Science of the Total Environment 493 (2014) 262–270
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Source of atmospheric heavy metals in winter in Foshan, China Ji-Hua Tan a,c, Jing-Chun Duan b,⁎, Yong-Liang Ma c, Fu-Mo Yang a, Yuan Cheng c, Ke-Bin He c, Yong-Chang Yu d, Jie-Wen Wang d a
College of Resource and Environment, University of Chinese Academy of Sciences, Beijing 100049, China State Key Laboratory of Environmental Criteria and Risk Assessment, Chinese Research Academy of Environmental Sciences, Beijing 100012, China State Key Joint Laboratory of Environmental Simulation and Pollution Control, School of Environment, Tsinghua University, Beijing 100084, China d Foshan Environmental Protection Bureau, Foshan 528000, China b c
H I G H L I G H T S • • • • •
Foshan city is the most air polluted city in Pearl rivel Delta. Atmospheric Zn, Pb and As were significantly high in this area. Two special sources were identified during the sampling period. Ceramic industry is the most important source of atmospheric heavy metal. EFs analysis indicated that local emissions were the main sources.
a r t i c l e
i n f o
Article history: Received 19 February 2014 Received in revised form 9 May 2014 Accepted 31 May 2014 Available online xxxx Editor: Xuexi Tie Keywords: Chemical composition PM2.5 PMF Foshan Control policy
a b s t r a c t Foshan is a ceramics manufacturing center in the world and the most polluted city in the Pearl River Delta (PRD) in southern China measured by the levels of atmospheric heavy metals. PM2.5 samples were collected in Foshan in winter 2008. Among the 22 elements and ions analyzed, 7 heavy metals (Zn, V, Mn, Cu, As, Cd and Pb) were studied in depth for their levels, spatiotemporal variations and sources. The ambient concentrations of the heavy metals were much higher than the reported average concentrations in China. The levels of Pb (675.7 ± 378.5 ng/m3), As (76.6 ± 49.1 ng/m3) and Cd (42.6 ± 45.2 ng/m3) exceeded the reference values of NAAQS (GB3095-2012) and the health guidelines of the World Health Organization. Generally, the levels of atmospheric heavy metals showed spatial distribution as: downtown site (CC, Chancheng District) N urban sites (NH and SD, Nanhai and Shunde Districts) N rural site (SS, Shanshui District). Two sources of heavy metals, the ceramic and aluminum industries, were identified during the sampling period. The large number of ceramic manufactures was responsible for the high levels of atmospheric Zn, Pb and As in Chancheng District. Transport from an aluminum industry park under light north-west winds contributed high levels of Cd to the SS site (Shanshui District). The average concentration of Cd under north-west wind was 220 ng/m3, 20.5times higher than those under other wind directions. The high daily maximum enrichment factors (EFs) of Cd, Pb, Zn, As and Cu at all four sites indicated extremely high contamination by local emissions. Back trajectory analysis showed that the heavy metals were also closely associated with the pathway of air mass. A positive matrix factorization (PMF) method was applied to determine the source apportionment of these heavy metals. Five factors (industry including the ceramic industry and coal combustion, vehicle emissions, dust, transportation and sea salt) were identified and industry was the most important source of atmospheric heavy metals. The present paper suggests a control policy on the four heavy metals Cd, Pb, Zn, and Cu, and suggests the inclusion of As in the ceramic industry emission standard in the future. © 2014 Elsevier B.V. All rights reserved.
1. Introduction The association of airborne particulate matter (PM) with adverse health effects has been recognized for a long time (Fang et al., 2005, ⁎ Corresponding author. E-mail address: [email protected] (J.-C. Duan).
http://dx.doi.org/10.1016/j.scitotenv.2014.05.147 0048-9697/© 2014 Elsevier B.V. All rights reserved.
2010; Gao et al., 2005; Yang et al., 2011; Franck et al., 2011; Steinle et al., 2013). Heavy metals are part of PM and pose serious risks to human health, resulting in a variety of human dysfunction (PineiroIglesias et al., 2003) and various diseases. Atmospheric heavy metals are removed by wet and dry deposition which can then cause negative effects on the biogeochemical cycling (Wong et al., 2003). Natural emissions, traffic and industrial emissions are the principal sources of heavy
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270
263
2. Materials and methods
2th to 31th Dec 2008 at three urban sites (CC, NH and SD) and a rural site (SS) (Fig. 1). The CC site is situated in Chancheng district, the metropolitan center of Foshan city, which is surrounded by main traffic roads, residential buildings and factories. Most of the ceramic factories in Foshan city area round this site. The NH site is located in Nanhai district, which is only 2 km from the neighboring Guangzhou city and is surrounded by main traffic roads and residential buildings. The SD site is located in Shunde district, which is China's leading home appliance production base. Finally, the SS site is located in Shanshui forest park in Shanshui district with fewer nearby anthropogenic sources. The sampler trapped PM2.5 on quartz filters (Whatman QM-A, 20.3 × 25.4 cm) for 24 h and a total of 112 samples were collected. All quartz filters were annealed at 500 °C for 4 h to remove trace organics before usage. Before and after collection, filters were wrapped in baked aluminum foil. Prior to weighing, the filters were stored under a constant temperature and humidity condition of 25 °C and 50% for 24 hours. After sampling, each filter was wrapped in aluminum foil, then placed in an airtight bag, and taken to the laboratory at the local Environmental Protection Bureau (EPB) for storage at − 30 °C. Upon the completion of sampling periods, the samples were transported in an ice box to the Key Joint Laboratory of Environment Simulation and Pollution Control at the School of Environment, Tsinghua University, for analysis. The lag period between sampling and analysis was about 2 months. Two types of blank samples were prepared: exposure blanks and filter blanks. The exposure blank filters were placed in the sampler for 24 h but had no air was drawn through them; hence, they were subjected to the same handling procedures as the actual samples. The filter blank samples were used to check for the contamination of the filters without going through the sampling procedures. On average, an exposure blank and a filter blank were taken for every 10 actual samples. Meteorology data (Fig. 2) were recorded at the CC site by the equipment of “Davis Complete Weather Wizard III Cable Basic Weather Monitor” which provided ambient temperature, humidity, pressure, wind speed, wind direction at the CC site every 15 min during the sampling days. Visibility was measured by Belfort Model 6000 Visibility Sensor (Belfort Instrument, USA).
2.1. Sample collection
2.2. Sample analysis
PM2.5 samples were collected simultaneously by high volume (1.13 m 3 /min) samplers (Thermo fisher instruments, USA) from
A small portion of each filter was cut using stainless steel scissors and placed in Teflon tubes with a 4 ml mixture of concentrated high-
metals in the ambient air (Lee and Hieu, 2011; Park et al., 2008). Studies showed that most heavy metals were enriched in PM2.5 (Duan et al., 2012b; Lu et al., 2012). For example, the ratios of heavy metals in PM2.5 to those in PM10 in Beijing, China ranked as Pb (88.5%) N Cd (81.8%) N Zn (81.5%) N As (77.1%) N Cu (75.9%) N Cr (71.7%) N Ni (67.9%) N Mn (63.3%) N V (46.8%) (Duan and Tan, 2013). Foshan is one of the most important manufacturing bases not only for China, but also for the world (Guo et al., 2011). The city is characterized by a high density of ceramics plants, accounting for around 30% world production of ceramics and consuming about 40% coal in Foshan in 2008 (Shen and Wei, 2012). With the substantial economic development, large amounts of pollutants were emitted into the atmosphere, which led to a rapid deterioration of air quality in and around this region. According to the Pearl River Delta Regional Air Quality Monitoring Network, the average concentrations of SO2, NOx and PM10 in Foshan city were significantly higher than those in other areas of Pearl River Delta (PRD). Many studies (Nie et al., 2010; Chen et al., 2012; He et al., 2009) showed that heavy metal pollution was a serious problem in Foshan. Heavy metal (Cd, Pb, Zn, and Cu) level invegetable soil in Foshan were significantly higher than those in the other areas throughout Guangdong Province and the whole country (Nie et al., 2010). Child blood lead levels (BLLs) in Foshan city were much higher than those in the developed countries (Chen et al., 2012; He et al., 2009). Several studies have been conducted on atmospheric heavy metals in the PRD region (Li et al., 2003; Wang et al., 2006a); however, no literature on characteristics and sources of atmospheric heavy metals has been investigated in Foshan. This study was supported by the Foshan City Government to investigate the sources of serious pollution in winter in Foshan, and to put forward suggestions on control policies. Among the 22 elements and ions measured during the study, 7 heavy metals (Zn, V, Mn, Cu, As, Cd and Pb) were studied in-depth for their levels, spatiotemporal variations and sources. The objectives of the present study are to: (1) investigate the levels and their sources of atmospheric heavy metals (Pb, As, Zn and Cd) in Foshan; (2) put forward emission control policy suggestions for atmospheric heavy metals in Foshan.
SS
NH CC
SD
Pearl
River
Delta Fig. 1. Map of sampling sites.
Sampling site
264
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270
Fig. 2. The meteorological conditions and heavy metals during the sampling period.
purity hydrochloric acid and nitric acid with a volume ratio of 3:1. Blank filter and sample replicates were randomly inserted for quality control. The digested solution was diluted to 10 ml using ultra-pure water to perform the metal analysis by inductively coupled plasmamass spectrometry (ICP-MS) (Thermo, X serial). The calibration was made using multi-element (metal) standards (certified reference materials (CRMs)) in a 3% (v/v) HNO3 solution. Since quartz filter is not appropriate for some elemental analysis, such as Al and Ca were not included in the analysis, only elements with low uncertainties were included in this paper. Six blank filters were treated and analyzed in the same way as for the actual samples. The detection limits were Ti (3.16 ppb), Zn (10 ppb), V (0.5 ppb), Mn (1.50 ppb), Cu (1.47 ppb), As (0.38 ppb), Rb (0.33 ppb), Cd (0.03 ppb), Cs (0.5 ppb), Ba (1.75 ppb), Tl (0.006 ppb), Pb (5.88 ppb), Fe (11.32 ppb) and Bi (0.33 ppb). Ion chromatography (Dionex 1400) was used to measure inorganic ions. The concentrations of water soluble inorganic ions in field blanks were 0.07 μg/m3, 0.08 μg/m3, not detected (nd), 0.16 μg/m3, nd, 2− 0.12 μg/m3, nd, 0.05 μg/m3 and 0.08 μg/m3 for F−, Cl−, NO− 3 , SO4 , + + + 2+ 2+ NH4 , Na , K , Ca and Mg , respectively. The relative standard deviation of each ion was less than 6% for the reproducibility test. The detailed analysis procedure of the heavy metal and water soluble inorganic species could be found elsewhere (Duan et al., 2012a; Tan et al., 2009). 2.3. PMF analysis The PMF model was developed by Paatero and Tapper (1994) and Paatero (2004). EPA PMF is one of the receptor models that the US EPA's Office of Research and Development (ORD) have developed. The algorithms used in EPA PMF model to compute profiles and contributions have been peer reviewed by leading scientists in the air quality management community and have been certified scientifically robust. In this study, 13 elements and 9 ions were included in the EPA PMF 3.0 model. The input data including concentrations of chemical species and equation-based uncertainties. The equation-based uncertainty
includes detection limits and error fractions (5%). If the concentration is less than or equal to the method detection limit (MDL) provided, the uncertainty (Unc) is calculated using the following equation (Polissar et al., 1998). Unc ¼
5 MDL 6
ð1 1Þ
If the concentration is greater than the MDL provided, the calculation is Unc ¼
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ðErrorFraction concentrationÞ2 þ ðMDLÞ2
ð1 2Þ
2.4. Back trajectory calculation In order to better understand the transport of airborne particles from distant sources, the HYSPLIT model (HYbrid Single-Particle Lagrangian Integrated Trajectory, Version 4.8, windows-based version), a comprehensive modeling system developed by the National Oceanic and Atmospheric Administration (NOAA) Air Resource Laboratory (Draxler and Rolph, 2003), was used to generate 72-h back trajectories every 6 h from 12/1 to 12/31 at the CC site. The relative ground height was 300 m and every trajectory was colored with the concentration of PM2.5 by Igorpro 6.2 (trajectories colored with atmospheric heavy metals were not presented here for their similarity with PM2.5). Because the synoptic atmospheric conditions at the four sampling sites were quite similar, only the back trajectory arriving at the CC site was calculated in this study. 3. Results and discussion 3.1. Concentration and spatiotemporal variations of heavy metals As shown in Table 1 and Fig. 2, the averaged concentration of PM2.5 was 136.4 μg/m3 during sampling period at the four sites in Foshan City,
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270
265
Table 1 The concentration of chemical composition at four sites in Foshan city. Species
Average
PM2.5 (μg/m3) Zn (ng/m3) Pb (ng/m3) V (ng/m3) Mn (ng/m3) Cu (ng/m3) As (ng/m3) Cd (ng/m3)
136.4 2214 675.7 40 200.6 283.8 76.6 42.6
± ± ± ± ± ± ± ±
CC 40.5 1438 378.5 29.1 129.4 174.7 49.1 45.2
154.4 3239 812.4 68.3 242.3 256.3 140.7 40.8
NH ± ± ± ± ± ± ± ±
47.8 1525 479.7 46.6 86.0 134.3 94.6 25.2
much higher than the National Ambient Air Quality Standard (NAAQS) of China for PM2.5 (75 μg/m3 daily and 35 μg/m3 annually GB30952012). It shows serious PM pollutions in winter in Foshan. The levels of PM2.5 in Foshan were comparable with those in Indian cities (100– 400 μg/m3); however, they were much higher than those reported in Western Europe and North America (10–55 μg/m3) (Sharma and Maloo, 2005). Table 1 showed the average concentrations of the heavy metals at the four sites. The time series of the heavy metals were illustrated in Fig. 2. The average concentration of atmospheric Pb concentration was 675.7 ± 378.5 ng/m3, which fall below the limits of the current national ambient air quality standard (NAAQS-1996) of 1000 ng/m3(Table 2); it however exceeds the new NAAQS (GB3095-2012) and the WHO guideline of 500 ng/m3. The averaged atmospheric V concentration was 40 ± 29.1 ng/m3, which was much lower than the guideline of WHO 1000 ng/m3. The averaged atmospheric As and Cd concentrations were 76.6 ± 49.1 ng/m3 and 42.6 ± 45.2 ng/m3, respectively, which were much higher than the reported average concentrations in China (Duan and Tan, 2013), and were much higher than the limits of the new NAAQS (GB3095-2012) and the WHO guideline. The averaged atmospheric concentrations of Mn, Cu and Zn were 200.6 ± 129.4 ng/m3, 283.8 ± 174.7 ng/m3 and 2214 ± 1438 ng/m3, respectively, but as yet there are no NAAQS for atmospheric Mn, Cu and Zn in China (GB3095-2012). The atmospheric concentrations of the heavy metals in Foshan in winter are ranked as: Zn N Pb N Cu N Mn N As N V N Cd, and all are above the national average air concentrations in China (Duan and Tan, 2013). The sum of concentrations of all atmospheric heavy metals in Foshan in winter ranked as: CC N NH N SD N SS. This ranking is consistent with the unit area energy consumption in Foshan (Shen and Wei, 2012). The highest average concentration of Pb, Zn, V and arsenic was observed at the CC site, Cd at the SS site, Mn and Cu at the NH site.
3.2. Source analysis of heavy metals 3.2.1. Enrichment factor of heavy metals Enrichment factor (EF) can be used to differentiate between the metals originating from human activities and those from natural procedure, and to assess the degree of anthropogenic influence. There is no widely accepted rule for the choice of the reference element; however, Si, Al and Fe are usually used for this purpose. In this study, Fe is used as the reference element and the abundance of elements in the upper
158.9 2157 679.1 32.1 244.3 429.5 51.2 36.0
SD ± ± ± ± ± ± ± ±
37.5 1069 251.4 21.6 80.3 178.4 34.7 10.9
118.8 1989 709.9 36.4 190.4 273.1 59.6 25.6
SS ± ± ± ± ± ± ± ±
29.6 983 296.3 20.0 130.6 122.3 23.7 9.8
107.4 1328 436.7 19.2 99.5 128.5 53.2 80.2
± ± ± ± ± ± ± ±
30.4 711.5 127.4 10.8 28.4 71.5 27.8 112.2
continental crust (UCC) is taken from a previous study (Taylor and McLennan, 1985). The EFCrust of element E in aerosols is defined as below (Wang et al., 2006b): EFCrust ¼ ½E=RAir = ½E=RCrust where R is a reference element of crustal material and [E/R]Air is the concentration ratio of E to R in aerosol, and [E/R]Crust is the concentration ratio of E to R in crust. The EFs of the heavy metals at the four sampling sites were calculated and displayed in Fig. 3. The EFs of Mn and V were lower than 10, confirming its mainly natural source. Enrichment factors of Cd, Pb, Zn, As and Cu were much higher than 10, indicating anthropogenic sources. EFs can also assist the determination of the degree of metal contamination. The mean EFs decreased in the order of Cd N Pb N Zn N As N Cu N V N Mn, suggesting that the decreasing order of their overall contamination degrees in Foshan. Cd, Pb, Zn, As and Cu had mean EFs higher than 40, suggesting extremely high contamination according to the five contamination categories on the basis of the enrichment factor (Loska and Wiechuya, 2003). Mn (CC and NH) and V (CC, NH and SD) had mean EFs higher than 5, indicating significant contamination; while Mn (SS and SD) and V (SS) with mean EFs between 2 and 5, were classified as moderately contaminated. The maximum EFs may reflect the degree to which local pollution affects each metal (Han et al., 2006). The highest daily maximum EFs of Cd were 8973, 18357, 3767 and 5605 at the CC, SS, NH and SD sites, respectively, indicating extremely high Cd contamination of local emission at the SS site. The highest daily maximum EFs of arsenic were 674.35, 141.81, 308.79 and 215.36, respectively, indicating extremely high Arsenic contamination of local emission at the CC site. 3.2.2. Trajectory analysis 3.2.2.1. Back trajectory analysis. Fig. 4 shows the back trajectories ending at Foshan city during the sampling period (2008/12/1–2008/12/31). In general, three categories of air masses could be identified: 1) NE—air masses coming from the northeast that originated from the continental inland areas and passing through Guangzhou city; 2) N—air masses from the north or northeast that originated from continental inland areas and passing through Qingyun city; and 3) E—air masses from the east that originated from continental inland areas of southeast China, reaching the PRD region through southeast China coast. Due to the Asian winter monsoon, the PRD was generally dominated by the
Table 2 Limits for heavy metals in China, WHO and other regions and countries.
NAAQS GB3095-1996 NAAQS GB3095-2012 WHO EU (DIRECTIVE 2004/107/EC) India
Hg
Pb
50 1000
1000 500 500 500
Concentrations corresponding to an excess lifetime risk of 1:100,000.
V
1000
As 6 6.6a 6 6
Mn
150
Ni
a
25 20 20
Cr (VI)
Cd
Year
0.025 0.25a
5 5 5
1996 2012 2000 2004 2009
266
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270
100,000 CC
SD
NH
SS
Average
10,000
Maximum 75th Percentile Median 25th Percentile
Minimum
1,000
100
10
1 Zn
Pb
Cu
Mn
As
V
Cd
Fig. 3. Boxplot of enrichment factors for metals at the four sampling sites in Foshan city.
air masses from the NE and N sector during the sampling period, and only occasionally from the E sector. The average concentrations of heavy metal in the three categories of air masses are shown in Table 3. The NE Category had the highest concentrations of heavy metals, including Zn, Pb, V, Mn, Cu, As and Cd, which were 4384 ng/m 3, 1502 ng/m3, 115 ng/m 3 , 313 ng/m3 , 363 ng/m3, 164 ng/m 3 and 59 ngm/m3 , respectively. The results showed that significant amounts of heavy metals were transported from the northeast inland areas, passing through Guangzhou city to Foshan city. On the contrary, the cleaner air mass from Central Asia contained the lower heavy metal during the N period. It should be noted that despite similar meteorological conditions at the four
sampling sites, most of the heavy metals at the CC site were much higher than those at the other three sites. In particular, the level of Cd was significantly higher during the episode, suggesting that local emissions play an important role in heavy metal pollution in these areas on top of the influence of the long-range transport of pollutants. 3.2.2.2. Local source of heavy metal. Though with fewer nearby anthropogenic sources, extremely high levels of Cd were found at the SS site. Fig. 5 shows the concentrations of PM2.5 and Cd during a pollution episode (12/15–12/21). On the 15th of December, the concentration of Cd (12.1 ng/m3) at the SS site was obviously lower than those at the other
Fig. 4. The back trajectories of air masses arriving at CC site from 12/1 to 12/31.
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270 Table 3 The mean concentrations of heavy metal at the CC site under different air masses. Air mass
n
NE
8
N
15
S
5
Mean S.D. Mean S.D. Mean S.D.
Zn
Pb
Cu
Mn
As
V
Cd
4384 2117 2967 1635 2476 1254
1502 486 917 455 892 344
115 44 49 34 45 26
313 61 208 80 210 74
363 156 219 108 196 101
164 52 101 46 155 144
59 31 34 23 31 16
three sites (34.4 ng/m3); however, it increased sharply then and exceeded the values of other sites significantly on December 17th. At the same time, the concentration of PM2.5 at the SS site was the lowest among the four sites. Based on the EFs analysis, the highest daily maximum EFs of Cd was found at the SS site on December 19th, indicating extremely high Cd contamination from local emissions. Further investigation found that high concentrations of Cd were always related to north-west winds, and low concentrations of Cd with other wind directions. However, as shown in Fig. 5, this emission source had negligible influences on the levels and EFs of Cd at the other three sites, indicating a localized emission source of Cd near the SS site. This localized emission is apparently linked to a large industrial park of smelting aluminum and other metals about 2.5 km to the north-west of the SS site. Previous studies have indicated that smelting aluminum industry is an important emission source of Cd (Tian et al., 2012), and the levels of Cd in vegetable soil in this area was significantly higher than the average value in China (Nie et al., 2010). The high soil Cd in this area may be related with the deposition of Cd. 3.2.3. PMF analysis In order to improve the model calculation, additional elements and ions were included with the heavy metals in the EPA PMF analysis: Ti, Zn, V, Mn, Cu, As, Rb, Cd, Cs, Ba, Tl, Pb, Bi, Na+, K +, Mg2 +,
267
2− − − Ca2 +, NH+ and NO− 4 , F , Cl , SO 4 3 . A critical step in PMF analysis is the determination of the number of factors. There are two principles to determine the numbers of factors: 1) the theoretical value of Q should be equal to the number of data points in the data set; 2) the source profiles also have to make physical sense. In this study, five factors are obtained by EPA PMF3.0 base on these two principles and identified as the following possible sources: industry (include ceramics industry and coal combustion), vehicle, dust, transportation and sea salt. Factor profiles (% of species and contributions) are illustrated in Fig. 6. (See Fig. 5.) Factor 1 has high contributions to F -(64.2%), As (60.4%), Cs (59.0%), Tl (56.7%), Cd (55.1%) and Pb (48.4%), and less than 20% to water soluble inorganic species. Atmospheric fluoride is recognized as a tracer of ceramics industry (Jordán et al., 2006) and was 1471 ng/m3 at the CC site; however, less than 400 ng/m3 at other sampling sites. High levels of Pb, Zn and As were emitted from coal combustion (Duzgoren-Aydin, 2007; Kang et al., 2011; Tian et al., 2010; Duan et al., 2012a,b) and are also commonly used as ceramic flux to lower the high melting point of silica. Peng et al. (2007) found that the concentrations of Pb and Zn are 2870 ng/m 3 and 6750 ng/m3in the flue gas of the ceramic in Foshan. Similar results were also observed in Spain (Minguillón et al., 2009). Foshan is an important manufacturing base in China, and coal is still the primary energy source in Foshan city (Shen and Wei, 2012) and production of ceramic in Foshan city contributes about 30% total world production. So ceramic industry and coal combustion were the most important industrial emission source considering the industry and energy structure in Foshan City. Factor 2 contributes a high percentage to K+(51.1%) and most of the 2− (62.2%) and NO− ions NH+ 4 (64.8%), SO4 3 (54.8%), and contributes less than 20% to main heavy metals, indicating that it's a mixture of aged 2− and NO− pollutants of long range transport. NH+ 4 , SO4 3 are indicators of secondary aerosol and long range transport. K+ is a good marker of
350
SS
SD
NH
CC
Concentration (µg/m3)
300
PM 2.5
250 200 150 100 50 0 350
Aluminum industrypark in Shihui city
Concentration (ng/m3)
300
Cd
Shanshui site
250 200 150 100 50 0 2008/12/15
2008/12/16
2008/12/17
2008/12/18
2008/12/19
2008/12/20
Fig. 5. PM2.5 and Cd at the four sampling sites during a pollution episode.
2008/12/21
268
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270
Fig. 6. Factor profiles (% of species and contributions) obtained from EPA PMF model.
biomass burning (Duan et al., 2004), which was often reported and has been identified as an important contributor to PM2.5 in the PRD (Hagler et al., 2006). Factor 3 contributes significantly to Cu (59.6%), Mg (56.5%), Mn (52.7%), Ca (56%) and Ba (48.2%), which are good indicators of crust related dust (Duan et al., 2012b). Factor 4 has high contributions to V (60.5%), Pb (26.47) and Zn (22.2%) and much less to other species (b 10%). V is a good indicator of combustion of petroleum. Combustion of fossil fuels is the main anthropogenic V emission accounting for 85% of total V emissions in China (Duan and Tan, 2013). Factor 5 contributes high percentages to Cl− (70.4%) and Na (57.8%), indicating sea salt impacts, but much less to most other heavy metals (b10%). Source contributions of As, Pb, Cd, Zn and Mn at the four sites calculated from EPA PMF3.0 model are shown in Fig. 7. It may be concluded that industries, including the ceramics industry and coal combustion, contributed about 73% to these five heavy metals and are the most important sources of heavy metal at the CC site. Industry (52%) and vehicle (17%) were the most important sources at the SS site. In the PMF analysis, the predicted Cd could only explain about 20% of the measured Cd at the SS site when high level of Cd was transported under northwest wind. Industry contributed 37% and 35% the atmospheric heavy metals at the NH and SD site, respectively.
3.3. Control policy and prospect of atmospheric heavy metals in Foshan The results presented above show that the ceramics industry has caused serious environmental problems in Foshan city, especially in Chanchen district. Potential adverse effects from the pollution have become a concern of the public and the local government. Regulations on ceramics industry in Foshan city were promulgated in 2008. Cleaner production, reducing energy consumption, usage of natural gas, liquid petroleum gas, low sulfur coal and oil, and pollutant discharge permit were included in these regulations. Since the implementation of the regulations on ceramics industry, the concentration of SO2, NOx and PM10 had decreased 23%, 12% and 7.5% from 2008 to 2009, respectively. However, the implemented control technologies have limited effects on levels of the atmospheric heavy metals. Problems still remain in the heavy metal emission control. To effectively reduce the ambient air levels of heavy metals in the city, a few important policies must be enacted and implemented. Firstly, emission standards need to be improved. Since 2010, the MEP of China has sped up the revision of air pollutant emission standards, and strengthened the control of heavy metal pollutant emissions. However, As was not included in the new emission standards for ceramics, aluminum, lead and zinc production, which are the principal sources of air pollutants in Foshan city. As needs to be included in the emission standards.
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270
269
Fig. 7. Source contributions at four sites calculated from EPA PMF model.
Secondly, current control technology for heavy metal emissions in use in Foshan is out of date and needs upgrading. Atmospheric heavy metals were generally controlled by dust control measurements. Several studies have shown that emission control facilities, such as electrostatic precipitators, bag filter, cyclone, wet precipitators and wet desulfurization are capable of removing heavy metals to some extent (Tian et al., 2012). However, these control technologies are still not widely used in many industries in Foshan City. Furthermore, arsenic and some arsenic compounds cannot be removed by these dust control technologies as they sublime upon heating at atmospheric pressure, and are emitted directly in gaseous form. To support the above goals, further work is suggested as follows: 1) conduct comprehensive surveys on the levels and sources of heavy metals in the city; 2) include As in the emission standards in the future; 3) develop and deploy control technologies for As emissions from industries; and 4) apply dust control technology such as wet systems and bag filters to reduce PM emissions from the industries, with anticipated co-benefits of heavy metal emission reduction in Foshan. 4. Conclusion Atmospheric heavy metals pollution in Foshan city is serious, especially for As, Pb, Cd and Zn. Atmospheric heavy metal concentrations are generally higher at the urban sites and ranked as CC N NH N SD N SS location wise. The EFs and back trajectory analysis indicated that local emissions were the main sources. Heavy metals were also significantly associated with the pathway of air mass, suggesting likely impact of long-range transport. A PMF method was applied for source apportionment of heavy metals; five factors (industry, vehicle, dust, transportation and sea salt) were identified. Industry was the most important source of atmospheric heavy metals and contributed 73%, 52%, 37% and 35% the total concentrations of atmospheric heavy metals at the CC, SS, NH and SD sites, respectively. Two special industrial sources (ceramics and aluminum industry) of heavy metals were identified during the sampling period. Atmospheric emission of Cd from an aluminum industry park was most likely the source for the extremely high concentrations of Cd at the SS site. Based on these results, several suggestions were made for the heavy metals pollution control policies: As should be included in
the emission standards in the future; control technology for As should be developed and deployed; dust control technology such as wet systems handbag filters should be widely applied to reduce heavy metal emission. Acknowledgments This work was supported by the National Natural Scientific Foundation of China (No. 41105111; No. 41275134). We are very grateful to Li Jinwei for his help in analysis. We acknowledge the contribution of all the staff who had participated in the sampling and chemical analysis. References Chen JM, Tong YP, Xu JZ, Liu XL, Li YL, Tan MG, et al. Environmental lead pollution threatens the children living in the Pearl River Delta region, China. Environ Sci Pollut Res 2012;19:3268–75. Duan JC, Tan JH. Atmospheric heavy metals and Arsenic in China: situation, sources and control policies. Atmos Environ 2013;74:93–101. Duan FK, Liu XD, Yu T, Cachier H. Identification and estimate of biomass burning contribution to the urban aerosol organic carbon concentrations in Beijing. Atmos Environ 2004;38:1275–82. Duan JC, Tan JH, Wang SL, Chai FH, He KB, Hao JM. Roadside, urban, and rural comparison of size distribution characteristics of PAHs and carbonaceous components of Beijing, China. J Atmos Chem 2012a;69:337–49. Duan JC, Tan JH, Wang SL, Hao JM, Chai FH. Size distributions and sources of elements in particulate matter at curbside, urban and rural sites in Beijing. J Environ Sci 2012b;24:87–94. Duzgoren-Aydin NS. Sources and characteristics of lead pollution in the urban environment of Guangzhou. Sci Total Environ 2007;385:182–95. Draxler RR, Rolph GD. HYSPLIT (HYbrid Single-Particle Lagrangian Integrated Trajectory) Silver Spring. MD: NOAA Air Resources Laboratory; 2003. Fang G, Wu Y, Huang S, Rau J. Review of atmospheric metallic elements in Asia during 2000–2004. Atmos Environ 2005;39:3003–13. Fang GC, Huang YL, Huang JH. Study of atmospheric metallic elements pollution in Asia during 2000–2007. J Hazard Mater 2010;180:115–21. Franck U, Odeh S, Wiedensohler A, Wehner B, Herbarth O. The effect of particle size on cardiovascular disorders—the smaller the worse. Sci Total Environ 2011;409(20): 4217–21. Gao M, Yu Q, Chen LM. Health effects of airborne particulate matter trace elements. Biomed Environ Sci 2005;18:349–55. Guo SJ, Tan JH, Ma YL, Yang FM, Yu YC, Wang JW. Characteristics of atmospheric nonmethane hydrocarbons during high PM10 episodes and normal days in Foshan, China. Atmos Res 2011;101:701–10. Hagler GS, Bergin MH, Salmon LG, Yu JZ, Wan ECH, Zheng M, et al. Source areas and chemical composition of fine particulate matter in the Pearl River Delta region of China. Atmos Environ 2006;40:3802–15.
270
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270
Han YM, Du PX, Cao JJ, Posmentier ES. Multivariate analysis of heavy metal contamination in urban dusts of Xi'an, Central China. Sci Total Environ 2006;355(1):176–86. He K, Wang S, Zhang J. Blood lead levels of children and its trend in China. Sci Total Environ 2009;407:3986–93. Jordán M, Álvarez C, Sanfeliu T. Spherical particles as tracers of atmospheric ceramic industry. Environ Geol 2006;51:447–53. Kang Y, Liu GJ, Chou CL, Wong MH, Zheng LG, Ding R. Arsenic in Chinese coals: distribution, modes of occurrence, and environmental effects. Sci Total Environ 2011; 412:1–13. Lee BK, Hieu NT. Seasonal variation and sources of heavy metals in atmospheric aerosols in a residential area of Ulsan, Korea. Aerosol Air Qual Res 2011;11:679–88. Li XD, Wong CSC, Zhang G, Qi SH, Peng XZ. Annual and seasonal atmospheric deposition of heavy metals in the Pearl River Delta, South China. Abstr Pap Am Chem Soc 2003; 225. [U821-U821]. Loska K, Wiechuya D. Application of principle component analysis for the estimation of source of heavy metal contamination in surface sediments from the Rybnik Reservoir. Chemosphere 2003;51:723–33. Lu SL, Zhang R, Yao ZK, Yi F, Ren JJ, Wu MH, et al. Size distribution of chemical elements and their source apportionment in ambient coarse, fine, and ultrafine particles in Shanghai urban summer atmosphere. J Environ Sci China 2012;24:882–90. Minguillón MC, Monfort E, Querol X, Alastuey A, Celades I, Miró JV. Effect of ceramic industrial particulate emission control on key components of ambient PM10. J Environ Manage 2009;90:2558–67. Nie CR, Lin CX, Du RY Liu Y, Lu WZ, Wu YG, et al. Heavy metal characteristics of vegetables and their soils in Foshan city. J Foshan Univ (Nat Sci Ed) 2010;28:1–5. Paatero P, Tapper U. Positive matrix factorization: a non-negative factor model with optimal utilization of error estimates of data values. Environmetrics 1994;5:111–26. Paatero P. User’s guide for positive matrix factorization programs PMF2 and PMF3, Part 1: tutorial. Department of Physics, Finland: University of Helsinki; 2004. Park K, Heo Y, Putra HE. Ultrafine metal concentration in atmospheric aerosols in urban Gwangju, Korea. Aerosol Air Qual Res 2008;8:411–22. Peng SJ, Ouyang XJ, Li DJ, Yi ZG. Metal elements in Tsp form ambient air of different ceramic workshops in Foshan. Earth Environ 2007;1:011.
Pineiro-Iglesias M, Lopez-Mahia P, Muniategui-Lorenzo S, Prada-Rodriguez D, Querol X, Alastuey A. A new method for the simultaneous determination of PAH and metals in samples of atmospheric particulate matter. Atmos Environ 2003;37:4171–5. Polissar AV, Hopke PK, Paatero P. Atmospheric aerosol over Alaska - 2. Elemental composition and sources, J Geophys Res 1998;103(D15):19045–57. Sharma M, Maloo S. Assessment of ambient air PM10 and PM2.5 and characterization of PM10 in the city of Kanpur, India. Atmos Environ 2005;39:6015–26. Shen J, Wei C. Relocation mechanisms of the ceramics industry impacted by the environmental regulations in Foshan City. Acta Geog Sin 2012;67:467–78. Steinle S, Reis S, Sabel CE. Quantifying human exposure to air pollution—moving from static monitoring to spatio-temporally resolved personal exposure assessment. Sci Total Environ 2013;443:184–93. Tan JH, Duan JC, Chen DH, Wang XH, Guo SJ, Bi XH, et al. Chemical characteristics of haze during summer and winter in Guangzhou. Atmos Res 2009;94:238–45. Taylor SR, McLennan SM. The continental crust: its composition and evolution. Oxford: Blackwell; 1985. Tian HZ, Wang Y, Xue ZG, Cheng K, Qu YP, Chai FH, et al. Trend and characteristics of atmospheric emissions of Hg, As, and Se from coal combustion in China, 1980–2007. Atmos Chem Phys 2010;10:11905–19. Tian HZ, Cheng K, Wang Y, Zhao D, Lu L, Jia WX, et al. Temporal and spatial variation characteristics of atmospheric emissions of Cd, Cr, and Pb from coal in China. Atmos Environ 2012;50:157–63. Wang XH, Bi XH, Sheng GY, Fu HM. Hospital indoor PM10/PM2.5 and associated trace elements in Guangzhou, China. Sci Total Environ 2006a;366:124–35. Wang XL, Sato T, Xing BS. Size distribution and anthropogenic sources apportionment of airborne trace metals in Kanazawa, Japan. Chemosphere 2006b;65:2440–8. Wong CSC, Li XD, Zhang G, Qi SH, Peng XZ. Atmospheric deposition of heavy metals in the Pearl River Delta, China. Atmos Environ 2003;37:767–76. 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. Atmos Chem Phys 2011;11:5207–19.
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Source of atmospheric heavy metals in winter in Foshan, China Ji-Hua Tan a,c, Jing-Chun Duan b,⁎, Yong-Liang Ma c, Fu-Mo Yang a, Yuan Cheng c, Ke-Bin He c, Yong-Chang Yu d, Jie-Wen Wang d a
College of Resource and Environment, University of Chinese Academy of Sciences, Beijing 100049, China State Key Laboratory of Environmental Criteria and Risk Assessment, Chinese Research Academy of Environmental Sciences, Beijing 100012, China State Key Joint Laboratory of Environmental Simulation and Pollution Control, School of Environment, Tsinghua University, Beijing 100084, China d Foshan Environmental Protection Bureau, Foshan 528000, China b c
H I G H L I G H T S • • • • •
Foshan city is the most air polluted city in Pearl rivel Delta. Atmospheric Zn, Pb and As were significantly high in this area. Two special sources were identified during the sampling period. Ceramic industry is the most important source of atmospheric heavy metal. EFs analysis indicated that local emissions were the main sources.
a r t i c l e
i n f o
Article history: Received 19 February 2014 Received in revised form 9 May 2014 Accepted 31 May 2014 Available online xxxx Editor: Xuexi Tie Keywords: Chemical composition PM2.5 PMF Foshan Control policy
a b s t r a c t Foshan is a ceramics manufacturing center in the world and the most polluted city in the Pearl River Delta (PRD) in southern China measured by the levels of atmospheric heavy metals. PM2.5 samples were collected in Foshan in winter 2008. Among the 22 elements and ions analyzed, 7 heavy metals (Zn, V, Mn, Cu, As, Cd and Pb) were studied in depth for their levels, spatiotemporal variations and sources. The ambient concentrations of the heavy metals were much higher than the reported average concentrations in China. The levels of Pb (675.7 ± 378.5 ng/m3), As (76.6 ± 49.1 ng/m3) and Cd (42.6 ± 45.2 ng/m3) exceeded the reference values of NAAQS (GB3095-2012) and the health guidelines of the World Health Organization. Generally, the levels of atmospheric heavy metals showed spatial distribution as: downtown site (CC, Chancheng District) N urban sites (NH and SD, Nanhai and Shunde Districts) N rural site (SS, Shanshui District). Two sources of heavy metals, the ceramic and aluminum industries, were identified during the sampling period. The large number of ceramic manufactures was responsible for the high levels of atmospheric Zn, Pb and As in Chancheng District. Transport from an aluminum industry park under light north-west winds contributed high levels of Cd to the SS site (Shanshui District). The average concentration of Cd under north-west wind was 220 ng/m3, 20.5times higher than those under other wind directions. The high daily maximum enrichment factors (EFs) of Cd, Pb, Zn, As and Cu at all four sites indicated extremely high contamination by local emissions. Back trajectory analysis showed that the heavy metals were also closely associated with the pathway of air mass. A positive matrix factorization (PMF) method was applied to determine the source apportionment of these heavy metals. Five factors (industry including the ceramic industry and coal combustion, vehicle emissions, dust, transportation and sea salt) were identified and industry was the most important source of atmospheric heavy metals. The present paper suggests a control policy on the four heavy metals Cd, Pb, Zn, and Cu, and suggests the inclusion of As in the ceramic industry emission standard in the future. © 2014 Elsevier B.V. All rights reserved.
1. Introduction The association of airborne particulate matter (PM) with adverse health effects has been recognized for a long time (Fang et al., 2005, ⁎ Corresponding author. E-mail address: [email protected] (J.-C. Duan).
http://dx.doi.org/10.1016/j.scitotenv.2014.05.147 0048-9697/© 2014 Elsevier B.V. All rights reserved.
2010; Gao et al., 2005; Yang et al., 2011; Franck et al., 2011; Steinle et al., 2013). Heavy metals are part of PM and pose serious risks to human health, resulting in a variety of human dysfunction (PineiroIglesias et al., 2003) and various diseases. Atmospheric heavy metals are removed by wet and dry deposition which can then cause negative effects on the biogeochemical cycling (Wong et al., 2003). Natural emissions, traffic and industrial emissions are the principal sources of heavy
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270
263
2. Materials and methods
2th to 31th Dec 2008 at three urban sites (CC, NH and SD) and a rural site (SS) (Fig. 1). The CC site is situated in Chancheng district, the metropolitan center of Foshan city, which is surrounded by main traffic roads, residential buildings and factories. Most of the ceramic factories in Foshan city area round this site. The NH site is located in Nanhai district, which is only 2 km from the neighboring Guangzhou city and is surrounded by main traffic roads and residential buildings. The SD site is located in Shunde district, which is China's leading home appliance production base. Finally, the SS site is located in Shanshui forest park in Shanshui district with fewer nearby anthropogenic sources. The sampler trapped PM2.5 on quartz filters (Whatman QM-A, 20.3 × 25.4 cm) for 24 h and a total of 112 samples were collected. All quartz filters were annealed at 500 °C for 4 h to remove trace organics before usage. Before and after collection, filters were wrapped in baked aluminum foil. Prior to weighing, the filters were stored under a constant temperature and humidity condition of 25 °C and 50% for 24 hours. After sampling, each filter was wrapped in aluminum foil, then placed in an airtight bag, and taken to the laboratory at the local Environmental Protection Bureau (EPB) for storage at − 30 °C. Upon the completion of sampling periods, the samples were transported in an ice box to the Key Joint Laboratory of Environment Simulation and Pollution Control at the School of Environment, Tsinghua University, for analysis. The lag period between sampling and analysis was about 2 months. Two types of blank samples were prepared: exposure blanks and filter blanks. The exposure blank filters were placed in the sampler for 24 h but had no air was drawn through them; hence, they were subjected to the same handling procedures as the actual samples. The filter blank samples were used to check for the contamination of the filters without going through the sampling procedures. On average, an exposure blank and a filter blank were taken for every 10 actual samples. Meteorology data (Fig. 2) were recorded at the CC site by the equipment of “Davis Complete Weather Wizard III Cable Basic Weather Monitor” which provided ambient temperature, humidity, pressure, wind speed, wind direction at the CC site every 15 min during the sampling days. Visibility was measured by Belfort Model 6000 Visibility Sensor (Belfort Instrument, USA).
2.1. Sample collection
2.2. Sample analysis
PM2.5 samples were collected simultaneously by high volume (1.13 m 3 /min) samplers (Thermo fisher instruments, USA) from
A small portion of each filter was cut using stainless steel scissors and placed in Teflon tubes with a 4 ml mixture of concentrated high-
metals in the ambient air (Lee and Hieu, 2011; Park et al., 2008). Studies showed that most heavy metals were enriched in PM2.5 (Duan et al., 2012b; Lu et al., 2012). For example, the ratios of heavy metals in PM2.5 to those in PM10 in Beijing, China ranked as Pb (88.5%) N Cd (81.8%) N Zn (81.5%) N As (77.1%) N Cu (75.9%) N Cr (71.7%) N Ni (67.9%) N Mn (63.3%) N V (46.8%) (Duan and Tan, 2013). Foshan is one of the most important manufacturing bases not only for China, but also for the world (Guo et al., 2011). The city is characterized by a high density of ceramics plants, accounting for around 30% world production of ceramics and consuming about 40% coal in Foshan in 2008 (Shen and Wei, 2012). With the substantial economic development, large amounts of pollutants were emitted into the atmosphere, which led to a rapid deterioration of air quality in and around this region. According to the Pearl River Delta Regional Air Quality Monitoring Network, the average concentrations of SO2, NOx and PM10 in Foshan city were significantly higher than those in other areas of Pearl River Delta (PRD). Many studies (Nie et al., 2010; Chen et al., 2012; He et al., 2009) showed that heavy metal pollution was a serious problem in Foshan. Heavy metal (Cd, Pb, Zn, and Cu) level invegetable soil in Foshan were significantly higher than those in the other areas throughout Guangdong Province and the whole country (Nie et al., 2010). Child blood lead levels (BLLs) in Foshan city were much higher than those in the developed countries (Chen et al., 2012; He et al., 2009). Several studies have been conducted on atmospheric heavy metals in the PRD region (Li et al., 2003; Wang et al., 2006a); however, no literature on characteristics and sources of atmospheric heavy metals has been investigated in Foshan. This study was supported by the Foshan City Government to investigate the sources of serious pollution in winter in Foshan, and to put forward suggestions on control policies. Among the 22 elements and ions measured during the study, 7 heavy metals (Zn, V, Mn, Cu, As, Cd and Pb) were studied in-depth for their levels, spatiotemporal variations and sources. The objectives of the present study are to: (1) investigate the levels and their sources of atmospheric heavy metals (Pb, As, Zn and Cd) in Foshan; (2) put forward emission control policy suggestions for atmospheric heavy metals in Foshan.
SS
NH CC
SD
Pearl
River
Delta Fig. 1. Map of sampling sites.
Sampling site
264
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270
Fig. 2. The meteorological conditions and heavy metals during the sampling period.
purity hydrochloric acid and nitric acid with a volume ratio of 3:1. Blank filter and sample replicates were randomly inserted for quality control. The digested solution was diluted to 10 ml using ultra-pure water to perform the metal analysis by inductively coupled plasmamass spectrometry (ICP-MS) (Thermo, X serial). The calibration was made using multi-element (metal) standards (certified reference materials (CRMs)) in a 3% (v/v) HNO3 solution. Since quartz filter is not appropriate for some elemental analysis, such as Al and Ca were not included in the analysis, only elements with low uncertainties were included in this paper. Six blank filters were treated and analyzed in the same way as for the actual samples. The detection limits were Ti (3.16 ppb), Zn (10 ppb), V (0.5 ppb), Mn (1.50 ppb), Cu (1.47 ppb), As (0.38 ppb), Rb (0.33 ppb), Cd (0.03 ppb), Cs (0.5 ppb), Ba (1.75 ppb), Tl (0.006 ppb), Pb (5.88 ppb), Fe (11.32 ppb) and Bi (0.33 ppb). Ion chromatography (Dionex 1400) was used to measure inorganic ions. The concentrations of water soluble inorganic ions in field blanks were 0.07 μg/m3, 0.08 μg/m3, not detected (nd), 0.16 μg/m3, nd, 2− 0.12 μg/m3, nd, 0.05 μg/m3 and 0.08 μg/m3 for F−, Cl−, NO− 3 , SO4 , + + + 2+ 2+ NH4 , Na , K , Ca and Mg , respectively. The relative standard deviation of each ion was less than 6% for the reproducibility test. The detailed analysis procedure of the heavy metal and water soluble inorganic species could be found elsewhere (Duan et al., 2012a; Tan et al., 2009). 2.3. PMF analysis The PMF model was developed by Paatero and Tapper (1994) and Paatero (2004). EPA PMF is one of the receptor models that the US EPA's Office of Research and Development (ORD) have developed. The algorithms used in EPA PMF model to compute profiles and contributions have been peer reviewed by leading scientists in the air quality management community and have been certified scientifically robust. In this study, 13 elements and 9 ions were included in the EPA PMF 3.0 model. The input data including concentrations of chemical species and equation-based uncertainties. The equation-based uncertainty
includes detection limits and error fractions (5%). If the concentration is less than or equal to the method detection limit (MDL) provided, the uncertainty (Unc) is calculated using the following equation (Polissar et al., 1998). Unc ¼
5 MDL 6
ð1 1Þ
If the concentration is greater than the MDL provided, the calculation is Unc ¼
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ðErrorFraction concentrationÞ2 þ ðMDLÞ2
ð1 2Þ
2.4. Back trajectory calculation In order to better understand the transport of airborne particles from distant sources, the HYSPLIT model (HYbrid Single-Particle Lagrangian Integrated Trajectory, Version 4.8, windows-based version), a comprehensive modeling system developed by the National Oceanic and Atmospheric Administration (NOAA) Air Resource Laboratory (Draxler and Rolph, 2003), was used to generate 72-h back trajectories every 6 h from 12/1 to 12/31 at the CC site. The relative ground height was 300 m and every trajectory was colored with the concentration of PM2.5 by Igorpro 6.2 (trajectories colored with atmospheric heavy metals were not presented here for their similarity with PM2.5). Because the synoptic atmospheric conditions at the four sampling sites were quite similar, only the back trajectory arriving at the CC site was calculated in this study. 3. Results and discussion 3.1. Concentration and spatiotemporal variations of heavy metals As shown in Table 1 and Fig. 2, the averaged concentration of PM2.5 was 136.4 μg/m3 during sampling period at the four sites in Foshan City,
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270
265
Table 1 The concentration of chemical composition at four sites in Foshan city. Species
Average
PM2.5 (μg/m3) Zn (ng/m3) Pb (ng/m3) V (ng/m3) Mn (ng/m3) Cu (ng/m3) As (ng/m3) Cd (ng/m3)
136.4 2214 675.7 40 200.6 283.8 76.6 42.6
± ± ± ± ± ± ± ±
CC 40.5 1438 378.5 29.1 129.4 174.7 49.1 45.2
154.4 3239 812.4 68.3 242.3 256.3 140.7 40.8
NH ± ± ± ± ± ± ± ±
47.8 1525 479.7 46.6 86.0 134.3 94.6 25.2
much higher than the National Ambient Air Quality Standard (NAAQS) of China for PM2.5 (75 μg/m3 daily and 35 μg/m3 annually GB30952012). It shows serious PM pollutions in winter in Foshan. The levels of PM2.5 in Foshan were comparable with those in Indian cities (100– 400 μg/m3); however, they were much higher than those reported in Western Europe and North America (10–55 μg/m3) (Sharma and Maloo, 2005). Table 1 showed the average concentrations of the heavy metals at the four sites. The time series of the heavy metals were illustrated in Fig. 2. The average concentration of atmospheric Pb concentration was 675.7 ± 378.5 ng/m3, which fall below the limits of the current national ambient air quality standard (NAAQS-1996) of 1000 ng/m3(Table 2); it however exceeds the new NAAQS (GB3095-2012) and the WHO guideline of 500 ng/m3. The averaged atmospheric V concentration was 40 ± 29.1 ng/m3, which was much lower than the guideline of WHO 1000 ng/m3. The averaged atmospheric As and Cd concentrations were 76.6 ± 49.1 ng/m3 and 42.6 ± 45.2 ng/m3, respectively, which were much higher than the reported average concentrations in China (Duan and Tan, 2013), and were much higher than the limits of the new NAAQS (GB3095-2012) and the WHO guideline. The averaged atmospheric concentrations of Mn, Cu and Zn were 200.6 ± 129.4 ng/m3, 283.8 ± 174.7 ng/m3 and 2214 ± 1438 ng/m3, respectively, but as yet there are no NAAQS for atmospheric Mn, Cu and Zn in China (GB3095-2012). The atmospheric concentrations of the heavy metals in Foshan in winter are ranked as: Zn N Pb N Cu N Mn N As N V N Cd, and all are above the national average air concentrations in China (Duan and Tan, 2013). The sum of concentrations of all atmospheric heavy metals in Foshan in winter ranked as: CC N NH N SD N SS. This ranking is consistent with the unit area energy consumption in Foshan (Shen and Wei, 2012). The highest average concentration of Pb, Zn, V and arsenic was observed at the CC site, Cd at the SS site, Mn and Cu at the NH site.
3.2. Source analysis of heavy metals 3.2.1. Enrichment factor of heavy metals Enrichment factor (EF) can be used to differentiate between the metals originating from human activities and those from natural procedure, and to assess the degree of anthropogenic influence. There is no widely accepted rule for the choice of the reference element; however, Si, Al and Fe are usually used for this purpose. In this study, Fe is used as the reference element and the abundance of elements in the upper
158.9 2157 679.1 32.1 244.3 429.5 51.2 36.0
SD ± ± ± ± ± ± ± ±
37.5 1069 251.4 21.6 80.3 178.4 34.7 10.9
118.8 1989 709.9 36.4 190.4 273.1 59.6 25.6
SS ± ± ± ± ± ± ± ±
29.6 983 296.3 20.0 130.6 122.3 23.7 9.8
107.4 1328 436.7 19.2 99.5 128.5 53.2 80.2
± ± ± ± ± ± ± ±
30.4 711.5 127.4 10.8 28.4 71.5 27.8 112.2
continental crust (UCC) is taken from a previous study (Taylor and McLennan, 1985). The EFCrust of element E in aerosols is defined as below (Wang et al., 2006b): EFCrust ¼ ½E=RAir = ½E=RCrust where R is a reference element of crustal material and [E/R]Air is the concentration ratio of E to R in aerosol, and [E/R]Crust is the concentration ratio of E to R in crust. The EFs of the heavy metals at the four sampling sites were calculated and displayed in Fig. 3. The EFs of Mn and V were lower than 10, confirming its mainly natural source. Enrichment factors of Cd, Pb, Zn, As and Cu were much higher than 10, indicating anthropogenic sources. EFs can also assist the determination of the degree of metal contamination. The mean EFs decreased in the order of Cd N Pb N Zn N As N Cu N V N Mn, suggesting that the decreasing order of their overall contamination degrees in Foshan. Cd, Pb, Zn, As and Cu had mean EFs higher than 40, suggesting extremely high contamination according to the five contamination categories on the basis of the enrichment factor (Loska and Wiechuya, 2003). Mn (CC and NH) and V (CC, NH and SD) had mean EFs higher than 5, indicating significant contamination; while Mn (SS and SD) and V (SS) with mean EFs between 2 and 5, were classified as moderately contaminated. The maximum EFs may reflect the degree to which local pollution affects each metal (Han et al., 2006). The highest daily maximum EFs of Cd were 8973, 18357, 3767 and 5605 at the CC, SS, NH and SD sites, respectively, indicating extremely high Cd contamination of local emission at the SS site. The highest daily maximum EFs of arsenic were 674.35, 141.81, 308.79 and 215.36, respectively, indicating extremely high Arsenic contamination of local emission at the CC site. 3.2.2. Trajectory analysis 3.2.2.1. Back trajectory analysis. Fig. 4 shows the back trajectories ending at Foshan city during the sampling period (2008/12/1–2008/12/31). In general, three categories of air masses could be identified: 1) NE—air masses coming from the northeast that originated from the continental inland areas and passing through Guangzhou city; 2) N—air masses from the north or northeast that originated from continental inland areas and passing through Qingyun city; and 3) E—air masses from the east that originated from continental inland areas of southeast China, reaching the PRD region through southeast China coast. Due to the Asian winter monsoon, the PRD was generally dominated by the
Table 2 Limits for heavy metals in China, WHO and other regions and countries.
NAAQS GB3095-1996 NAAQS GB3095-2012 WHO EU (DIRECTIVE 2004/107/EC) India
Hg
Pb
50 1000
1000 500 500 500
Concentrations corresponding to an excess lifetime risk of 1:100,000.
V
1000
As 6 6.6a 6 6
Mn
150
Ni
a
25 20 20
Cr (VI)
Cd
Year
0.025 0.25a
5 5 5
1996 2012 2000 2004 2009
266
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270
100,000 CC
SD
NH
SS
Average
10,000
Maximum 75th Percentile Median 25th Percentile
Minimum
1,000
100
10
1 Zn
Pb
Cu
Mn
As
V
Cd
Fig. 3. Boxplot of enrichment factors for metals at the four sampling sites in Foshan city.
air masses from the NE and N sector during the sampling period, and only occasionally from the E sector. The average concentrations of heavy metal in the three categories of air masses are shown in Table 3. The NE Category had the highest concentrations of heavy metals, including Zn, Pb, V, Mn, Cu, As and Cd, which were 4384 ng/m 3, 1502 ng/m3, 115 ng/m 3 , 313 ng/m3 , 363 ng/m3, 164 ng/m 3 and 59 ngm/m3 , respectively. The results showed that significant amounts of heavy metals were transported from the northeast inland areas, passing through Guangzhou city to Foshan city. On the contrary, the cleaner air mass from Central Asia contained the lower heavy metal during the N period. It should be noted that despite similar meteorological conditions at the four
sampling sites, most of the heavy metals at the CC site were much higher than those at the other three sites. In particular, the level of Cd was significantly higher during the episode, suggesting that local emissions play an important role in heavy metal pollution in these areas on top of the influence of the long-range transport of pollutants. 3.2.2.2. Local source of heavy metal. Though with fewer nearby anthropogenic sources, extremely high levels of Cd were found at the SS site. Fig. 5 shows the concentrations of PM2.5 and Cd during a pollution episode (12/15–12/21). On the 15th of December, the concentration of Cd (12.1 ng/m3) at the SS site was obviously lower than those at the other
Fig. 4. The back trajectories of air masses arriving at CC site from 12/1 to 12/31.
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270 Table 3 The mean concentrations of heavy metal at the CC site under different air masses. Air mass
n
NE
8
N
15
S
5
Mean S.D. Mean S.D. Mean S.D.
Zn
Pb
Cu
Mn
As
V
Cd
4384 2117 2967 1635 2476 1254
1502 486 917 455 892 344
115 44 49 34 45 26
313 61 208 80 210 74
363 156 219 108 196 101
164 52 101 46 155 144
59 31 34 23 31 16
three sites (34.4 ng/m3); however, it increased sharply then and exceeded the values of other sites significantly on December 17th. At the same time, the concentration of PM2.5 at the SS site was the lowest among the four sites. Based on the EFs analysis, the highest daily maximum EFs of Cd was found at the SS site on December 19th, indicating extremely high Cd contamination from local emissions. Further investigation found that high concentrations of Cd were always related to north-west winds, and low concentrations of Cd with other wind directions. However, as shown in Fig. 5, this emission source had negligible influences on the levels and EFs of Cd at the other three sites, indicating a localized emission source of Cd near the SS site. This localized emission is apparently linked to a large industrial park of smelting aluminum and other metals about 2.5 km to the north-west of the SS site. Previous studies have indicated that smelting aluminum industry is an important emission source of Cd (Tian et al., 2012), and the levels of Cd in vegetable soil in this area was significantly higher than the average value in China (Nie et al., 2010). The high soil Cd in this area may be related with the deposition of Cd. 3.2.3. PMF analysis In order to improve the model calculation, additional elements and ions were included with the heavy metals in the EPA PMF analysis: Ti, Zn, V, Mn, Cu, As, Rb, Cd, Cs, Ba, Tl, Pb, Bi, Na+, K +, Mg2 +,
267
2− − − Ca2 +, NH+ and NO− 4 , F , Cl , SO 4 3 . A critical step in PMF analysis is the determination of the number of factors. There are two principles to determine the numbers of factors: 1) the theoretical value of Q should be equal to the number of data points in the data set; 2) the source profiles also have to make physical sense. In this study, five factors are obtained by EPA PMF3.0 base on these two principles and identified as the following possible sources: industry (include ceramics industry and coal combustion), vehicle, dust, transportation and sea salt. Factor profiles (% of species and contributions) are illustrated in Fig. 6. (See Fig. 5.) Factor 1 has high contributions to F -(64.2%), As (60.4%), Cs (59.0%), Tl (56.7%), Cd (55.1%) and Pb (48.4%), and less than 20% to water soluble inorganic species. Atmospheric fluoride is recognized as a tracer of ceramics industry (Jordán et al., 2006) and was 1471 ng/m3 at the CC site; however, less than 400 ng/m3 at other sampling sites. High levels of Pb, Zn and As were emitted from coal combustion (Duzgoren-Aydin, 2007; Kang et al., 2011; Tian et al., 2010; Duan et al., 2012a,b) and are also commonly used as ceramic flux to lower the high melting point of silica. Peng et al. (2007) found that the concentrations of Pb and Zn are 2870 ng/m 3 and 6750 ng/m3in the flue gas of the ceramic in Foshan. Similar results were also observed in Spain (Minguillón et al., 2009). Foshan is an important manufacturing base in China, and coal is still the primary energy source in Foshan city (Shen and Wei, 2012) and production of ceramic in Foshan city contributes about 30% total world production. So ceramic industry and coal combustion were the most important industrial emission source considering the industry and energy structure in Foshan City. Factor 2 contributes a high percentage to K+(51.1%) and most of the 2− (62.2%) and NO− ions NH+ 4 (64.8%), SO4 3 (54.8%), and contributes less than 20% to main heavy metals, indicating that it's a mixture of aged 2− and NO− pollutants of long range transport. NH+ 4 , SO4 3 are indicators of secondary aerosol and long range transport. K+ is a good marker of
350
SS
SD
NH
CC
Concentration (µg/m3)
300
PM 2.5
250 200 150 100 50 0 350
Aluminum industrypark in Shihui city
Concentration (ng/m3)
300
Cd
Shanshui site
250 200 150 100 50 0 2008/12/15
2008/12/16
2008/12/17
2008/12/18
2008/12/19
2008/12/20
Fig. 5. PM2.5 and Cd at the four sampling sites during a pollution episode.
2008/12/21
268
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270
Fig. 6. Factor profiles (% of species and contributions) obtained from EPA PMF model.
biomass burning (Duan et al., 2004), which was often reported and has been identified as an important contributor to PM2.5 in the PRD (Hagler et al., 2006). Factor 3 contributes significantly to Cu (59.6%), Mg (56.5%), Mn (52.7%), Ca (56%) and Ba (48.2%), which are good indicators of crust related dust (Duan et al., 2012b). Factor 4 has high contributions to V (60.5%), Pb (26.47) and Zn (22.2%) and much less to other species (b 10%). V is a good indicator of combustion of petroleum. Combustion of fossil fuels is the main anthropogenic V emission accounting for 85% of total V emissions in China (Duan and Tan, 2013). Factor 5 contributes high percentages to Cl− (70.4%) and Na (57.8%), indicating sea salt impacts, but much less to most other heavy metals (b10%). Source contributions of As, Pb, Cd, Zn and Mn at the four sites calculated from EPA PMF3.0 model are shown in Fig. 7. It may be concluded that industries, including the ceramics industry and coal combustion, contributed about 73% to these five heavy metals and are the most important sources of heavy metal at the CC site. Industry (52%) and vehicle (17%) were the most important sources at the SS site. In the PMF analysis, the predicted Cd could only explain about 20% of the measured Cd at the SS site when high level of Cd was transported under northwest wind. Industry contributed 37% and 35% the atmospheric heavy metals at the NH and SD site, respectively.
3.3. Control policy and prospect of atmospheric heavy metals in Foshan The results presented above show that the ceramics industry has caused serious environmental problems in Foshan city, especially in Chanchen district. Potential adverse effects from the pollution have become a concern of the public and the local government. Regulations on ceramics industry in Foshan city were promulgated in 2008. Cleaner production, reducing energy consumption, usage of natural gas, liquid petroleum gas, low sulfur coal and oil, and pollutant discharge permit were included in these regulations. Since the implementation of the regulations on ceramics industry, the concentration of SO2, NOx and PM10 had decreased 23%, 12% and 7.5% from 2008 to 2009, respectively. However, the implemented control technologies have limited effects on levels of the atmospheric heavy metals. Problems still remain in the heavy metal emission control. To effectively reduce the ambient air levels of heavy metals in the city, a few important policies must be enacted and implemented. Firstly, emission standards need to be improved. Since 2010, the MEP of China has sped up the revision of air pollutant emission standards, and strengthened the control of heavy metal pollutant emissions. However, As was not included in the new emission standards for ceramics, aluminum, lead and zinc production, which are the principal sources of air pollutants in Foshan city. As needs to be included in the emission standards.
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270
269
Fig. 7. Source contributions at four sites calculated from EPA PMF model.
Secondly, current control technology for heavy metal emissions in use in Foshan is out of date and needs upgrading. Atmospheric heavy metals were generally controlled by dust control measurements. Several studies have shown that emission control facilities, such as electrostatic precipitators, bag filter, cyclone, wet precipitators and wet desulfurization are capable of removing heavy metals to some extent (Tian et al., 2012). However, these control technologies are still not widely used in many industries in Foshan City. Furthermore, arsenic and some arsenic compounds cannot be removed by these dust control technologies as they sublime upon heating at atmospheric pressure, and are emitted directly in gaseous form. To support the above goals, further work is suggested as follows: 1) conduct comprehensive surveys on the levels and sources of heavy metals in the city; 2) include As in the emission standards in the future; 3) develop and deploy control technologies for As emissions from industries; and 4) apply dust control technology such as wet systems and bag filters to reduce PM emissions from the industries, with anticipated co-benefits of heavy metal emission reduction in Foshan. 4. Conclusion Atmospheric heavy metals pollution in Foshan city is serious, especially for As, Pb, Cd and Zn. Atmospheric heavy metal concentrations are generally higher at the urban sites and ranked as CC N NH N SD N SS location wise. The EFs and back trajectory analysis indicated that local emissions were the main sources. Heavy metals were also significantly associated with the pathway of air mass, suggesting likely impact of long-range transport. A PMF method was applied for source apportionment of heavy metals; five factors (industry, vehicle, dust, transportation and sea salt) were identified. Industry was the most important source of atmospheric heavy metals and contributed 73%, 52%, 37% and 35% the total concentrations of atmospheric heavy metals at the CC, SS, NH and SD sites, respectively. Two special industrial sources (ceramics and aluminum industry) of heavy metals were identified during the sampling period. Atmospheric emission of Cd from an aluminum industry park was most likely the source for the extremely high concentrations of Cd at the SS site. Based on these results, several suggestions were made for the heavy metals pollution control policies: As should be included in
the emission standards in the future; control technology for As should be developed and deployed; dust control technology such as wet systems handbag filters should be widely applied to reduce heavy metal emission. Acknowledgments This work was supported by the National Natural Scientific Foundation of China (No. 41105111; No. 41275134). We are very grateful to Li Jinwei for his help in analysis. We acknowledge the contribution of all the staff who had participated in the sampling and chemical analysis. References Chen JM, Tong YP, Xu JZ, Liu XL, Li YL, Tan MG, et al. Environmental lead pollution threatens the children living in the Pearl River Delta region, China. Environ Sci Pollut Res 2012;19:3268–75. Duan JC, Tan JH. Atmospheric heavy metals and Arsenic in China: situation, sources and control policies. Atmos Environ 2013;74:93–101. Duan FK, Liu XD, Yu T, Cachier H. Identification and estimate of biomass burning contribution to the urban aerosol organic carbon concentrations in Beijing. Atmos Environ 2004;38:1275–82. Duan JC, Tan JH, Wang SL, Chai FH, He KB, Hao JM. Roadside, urban, and rural comparison of size distribution characteristics of PAHs and carbonaceous components of Beijing, China. J Atmos Chem 2012a;69:337–49. Duan JC, Tan JH, Wang SL, Hao JM, Chai FH. Size distributions and sources of elements in particulate matter at curbside, urban and rural sites in Beijing. J Environ Sci 2012b;24:87–94. Duzgoren-Aydin NS. Sources and characteristics of lead pollution in the urban environment of Guangzhou. Sci Total Environ 2007;385:182–95. Draxler RR, Rolph GD. HYSPLIT (HYbrid Single-Particle Lagrangian Integrated Trajectory) Silver Spring. MD: NOAA Air Resources Laboratory; 2003. Fang G, Wu Y, Huang S, Rau J. Review of atmospheric metallic elements in Asia during 2000–2004. Atmos Environ 2005;39:3003–13. Fang GC, Huang YL, Huang JH. Study of atmospheric metallic elements pollution in Asia during 2000–2007. J Hazard Mater 2010;180:115–21. Franck U, Odeh S, Wiedensohler A, Wehner B, Herbarth O. The effect of particle size on cardiovascular disorders—the smaller the worse. Sci Total Environ 2011;409(20): 4217–21. Gao M, Yu Q, Chen LM. Health effects of airborne particulate matter trace elements. Biomed Environ Sci 2005;18:349–55. Guo SJ, Tan JH, Ma YL, Yang FM, Yu YC, Wang JW. Characteristics of atmospheric nonmethane hydrocarbons during high PM10 episodes and normal days in Foshan, China. Atmos Res 2011;101:701–10. Hagler GS, Bergin MH, Salmon LG, Yu JZ, Wan ECH, Zheng M, et al. Source areas and chemical composition of fine particulate matter in the Pearl River Delta region of China. Atmos Environ 2006;40:3802–15.
270
J.-H. Tan et al. / Science of the Total Environment 493 (2014) 262–270
Han YM, Du PX, Cao JJ, Posmentier ES. Multivariate analysis of heavy metal contamination in urban dusts of Xi'an, Central China. Sci Total Environ 2006;355(1):176–86. He K, Wang S, Zhang J. Blood lead levels of children and its trend in China. Sci Total Environ 2009;407:3986–93. Jordán M, Álvarez C, Sanfeliu T. Spherical particles as tracers of atmospheric ceramic industry. Environ Geol 2006;51:447–53. Kang Y, Liu GJ, Chou CL, Wong MH, Zheng LG, Ding R. Arsenic in Chinese coals: distribution, modes of occurrence, and environmental effects. Sci Total Environ 2011; 412:1–13. Lee BK, Hieu NT. Seasonal variation and sources of heavy metals in atmospheric aerosols in a residential area of Ulsan, Korea. Aerosol Air Qual Res 2011;11:679–88. Li XD, Wong CSC, Zhang G, Qi SH, Peng XZ. Annual and seasonal atmospheric deposition of heavy metals in the Pearl River Delta, South China. Abstr Pap Am Chem Soc 2003; 225. [U821-U821]. Loska K, Wiechuya D. Application of principle component analysis for the estimation of source of heavy metal contamination in surface sediments from the Rybnik Reservoir. Chemosphere 2003;51:723–33. Lu SL, Zhang R, Yao ZK, Yi F, Ren JJ, Wu MH, et al. Size distribution of chemical elements and their source apportionment in ambient coarse, fine, and ultrafine particles in Shanghai urban summer atmosphere. J Environ Sci China 2012;24:882–90. Minguillón MC, Monfort E, Querol X, Alastuey A, Celades I, Miró JV. Effect of ceramic industrial particulate emission control on key components of ambient PM10. J Environ Manage 2009;90:2558–67. Nie CR, Lin CX, Du RY Liu Y, Lu WZ, Wu YG, et al. Heavy metal characteristics of vegetables and their soils in Foshan city. J Foshan Univ (Nat Sci Ed) 2010;28:1–5. Paatero P, Tapper U. Positive matrix factorization: a non-negative factor model with optimal utilization of error estimates of data values. Environmetrics 1994;5:111–26. Paatero P. User’s guide for positive matrix factorization programs PMF2 and PMF3, Part 1: tutorial. Department of Physics, Finland: University of Helsinki; 2004. Park K, Heo Y, Putra HE. Ultrafine metal concentration in atmospheric aerosols in urban Gwangju, Korea. Aerosol Air Qual Res 2008;8:411–22. Peng SJ, Ouyang XJ, Li DJ, Yi ZG. Metal elements in Tsp form ambient air of different ceramic workshops in Foshan. Earth Environ 2007;1:011.
Pineiro-Iglesias M, Lopez-Mahia P, Muniategui-Lorenzo S, Prada-Rodriguez D, Querol X, Alastuey A. A new method for the simultaneous determination of PAH and metals in samples of atmospheric particulate matter. Atmos Environ 2003;37:4171–5. Polissar AV, Hopke PK, Paatero P. Atmospheric aerosol over Alaska - 2. Elemental composition and sources, J Geophys Res 1998;103(D15):19045–57. Sharma M, Maloo S. Assessment of ambient air PM10 and PM2.5 and characterization of PM10 in the city of Kanpur, India. Atmos Environ 2005;39:6015–26. Shen J, Wei C. Relocation mechanisms of the ceramics industry impacted by the environmental regulations in Foshan City. Acta Geog Sin 2012;67:467–78. Steinle S, Reis S, Sabel CE. Quantifying human exposure to air pollution—moving from static monitoring to spatio-temporally resolved personal exposure assessment. Sci Total Environ 2013;443:184–93. Tan JH, Duan JC, Chen DH, Wang XH, Guo SJ, Bi XH, et al. Chemical characteristics of haze during summer and winter in Guangzhou. Atmos Res 2009;94:238–45. Taylor SR, McLennan SM. The continental crust: its composition and evolution. Oxford: Blackwell; 1985. Tian HZ, Wang Y, Xue ZG, Cheng K, Qu YP, Chai FH, et al. Trend and characteristics of atmospheric emissions of Hg, As, and Se from coal combustion in China, 1980–2007. Atmos Chem Phys 2010;10:11905–19. Tian HZ, Cheng K, Wang Y, Zhao D, Lu L, Jia WX, et al. Temporal and spatial variation characteristics of atmospheric emissions of Cd, Cr, and Pb from coal in China. Atmos Environ 2012;50:157–63. Wang XH, Bi XH, Sheng GY, Fu HM. Hospital indoor PM10/PM2.5 and associated trace elements in Guangzhou, China. Sci Total Environ 2006a;366:124–35. Wang XL, Sato T, Xing BS. Size distribution and anthropogenic sources apportionment of airborne trace metals in Kanazawa, Japan. Chemosphere 2006b;65:2440–8. Wong CSC, Li XD, Zhang G, Qi SH, Peng XZ. Atmospheric deposition of heavy metals in the Pearl River Delta, China. Atmos Environ 2003;37:767–76. 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. Atmos Chem Phys 2011;11:5207–19.
