Individual particle analysis of aerosols collected at Lhasa City in the Tibetan Plateau

J O U RN A L OF E N V I RO N M EN TA L S CI EN CE S 29 (2 0 1 5 ) 1 6 5–1 7 7

Available online at www.sciencedirect.com

ScienceDirect www.journals.elsevier.com/journal-of-environmental-sciences

Individual particle analysis of aerosols collected at Lhasa City in the Tibetan Plateau Bu Duo1,2 , Yunchen Zhang1 , Lingdong Kong1 , Hongbo Fu1,⁎, Yunjie Hu1 , Jianmin Chen1,⁎, Lin Li3 , A. Qiong3 1. Shanghai Key Laboratory of Atmospheric Particle Pollution and Prevention, Department of Environmental Science & Engineering, Fudan University, Shanghai 200433, China 2. Department of Chemistry & Environmental Science, Tibet University, Lhasa 850000, China 3. Environmental Monitoring Center Station of Tibet Autonomous Region, Lhasa 850000, China

AR TIC LE I N FO

ABS TR ACT

Article history:

To understand the composition and major sources of aerosol particles in Lhasa City on the

Received 9 May 2014

Tibetan Plateau (TP), individual particles were collected from 2 February to 8 March, 2013 in

Revised 29 June 2014

Tibet University. The mean concentrations of both PM2.5 and PM10 during the sampling were

Accepted 18 July 2014

25.7 ± 21.7 and 57.2 ± 46.7 μg/m3, respectively, much lower than those of other cities in East

Available online 3 February 2015

and South Asia, but higher than those in the remote region in TP like Nam Co, indicating minor urban pollution. Combining the observations with the meteorological parameters

Keywords:

and back trajectory analysis, it was concluded that local sources controlled the pollution

Individual particles

during the sampling. Transmission electron microscopy (TEM) combined with energy-

Fireworks

dispersive X-ray spectra (EDS) was used to study 408 particles sampled on four days. Based

Biomass burning

on the EDS analysis, a total of 8 different particle categories were classified for all 408

Lhasa

particles, including Si-rich, Ca-rich, soot, K-rich, Fe-rich, Pb-rich, Al-rich and other particles.

TEM

The dominant elements were Si, Al and Ca, which were mainly attributed to mineral dust in the earth's crust such as feldspar and clay. Fe-, Pb-, K-, Al-rich particles and soot mainly originated from anthropogenic sources like firework combustion and biomass burning during the sampling. During the sampling, the pollution mainly came from mineral dust, while the celebration ceremony and religious ritual produced a large quantity of anthropogenic metal-bearing particles on 9 and 25 February 2013. Cement particles also had a minor influence. The data obtained in this study can be useful for developing pollution control strategies. © 2015 The Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. Published by Elsevier B.V.

Introduction The Tibetan Plateau (TP) is the largest and highest plateau on earth, with an average elevation of over 4000 m above sea level (Cong et al., 2007), and an area of approximately 2.5 × 106 km2 with a size of about one quarter of the Chinese

territory, which extends over the area of 27°–45°N, 70°–105°E. It has the largest area of snow and ice in the mid-latitudes and serves as “the world's water tower”, acting as water storage for South and East Asia. Due to its particular orography, location and large extent, the TP can significantly influence the atmospheric circulation, energy budget, and hydrological

⁎ Corresponding authors. E-mail: [email protected] (Hongbo Fu), [email protected] ( Jianmin Chen).

http://dx.doi.org/10.1016/j.jes.2014.07.032 1001-0742/© 2015 The Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. Published by Elsevier B.V.

166

J O U RN A L OF E N V I RO N ME N TA L S CI EN CE S 29 (2 0 1 5 ) 1 65–1 7 7

cycles in Asia or even the globe through dynamical and thermal forcing. For example, numerical simulations have suggested that the low pressure over the TP as an elevated heat source in summer induces a supply of moist, warm air from the Indian Oceans to the continent, affecting Asia summer monsoon circulation profoundly (Liu et al., 2008). According to earlier sunphotometer observations, very low annual aerosol optical depth (AOD) was recorded on the central TP (Cong et al., 2009). Thus, the atmosphere over the TP is minimally disturbed by anthropogenic activities due to the sparse population and limited industries. However, the TP is located close to regions in South and East Asia that have been and are predicted to continue to be the largest sources of black soot in the world (Cong et al., 2013). Widespread air pollution can enter the interior TP through atmospheric circulation. Based on the GEOS–Chem model, Kopacz et al. (2011) proposed that the TP receives most black carbon (BC) from western and central China, as well as from India, Nepal, and Pakistan. In the extreme cases, they found that sources as varied as African biomass burning and Middle Eastern fossil fuel combustion could significantly contribute to the BC reaching the Himalayas and TP. The emission inventory in conjunction with air mass trajectories demonstrated that BC in the Nam Co lake region was most likely transported from South Asia (Cong et al., 2013). Meanwhile, dust particles, arising from the numerous surrounding sources, can be transported to and stacked over the TP, especially on the northern side of the surrounding area where the Taklimakan and Gobi deserts are located (Xia et al., 2008; Ferrat et al., 2013). Model studies by Lau et al. (2006) show that the highly elevated surface air over the plateau may act as an “elevated heat pump” and alter the regional climate significantly through the absorption of solar radiation by dust coupled with black carbon. Furthermore, these colored particles incorporated in snowflakes darken snow and ice surfaces, increasing surface melt. As the world's largest ice storage site after the Arctic and Antarctic, the glaciers on the TP are experiencing shrinkage largely caused by increasing atmospheric BC deposition (Xu et al., 2009a,b). Qian et al. (2011) suggested that the changes of surface albedo and radiative fluxes over the TP induced by the deposition of dust and black carbon on the snow of the TP could significantly modify the hydrological cycle and monsoon climate in India and East Asia. Analysis of ice cores from the Tibet–Himalaya region also demonstrated the aerosols' impact on climate and glacier melt (Chen et al., 2013). Thus, the TP impacts on regional and global climate could be modified by the aerosols transported over the TP. Based on satellite data, Yang et al. (2012) proposed that the solar dimming over the TP is mainly due to the increase in water vapor amounts, which in turn are related to the rapid warming and the increase in convective available potential energy. Short-term ground-based remote sensing aerosol optical depth (AOD) in Lhasa showed that the daily average at 500 nm was less than 0.2 during summer (You et al., 2013). On the basis of ice core records, the temporal trend of BC fluxes clearly showed a recent rise, reflecting increased emissions from anthropogenic activities. Xu et al. (2009a,b) further proved that black soot aerosols deposited on Tibetan glaciers have been a significant contributing factor to observed rapid glacier retreat. Kang et al. (2010) present a history of atmospheric dust loading variability, reconstructed using an ice core record from Mt. Geladaindong,

and suggested that dust loading over the TP was closely related to atmospheric circulation. In general, a number of atmosphere research programs have been conducted on the TP, but most investigations have depended on remote sensing data or ice core glaciochemistry. Elemental analysis of aerosol collected from a remote region of the central TP showed that the mean elemental concentrations were comparable with those from other remote sites and significantly lower than those from megacities, such as Beijing and Shanghai (Cong et al., 2007). Several anthropogenic heavy metals are transported long distances atmospherically from South Asia. In an early research study, Zhang et al. (2001a,b) also investigated some aerosol samples collected at Lhasa and Gongga in the summers of 1998 and 1999. They found that soot particles emitted from vegetation burning were dominant in the urban area of Lhasa city. Elements of PM2.5 collected at Lijiang, at the southeastern edge of the TP, were mainly from crustal sources, biomass burning emissions and regional traffic-related sources (Zhang et al., 2013). Indoor air pollution in the nomadic tents on the TP, mainly due to yak dung combustion, has also become a growing concern for the health of herdsmen. Both trace metals and PAHs within Tibetan tents were much higher than those of outdoor air in the study area, and outdoor air quality of the study area was affected by indoor air pollutants to some extent (Li et al., 2012). In general, several aerosol-related research programs have been conducted on the TP. But due to the low atmospheric pressure, some sampling instruments could be affected, such as pumps, and the harsh climate makes it difficult to carry out long-term sampling. Thus, TP surface observation has been limited due to the high elevation and harsh climate, which makes it difficult to judge whether model results are realistic. Therefore, in light of the vast area of the TP and the very limited field data available, more investigation is needed to understand the aerosol chemical composition over the TP. Lhasa (29.65°N, 91.13°E) is a famous city for tourism, and it is also the Qinghai–Tibet railway terminal. Increasing numbers of tourists are visiting the city, consequently influencing the environment, e.g., its water resources, precipitation chemistry and atmosphere. In this work, atmospheric aerosol particles in the Lhasa urban area were collected in the winter of 2013. The particles were analyzed with electron microscopes coupled with an energy dispersive X-ray spectrometer. Individual particle morphology and elemental composition were investigated. On the basis of the back trajectory, the possible sources of aerosols were analyzed. The impact of firework/firecracker combustion and biomass burning attributed to the celebration ceremony and religious ritual in the Lhasa urban area was analyzed. This work could fill the gap that exists in our knowledge on the aerosol characteristics in the special conditions over this huge high elevation area.

1. Experimental section 1.1. Sampling The sampling was conducted at the top of a five-story building (29°38′42.72″N, 91°08′39.10″E), 17 m above the ground, 3667 m of altitude, at Tibet University in Chengguan District of Lhasa, lying in the Jiangsu Road and East Jiangsu Road, with some

J O U RN A L OF E N V I RO N M EN TA L S CI EN CE S 29 (2 0 1 5 ) 1 6 5–1 7 7

minor residential emission sources but few industrial and mobile sources around the sampling site (shown in Fig. 1). Aerosol samples were collected over the period of 2 February to 8 March 2013. 43 samples were collected during the sampling period at the rate of 1–4 samples daily. One single-stage cascade impact or with 1.0-mm-diameter jet nozzle was used to collected particles on Transmission electron microscopy (TEM) copper grids for TEM analysis. The sampler inlet was mounted at the height of 2 m vertically above the floor. Air flow rate was set at 1.0 L/min, and sampling duration time depended on the ambient visibility, ranging from 30 sec to 30 min. Assuming the density to be 2 g/cm3 for particles, the cut-off aerodynamic diameter is 0.5 μm at this flow rate. Collection efficiency is highest for particles of ~1.0 μm diameter, which are likely to have the longest lifetime in the atmosphere. After sampling finished, samples were preserved in plastic carriers, then sealed in a plastic bag to minimize exposure to the atmosphere.

1.2. TEM analysis A JEOL-2010F field emission high-resolution transmission electron microscope (FEHRTEM) with an Oxford energy-dispersive X-ray Spectrometer (EDS) was applied to obtain the morphology

167

and composition of particles collected on the TEM copper grids. The instrument can detect and semi-quantify elements with atomic weight larger than 11, as well as the crystallography of compounds (Fu et al., 2012). As the TEM copper grids were coated with carbon films (carbon type-B, 300-mesh copper, Tianld Co., China), Cu and C were ruled out of consideration. Exposure time of 15 sec was chosen for EDS analysis, to minimize the beam damage to sensitive particles. Selected area electron diffraction (SAED) patterns or HRTEM images were used to examine the structure of crystalline particles (Li et al., 2011). PCPDFWIN (version 2.02) software was used to index the metal compounds by comparing the d-spacings of SAED patterns or HRTEM images in some areas of interest in the particles with crystallographic data from International Centre for Diffraction Data (ICDD) inorganic compound powder diffraction file (PDF) database. Thus, in combination with chemical composition, it was possible to identify the speciation almost unequivocally. The X-ray counts from the carbon coating of TEM grids were estimated by analyzing blank areas between particles. The net X-ray counts of each particle analyzed were calculated, and their proportions were normalized to 100%. The distribution of aerosol particles on TEM grids was not uniform, with coarser particles occurring near the center, and finer particles occurring on the

Fig. 1 – Schematic map of the sampling site in Lhasa.

168

J O U RN A L OF E N V I RO N ME N TA L S CI EN CE S 29 (2 0 1 5 ) 1 65–1 7 7

Table 1 – Meteorological parameters during collecting samples. Sampling time (BST a) Date

Starting

7-Feb 9-Feb 24-Feb 25-Feb

16:00 23:57 11:15 9:30

a

Duration 25 10 30 30

Relative humidity (%)

Temp. (°C)

12 – 20 46

12 – 4 0

min min min min

Wind Speed (km/hr)

Direction

7.2 – 3.6 –

225 – 90 –

Visibility (km) 30 – 30 30

No.

Feb. Feb. Feb. Feb.

7 9 24 25

Beijing standard time (8 hr prior to GMT).

periphery (Fu et al., 2012; Li and Shao, 2009). Therefore, to ensure that the analyzed particles were more representative, three areas were chosen from the center and periphery of the sampling spot on each grid. Every particle in the selected area was analyzed. 408 individual aerosol particles in the samples were analyzed.

1.3. Composition statistics Particles can be distinguished by their features including morphology, mixing state, composition, beam sensitivity and structure (Li and Shao, 2009; Fu et al., 2012). Eight main categories of particles were recognized in the study: K-rich, Si-rich, Ca-rich, soot, Pb-rich, Fe-rich, Al-rich particles and others. Others were Mg- and Ti-rich particles. There was no special ceremony on 7 and 24 February, representing the normal particle pollution in Lhasa. There was firework /firecracker combustion on 9 and 25 February, and also large-scale biomass burning on 9 February, because of the New Year celebration and religious ritual. So we

chose the four days (7, 9, 24 and 25 February) to study the normal and special particle pollution in Lhasa. The meteorological parameters on 7, 9, 24 and 25 February are shown in Table 1. During 7, 9, 24 and 25 February, Pb-, Fe- and Al-rich particles account for more than 91% of all the metal-rich particles, leading to insignificant quantitative contribution by the remnants. Statistics are based on the number concentration. The numbers of particles in the four samples are shown in Table 2.

1.4. Back trajectory The NOAA/ARL (U.S. National Oceanic and Air Administration/Air Resources Laboratory) Hybrid Single-Particle Lagrangian Integrated Trajectory model (available at http://www.arl. noaa.gov/ready/hysplit4.html) was employed to determine 4 back trajectories arriving at Lhasa at 100, 500, 1000 m, respectively. Each trajectory represented the past 72 hr of the air mass, with its arrival time at 00:00 UTC every day (Draxler and Hess, 1998).

Table 2 – Classification of particles in Lhasa based on EDS.

-

169

J O U RN A L OF E N V I RO N M EN TA L S CI EN CE S 29 (2 0 1 5 ) 1 6 5–1 7 7

2. Results and discussion 2.1. Particle concentration During the sampling, particle concentrations generally stayed at low levels with several small peaks occurring occasionally, as shown in Fig. 2. The mean concentrations of PM2.5 and PM10 during this period were 25.7 ± 21.7 and 57.2 ± 46.7 μg/m3, respectively. The daily PM10 average concentration during the sampling was lower than Grade (II) of NAAQS in China (150 μg/m3). The daily PM2.5 average concentration during the sampling was lower than Grade (I) of NAAQS in China (35 μg/m3). The PM10 concentrations indicated that Lhasa had a relatively low particulate matter loading compared with other Asian cities, such as Beijing (172 μg/m3), Wuhan (156–197 μg/m3), Lahore, Pakistan (340 μg/m3) and Agra, India (154 μg/m3), while the situation was similar to some European cities, such as Bern, Switzerland (40 μg/m3), and Budapest, Hungary (48–54 μg/m3) (Cong et al., 2011). However, compared with the remote region in TP, where the annual average concentrations of total suspended particulate measured at Nam Co was 6.74 μg/m3 (Cong et al., 2009), which was much lower than Lhasa, a minor level of urban pollution in Lhasa was indicated. However, on 7, 9, 14 and 28 February, the particle concentrations sharply climbed up and lasted for several hours. The highest PM10 concentrations were 322.8, 244.5, 437.4 and 335.9 μg/m3 on 7, 9, 14 and 28 February, respectively. On 7, 14 and 28 February, the average hourly ratio of PM2.5/PM10 reached the low value of 0.32, 0.31 and 0.32, respectively, indicating the characteristic of coarse particle pollution (Fu et al., 2014). There was dusty weather on 28 February, so we supposed that the locally raised dusty weather contributed to the coarse particle pollution on 28 February. On 9 February, the average hourly ratio of PM2.5/PM10 reached 0.67, almost 1.5 times higher than the average ratio of PM2.5/PM10

(0.42) during the sampling, indicating the characteristic of fine particle pollution. On 9 February, there was large-scale firework/ firecracker combustion to celebrate the New Year in Lhasa. Moreno et al. (2007) suggested that the particles formed by firework combustion would yield a high PM2.5/PM10 ratio, which is consistent with our observation. Therefore the firework/ firecracker combustion on February 9 was assumed to contribute to the fine particle pollution. Fig. 2 also shows the concentrations of SO2, NO2, CO and O3 during the sampling period. The mean concentrations of SO2, NO2, CO and O3 were 8.29 ± 3.37, 19.95 ± 10.85, 1109.04 ± 306.54 and 72.10 ± 17.33 μg/m3, respectively. The SO2 concentration was lower than the Grade (I) of NAAQS in China (150 μg/m3, hourly average data). The NO2 concentration was lower than the Grade (I) of NAAQS in China (120 μg/m3, hourly average data). The CO concentration was much lower than the Grade (III) of NAAQS in China (2000 μg/m3, hourly average data). The O3 concentration was lower than the Grade (I) of NAAQS in China (120 μg/m3, hourly average data). Thus, the SO2, NO2 and O3 pollution in Lhasa was very mild due to the few industries and little population in Lhasa; but the CO concentration was relatively large. CO is mostly contributed by the incomplete combustion of carbon. Thus the firework/ firecracker combustion and biomass burning were assumed to impact the CO concentration.

2.2. Meteorological interpretation Fig. 3 shows the surface meteorological parameters, including wind speed, wind direction, relative humidity, dew point and temperature every 3 hr. During the sampling, winds were generally weak (0–5 (1.29) m/sec) (Number in the square brackets denotes the arithmetic average, similar hereinafter), disfavoring pollutant dispersion. Surface meteorology showed a dominant southwest wind during 5–25 February. During the

100 0

CO (μg/m3)

3000

250 200 150 100 50 0

2000 1000

PM10 (μg/m3)

0 500 400 300 200 100 0 2/2

O3 (μg/m3)

0

200 PM2.5

150 100 50

PM2.5 (μg/m3)

SO2 (μg/m3)

200 60

NO2 (μg/m3)

300

120

0 2/6

2/11

2/16

2/21 Date (m/dd)

2/26

3/3

3/8

Fig. 2 – Time series of the meteorological parameters including concentrations of SO2, NO2, CO, O3, PM2.5 and PM10 (μg/m3) during February 2 to March 8.

170

J O U RN A L OF E N V I RO N ME N TA L S CI EN CE S 29 (2 0 1 5 ) 1 65–1 7 7

North 1m/sec

Temperature (°C)

Dew point (°C) RH (%)

60

60

40

40

20

20

0 2/2

2/6

2/11

2/16

2/21 Date (m/dd)

2/26

3/3

Dew point (°C), RH (%)

80

80

0 3/8

Fig. 3 – Time series of the meteorological parameters including wind direction and speed (m/sec), relative humidity (%), temperature (°C) and dew point (°C) during 2 February to 8 March.

other period of sampling, wind was mainly from the southeast direction. During the sampling, the visibility was mostly beyond 30 km, with low relative humidity (RH) (4–74 (12.6)%). Huang and Yang (2013) pointed out that particle concentration is negatively related to the visibility, so the low concentrations of PM10 and PM2.5 during the sampling are assumed to lead to the high visibility.

2.3. Back trajectory interpretation Surface meteorology showed a dominant southwest wind and air backward trajectories starting at three altitudes of 100, 500 and 1000 m, most of which flowed from Northern India and Nepal, and passed over the Himalaya Mountains on 7, 9, 24 and 25 February, as shown in Fig. 4. The 7th February possessed short

32

32 02-09

02-07 Lhasa

28

26

Lhasa

30

6000

Latitude

Latitude

30

28

26

6000 5000

5000

4000

4000

24

3000

24

3000

2000

2000

1000

1000

22

0

84

86

88 longitude

90

22

92

32

86

88 longitude

90

02-25 Lhasa

30

28

26

6000

Latitude

30

Lhasa

28

26

6000 5000

5000

4000

4000

24

3000

24

3000 2000

2000

1000

1000

22

92

32 02-24

Latitude

0

84

0

84

86

88 longitude

90

92

22

0

84

86

88 longitude

90

92

Fig. 4 – Three-day back trajectories of analyzed samples ending at the height 100, 500 and 1000 m, with the height indicated by color-coded line.

J O U RN A L OF E N V I RO N M EN TA L S CI EN CE S 29 (2 0 1 5 ) 1 6 5–1 7 7

trajectories at height of 500 and 1000 m that originated from central Nepal, while the trajectory at height of 100 m was much longer and originated from north India. This suggests that the air parcel moved slowly from the Himalayan Mountains in Nepal. On 9 February, all air masses originated from central Nepal with low wind speed. On 24 February, the air masses at 100 and 500 m, which originated from east Nepal, moved very slowly when they arrived at Lhasa, and may have impacted the ground-level air quality in Lhasa. The day on 25 February possessed a short trajectory at 100 m height that originated from Bhutan, while the trajectories at 500 and 1000 m height originated from southwest India. On 7, 9, 24 and 25 February, the average wind speed reached as high as 1.13, 1.43, 1.38 and 1.38 m/sec in Lhasa, with gust speeds over 3.0 m/sec, respectively. These suggest that the air parcel mostly moved slowly from the south Tibetan Plateau. Cong et al. (2009) suggested that the particle concentration at the remote region in TP was low, and Nepal is known as a low-level industrialized country. Thus, the particle pollution via long-range transport from outside sources had little impact on Lhasa air pollution, and pollution was mainly contributed by local sources as favored by calm weather.

2.4. Particle analysis Based on the EDS analysis, the modified mass fraction method of X element can be used to distinguish the composition of particles (Okada et al., 2005). P (X) = X / (Na + Mg + Al + Si + P + S + K + Ca + Mn + Fe + Ba + Zn + Pb + Ti + Cl + V). Si-. Ca-, K-, Fe-, Pb-, Al-rich particles and soot were the most abundant particles observed under the TEM. They constituted more than 98% of the 408 particles in the study. Particles rich in other elements were too rare to be significant for statistics. Fe- and Pb-rich particles constituted 68% of metal-rich particles, while the Fe-, Zn- and Pb-rich particles constituted 60% of metal-rich particles in Mexico City (Adachi and Buseck, 2010). It is well documented that the major sources of Zn in atmospheric particles around the globe are combustion of fossil fuels, industrial metallurgical process, waste incineration and trafficrelated sources (Nriagu and Pacyna, 1988; Chueinta et al., 2000; Rogge et al., 1993). It is well known that the Tibetan Plateau is a relatively pristine region. No Zn-rich particles were observed in our study because of the low local emission of industrial and traffic-related pollutants in Lhasa. Instead of Zn-rich particles, Al-rich particles with high relative Al content (>59%) and minimal Si (<20%) constituted 24% of metal-rich particles. Fig. 5 shows the typical particles collected on 7 February, and typical individual particles are shown in Fig. 6. The particle morphology and the proportion of each particle category of samples collected on 7 and 24 February were similar. The sampling conditions of the two days were also similar, representing the normal air pollution in Lhasa. On 7 and 24 February, Si-rich particles were the most abundant particles in the observation. They constituted 64.7% and 62.5% of total particles, respectively. Ca-rich particles were detected in 14.7% and 25% of total particles, respectively. Mine dust contributed to a high particulate mass loading, especially in Si- and Ca-rich particles, overwhelming and bringing down the metal particle percentage. Fe-rich particles constituted 17.6% of total particles on 7 February.

171

Si-rich Ca-rich Fe-rich C-rich K-rich Pb-rich Al-rich

1 μm

Fig. 5 – Particles detected on 7 February. Si-, Ca-rich and Fe-Si particles were dominant species. A few soot particles were detected. Few metal particles were detected.

Fig. 7 shows the typical particles collected on 9 February, and typical individual particles are shown in Fig. 8. The percentage of metal-rich particles collected on 9 February was 31.7%, which is much larger than those on 7 (20.5%) and 24 (0%) February. The particles collected on 9 February also contained more diverse species of other metal-rich particles. This may be largely attributed to the diffusive inhabitation of complex local emission by stable meteorological conditions. K-rich particles constituted 41.7% of the total 118 particles on 9 February, due to the biomass burning during the exorcism ritual and fireworks of New Year. In addition to abundant potassium (the dominant metal component of the firework propellant), the particles emitted by the fireworks may con tain a complex mixture of different trace metals, some of which will be present in concentrations far above their normal ranges (Moreno et al., 2007). The metal-rich particles collected on 9 February also contained a little manganese and zinc. Pband Al-rich particles constituted 8.3% and 16.7% of all 118 particles on the sample collected on 9 February. Fig. 9 shows the typical particles collected on 25 February, and typical individual particles are shown in Fig. 10. Metal-rich particles constituted a large percentage of total particles collected on 25 February, but there was a great deal of diversity in the ratios among Fe-, Al- and Pb-bearing particles on 9 and 25 February. On 25 February, Fe-rich particles were the dominant species, comprising almost all the metal particles, while Pb- and Al-rich particles constituted 79% of metal particles on 9 February. The firecrackers of the Lantern Festival may have contributed to the Fe-rich particles on 25 February. Si-. Ca-, Fe- were further classified into several categories based on their abundance and composition. Table 2 shows the classification of particles based on the EDS data. Four major categories of Si-rich particles were differentiated: high silica content particles, “Si + Al” particles, “Si + Fe” particles and “Si + Ca” particles. Ca-rich particles were divided into high calcium content particles, “Ca + Si” particles and “Ca + Fe + Si + Al” particles. Fe-rich particles were further grouped into high iron content particles (Fe > 85%) and “Fe + Si” particles.

172

J O U RN A L OF E N V I RO N ME N TA L S CI EN CE S 29 (2 0 1 5 ) 1 65–1 7 7

a

b

c

d

e

f

g

h

i

j

k l

Fig. 6 – Individual particles detected on 7 February. Their compositions were determined by EDS. Mass percentages of interested elements are presented in the square brackets. Some particles have SAED. (a) rectangular SiO2 particle; (b) rectangular palisade Si-rich particle; (c) triangular Si–Al particle; (d) irregular Si–Al particle; (e) irregular Si–Ca particle with Al; (f ) irregular Si–Fe particle with Al; (g) long striped Ca-rich particle; (h) spherical Ca–Si–Fe–Al particle; (i) irregular Fe–Si particle with Mg; ( j) irregular Fe–Si particle with Al; (k) irregular Fe–Si particle with Mn; (l) soot particle.

Si-rich Ca-rich Fe-rich C-rich K-rich Pb-rich Al-rich 1 μm

1 μm Fig. 7 – Particles detected on 9 February. K-, Al-rich and soot particles were dominant species. A few Pb- and Fe-rich particles were detected. Few earth's crust elements were detected.

The Si-rich particles constituted more than 36.8% of total observed particles. High silica content particles made up 11.7% of Si-rich particles. The particles for which Si > 65% are assumed to be mostly SiO2 (Okada et al., 2005). Their morphology displayed various shapes, such as rectangle (shown in Fig. 6a), palisade (shown in Fig. 6b), irregular shape (shown in Fig. 10a) and trapezium (shown in Fig. 10b). “Si + Al” particles were the dominant species, which constituted 65% of all Si-rich particles. The “Si + Al” particles lead to an expected range of Si/Al ratios between 1 and 4 (Cwiertny et al., 2008). They mainly come from the feldspar and clay minerals. Their morphology also displays a variety of shapes, such as triangle (shown in Fig. 6c), irregular shape (shown in Fig. 6d; Fig. 10c) and long stripe (shown in Fig. 10d). “Si + Fe” and “Si + Ca” particles constituted 16.7% and 5% of Si-rich particles, respectively. They were mostly irregular particles (shown in Fig. 6f and e; Fig. 8a; Fig. 10g), with a few spherical particles (shown in Fig. 8b). As a whole, most Si-rich particles contained silicon and aluminum, which suggests that Si-rich particles mainly came from silicate particles in the earth's crust. The Ca-rich particles constituted 15.4% of all particles in our study. The Ca-rich particles constituted 6%–9% of all particles collected near the Taklamakan desert, while the

173

J O U RN A L OF E N V I RO N M EN TA L S CI EN CE S 29 (2 0 1 5 ) 1 6 5–1 7 7

a

c

b

d

g h

f e

j

i

k

l

Fig. 8 – Individual particles detected on 9 February. (a) irregular Si–Fe particle; ( b) spherical Si–Fe particle; (c) irregular Ca-rich particle which is supposed to be calcite; (d) irregular Ca-rich particle which is supposed to be calcite; (e) spherical K-rich particle with S, implying biomass burning source; (f ) spherical Mg-rich particle with S; (g) spherical K-rich particle with Cl; ( h) spherical K-rich particle with Cl, adhering to clustered Al-rich particles and soot; (i) spherical K-rich particle with Cl, adhering to clustered Al-rich particles and soot; ( j) spherical K-rich particle with S, adhering to clustered Al-rich particles and clustered Pb-rich particle; (k) clustered Pb-rich particle with Cl, K and S; (l) Irregular Al-rich particle with K.

proportion of Ca-rich particles collected in Urumqi was 18% (Okada and Kaj, 2004). Our result was similar to the particles collected in Urumqi, indicating that not only natural sources

but also anthropogenic sources contributed to the Ca-rich particles in Lhasa. High calcium content particles, in which Ca ≥ 65%, were detected in 43.5% of Ca-rich particles. They

Si-rich Ca-rich Fe-rich C-rich K-rich Pb-rich Al-rich

1 μm

1 μm

Fig. 9 – Particles detected on 25 February. Fe-, Si- and Ca-rich particles were dominant species. A few soot and Ti-rich particles were detected.

174

J O U RN A L OF E N V I RO N ME N TA L S CI EN CE S 29 (2 0 1 5 ) 1 65–1 7 7

a

b

c

e

f

g

h

j

k

l

i

d

Fig. 10 – Individual particles detected on 25 February. (a) irregular SiO2 particle; ( b) trapezoid SiO2 particle; (c) irregular Si–Al particle; (d) long striped Si–Al particle; (e) irregular Ca-rich particle which is supposed to be calcite; (f ) spherical Ca–Si–Fe–Al particle; (g) irregular Si–Fe particle with Ti; ( h) irregular Fe oxide particle; (i) irregular Fe oxide particle; ( j) irregular zero-valent Fe particles encompassed by organic coating; (k) spherical Fe-rich particle with Al and Mg; ( l) long striped Ti-rich particle with V.

mainly came from calcite minerals in the earth's crust (Li et al., 2008). A few long-striped high-calcium-content particles were observed (shown in Fig. 6g), but most of the high-calcium-content particles were irregular (shown in Fig. 8c and d, Fig. 10e). “Ca + Si” and “Ca + Fe + Si + Al” particles constituted 34.8% and 21.7% of Ca-rich particles, respectively, and their morphologies were observed as spherical particles under the TEM (shown in Fig. 6h; Fig. 10f). They were thought to have originated from the cement fly ash from construction sites (Zhang et al., 2005; Li et al., 2008). In recent years, the urbanization of Lhasa city increased rapidly, and the construction industry consumes a large amount of cement. Thus, an anthropogenic source also contributed to the Ca-rich particles in Lhasa. The Fe-rich particles constituted 13.5% of total particles. High-iron-content particles with irregular morphology (shown in Fig. 10h, i and j) constituted 68.2% of all Fe-rich particles with very high relative Fe content (>85%) and minimal Si and Al (<10%). These data likely pertained to particles containing Fe in some oxide forms (Cwiertny et al., 2008). Steelwork emissions strongly affect the abundance of iron oxide particles, such as magnetite (Fe3O4) and hematite (α-Fe2O3), even several kilometers away from the sampling site (Choël et al., 2007). But there was no steel work source in the area surrounding the sampling site. 75.9% of all high-iron-content particles were

detected on 25 February. Persistent steelwork emissions were assumed not to exist. Particles with sphere morphology are commonly produced by a gas-to-particle transformation followed by condensational growth under a high-temperature environment (Ault et al., 2012). Shown in Fig. 10k, spherical Fe-rich particles were observed in the individual particle analysis. Furthermore, Li et al. (2013) observed a significant increase of Fe-rich particles after firework/firecracker combustion during the Chinese New Year, which is similar to our results. Therefore, the high-iron-content particles on 25 February were proposed to have been emitted from the firecracker combustion of the Lantern Festival. Silicate particles with high iron content contributed “Fe + Si” particles (Zhang et al., 2011). According to the TEM analysis, most of these were irregular particles (shown in Fig. 6i, j and k). Therefore, Fe-rich particles collected during the sampling mainly came from firework/firecracker combustion, while a few “Fe + Si” particles (Fe >45%, Si < 25%) may have come from mineral dust. All of the Pb-, K- and Al-rich particles were detected on 9 February. Fig. 8j shows clustered Pb-rich particles adhering to Al-rich particles and spherical K-rich particles. Fig. 8k shows an individual Pb-rich particle cluster. Besides the natural background, the major sources of anthropogenic Pb in the atmosphere include the combustion of leaded gasoline by automobiles, industrial discharges and coal burning (Mukai et

J O U RN A L OF E N V I RO N M EN TA L S CI EN CE S 29 (2 0 1 5 ) 1 6 5–1 7 7

al., 2001; Widory et al., 2010). But there were few mobile or industrial sources surrounding the sampling site, and no Pb-rich particles were detected on the other days. Persistent sources of lead pollution were thus concluded not to exist. Mukai et al. (2001) suggested that coal combustion considerably contributed to atmospheric lead in Chinese cities. However, the current power demands in Lhasa rely on renewable energy forms, such as solar (4.99% of the total energy consumption), biotic (37.4%) geothermal (5.39%) and water energy (28.67%), and very limited coal (5.31%) is used in Lhasa (Hua, 2009). The influence of coal burning was not considered to be a major source in the city. It has been documented that Pb was used to achieve a steady and reproducible burning rate, and increased a great deal during the burning of firecrackers (Conkling, 1985; Gao et al., 2002; Wang et al., 2007). Therefore the Pb-rich particles on 9 February were strongly suggested to be emitted by the fireworks/firecrackers in the New Year celebration. Biomass burning is a global phenomenon that is one of the major sources of airborne particulate matter, and exerts an important impact on the environment and climate on a global scale (Houghton et al., 2001; Fiedler et al., 2010; Fu et al., 2012; Fu et al., 2014). Measurements of K have been widely used as a source tracer for biomass burning emissions and have been applied in source appointment studies (Khalil and Rasmussen, 2003; Hays et al., 2005). Most K-rich particles on 9 February were spherical, as shown in Fig. 8e and g. Some K-rich particles were observed adhering to the other particles, such as Al-rich particles, Pb-rich particles and soot (shown in Fig. 8h, i and j). No K-rich particles were observed on 7, 24 and 25 February. Therefore the K-rich particles collected on 9 February originated from the biomass burning during the traditional Tibetan exorcism ritual and firework/firecracker combustion of the New Year celebration. A few K-rich particles collected on 9 February contained S, and thus were most likely K2SO4 (Li et al., 2013). In addition, we found significant amounts of Cl in K-rich particles. Wilkin et al. (2007) and Shi et al. (2011) found a high content of perchlorate particles in urban cites after firecracker/ firework activity. It was well documented that fresh K-rich particles from Chinese firecrackers generated in the laboratory contain abundant Cl but little S (Li et al., 2013). Indeed, the fresh emissions of fireworks/firecrackers contain complex chlorideaerosols such as KCl, KClO4, or other chlorine-organics (Fleischer et al., 1999; Wilkin et al., 2007 and Shi et al., 2011). However, the particles yielded by fireworks contain abundant S but little Cl after long-range transport (Li et al., 2013). Therefore, the K-rich particles collected on 9 February could be fresh particles from a local source. The individual K-rich particles or K-rich particles just adhering to soot are suggested to be contributed by torch burning during the Weisang religious ritual at Lhasa city. K-rich particles adhering to Pb- and Al-rich particles were suggested to be yielded by the firework/firecracker combustion in the New Year celebration. The most frequent metal particles observed in particles emitted by fireworks were Al-rich, Mg-rich, and Fe-rich (Li et al., 2013). We detected Al-rich particles only on 9 February. Persistent sources of lead pollution are thus concluded not to exist. Al-rich particles with high relative Al content (>59%) and minimal Si (< 20%) mostly adhering to K-rich particles and soot. Therefore, Al-rich particles collected on February 9 may

175

have been contributed by the fireworks. As a whole, Pb-, Kand Al-rich particles during the sampling mainly came from anthropogenic sources. Soot constituted 9.8% of all particles. Typical soot particles are shown in Fig. 6l; Fig. 8h and i. Through the morphological observation, soot particles were seen to be mainly clustered. Individual C-rich particles were mostly ellipsoids and spheres, which were assumed to be soot. 81.3% of soot particles were detected on 9 and 25 February. The black carbon (BC) concentrations were elevated during the burning of firecrackers/fireworks (Li et al., 2013). Therefore, fireworks and biomass burning were proposed to contribute to the soot particles observed. In addition, Mg- and Ti-rich particles were detected during the sampling. They only constituted 1.7% of all particles. Mg-rich particles (shown in Fig. 8f) may have come from trace mineral dust (Li et al., 2008). Ti-rich particles are shown in Fig. 10l. Ti is a crustal element, with frequent occurrence in mineral and fly ash (Chen et al., 2006; Murr and Bang, 2003). Ti-rich particles indexed in the form of the crystalline rutile (TiO2) have been widely reported in field measurements (Chou et al., 2008; Fu et al., 2014). During the sampling, the PM10 concentrations of Lhasa were much lower than those of other cities in east and south Asia, but higher than those in the remote region in TP like Nam Co, indicating a minor level of urban pollution. The back trajectories showed short paths for winds during the sampling, indicating local-source pollution. According to the TEM EDS analysis, most particles collected during the sampling came from mineral dust in the earth's crust, especially the Si–Al particles. During some specified periods, a celebration ceremony and religion ritual led to large-scale firework/firecracker combustion and biomass burning, which contributed to K-rich and metal-bearing particles and led to fine particle pollution. Cl-bearing particles from the fireworks/firecrackers are processed in the atmosphere via heterogeneous reactions during long-range transport, with the Cl− being replaced by secondary SO2− 4 (Li et al., 2010). Cui et al. (2008) suggested that the heterogeneous reaction rate of SO2− 4 is positively related to the temperature. The thick water layer attributed to high RH favors surface reaction, such as the heterogeneous reaction of SO2 to H2SO4 (Li et al., 2011). So the low RH (12.6%) and temperature and calm weather in Lhasa during the sampling significantly reduced heterogeneous reactions in the atmosphere. Therefore the particles from firework combustion contained much more Cl than S. Pb- and Al-rich particles were mostly clustered, adhering to K-rich particles and soot, indicating that they mainly came from the firework/ firecracker combustion. A few “Ca + Si” and “Ca + Fe + Si + Al” particles are suggested to come from cement particles. Therefore, dominating mineral dust pollution was observed in our study, with much anthropogenic pollution during some specified periods.

3. Conclusions Single particles were collected in the urban area of Lhasa. The aerosol in Lhasa was mainly contributed by local sources during the sampling. The dominant elements were Si, Al and Ca, which mainly came from mineral dust in the earth's crust like feldspar and clay. Fe-, Pb-, K-, Al-rich particles and

176

J O U RN A L OF E N V I RO N ME N TA L S CI EN CE S 29 (2 0 1 5 ) 1 65–1 7 7

soot mainly came from anthropogenic sources like firework/ firecracker combustion and biomass burning. Lhasa is a non-industrialized city with a small population of 559,400 (2012). Therefore the industrial and transport-related pollution was slight in Lhasa. During the sampling, the pollution in Lhasa mainly came from mineral dust in the earth's crust, while the celebration ceremony and religious ritual produced a large quantity of metal-bearing particles during some specified periods. The cement factory in the suburban area also exerted a minor influence. The metal compounds within individual aerosol particles may enhance the toxicity of such fine particles (Oberdörster et al., 2005) and the combinations of different metals in individual particles could exhibit a synergistic effect on human health (Dye et al., 1999). In Lhasa, firecracker/firework activities and biomass burning were not prohibited due to religious reasons. It is recommended that relevant epidemiological and toxicological studies be conducted to investigate the potential health effects of the firecracker/firework displays in order to provide scientific evidence for more strict legislation. The data obtained in this study can be useful for developing pollution control strategies in the city.

Acknowledgments Financial support was provided by the National Natural Science Foundation of China (Nos. 21177026, 21190053, 40975074), the Ministry of Education of new century talent project (NCET11-0104), the Doctoral Fund of Ministry of Education (No. 2013007111008) and the Pujiang Talent Program of Shanghai (No. PJ[2010]00317).

REFERENCES Adachi, K., Buseck, P.R., 2010. Hosted and free-floating metal-bearing atmospheric nanoparticles in Mexico City. Environ. Sci. Technol. 44 (7), 2299–2304. Ault, A.P., Peters, T.M., Sawvel, E.J., Casuccio, G.S., Willis, R.D., Norris, G.A., et al., 2012. Single-particle SEM-EDX analysis of iron-containing coarse particulate matter in an urban environment: Sources and distribution of iron within Cleveland, Ohio. Environ. Sci. Technol. 46 (8), 4331–4339. Chen, Y., Shah, N., Huggins, F.E., Huffman, G.P., 2006. Microanalysis of ambient particles from Lexington, KY, by electron microscopy. Atmos. Environ. 40 (4), 651–663. Chen, S.Y., Huang, J.P., Zhao, C., Qian, Y., Leung, L.R., Yang, B., 2013. Modeling the transport and radiative forcing of Taklimakan dust over the Tibetan Plateau: a case study in the summer of 2006. J. Geophys. Res. 118 (2), 797–812. Choël, M., Deboudt, K., Flament, P., Aimoz, L., Mériaux, X., 2007. Single-particle analysis of atmospheric aerosols at Cape Gris-Nez, English Channel: influence of steel works on iron apportionment. Atmos. Environ. 41 (13), 2820–2830. Chou, C., Formenti, P., Maille, M., Ausset, P., Helas, G., Harrison, M., et al., 2008. Size distribution, shape, and composition of mineral dust aerosols collected during the African Monsoon Multidisciplinary Analysis Special Observation Period 0: dust and biomass-burning experiment field campaign in Niger, January 2006. J. Geophys. Res. Atmos. 113 (D23). http://dx.doi. org/10.1029/2008JD009897 (1984-2012).

Chueinta, W., Hopke, P.K., Paatero, P., 2000. Investigation of sources of atmospheric aerosol at urban and suburban residential areas in Thailand by positive matrix factorization. Atmos. Environ. 34 (20), 3319–3329. Cong, Z.Y., Kang, S.C., Liu, X.D., Wang, G.F., 2007. Elemental composition of aerosol in the Nam Co region, Tibetan Plateau, during summer monsoon season. Atmos. Environ. 41 (6), 1180–11187. Cong, Z.Y., Kang, S.C., Smirnov, A., Holben, B., 2009. Aerosol optical properties at Nam Co, a remote site in central Tibetan Plateau. Atmos. Res. 92 (1), 42–48. Cong, Z.Y., Kang, S.C., Luo, C.L., Li, Q., Huang, J., Gao, S.P., et al., 2011. Trace elements and lead isotopic composition of PM10 in Lhasa, Tibet. Atmos. Environ. 45 (34), 6210–6215. Cong, Z.Y., Kang, S.C., Gao, S.P., Zhang, Y.L., Li, Q., Kawarmura, K., 2013. Historical trends of atmospheric black carbon on Tibetan Plateau as reconstructed from a 150-year lake sediment record. Environ. Sci. Technol. 47 (6), 2579–2586. Conkling, J.A., 1985. Chemistry of Pyrotechnics: Basic Principles and Theory. Marcel Dekker, Inc., New York. Cui, H.X., Cheng, T.T., Chen, J.M., Xu, Y.F., Fang, W., 2008. A simulated heterogeneous reaction of SO2 on the surface of hematite at different temperatures. Acta Phys. -Chim. Sin. 24 (12), 2331–2336. Cwiertny, D.M., Baltrusaitis, J., Hunter, G.J., Laskin, A., Scherer, M.M., Grassian, V.H., 2008. Characterization and acid-mobilization study of iron-containing mineral dust source materials. J. Geophys. Res. 113, D05202. http://dx.doi. org/10.1029/2007JD009332. Draxler, R.R., Hess, G.D., 1998. An overview of the HYsplit-4 modeling system for trajectories, dispersion, and deposition. Aust. Meteorol. Mag. 47, 295–308. Dye, J.A., Adler, K.B., Richards, J.H., Dreher, K.L., 1999. Role of soluble metals in oil fly ash-induced airway epithelial injury and cytokine gene expression. Am. J. Physiol. Lung Cell. Mol. Physiol. 277 (3Pt1), L498–L510. Ferrat, M., Langmann, B., Cui, X., Gomes, J., Weiss, D.J., 2013. Numerical simulations of dust fluxes to the eastern Qinghai–Tibetan Plateau: comparison of model results with a Holocene peat record of dust deposition. J. Geophys. Res. 118 (10), 4597–4609. http://dx.doi.org/10.1002/jgrd.50275. Fiedler, V., Arnold, F., Ludmann, S., Minikin, A., Pirjola, L., Dornbrack, A., et al., 2010. African biomass burning plumes over the Atlantic: aircraft based measurements and implications for H2SO4 and HNO3 mediated smoke particle activation. Atmos. Chem. Phys. Discuss. 10, 7699–7743. Fleischer, O., Wichmann, H., Lorenz, W., 1999. Release of polychlorinated dibenzo-p-dioxins and dibenzofurans by setting off fireworks. Chemosphere 39 (6), 925–932. Fu, H., Zhang, M., Li, W., Chen, J., Wang, L., Quan, X., et al., 2012. Morphology, composition and mixing state of individual carbonaceous aerosol in urban Shanghai. Atmos. Chem. Phys. 12 (2), 693–707. Fu, H.B., Shang, G.F., Lin, J., Hu, Y.J., Guo, L., Zhang, Y.C., et al., 2014. Fractional iron solubility of aerosol particles enhanced by biomass burning and ship emission in Shanghai, East China. Sci. Total Environ. 481, 377–391. Gao, Y., Nelson, E.D., Field, M.P., Ding, Q., Li, H., Sherrell, R.M., et al., 2002. Characterization of atmospheric trace elements on PM2.5 particulate matter over the New York–New Jersey harbor estuary. Atmos. Environ. 36, 1077–1086. Hays, G.C., Richardson, A.J., Robinson, C., 2005. Climate change and marine plankton. Trends Ecol. Evol. 20 (6), 337–344. Houghton, J.T., Ding, Y., Griggs, D., 2001. Climate Change: The Intergovernmental Panel on Climate Change Scientific Assessment. Cambridge University Press, IPCC, Cambridge, UK, p. 7701. Hua, H.R., 2009. Energy research in Lhasa region. Energy Research and Utilization. 1 pp. 30–32.

J O U RN A L OF E N V I RO N M EN TA L S CI EN CE S 29 (2 0 1 5 ) 1 6 5–1 7 7

Huang, Y.L., Yang, X., 2013. Influence of fine particulate matter on atmospheric visibility. Chin. Sci. Bull. 58 (13), 1165–1170. Kang, S.C., Zhang, Y.L., Zhang, Y.J., Grigholm, B., Kaspari, S., Qin, D.H., et al., 2010. Variability of atmospheric dust loading over the central Tibetan Plateau based on ice core glaciochemistry. Atmos. Environ. 44 (25), 2980–2989. Khalil, M.A.K., Rasmussen, R.A., 2003. Tracers of wood smoke. Atmos. Environ. 37, 1211–1222. Kopacz, M., Mauzerall, D.L., Wang, J., Leibensperger, E.M., Henze, D.K., Singh, K., 2011. Origin and radiative forcing of black carbon transported to the Himalayas and Tibetan Plateau. Atmos. Chem. Phys. 11 (6), 2837–2852. Lau, K.M., Kim, M.K., Kim, K.M., 2006. Asian summer monsoon anomalies induced by aerosol direct forcing: the role of the Tibetan Plateau. Clim. Dyn. 26 (7–8), 855–864. Li, W.J., Shao, L.Y., 2009. Transmission electron microscopy study of aerosol particles from the brown hazes in northern China. J. Geophys. Res. 114, D09302. http://dx.doi.org/10.1029/ 2008JD011285. Li, W.J., Shao, L.Y., Shi, Z.B., Li, J.J., Yang, S.S., 2008. Physical and chemical characteristics of individual mineral particles in an urban fog episode. China Environ. Sci. 29 (1), 253–258. Li, W.J., Shao, L.Y., Buseck, P.R., 2010. Haze types in Beijing and the influence of agricultural biomass burning. Atmos. Chem. Phys. 10 (17), 8119–8130. Li, W.J., Zhou, S.Z., Wang, X.F., Xu, Z., Yuan, C., Yu, Y.C., et al., 2011. Integrated evaluation of aerosols from regional brown hazes over northern China in winter: concentrations, sources, transformation, and mixing states. J. Geophys. Res. Atmos. 116. http://dx.doi.org/10.1029/2010JD015099 (1984–2012). Li, C.L., Kang, S.C., Chen, P.F., Zhang, Q.G., Fang, G.C., 2012. Characterizations of particle-bound trace metals and polycyclic aromatic hydrocarbons (PAHs) within Tibetan tents of south Tibetan Plateau, China. Environ. Sci. Pollut. Res. 19 (5), 1620–1628. Li, W.J., Shi, Z.B., Yan, C., Yang, L.X., Dong, C., Wang, W.X., 2013. Individual metal-bearing particles in a regional haze caused by firecracker and firework emissions. Sci. Total Environ. 443, 464–469. Liu, Z., Liu, D., Huang, J., Vaughan, M., Uno, I., Sugimoto, N., Kittaka, C., et al., 2008. Airborne dust distributions over the Tibetan Plateau and surrounding areas derived from the first year of CALIPSO lidar observations. Atmos. Chem. Phys. 8 (16), 5045–5060. Moreno, T., Querol, X., Alastuey, A., Minguillon, M.C., Pey, J., Rodriguez, S., et al., 2007. Recreational atmospheric pollution episodes: inhalable metalliferous particles from firework displays. Atmos. Environ. 41 (5), 913–922. Mukai, H., Tanaka, A., Fujii, T., Zeng, Y.Q., Hong, Y.T., Tang, J., et al., 2001. Regional characteristics of sulfur and lead isotope ratios in the atmosphere at several Chinese urban sites. Environ. Sci. Technol. 35 (6), 1064–1071. Murr, L.E., Bang, J.J., 2003. Electron microscope comparisons of fine and ultra-fine carbonaceous and non-carbonaceous, airborne particulates. Atmos. Environ. 37 (34), 4795–4806. Nriagu, J.O., Pacyna, J.M., 1988. Quantitative assessment of worldwide contamination of air, water and soils by trace metals. Nature 333 (6169), 134–139. Oberdörster, G., Oberdorster, E., Oberdorster, J., 2005. Nanotoxicology: an emerging discipline evolving from studies of ultrafine particles. Environ. Health Perspect. 113 (7), 823–839. Okada, K., Kaj, K., 2004. Atmospheric mineral particles collected at Qira in the Taklamakan Desert, China. Atmos. Environ. 38 (40), 6927–6935.

177

Okada, K., Qin, Y., Kai, K., 2005. Elemental composition and mixing properties of atmospheric mineral particles collected in Hohhot, China. Atmos. Res. 73 (1–2), 45–67. Qian, Y., Glanner, M.G., Leung, L.R., Wang, W., 2011. Sensitivity studies on the impacts of Tibetan Plateau snowpack pollution on the Asian hydrological cycle and monsoon climate. Atmos. Chem. Phys. 11 (5), 1929–1948. Rogge, W.F., Hiddlemann, L.M., Mazurek, M.A., Cass, G.R., Simoneit, B.R.T., 1993. Sources of fine organic aerosol 2. Noncatalyst and catalyst-equipped automobile and heavy-duty diesel trucks. Environ. Sci. Technol. 27, 636–651. Shi, Y.L., Zhang, N., Gao, J.M., Li, X., Cai, Y.Q., 2011. Effect of fireworks display on perchlorate in air aerosols during the Spring Festival. Atmos. Environ. 45 (6), 1323–1327. Wang, Y., Zhuang, G.S., Xu, C., An, Z.S., 2007. The air pollution caused by the burning of fireworks during the lantern festival in Beijing. Atmos. Environ. 41 (2), 417–431. Widory, D., Liu, X., Dong, S., 2010. Isotopes as tracers of sources of lead and strontium in aerosols (TSP & PM2.5) in Beijing. Atmos. Environ. 44, 3679–3687. Wilkin, R.T., Fine, D.D., Burnett, N.G., 2007. Perchlorate behavior in a municipal lake following fireworks displays. Environ. Sci. Technol. 41 (11), 3966–3971. Xia, X.G., Wang, P.C., Wang, Y.S., Li, Z.Q., Xin, J.J., Liu, J., et al., 2008. Aerosol optical depth over the Tibetan Plateau and its relation to aerosols over the Taklimakan Desert. Geophys. Res. Lett. 35, L16804. http://dx.doi.org/10.1029/2008GL034981. Xu, B., Cao, J., Hansen, J., Yao, T., Joswia, D.R., Wang, N., et al., 2009a. Black soot and the survival of Tibetan glaciers. Proc. Natl. Acad. Sci. U. S. A. 106 (52), 22114–22118. Xu, B.Q., Wang, M., Joswiak, D.R., Cao, J.J., Yao, T.D., Wu, G.J., et al., 2009b. Deposition of anthropogenic aerosols in a southeastern Tibetan glacier. J. Geophys. Res. 114, D17209. http://dx.doi.org/ 10.1029/2008JD011510. Yang, K., Ding, B.H., Qin, J., Tang, W.J., Liu, N., Lin, C.G., 2012. Can aerosol loading explain the solar dimming over the Tibetan Plateau? Geophys. Res. Lett. 39, L20710. http://dx.doi.org/10. 1029/2012GL053733. You, Q.L., Lorenzo-Sanchez, A., Wild, M., Folini, D., Fraedrich, K., Ren, G., et al., 2013. Decadal variation of surface solar radiation in the Tibetan Plateau from observations, reanalysis and model simulations. Clim. Dyn. 40 (7–8), 2073–2086. Zhang, D.Z., Iwasaka, Y., Shi, G., 2001a. Soot particles and their impacts on the mass cycle in the Tibetan atmosphere. Atmos. Environ. 35 (34), 5883–5894. Zhang, X.Y., Arimoto, R., Cao, J.J., An, Z.S., Wang, D., 2001b. Atmospheric dust aerosol over the Tibetan Plateau. J. Geophys. Res. 106 (D16), 18471–18476. Zhang, D.Z., Shi, G.Y., Iwasaka, Y., Hu, M., Zang, J.Y., 2005. Anthropogenic calcium particles observed in Beijing and Qingdao, China. Water Air Soil Pollut. Focus 5 (3–6), 261–276. Zhang, X.L., Wu, G.J., Yu, Y.H., Zhang, C.L., 2011. Mineral composition and morphology of individual dust fall particulates over Lhasa in summer. Acta Petrol. Mineral. 30 (1), 127–134. Zhang, N.N., Cao, J.J., Xu, H.M., Zhu, C.S., 2013. Elemental compositions of PM2.5 and TSP in Lijiang, southeastern edge of Tibetan Plateau during pre-monsoon period. Particuology 11 (1), 63–69.