S CI EN C E OF TH E T OTAL EN V I RO N M EN T 4 0 7 ( 2 0 09 ) 29 3 1–2 93 8
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
w w w. e l s e v i e r. c o m / l o c a t e / s c i t o t e n v
Source apportionment of polycyclic aromatic hydrocarbons
(PAHs) in surface sediments of the Huangpu River,
Shanghai, China
Ying Liu a , Ling Chen a,⁎, Qing-hui Huang b , Wei-ying Li b , Yin-jian Tang a , Jian-fu Zhao a
a
State Key Laboratory of Pollution Control and Resources Reuse, College of Environmental Science and Engineering,
Tongji University, Shanghai, 200092 China
b
Key Laboratory of Yangtze River Water Environment, Ministry of Education, College of Environmental Science and Engineering,
Tongji University, Shanghai, 200092 China
AR TIC LE D ATA
ABSTR ACT
Article history:
We applied cluster analysis and principal component analysis (PCA) with multivariate linear
Received 17 April 2008
regression (MLR) to apportion sources of polycyclic aromatic hydrocarbons (PAHs) in surface
Received in revised form
sediments of the Huangpu River in Shanghai, China, based on the measured PAH
30 November 2008
concentrations of 32 samples collected at eight sites in four seasons in 2006. The results
Accepted 14 December 2008
indicate that petrogenic and pyrogenic sources are the important sources of PAHs. Further
Available online 5 February 2009
analysis shows that the contributions of coal combustion, traffic-related pollution and spills
of oil products (petrogenic) are 40%, 36% and 24% using PCA/MLR, respectively. Pyrogenic
Keywords:
sources (coal combustion and traffic related pollution) contribute 76% of anthropogenic
PAHs
PAHs to sediments, which indicates that energy consumption is a predominant factor of
Surface sediments
PAH pollution in Shanghai. Rainfall, the monsoon and temperature play important roles in
Source apportionment
the distinct seasonal variation of PAH pollution, such that the contamination level of PAHs
PCA/MLR
in spring is significantly higher than in the other seasons.
Cluster analysis
Brief: We apportion PAHs in surface sediments of the Huangpu River and show that coal
combustion, traffic-related pollution, and petroleum spillage are the major sources.
© 2008 Elsevier B.V. All rights reserved.
1.
Introduction
Polycyclic aromatic hydrocarbons (PAHs) containing two or
more fused benzene rings form one of the most important
classes of environmental pollutants. Due to the persistent,
toxic, mutagenic and carcinogenic characteristics of PAHs
(Zedeck, 1980; NRC, 1983), some of them are on the US EPA list
of priority pollutants. Pyrogenic and petrogenic sources are
two major origins of anthropogenic PAHs in the environment.
Pyrogenic PAHs are formed as trace contaminants by the
incomplete combustion of organic matter, such as wood, fossil
fuels, asphalt, and industrial waste. Crude and refined
petroleum contain petrogenic PAHs, and are also important
sources of PAHs. Once produced, PAHs can be widely
dispersed into the environment by atmospheric transport or
through stream pathways, and eventually accumulate in soils
and aquatic sediments.
The Huangpu River, the most important shipping artery of
Shanghai, arises in the lake district of the Shanghai Municipality (Dianshan Lake) and flows northeast past Shanghai into
the Yangtze River. Although Shanghai is one of the most
comprehensively industrial and commercial cities in China,
ranking first in population and population density, a few
studies have reported on the source apportionment of
sedimentary PAHs in Shanghai. Ren et al. (2006) reported the
distribution and sources of PAHs from dust collected in
⁎ Corresponding author. Tel./fax: +86 21 6598 4261.
E-mail address: chenling@tongji.edu.cn (L. Chen).
0048-9697/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.scitotenv.2008.12.046
2932
S CI EN CE OF T H E T OTAL EN V I RO N M EN T 4 0 7 ( 2 0 09 ) 29 3 1–2 93 8
Shanghai, and affirmed that vehicle exhaust was the main
source. Liu et al. (2007a) also characterized the PAH sources,
identifying road dust PAHs in central Shanghai areas, and
suggested that road dust PAHs mainly came from the mixing
of traffic and coal combustion and that PAH levels in most
samples in the winter were almost always higher than those
in the summer. Feng et al. (2006) investigated the characteristics of organic matter in PM2.5 in the atmosphere of Shanghai
and found a strong presence of combustion engine exhaust
emissions.
Knowledge regarding the sources and pathways of pollutants in aquatic sediments is important for effective pollution
abatement. While diagnostic ratios of PAHs have been widely
applied to identify sources in various environments (Soclo
et al., 2000; Yunker et al., 2002; Rocher et al., 2004; Zhang et al.,
2004; Wang et al., 2006; Li et al., 2006a), their use is limited due
to a lack of reliability. More sophisticated statistical
approaches have been demonstrated, including cluster analysis, principal components analysis (PCA), and chemical
mass balance (CMB). However, there are limitations in
requiring an input of source emission profiles to calculate
source contributions when a CMB model has been used to
identify and quantify sources of pollutants (Duval and
Friedlander, 1981; Li et al., 2001, 2003). PCA, which can provide
information on source contributions, in conjunction with
multivariate linear regression (MLR), has been performed to
identity and apportion PAH sources in the air, soil, and
sediment in many cities (Harrison et al., 1996; Larsen and
Baker, 2003; Li et al., 2006b; Zuo et al., 2007).
In our previous work, we reported the concentrations,
spatial distribution and sources of PAHs in surface sediments
of the Yangtze estuary, Huangpu River and Suzhou River in
Shanghai, China, and identified pyrogenic sources as important contributors of sedimentary PAHs in the Huangpu River
(Liu et al., 2008). The purpose of this work is to further identify
the major sources of sedimentary PAHs in the Huangpu River
by cluster analysis and principal component analysis, and to
carry out quantitative sources apportionment and to discuss
seasonal variations of PAH pollution based on the PCA/MLR.
Methylnaphthalenes and 18 PAHs, including 16 PAHs identified by the US EPA as priority pollutants, were monitored in
surface sediments. A total organic carbon (TOC) analysis was
also carried out to normalize the sedimentary PAH concentrations of Huangpu River to reduce the effect of sediment
property on PAH concentration. The results of this study will
provide valuable information for regulatory actions to
improve the environmental quality of Huangpu River,
Shanghai.
2.
Experimental methods
2.1.
Sample collection and analysis
Eight sampling stations were selected along the Huangpu
River in Fig. 1; details of the sampling stations are listed
elsewhere (Liu et al., 2008). Surface sediment samples were
collected at 8 sampling stations using a grab dredge in April
(spring), August (summer), October (autumn) and December
(winter) of 2006. A total of 32 samples were used in this work.
Surface sediment samples were air-dried in the dark, sieved to
<0.076 mm (200 mesh) after removing stones and residual
roots, and stored at −4 °C until analysis.
16 PAHs characterized by the US EPA as priority pollutants
were analyzed, including naphthalene (Nap), acenaphthylene
(AcNy), fluorene (Fl), acenaphthene (AcNe), phenanthrene
(PhA), anthracene (An), fluoranthene (FlA), pyrene (Py), benz
[a]anthracene (BaA), chrysene (Chy), benzo[b]fluoranthene
(BbF), benzo[k]fluoranthene (BkF), benzo[a]pyrene (BaP),
indeno[1,2,3-cd]pyrene (IP), benzo[ghi]perylene (BghiP), and
dibenz[a,h]anthracene (DBahA). In addition to the 16 priority
PAHs, benzo[e]pyrene (BeP), perylene (Pery), 1-methylnaphthalene and 2-methylnapthalene were also analyzed.
Two isomers of methylnaphthalenes were pooled as total
methylnaphthalenes (MNap).
Sample extraction and cleanup were carried out according
to Method 3540C and Method 3630C published by US EPA
(USEPA, 1996). After extraction and cleanup, samples were
concentrated and adjusted to 1 mL volume for analysis. PAH
analysis was carried out by high performance liquid chromatography (HPLC) with a photodiode array detector. Identification of PAHs was based on retention time and the ultraviolet
spectra of PAH standards. The quantification was performed
by the external standard method. The ultraviolet measuring
wavelengths include 218 nm (NaP), 223 nm (MNaP), 226 nm
(AcNe and AcNy), 249 nm (IP), 254 nm (Fl, PhA and An), 266 nm
(Chy), 286 nm (FlA and BaA), 300 nm (BbF, BkF, BaP, DBahA and
BghiP), 330 nm (BeP), 334 nm (Py) and 433 nm (Pery). Detailed
procedures for sample preparation, extraction, cleanup,
measurement and quality control are described elsewhere
(Liu et al., 2007b, 2008). Method detection limits were 1–19 ng/
g-dw, and spiked recoveries of PAHs were 87–113%. All of the
samples taken were analyzed in triplicate, and the relative
standard deviation was less than 20%.
Total organic carbon (TOC) analysis was performed with
the Shimadzu TOC-Vcpn analyzer with the solid sample
module (SSM-5000A). The overall precision of measurements
was less than 3% (n = 3).
Fig. 1 – Sediment sampling locations in the Huangpu River.
2933
S CI EN C E OF TH E T OTAL EN V I RO N M EN T 4 0 7 ( 2 0 09 ) 29 3 1–2 93 8
PCA factor scores and the standardized normal deviation of
total PAH concentrations (normalization first to organic
carbon and then scaled to mean and standard deviation) as
independent and dependent variables, respectively (Larsen
and Baker, 2003). The regression was run using a forward
stepwise method. The standardized regression coefficients
were used to represent the relative contributions from various
sources (Larsen and Baker, 2003; Zuo et al., 2007).
Fig. 2 – Hierarchical dendogram for 18 PAHs in the Huangpu
River sediments using average linkage between groups and
Pearson correlation as measure interval.
2.2.
Statistical analyses
Before statistical analysis of data, we replaced undetectable
values by a random number between zero and the limit of
detection, and eliminated AcNy as it was undetectable in most
of the samples. Different hydraulic conditions at the different
sampling locations lead to different deposition rates and to
different sediment properties, e.g., TOC content and particle
size distribution. Many researchers have indicated that
organic carbon is an important controlling factor of the
sorption of PAHs on sediments (Karickhoff et al., 1979; Wang
et al., 2001; Zakaria et al., 2002). In order to reduce this effect,
we sieved sediment samples to <0.076 mm (200 mesh) before
analysis, and normalized PAH concentrations to TOC contents. At the same time, normalization to TOC produced a
normalized dataset for the following statistical analysis.
Statistical analyses, including the Kolmogorov–Smirnov
(K–S) test, ANOVA, cluster analysis, principal components
analysis, and multivariate linear regression, were performed
using SPSS 13.0 for Windows. The K–S test was carried out to
test the frequency distribution of PAH data, and all of the
variables after normalization to TOC achieved a normal
distribution with P > 0.05. A repeated measures one-way
ANOVA procedure was performed to test the significant
differences of the PAH dataset. The contents of the individual
PAHs were hierarchically clustered using weighted average
linkage between the groups and the Pearson correlation for
the cluster intervals (Zhang et al., 2006). PCA, as a multivariate
analytical tool, was used to reduce the set of original variables
(measured PAH contents in sediment samples) and to extract
a small number of latent factors (principal components) to
analyze the relationships among the observed variables. In
detail, all factors with eigenvalues over 1 were extracted
according to KMO and Bartlett's test of sphericity, and were
rotated using the Varimax method. MLR was conducted using
3.
Results and discussion
3.1.
Source estimates from cluster analysis
Cluster analysis was performed to identify the homogeneous
groups of individual PAHs in the Huangpu River sediments.
The result of the cluster analysis is shown in the hierarchical
dendogram (Fig. 2), which distinguishes the 18 individual
PAHs into three major groups. The first group, which includes
MNap, Nap, Fl and AcNe, belongs to the low molecular weight
PAHs with 2–3 rings or alkyl-substituted PAHs, which are
abundant in petrogenic sources mainly caused by petroleum
spills, e.g., fresh or used crankcase oil, crude and fuel oil (Marr
et al., 1999; Utvik et al., 1999; Dobbins et al., 2006; Gonzalez
et al., 2006; Ye et al., 2006). The second major group is
subdivided into two subgroups. The first subgroup contains
BbF, BkF, BaP, BeP, DBahA and BghiP, which are the high
molecular weight PAHs with 5–6 rings. The second subgroup
consists of PhA, FlA, Py, BaA, Chy and IP, most of which are 4
ring PAHs. Both of these subgroups are usually detected in
pyrogenic source, e.g., combustion of coal, wood, vehicle fuel
and waste tire (Levendis et al., 1998; Zakaria et al., 2002; Wang
et al., 2007). The third major group contains only two
components of An and Pery, and currently has an unknown
source, which is further discussed in the PCA.
3.2.
Diagnostic ratios of PAHs
The relative abundances or diagnostic ratios are useful
indicators of PAH sources because isomer pairs are diluted to
a similar extent upon mixing with natural particulate matter,
and are distributed similarly to other phases as they have
comparable thermodynamic partitioning and kinetic mass
transfer coefficients (Dickhut et al., 2000). Diagnostic ratios of
PAHs, such as the ratio of LMW (2–3 ring PAHs) to HMW (4–6
ring PAHs), An/(PhA + An) and FlA/(FlA + Py), can be used to
identify the possible emission sources, as summarized in
Table 1 – The range of diagnostic ratios for PAHs sources
Diagnostic ratio Petrogenic Pyrogenic
LMW/HMW
>1
<1
An/(PhA + An)
<0.1
>0.1
FlA/(FlA + Py)
<0.4
>0.4
References
Soclo et al. (2000),
Rocher et al. (2004),
Wang et al. (2006)
Yunker et al. (2002),
Zhang et al. (2004),
Li et al. (2006a)
Yunker et al. (2002),
Zhang et al. (2004),
Li et al. (2006a)
2934
S CI EN CE OF T H E T OTAL EN V I RO N M EN T 4 0 7 ( 2 0 09 ) 29 3 1–2 93 8
Fig. 3 – PAH cross plots for the ratios of An/(An + PhA) vs. FlA/
(FlA + Py).
Table 1. The ratios of LMW/HMW in the Huangpu River
sediments range from 0.12 to 0.59 with a mean of 0.33. The
ratio of LMW/HMW is < 1, indicating a predominance of
combustion source (Soclo et al., 2000; Rocher et al., 2004;
Wang et al., 2006). In Fig. 3, the ratios of An/(An + PhA) range
ttfrom 0.12 to 0.39 with a mean of 0.27, and the ratios of FlA/
(FlA + Py) range from 0.39 to 0.57 with a mean of 0.49. These
are similar to measures for combustion, especially coal
combustion of power plants and liquid fossil fuel (vehicle
and crude oil) combustion (Yunker et al., 2002; Zhang et al.,
2004; Li et al., 2006a). Therefore, pyrogenic sources are the
major sources of PAHs in the Huangpu River sediments.
3.3.
Source estimates from principal components analysis
The purpose of PCA is to represent the total variability of the
original PAH data with a minimum number of factors. By
critically evaluating the factor loadings, an estimate of the
chemical source responsible for each factor can be made
(Larsen and Baker, 2003). The rotated factors of 18 normalized
PAHs (by TOC) from the Huangpu River sediments are
presented in Table 2. The three factors account for 83.9% of
the variability in the data. Factor 1, which explains 43.9% of
total variance, is dominated by PhA, FlA, Py, Chy, BaA, BbF,
BkF, BaP, DBahA, IP, BghiP and BeP. Factor 2, contributing
20.1% of total variance, is highly weighted by An and Pery.
Factor 3, which explains 19.9% of total variance, is dominated
by MNap, Nap, Fl and AcNe. The result of PCA is similar to that
of the cluster analysis above. Factor 3, corresponding to the
first group, represents a petrogenic source; Factor 1, corresponding to the second group, represents pyrogenic source
PAHs; and Factor 2, corresponding to the third group,
represents an unknown source. In Table 2, the loadings of
An and Pery are 0.87 and 0.89, respectively. It is unusual to
have An and Pery covary in environmental samples. Pery is a
natural compound formed from biogenic precursors (e.g.,
perylenequinone pigments) during early diagenesis, while
trace concentrations of perylene are generated through
combustion of fossil fuel (Hites et al., 1980; Venkatesan,
1988; Boonyatumanond et al., 2006; Ye et al., 2006). The
biogenic production of An is negligible in comparison to Pery,
because An is susceptible to biogradation (Santos et al., 2008),
hence Pery might be primarily combustion generated. Therefore, Factor 3, namely the third group, is believed to be an
unknown combustion source, although we cannot explain
why An was classified in this group.
According to the cluster analysis, the pyrogenic source can
be subdivided into two subgroups, which represent two kinds
of different pyrogenic sources. However, the results of PCA
cannot differentiate the two subsets of pyrogenic sources,
even if the number of principal components is set as 4. Since
pyrogenic sources of PAHs are the main objectives investigated in order to control PAH pollution in the Huangpu River
sediments, the unknown source can be ignored in this
investigation. We therefore removed data about An and Pery
from the data matrix and performed the PCA again in order to
further investigate the pyrogenic sources of PAHs.
The rotated factors of the 16 PAHs without Pery and An are
shown in Table 3. There are again three factors, accounting for
84.8% of the variability in the data.
The first factor is responsible for 35.0% of the total variance.
This factor is heavily weighted in BkF, BaP, DBahA, BghiP and
BeP, along with moderate loadings for Py, BbF and IP. These PAH
components, the high molecular weight PAHs with 5–6 rings,
basically belong to the first subgroup of Group 2 of the cluster
analysis. The source this factor represents appears to be road
dust collected from the Shanghai urban area (Ren et al., 2006; Liu
et al., 2007a) and is vehicular (gasoline and diesel) in nature
(Harrison et al., 1996; Larsen and Baker, 2003; Ye et al., 2006; Zuo
et al., 2007). BghiP has been identified as a tracer of auto
emissions because it was found to be enriched in a traffic tunnel
along with BaP (Harrison et al., 1996; Larsen and Baker, 2003;
Table 2 – Rotated component matrix of 18 PAHs from the
Huangpu River sediment a
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
PAH
MNap
Nap
Fl
AcNe
PhA
An
FlA
Py
Chy
BaA
BbF
BkF
BaP
DBahA
IP
BghiP
BeP
Pery
Estimated source
Variance (%)
a
b
Principal components
1
2
3
0.03
0.03
0.30
0.32
0.69
0.20
0.90
0.91
0.88
0.93
0.82
0.72
0.74
0.68
0.79
0.70
0.76
− 0.17
−0.08
−0.10
0.46
0.35
0.10
0.87
−0.26
0.23
−0.12
−0.04
0.34
0.36
0.58
0.59
0.18
0.64
0.45
0.89
0.95 b
0.94
0.74
0.82
0.49
0.08
0.18
0.18
0.22
0.15
0.10
−0.20
−0.12
−0.17
0.18
0.15
0.21
0.14
Pyrogenic
43.9
Unknown
20.1
Petrogenic
19.9
Rotation method: Varimax with Kaiser normalization.
Bold loadings > 0.70.
S CI EN C E OF TH E T OTAL EN V I RO N M EN T 4 0 7 ( 2 0 09 ) 29 3 1–2 93 8
Table 3 – Rotated component matrix of 16 PAHs from the
Huangpu River sediment a
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
PAH
MNap
Nap
Fl
AcNe
PhA
FlA
Py
Chy
BaA
BbF
BkF
BaP
DBahA
IP
BghiP
BeP
Estimated source
Variance (%)
a
b
Principal components
1
2
3
− 0.17
− 0.19
0.41
0.34
0.36
0.23
0.65
0.31
0.43
0.69
0.77
0.91
0.88
0.56
0.88
0.78
0.13
0.15
0.11
0.17
0.59
0.92
0.68
0.86
0.85
0.55
0.32
0.30
0.24
0.58
0.28
0.39
0.93 b
0.91
0.80
0.87
0.49
0.12
0.21
0.18
0.13
0.15
− 0.11
− 0.01
− 0.06
0.21
0.26
0.29
Traffic
35.0
Coal
27.0
Petrogenic
22.8
Rotation method: Varimax with Kaiser normalization.
Bold loadings > 0.70.
(Levendis et al., 1998). Some researchers have also reported PhA,
FlA, Py as predominant in coal combustion profiles (Harrison
et al., 1996; Zuo et al., 2007). In Shanghai, coal is the most
important energy source and is used widely for industrial and
domestic purposes, especially in the steel and power industry.
The Shanghai municipal electric power supply is mainly derived
from coal-burning power plants. It is reasonable to assign this
factor to coal combustion.
The third factor is responsible for 22.8% of the total
variance, and is heavily weighted in MNap, Nap, Fl and
AcNe, the same as the result of the former PCA. This factor
is suggested to be indicative of volatilization or spill of
petroleum-related products (Marr et al., 1999; Utvik et al.,
1999; Zakaria et al., 2002; Luca et al., 2004; Dobbins et al., 2006;
Gonzalez et al., 2006; Wang et al., 2006; Ye et al., 2006), e.g.,
from the waterway transportation industry. This factor is
believed to be the petrogenic source of PAHs.
In general, the first PCA of PAH data shows that there are
three PAH sources in the Huangpu River sediments, namely
pyrogenic, petrogenic and an unknown combustion source.
The second PCA of data without Pery and An (the unknown
source) divides the pyrogenic source of PAHs into two subsets,
one traffic-related and the other due to coal combustion.
3.4.
Boonyatumanond et al., 2007). The higher level of BkF relative to
other PAHs is suggested to indicate diesel vehicles (Venkataraman et al., 1994; Larsen and Baker, 2003). On the other hand, this
factor is moderately weighted in IP, which has also been found
in both diesel and gas engine emissions (May and Wise, 1984;
Larsen and Baker, 2003) and gasoline vehicle soot (Boonyatumanond et al., 2007). Therefore, this factor is selected to
represent the traffic-related source of PAHs.
The second factor is responsible for 27.0% of the total
variance. This factor is predominately composed of FlA, Chy
and BaA (4-ring PAHs) with moderate loadings of PhA, Py, BbF
and IP. There are similar PAH components between this factor
and the second subgroup of Group 2 in the cluster analysis. The
literature reports that 4-ring PAHs are abundant in the road dust
in Bangkok city (Boonyatumanond et al., 2007) and Kuala
Lumpur city (Zakaria et al., 2002), but the abundance of 4-ring
PAHs is lower in the road dust in Shanghai city (Ren et al., 2006;
Liu et al., 2007a) compared with these. The ratio of PAHs (4 rings)
to PAHs (5–6 rings), abbreviated PAHs(4)/PAHs(5 + 6), is greater
than 1 in Bangkok and Kuala Lumpur, according to the PAH
profiles of road dust (Zakaria et al., 2002; Boonyatumanond et al.,
2007). However, PAHs(4)/PAHs(5+ 6) is 0.3–0.8 in the central
Shanghai area (Liu et al., 2007a) and 0.5–1.1 in the Shanghai
urban area (Ren et al., 2006). In this paper, we show that PAHs(4)/
PAHs(5 + 6) in the sediment, 0.8–1.4 with a mean of 1.1, is more
than that in the road dust from Shanghai. This indicates that
there should be an additional PAH source (not traffic-related)
leading to the higher ratio of PAHs(4)/PAHs(5 + 6). Duval and
Friedlander (1981) considered PhA, FlA, Py, BaA and Chy as
markers of coal combustion. According to PAH data from the
combustion of pulverized coal and tire crumbs at 1000 °C (near
the temperature of coal combustion in the coal-burning power
plant), PhA, FlA and Py are the dominant PAHs with lower
concentrations of 5–6 ring PAHs detected in furnace effluents
2935
Contribution of PAH sources
The ultimate goal of source apportionment is to determine the
percent contribution of different PAH sources for a given
samples. We unveiled the major sources of sedimentary PAHs
in the Huangpu River using diagnostic ratios, cluster analysis
and PCA. We then calculated the percent contributions of the
major sources using multivariate linear regression (MLR) from
the PCA factor scores and the standardized normal deviation
of total PAH concentrations as our independent and dependent variables, respectively. Several authors (Harrison et al.,
1996; Larsen and Baker, 2003; Zuo et al., 2007) have reported
applying PCA/MLR to apportion sources of PAHs in the urban
atmosphere and surface soils.
The mean percent contribution of source i is the ratio of the
regression coefficient for factor i to the sum of all the
regression coefficients, according to the description in the
literature (Larsen and Baker, 2003). The factor scores are from
the result of PCA without Pery and An. The R squared value for
MLR is 0.983 and the P values for the regression coefficients
are less than 0.05. Thus, the mean contribution percents are
40% for the vehicular source, 36% for the coal combustion
source, and 24% for the petrogenic source.
The traffic-related source (40%) is the first contributor to
the PAHs. There are two origins of traffic-related PAHs in the
Huangpu River. For one thing, traffic-related PAHs in road dust
can enter the sediments through urban runoff (Zakaria et al.,
2002; Murakami et al., 2005; Boonyatumanond et al., 2006,
2007). In recent decades, a rapid increase in motor vehicles in
Shanghai has aggravated PAH pollution in the Huangpu River
sediments. For example, the number of taxis increased from
11,298 in 1990 to 48,022 in 2006 (Yin, 2007). Exhaust from cargo
vessels and passenger ferries in the Huangpu River also plays
an important role as a PAH contributor, and should not be
ignored in the discussion of sedimentary PAHs. In fact, cargo
vessels are responsible for most coal transportation to coal-
2936
S CI EN CE OF T H E T OTAL EN V I RO N M EN T 4 0 7 ( 2 0 09 ) 29 3 1–2 93 8
burning power plants located along the Huangpu River.
Passenger ferries also still serve the public, despite the rapid
development of traffic facilities across the Huangpu River.
Furthermore, compared with automobiles, more PAHs are
emitted from vessels and ferries because most of them are not
equipped with catalytic converters (Rogge et al., 1993).
Coal combustion (36%) is the second contributor to PAHs. In
recent years, most coal in Shanghai is used to generate electric
power due to a control on the emission of SO2, NOx and
suspended particles. However, coal consumption in the power
industry in Shanghai is continually increasing, from 3.0 million
tons in 1980 to 10.6 million tons in 1990 to 25.0 million tons in
2005 (SLCO, 2008a), causing a high percent contribution of coal
combustion. Several large-scale coal-burning power plants are
located along the Huangpu River, leading to an easy input of
coal combustion PAHs into the Huangpu River.
Petrogenic sources (24%) form the third contributor to PAHs.
Petrogenic sources include crude oil and refined products (e.g.,
crude and fuel oil). Zakaria et al. (2002) thought that there were
two major input routes to aquatic environments, namely
(1) spillage and dumping of waste crankcase oil and (2) leakage
of crankcase oils from vehicles onto road surfaces, with
subsequent washout by street runoff. In China, waste crankcase
oil has been included in the national list of hazardous wastes,
and its dumping is illegal and punished by law. In general, most
used crankcase oil is recycled by garages in vehicle maintenance. However, if crankcase oil is replaced in private, it may
be improperly stored or deposited, allowing for indiscriminate
spillage to the ground and streets or even for oil to be poured
directly into drains or the water environment, as there is
currently no efficient recycling program for used crankcase oil in
Shanghai. In addition, vessel maintenance and fuel supply may
lead to input of petrogenic PAHs into the sediment.
3.5.
Seasonal variation
Contamination of the bottom sediment is often a gradual
process. However, the repeated measures one-way ANOVA
results indicate a distinct seasonal variation, that is, the
concentrations of total PAHs are significantly higher in the
spring than in the other seasons (at the significant level of
0.05). The mean concentrations of total PAHs in different
seasons are listed in Table 4. In order to further investigate the
causes, the contributions of three important sources in
different seasons are calculated in Table 4, according to the
following formula of Larsen and Baker (2003).
Contribution of source i ðng=mg TOCÞ = mean
X Bi + Bi dPAHs FSi
Bi =
X
16
PAHs
where Bi/ΣBi is the ratio of the regression coefficient for factor i
to the sum of all of the regression coefficients, FSi is the factor
score for factor i, and δPAHs is the standard deviation of total
PAH concentrations after normalization to organic carbon.
A higher concentration of total PAHs in the spring results
from a higher contribution of coal combustion and petrogenic
sources. The repeated measures one-way ANOVA results
suggest that the contributions of coal combustion in spring are
significantly higher than those in the summer and autumn, and
that those of the petrogenic source in spring are higher than in
the other seasons (at the significant level of 0.05). As for coal
combustion, 71.3 BkW h of electric power were generated during
2006, most from coal-burning power plants in Shanghai (SMEPC,
2008). This indicates that the consumption of electric power can
reflect the consumption of coal in Shanghai. Statistical data in
Table 4 show that the consumption of electric power in the
summer and winter (19.9 and 18.4 BkW h) is more than that in
the spring and autumn (16.4 and 16.9 BkW h). This results in
higher PAH pollution from coal combustion in the summer and
winter, which disagrees with the result of PAH source apportionment (higher PAH pollution in the spring). There are two
reasons leading to this disagreement. For one thing, winter in
Shanghai is a dry season with 102 mm of average seasonal
rainfall (~9% of average annual rainfall), and a rainy period
follows from April 15th to May 15th with ~14% of average annual
rainfall (157 mm) (SLCO, 2008b), leading to PAHs in the air and
soil generated in the winter being brought into the Huangpu
River in the spring through rainfall and surface runoff. In
addition, as a coastal city, Shanghai is affected by summer
monsoons, which bring in clean oceanic winds in the summer
that dilute local air pollutants (Feng et al., 2006); PAHs from coal
combustion are typical air pollutants. As for petrogenic PAHs,
the lower temperature in the winter and spring is an important
factor, because low temperature decreases the evaporation of
petrogenic PAHs (Feng et al., 2006). Meanwhile, rainfall in the
spring brings particles with petrogenic and coal combustion
PAHs into the Huangpu River. For the traffic-related source, the
contribution in the spring is significantly lower than that in
Table 4 – Seasonal differentiation of PAHs sources
Season a
Spring
Summer
Autumn
Winter
a
Temperature
(°C) b
Electric power
(BkW h) c
Total PAHs
(ng/mg-TOC) d
8.4–19
23–27
12–23
3.6–6.2
16.4
19.9
16.9
18.4
137.2
102.5
105.1
103.4
Pollution source of PAHs (ng/mg-TOC) e
Traffic
Coal combustion
Petrogenic
40.2
33.7
57.2
48.8
54.1
34.5
33.1
40.1
35.3
28.7
21.1
21.6
Spring in Shanghai is March–May, summer is June–August, autumn is September–November, and winter is December–next February.
Data were official statistics from 1951 to 1990 (SLCO, 2008b).
c
Seasonal loads of Shanghai electric power grid, and unit is billion kilowatt-hours. Data from Match 2006 to February 2007 were provided by
Shanghai Municipal Electric Power Company (SMEPC, 2008).
d
Mean of total PAHs concentrations of 8 sampling locations; total PAHs is sum of observed concentrations of 16 PAHs listed in Table 3.
e
Mean source contributions of 8 sampling locations.
b
S CI EN C E OF TH E T OTAL EN V I RO N M EN T 4 0 7 ( 2 0 09 ) 29 3 1–2 93 8
2937
variate analysis methods show that the contributions of coal
combustion, traffic-related pollution and spill of oil products
are dominant in the Huangpu River sediments. The results of
diagnostic ratios show that pyrogenic sources are the major
source of PAHs. PCA/MLR further apportions the sources'
contributions, and the results show that the contributions of
coal combustion, traffic-related pollution and spill of oil
product are 40%, 36% and 24%, respectively. The pyrogenic
sources (coal combustion and traffic-related pollution) contribute 76% of anthropogenic PAHs to the Shanghai sediments.
Energy consumption is a predominant reason for PAH pollution in Shanghai.
Sedimentary PAH pollution is significantly higher in the
spring than in the other seasons. The higher concentrations in
the spring are attributed to the higher contribution of coal
combustion and petrogenic sources. Rainfall, monsoon and
temperature play important roles in producing the distinct
seasonal variation of sedimentary PAHs.
Acknowledgements
This work was supported by the National Natural Science
Foundation of China (Nos. 20477030 and 40601095) and the
Shanghai Science and Technology Commission (Nos.
05JC14059 and 04JC14072).
REFERENCES
Fig. 4 – Score plots of principal components analysis.
the autumn (at the significant level of 0.05). Li et al. (2006b)
suggested that when the Chinese traditional Spring Festival
takes place in the spring with few vehicles and vessels on the
move (people taking vacation out of the city), the traffic-related
pollution is low.
Factor score plots of principal components analysis also
show a seasonal variation of sedimentary PAHs in the Huangpu
River. Fig. 4a and b shows score plots of PC1 vs. PC2 and PC1 vs.
PC3, respectively. Fig. 4 illustrates that samples collected in the
spring and autumn are clustered in respective areas of the
diagram. Samples collected in the autumn have higher traffic
factor scores (PC1) and lower factor scores of coal combustion
(PC2) and petrogenic sources (PC3). On the other hand, scores in
the spring are higher in the coal combustion (PC2) and
petrogenic factors (PC3), and lower in the traffic factor (PC1).
This indicates that there is a distinct variation between spring
and autumn for the contributions of PAH sources.
4.
Conclusion
The combination of cluster analysis and principal component
analysis is effective for identifying PAHs sources. Both multi-
Boonyatumanond R, Wattayakorn G, Togo A, Takada H. Distribution
and origins of polycyclic aromatic hydrocarbons (PAHs) in
riverine, estuarine, and marine sediments in Thailand. Mar
Pollut Bull 2006;52:942–56.
Boonyatumanond R, Murakami M, Wattayakorn G, Togo A, Takada
H. Sources of polycyclic aromatic hydrocarbons (PAHs) in street
dust in a tropical Asian mega-city, Bangkok, Thailand. Sci Total
Environ 2007;384:420–32.
Dickhut RM, Canuel EA, Gustafson KE, Liu K, Arzayus KM, Walker
SE, et al. Automotive sources of carcinogenic polycyclic
aromatic hydrocarbons associated with particulate matter in
the Chesapeake Bay Region. Environ Sci Technol
2000;34:4635–40.
Dobbins RA, Fletcher RA, Benner Jr BA, Hoeft S. Polycyclic aromatic
hydrocarbons in flames, in diesel fuels, and in diesel emissions.
Combust Flame 2006;144:773–81.
Duval MM, Friedlander SK. Source resolution of polycyclic aromatic
hydrocarbons in the Los Angeles atmospheres. Application of a
CMB with first order decay. Washington, DC: U.S. Government
Printing Office; 1981. U.S. EPA Report EPA-600/2-81-161.
Feng J, Chan CK, Fang M, Hu M, He L, Tang X. Characteristics of
organic matter in PM2.5 in Shanghai. Chemosphere
2006;64:1393–400.
Gonzalez JJ, Vinas L, Franco MA, Fumega J, Soriano JA, Grueiro G, et al.
Spatial and temporal distribution of dissolved/dispersed
aromatic hydrocarbons in seawater in the area affected by the
Prestige oil spill. Mar Pollut Bull 2006;53:250–9.
Harrison RM, Smith DJT, Luhana L. Source apportionment of
atmospheric polycyclic aromatic hydrocarbons collected from
an urban location in Birmingham, U.K. Environ Sci Technol
1996;30:825–32.
Hites RA, Laflamme RF, Windsor Jr JG, Farrington JW, Deuser WG.
Polycyclic aromatic hydrocarbons in an anoxic sediment core
2938
S CI EN CE OF T H E T OTAL EN V I RO N M EN T 4 0 7 ( 2 0 09 ) 29 3 1–2 93 8
from the Pettaquamscutt River. Geochim Cosmochim Acta
1980;44:873–8.
Karickhoff SW, Brown DS, Scott TA. Sorption of hydrophobic
pollutants on natural sediments. Water Res 1979;13:241–8.
Larsen III RK, Baker JE. Source apportionment of polycyclic aromatic
hydrocarbons in the urban atmosphere: a comparison of three
methods. Environ Sci Technol 2003;37:1873–81.
Levendis YA, Atal A, Carlson JB. On the correlation of CO and PAH
emissions from the combustion of pulverized coal and waste
tires. Environ Sci Technol 1998;32:3767–77.
Li K, Christensen ER, Van Camp RP, Imamoglu I. PAHs in dated
sediments of Ashtabula River, Ohio, USA. Environ Sci Technol
2001;35:2896–902.
Li A, Jang J-K, Scheff PA. Application of EPA CMB8.2 model for
source apportionment of sediment PAHs in Lake Calumet,
Chicago. Environ Sci Technol 2003;37:2958–65.
Li G, Xia X, Yang Z, Wang R, Voulvoulis N. Distribution and sources
of polycyclic aromatic hydrocarbons in the middle and lower
reaches of the Yellow River, China. Environ Pollut
2006a;144:985–93.
Li J, Zhang G, Li XD, Qi SH, Liu GQ, Peng XZ. Source seasonality of
polycyclic aromatic hydrocarbons (PAHs) in a subtropical city,
Guangzhou, South China. Sci Total Environ 2006b;355:145–55.
Liu M, Cheng SB, Ou DN, Hou LJ, Gao L, Wang LL, et al.
Characterization, identification of road dust PAHs in central
Shanghai areas, China. Atmos Environ 2007a;41:8785–95.
Liu Y, Chen L, Tang Y, Huang Q, Zhao J. Determination of trace
polycyclic aromatic hydrocarbons in surface sediments of
Huangpu River by high performance liquid chromatography.
Chin J Chromatogr 2007b;25:356–61.
Liu Y, Chen L, Zhao JF, Huang QH, Zhu ZL, Gao HW. Distribution
and source of polycyclic aromatic hydrocarbons in surface
sediments of rivers and an estuary in Shanghai, China. Environ
Pollut 2008;154:298–305.
Luca GD, Furesi A, Leardi R, Micera G, Panzanelli A, Piu PC, et al.
Polycyclic aromatic hydrocarbons assessment in the sediments
of the Porto Torres Harbor (Northern Sardinia, Italy). Mar Chem
2004;86:15–32.
Marr LC, Kirchstetter TW, Harley RA, Miguel AH, Hering SV,
Hammond SK. Characterization of polycyclic aromatic
hydrocarbons in motor vehicle fuels and exhaust emissions.
Environ Sci Technol 1999;33:3091–9.
May WE, Wise SA. Liquid chromatographic determination of
polycyclic aromatic hydrocarbons in air particulate extracts.
Anal Chem 1984;56:225–32.
Murakami M, Nakajima F, Furumai H. Size- and density-distributions
and sources of polycyclic aromatic hydrocarbons in urban road
dust. Chemosphere 2005;61:783–91.
NRC. Polycyclic aromatic hydrocarbons: evaluation of sources and
effects. Washington, D.C.: National Academy Press; 1983.
Ren Y, Zhang Q, Chen J. Distribution and source of polycyclic
aromatic hydrocarbons (PAHs) on dust collected in Shanghai,
People's Republic of China. Bull Environ Contam Toxicol
2006;76:442–9.
Rocher V, Azimi S, Moilleron R, Chebbo G. Hydrocarbons and
heavy metals in the different sewer deposits in the Le Marais'
catchment (Paris, France): stocks, distributions and origins. Sci
Total Environ 2004;323:107–22.
Rogge WF, Hildemann LM, Mazurek MA, Cass GR, Simoneit BRT.
Sources of fine organic aerosol. 2. Noncatalyst and
catalyst-equipped automobiles and heavy-duty diesel trucks.
Environ Sci Technol 1993;27:636–51.
Santos EC, Jacques RJS, Bento FM, Peralba MDCR, Selbach PA, Sa
ELS, et al. Anthracene biodegradation and surface activity by
an iron-stimulated Pseudomonas sp. Bioresour Technol
2008;99:2644–9.
SLCO (Shanghai Local Chronicle Office). 2008a. from:http://www.
shtong.gov.cn/node2/node2245/node4441/node58151/index.
html,2008, April.
SLCO (Shanghai Local Chronicle Office). 2008b. from http://www.
shtong.gov.cn/node2/node2245/node65523/node65528/
node65574/node65678/userobject1ai61022.html,2008, April.
SMEPC (Shanghai Municipal Electric Power Company). 2008. from:
http://sd.smepc.com/list_1_11086960665005265648.htm,2008,
April.
Soclo HH, Garrigues P, Ewald M. Origin of polycyclic aromatic
hydrocarbons (PAHs) in coastal marine sediments: case studies
in Cotonou (Benin) and Aquitaine (France) areas. Mar Pollut
Bull 2000;40:387–96.
USEPA. Test methods for evaluating solid waste, physical/chemical
methods SW-846. Washington, DC, USA: Office of solid waste
and emergency response; 1996.
Utvik TIR, Durell GS, Johnsen S. Determining produced water
originating polycyclic aromatic hydrocarbons in North Sea
waters: comparison of sampling techniques. Mar Pollut Bull
1999;38:977–89.
Venkataraman C, Lyons JM, Friedlander SK. Size distributions of
polycyclic aromatic hydrocarbons and elemental carbon. 1.
Sampling, measurement methods, and source characterization.
Environ Sci Technol 1994;28:555–62.
Venkatesan MI. Occurrence and possible sources of perylene in
marine sediments — a review. Mar Chem 1988;25:1-27.
Wang X-C, Zhang Y-X, Chen RF. Distribution and partitioning of
polycyclic aromatic hydrocarbons (PAHs) in different size
fractions in sediments from Boston Harbor, United States. Mar
Pollut Bull 2001;42:1139–49.
Wang X-C, Sun S, Ma H-Q, Liu Y. Sources and distribution of
aliphatic and polyaromatic hydrocarbons in sediments of
Jiaozhou Bay, Qingdao, China. Mar Pollut Bull 2006;52:129–38.
Wang Z, Li K, Lambert P, Yang C. Identification, characterization and
quantitation of pyrogenic polycylic aromatic hydrocarbons and
other organic compounds in tire fire products. J Chromatogr A
2007;1139:14–26.
Ye B, Zhang Z, Mao T. Pollution sources identification of polycyclic
aromatic hydrocarbons of soils in Tianjin area, China.
Chemosphere 2006;64:525–34.
Yin Y. Shanghai yearbook (2007). Shanghai, China: Shanghai
Nianjian Press; 2007.
Yunker MB, Macdonald RW, Vingarzan R, Mitchell HR, Goyette D,
Sylvestre S. PAHs in the Fraser River basin: a critical appraisal
of PAH ratios as indicators of PAH source and composition. Org
Geochem 2002;33:489–515.
Zakaria MP, Takada H, Tsutsumi S, Ohno K, Yamada J, Kouno E, et al.
Distribution of polycyclic aromatic hydrocarbons (PAHs) in rivers
and estuaries in Malaysia: a widespread input of petrogenic
PAHs. Environ Sci Technol 2002;36:1907–18.
Zedeck MS. Polycyclic aromatic hydrocarbons: a review. J Environ
Pathol Toxicol 1980;3:537–67.
Zhang Z, Huang J, Yu G, Hong H. Occurrence of PAHs, PCBs and
organochlorine pesticides in the Tonghui River of Beijing,
China. Environ Pollut 2004;130:249–61.
Zhang HB, Luo YM, Wong MH, Zhao QG, Zhang GL. Distributions
and concentrations of PAHs in Hong Kong soils. Environ Pollut
2006;141:107–14.
Zuo Q, Duan YH, Yang Y, Wang XJ, Tao S. Source apportionment of
polycyclic aromatic hydrocarbons in surface soil in Tianjin,
China. Environ Pollut 2007;147:303–10.