duvall, 2014). The PMF model is a commonly applied receptor model because knowledge of the source profiles and photochemistry reaction mechanisms is not necessary. PMF is a multivariate factor analysis tool that decomposes the matrix $\Chi_{i j}$ of the sample data into two matrices: factor contributions $\left(G_{i k}\right)$ and factor profiles $(\mathsf{F}_{k j})$ . $\mathrm{E}_{i j}$ is the matrix of residuals representing the differences between observation and the model (Eq. (1)). The PMF solution, which minimizes the object function Q (Eq. (2)) by the least squares method, is based on the uncertainty $\left(\upsigma_{i j}\right)$ of the observed value $(\mathbf{X}_{i j};$ Paatero and Tapper, 1994; Paatero, 1997):
$$
\mathsf{E}_{i j}=\mathsf{X}_{i j}-\sum_{k=1}^{p}\mathsf{G}_{i k}\mathsf{F}_{k j}\quad(\mathrm{~}i=1,\;...,m,\,j=1,...,n)
$$
$$
\mathsf Q(E)=\sum_{i=1}^{m}\sum_{j=1}^{n}\bigl(\mathsf E_{i j}/\mathsf{\sigma}_{i j}\bigr).
$$
This study used U.S. Environmental Protection Agency (EPA) PMF v5.0 software to conduct the simulation. The concentrations of $\mathsf{N H}_{4}^{+}$ , Cl⁻, $\mathsf{N O}_{3}^{\phantom{\,}}$ , $S0_{4}$ 2⁻, OM, EC, Na, Mg, Al, K, Ca, V, Fe, Zn, and $\mathsf{P b}$ species (totally accounting for $95.3\%$ of the $\mathsf{P M}_{2.5}$ mass on average) in the 301 samples from the five sites were input into the PMF model. Other metallic elements having very low concentrations and little implication for pollution sources were not included in the model. When the concentrations of species were lower than the detectable level, one-half of the detectable level of the corresponding species was substituted for model input. The uncertainties of the species concentration were set to $20\%$ , as a maximum estimation of the deviations of the sampling and analysis, in this model. The PMF modeling results will be detailed in Section 3.2.
3. Results and discussion
3.1. Chemical composition and spatiotemporal distribution of $P M_{2.5}$
The annual average mass concentration of $\mathsf{P M}_{2.5}$ during the entire year of observation at all five sites in 2014 was $53~\upmu\mathrm{g}/\mathrm{m}^{3}$ , which is 0.5 times higher than the Grade II national standards for air quality, at $35~\ensuremath{\,\upmu\mathrm{g}}/\ensuremath{\ensuremath{\mathrm{m}}}^{3}$ . Fig. 3 (a) shows the annual average chemical composition of $\mathsf{P M}_{2.5}$ in Dongguan. OM was dominant, accounting for $38\%$ ; this indicates that the organic pollution was severe in Dongguan. The other major nonmetallic components of $\mathsf{P M}_{2.5}$ were $\mathsf{S O}_{4}^{2}$ , $\mathsf{N H4}^{+}$ , EC, $\mathrm{NO}{_3}^{-}$ , and $C1^{-}$ , contributing $23\%$ , $11\%$ , $8\%$ , $8\%$ , and $1\%$ of the total $\mathsf{P M}_{2.5}$ , respectively. All 15 metallic elements were determined by ICP-MS, accounting for $6\%$ . Among these, the contributions of K and Na elements were higher, both exceeding $1\%$ . In addition, undetected components such as $\mathsf{S i O}_{2}$ , water, and oxygen components of metallic elements accounted for $6\%$ . Compared with the results of long-term $\mathsf{P M}_{2.5}$ observation reported in recent years in other cities in China (Table 2), the annual average $\mathsf{P M}_{2.5}$ mass concentrations in the PRD's cities were significantly lower than that in Tianjin, Zhejiang, and Sichuan. However, the $\mathsf{P M}_{2.5}$ annual average concentration in Dongguan was in the medium pollution level in the PRD region. EC was higher both in Shenzhen and Dongguan because these port cities are significantly affected by heavy-duty diesel vehicles and port cargo ships.
Seasonal variations in the species concentration of $\mathsf{P M}_{2.5}$ are shown in Fig. 3 (b). The $\mathsf{P M}_{2.5}$ pollution level was distinctly lower in spring and summer and higher in autumn and winter. The highest seasonal average concentration was $84\,\upmu\mathrm{g}/\mathrm{m}^{3}$ in winter, which is 1.6 times higher than the lowest value, $32~\upmu\mathrm{g}/\mathrm{m}^{3}$ in summer. Major components of $\mathsf{P M}_{2.5}$ such as OM, $\mathrm{SO}_{4}^{2.}$ , $\mathsf{N O}_{3}^{\phantom{\,}-}$ , $\mathsf{N H}_{4}^{+}$ and EC showed seasonal variations similar to those of $\mathsf{P M}_{2.5}$ . Moreover, the seasonal variations in $\mathsf{P M}_{2.5}$ concentration correlated with the meteorological conditions. The back trajectories of the air mass in Dongguan in Fig. 2 show that air mass in winter effectively carried pollutants from the northern mainland to Dongguan, resulting in more severe air pollution. In the winter of 2014, the northern provinces covered by the back trajectories in Fig. 2 had a $\mathsf{P M}_{2.5}$ level of $52{-}98~\upmu\mathrm{g}/\mathrm{m}^{3}$ while the province of Guangdong, where Dongguan is located, had a much lower $\mathsf{P M}_{2.5}$ level of $48~\upmu\mathrm{g}/\mathrm{m}^{3}$ , as the official monitoring data indicated (www.zhb.gov.cn). He et al. (2011) analyzed the relationship between the back trajectories of air mass and $\mathrm{PM}_{1}$ pollution in the neighboring cityeShenzhen, and found that the northerly back trajectories corresponded to a $\mathsf{P M}_{1}$ pollution level that was 1.6e4.3 times of that associated with the southerly back trajectories. The air mass of high humidity from the northeast coastal areas in spring reduced the air pollution in Dongguan. However, the lowest pollutant concentration appeared in summer owing to the clean air mass from the South China Sea. The air mass originated from the northern mainland and northeast coastal areas in fall; therefore, the pollution level in fall was higher than that in spring and lower than that in winter. Such seasonal variation of $\mathsf{P M}_{2.5}$ pollution in this region was also revealed in the literature (Hagler et al., 2006; Huang et al., 2014).
Fig. 3(c) gives the spatial distribution of the annual average concentrations of total $\mathsf{P M}_{2.5}$ and its components at the five sites shown in Fig. 1. The $\mathsf{P M}_{2.5}$ pollution levels increased from east to west in Dongguan. At QX, SSL, CA, and NC from east to west and southwestern MC, the concentrations of total $\mathsf{P M}_{2.5}$ were 43, 55, 56, 54, and $56\,\upmu\mathrm{g}/\mathrm{m}^{3}$ , respectively. The annual average concentration of $\mathsf{P M}_{2.5}$ still exceeded $35\,\upmu\mathrm{g}/\mathrm{m}^{3}$ at QX, an eastern mountain area, even though the pollution emissions of local sources were negligible. The results show that the background concentration of fine particulate matter was high in Dongguan and was influenced significantly by regional transportation of pollutants. The three suburban sites with more industry, MC, CA and SSL, and the downtown site, NC, had similar annual average $\mathsf{P M}_{2.5}$ pollution levels. The annual average concentration of $\mathsf{P M}_{2.5}$ at the four sites was about 0.3 times higher than that at QX, implying that this extra $30\%\ \mathrm{PM}_{2.5}$ concentration was tightly related to local emissions in the downtown and industrial areas.
3.2. Source apportionment of $P M_{2.5}$ based on PMF model
In this study, the PMF model was used to apportion the sources of $\mathsf{P M}_{2.5}$ . Three to twelve factors in PMF model were tested, and the eight-factor output by this model was the most reasonable solution for explaining all identified source profiles in Dongguan, as profiled in Fig. 4. The solutions with factors more than eight produced unexplained factors. Therefore, the eight-factor solution (base run, $\mathrm{Q}_{\mathrm{true}}/\mathrm{Q}_{\mathrm{exp}}=1.06)$ was chosen as the optimal solution for this study, and the seasonal variation and spatial distribution of each source are shown in Fig. $5({\mathsf{a}})$ and (b), respectively. The characteristics of each factor identified by PMF are discussed below, and we will then try to estimate the amount of secondary organic aerosol (SOA) based on the factor profiles resolved by PMF.
3.2.1. Fugitive dust
The first factor had the highest characteristic value (the species contribution $(\%)$ in Fig. 4) of Al and Mg. Al is a tracer of soil dust; therefore, this factor is closely related to a fugitive dust source. The high concentrations of sulfate and ammonium associated with this factor could be attributed to the heterogeneous reactions of $S0_{2}$ on alkaline particles (Cheng et al., 2016). The contribution of this source to $\mathsf{P M}_{2.5}$ showed seasonal variation: higher in fall and winter, and lower in spring and summer, which was mostly a result of high rainfall and RH in these two seasons (in Table 1). The spatial distribution of this source showed a significant difference. The contribution of fugitive dust was the lowest at the downtown area because more pavement was present. The results show that this source is strongly characteristic of a local source.
3.2.2. Secondary sulfate
The second factor had significant characteristic values of $S0_{4}^{2-}$ and $\mathsf{N H4}$ . $\mathrm{SO}_{4}^{2-}$ in $\mathsf{P M}_{2.5}$ is produced from a secondary reaction process of $S0_{2}$ . VOCs in the atmosphere were simultaneously oxidized to OM during the secondary reaction process of $S0_{2}$ ; therefore, OM in this factor was also identified as secondary OM. The seasonal variation showed that contributions of this source were higher in fall and winter and lower in spring and summer, which was mainly a result of regional transport from the polluted northern continent in fall and winter (Hagler et al., 2006). The spatial distributions at the five sites were almost consistent. These results reveal that secondary sulfate is a typical regional transportation source in Dongguan. Moreover, the regional characteristics of this source were also identified in neighboring Shenzhen (Huang et al., 2014).
3.2.3. Coal burning
The third factor had the highest characteristic value of Cl⁻. Cl⁻ in $\mathsf{P M}_{2.5}$ originates mainly from ammonium chloride, which is formed by the rapid combination of HCl and ammonia in the atmosphere. In China, a country with huge coal-consumption, the biggest potential source of HCl is coal burning (Yudovich and Ketris, 2006; Wang et al., 2015). This factor also had higher mass contributions from $\mathrm{SO}_{4}^{2.}$ , $\mathsf{N O}_{3}^{\phantom{\,}-}$ , and EC, further confirming that it was a combustion-related source. Considering there are many coalcombustion power plants around Dongguan, this factor is recognized as a coal-burning source. The contribution of this source to $\mathsf{P M}_{2.5}$ was highest in winter, lowest in summer, and almost equal in both spring and fall. Contributions of this source at the five sites presented obvious differences, with higher contributions in the downtown and industrial areas being 1.6 times higher than the lowest one at the background site, signifying that the coal burning contribution is affected significantly by the nearby power plants.
3.2.4. Biomass combustion For the fourth factor, K, Zn, and Pb had prominent characteristic values. K in $\mathsf{P M}_{2.5}$ is a tracer for biomass combustion (Yamasoe et al., 2000), and OM had the biggest mass contribution to this factor. Therefore, this source was interpreted as biomass combustion. Open-air biomass burning often mixes with garbage burning in Chinese rural areas. The smoke of garbage-burning emission might contain Zn and Pb. Huang et al. (2014) and Yuan et al. (2006a) apportioned K and $\boldsymbol{\mathrm{Pb}}$ to the same factor in Shenzhen and Hong Kong, which was interpreted as regional biomass- and garbageburning sources. The contributions of this source were obviously higher in fall and winter and lower in spring and summer. The spatial variations in the source contribution were negligible at the five sites. These results indicate that the biomass combustion source has an apparent characteristic of regional transportation.
3.2.5. Ship emissions
The fifth factor had the highest characteristic value of V, which is a trace element of heavy oil-burning emission (Chow and Watson, 2002). This type of emission is closely related to ships because only ocean-going vessels generally use heavy fuel oil in the PRD region. Huang et al. (2014) used the PMF model to identify major sources of $\mathsf{P M}_{2.5}$ in Shenzhen, and they identified heavy oil-burning emission with a characteristic value of V. The activities of ships in the PRD occur mainly in southern seas. The predominant wind direction in Dongguan was south in spring and summer and north in fall and winter (Fig. 2). Owing to the monsoon, the contribution of the ship emissions source was higher in spring and summer and lower in fall and winter. The contributions of this source to the five sites showed little difference in spatial distribution, indicating that this source had a certain regional origin from the sea. However, MC and CA showed relatively higher contributions from ship emissions because they are close to ports.
3.2.6. Industrial emissions
Ca and EC showed obvious characteristic values in this factor, implying a likely source of smoke dust from inferior coal and diesel oil combustion (Hopke et al., 1976; Lewis and Macias, 1980; Wu et al., 2013), which are often used in middle-small factories in China. The spatial distribution of this factor showed significant variation, with the highest contribution at MC, an intensive industrial area, and the lowest contribution at NC, a downtown area, indicating this factor was local industry-related rather than life activity-related. These characteristics correlate with the emissions of many industrial boilers and furnaces in Dongguan; therefore, this factor was determined as an industrial emissions source. The seasonal variations in this source were lower in summer, while similarly higher in the other three seasons. This seasonal variation pattern was not only a result of seasonal meteorological conditions, but also a result the complex industrial activity variations during the year.
3.2.7. Secondary nitrate
For the seventh factor, $\bar{\mathsf{N O}_{3}}$ had the highest characteristic value, indicating that this source originated from a secondary reaction process of $\mathsf{N O}_{\mathtt{X}}$ . OM in this factor was identified as secondary OM, which was produced from the secondary reaction process of volatile organic compounds (VOCs) that occurs during the process of $\Nu0_{\mathrm{x}}$ oxidation. The contribution of this source was prominent in winter but particularly low in summer. This was partly due to high ambient temperature in summer, which would lead semi-volatile $\mathrm{NH}_{4}\mathrm{NO}_{3}$ to evaporate (Huang et al., 2014). In winter, the air mass from the northern mainland brought more nitrates in $\mathsf{P M}_{2.5}$ to Dongguan; thus, the contribution of secondary nitrate was even higher in winter. The spatial distribution of secondary nitrate at the five sites showed that all suburban areas with more industry, MC, SSL and CA, and the downtown area, NC, had a higher contribution of over two times that at the background site, QX. These results show that the secondary nitrate was generated more in local areas than from regional transport in terms of a whole year.
3.2.8. Vehicle emissions
The eighth factor had the highest characteristic values of OM and EC, signifying a dominant combustion source. Compared with other factors, this factor had weaker seasonal and spatial variations, corresponding to ubiquity of it in the atmosphere near the ground. Therefore, this factor was identified as a vehicle emissions source. The relatively lower contribution in spring could be a result of a relatively higher wind speed in this season (in Table 1), and the contribution was relatively lower at the background site.
In the above PMF modeling output, it should be noted that an important $\mathsf{P M}_{2.5}$ origin, i.e., secondary organic aerosol (SOA), did not appear, even if in the PMF output with nine or more factors. Since the types of OM are highly complex and contain thousands of compounds, the source apportionment of OM remains a hard scientific problem and secondary organic aerosol (SOA) cannot be easily quantified by traditional application of CMB and PMF models on $\mathsf{P M}_{2.5}$ source apportionment. In the PMF model results, OM was usually apportioned into various factors such as primary characteristic factors like vehicle emissions and biomass combustion and secondary characteristic factors like secondary sulfate and secondary nitrate. Yuan et al. (2006b) proposed a valid method to estimate the contribution of SOA by extracting the OM in the secondary characteristic factors. This method is proved to be also valid in this study and thus adopted since there are significant organic matter contributions in Factor II secondary sulfate and Factor VII secondary nitrate in Fig. 4. The PMF analysis for organic aerosol measured by on-line aerosol mass spectrometer (AMS) also revealed that SOA in this region could be divided into two fractions: one was semi-volatile oxygenated organic aerosol (SV-OOA) correlated well to nitrate; while the other was low-volatile oxygenated organic aerosol (LV-OOA) correlated well to sulfate (He et al., 2011). Therefore, for each individual $\mathsf{P M}_{2.5}$ sample, its SOA mass contribution can be estimated by Eq. (3):
$$
\mathsf{O M}_{\mathsf{S O A}}=\mathsf{O M}_{\mathsf{s u l}}+\mathsf{O M}_{\mathrm{nit}}
$$
where $0\mathtt{M}_{\mathtt{S O A}}$ is the SOA mass contribution for the sample; $\mathsf{O M}_{\mathrm{sul}}$ is the OM contribution contained in the factor of secondary sulfate resolved by PMF; $\mathsf{O M}_{\mathrm{nit}}$ is the OM contribution contained in the factor of secondary nitrate resolved by PMF. The results through this approach show that SOA accounted for $27\%$ of the total OM. The contributions of SOA were still higher in fall and winter and lower in spring and summer. The spatial distribution at the five sites showed a small difference, with slightly lower values noted only at the background site, QX. Thus, SOA had a regional pollution characteristic, which was also indicated by He et al. (2011).
Fig. 6(a) shows the annual average $\mathsf{P M}_{2.5}$ source structure in Dongguan in 2014 based on the above results of source apportionment using the PMF model. Vehicle emissions, secondary sulfate, biomass combustion, and SOA were the main sources of contributions to $\mathsf{P M}_{2.5}$ in Dongguan, accounting for $21\%$ , $20\%$ , $11\%$ and $10\%$ , respectively. The total contribution of these four major sources exceeded $60\%$ . The other sources, ship emissions, fugitive dust, secondary nitrate, industrial emissions, and coal burning, each contributed $5\%{-}8\%$ to the total $\mathsf{P M}_{2.5}.$ , In addition, because the undetected masses were not input into the PMF model, the unknown sources accounted for $4\%$ of the total $\mathsf{P M}_{2.5}$ . Fig. 6(b) explains the corresponding relationship between the model-input total mass of the fifteen species and the model-reconstructed total mass of all the factors. The correlation was very high $(\mathsf{R}^{2}=0.98$ slope $=1.02$ ), which indicates highly satisfactory results of model fitting. The eight factors of the model output can also effectively simulate the main sources of $\mathsf{P M}_{2.5}$ in Dongguan.
3.3. Sources characteristics of $P M_{2.5}$ on different pollution levels
In this study, $\mathsf{P M}_{2.5}$ samples were simultaneously collected at five sites for totally 66 days in 2014. The average concentration of $\mathsf{P M}_{2.5}$ at the five sites on the same day was regarded as the daily average concentration of $\mathsf{P M}_{2.5}$ in Dongguan. Therefore, based on the daily average concentration limit of Grade II national standards for air quality $(75\ \upmu\mathrm{g}/\mathrm{m}^{3})$ , the sampling days in Dongguan were divided into two types: exceeding standard days (ESD, $\mathsf{P M}_{2.5}$ greater than $75~\upmu\mathrm{g}/\mathrm{m}^{3})$ , and non-exceeding standard days (NSD, $\mathsf{P M}_{2.5}$ less than or equal to $75\ \upmu\mathrm{g}/\mathrm{m}^{3})$ . According to this classification of sampling days, ESD appeared only in fall and winter during the entire sampling period, although NSD appeared in every season. Therefore, the special source structure of $\mathsf{P M}_{2.5}$ in ESD could be understood by analysis of the source structure of $\mathsf{P M}_{2.5}$ in ESD and NSD both in fall and winter (Fig. 7). Compared with those in NSD, secondary nitrate, SOA, and biomass combustion increased the mostly in ESD, each by 3.4, 1.6, and 1.3 times, respectively. Other sources with an increasing trend in ESD included fugitive dust (0.9), secondary sulfate (0.8) and coal burning (0.5). On the basis of the growth ratio of each source, the precursors of secondary nitrateeNOx, the precursors of SOAeVOCs, and the primary biomass burning should be given more attention to control the ESD pollution levels. Considering the dominant regional sources of SOA and biomass burning, as discussed before, the control measures should be implemented on a regional rather than local scale.
4. Conclusions
In this study, $\mathsf{P M}_{2.5}$ was analyzed in four seasons at five sites in Dongguan, a typical industrial city in southern China, and the major conclusions are presented in the following points.
(1) The annual average concentration of $\mathsf{P M}_{2.5}$ in Dongguan was $53\ \upmu\mathrm{g}/\mathrm{m}^{3}$ in 2014. OM, $\mathsf{S}0_{4}^{2.}$ , $\mathsf{N H4}^{+}$ , EC, $\mathrm{NO}_{3}{}^{-}$ , Cl⁻, and metallic elements contributed $38\%$ , $23\%$ , $11\%$ , $8\%$ , $8\%$ , $1\%$ , and $6\%$ of the $\mathsf{P M}_{2.5}$ total mass, respectively. The seasonal variation of $\mathsf{P M}_{2.5}$ was generally higher in fall and winter and lower in spring and summer. The spatial distribution of $\mathsf{P M}_{2.5}$ at the five sites was generally higher in the west and lower in the east.
(2) According to the results of source analysis, Vehicle emissions, secondary sulfate, biomass combustion, and SOA were the main sources of contributions to $\mathsf{P M}_{2.5}$ in Dongguan, accounting for $21\%$ , $20\%$ , $11\%$ , and $10\%$ , respectively. Ship emissions, fugitive dust, secondary nitrate, industrial emissions, and coal burning each contributed $5\%{-}8\%$ to the total $\mathsf{P M}_{2.5}.$ . The temporalespatial variation in the different sources indicated that secondary sulfate, biomass combustion, SOA, and ship emissions have obvious regional characteristics, and vehicle emissions, secondary nitrate, coal burning, fugitive dust, and industrial emissions have obvious local characteristics.
(3) Secondary nitrate, SOA, and biomass burning were found to increase in concentration the mostly on ESD days in fall and winter. Other sources with an increasing trend in ESD included fugitive dust (0.9), secondary sulfate (0.8) and coal burning (0.5). Therefore, to decrease the high pollution levels, more attention should be given to secondary nitrate, SOA, and biomass burning.
Acknowledgements
This work was supported by the National Natural Science Foundation of China (U1301234), the Dongguan Environmental Monitoring Central Station, and the Ministry of Science and Technology of China (2014BAC21B03).
References
Charlson, R.J., Schwartz, S.E., Hales, J.M., Cess, R.D., Coakley, J.A., Hansen, J.E., Hofmann, D.J., 1992. Climate forcing by anthropogenic aerosols. Science 255 (5043), 423e430.
Cheng, Y., Zheng, G., Wei, C., Mu, Q., Zheng, B., Wang, Z., Gao, M., Zhang, Q., He, K., Carmichael, G., Po schl, U., Su, H., 2016. Reactive nitrogen chemistry in aerosol water as a source of sulfate during haze events in China. Sci. Adv. 2 (12), e1601530.
Chow, J.C., Watson, J.G., 2002. Review of $\mathrm{PM}_{2.5}$ and $\mathrm{PM_{10}}$ Apportionment for fossil fuel combustion and other sources by the chemical mass balance receptor model. Energy & Fuels 16, 222e260.
Donkelaar, A.V., Randall, V.M., Brauer, M., Kahn, R., Levy, R., Verduzco, C., Villeneuve, P.J., 2010. Global estimates of ambient fine particulate matter concentrations from satellite-based aerosol optical depth: development and application. Environ. Health Perspect. 118 (6), 847e855.
Gao, B., Guo, H., Wang, X.M., Zhao, X.Y., Ling, Z., Zhang, Z.H., Liu, T.Y., 2013. Tracerbased source apportionment of polycyclic aromatic hydrocarbons in $\mathrm{PM}_{2.5}$ in Guangzhou, southern China, using positive matrix factorization (PMF). Environ. Sci. Pollut. Res. 20 (4), 2398e2409.
Gong, Z.H., Lan, Z.J., Xue, L., Zeng, L.W., He, L.Y., Huang, X.F., 2012. Characterization of submicron aerosols in the urban outflow of the central Pearl River Delta region of China. Front. Environ. Sci. Eng. 6 (5), 725e733.
Hagler, G.S.W., Bergin, M.H., Salmon, L.G., Yu, J.Z., Wan, E.C.H., Zheng, M., Zeng, L.M., Kiang, C.S., Zhang, Y.H., Lau, A.K.H., Schauer, J.J., 2006. Source areas and chemical composition of fine particulate matter in the Pearl River Delta region of China. Atmos. Environ. 40 (20), 3802e3815.
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 (12), 4461e4466.
He, L.Y., Huang, X.F., Xue, L., Hu, M., Lin, Y., Zheng, J., Zhang, R.Y., Zhang, Y.H., 2011. Submicron aerosol analysis and organic source apportionment in an urban atmosphere in Pearl River Delta of China using high-resolution aerosol mass spectrometry. J. Geophys. Res. 116 (D12304).
Hopke, Philip K., Gladney, Ernest S., Gordon, Glen E., Zoller, William H., Jones, Alun G., 1976. The use of multivariate analysis to identify sources of selected elements in the Boston urban aerosol. Atmos. Environ. (1967) 10 (11), 1015e1025.
Huang, X.F., Yun, H., Gong, Z.H., Li, X., He, L.Y., Zhang, Y.H., Hu, M., 2014. Source apportionment and secondary organic aerosol estimation of $\mathrm{PM}_{2.5}$ in an urban atmosphere in China. Sci. China Earth Sci. 57 (6), 1352e1362.
Karagulian, Federico, Belis, Claudio A., Dora, Carlos Francisco C., PrüssUstün, Annette M., Bonjour, Sophie, Adair-Rohani, Heather, Amann, Markus, 2015. Contributions to cities' ambient particulate matter (PM): a systematic review of local source contributions at global level. Atmos. Environ. 120, 475e483.
Kuang, B.Y., Lin, P., Huang, X.H.H., Yu, J.Z., 2015. Sources of humic-like substances in the Pearl River Delta, China: positive matrix factorization analysis of $\mathrm{PM}_{2.5}$ major components and source markers. Atmos. Chem. Phys. 15 (4), 1995e2008.
Lewis, Charles W., Macias, Edward S., 1980. Composition of size-fractionated aerosol in Charleston, West Virginia. Atmos. Environ. (1967) 14 (2), 185e194.
Lim, S.S., Vos, T., Flaxman, A.D., Danaei, G., Shibuya, K., Adair-Rohani, H., Amann, M., et al., 2012. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 380 (9859), 2224e2260.
Norris, G., Duvall, R., September 2014. EPA Positive Matrix Factorization (PMF) 5.0 Fundamentals and User Guide.
Paatero, Pentti, 1997. Least squares formulation of robust non-negative factor analysis. Chemom. Intell. Lab. Syst. 37 (1), 23e35.
Paatero, Pentti, Tapper, Unto, 1994. Positive matrix factorization: a non-negative factor model with optimal utilization of error estimates of data values. Environmetrics 5 (2), 111e126.
Sarnat, J.A., Marmur, A., Klein, M., Kim, E., Russell, A.G., Sarnat, S.E., Mulholland, J.A., Hopke, P.K., Tolbert, P.E., 2008. Fine particle sources and cardiorespiratory morbidity: an application of chemical mass balance and factor analytical source-apportionment methods. Environ. Health Perspect. 116 (4), 459e466.
Tan, J.H., Duan, J.C., Ma, Y.L., He, K.B., Cheng, Y., Deng, S.X., Huang, Y.L., Si-Tu, S.P., 2016. Long-term trends of chemical characteristics and sources of fine particle in Foshan City, Pearl River Delta: 2008e2014. Sci. Total Environ. 565, 519e528.
Viana, M., Kuhlbusch, T.A.J., Querol, X., Alastuey, A., Harrison, R.M., Hopke, P.K., Winiwarter, W., Vallius, M., Szidat, S., Pr evo^t, A.S.H., Hueglin, C., Bloemen, H., Wåhlin, P., Vecchi, R., Miranda, A.I., Kasper-Giebl, A., Maenhaut, W., Hitzenberger, R., 2008. Source apportionment of particulate matter in Europe: a review of methods and results. J. Aerosol Sci. 39 (10), 827e849.
Wang, J.Z., Ho, S.S.H., Ma, S.X., Cao, J.J., Dai, W.T., Liu, S.X., Shen, Z.X., Huang, R.J., Wang, G.H., Han, Y.M., 2016. Characterization of $\mathrm{PM}_{2.5}$ in Guangzhou, China: uses of organic markers for supporting source apportionment. Sci. Total Environ. 550, 961e971.
Wang, Q.Q., Huang, X.H.H., Zhang, T., Zhang, Q., Feng, Y., Yuan, Z., Wu, D., Lau, A.K.H., Yu, J.Z., 2015. Organic tracer-based source analysis of $\mathrm{PM}_{2.5}$ organic and elemental carbon: a case study at Dongguan in the Pearl River Delta, China. Atmos. Environ. 118, 164e175.
Watson, John G., Antony Chen, L.-W., Chow, Judith C., Doraiswamy, Prakash, Lowenthal, Douglas H., 2008. Source apportionment: findings from the U.S. Supersites program. J. Air & Waste Manag. Assoc. 58 (2), 265e288.
Wu, H., Zhang, C.Y., Wang, J., et al., 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), 583e589.
Yamasoe, Ma rcia A., Artaxo, Paulo, Miguel, Antonio H., Allen, Andrew G., 2000. Chemical composition of aerosol particles from direct emissions of vegetation fires in the Amazon Basin: water-soluble species and trace elements. Atmos. Environ. 34 (10), 1641e1653.
Yuan, Z.B., Lau, A.K.H., Zhang, H.Y., Yu, J.Z., Louie, P.K.K., Fung, J.C.H., 2006a. Identification and spatiotemporal variations of dominant PM10 sources over Hong Kong. Atmos. Environ. 40 (10), 1803e1815.
Yuan, Z.B., Yu, J.Z., Lau, A.K.H., Louie, P.K.K., Fung, J.C.H., 2006b. Application of positive matrix factorization in estimating aerosol secondary organic carbon in Hong Kong and its relationship with secondary sulfate. Atmos. Chem. Phys. 6 (1), 25e34.
Yudovich, Y.E., Ketris, M.P., 2006. Chlorine in coal: a review. Int. J. Coal Geol. 67, 127e144.
Zhang, H.L., Wang, Y.G., Hu, J.L., Ying, Q., Hu, X.M., 2015. Relationships between meteorological parameters and criteria air pollutants in three megacities in China. Environ. Res. 140, 242e254.
Zhou, J.B., Xing, Z.Y., Deng, J.J., Du, K., 2016. Characterizing and sourcing ambient $\mathrm{PM}_{2.5}$ over key emission regions in China I: water-soluble ions and carbonaceous fractions. Atmos. Environ. 135, 20e30.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.