Aerosol and Air Quality Research, 13: 308–323, 2013 Copyright © Taiwan Association for Aerosol Research ISSN: 1680-8584 print / 2071-1409 online doi: 10.4209/aaqr.2010.04.0025

Physicochemical Characteristics and Source Apportionment of Atmospheric Aerosol Particles in Kinmen-Xiamen Airshed

Tsung-Chang Li1, Wei-Hsiang Chen1, Chung-Shin Yuan1,2*, Shui-Ping Wu2, Xin-Hong Wang2

1 Institute of Environmental Engineering, National Sun Yat-sen University, Kaohsiung 80424, Taiwan 2 State Key Laboratory of Marine Environmental Science, Xiamen University, Xiamen 361001, China

ABSTRACT

The objective of this study was to characterize the chemical properties of atmospheric particles sampled in the Kinmen- Xiamen Airshed located on the west bank of the Taiwan Strait. Seven particulate matter (PM) sampling sites in the Kinmen- Xiamen Airshed, including three sites at Kinmen Island and four in urban Xiamen, were selected for this particular study. Regular sampling was conducted to collect PM10 with high-volume samplers twice a month from March 2008, while intensive sampling was conducted to collect PM2.5 and PM2.5–10 with dichotomous samplers and PM10 with high-volume samplers in the spring and winter of 2008–2009. After sampling, the metallic contents of PM10 were analyzed with an inductively coupled plasma-atomic emission spectrometer (ICP-AES). Ionic species and carbonaceous contents of PM10 were analyzed with an ion chromatograph (IC) and elemental analyzer (EA), respectively. Finally, the source identification and apportionment of PM were analyzed by principal component analysis (PCA) and receptor modeling (CMB), respectively. The results from PM10 sampling indicated that atmospheric aerosol particles had a tendency to accumulate in Xiamen Bay all year round, particularly in spring and winter. The five sampling sites at the center of Xiamen Bay had relatively higher PM10 concentrations than the two sampling sites outside Xiamen Bay, suggesting that local emissions from Xiamen Bay were more significant than emissions transported over a long distance by the Northeastern Monsoon. The phenomenon of superimposition was regularly observed during air pollution episodes at Xiamen Bay. Moreover, the results of 2– – + chemical analysis showed that the main chemical components of the PM were SO4 , NO3 , NH4 , OC, and EC and crustal elements (Ca, Mg, Fe, and Al) in the aerosol particles in the Kinmen-Xiamen Airshed. The neutralization ratios (NR) of PM were generally smaller than unity, indicating that the atmospheric particulates were mostly acidic. The averaged sulfur oxidation ratio (SOR) ranged from 0.20 to 0.51, and the nitrogen oxidation ratio (NOR) ranged from 0.10 to 0.41 for all seasons. The ratios of sulfur and nitrogen oxidation were generally higher than 0.25 and 0.10, respectively, suggesting that secondary sulfate and nitrate aerosols came mainly from across-boundary transportation and could be further accumulated in the Kinmen-Xiamen Airshed. The results from CMB receptor modeling showed that the major sources of atmospheric PM10 in the Kinmen-Xiamen Airshed were soil dusts, secondary aerosols, the petroleum industry, motor vehicle exhausts, the iron and steel industry, the cement industry, diesel vehicle exhausts, marine aerosols, and biomass burning.

Keywords: Kinmen-Xiamen Airshed; Tempospatial distribution; Physicochemical properties; Source apportionment.

INTRODUCTION 150.0 km2, respectively. There are huge industrial emissions in metro Xiamen, including these three coal-fired power Xiamen Bay has subtropical monsoon weather with an plants, stone processors, ceramic industry, porcelain products, annual average temperature of about 21°C, rainfall of clothing manufacturers which are next to the coastal region 1,043 mm, and prevailing winds from the northeastern at Xiamen Bay. There are no large-scale industrial emission monsoon (from October to April of next year) and the sources on Kinmen Island. According to source emission southwestern monsoon (from May to August). Xiamen and data of 2009, the stationary sources in urban Xiamen were Kinmen Islands located in Xiamen Bay are a scenic coastal approximately twenty times higher than those on Kinmen cities in south-eastern China, covering areas of 132.5 and Island. Particulate matter with aerodynamic diameter (dpa) ≤ 10 μm (PM10), particularly the fine particle fraction (dpa ≤ 2.5 μm), has been shown to be associated with health problems, * Corresponding author. Tel.: 886-7-5252000 ext. 4409; such as asthma (Anderson et al., 1992; Dockery et al., 1993; Fax: 886-7-5254449 Dockery and Pope, 1994; Pope et al., 2002). Previous E-mail address: [email protected] studies reported that the physicochemical characteristics of

Li et al., Aerosol and Air Quality Research, 13: 308–323, 2013 309 atmospheric aerosols correlate closely with ambient air (118°02′N, 24°22′E) and Jinjing Elementary School (A4) quality, atmospheric visibility reduction. (Yuan et al., 2002; (118°36′N, 24°34′E) are located outside Xiamen Bay. The Lee et al., 2005; Yuan et al., 2006), and human health. Our other five sampling sites at the main campus of Xiamen recent studies indicated that the main chemical composition University (A2) (118°05′N, 24°26′E), Daderng High School of PM10 are secondary aerosols in water-soluble ionic (A3) (118°19′N, 24°33′E), Lieyu Junior High School (B1) species, crustal elements in metallic contents, and carbons (118°14′N, 24°25′E), Jinding Elementary School (B2) in Xiamen City in winter and spring. Ammonium sulfate (118°20′N, 24°26′E), and Jinsha Elementary School (B3) and nitrate are generally the major common components of (118°24′N, 24°29′E) are located inside Xiamen Bay. secondary aerosols in the atmosphere, converting from sulfur Sampling location, altitude, and surrounding environment of dioxide (SO2) and nitrogen oxides (NOx), respectively. The each site are summarized in Table 1. Sampling sites A2 and sulfur oxidation ratio (SOR) expresses the degree of A4 located in the downtown Xiamen and Jinjing, are next to oxidation of sulfur in terms of the ratio of sulfate to total a street and likely to be influenced by direct emissions from sulfur (sulfate plus SO2). Similarly, the nitrogen oxidation vehiclular exhausts and textile industries. Site A1 located ratio (NOR) expresses the degree of oxidation of nitrogen in the Zhangzhou campus of Xiamen University at the foot in terms of the ratio of nitrate to total nitrogen (nitrate plus of hills has a rural area characters. Air pollutants from NOx). High SOR and NOR suggest that photochemical Zhangzhou Harbor, Xiamen Harbor, Songyu power plant and oxidation tends to form secondary aerosols in the atmosphere the Xiamen metropolitan area would influence the ambient (Colbeck and Harrison, 1984; Ohta and Okita, 1990). air quality at site A1 during the northeastern monsoon periods, In recent years, the ambient air quality of the Kinmen- whereas air pollutants emitted from Houshi power plant Xiamen Airshed has deteriorated gradually, and PM10 is could be transported to the downwind sites. Site A3 located responsible for the poor air quality in spring and winter. at an island of Dadeng in northern Xiamen is a tourism High percentages of poor air quality (Pollutants Standard township with a cargo harbon. Kinmen Islands is recognized Index, PSI > 100), ranging from 5.5 to 14.0% at Kinmen as a national park and its local emission sources are well Island in the years 2002–2008, show that it had the worst air controlled. Thus, sites B1, B2, and B3 in Kinmen Islands quality among seven Air Quality Zones (AQZ) in Taiwan. are more likely to be influenced by regional air pollution, PSI is the abbreviation of Pollutant Standards Index based especially from the upwind northern and northeastern on five criteria air pollutants: PM10, sulfur dioxide, carbon industrial areas in Jinjiang River and Jinjing during the monoxide, nitrogen dioxide, and ozone. For each air pollutant, northeastern monsoon seasons. a value of 100 is assigned to be the maximum permitted Regular and intensive PM sampling were conducted concentration of that air pollutant. After determining the from March 2008 to February 2009. Regular sampling was sub-index value of each air pollutant, the highest sub-index conducted to collect PM10 with a high-volume sampler for of the five air pollutants is reported as the Pollutant Standards 24 hours at each site twice a month from March 2008, while 3 Index (PSI) of the day. High levels of PM10 (> 125 mg/m ) intensive sampling was conducted to collect 24-hr PM2.5 found in Kinmen Islands during the northeast monsoon and PM2.5–10 with a dichotomous samplers at the Jinding seasons are likely blow from the upwind emission sources. Elementary School site (B2) and 24-hr PM10 with a high- Similar seasonal variations of low PM10 levels from June to volume sampler at each site on January 6–10 and March August and high PM10 levels from October to March of 16–20, 2009. Ambient particles with aerodynamic diameters next year, in the Xiamen Bay. Moreover, more than 700 below 10 μm (PM10) were divided into two separate fractions ceramic and title factories mainly with coal-fired furnaces (i.e., PM2.5 and PM2.5–10) using a virtual impactor with a 10 in Jinjiang City located at the northeastern region of the μm cutpoint at the inlet of the sampler. These two fractions Xiamen Bay, where highly influenced the air quality of were classified as fine (PM2.5) and coarse (PM2.5–10) particles. Xiamen Bay under the northeastern winds (Wu, 2011). In this study the space analysis software (SURFER) was Consequently, this study investigated the tempospatial used to assess the PM10 hot spots in Kinmen-Xiamen distribution of PM10, including the mass concentration and Airshed and to draw the concentration contours of each physicochemical properties in the Kinmen-Xiaman Airshed. month from 2008 to 2009. The SURFER model has been The source identification and apportionment of PM10 were commonly used to describe the spatial distribution of PM10 further analyzed by principal component analysis (PCA) for metropolitan district, industrial area, rural area, etc. and receptor modeling (CMB). The SURFER model is (Otvos et al., 2003; Shaocai et al., 2004; Tsai et al., 2010). seldom used in combination with the CMB model to look for PM10 hot spots and estimate the contribution of various Chemical Analysis sources, and this is the aim of the current study. After sampling, quartz filters were temporarily stored at 4°C and then transported back to the Air Pollution EXPERIMENTAL METHODS Laboratory in the Institute of Environmental Engineering at National Sun Yat-Sen University for further conditioning, Sampling Protocol weighing, and chemical analysis. All PM10 sampling filters The locations of the seven PM10 sampling sites are collected by high-volume samplers were analyzed for illustrated in Fig. 1, including three sites on Kinmen Island chemical composition. One part of the quartz filter was and four sites in urban Xiamen. Among these, two sampling analyzed for water-soluble ionic species. The filter analyzed sites at Zhangzhou campus of Xiamen University (A1) for ionic species was put inside a 15-mL PE bottle for each

310 Li et al., Aerosol and Air Quality Research, 13: 308–323, 2013

P6

P7 P5 P8 Metro P4 Xiamen Xiamen Bay Kinmen ●P2 Island Taiwan Strait ●P3

●P1 ◆Sampling sites ●Power plants

Fig. 1. Location of PM10 sampling sites and major industrial complexes surrounding the Kinmen-Xiamen Airshed. (P1–P3: power plant; P1: Houshi power plant; P2: Songyu power plant; P3: Tashan power plant; P4: petroleum refinery; P5: petrochemical and mechanical industries; P6: light industries; P7: ceramics and stone process industries; P8: textile industry).

Table 1. Sampling location and environmental description for seven sampling sites. Site Sampling location Latitude Longitude Altitude (m) Site description Zhangzhou campus of A1 24°22'48'' 118°02'28'' 45 Rural area near hills Xiamen University Main campus of Xiamen Busy traffic and compact A2 24°26'08'' 118°05'25'' 21 University residential district Township with a chargo A3 Daderng High School 24°33'33'' 118°19'49'' 14 harbor close to farmland Jinjing Elementary Residential, factory and traffic A4 24°34'34'' 118°36'10'' 18 School mixture area Lieyu Junior High B1 24°25'50'' 118°14'30'' 34 Open area and naked field School Jinding Elementary B2 24°26'53'' 118°20'14'' 30 Farmland and open area School Jinsha Elementary Township surrounding by B3 24°29'19'' 118°24'43'' 28 School open area

sample. Distilled de-ionized water (D.I. H2O) was added Coupled Plasma-Atomic Emission Spectrometer, ICP-AES into each bottle and vibrated in an ultrasonic process for (Perkin Elmer, Model Optima 2000DV). approximately 60 mins. An Ion Chromatographer (DIONEX, Two parts of quartz filters were further used to measure Model DX-120) was used to analyze the concentration of the carbonaceous contents of the PM. The carbonaceous – – 2– – + major anions (F , Cl , SO4 , and NO3 ) and cations (NH4 , contents, including elemental, organic, and total carbons K+, Na+, Ca2+ and Mg2+). (OC, EC, and TC), were measured with an Elemental Another part of the quartz filter analyzed for metals was Analyzer (Carlo Erba, Model 1108). Before sampling, initially digested in a 20 mL mixed acid solution (HNO3: quartz filters were pre-heated to 900°C for 1.5 hr to remove HClO3 = 3:7) at 150–200°C for 2 hrs, and then diluted to potential carbon impurities from the filters. The preheating 25 ml with distilled de-ionized water (D.I. H2O). During procedure could minimize the background carbon in the the digestion, D.I. H2O was added to the residual solution quartz filters and matrix, which might cause interference two or more times in order to eliminate the acid content of with the analytical results, leading to an overestimation of the digestion solution. The metallic species of the PM, the carbonaceous content of the PM. The Elemental Analyzer including Na, Ca, Al, Fe, Mg, K, Zn, Cr, Ti, Mn, Ba, Sr, Ni, (EA) was operated using the procedure of oxidation at Pb, and Cu, were then analyzed with an Inductively 1020°C and reduction at 500°C, with continuous heating for

Li et al., Aerosol and Air Quality Research, 13: 308–323, 2013 311

15 mins. Additionally, one-eighth of the quartz filter was solution expresses each receptor concentration of a chemical heated in advance using nitrogen gas at 340–345°C for at species as a linear summation of the products of source least 30 mins to expel the organic carbon (OC) fraction, after profiles and source contributions. Source profiles (the which the amount of elemental carbon (EC) was determined. fractional amount of each species in the emissions from each Another one-eighth of the quartz filter was analyzed without source type) and receptor concentrations, each with realistic heating to determine total carbon (TC). The amount of OC uncertainty estimates, serve as input data to the CMB receptor was then estimated by subtracting EC from TC. model. The model output consists of the contribution from each source type to the total ambient aerosol mass, as well Quality Assurance and Quality Control as to individual chemical species concentration. The CMB8 The quality assurance and quality control (QA/QC) for model results are evaluated by using several fit indices, both PM sampling and chemical analysis were conducted such as R2 (≥ 0.8), χ2 (≤ 4.0), T statistics (≥ 2.0), and in this study. Prior to conducting PM sampling, the flow percentage of mass accounted for 0.8–1.2. The source rate of each PM sampler was carefully calibrated with a profiles used in this study were reported by USEPA, film flowmeter (MCH-01 SENSIDYNE Inc.). Quartz filters Southern California Air Quality Study, and the researches were then carefully handled and placed on the PM10 were studied the chemical composition and source profile in samplers to prevent potential cracking during the sampling Taiwan. Table 2 summarizes the source profiles which were procedure. After sampling, aluminum foil was used to fold used in this case of PM10 in the Kinmen-Xiamen Airshed. the quartz filters, which were then temporarily stored at 4°C and transported back to the central laboratory for chemical Chemical Transformation of SO2 and NOx analysis. The sampling and analytical procedure was similar Sulfate and nitrate are the major components contained to that described in various previous studies (Cheng and in the atmospheric aerosols in urban areas. To determine 2– Tsai, 2000; Lin, 2002; Yuan et al., 2006; Tsai et al., 2008; the degree of atmospheric transformation of SO2 to SO4 – 2010, 2011). Both field and transportation blanks were and NOx to NO3 , the sulfur and nitrogen oxidation ratios used for PM sampling, while reagent and filter blanks were (i.e., SOR and NOR) were employed, and these are defined used for chemical analysis. The determination coefficient as follows (Colbeck and Harrison, 1984; Ohta and Okita, (R2) of the calibration curve for each chemical species was 1990; Kaneyasu et al., 1999): required to be higher than 0.995. Background contamination was routinely monitored by using operational blanks SOR = Snss-SO4/(Snss-SO4 + SSO2) (1) (unexposed filters), that were proceeded simultaneously with field samples. The background interference was insignificant NOR = NNO3/(NNO3 + NNO2) (2) in this study, and can thus be ignored. At least 10% of the samples were analyzed by spiking with a known amount of where nss-SO4 is the excess sulfate that was calculated by 2– metallic and ionic species to calculate their recovery subtracting the amount of SO4 of marine from that of 2– efficiencies. SO4 in the atmosphere (Kaneyasu et al., 1995; Cheng et al., 2000). The units of nss-SO4, SSO2, NNO3, and NNO3- are 3 Principal Component Analysis (PCA) and Chemical Mass neq/m . The average concentrations of SO2 and NOx during Balanced (CMB) Receptor Model each sampling period were obtained from the ambient air The concentrations of ionic species, metallic elements, quality monitoring station in Kinmen Island. and carbonaceous contents for PM10 samples were used to calculate varimax rotated principal component analysis to RESULTS AND DISCUSSION identify the number of principal components having eigenvalues > 1.0 (Tandon et al., 2008; Deshmukh et al., PM Concentration and Size Distribution 2011). The source apportionment of ambient PM10 was As illustrated in Table 3 and Fig. 2, regular sampling of assessed using a receptor model based on the chemical PM10 concentrations at seven sampling sites at Xiamen Bay mass balance (CMB) (Ke et al., 2007; Kothai et al., 2008; was conducted from March 2008 to February 2009. The Wang et al., 2008; Yatkin and Bayram, 2008). Since the average concentration of 24-hr PM10 varied from 43.5 to detailed descriptions of CMB receptor model (e.g., CMB8) 148.53 μg/m3, with an average of 91.1 ± 25.9 μg/m3 in spring; are available elsewhere, only a brief summary is presented from 30.8 to 83.2 μg/m3, with an average of 48.5 ± 14.6 below. μg/m3 in summer; from 42.0 to 90.7 μg/m3, with an average The CMB receptor model uses the emission profiles of of 66.8 ± 14.7 μg/m3 in fall; and from 45.49 to 136.4 μg/m3, prominent sources to estimate their contribution to a with an average of 93.5 ± 45.9 μg/m3 in winter. specific receptor. It is assumed that the total concentration Among the seven sampling sites, the average of a particular chemical species at the receptor is the linear concentrations of 24-hr PM10 frequently violated the ambient summation of each individual contribution from various air quality standard of 125 μg/m3 in winter. The highest sources. The CMB receptor model uses the results of the PM10 concentrations mostly occurred in winter, while the least-square regression analysis of the aerosol chemical lowest generally occurred in summer. The PM10 concentration composition to estimate the most appropriate contributions contour shows that the highest PM10 concentration was of source apportionment. Therefore, this model consists of observed along the northern coast of Xiamen Bay. The results a least-square solution to a set of linear equations. This from PM10 sampling indicate that PM10 had a tendency to

312 Li et al., Aerosol and Air Quality Research, 13: 308–323, 2013

Table 2. Source Profiles of PM10 were used in chemical mass balance. Code Source Profile Research SCT004 PBPRI1 Petroleum cracking Plant U.S EPA. 1991 SCT007 PP004 Industrial Boilers (Oil) Cheng et al., 2001 SCT008 PP005 Industrial Boilers (Coal) Cheng et al., 2001 SCT009 PETRO1 Petroleum Industry U.S EPA. 1991 SCT010 STEEL1 Steel Industry Chiang et al., 1993 SCT011 STEEL2 Coke Plant Chiang et al., 1993 SCT012 STEEL3 Sinter Plant Chiang et al., 1993 SCT013 STEEL4 Electric Arc Furnace Yuan et al., 2003 SCT020 CEMENT Cement Industry Chiang et al., 1993 SCT023 VEHICLE2 Vehicular Exhausts J.C Chow. 1991 SCT024 VEHICLE3 Diesel Exhausts J.C Chow. 1991 SCT025 DUST1 Paved Road dust in South Taiwan Cheng et al., 1998 SCT026 DUST2 Paved Road dust in Central Taiwan Cheng et al., 1998 SCT027 DUST3 Paved Road dust in South Taiwan Yuan et al., 1991 SCT028 DUST4 Paved Road dust in Central Taiwan Chiang et al., 1993 SCT029 DUST5 Unpaved Road dust in Central Taiwan Chiang et al., 1993 SCT030 SOIL1 Dust U.S EPA. 1991 SCT033 MARIN1 Marin in Central Taiwan Cheng et al., 1998 SCT034 MARIN2 Marin in South Taiwan Chen et al., 1998 SCT035 VB001 Biomass Burning Cheng et al., 1999 SCT036 SO4 Secondary Sulfate Wang et al., 2006 SCT037 NO2 Secondary Nnitrate Wang et al., 2006 The source profiles used in this study were mainly obtained from the researcher’s finding of the chemical composition of PM10 emitted from various emission sources. Only limited source profiles are referred from USEPA and Southern California Air Quality Study, and local emission source profiles.

Table 3. The mass concentration for different fraction of particles in the Kinmen and Xiamen region. 3 Sampling Perios Seasons n Sampling Dates PM10 (μg/m ) 2008.03.05; 2008.03.20 Spring 41 2008.04.07; 2008.04.23 92.1 ± 33.6 (March, April, and May) 2008.05.07; 2008.05.23 2008.06.05; 2008.06.24 Summer 42 2008.07.10; 2008.07.20 48.5 ± 22.8 (June, July, and August) 2008.08.05; 2008.08.20 Regular Sampling Fall 2008.09.04; 2008.09.20 (September, October, and 42 2008.10.05; 2008.10.20 66.1 ± 21.0 November) 2008.11.06; 2008.11.20 Winter 2008.12.04; 2008.12.20 (December, January, and 42 2009.01.06; 2009.01.14 92.0 ± 50.1 February) 2009.02.20; 2009.02.28 21 2008.03.05–2008.03.07 88.8 ± 33.6 Intensive Sampling Continuous sampling 35 2009.01.06–2009.01.10 71.3 ± 21.6 35 2009.03.16–2009.03.20 111.9 ± 56.1 accumulate in Xiamen Bay all year round, particularly in during air pollution episodes at Xiamen Bay. Local emissions spring and winter. The five sampling sites (A2, A3, B1, B2, from the Xiamen Bay were thus more significant than and B3) located at the center Xiamen Bay had relatively emissions transported over a long distance by the Northeastern higher PM10 concentration than the other two sampling sites Monsoons. (A1 and A4) outside Xiamen Bay. The PM10 concentrations During the intensive sampling periods, fine (PM2.5) and at the upwind sites, including site A4, were not much coarse (PM2.5–10) particles were simultaneously sampled at higher than those at the downwind site A1 during the site B2 with a dichotomous sampler. The concentrations of Northeastern Monsoons in spring and winter. The highest fine and coarse particles and their mass ratios (PM2.5/ PM10 concentrations occurred at site A3 in the Kinmen- PM2.5–10) are summarized in Table 4. During the first Xiamen Airshed, especially in December. The results suggest intensive sampling period in winter, the highest PM10 and 3 that a superimposition phenomenon was regularly observed PM2.5 concentrations were 115.2 and 56.0 μg/m , respectively.

Li et al., Aerosol and Air Quality Research, 13: 308–323, 2013 313

(1) Spring 400 350

300 2005 250 2006 200 2007 150 2008 Wind Direction Wind 100

50

0 3/1 3/15 3/29 4/12 4/26 5/10 5/24 (2) Summer 400 350 300

n 2005 250 2006 200 2007 150 2008 Wind Directio 100

50 0 6/1 6/15 6/29 7/13 7/27 8/10 8/24 (3) Fall 400 350

300

250 2004 2005 200 2006 150 2007 Wind Direction Wind 100 2008

50

0 9/1 9/15 9/29 10/13 10/27 11/10 11/24 (4) Winter 400 350

300 2004 250 2005 200 2006 2007 150

Wind Direction 2008 100

50

0 12/1 12/15 12/29 1/12 1/26 2/9 2/23 Fig. 2. Daily variation of wind direction in the Kinmen-Xiamen Airshed from 2004 to 2008.

314 Li et al., Aerosol and Air Quality Research, 13: 308–323, 2013

Table 4. Concentrations of fine and coarse particles and their mass ratio during the intensive sampling periods. 3 3 3 Sampling Periods Sampling Dates PM2.5 (μg/m ) PM2.5–10 (μg/m )PM10 (μg/m ) PM2.5/PM10 (%) January 6 56.0 46.8 102.8 54.5 First Intensive January 7 39.8 29.8 69.6 57.2 Sampling January 8 42.6 46.2 88.8 48.0 (Winter) January 9 45.8 43.5 89.3 51.3 January 10 47.1 68.1 115.2 40.9 March 16 100.5 43.6 144.0 69.8 Second Intensive March 17 65.6 33.0 98.7 66.5 Sampling March 18 45.9 26.5 72.4 63.5 (Spring) March 19 36.0 13.6 49.7 72.6 March 20 121.2 51.0 172.3 70.4

During the second sampling period in spring, the highest (37.6–192.9 μg/m3) were higher than those in summer (23.5– 3 3 PM10 and PM2.5 concentrations were 172.3 and 121. 2 μg/m , 83.2 μg/m ), because the weather system in the Kinmen- respectively. The mass ratios of PM2.5 to PM10 ranged from Xiamen Airshed is dominated by Northeastern Monsoons, 40.9 to 57.2% and from 63.5 to 72.6% for the first and which blow air pollutants from the eastern coast of China second intensive sampling periods, respectively. Generally to Xiamen Bay from approximately March to February. speaking, fine particles were more common than coarse However, local emissions from the surrounding region of particles in the Kinmen-Xiamen Airshed, suggesting that Xiamen Bay were as important as long-range transportation high PM10 concentration was probably attributed to local by Northeastern Monsoons, and thus a superimposition anthropogenic sources (i.e., industrial and vehicular emissions) phenomenon was regularly observed during the air pollution adjacent to the sampling sites during the intensive sampling episodes at Xiamen Bay. periods. In winter and spring, the PM10 concentration at site A3 was the highest, followed by those at site A1 located at Tempospatial Variation of Ambient PM10 Xiamen Island and site B1 located at Kinmen Island. In The PM10 monitoring data at seven sampling sites was summer, the PM10 concentration was relatively higher at used to investigate the spatial distribution and temporal sites A1 and B3. In fall, the PM10 concentration was higher variation of PM10 concentration. Figs. 2 and 3 illustrates at sites A1, A2, and A3, located in urban Xiamen. It should the daily variation of wind direction and monthly variations be noted that the sampling sites with high PM10 concentrations of the PM10 concentration contour of each sampling sites in were always adjacent to either industrial areas or main traffic the Kinmen-Xiamen Airshed. In fall, winter, and spring arteries. Fig. 4 illustrates the concentration contours and (from September 2008 to May 2009), the Daderng High prevailing wind direction in the Kinmen-Xiamen Airshed. School (A3) site and northern Xiamen island generally had The concentration contours of PM10 show that the PM10 hot the highest PM10 concentrations in Xiamen Bay. In summer spots of were always located between Daderng Island and (from June to August, 2008), the PM10 concentration was urban Xiamen in the Kinmen-Xiamen Airshed. The hot relatively higher in the region between urban Xiamen and spots of PM10 in the Kinmen-Xiamen Airshed were highly Kinmen Islands. High PM10 concentrations were always affected by the wind field of Xiamen Bay. The highest observed at the sites adjacent to industrial areas along the concentrations of PM10 were observed in January, mainly northern coast of Xiamen Bay. PM10 concentrations in winter due to the Northeastern Monsoons. Northern winds could

250 B1 B2 B3 A1 A2 A3 A4 ) 200 g/m3 μ 150

100

Concentration ( 50

0 Mar. Apr. May Jun. Jul. Aug. Sept. Oct. Nov. Dec. Jan. Feb. Month

Fig. 3. Monthly variation of PM10 concentrations at seven sampling sites in the Kinmen-Xiamen Airshed from March 2008 to February 2009.

Li et al., Aerosol and Air Quality Research, 13: 308–323, 2013 315

Fig. 4. Monthly variation of PM10 concentration contours and prevailing wind direction in the Kinmen-Xiamen Airshed from March 2008 to February 2009. transport atmospheric particles from the upwind emission were generally higher in spring and winter than those in sources (e.g., Jinjiang River Basin) to the downwind sites summer and fall. (Kinmen Island), causing a significant increase in atmospheric The metallic contents of PM10 sampled in the Kinmen- particle concentrations at Kinmen Island. The Hybrid Single- Xiamen Airshed are shown in Fig. 7. Crustal elements (Ca, Particle Lagrangian Integrated Trajectory is a widely used Mg, Fe, and Al) and anthropogenic elements (Zn, As, and model that plots the trajectory of a single air parcel from a Pb) contributed the major metallic contents of PM10. Among specific location and height above ground over a period of the crustal elements, Ca and Fe were the most abundant time. The 96-hour backward trajectories starting at metals. In particular, the concentration of Ca was much 118°14′N, 24°25′E at the heights of 100, 350, and 500 m, higher than that of other crustal elements (Fe, Mg and Al). respectively, to study the transport pathways of air parcels The highest concentrations of Ca in spring and winter were that arrived at Xiamen Bay on four different days are shown 3.4 and 5.7 μg/m3, respectively. Zn and Pb were the major in Fig. 5 (Chen et al., 2012). The level of atmospheric components among the other metallic elements (Cr, Mn, PM10 is affected by meteorological condition, thus PM10 Ni, and Cd). Pb is a toxic heavy metal (Barrat, 1990; Akhler concentrations in spring and winter was much higher than and Madany, 1993; Pirrone et al., 1996), and the highest those in fall and summer. Results from backward trajectories concentrations of Pb were observed at sites A2 and A3. showed that the concentrations of PM10 blown from the Observations showed that the incinerator combustion and north were generally higher than those from the south. heavy vessel travel were always heavy around Xiamen Bay, resulting in high Zn and Pb emissions. High Pb emissions Chemical Composition of PM10 may also came from incinerator, leaded-gasoline combustion. Figs. 6–8 illustrate the seasonal variations in the chemical High Pb concentrations could be also contributed form deuse composition of PM10 in the Kinmen-Xiamen Airshed. As heavy vessels which were quite busy around Xiamen Bay. shown in Fig. 6, the abundant water-soluble ionic species The oil burning for heavy vessels is one of the major cause 2– – + of PM10 were SO4 , NO3 , and NH4 at Xiamen Bay. The of chemical characteristic for Pb. The high Pb concentration main chemical species of PM10 were probably ammonium were also contributed form heavy vessel which were always sulfate ((NH4)2SO4) and ammonium nitrate (NH4NO3) (Yao heavy around Xiamen Bay. The oil of vessel were one of et al., 2003; Han et al., 2007; Kocak et al., 2007; Tsai et the cause of the high marker for the metal of Pb. al., 2012). The sulfate concentrations ranged from 3.0 to The concentrations of Ca and Fe at sites A2 and A3 in 37.0 μg/m3 with an average of 13.0 μg/m3, the nitrate the rural open lands and construction areas were higher concentrations ranged from 1.4 to 25.3 μg/m3 with an average than those at other sites. Al and Ca are the main elements of 9.2 μg/m3, and the ammonium concentrations ranged in the earth’s crust and particles emitted from the cement from 1.0 to 19.1 μg/m3 with an average of 6.5 μg/m3, which industry, and the wind-blown dusts and frictional works

316 Li et al., Aerosol and Air Quality Research, 13: 308–323, 2013

Spring Summer

Fall Winter

Fig. 5. Backward trajectories in all season in Kinmen-Xiaman Airshed with HYSPLIT model.

from construction sites increased the atmospheric loading The carbonaceous contents of PM10 sampled in the of dust particles (Huang et al., 1994). In addition, there are Kinmen-Xiamen Airshed are illustrated in Fig. 8. Elemental many stone processing plants located to the north of site carbon (EC), which has a chemical structure similar to A3. Moreover, anthropogenic metallic elements could be impure graphite, originates primarily from direct emissions transported to Xiamen Bay from the eastern coast of China from combustion. Organic carbon (OC) is emitted from through long-range transportation by the Northeastern primary anthropogenic sources and secondary organic Monsoons. aerosols formed by chemical reactions in the atmosphere.

Li et al., Aerosol and Air Quality Research, 13: 308–323, 2013 317

In this study, the concentrations of OC were always higher this study, the highest average OC/EC ratio of PM10 was than EC for all seasons at each sampling site. The mass observed at site A4, which was adjacent to major emission ratios of OC to EC (OC/EC) ranged from 1.1 to 2.7, and sources such as textile plants, industrial boilers, heavy oil were larger than unity for PM10 at all sampling sites. The and coal burning. Certain meteorological conditions (i.e. average OC/EC ratio of 1.6 at urban Xiamen was higher the Northeast Monsoons) and the huge amount of volatile than that of 1.4 at Kinmen Island. The results indicate that organic compounds emitted from various sources (e.g., OC was the major carbonaceous species of PM10 in the textile industry at Jinjing River Basin) could enhance the Kinmen-Xiamen Airshed. The OC/EC ratio can be used to formation of secondary organic aerosols. identify the formation of secondary organic aerosols when the OC/EC ratio exceeds 2.2 (Turpin et al., 1990; Chow et Chemical Transformation of SO2 and NOx 2– al., 1996). The order of OC/EC ratios for PM10 at all The chemical transformations of SO2 and NOx to SO4 – sampling sites was A4 > A2 > A3 > A1 > B3 > B2 > B1. In and NO3 (i.e., SOR and NOR) for PM10 are shown in Fig. 9.

25 (a)Spring ) B1 B2 B3 A1 A2 A3 A4 3 20 g/m μ 15

10

5 Concentration (

0 Na+ NH4+ K+ Mg2+ Ca2+ F- Cl- NO3- SO42- 25

) (b)Summer

3 B1 B2 B3 A1 A2 A3 A4 20 g/m μ 15

10

5 Concentration ( Concentration

0 Na+ NH4+ K+ Mg2+ Ca2+ F- Cl- NO3- SO42- 25

) B1 B2 B3 A1 A2 A3 A4 (c)Fall 3 20 g/m μ 15

10

5 Concentration ( Concentration

0 Na+ NH4+ K+ Mg2+ Ca2+ F- Cl- NO3- SO42- 25 B1 B2 B3 A1 A2 A3 A4 (d)Winter ) 3 20 g/m μ 15

10

5 Concentration ( Concentration

0 Na+ NH4+ K+ Mg2+ Ca2+ F- Cl- NO3- SO42-

Fig. 6. Tempospatial variation of ionic species of PM10 sampled in the Kinmen-Xiamen Airshed.

318 Li et al., Aerosol and Air Quality Research, 13: 308–323, 2013

6.0 B1 B2 B3 A1 A2 A3 A4 (a)Spring 5.0 ) 3 g/m

μ 4.0

3.0

2.0 Concentration ( Concentration 1.0

0.0 MgK CaTiCrMnFeZnAlCdAsPbCuNi 6.0 B1 B2 B3 A1 A2 A3 A4 (b)Summer

) 5.0 3 g/m

μ 4.0

3.0

2.0

Concentration ( 1.0

0.0 Mg K Ca Ti Cr Mn Fe Zn Al Cd As Pb Cu Ni 6.0 B1 B2 B3 A1 A2 A3 A4 (c)Fall

) 5.0 3 g/m

μ 4.0

3.0

2.0

Concentration ( 1.0

0.0 MgK CaTiCrMnFeZnAlCdAsPbCuNi 6.0 B1 B2 B3 A1 A2 A3 A4 (d)Winter 5.0 ) 3 g/m

μ 4.0

3.0

2.0 Concentration ( Concentration 1.0

0.0 MgK CaTiCrMnFeZnAlCdAsPbCuNi

Fig. 7. Tempospatial variation of metallic contents of PM10 sampled in the Kinmen-Xiamen Airshed.

The values of SOR were generally higher in spring than and Okita, 1990). The high SOR and NOR obtained in this 2– – those in other seasons, since the major emissions of SO2 study suggest that the formation of SO4 and NO3 from were mostly contributed from both local emissions and long- SO2 and NOx could occur in the atmosphere. The average range transportation. The average SOR of PM10 at Kinmen NOR of PM10 at Kinmen Island ranged from 0.10 to 0.41. Island ranged from 0.20 to 0.51. Previous studies reported A previous study reported that NOR is generally lower than that SOR and NOR are less than 0.25 and 0.10 for primary SOR (Colbeck and Harrison, 1984). The high NOR obtained – pollutants, respectively, while the chemical oxidation of in this study suggests that the formation of NO3 from NOx SO2 and NOx would occur in the atmosphere when SOR occurred in the atmosphere during the sampling period on and NOR are greater than 0.25 and 0.10, respectively (Ohta Kinmen Island.

Li et al., Aerosol and Air Quality Research, 13: 308–323, 2013 319

15 2.5 EC OC OC/EC

12 2 ) 3 g/m

μ 9 1.5

6 1 OC/EC Ratio OC/EC Concentration (

3 0.5

0 0 Spring- Spring- Summer- Summer- Fall- Fall- Winter- Winter- Kinmen Xiamen Kinmen Xiamen Kinmen Xiamen Kinmen Xiamen

Fig. 8. Tempsoparial variation of carbonaceous contents and their mass ratios (OC/EC) ratio of PM10 in the Kinmen- Xiamen Airshed.

0.6 SOR NOR 0.5

0.4

0.3 SOR or NOR or SOR 0.2

0.1

0 4-Apr. 23-Apr. 7-May 23-May 5-Jun. 24-Jun. 10-Jul. 20-Jul. 5-Aug. 20-Aug. 4-Sept. 5-Oct. 20-Oct. 5-Nov. 20-Nov. 4-Dec. 20-Dec. 5-Jan. 6-Jan. 7-Jan. 8-Jan. 9-Jan. 10-Jan. 14-Jan. 20-Feb. 28-Feb. 5-Mar. 16-Mar. 17-Mar. 18-Mar. 19-Mar. 20-Mar.

Fig. 9. The SOR and NOR ratios of PM10 in the Kinmen-Xiamen Airshed.

+ – Source Apportionment of PM10 secondary inorganic aerosols, while Na and Cl are from PCA has been commonly used as an exploratory tool to oceanic spray (Khemani et al., 1985). Pb is highly associated identify the major sources of aerosol emissions and to with vehicular exhausts, while Ca, Fe, and Al are associated statistically select independent source tracers. One of the with fugitive soils and road dusts (Kumar et al., 2001). Cr, objectives of this study was to distinguish the emission Pb, and Zn originate from anthropogenic sources, such as sources of PM10 sampled in the Kinmen-Xiamen Airshed. iron and steel plants, power plants, and industrial boilers., The concentrations of ionic species, metallic elements, and and Cu, Mn, and Pb are regarded as road dusts emitted carbonaceous contents of PM10 were used to calculate from heavy traffics and vehicular emissions (Kumar et al., varimax rotated principal component analysis to identify 2001; Davis et al., 2001). K is mainly derived from biomass the number of principal components having an eigenvalue burning. Ni is straight from lumber and heavy oil combustion. > 1.0 (Tandon et al., 2008) involved in the emission of PM10. Tables 5 and 6 summarize the results of PCA for the The source apportionment of PM10 was further assessed ionic species, metallic elements, and carbonaceous contents using a receptor model based on chemical mass balance of PM10 on Kinmen Island and in urban Xiamen. Five major (CMB) (Ke et al., 2007; Kothai et al., 2008; Wang et al., factors (KF1-KF5) contributing to PM10 were identified on 2008; Yatkin and Bayram, 2008). Kinmen Island (see Table 5). The highest component loading 2– + – 2– Previous investigation reported that SO4 , Al, OC, and associated with NH4 , NO3 , SO4 , EC, and OC was EC are from diesel vehicular exhausts (Wang et al., 2003; identified as Factor KF1, which accounted for 27.16% of + 2– – − 2− Cheng et al., 2010). NH4 , SO4 , and NO3 are from the variance. Moderate loading of OC, NO3 , SO4 , and

320 Li et al., Aerosol and Air Quality Research, 13: 308–323, 2013

Table 5. Principal component analysis of PM10 sampled on Table 6. Principal component analysis of PM10 at metro Kinmen Island. Xiamen.

Kinmen PM10 metro PM10 Island KF1 KF2 KF3 KF4 KF5 Xiamen XF1 XF2 XF3 XF4 Na+ 0.63 Na+ 0.81 + + NH4 0.95 NH4 0.95 Cl– 0.69 Cl– 0.82 – – NO3 0.91 NO3 0.94 2– 2– SO4 0.90 SO4 0.89 Mg Mg 0.52 K 0.71 K 0.54 0.50 Ca 0.87 Ca 0.83 Cr Cr 0.91 Mn 0.76 Mn Fe 0.79 Fe 0.79 Zn Zn 0.56 Al 0.80 Al 0.83 Pb 0.88 Pb Cu 0.64 Cu 0.77 Ni 0.69 Ni 0.78 EC 0.85 EC 0.79 OC 0.93 OC 0.89 Eigenvalues 4.89 3.56 2.41 2.09 1.51 Eigenvalues 7.71 2.98 2.53 1.26 Percentage of Percentage of 27.16 19.79 13.38 11.61 8.38 42.84 16.54 14.07 6.99 Variance (%) Variance (%) Accumulation (%) 27.16 46.95 60.32 71.93 80.31 Accumulation (%) 42.84 59.38 73.46 80.45

+ NH4 were also observed in Factor KF1, and these are XF2 explained 16.54% of the variance and was highly secondary aerosols and the byproducts of combustion. loaded with Na+ and Cl−. Hence, the second factor can be Hence, the first factor can be identified as industrial emission identified as oceanic spray. Factor XF3 explained 14.07% plus secondary aerosols, Factor KF2 was heavily loaded of the variance and was highly loaded with Cu, Ni, and K. with Ca, Fe, and Al with the percentage variance of 19.79%, K is derived from biomass burning, while Ni is straight which can be interpreted as the crustal materials. Thus, the from lumber and heavy oil combustion. The third factor second factor can be identified as road dusts. Factor KF3 can thus be identified as biomass burning and heavy oil explained 13.38% of the variance and was highly loaded combustion. Factor XF4 explained 6.99% of the variance with Pb, Ni, amd K, which are markers for biomass burning, and was highly loaded with Cr, which originates from heavy oil combustion (Kowalczyk et al., 1982; Harrison et anthropogenic sources, such as iron and steel plants, power al., 1996), and waste incinerator. The high concentrations plants, and industrial boilers. Hence, the fourth factor can of Pb were contributed mainly form heavy vessels which be identified as industrial emissions. According to the PCA were always heavily deuse around Xiamen Bay. The used results, the major sources in the Kinmen-Xiamen Airshed by vessels contains Pb which can be used as a markers for were secondary aerosols, crustal materials, biomass burning, vessels. The third factor can thus be identified as biomass oceanic spray, and heavy oil combustion. and fuel combustion and waste incinerator. Factor KF4 Table 7 summarizes the source apportionment of PM10 in explained 11.61% of the variance and was highly loaded with the Kinmen-Xiamen Airshed, and an obvious seasonal Na+ and Cl–, and thus the fourth factor can be identified as variation can be observed. In spring, soil dusts (19.93– oceanic spray (Khemani et al., 1985). Factor KF5 explained 24.51%) were the main source of PM10, followed by 8.38% of the variance and was highly loaded with Cu and secondary aerosols (16.39–22.86%), industrial boilers (11.94– Mn, and thus the fifth factor can be identified as industrial 19.48%), motor vehicular exhausts (8.48–14.41%), diesel emissions. vehicular exhausts (3.12–7.33%), petrochemical plants (2.21– Four factors (XF1-XF4) accounting for 80.45% of the 5.33%), oceanic spray (3.55–5.34%), steel plants (3.40– total variance were identified for PM10 in urban Xiamen 5.86%), biomass burning (2.12–8.11%), and cement plants (see Table 6). Factor XF1 explained 42.84% of the variance (1.65–4.97%) in the Kinmen-Xiamen Airshed. Among and presented high loadings of Ca, Fe, and Al, which can these, soil dusts, petrochemical plants, and vehicular exhausts be interpreted as the crustal contribution. Moderate loadings were the major sources in metro Xiamen, whereas soil − 2− + of OC, NO3 , SO4 and NH4 were also observed in Factor dusts and biomass burning dominated in Kinmen Island. In XF1. These are secondary aerosols and the byproducts of particular, the concentration of oceanic spray on Kinmen combustion, and thus the first factor can be identified as Island (3.89–5.34%) was similar to that observed in Xiamen the crustal contribution plus secondary aerosols. Factor Island (3.55–4.87%). In summer, soil dusts, vehicular

Li et al., Aerosol and Air Quality Research, 13: 308–323, 2013 321

Table 7. Source apportionment of PM10 sampled in the Kinmen-Xiamen Airshed. Spring Summer Emission Sources A1 A2 A3 A4 B1 B2 B3 A1 A2 A3 A4 B1 B2 B3 Industrial Boilers 19.06 19.48 16.26 17.11 11.94 12.82 13.31 14.73 5.54 9.79 8.15 7.35 6.36 13.09 Petroleum Industry 5.33 3.33 3.11 5.10 3.17 2.61 2.21 11.18 6.89 3.48 4.09 4.15 3.39 2.78 Cement Industry - 4.97 - - 2.56 1.65 2.22 - - - 5.18 3.15 2.47 3.04 Secondary Nitrate 8.75 6.93 9.21 6.22 5.97 6.60 6.51 11.00 13.71 10.90 6.43 4.18 5.05 3.98 Secondary Sulfate 9.26 11.77 13.65 13.40 10.42 10.59 10.67 16.28 10.76 13.91 9.73 9.64 8.92 6.69 Vehicular Exhausts 8.48 14.41 9.29 8.79 10.84 12.12 11.79 10.47 13.25 5.99 5.74 13.85 14.03 11.89 Diesel Exhausts 5.31 7.33 5.61 3.12 4.61 5.33 6.33 4.73 7.84 3.84 3.14 7.02 5.36 5.51 Oceanic Spray 3.55 4.87 3.72 4.54 5.34 3.89 4.55 4.31 3.12 6.07 5.77 8.14 6.15 7.35 Biomass Burning 2.12 4.34 3.20 2.12 4.67 6.40 8.11 ------Soil Dusts 23.80 19.93 22.31 20.67 24.51 20.56 21.51 25.82 19.56 23.31 25.13 24.12 23.28 20.30 Steel Industry 3.45 5.24 5.44 5.53 5.86 4.91 3.40 - - - 2.72 4.97 6.65 4.43 Others 10.88 1.39 8.20 13.41 10.10 10.52 9.39 0.43 14.12 17.38 23.92 13.44 18.34 20.94 Mass Percentage (%) 89.12 102.61 91.80 86.59 89.90 87.48 90.61 99.57 85.88 82.62 76.08 86.56 81.66 79.06 R2 0.87 0.84 0.87 0.87 0.87 0.89 0.93 0.93 0.87 0.83 0.81 0.88 0.85 0.93

Fall Winter Emission Sources A1 A2 A3 A4 B1 B2 B3 A1 A2 A3 A4 B1 B2 B3 Industrial Boilers 13.54 15.76 15.33 11.76 13.32 13.72 13.78 16.72 15.51 16.54 7.49 13.11 12.51 11.33 Petroleum Industry 8.09 6.13 3.53 3.66 6.37 3.08 2.02 5.95 6.19 5.16 3.60 5.51 7.26 6.87 Cement Industry 1.53 1.04 3.71 4.78 3.01 1.03 2.62 2.56 1.21 - - 4.01 2.31 1.69 Secondary Nitrate 9.37 9.82 8.48 9.50 8.57 7.88 8.17 7.64 5.13 9.84 9.81 7.12 12.04 10.88 Secondary Sulfate 9.51 10.07 12.02 12.50 9.05 8.43 9.24 11.97 10.29 14.37 16.17 13.44 17.12 17.71 Vehicular Exhausts 8.81 10.99 9.57 7.39 10.54 9.45 8.31 6.59 11.04 6.56 10.54 10.78 9.31 7.21 Diesel Exhausts 6.93 5.17 5.71 6.66 5.32 6.21 4.51 2.95 6.67 2.45 2.80 3.75 2.18 3.33 Oceanic Spray 3.18 3.73 3.11 2.45 6.50 3.59 4.51 3.15 2.87 3.33 1.89 5.13 2.30 4.13 Biomass Burning 1.11 3.16 2.36 3.33 2.59 6.61 5.35 3.67 - 2.33 - 6.48 8.68 6.26 Soil Dusts 21.92 21.20 20.71 20.92 18.36 20.84 18.52 24.61 23.13 25.79 28.04 17.62 11.89 12.47 Steel Industry 5.41 5.11 0.55 1.05 3.34 1.15 4.05 3.68 3.63 3.94 3.97 0.34 0.05 0.31 Others 10.60 7.81 14.92 15.99 13.03 18.01 18.92 10.51 14.33 9.69 15.68 12.71 14.36 17.82 Mass Percentage (%) 89.40 92.19 85.08 84.01 86.97 81.99 81.08 89.49 85.67 90.31 84.32 87.29 85.64 82.18 R2 0.88 0.81 0.91 0.93 0.83 0.82 0.83 0.93 0.88 0.81 0.84 0.93 0.91 0.81 “-“ such emission source is not apportioned. exhausts, and secondary aerosols were the major sources CONCLUSIONS that contributed to PM10. In fall, soil dusts, secondary aerosols, industrial boilers, and vehicular exhausts were the This study investigated the tempospatial distribution, major sources in metro Xiamen. The seasonal variation of physicochemical characteristics, and source apportionment the contribution of vehicular exhausts to PM10 in urban of atmospheric particles in the Kinmen-Xiamen Airshed. Xiamen was always higher than at other sites, since urban The results revealed that high PM10 concentrations were Xiamen had the heaviest traffic in the Kinmen-Xiamen usually found at the sampling sites adjacent to the industrial Airshed. Although vehicular exhausts were the dominant areas along the northern coast of Xiamen Bay. The PM10 source, biomass burning and secondary aerosols were also concentration contour showed that the highest PM10 significant sources causing an increase in PM10 at urban concentrations generally occurred in the region between and suburban areas in the Kinmen-Xiamen Airshed. In metro Xiamen and Kinmen Island. A superimposition winter, soil dusts, secondary aerosols, vehicular exhausts, phenomenon was regularly observed during air pollution and industrial boilers were the major sources of PM10. The episodes at Xiamen Bay, resulting from both local emissions Northern Monsoons transported suspended particles from from Xiamen Bay and long-range transportation from the the upwind emission sources (e.g., Jinjiang River Basin) to eastern coast of China by Northeastern Monsoons. the downwind sites (B1–B3) on Kinmen Island, causing a The mass ratios of fine and coarse particles (PM2.5/ significant increase in secondary aerosols mainly composed PM2.5–10) ranged from 40.9 to 57.2% and 63.5 to 72.6% of sulfate and nitrate. Moreover, the contribution of secondary during the first and second intensive sampling periods, aerosols to PM10 in metro Xiamen was generally higher respectively. Fine particles were more common than coarse than that in Kinmen Island. In contrast, the contribution of particles in the Kinmen-Xiamen Airshed, suggesting that biomass burning to PM10 in Kinmen Island was higher than the high PM10 concentration could mainly be attributed to that in metro Xiamen. local anthropogenic sources (i.e. industrial emissions and

322 Li et al., Aerosol and Air Quality Research, 13: 308–323, 2013 vehicular traffics) adjacent to the sampling sites in urban Cheng, Y., Lee, S.C., Ho, K.F., Chow, J.C., Watson, J.G., Xiamen. Chemical analysis indicated that the major water- Louie, P.K.K., Cao, J.J. and Hai, X. (2010). Chemically- 2– – + soluble ionic species of PM10 were SO4 , NO3 , and NH4 Speciated on-Road PM2.5 Motor Vehicle Emission Factors in Xiamen Bay, suggesting that PM10 was mainly composed in Hong Kong. Sci. Total Environ. 408: 1621–1627. of ammonium sulfate and ammonium nitrate, which were Chow, J.C., Watson, J.G. and Lu, Z. (1996). Descriptive generally seen in higher amounts in spring and winter than Analysis of PM2.5 and PM10 at Regionally Representative in summer and fall. The most abundant metals of PM10 Locations during SJVAQS/AUSPEX. Atmos. Environ. 30: were Ca, Mg, K, Fe, Al, Zn, Pb, Mn, and Cu, while OC 2079–2112. was higher than EC for all seasons in the Kinmen-Xiamen Colbeck, I. and Harrison R.M. (1984). Ozone-Secondary Airshed. The mass ratios of OC to EC (OC/EC) ranged Aerosol Visibility Relationships in Northwest England. from 1.1 to 2.7, and were larger than unity for PM10 at all the The Sci. Total Environ. 34: 87–100. sampling sites. SOR and NOR of PM10 on Kinmen Island Davis, A.P., Shokouhian, M. and Ni, S. (2001). Loading ranged from 0.20 to 0.51 and 0.10 to 0.41, respectively. Estimates of Lead, Copper, Cadmium, and Zinc in Urban 2– High SOR and NOR suggested that the formation of SO4 Runoff from Specific Sources. Chemosphere 44: 997–1009. – and NO3 from SO2 and NOx could occur in the atmosphere Deshmukh, D.K., Deb, M.K., Tsai, Y.I. and Mkoma, S.L. of the Kinmen-Xiamen Airshed. The results of the principal (2011). Water Soluble Ions in PM2.5 and PM1 Aerosols in component analysis and CMB receptor modeling showed Durg City, Chhattisgarh, India. Aerosol Air Qual. Res. that major sources of PM10 in the Kinmen-Xiamen Airshed 11: 696–708. were soil dusts, secondary aerosols, petrochemical industries, Dockery, D.W., Pope III, C.A., Xu, X., Spengler, J.D., Ware, vehicular exhausts, industrial boilers, oceanic spray, and J.H., Fay, M.E., Ferris, B.G. and Speizer, F.E. (1993). An biomass burning. Northern winds may have transported Association between Air Pollution and Mortality in Six suspended particles from the upwind emission sources U.S. Cities. New Engl. J. Med. 329: 1573–1759. (e.g., Jinjiang River Basin) to the downwind sites (Kinmen Dockery, D.W. and Pope, C.A. (1994). Acute Respiratory Island), causing a significant increase in the concentration Effects of Particulate Air Pollution. Auun. Rev. Publ. of atmospheric particle on Kinmen Island. Health 15: 107–132. Eddie, L., Chan, C.K. and P. Paatero (1999). Application of ACKNOWLEDGMENTS Positive Matrix Factorization in Source Apportionment of Particulate Pollutants in Hong Kong. Atmos. Environ. This study was performed under the auspices of 33: 3201–3212. Environmental Protection Bureau of Kinmen Government. Fung, Y.S. and Wong, L.W.Y. (1995). Apportionment of Air The authors are grateful to Professor Shui-Ping Wu and Pollution Sources by Receptor Models in Hong Kong. Professor Xin-Hong Wang in the State Key Laboratory of Atmos. Environ. 29:2041–2048. Marine Environmental Science, Xiamen University for their Han, L., Zhuang, G., Cheng, S. and Li, J. (2007). The Mineral assistance in sampling and analysis of PM for this study. Aerosol and Its Impact on Urban Pollution Aerosols over Beijing, China. Atmos. Environ. 41: 7533–7546. REFERENCES Harrison, R.M., Smith, D.T.J. and Luhana, L. (1996). Source Apportionment of Atmospheric Polycyclic Aromatic Akhler, S.M. and Madany, I.M. (1993). Heavy Metals in Hydrocarbons Collected from an Urban Location in Street and House Dust in Bahrain. Water Air Soil Pollut. Birmingham, UK. Environ. Sci. Technol. 30: 825–832. 66: 111–119. Huang, X., Olmez, I. and Aras, N.K. (1994). Emissions of Anderson, K.R., Avol, E.L., Edwards, S.A., Shamoo, D.A., Trace Elements from Motor Vehicles: Potential Marker Pen, R.C., Linn, W.S. and Hackney, J.D. (1992). Controlled Elements and Source Composition Profile. Atmos. Environ. Exposures of Volunteers to Respirable Carbon and 28: 1385–1391. Sulphuric Acid Aerosols. J. Air Waste Manage. Assoc. Kaneyasu, N. (1995). Seasonal Variation in Chemical 42: 770–776. Composition of Atmospheric Aerosols and Gaseous Barrat, R.S. (1990). An Assessment of Dust Analyses: With Species in Sapporo, Japan. Atmos. Environ. 29: 1559–1568. Particular Reference to Lead and Certain Other Metals. Ke, L., Ding, X., Tanner R.L., Schauer, J.J. and Zheng, M. Int. J. Environ. Anal. Chem. 40: 77–97. (2007). Source Contributions to Carbonaceous Aerosols Chen, B., Du, K., Wang, Y., Chen, J.S., Zhao, J.P., Wang, K., in the Tennessee Valley Region. Atmos. Environ. 41: Zhang, F.W. and Xu, L.L. (2012). Emission and Transport 8898–8923. of Carbonaceous Aerosols in Urbanized Coastal Areas in Khemani, L.T., Momin, G.A., Naik, M.S., Rao, P.S.P., China. Aerosol Air Qual. Res. 12: 371–378. Kumar, R. and Murty, B.V.R. (1985). Trace Elements Chen, K.S., Lin, C.F. and Chou, Y.M. (2001). Determination and Sea Salt Aerosols over the Sea Areas around the of Source Contributions to Ambient PM2.5 in Kaohsiung Indian Sub-continent. Atmos. Environ. 19:277–284. City Using Receptor Model. J. Air Waste Manage. Assoc. Kocak, M., Mihalopoulos, N. and Kubilay, N. (2007). 51: 489–498. Chemical Composition of the Fine and Coarse fraction Cheng M.T. and Tsai, Y.I. (2000). Characterization of of Aerosols in the Northeastern Mediterranean. Atmos. Visibility and Atmospheric Aerosols in Urban, Suburban, Environ. 41: 7351–7368. and Remote Areas. Sci. Total Environ. 263: 101–114. Kothai, P., Saradhi, I.V., Prathibha, P., Hopke, P. K., Pandit,

Li et al., Aerosol and Air Quality Research, 13: 308–323, 2013 323

G.G. and Puranik, V.D. (2008). Source Apportionment of Inland and Offshore Sites over Southeastern Coastal Coarse and Fine Particulate Matter at Navi Mumbai, Region of Taiwan Strait. Aerosol Air Qual. Res. 11: 664– India. Aerosol Air Qual. Res. 8: 423–436. 678. Kowalczyk, G.S., Gordon, G.E. and Rheingrover, S.W. Tsai, J.H., Lin, J.H., Yao, Y.C. and Chiang, H.L. (2012). (1982). Identification of Atmospheric Particulate Sources Size Distribution and Water Soluble Ions of Ambient in Washington, DC, Using Chemical Element Balances. Particulate Matter on Episode and Non-episode Days in Environ. Sci. Technol. 16:79–90. Southern Taiwan. Aerosol Air Qual. Res. 12: 263–274. Kumar, A.V., Patil, R. S. and Nambi, K.S.V. (2001). Source Turpin, B.J., Cary, R.A. and Huntzicker, J.J. (1990). An In- Apportionment of Suspended Particulate Matter at Two Situ, Time-Resolved Analyzed for Aerosol Organic and Traffic Junctions in Mumbai, India. Atmos. Environ. 35: Elemental Carbon. Aerosol Sci. Technol. 12: 161–171. 4245–4251. Wang, W.C., Chem, K.S., Chem, S.J., Lin, C.C., Tsai, J.H., Lin, J.J. (2002). Characterization of the Major Chemical Lai, C.H. and Wang, S.K. (2008). Characteristics and Species in PM2.5 in the Kaohsiung City, Taiwan. Atmos. Receptor Modeling of Atmospheric PM2.5 at Urban and Environ. 36: 1911–1920. Rural Sites in Pingtung, Taiwan. Aerosol Air Qual. Res. Ohta, S. and Okita, T. (1990). A Chemical Characterization 8: 112–129. of Atmospheric Aerosol in Sapporo. Atmos. Environ. 24A: Wang, X.H., Bi, X.H., Sheng, G.Y. and Fu, J.M. (2006). 815–822. Chemical Composition and Sources of PM10 and PM2.5 Otvos, E., Pazmandi, T. and Tuba, Z. (2003). First National Aerosols in Guangzhou, China. Environ. Monit. Assess. Survey of Atmospheric Heavy Metal Deposition in 119: 425–439. Hungary by the Analysis of Mosses. Sci. Total Environ. Wang, Y.F., Huang, K.L., Li, C.T., Mi, H.H., Luo, J.H. and 309: 151–160. Tsai, P.J. (2003). Emissions of Fuel Metals Content from Pirrone, N., Keeler, G.J., Nriagu, J.O. and Warner, P.O. a Diesel Vehicle Engine. Atmos. Environ. 37: 4965–4971. (1996). Historical Trends of Airborne Trace Metals in Wu, S.P., Qian, R.R., Li, T.C., Wang, X.H., Hong, H.S. Detroit from 1971 to 1992. Water Air Soil Pollut. 88: and Yuan, C.S. (2012). Seasonal Variation for the Ratio 145–165. of BaP to BeP at Different Sites in Great Xiamen Bay. J. Pope, C.A., Burnett, R.T., Thun, M.J., Calle, E.E., Krewski, Environ. Monit. 14: 1221–1230. D., Ito, K. and Thurston, G.D. (2002). Lung Cancer, Yao, X., Lau, P.S., Fang, M., Chan C.K. and Hu, M. Cardiopulmonary Mortality, and Long-term Exposure to (2003). Size Distributions and Formation of Ionic Species Fine Particulate Air Pollution. J. Am. Med. Assoc. 287: in Atmospheric Particulate Pollutants in Beijing, China: 1132–1141. l-Inorganic Ions. Atmos. Environ. 37: 2991–3000. Shaocai Y., Dennis, R.L., Bhave, P.V. and Eder, B.K., Yatkin, S. and Bayram, A. (2008). Source Apportionment (2004). Primary and Secondary Organic Aerosols over of PM10 and PM2.5 Using Positive Matrix Factorization the United States: Estimates on the Basis of Observed and Chemical Mass Balance in Izmir, Turkey. Sci. Total Organic Carbon (OC) and Elemental Carbon (EC), and Environ. 390: 109–123. Air Quality Modeled Primary OC/EC Ratios. Atmos. Yuan, C.S., Lee, C.G., Liu, S.H., Yuan, C., Yang, H.Y. and Environ.38: 5257–5268. Chen, C.T. (2002). Developing Strategies for Improving Swietlicki, E. and Krejci, R. (1996). Source Characterisation Urban Visual Air Quality. Aerosol Air Qual. Res. 2: 9–22. of the Central European Atmospheric Aerosol Using Yuan, C.S., Sau, C.C., Chen, M.C., Hung, M.H., Chang, Multivariate Statistical Methods. Nucl. Instrum. Methods S.W. and Lin Y.C. (2004). Mass Concentration and Size- Phys. Res., Sect. B 109–110: 519–525. Resolved Chemical Composition of Atmospheric Aerosols Tsai, H.H., Ti, T.H. Yuan, C.S., Hung, C.H. and Lin, C. Sampled at Pescardores Islands during Asian Dust Storm (2008). Effects of Sea-land Breezes on the Spatial and Periods in the Years of 2001 and 2002. Terr. Atmos. Temporal Distribution of Gaseous Air Pollutants around Ocean. Sci. 15: 857–879. the Coastal Region of Southern Taiwan. J. Environ. Eng. Yuan, C.S., Lee, C.G., Liu, S.H., Chang, J.C., Yuan, C. and Manage. 18: 387–396. Yang, H.Y. (2006). Correlation of Atmospheric Visibility Tsai, H.H., Yuan, C.S., Hung, C.H. and Lin, Y.C. (2010). with Chemical Composition of Kaohsiung Aerosols. Comparing Physicochemical Properties of Ambient Atmos. Res. 82: 663–679. Particulate Matter of Hot Spots in a Highly Polluted Air Quality Zone. Aerosol Air Qual. Res. 10: 331–344. Tsai, H.H., Yuan, C.S. Hung, C.H. and Lin, C. (2011). Received for review, November 16, 2011 Physicochemical Properties of PM2.5 and PM2.5–10 at Accepted, August 3, 2012