Pahs In Hongkong Soil

Environmental Pollution 141 (2006) 107e114 www.elsevier.com/locate/envpol

Distributions and Concentrations of PAHs in Hong Kong Soils H.B. Zhang a,b, Y.M. Luo a,c, M.H. Wong a,b,*, Q.G. Zhao a,c, G.L. Zhang a,c a

Soil and Environment Joint Open Laboratory between Institute of Soil Science, Chinese Academy of Sciences and Hong Kong Baptist University, Soil and Environmental Bioremediation Research Center, State Key Laboratory of Soil and Sustainable Agriculture, Nanjing 210008, China b Croucher Institute for Environmental Science and Department of Biology, Hong Kong Baptist University, Kowloon Tong, Hong Kong c Graduate School of the Chinese Academy of Sciences, Beijing 100039, China Received 10 December 2004; accepted 2 August 2005

Baseline information is provided on levels, distributions and possible sources of PAHs in Hong Kong soils. Abstract Surface soil (0e10 cm) samples from 53 sampling sites including rural and urban areas of Hong Kong were collected and analyzed for 16 EPA priority polycyclic aromatic hydrocarbons (PAHs). Total PAH concentrations were in the range of 7.0e410 mg kgÿ1 (dry wt), with higher concentrations in urban soils than that in rural soils. The three predominant PAHs were Fluoranthene, Naphthalene and Pyrene in rural soils, while Fluoranthene, Naphthalene and Benzo(b C k)fluoranthene dominated the PAHs of urban soils. The values of PAHs isomer indicated that biomass burning might be the major origin of PAHs in rural soils, but vehicular emission around the heavy traffic roads might contribute to the soil PAHs in urban areas. A cluster analysis was performed and grouped the detectable PAHs under 4 clusters, which could be indicative of the PAHs with different origins and PAHs affected by soil organic carbon contents respectively. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Hong Kong soil; Cluster analysis; PAH profile; PAH isomer ratio

1. Introduction Polycyclic aromatic hydrocarbons (PAHs) are a class of several hundred individual compounds containing at least 2 condensed rings, and 16 of them have been identified as ‘‘priority pollutants’’ by the United States Environmental Protection Agency (US EPA). Many of PAHs possessing of mutagenic, carcinogenic, and teratogenic properties are ubiquitously distributed in the environment (Xue and Warshawsky, 2004). Wild and Jones (1995) estimated that at least 90% of the environmental PAH burden in Great Britain is stored in soils. This estimate excludes contaminated sites such as gaswork sites, petroleum refineries, or wood preservation plants. * Corresponding author. Croucher Institute for Environmental Science and Department of Biology, Hong Kong Baptist University, Kowloon Tong, Hong Kong. Tel.: C852 3411 7746; fax: C852 3411 7743. E-mail address: [email protected] (M.H. Wong). 0269-7491/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.envpol.2005.08.031

The typical level of endogenous total PAHs in soils is in the range 1e10 mg kgÿ1 resulting from plant synthesis and natural fires (Edwards, 1983). However, concentrations and patterns of PAHs varied markedly between temperate and tropical soils due to the faster microbial degradation and photo-oxidation and enhanced volatilization in tropical soils (Smith et al., 1995; Wilcke et al., 1999a,b). Typical concentrations in tropical soils of Thailand are in the range 11e347 mg kgÿ1 dominated by naphthalene, phenanthrene and perylene (Wilcke et al., 1999b), while 108e54 500 mg kgÿ1 (the sum 14 PAHs) in temperate soil of Wales, dominated by high molecular weight PAHs such as fluoranthene, 1,2-dibenzanthracene and chrysene were reported (Jones et al., 1989). The major PAH source is incomplete combustion of organic materials (Jenkins et al., 1996; Wilcke et al., 2002; Kim et al., 2003). This occurs naturally in vegetation fire, volcanic activities, and diagenetic processes. Moreover, most PAHs in urban areas are produced by anthropogenic combustion of

108

H.B. Zhang et al. / Environmental Pollution 141 (2006) 107e114

wood and fossil fuels. Other sources of PAH include oil spill accidents and wastewater discharge from domestic and industrial activities (Tam et al., 2001). In addition to pyrolytic and petrogenic origins, there are some evidences showing biological PAH formation. Naphthalene and perylene are believed to be produced biologically by woody plants or termites in Amazonian rain forest (Wilcke et al., 2000, 2003). Concentrations of 4-, 5-, and 6-, ring PAHs are all found significantly increased in soils under oxygen deficient conditions, it is thus assumed that PAHs are formed in soils from plant materials and humus precursors under such conditions (Thiele and Brummer, 2002). A number of studies have been conducted on the sources and distributions of PAHs in atmosphere (Zheng et al., 1996; Zheng and Fang, 2000; Lee et al., 2001), sediments (river, wetland and marine) (Zheng and Richardson, 1999; Tam et al., 2001; Zheng et al., 2002; Wong and Poon, 2003), and marine shellfish (Richardson, 2003) in Hong Kong. However little is known about the sources and concentrations of PAHs in soils of Hong Kong. Soil as a repository of all types of chemical inputs, is important to the sources and pools of PAHs on the regional scale. The main objective of this study is to determine the distributions and concentrations of the 16 EPA priority PAHs in Hong Kong soils. It is also hoped to identify the possible sources of these pollutants.

sample consisted of 3 sub-samples collected from the surrounding of each site (within 1 m2). Among the 45 rural soil samples, most of them were collected from woodlands and grasslands, 7 were from farmlands and 2 from wetlands (Mai Po mangrove swamp), and the other 8 urban soil samples were collected mostly from urban parks located in Hong Kong Island, Kowloon, Sha Tin, Tai Po, Tun Men and Tsuen Wan, while one sample was colleted from the reclamation site in Ma On Shan seaside. All the samples were freeze-dried and sieved to !2 mm after removing stones and residual roots, then stored in desiccators prior to analysis of PAHs.

2.3. Sample extraction and cleanup Sample extraction, cleanup and analysis of PAHs were conducted in the State Key Laboratory of Pollution Control and Resource Reuse in Nanjing University, following the methods described by Ma et al. (2003). Soil samples (7.5 g dry weight) were placed into a centrifugal tube, with a mixture of 20 ml n-hexane, 5 ml methanol and 5 ml distilled water and sonicated for 1 h in an ultrasonic shaking apparatus. The mixture was then centrifuged and the extract collected. The same extraction was repeated once again by adding 20 ml n-hexane to the filter residue. The two extract solutions were then combined and anhydrous sodium sulfate was added for drying, and then concentrated to around 1 ml by rotary evaporation. The concentrated extracts were cleanup using a chromatographic column consisting of 1 g of activated silica gel (0.07 ml distilled water was added), and then eluted with 8 ml n-hexane, after that the eluent was further concentrated to about 1 ml. The solution was finally concentrated to around 0.1 ml under a gentle steam of pure nitrogen. 10 ml 9-phenylanthracene (20 ppm) was added to the solution as an internal standard prior to transferring to a glass of microvial for GC injection.

2.4. PAH analysis 2. Materials and methods 2.1. The study area Hong Kong is situated at the south-eastern tip of Mainland China, with a total area slightly over 1100 km2 covering Hong Kong Island, Kowloon, New Territories and a number of outlaying islands. Out of the total land area, three-quarters is countryside, including 4 main types of land use: woodlands, grasslands, farmlands or fallow lands and wetlands (AFCD, 2002). Woodlands and grasslands are mainly distributed in hills that largely unsettled and remained fairly natural, while the limited agricultural lands and fallow lands are scatted at the alluvial plain. Most wetlands can be found in the northwestern part of the New Territories, and half of which is mangrove swamp (Ashworth et al., 1993). Mai Po mangrove swamp is the most important wetland in Hong Kong because it has been designed as a site with international importance (a Rasmar site). The remaining lands (one-quarter) are urban areas with a population density over 26 000 persons per km2, which mainly centralized in the northern part of Hong Kong Island, southern part of Kowloon, Kwai Chung, and Tsuen Wan Districts. Most soils of Hong Kong are Oxisols commonly found in the humid tropics, characterized by strong weathering and leaching (Soil and Conservation Service, 1999). Hong Kong’s climate is sub-tropical, tending towards temperate for nearly half of the year and the annual average temperature and relative humidity were 24  C and 78% respectively in 2000 (Hong Kong Observatory, 2001). The mean annual rainfall ranges from around 1300 mm at Waglan Island to more than 3000 mm in the vicinity of Tai Mo Shan. About 80% of the rain falls between May and September. October to next April is relatively arid, and hill fire caused by tomb worship is often occurred in countryside in this period (AFCD, 2002).

The concentrations and profiles of PAH compounds were analyzed using an Agilent6890 gas chromatograph (HP-5 quartz capillary column: 30 m, 0.33 mm ID ! 0.25 mm film thickness) with a flame ionization detector (GC/FID). The oven temperate was initially set at 80  C and held for 1 min, ramped at 25  C/min to 160  C, 3  C/min from 160 to 300  C, and held for 2 min. High purity (99.99%) nitrogen gas was used as the carrier gas. Identification and quantification of 16 PAH compounds were based on matching their retention time with a mixture of PAH standards. The 16 PAH compounds were Naphthalene (Naph), Acenaphthylene (Acel), Acenaphthene (Ace), Fluorene (Flu), Phenanthrene (Phe), Anthracene (Ant), Fluoranthene (Fla), Pyrene (Pyr), Benzo(a)Anthracen (BaAnt), Chrysene (Chry), Benzo(b)Fluoranthene (BbFla), Benzo(k)Fluoranthene (BkFla), Benzo(a)Pyrene (BaP), Inde(1,2,3)Pyrene (IPyr), Dibenz(a,h)Anthracene (DahA), Benzo(g,h,i)Perylene (BghiP). Among the analyzed 16 EPA PAH compounds, only 15 were identified (DahA was not identified). Out of the detected 15 PAHs, the peaks of Benzo(b)Fluoranthene and Benzo(k)Fluoranthene were extremely close and therefore difficult to be distinguished, so these two compounds were calculated as one, namely Benzo(b C k)Fluoranthene (B(b C k)Fla). The 14 PAHs were then used for the calculation of the total PAHs.

2.5. Quality control The detection limit ranged from 0.005 to 0.09 mg kgÿ1. Three laboratory blanks were run with the samples and concentrations of PAHs in soils were corrected accordingly. The procedure was also checked for recovery efficiencies by analyzing soil samples spiked with PAH standards. The average recovery of the internal standards ranged from 90 to 116%. All readings were recorded in duplicate and the variation of PAH concentrations of replicated samples was less than 10%.

2.2. Soil sampling and preparation 2.6. Soil physical and chemical analyses Surface soils (0e10 cm) were collected from the New Territories, Kowloon, Hong Kong Island, and Lantau Island in December 2000. The sampling sites and land use of each sampling site are shown in Fig. 1. In total, 53 soil samples (consisted of 45 in rural areas and 8 in urban areas) were collected, and each

The methods of soil analyses methods were based on the methods described in Lu (2000). Soil pH was measured using 10.0 g air-dried soil (passed through a 60 mesh) suspended in 25 ml deionized water using a pH meter.

H.B. Zhang et al. / Environmental Pollution 141 (2006) 107e114

109

Fig. 1. Map of soil sampling sites in Hong Kong.

Total organic carbon (TOC) was measured using potassium bichromate oxidation process. Total nitrogen was determined by Kjeldahl method. The analysis of particle-size distribution was carried out using the pipette method, and classification of the soil texture was based on Soil Survey Division Staff (1993). Cation exchange capacity (CEC) was analyzed by ammonium acetate saturation.

2.7. Statistical analyses Statistical analyses including ANOVA test and cluster analysis were performed using SPSS 11.0 for windows. The KolmogoroveSmirnov (KeS) test was performed to test the frequency distribution of data set of PAHs concentrations. The variables of not normally distributed were log-transformed to achieve normal distribution prior to the statistical analyses conducted. Arithmetic means and geometric means were both provided for expressing the average concentrations of PAHs. ANOVA test companied with LSD method were adopted for multiple comparison. Cluster analysis was carried out for identifying homogeneous groups of individual PAHs in Hong Kong soils. The raw data was firstly performed with principal components analysis (PCA) before execution of the cluster analysis, and then the scores resulted from the PCA were hierarchically clustered using weighted average linkage between the groups and the Pearson correlation for the cluster intervals (Atanassova and Bru¨mmer, 2004).

soil CEC and TOC concentrations followed a similar pattern, with the lowest values found in urban soils. The higher contents of coarse fraction and lower contents of soil CEC and TOC of urban soils suggested the human disturbance in the urban areas (Jim, 2002). Although the C/N ratios of different soils were similar, the farmland soils had a lower C/N ratio on average, suggesting that the organic matter content was more N-rich. In general, urban soils were disturbed strongly by human activities, farmland and wetland soils were moderately affected, while woodland and grassland soils were not affected. 3.2. Concentrations and distributions of PAH in Hong Kong soils PAHs were detected in all soil samples, and the KeS test result indicated that concentrations of total PAH, Phe, Fla Table 1 The main characteristics of soils in Hong Kong Soil properties

Land-use patterns

Sand (%) Silt (%) Clay (%) Textural class USDA pH (H2O) CEC (cmol kgÿ1) TOC (g kgÿ1) Total nitrogen (g kgÿ1) C/N ratio

42.9 33.0 24.1 Loam

38.3 40.8 20.9 Loam

40.9 43.9 15.2 Loam

2.3 79.1 55.7 12.0 42.0 8.9 Silty clay Loamy sand

4.62 10.2 23.8 1.25

4.70 10.9 21.0 1.18

6.23 8.57 12.3 1.07

6.35 9.75 29.2 1.93

19.0

17.8

11.5

15.1

3. Results and discussion 3.1. Major characteristics of Hong Kong soils The properties of the collected surface soils, including soil texture, pH, CEC, TOC, C/N ratio are presented in Table 1. The texture ranged from silty clay in wetlands to loamy sand in urban areas, while most rural soils were loam. The pH (H2O) values of woodland and grassland soils ranged from strongly to moderately acidic, farmland and wetland soils were near neutral, while urban soils were more alkaline. The

Woodland Grassland Farmland Wetland

Urban areas

7.52 3.77 5.32 0.28 18.9

110

H.B. Zhang et al. / Environmental Pollution 141 (2006) 107e114

and Pyr were normally distributed (P ! 0.05) in soils. The total concentrations of PAH varied from 7.00 to 410 mg kgÿ1 (dry wt), with a arithmetic mean of 54.6 mg kgÿ1 (dry wt) and geometric mean of 37.6 mg kgÿ1(dry wt). It was higher than the total PAH concentrations (1e10 mg kgÿ1 dry wt) of typical endogenous soils resulted from plant synthesis and natural fires (Edwards, 1983). When comparing with the target value set by Dutch government for unpolluted soils (20e50 mg kgÿ1) (Van Brummelen, 1996), most rural soils (7.0 to 69.3 mg kgÿ1 dry wt) were within the target value. However, total PAHs concentrations of urban soils (especially samples from Hong Kong Island) were much higher (O100 mg kgÿ1). Fig. 2 is the map of total PAH concentrations in Hong Kong soils, showing that total PAH concentrations higher than 50 mg kgÿ1 were mostly found in urban areas, such as Hong Kong Island, Kowloon, and Tsuen Wan. The maximum concentration of 410 mg kgÿ1 was detected in the Zoological and Botanical Garden of Hong Kong Island, which might be due to the high traffic volumes of the area (Lee, 2001). The soil samples from Tusen Wan, a former industrial district also contained a rather high level (O200 mg kgÿ1), as expected. Table 2 compares the mean concentrations of PAHs in the 5 land use types. Soil PAH concentrations of all soil samples from rural areas including woodland, grassland, farmland and wetland were not significantly different, but they were all significantly (P ! 0.05) lower than those contained in urban soils with the sole exception of Naph, which was almost similar in all Hong Kong soils (around 6.0 mg kgÿ1 dry wt). The average total concentration of soil PAHs in urban areas was around 5 folds of those in rural soils, and concentrations of

B(b C k)Fla and Chry in urban soils were 36 and 18 folds respectively of those in rural soils. It is demonstrated that rural soil in Hong Kong are not or little polluted by PAHs that emitted from the urban areas due to the similar concentrations of soil PAHs among the four using lands of rural areas and much lower than those in urban areas. It is commonly observed that PAH concentrations in soils are affected by soil properties such as TOC, with higher concentrations accompanied with high TOC (Wilcke, 2000; Wilcke and Amelung, 2000; Tam et al., 2001; Ribes et al., 2003). In this study, such correlation was not revealed between soil properties and PAH concentrations. However, concentrations of Flu, Phe and BaP showed a significant variance between two TOC levels (Table 3). The relatively higher concentrations of these three PAH compounds were always detected in soils with TOC contents higher than 30 g kgÿ1 (dry wt), but it was not the case of the reverse, which indicated that soil PAHs in these soils were not biologically formed from the precursors of soil humus, which just make a role of prohibiting the PAHs degradation or volatilization from soil through sorption of the PAHs (Chiou et al., 1998; White and Pignatello, 1999). The poor correlations between PAHs and soil properties in this study might associate with the relatively lower concentrations of PAHs in Hong Kong soils compared with other reports all carried out in heavily PAH contaminated sediments (Kim et al., 1999; Yang, 2000). Simpson et al. (1996) showed that the relationship between total PAHs and organic carbon was only significant for highly contaminated sites where total PAH concentration was great than 2000 mg kgÿ1. A weak correlation between TOC and PAH

Fig. 2. Map of total PAHs concentrations in Hong Kong soil.

H.B. Zhang et al. / Environmental Pollution 141 (2006) 107e114

111

Table 2 Average concentrations (mg kgÿ1 dry weight) of PAHs in soils of Hong Kong PAH

Land-use patterns Woodland

Naph Acel Ace Flu Phe Ant Fla Pyr BaAnt Chry B(b C k)Fla BaP Ipry BghiP SPAHs Numbers of samples

Grassland

Farmland

Wetland

Urban Areas

A.M.

G.M.

A.M.

G.M.

A.M.

G.M.

A.M.

G.M.

A.M.

G.M.

6.56 1.13 ND 2.43 2.05 1.06 8.76 3.31 1.51 0.35 1.45 3.39 0.04 2.24 34.3 17

5.36 1.84 ND 2.04 1.62 1.66 6.98 3.75 2.78 2.04 1.90 4.78 0.61 3.10 30.1

4.62 1.50 Trace 2.9 2.11 0.90 8.09 1.81 1.16 1.02 0.99 4.66 0.89 4.67 35.4 19

3.97 1.61 Trace 2.13 1.78 2.13 8.30 3.40 2.76 4.35 3.57 7.33 1.56 11.2 31.3

4.97 2.73 ND 1.24 1.61 0.94 10.3 5.33 1.24 0.07 ND 0.13 ND 2.54 31.1 7

3.43 2.84 ND 1.06 1.74 3.11 9.25 7.14 1.90 0.52 ND 0.90 ND 2.81 26.91

5.60 1.90 ND 1.30 2.45 ND 3.40 11.3 2.15 2.15 0.45 0.85 2.05 ND 33.6 2

5.49 3.80 ND 1.30 4.90 ND 3.38 7.56 4.30 4.30 0.89 1.70 4.10 ND 29.4

6.24 2.73 0.53 4.56 16.7 3.57 28.0 27.1 8.97 16.2 26.7 9.90 8.26 9.82 169 8

5.99 3.29 1.05 4.54 12.7 3.95 23.9 21.2 9.28 11.1 24.2 10.7 7.18 7.88 134

A.M.: Arithmetic means; G.M.: Geometric means; ND: under detection limit.

3.3. PAH profile in Hong Kong soils The PAH profile of Hong Kong soils is presented in Table 2. Fla (four-ring PAH), Naph (two-ring PAH) and Pyr (four-ring PAH) were dominated in rural areas, each accounting for 25.1, 15.7 and 10.6% of the total concentrations respectively, while Fla, Pyr and B(b C k)Fla(five-ring PAH) were dominated in urban soils, and each accounting for 16.6, 16.0 and 15.8% of the total concentration respectively. Fig. 3 further shows that in urban soils, high weight molecular PAH (O5 rings PAH) dominated the PAH profiles, with average concentrations higher than that of low weight molecular PAH(2e3 rings PAH), compared with rural soils, in which average concentrations of 2e3 rings PAHs were higher or equal to that of S5 rings PAH. The different PAH sources of rural and urban soils are due to the fact that BbFla, BkFla and other 5 rings PAH

Table 3 Compare means of selected PAHs based on two different Total Organic Carbon levels Selected PAHs (mkgÿ1 dry weight)

Flu Phe BaP

ÿ1

TOC Levels (g kg ) !30

Sig. O30

N

Mean

Std. Deviation

N

Mean

Std. Deviation

35 35 35

1.93 1.65 2.06

1.80 1.38 3.26

11 11 11

4.12 3.16 6.96

5.50 3.10 4.92

such as Chry and BaP are typical markers for fossil fuel combustion (Wilcke, 2003). Profiles of PAHs also varied within rural areas, e.g. Fla, Naph, and BaP were dominated in woodland soils whereas Fla, BghiP and BaP in grassland soils. BaP in these two soils were relatively higher than other soils due to their higher concentrations of TOC, which might act as a strong adsorption matrix for BaP (Chiou et al., 1998). The three predominant soil PAHs in the human affected farmland and wetland soils were Naph, Fla and Pyr. The concentration of Pyr was significantly (P ! 0.05) higher in these two soils than that in woodland or grassland soils, as it is one of the major combustion products of fossil fuels and other organic matters (Wilcke et al., 2000). Oil spill and leakage from boats and ships, and discharge from municipal and industrial wastewater and runoff might be attributing to relatively higher Pyr concentration in the wetland soils (Tam et al., 2001). Bioformation of Pyr from plant materials and humus precursors takes place under reducing conditions, and wetland soil is often waterlogged, which may account partly

160.0 2-3 rings PAH

140.0

PAHs Conc.(ug/kg)

was also observed in Welsh rural soils (Jones et al., 1989). Therefore, it was assumed that in such a not heavily PAHs contaminated site just like Hong Kong soil, concentration of PAHs will not be significantly affected by soil TOC concentration until it reached a certain level.

120.0

4 rings PAH 5 rings PAH

100.0 80.0 60.0 40.0 20.0

0.045 0.027 0.000

0.0

Woodland Grassland

Farmland

Wetland Urban Area

Fig. 3. Soil PAHs concentrations in different land utilizations.

H.B. Zhang et al. / Environmental Pollution 141 (2006) 107e114

3.4. Sources of PAHs in Hong Kong soils

Petroleum Combustion

Petroleum

Grass/Wood/Coal Combustion

Rural soil Urban soil

1.00 0.80

Combustion

0.60 0.40 0.20 0.00 0.00

Petroleum 0.40

0.20

0.60

0.80

1.00

1.20

Fla/(Fla+Pyr) Fig. 4. PAH cross plot for the ratio of Ant/(Ant C Phe) vs. Fla/(Fla C Pyr).

and urban areas, a PAHs pattern of aerosol derived from Zheng et al. (1996) was used to compare with the data obtained from soils samples collected in the present study (Fig. 5). Taking into account of different units used in these two studies, a relative concentration in percentage of individual PAH to total PAHs were employed. The PAH pattern in rural soils was deviated from that found in aerosol characterized by vehicular exhaust contamination. While the low molecular weight PAHs such as Phe and Ant may be degraded in soils or lost by volatilization after deposition, this is hardly the case for the high-molecular weight PAHs (Wild and Jones, 1995). Even the pattern of specific vehicular exhaust, high-molecular weight PAHs in rural soils is different from that in aerosol of Hong Kong, however, which in urban soil is different to that in rural soil. The specific vehicular exhaust original PAHs such as Chry, B(b C k)Fla and BaPyr in urban soil were shown the resemble trend with that in Hong Kong aerosol. Therefore, PAHs in Hong Kong urban soils may mainly origin from the emission of vehicular exhaust, while that in rural soils, biomass combustion may be the predominant source.

30.0 25.0

rural soil urban soil aerosol

20.0 15.0 10.0 5.0 0.0 B(g,h,i)P

Ipyr

BaPyr

B(b+k)Fla

Chry

BaAnt

Pyr

Fla

Ant

Phe

It is essential to identify the origin and potential sources of soil PAH in order to assess the environmental risk caused by PAHs. A number of studies demonstrated the usefulness of PAH isomer ratios in source identification (Zheng and Fang, 2000; Lee et al., 2001; Zheng et al., 2002). The ratio of Ant/ (Ant C Phe) is one of the frequently used PAH isomer ratios in distinguishing between combustion and petroleum sources. The ratio !0.1 is taken as an indication of petroleum while a ratio O0.1 indicates a dominance of combustion (Budzinski et al., 1997). However, Fraser et al. (1998) demonstrated that Ant undergoes more rapid photochemical reaction in the atmosphere than Phe, suggesting that the original composition information will not persevered during atmospheric transport. In order to assess sources of PAHs in soils more accurately, the ratio of Fla/(Fla C Pyr) are used as Fla and Pyr isomer pair degrade photolytically at comparable rates (Behymer and Hites, 1988). Yunker et al. (2002) suggested that a Fla/(Fla C Pyr) ratio !0.4 indicates petroleum input, ratio between 0.4e0.5 liquid fossil fuel (vehicle and crude oil) combustion and ratio O0.5 grass, wood or coal combustion. In this study, the Ant/(Ant C Phe) ratios ranged from 0e1.0, while Ant in many samples were not detected, it is therefore difficult to judge the sources of PAHs in these soils samples only with a Ant/(Ant C Phe) ratio of 0 which may be due to the photolytic degradation prior to Phe before deposition onto soils. The ratios of Fla/(Fla C Pyr) in this study ranged from 0.16e1.0. The minimal value was found in a farmland soil sample as shown in Fig. 4, indicating that the input of PAHs was derived from petroleum. The sample site was located at a farm near the Tolo Harbour in Ma On Shan area, and irrigation with domestic or industrial wastewater containing petroleum might be the main source of soil PAHs. Soil samples with Fla/(Fla C Pyr) ranged from 0.4e0.5 was only found once in urban soils, while most other urban soil samples had a Fla/(Fla C Pyr) ratio near 0.5 and an Ant/(Ant C Phe) ratio O0.1, which suggested that petroleum combustion was the predominant sources of soil PAHs in urban areas (Fraser et al., 1998). However, most rural soils had Fla/(Fla C Pyr) ratio O0.5, indicating grass, wood or coal combustion was the main PAHs source in Hong Kong rural soils. Being densely populated with a heavy traffic, it has been revealed that vehicular exhaust is the main source of aerosol PAHs in Hong Kong (Zheng et al., 1996). In order to further demonstrate the differences in sources of soil PAHs in rural

1.20

Ant/(Ant+Phe)

for the relatively higher concentration of Pyr (Thiele and Brummer, 2002). Fla and Naph, which were the major PAHs produced in vegetation fires (Freeman and Cattell, 1990; Kim et al., 2003; Lemieux, 2004) were almost dominated in all rural soils. This is due to the frequently occurred hill fire in the countryside (from October to next April) resulting from the activities of grave worship and barbecue (AFCD, 2002). It is therefore believed that vegetation fire is the major origin of PAHs in the soils of Hong Kong countryside.

Percentages of individual PAHs to total PAHs (%)

112

Individual PAHs Fig. 5. A relative concentration in percentage of individual PAHs to total concentrations PAHs in the soil of this study and in aerosol from vehicular emission in Mong Kok, Hong Kong (Zheng et al., 1996).

H.B. Zhang et al. / Environmental Pollution 141 (2006) 107e114

CASE Label Ace BaAnt Phe IPyr Pyr B(b+k)Fla Chry Ant Fla Naph Acel B (g,h,i) P BaPyr Flu

0

Rescaled 5

Distance 10

Cluster 15

113

Combine 20

25

Num 1 4 13 11 14 6 8 3 9 12 2 7 5 10

Fig. 6. Hierarchical dendogram for 14 individual PAHs in Hong Kong soils using average linkage between groups and Pearson correlation as measure interval.

3.5. Cluster analysis Cluster analysis was performed to identify the homogeneous groups of individual PAHs in Hong Kong soils. The results presented in the dendrogram (Fig. 6) distinguished the 14 individual PAHs into two major groups. The first major group can be subdivided into two subgroups. The first subgroup consisted of Ace, BaAnt, Phe, IPyr, Pyr, B(b C k)Fla, Chry and Ant, which were all low in rural soils, but higher in urban soils, in particular B(b C k)Fla and Chry (which were 36 and 17 times higher in urban soils than those in rural soils). Therefore, this subgroup represents the typical marks of PAHs in Hong Kong urban soils. The second subgroup composed of Fla and Naph, which were both anthropogenic and biopedogenic origins (Atanassova and Bru¨mmer, 2004). It was found that Fla was the dominated individual PAH both in rural and urban soils, and the average concentration of Naph in rural soils was approximately the same to that in urban soils. Naph is one of the major PAHs produced in vegetation fires (Freeman and Cattell, 1990), and Lee et al. (2001) studied PAHs in Hong Kong atmosphere and found that Naph in atmosphere is mainly existed in gaseous phase and can be transported to a long distance. These may explain why concentrations of Naph in both rural and urban soils were within the same range. The second major group also contained two subgroups. The first one with Acel and B(ghi)P, which were found in some rural soils with a high concentration as in urban soils. They may be produced from biomass burning (Jenkins et al., 1996). The second one comprised of BaPyr and Flu, with significant correlations with soil TOC, indicating that soil organic carbon is also a factor controlling the concentration of PAHs in Hong Kong soils. 4. Conclusions The concentrations of PAHs in Hong Kong soils ranged from 7.0e69.3 mg kgÿ1 (dry wt) in rural soils and 42.9e410 mg kgÿ1 in urban soils, respectively. Fluoranthene, Naphthalene and

Pyrene were dominated in rural soils, while Fluoranthene, Naphthalene and Benzo(b C k)fluoranthene in urban soils. Higher weight molecular PAHs were observed in urban soils. The profile of PAHs varied slightly among different types of land use under rural soils. The present results indicated that the biomass burning might be the major source of PAHs in rural soils. However, in urban soils, vehicular emission might play an important role to soil PAH contamination. Using the cluster analysis, 4 clusters were distinguished, in which, two clusters each representing the typical marks of PAHs in urban and rural soils, one cluster indicating that PAHs were derived from both anthropogenic and biopedogenic origins, and the last one showing the accumulation of PAHs was related to higher organic carbon concentrations in Hong Kong soils.

Acknowledgements The authors are grateful for the financial support given by the Major State Basic Research Development Program of China (973 Program), No. 2002CB410810/09, and by the Strategic Research Fund of the Hong Kong Baptist University.

References Agriculture, Fisheries and Conservation Department (AFCD), 2002. Department annual report 2001e2002. Agriculture, Fisheries and Conservation Department, Hong Kong. Ashworth, J.M., Corlett, R.T., Dudgeon, D., Melville, D.S., Tang, W.S.M., 1993. Hong Kong and Fauna: Computing Conservation. Hong Kong ecological database. World Wide Fund for Nature Hong Kong, 24. Atanassova, I., Bru¨mmer, G.W., 2004. Polycyclic aromatic hydrocarbons of anthropogenic and biopedgenic origin in a colluviated hydromorphic soil of Western Europe. Geoderma 120 (1e2), 27e34. Behymer, T.D., Hites, R.A., 1988. Photolysis of polycyclic aromatic hydrocarbons adsorbed on fly ash. Environ. Sci. Technol. 22, 1311e1319. Budzinski, H., Jones, I., Bellocq, J., Pierard, C., Garrigues, P., 1997. Evaluation of sediment contamination by polycyclic aromatic hydrocarbons in the Gironde estuary. Mar. Chem. 58, 85e97.

114

H.B. Zhang et al. / Environmental Pollution 141 (2006) 107e114

Chiou, C.T., Mcgroddy, S.E., Kile, D.E., 1998. Partition characteristics of polycyclic aromatic hydrocarbons on soils and sediments. Environ. Sci. Technol. 32, 264e269. Edwards, N.T., 1983. Polycyclic aromatic hydrocarbons (PAHs) in the terrestrial environment e a review. J. Environ. Qual. 12, 427e441. Fraser, M.P., Gass, G.R., Simoneit, B.R., Rasmussen, R.A., 1998. Air quality model evaluation data for organics. 5. C6eC22 non-polar and semipolar aromatic compounds. Environ. Sci. Technol. 32, 1760e1770. Freeman, D.J., Cattell, F.C.R., 1990. Woodburning as a sources of atmospheric polycyclic aromatic hydrocarbons. Environ. Sci. Technol. 24, 1581e1585. Hong Kong Observatory, 2001. Summary of meteorological observations in Hong Kong 2000. Hong Kong Observatory, Hong Kong. Jenkins, B.M., Jones, A.D., Turn, S.Q., Williams, R.B., 1996. Emission factors for polycyclic aromatic hydrocarbons from biomass burning. Environ. Sci. Technol. 30, 2462e2469. Jim, C.Y., 2002. Soil recovery from human disturbance in tropical woodlands in Hong Kong. Catena 743, 1e19. Jones, K.C., Stratford, J.A., Waterhouse, K.S., Vogt, N.B., 1989. Organic contaminants in Welsh soils: polynuclear aromatic hydrocarbons. Environ. Sci. Technol. 23, 540e550. Kim, G.B., Maruya, K.A., Lee, R.F., Lee, J.H., Koh, C.H., Tanabe, S.S., 1999. Distribution and sources of polycyclic aromatic hydrocarbons in sediments from Kyeonggi Bay, Korea. Mar. Pollut. Bull. 28, 7e15. Kim, E.J., Oh, J.E., Chang, Y.S., 2003. Effects of forest fire on the level and distribution of PCDD/Fs and PAHs in soil. Sci. Tot. Environ. 311, 177e189. Lee, S.C., Ho, K.F., Chan, L.Y., Zielinska, B., Chow, J.C., 2001. Polycyclic aromatic hydrocarbons (PAHs) and carbonyl compounds in urban atmosphere of Hong Kong. Atmospheric Environ. 35, 5949e5960. Lemieux, P.M., Lutes, C.C., Santoianni, D.A., 2004. Emission of organic air toxics from open burning: a comprehensive review. Prog. Energy Combust. Sci. 30, 1e32. Lu, R.K., 2000. Analytical Methods of Soil and Agri-Chemistry. Scientific and Technology Press, Beijing. Ma, L.L., Lao, W.J., Wang, X.T., Liu, H., Chu, S.G., Xu, X.B., 2003. Analytical method for trace semi-volatile organic compounds in the soil of Beijing suburbs. Chinese J. Anal. Chem. 9, 1025e1029 (in Chinese). Ribes, A., Grimalt, J.O., Garcia, C.J.T., Cuevas, E., 2003. Polycyclic aromatic hydrocarbons in mountain soils of the subtropical Atlantic. J. Environ. Qual. 32, 977e987. Richardson, B.J., Zheng, G.J., Tse, E.S.C., Luca-Abbott, S.B.D., Siu, S.Y.M., Lam, P.K.S., 2003. A comparison of polycyclic aromatic hydrocarbon and petroleum hydrocarbon uptake by mussels (Perna viridis) and semipermeable membrane devices (SPMDs) in Hong Kong coastal waters. Environ. Pollut. 122, 223e227. Simpson, C.D., Mosi, A.A., Cullen, W.R., Reimer, K.J., 1996. Composition and distribution of polycyclic aromatic hydrocarbons in surficial marine sediments from Kitimat Harbour, Canada. Sci. Total Environ. 181, 265e278. Smith, D.J.T., Edelhauser, E.C., Harrison, R.M., 1995. Polynuclear aromatic hydrocarbon concentrations in road dust and soil samples collected in the United Kingdom and Pakistan. Environ. Technol. 16, 35e53. Soil Conservation Service, 1999. Soil taxonomy: a basic system of soil classification for making and interpretation soil survey, second ed. Agriculture Handbook, vol. 436. United States Department of Agriculture, Natural Resources Conservation Division, Washington, DC. Soil Survey Division Staff, 1993. Soil Survey Manual, Revised Edition. Agriculture Handbook, vol. 18. United States Department of Agriculture, Washington DC. Tam, N.F.Y., Ke, L., Wang, X.H., Wong, Y.S., 2001. Contamination of polycyclic aromatic hydrocarbons in surface sediments of mangrove swamps. Environ. Pollut. 114, 255e263.

Thiele, S., Bru¨mmer, G.W., 2002. Bioformation of polycyclic aromatic hydrocarbons in soil under oxygen deficient conditions. Soil Biol. Biochem. 34, 733e735. Van Brummelen, T.C., Verweij, R.A., Wedzinga, S.A., Van Gestel, C.A.M., 1996. Enrichment of polycyclic hydrocarbons in forest soils near a blast furnace plant. Chemosphere 32, 293e314. White, J.C., Pignatello, J.J., 1999. Influence of bisolute competition on the desorption kinetics of polycyclic aromatic hydrocarbons in soil. Environ. Sci. Technol. 33, 4292e4298. Wilcke, W., 2000. Polycyclic aromatic hydrocarbons (PAHs) in soil e a Review. J. Plant Nutr. Soil Sci. 163, 229e248. Wilcke, W., Amelung, W., 2000. Persistent organic pollutants in native grassland soils along a climosequence in North America. Soil Sci. Soc. Am. J. 64, 2140e2148. Wilcke, W., Lilienfein, J., Lima, S.D.C., Zech, W., 1999a. Contamination of highly weathered urban soils in Uberlandia, Brazil. J. Plant Nutr. Soil Sci. 162, 539e548. Wilcke, W., Muller, S., Kanchanakool, N., Niamskul, C., Zech, W., 1999b. Polycyclic aromatic hydrocarbons (PAHs) in hydromorphic soils of the tropical metropolis Bangkok. Geoderma 91, 297e309. Wilcke, W., Amelung, W., Martius, C., Garcia, M.V.B., Zech, W., 2000. Biological sources of polycyclic aromatic hydrocarbons (PAHs) in the Amazonian rain forest. J. Plant Nutr. Soil Sci. 163, 27e30. Wilcke, W., Krauss, M., Amelung, W., 2002. Carbon isotope signature of polycyclic aromatic hydrocarbons (PAHs): evidence for different sources in tropical and temperate environments. Environ. Sci. Technol. 36, 3530e 3536. Wilcke, W., Amelung, W., Krauss, M., Martius, C., Bandeira, A., Garcia, M., 2003. Polycyclic aromatic hydrocarbon (PAH) patterns in climatically different ecological zones of Brazil. Organic Geochem. 34, 1405e1417. Wild, S.R., Jones, K.C., 1995. Polynuclear aromatic hydrocarbons in the United Kingdom environment: a preliminary source inventory and budget. Environ. Pollut. 88, 91e108. Wong, M.H., Poon, B.H.T., 2003. Sources, fates and effects of persistent organic pollutants in China, with emphasis on the Peal River Delta. In: Fiedler, H. (Ed.), The Handbook of Environmental Chemistry, vol. 3. Springer-Verlag, Berlin/Heidelberg, pp. 356e359. Xue, W.L., Warshawsky, D., 2004. Metabolic activation of polycyclic and heterocyclic aromatic hydrocarbons and DNA damage: a review. Toxicol. Appl. Pharmacol. 206 (1), 73e93. Yang, G.P., 2000. Polycyclic aromatic hydrocarbons in the sediments of the South China Sea. Environ. Pollut. 108, 163e171. Yunker, M.B., Macdonald, R.W., Vingarzan, R., Mitchell, R.H., Goyette, D., Sylvestre, S., 2002. PAHs in the Fraser river basin: a critical appraisal of PAH ratios as indicators of PAH sources and composition. Organic Geochem. 33, 489e515. Zheng, M., Fang, M., 2000. Particle-associated polycyclic aromatic hydrocarbons in the atmosphere of Hong Kong. Water Air Soil Pollut. 117, 175e189. Zheng, G.J., Richardson, B.J., 1999. Petroleum hydrocarbons and polycyclic aromatic hydrocarbons (PAHs) in Hong Kong marine sediments. Chemosphere 38, 2625e2632. Zheng, M., Wan, T.S.M., Fang, M., Wang, F., 1996. Characterization of the non-volatile organic compounds in the aerosols of Hong Kong-Identification, abundance and origin. Atmospheric Environ. 31, 227e237. Zheng, G.J., Man, B.K.W., Lam, J.C.W., Lam, M.H.W., Lam, P.K.S., 2002. Distribution and sources of polycylic aromatic hydrocarbons in the sediment of a sub-tropical coastal wetland. Water Res. 36, 1457e1468.