2c

Environ Geochem Health (2007) 29:11–19 DOI 10.1007/s10653-006-9052-2

ORIGINAL PAPER

Some characteristics of the distribution of heavy metals in urban topsoil of Xuzhou, China Xue-Song Wang Æ Yong Qin

Received: 17 August 2005 / Accepted: 9 May 2006 / Published online: 23 December 2006
Springer Science+Business Media B.V. 2006

Abstract An assessment is presented of distribution characteristics of heavy metals in the urban topsoil from the city of Xuzhou. The concentrations of Ag, Al, As, Au, Ba, Be, Bi, Cd, Co, Cr, Cu, Fe, Ga, Hg, Li, Mn, Mo, Ni, Pb, Pd, Pt, Sb, Sc, Se, Sn, V and Zn have been determined from 21 soil samples. Examination of lognormal distribution plots indicates that the diagrams of Al, Be, Fe, Ga, Li, and V are almost linear suggesting that these metals are almost unaffected by anthropogenic activities while the plots for As, Cd, Cu, Pb, Pd, Pt, Se, Zn and others are not linear probably due to anthropogenic activities from which these metals are delivered to the soils. Al is used for mineralogical normalization of these data. An evaluation of background values for topsoil is also carried out by means of lognormal distribution plots. The results show our background values obtained from the lognormal distribution plots are comparable to those values of uncontaminated soils of Xuzhou obtained by

X.-S. Wang (&) Department of Chemical Engineering, Huaihai Institute of Technology, Lianyungang, Jiangsu 222005, China e-mail: [email protected] Y. Qin School of Mineral Resources and Geo-Science, China University of Mining and Technology, Xuzhou, Jiangsu 221008, China

previous work except for Cd and Hg. At present, no explanation for the exceptions Cd and Hg can be given. Keywords Background value Æ Enrichment factor Æ Heavy metals Æ Lognormal distribution plot Æ Normalization Æ Urban soils

Introduction Heavy metals continue to receive increasing attention due to the better understanding of their toxicological relevance in ecosystems and human health. Numerous studies have been conducted to characterize the metal content in different substrates: soil, air, food, water, paints, dust, teeth and others. The study of heavy metal content in soil is of a great importance due to the fact that soils effectively act as a reservoir which, after temporary storage of metals, can act as a source and a sink for metal contamination (Martinez Garcia et al., 2001). Wind erosion with ensuing airborne soil particles is one of the main sources contributing to atmosphere particulate matter. Urban soils are known to have peculiar characteristics such as unpredictable layering, poor structure, and higher concentrations of heavy metals (Tiller, 1992). In recent years, public attention has been focused increasingly on heavy metal contamination and its effect on man and

123

12

creatures in urban soils. Around the world a great many studies have evaluated the concentrations in urban soil (Cal-Prieto et al., 2001; De Miguel et al., 1998; Li, Poon, & Liu, 2001; Manta et al., 2002; Martlnez Garcla et al., 2001; Ordo´n˜ez et al., 2003; Paterson, Sanka, & Clark, 1996; Wilcke et al., 1998). However, the absolute concentration of heavy metals in soils does not necessarily indicate the degree of contamination by anthropogenic sources as metals from natural sources also accumulate in these soils. The differences in soil composition such as its grain-size distribution and mineralogy would affect the natural heavy metal concentrations. In order to derive correct results from data on heavy metal content in soils, granulometric measurement, and/or metal/reference elements ratios are useful approaches towards complete normalization of granular and mineralogical variations, and identification of anomalous metal concentrations in soils. This approach has been successfully used to compensate for the natural variability of metals in the marine and coastal sediments to detect and quantify anthropogenic contamination (Covelli & Fontolan, 1997; Din, 1992; Schropp et al., 1990; Sedemann, 1991; Summers et al., 1996). Nevertheless the suitability of this approach to soils is not clear in spite of the need to monitor the degree of metal contamination in urban soils. It has also been demonstrated that this kind of normalization of data cannot always be used successfully for the explanation of heavy metal distribution and the contamination sources (Celo et al., 1999). Lognormal distribution plots are another way to further elucidate data. Until now, the distribution characteristics of heavy metals in Xuzhou urban topsoil have not been investigated. In this study, the metal-reference-metal normalization technique as well as lognormal distribution plots is used to determinate background levels and to estimate different source for different metals.

Environ Geochem Health (2007) 29:11–19

11622¢–11840¢ E. Xuzhou (China) is an industrial city and current urban population exceeds 1,200,000 inhabitants. The main wind direction is from the northwest, although in general, wind velocities are low, even approaching zero at times. This enhances the deposition of particulates within the city. A total of 21 topsoil samples (depth = 0–10 cm) were collected within the city of Xuzhou (Fig. 1). At each sampling point, three sub-samples, with a 20 · 20 cm surface, were taken and then mixed to obtain a bulk sample. Such a sampling strategy was adopted in order to reduce the possibility of random influence of urban waste not clearly visible. All the samples were collected with a stainless steel spatula and kept in PVC packages. These soil samples are mainly composed of parks/recreational, industrial/ manufacturing, gardens and roadside soils. The soils samples were air-dried and sieved through a 2-mm sieve. Soils ground to 200 mesh (approximately 80 lm in diameter) were decomposed with mixed acids (hydrofluoric acid, nitric acid and hydrochloric acid) in sealed Teflon vessels under microwave irradiation for the ICPAES (inductively coupled plasma atom emission spectrometry) and ICP-MS (inductively coupled plasma mass spectrometry) measurements. Boric acid was then added to mask the free fluoride ion which attacks glassware of ICP-AES and ICP-MS and to dissolve sparingly soluble fluorides (Fujikawa, Fukui, & Kudo, 2000). The recovery rates were 89–105% for the measurements of Au, Ba, Be, Cd, Co, Cu, Li, Mn, Mo, Ni, Pd, Pt, Sc, V, and Zn by ICP-MS for certified values of standard soil samples, ESS-1 and ESS-2 , provided by China Environmental Monitoring General Station, China. The recovery was 100–120% for the measurements of Al, Ag, As, Bi, Cr, Ga, Fe, Hg, Pb, Sb, Se and Sn by ICP-AES, also based on the two certified values of standard soil samples.

Results and discussion Materials and methods Heavy metal concentrations The study area is located northwestern part of Jiangsu, one of the provinces of China, the geographical position being 3343¢–3458¢ N,

123

Summary statistics (mean, standard derivation, median, maximum and minimum values) for the

Environ Geochem Health (2007) 29:11–19

13

Fig. 1 Map of the Xuzhou city (Jiangsu) with location of sampling sites of topsoil

analyzed 27 elements in all the studied samples are presented in Table 1. For comparison mean values for uncontaminated soils of Xuzhou are reported in the same table. The results demonstrate a general enrichment of heavy metals in the topsoil with respect to background values. Ag, As, Cd, Cu, Hg, Pb, Sb and Zn concentrations are higher for topsoil compared to their levels in uncontaminated soil (China). Ba, Bi, Cr, Li, Ni, Sc, Se and Sn concentrations exhibit generally low levels, close to those reported for uncontaminated soils. Be, Co, Ga, Mn, Mo and V concentrations are in normal ranges compared to the uncontaminated soils. Unleaded gasoline was introduced in China during the late 1980s, when a new generation of cars was equipped with catalytic converters (Pd, Pt, Rh). These converters reduce the emission of CO, non-combusted hydrocarbons and NOx, favouring

redox reactions and leading to the formation of less dangerous compounds (CO2, N2, H2O) (Varrica et al., 2003). However the introduction of catalytic converters is now causing a considerable rise in the concentration of these elements in several natural matrices such as soil, urban water and vegetation near areas subjected to intense vehicular traffic (Barefoot, 1997; Helmers, Schwarzer, & Schuter, 1998; Varrica et al., 2003). Pt and Pd concentrations in the studied soils vary in the range 1.0–4.7 and 1.1–3.7 ng g–1, respectively. Au concentrations have arithmetic mean concentration 4.4 ng/g, varying in the range 0.8–24 ng g–1 in the urban topsoils of Xuzhou. Enrichment factors (EFs) Quantifying the level of contamination is useful and can be done using enrichment factors. This

123

14

Environ Geochem Health (2007) 29:11–19

Table 1 Descriptive statistics of the analytical data found in soils of Xuzhou

a

Wang (2005)

Element

Mean

Standard derivation

Median

Max.

Min.

Background valuea

Al (%) Fe (%) Ag (mg kg–1) Se (mg kg–1) Sc (mg kg–1) Ga (mg kg–1) Ba (mg kg–1) Li (mg kg–1) Bi (mg kg–1) V (mg kg–1) Pb (mg kg–1) Cu (mg kg–1) Zn (mg kg–1) Cd (mg kg–1) Ni (mg kg–1) Co (mg kg–1) Cr (mg kg–1) As (mg kg–1) Hg (mg kg–1) Sb (mg kg–1) Mn (mg kg–1) Mo (mg kg–1) Be (mg kg–1) Sn (mg kg–1) Pt (lg kg–1) Au (lg kg–1) Pd (lg kg–1)

6.05 3.37 0.28 0.42 17.4 15.2 485 35.6 0.42 76 43.3 38.2 144.1 0.54 34.3 11.7 78.4 39.8 0.29 3.46 543.1 1.51 1.79 5.13 2.5 4.4 2.3

0.64 0.48 0.24 0.27 2.63 1.83 54.1 5.67 0.39 10.3 26.1 16.2 90.1 0.6 17.6 2.59 21.6 123 0.32 11.3 116 0.9 0.19 2.3 1.1 4.8 0.6

5.87 3.29 0.19 0.31 17.0 15.0 470 36.0 0.32 74 36 32 102 0.42 30 11 72 13 0.18 0.96 508 1.2 1.7 4.2 2.9 3.5 2.8

5.13 4.15 1.10 1.0 25.0 18.0 628 47 2.1 101 120 80 380 2.9 104 19 162 577 1.3 53 902 4.9 2.2 11 4.7 24 3.7

8.04 2.66 0.06 0.13 13.0 12.0 425 25 0.23 62 16 17 53 0.11 23 8.9 63 8.7 0.02 0.79 430 0.71 1.5 2.2 1.0 0.8 1.1

5.30 2.36 0.094 0.11 14 12 413 27 0.23 61 16 15 46 0.067 24 9.5 60 10 0.026 0.83 417 0.58 1.5 2.9 1.0 0.60 1.9

ratio describes the magnitude at which the metal is enriched above what is considered background value and is defined below: EF ¼

ðM=RÞsoil ðM=RÞbackground

ð1Þ

where M is the concentration of the metal of concern, R is the reference element concentration, soil refers to the metal-reference ratio for the soil, and background refers to the metal-reference ratio for uncontaminated soils (Table 1). The appropriate reference metals are those not subject to contamination by anthropogenic sources. In this evaluation, Ga was chosen as the reference element. If an enrichment factor is greater than unity, this indicated that the metal is more abundant in the soil relative to that found in the regional background. However, enrichment factors less than 5 may be not considered significant although they are an indicator of metal accumulation, because such small enrichments may arise from differences in the composition of

123

local soil material and the reference background value used in the EF calculations. If the EF values are greater than 5, in this case they are considered to be soil contamination for related metals (Tokalioglu, Kartal, & Birol, 2003). Figure 2 represents the distributions of enrichment factors of metals. The mean values of enrichment factors are greater than 5 for Hg (0.71–40), Cd (1.36–30.5), Au (1.23–28.23), vary between 2 and 5 for As (0.8–46.1), Sb (0.75–51.1), Se (1.09–7.79), Cu (1.04–3.76), Pb (0.92–6.01), Ag (0.58–8.26), Zn (1.06–5.83), Mo (1.13–5.96), Pt (0.67–3.78) and vary between 1 and 2 for Ni (0.77–3.05), Sn (0.70–2.76), Cr (0.78–2.16), Fe (0.88–1.52), Pd (0.43–1.51), Bi (0.86–7.30), Mn (0.81–1.44) and Li (0.85–1.26) and vary between 0 and 1 for Ba (0.68–1.21), Sc (0.7–1.26), Co (0.74 –1.33), V (0.72–1.14) and Be (0.8–1.13). From the above observation, it is inferred that some of the urban topsoils are significantly contaminated by Au, Cd, and Hg and the soils are not significantly contaminated by Ba, Be, Bi, Co, Cr, Fe, Li, Mn, Ni, Pd, Sc, Sn and V.

Environ Geochem Health (2007) 29:11–19

Enrichment factor

50 40 30 20 10 0 HgAsCdSeSbZnCuNiPbAgSnCrBaLi BiScMnCoMoAuPdFe V BePt

Metals

Fig. 2 Box-plots of Enrichment factors (Ga as reference element) Note: The box in the box plot gives the interquartile range of the values, the line in the box the median value. Circle outside the box indicate outlier (more than 1.5 box lengths from the edge of the box), and asterisk represents extreme value (more than three box lengths from the edge of the box)

Lognormal distribution plots Acero et al. (2003) suggested that background values, serving as thresholds between contaminated and uncontaminated areas, could be derived from the log-distribution curves. For metal distributions in which linearity is not observed, a break-point in the curve inflection is arbitrarily fixed and the corresponding concentration is used to represent an operational threshold between contaminated and non-contaminated (or slightly contaminated) areas. The inflection point in the lognormal distribution plot was identified according to the procedure applied by Acero et al. (2003). Previous research has shown that Be, Ga, Fe, Li, V, etc. are mainly derived from soil parent materials, Cd, Cu, Pb, Zn, etc. from traffic emissions and As, Sb, etc. from coal combustion (Wang, Qin, &Sun, 2005). In this present study, we only chose the representative metals derived from different sources to discuss the lognormal distribution plots. The lognormal distribution plots for Al, As, Be, Cd, Cu, Fe, Ga, Li, Pb, Pd, Pt, Se, V, and Zn are shown in Fig. 3. The ordinate axis is heavy metal concentration in the soils, and the abscissa is the percentage of cumulative probability. A straight line corresponds to a

15

lognormal distribution, and the slope is related to the standard deviation, r. The steeper the slope, the larger the value for r. Data points along the same line indicate that those soils are of the same type and those that deviate are anomalous (Kaminski & Landsberger, 2000). According to the distribution plots and corresponding correlation coefficients (Fig. 3), these twelve elements can broadly be classified into two groups: As, Cd, Cu, Pb, Pd, Pt, Se and Zn with poor linearity suggesting significant impact of contamination in their distribution. The inflection points may be relatively easily noted for these elements from their lognormal distribution plots. Al, Be, Fe, Ga, Li and V with better linearity (correlation coefficients are 0.989, 0.973, 0.980, 0.989, 0.984, and 0.978, respectively) than the above group indicated that their distributions were less affected by the anthropogenic contamination source. It is difficult to demarcate the inflection points in the distributions of these elements. Normalization of data Metal-reference metal normalization may be used in lieu of grain size normalization. Aluminum is a major constituent of fine-grained aluminosilicate with which the bulk of the heavy metals are associated (Loring, 1991; Schropp et al., 1990; Shine, Ika, & Ford, 1995). In our study area, Al concentration distribution is probably not significantly affected by human activities. This is why we decided to normalize the data, using Al as reference metal. The correlation coefficients for the Al-metal relation are summarized in Table 2. Inter-element relationships can provide interesting information on element sources and pathways (Manta et al., 2002). Aluminium in Xuzhou topsoil is mainly derived from soil parent materials. The correlation for Ba, Be, Co, Fe, Ga, Li, Mn and V may be strong, suggesting that the topsoil in the city of Xuzhou is also probably derived from soil parent materials and not seriously contaminated by these metals. On the other hand, no significant correlation exists between Ag, Al, As, Au, Bi, Cd, Cr, Cu, Hg, Mo, Ni, Pb, Sb, Sc, Se, Sn and Zn, indicating that these metals may probably from anthropogenic inputs. The correlation

123

100 90 80 70 60 50

1 cd (mg kg-1)

Fig. 3 Log-normal distribution plots for Pb, Cd, Cu, Zn, Li, Ga, Be, V, Se, As, Al, Fe, Pd and Pt

Environ Geochem Health (2007) 29:11–19

Pb (mg kg-1)

16

40 30

0.1

20

1

10 40 70 95 99.5 Probability (cum. %)

80 70 60 50

10

40 70 95 99.5 Probability (cum. %)

40 30 20

11

04

07 09

5 99. 59 9.99 9

200

100 90 80 70 60 50

11

04 07 09

probability (cum.%)

59 9.5 99.999

probability (cum.%)

50

100

R=0.978 P< 0.0001

R=0.984 p< 0.0001

90

40

V (mg kg-1)

Li (mg kg-1)

99.999

300 Zn (mg kg-1)

Cu (mg kg-1)

1

30

80 70 60

10

40 70 95 99.5 probability (cum.%)

99.999

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2

0.1 1

10 40 70 95 99.5 probability (cum.%)

coefficients between Pt, Pd and Al are negative (R = –0.613, –0.235, respectively). No explanation for the exceptions Pt and Pd can be given. Background levels The total concentrations of heavy metals are not the most important information for assessing

123

1

As (mg kg-1)

Se (mg kg-1)

1

99.999

10

40 70 95 99.5 probability (cum.%)

99.999

10 9 8 1

10

40 70 95 99.5 99.999 probability (cum.%)

heavy metal contamination of certain areas. The determination of natural levels of these metals is of the same importance. There are several methods to establish background values of metals (Acero et al., 2003; Celo et al., 1999; Herreweghe et al., 2003; Stercheman et al., 2000). For the background values of our study area, we have selected the method used by Celo et al. (1999). For the metal

Environ Geochem Health (2007) 29:11–19

17

Fig. 3 Continued 10 Pt (mg kg-1)

Ga (mg kg-1)

1

R=0.980 p=5.70E-4

R=0.989 p< 0.0001

1 1

10

40 70

95

0.1

99.5 99.999

1

10

Probability (cum. %)

R=0.989 p< 0.0001

4

40 70 95

99.5

99.999

Probability (cum. %)

5 4

Pt (ng g-1)

Fe (%)

3

3

1

10

40 70

95 99.5

2

1 0.9 0.8

99.999

1

10

Probability (cum. %)

40 70

95 99.5

99.999

Probability (cum. %)

8 R=0.973 p< 0.0001

Pd (ng g-1)

Al (%)

7 6

1

5 1

10

40 70

95 99.5

99.999

Probability (cum. %)

whose lognormal distribution is not linear, a break-point in the curve is arbitrarily fixed and this concentration is used to represent an operational threshold between the contaminated and noncontaminated (or slightly contaminated) areas. For Al, Be, Fe, Ga, Li, V, etc, linearity is observed in the lognormal distribution plots (Fig. 3) and the inflection point is hardly to be identified. Therefore, the corresponding concentration can not be obtained. For metal distributions in which linearity is observed, a break-point in the curve inflection is arbitrarily fixed and the corresponding concentration is used to represent an operational threshold between contaminated and non-contaminated (or slightly contaminated) areas. The background values obtained by this procedure are summarized in Table 3. In general,

1

10

40 70

95 99.5

99.999

Probability (cum. %)

these background values are comparable to those of Xuzhou uncontaminated soils except for Cd and Hg (Table 1).

Conclusions In order to obtain some characteristics of the distribution of heavy metals in Xuzhou urban soils, lognormal distribution plots as well as metal-reference-metal normalization procedures have been employed in this investigation. The lognormal distribution plots follow two different patterns for different metals. For Al, Be, Ga, Fe, Li and V the diagrams are almost linear suggesting that these metals are almost derived from soil parent materials and unaffected by anthropogenic activities. The plots for As, Cd, Cu, Pb, Pd, Pt, Se

123

18 Table 2 The Pearson’s metal–aluminum

Environ Geochem Health (2007) 29:11–19 correlation

coefficients

(R)

R Fe Ag Se Sc Ga Ba Li Bi V Pb Cu Zn Cd Ni Co Cr As Hg Sb Mn Mo Be Sn Pt Au Pd

0.446* –0.067 –0.184 –0.397 0.593** 0.453* 0.757** –0.051 0.933** –0.045 0.105 0.038 0.154 0.180 0.795** –0.006 –0.058 0.127 –0.070 0.870** –0.089 0.862** 0.294 –0.613** –0.164 –0.235

* correlation is significant at the 0.05 level. **correlation is significant at the 0.01 level Table 3 Background values (mg kg–1) of heavy metals obtained using lognormal distribution plots (unit of Au and Pd is ng g–1) Our values Ag Se Sc Ba Bi As Hg Sb Mn Sn Pb Cu Zn Cd Ni Co Cr Au Pd Mo

123

0.10 0.14 17 420 0.23 13 0.12 1.0 421 1.4 18 17 56 0.21 28 11 72 0.7 1.3 0.8

and Zn are not linear, perhaps because of intensive anthropogenic activities from which these metals are delivered to the soils. Using aluminum as a normalizer can broadly differentiate natural source and anthropogenic source for these metals studied.

References Acero, P., Mandada, J. M. A., Gomez, J., Gimeno, M. J., Auque, L. F., & Torrigo, F. J. (2003). Environmental impact of heavy metal dispersion in the Huerva River (Iberian range, NE Spain). Environmental Geology, 43, 950–956. Barefoot, R. R. (1997). Determination of platinum at trace levels in environmental and biological materials. Environmental Science & Technology, 31, 309–314. Cal-Prieto, M. J., Carlosena, A., Andrade, J. M., Martinez, M. L., Muniatequi, S., Lopez-Mahia, P., & Prada, D. (2001). Antimony as tracer of the anthropogenic influence on soils and estuarine sediments. Water, Air and Soil Pollution, 129, 248–333. Celo, V., Babi, D., Baraj, B., & Cullaj, A. (1999). An assessment of heavy metal pollution in sediment along the Albanian coast. Water, Air and Soil Pollution, 111, 235–250. Covelli, S., & Fontolan, G. (1997). Application of a normalization procedure in determining regional geochemical baseline. Environmental Geology, 30, 34–45. De Miguel, E., De Grado, J., Liamas, J. F., Martin-Dorado, A., & Mazadiego, L. F. (1998). The overlooked contribution of compost application to the trace element load in the urban soil of Madrid (Spain). The Science of Total Environment, 215, 113–122. Din, Z. B. (1992). Use of aluminum to normalize heavy metal data from estuarine and coastal sediments of strait’s of Melaka. Marine Pollution Bulletin, 24, 484–491. Fujikawa, Y., Fukui, M., & Kudo, A. (2000). Vertical distributions of trace metals in natural soil horizons from Japan. part 1. effect of soil types. Water, Air, and Soil Pollution, 124, 1–21. Helmers, E., Schwarzer, M., & Schuter, M. (1998). Comparison of palladium and platinum in environmental matrices: palladium pollution by automobile emissions? Environmental Science & Pollution Research, 5, 44–50. Herreweghe, S. V., Swennen, R., Vandecasteele, C., & Sprovieri, M. (2003). Solid phase speciation of arsenic by sequential extraction in standard reference materials and industrially contaminated soil samples. Environmental pollution, 122, 323–342. Kaminski, M. D., & Landsberger, S. (2000). Heavy metals in urban soils of East St. Louis, IL, part 1: total concentration of heavy metals in soils. Jouranl of Air & Waste Management Association, 50, 1667–1679.

Environ Geochem Health (2007) 29:11–19 Li, X., Poon, C. S., & Liu, P. S. (2001). Heavy metal contamination of urban soils and street dusts in Hong Kong. Applied Geochemistry, 16, 1361–1368. Loring, D. H. (1991). Normalization of heavy metal data from estuarine and coastal sediments. Journal of Marine Science, 48, 101–115. Manta, D. S., Angelone, M., Bellanca, A., & Neri, R. (2002). Heavy metals in urban soils: a case study from the city of Palermo (Sicily), Italy. The Science of Total Environment, 300, 229–243. Martinez Garcia, M. J., Moreno-Grau, S., Martinez Garcia, J. J., Moreno, J., Bayo, J., Guillen Derez, J. J., & Moreno-Clavel, J. (2001). Distribution of the metals lead, cadmium, copper, and zinc in the topsoil of Cartagena, Spain. Water, Air and Soil Pollution, 131, 329–347. Ordo´n˜ez, A., Loredo, J., De Miguel, E., & Charlesworth, S. (2003). Distribution of heavy metals in the street dusts and soils of an industrial city in Northern Spain. Archive of Environmental Contamination and Toxicology, 44, 160–170. Paterson, E., Sanka, M., & Clark, L. (1996). Urban soils as pollutant sinks—A case study from Aberdeen, Scotland. Applied Geochemistry, 11(1, 2), 129–131. Schropp, S. J., Lewis, F. G., Windom, H. L., Ryan, J. D., Calder, F. D., & Burney, L. C. (1990). Interpretation of metal concentrations in estuarine sediment of Florida using aluminum as a reference element. Estuarine, 13, 227–235. Sedemann, D. E. (1991). Metal pollution in sediments of Jamaica Bay, New York, USA—an urban estuary. Environmental Management, 15, 73–81. Shine, J. P., Ika, P. V., Ford, & T. E. (1995). Multivariate statistical examination of spatial and temporal pattern

19 of heavy metal contamination in New Bedford Harbour marine sediments. Environmental Science & Technology, 29, 1781–1788. Sterckeman, T., Douay, F., Proix, N., & Fourrier, H. (2000). Vertical distribution of Cd, Pb and Zn in soils near smelters in the North of France. Environmental Pollution, 107, 377–389. Summers, J. K., Wade, T. L., & Engle, V. D., & Malaeb, Z. A. (1996). Normalization of metal concentrations in estuarine sediment form Gulf of Mexico. Estuarine, 19, 581–594. Tiller, K. G. (1992). Urban soil contamination in Australia. Australian Journal of Soil Research, 30, 937–957. Tokalioglu, S., Kartal, S., & Birol, G. (2003). Application of a three-stage sequential extraction procedure for the determination of extractable metal contents in highway soils. Turk. J. Chem, 27, 333–346. Varrica, D., Dongarra, G., Sabatino, G., & Monna, F. (2003). Inorganic geochemistry of roadway dust from the metropolitan area of Palermo, Italy. Environmental Geology, 44, 222–230. Wang X. S. (2005). Accumulation-leaching characteristics of heavy metals in urban soils and its magnetic responses: a case study from the city of Xuzhou, China. Doctor’ Thesis, China University of Mining and Technology. Wang, X. S., Qin, Y., & Sun, S. X. (2005). Accumulation and sources of heavy metals in urban topsoils: a case study from the city of Xuzhou, China. Environmental Geology, 48, 101–107. Wilcke, W., Muller, S., Kanchanakool, N., & Zech, W. (1998). Urban soil contamination in Bangkok: heavy metal and aluminium partitioning in topsoils. Geoderma, 86, 211–228.

123