Organochlorine Pesticides and Polychlorinated Biphenyl Congeners in Lanner Falco biarmicus feldeggii Schlegel Chicks and Lanner Prey in Sicily, Italy

lable at ScienceDirect

Environmental Pollution 157 (2009) 618–624

Contents lists avai

Environmental Pollution

journal homepage: www.elsevier .com/locate/envpol

Organochlorine pesticides and polychlorinated biphenyls in riverine runoffof the Pearl River Delta, China: Assessment of mass loading, input sourceand environmental fate

Yu-Feng Guan a,b, Ji-Zhong Wang a,b, Hong-Gang Ni a,b, Eddy Y. Zeng a,*

a State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, P.O. Box 1131, Wushan, Guangzhou 510640, Chinab Graduate School, Chinese Academy of Sciences, Beijing 100049, China

Mass loadings, input sources and environmental fate of organochlorin

a r t i c l e i n f o

Article history:Received 22 May 2008Received in revised form 13 August 2008Accepted 17 August 2008

Keywords:Organochlorine pesticidesPolychlorinated biphenylsRiverine runoffMass loadingFatePearl River Delta

* Corresponding author. Tel.: þ86 20 85291421; faxE-mail address: [email protected] (E.Y. Zeng).

0269-7491/$ – see front matter � 2008 Elsevier Ltd.doi:10.1016/j.envpol.2008.08.011

a b s t r a c t

A large-scale sampling program was conducted to simultaneously collect water samples at the eightmajor riverine runoff outlets of the Pearl River Delta (PRD), South China to assess the importance ofriverine runoff in transporting anthropogenic pollutants from terrestrial sources to the coastal ocean. Theconcentrations of

P21OCPs (sum of 21 OCP components) and

P20PCBs (sum of 20 PCB congeners) were

2.57–41.2 and 0.12–1.47 ng/L, respectively. Compositional distributions of DDTs suggested the possibilityof new input sources in the study area, but contributions from dicofol seemed considerably low. Theannual inputs of

P21OCPs and

P20PCBs were 3090 and 215 kg, with those of total HCHs and DDTs being

1110 and 1020 kg, respectively. A mass balance consideration indicated that riverine runoff is the majormode carrying OCPs from the PRD to the coastal ocean, and the majority of OCPs is further dissipated toopen seas.

� 2008 Elsevier Ltd. All rights reserved.

1. Introduction

Organochlorine pesticides (OCPs) and polychlorinated biphe-nyls (PCBs), classical examples of persistent organic pollutants(POPs), have been of worldwide concern owing to their persistence,bioaccumulative ability, and potential negative impacts on humansand animals. Organochlorine pesticides and PCBs are ubiquitouscontaminants and have been detected quite far from the pollutantsource, because of long-range transport stemming from atmo-spheric exchange, water currents and animal migration (Zhanget al., 2007). Because of the massive oceanic areas on the Earth,water bodies play an important role in the spatial distribution ofPOPs. Within coastal zones, riverine runoff is an important vector intransporting POPs from terrestrial sources to oceans, such as in thecases of polycyclic aromatic hydrocarbons and polybrominateddiphenyl ethers (Guan et al., 2007; Wang et al., 2007). Clearly,determining the amounts of POPs transported by riverine runoffcan shed lights into the cross-boundary transfer and help esti-mating the regional contributions of POPs to the global inventory.

: þ86 20 85290706.

All rights reserved.

As one of the most prosperous regions in China, the Pearl RiverDelta (PRD) has been undergoing rapid industrialization andurbanization, as well as enhanced agricultural development. Largeamounts of organic pollutants are also generated, and transportedto the coastal waters and eventually to the global oceans. Althoughthe application of technical dichlorodiphenyltrichloroethanes(DDTs) and hexachlorocyclohexanes (HCHs) in agriculture haspresumably been banned in China since 1983, new inputs of DDTsand HCHs have been identified and attributed to local usage oflindane, dicofol, and DDT-containing anti-fouling paints (Li et al.,2007; Qiu et al., 2005). The use of PCBs has been banned since theearly 1970s and the total amount of technical PCBs produced inChina accounted for less than 0.6% of the total global production(Mai et al., 2005). However, the atmospheric concentrations of PCBsin China were higher than those in several other Asian countries,such as Singapore, Japan, and Korea, and those in the PRD wereamong the highest in China (Jaward et al., 2005). In the last severaldecades, numerous efforts have been made to characterize thelevels and distribution of persistent chlorinated hydrocarbons inthe PRD and adjacent South China Sea (SCS) (Fu et al., 2003; Luoet al., 2004; Mai et al., 2005; Meng et al., 2007; Yang et al., 2004b,1997; Yu et al., 2008). However, no systematic study has beenconducted to determine the mass emissions of OCPs and PCBs fromthe PRD to the coastal ocean via riverine runoff.

Y.-F. Guan et al. / Environmental Pollution 157 (2009) 618–624 619

The main object of this study was to examine the occurrenceand mass loadings of OCPs and PCBs in riverine runoff of the PRD,from which the total inputs of OCPs and PCBs from the PRD to thecoastal ocean were estimated and compared with those fromvarious major rivers in China and other countries.

2. Methods and materials

2.1. Sample collection and extraction

The locality of the sampling sites is displayed in Figure S1 of the Supplementarymaterial (‘‘S’’ designates tables and figures in the Supplementary material there-after). The eight riverine runoff outlets for sampling are divided into four easternoutlets, Humen, Jiaomen, Honqilimen, and Hengmen, and four western outlets,Modaomen, Jitimen, Hutiaomen, and Yamen. The detailed procedures of samplecollection are presented elsewhere (Ni et al., 2008a). Briefly, sample collection wascarried out monthly, from March 2005 to February 2006, approximately 1 h beforethe intra-day lower tide. The volume of each sample was 40 L, composited fromsubsamples collected at various points along a river cross section, and stored into10-L brown glass bottles. Water samples were filtered using a vermicular system(pre-cleaned with acetone), and suspended particulate matter (SPM) was collectedwith 0.7 mm GF/F glass fiber filters (142 mm diameter, Whatman International,Maidstone, England) which were kilned at 450 �C for 4 h prior to use. SPM-loadedfilters were wrapped in pre-cleaned aluminum foil and immediately stored at�20 �C until analyzed. Filtrates were processed immediately upon filtration. Theprocedures of extraction and chromatographic separation were described in detailelsewhere (Luo et al., 2004; Meng et al., 2007) and are presented in the Supportinginformation.

2.2. Instrumental analysis

The concentrations of OCPs and PCBs were determined with a Hewlett–Packard5890 gas chromatograph (GC) coupled to a 5972 mass spectrometer (MS) anda Varian 3800 GC interfaced with a Saturn 2000 MS, respectively, in the selective ionmonitoring mode. Injection of 1 mL sample was conducted manually on the Hewlett–Packard GC/MS and automated with a Varian CP-8410 autosampler on Varian GC/MSsystem, both in the splitless mode. The temperature programs and quantitation ionsemployed were detailed elsewhere (Meng et al., 2007). A total of 21 OCP compoundswere quantified, i.e., a-HCH, b-HCH, g-HCH, d-HCH, heptachlor, heptachlor epoxide,aldrin, dieldrin, endrin, endrin ketone, endrin aldehydes, o0p-DDT, o0p-DDE, o0p-DDD, p0p-DDT, p0p-DDE, p0p-DDD, endosulfan I, endosulfan II, endosulfan sulfate, andmethoxychlor.

P21OCPs is defined as the sum of these 21 OCP components. For

PCBs, the chromatographic peaks were quantified by a secondary Aroclor mixture of1242:1248:1254:1260 (1:1:1:1) as PCB 5, 18, 31(28), 44, 52, 66, 87, 99, 101, 110, 118,123, 138, 141, 153, 170, 180, 183, and 187 according to the IUPAC nomenclature.P

20PCBs is defined as the sum of these 20 PCB congeners.

2.3. Quality assurance and quality control

Quality assurance and quality control procedures have been described in detailelsewhere (Guan et al., 2007; Luo et al., 2004). Procedure blanks (n¼ 3), laboratoryblanks (n¼ 4), and duplicate samples (three for every outlet) were processedalternately throughout the sampling analysis. In sample analysis for OCPs, only lowconcentrations of a-HCH (26 pg/L) and p0p-DDT (46 pg/L) were found in oneprocedural blank, which are below the respective reporting limit for OCPs (seebelow). However, PCB 18, 31, 44, 52, 66, 87, 123, and 153 were detected in procedureblanks and laboratory blanks. In particular the concentrations of PCB 18 (a mean of38 pg/L), 31 (a mean of 41 pg/L), 52 (a mean of 21 pg/L), and 44 (a mean of 45 pg/L)were higher than the lowest concentration of the calibration standards used insample analysis for PCBs. Therefore, the average concentrations in blank sampleswere subtracted from the concentrations of PCB 18, 31, 44, and 52, and the reportinglimit for these PCB congeners was set at 45 pg/L. For other PCB congeners, the lowestconcentration level (15 pg/L) of the calibration standards was defined as thereporting limit. Similarly, the reporting limit for OCPs was set at 50 pg/L. Thesurrogate recoveries of PCB 67(for OCPs), PCB 191 (for OCPs), 13C-PCB 141 (for PCBs),and PCB 209 (for PCBs) in all filtrate samples were 93�15%, 76�14%, 80�19%, and70�15%, respectively, and were 85�14%, 85�19%, 63�14%, and 71�18%,respectively, in all SPM samples. Reported concentrations were not surrogaterecovery corrected.

2.4. Data analysis

The monthly fluxes ofP

21OCPs andP

20PCBs from the eight runoff outlets fromMarch 2005 to February 2006 were calculated by

Fi;j ¼ Ci;j � Qi;j � 10�9 (1)

where Fi,j (kg/month) is the flux ofP

21OCPs orP

20PCBs from the ith outlet duringperiod j; Ci,j (ng/L) is the sum concentrations of

P21OCPs or

P20PCBs in both the

aqueous and SPM phases; and Qi,j (m3) is the total discharge amount from the ithoutlet during time period j. The total annual flux (Fi) from the ith outlet was thenestimated by

Fi ¼X12

j¼ 1

Fi;j (2)

For any concentration below the reporting limit, zero and half the reporting limitwas employed for the concentration and flux calculations, respectively. All OCPs andPCBs data were analyzed with Excel 2003 and SigmaPlot 9.0.

3. Results and discussion

3.1. Occurrence of OCPs and PCBs in riverine runoff

DDTs (sum of DDT, DDE, and DDD) and HCHs (sum of a-HCH, b-HCH, g-HCH, and d-HCH) are two groups of OCPs that have beenfrequently detected in waters of the PRD (Luo et al., 2004; Yu et al.,2008). In this study, the concentrations of DDTs and HCHs consti-tuted greater than 36% and 35% of total OCP concentrations in watersamples (Table S1). Detailed concentration data are presented inTable S1 and a summary is given in Table 1, and a comparison of thelevels of OCPs and PCBs in water bodies of China and other coun-tries is presented in Table S2.

The levels ofP

21OCPs andP

20PCBs were 2.57–41.2 ng/L witha mean of 11.0 ng/L and 0.12–1.47 ng/L with a mean of 0.77 ng/L,respectively, whereas those of DDTs and HCHs were 1.08–19.6 ng/L(mean: 3.89 ng/L) and 0.50–14.8 ng/L (mean: 3.69 ng/L), respec-tively. In general, the levels of DDTs and HCHs were at the high endof the global range, while the level of PCBs was at the low end of theglobal values (Table 2 and S2). It is notable that the concentrations ofDDTs and HCHs generally showed a seaward decreasing trend fromGuangzhou Channel of the Pearl River to the SCS (Fig. 1 and S2)(Luo et al., 2004; Yang et al., 2004a; Yu et al., 2008; Zhang et al.,2007), strongly suggesting that riverine runoff is the importantmode to transport these contaminants from terrestrial sources tothe coastal ocean.

The monthly concentrations ofP

21OCPs varied moderately atthe eight sampling locations (Fig. 2a). The highest mean concen-tration of

P21OCPs occurred in the water samples collected from

the Humen Outlet. This was probably due to the fact that watercourses draining into the Humen Outlet run through several largeurban areas including Guangzhou, Huizhou, and Dongguan. Theconcentrations of

P21OCPs were also fairly high in the water

samples from the Yamen Outlet, similar to the results for PBDEs(Guan et al., 2007), although no plausible explanation can be givenpresently. The distributions of DDTs and HCHs were somewhatsimilar to those of

P21OCPs, but the second highest mean

concentration of DDTs occurred in water samples from the Hon-gqilimen Outlet (Fig. 2a). This may be consistent with the fact thatthe water courses draining into the Hongqilimen Outlet runthrough several large-scale fruit farms. On the other hand, themean concentration of

P20PCBs did not vary much spatially

(Fig. 2a). Although PCBs and PBDEs possess similar properties anda portion of PCBs may have also been derived from e-waste, thespatial distribution of PCBs from this study was substantiallydifferent from that of PBDEs (Guan et al., 2007). This suggests thatthe majority of

P20PCBs in waters of the PRD may have stemmed

from historical usage of PCBs.Temporally, the concentrations of

P21OCPs varied considerably,

but no clear difference between dry (from October to February) andwet (from March to September) weather seasons was observed(Fig. 2b, p> 0.05). The temporal distributions of DDTs and HCHs,two major components of OCPs, were quite different, e.g. theconcentrations of HCHs were higher in April and June than in othermonths, whereas those of DDTs were higher in July and December.Besides any possible effects of precipitations, fresh inputs of DDTs

Table 1Average concentrations (ng/L) and total inputs (kg) of

P21OCPs,

P20PCBs, DDTs, and HCHs, as well as annual water discharge Qi, for the eight major outlets of the Pearl River

Delta, China

Average concentrationa Qib Total input

HCHs DDTsP

21OCPsP

20PCBs 109 m3 HCHs DDTsP

21OCPsP

20PCBs

Humen 4.60 5.53 13.3 0.93 45.5 252 251 704 31.8Jiaomen 3.24 2.85 10.3 1.26 49.8 182 135 500 36.3Hongqilimen 2.71 4.63 11.2 0.86 30.9 91.1 143 336 17.0Hengmen 4.34 3.54 10.9 1.40 39.4 154 142 417 45.0Modaomen 3.39 3.66 9.67 1.07 73.2 272 208 697 54.5Jitimen 3.69 4.11 11.3 1.23 9.9 35.7 42.5 115 7.26Hutiaomen 2.41 3.20 8.62 1.38 10.6 32.6 36.1 87.6 10.1Yamen 5.12 3.59 12.5 1.15 17.2 92.5 65.7 230 12.7Total 276 1110 1020 3090 215

a Sum of the concentrations in both the aqueous and SPM phases.b Qi ¼

P12j¼ 1 Qj;i .

Y.-F. Guan et al. / Environmental Pollution 157 (2009) 618–624620

and HCHs, which have been reported to occur in the PRD (An et al.,2005; Li et al., 2007), may also influence the distribution of DDTsand HCHs. In addition, no obvious temporal variation in the meanconcentrations of

P20PCBs was observed. Slightly higher concen-

trations ofP

20PCBs occurred during the period of August toOctober compared to other sampling times. This is also consistentwith the above assessment that historical usage may have been themain source of PCBs in the PRD.

3.2. Compositional profiles of OCPs and PCBs

The ratios of DDT/(DDDþDDE) and DDE/DDD have been used toindicate the degree of DDT decomposition and to identify any freshinput of DDTs (Lee et al., 2001). In this study, the ratios of DDT/(DDDþDDE) ranged from 0.16 to 27.9, with a mean of 3.76 (Table 2and S5). In addition, 95% of the DDT/(DDDþDDE) values washigher than unity (Table 2 and S5). These results suggest theoccurrence of fresh DDT inputs in the aquatic environment of thePRD. Furthermore, the mean ratios of DDE/DDD ranged from 0.05 to0.25 (Table 2), implying that DDT residues may have been mainlydechlorinated to DDD.

Qiu et al. (2005) suggested that dicofol, used on cotton and fruittrees and characterized with high o,p0-DDT/p,p0-DDT ratio, could bethe new source of DDT residues in the southern and eastern

DDTs

Guangzhou

Zhujiang

PRD OutletPRE

SCS

Guangzhou

Zhujiang

PRD OutletPRE

SCS

Co

ncen

tratio

n (n

g/L

)

0

2

4

28

32 HCHs

Fig. 1. Concentrations of DDTs and HCHs in waters of Baiertang of Guangzhou (Luoet al., 2004), Zhujiang (Pearl River artery) (Yang et al., 2004a), Human riverine runoffoutlet (this study), the Pearl River Estuary (Yu et al., 2008), and the South China Sea(Zhang et al., 2007). (The sampling locations are displayed in Figure S2 of theSupplementary material).

provinces of China. However, a recent study found that the ratios ofo,p0-DDT/p,p0-DDT were mostly lower than one in surface water ofthe northern SCS (Zhang et al., 2007), suggesting that other sourcesthan dicofol may have mainly contributed to the loadings of DDTsnear the coastal region off the PRD. The mean ratios of o,p0-DDT/p,p0-DDT obtained from this study ranged from 0 to 0.24 witha mean of 0.05 (Table 2), consistent with the results of Zhang et al.(2007). It was estimated that the amount of dicofol sold in threelarge urban centers (Guangzhou, Foshan, and Dongguan; Figure S1)in the PRD was approximately 35 metric tons in 2003 (An et al.,2005). Assuming an annual consumption rate of 35 tons/yr and therelative abundance of DDT residues in dicofol at 5% (An et al., 2005),the total amount of DDT generated via the use of dicofol in the PRDregion was 26 tons in the past 15 yr. On the other hand, the DDTinventory in soils of the three urban centers was approximately338 tons, of which 230 tons were attributed to p0p-DDT (Ma, 2007).It is apparent that the amount of DDTs generated via the use ofdicofol was considerably small relative to the quantity of histori-cally used technical DDTs. In addition, technical DDTs remain inuse in some parts of China for Malaria control and as additives inanti-fouling paints for fishing ships (Li et al., 2007).

The compositional data for HCHs are also presented in Table 2.The relative abundances of a-HCH and g-HCH were 36% and 13%,respectively, i.e. the ratio of a- to g-HCH was

P3, implying that

technical HCHs and lindane were two predominant products usedin this region (Yu et al., 2008). Because b-HCH is more persistentand less volatile than other HCH isomers, it tends to accumulatemore readily in the environment than other HCH isomers (Mid-deldorp et al., 1996; Quintero et al., 2005). It is therefore notsurprising that b-HCH accounted for almost 40% of total HCHs, thehighest among all HCH isomers (Table 2).

Tri- to penta-PCBs accounted for approximately 90% of totalPCBs and no octa- to deca-PCBs were detected. The concentrationsof various homologues decreased in the order of tri-PCBs> tetra-PCBs> penta-PCBs> hexa-PCBs> hepta-PCBs. In addition, therelative abundances of tri-PCBs at all outlets were more than half oftotal PCBs. The relative abundances of penta-PCBs at the foureastern outlets, Humen, Jiaomen, Honqilimen, and Hengmen, weregreater than those at the four western outlets, Modaomen, Jitimen,Hutiaomen, and Yamen, whereas a reverse trend was observed fortetra-PCBs (Table 2). These are probably reflective of the amountsand types of technical PCBs used in the PRD.

3.3. Partitioning of OCPs between the dissolved andparticulate phases

Total concentrations of HCHs and DDTs in the dissolved andparticulate phases are presented in Tables S3 and S4. These dataindicate that most HCHs were partitioned in the aqueous phase,

Table 2Compositional profiles of DDTs, HCHs, and PCBs in riverine runoff from eight major outlets of the Pearl River Delta, China

DDE/DDD DDT/(DDDþDDE) o0p-DDT/p0p-DDT a-HCH/HCHs b-HCH/HCHs g-HCH/HCHs tri-PCB tetra-PCB penta-PCB hexa-PCB hepta-PCB

HM 0.25 3.99 0.08 0.39 0.30 0.13 0.46 0.17 0.14 0.21 0.01JM 0.05 2.48 0.24 0.37 0.35 0.15 0.59 0.20 0.11 0.09 0.01HQ 0.06 5.95 0.00 0.31 0.42 0.13 0.55 0.22 0.13 0.07 0.03HE 0.13 3.65 0.01 0.32 0.49 0.10 0.60 0.21 0.10 0.07 0.02MD 0.09 3.63 0.00 0.41 0.37 0.11 0.58 0.22 0.10 0.08 0.01JT 0.17 4.22 0.00 0.35 0.44 0.11 0.61 0.24 0.06 0.08 0.01HT 0.18 4.25 0.00 0.40 0.35 0.15 0.58 0.26 0.07 0.06 0.02YM 0.19 1.95 0.08 0.31 0.45 0.15 0.56 0.23 0.14 0.06 0.01Mean 0.14 3.76 0.05 0.36 0.40 0.13 0.57 0.22 0.11 0.09 0.02

Y.-F. Guan et al. / Environmental Pollution 157 (2009) 618–624 621

accounting for more than 85% of total HCHs, whereas the majority(w67%) of DDTs was associated with particulates. The partitioningof OCPs between the dissolved and particulate phases can becharacterized by the in-situ organic carbon-normalized partitioncoefficient (Koc

0) (Kumata et al., 2002; Xue et al., 2006):

Koc0 ¼

�Cp=Cd

��foc ¼ Kp=foc (3)

where Cp and Cd are the analyte concentrations in the particulateand dissolved phases, respectively, foc is the mass fraction of organiccarbon in particulates, and Kp¼ Cp/Cd. Under the condition ofequilibrium partitioning, the partition coefficient Koc can be

March

April

May June Ju

ly

Augus

t

Septem

ber

Octobe

r

Novem

ber

Decem

ber

Janu

ary

Februa

ry

Co

ncen

tratio

n (n

g/L

)

0

5

10

15

20

a

Humen

Jiaomen

Hongqilimen

Hengmen

ModaomenJiti

men

HutiaomenYamen

Co

ncen

tratio

n (n

g/L

)

0

5

10

15

20∑21OCPs

HCHs

DDTs

∑20PCBs

∑21OCPsHCHs

DDTs

∑20PCBs

b

Fig. 2. Spatial (a) and temporal (b) concentration distributions ofP

21OCPs andP20PCBs, as well as DDTs and HCHs, in water samples from the eight major runoff

outlets of the Pearl River Delta, China.

predicted from octanol–water partition coefficient Kow by (Kumataet al., 2002; Seth et al., 1999):

log Koc ¼ 1:03 log Kow � 0:61 (4)

where Kow for the target analytes investigated in this study can beobtained from the literature (UNEP Chemicals). As shown in Table 3,the log Koc

0 values of DDTs are larger than those of HCHs, indicatingthat DDTs have greater affinity for particulates than HCHs. Inaddition, the observed log Koc

0 values were substantially larger thanlog Koc predicted by Eq. (4) for HCHs, whereas they were similar forDDTs. This indicates that partitioning of HCHs was not equilibriumbetween the dissolved and particulate phases, and thereforea portion of HCHs may tend to partition into the dissolved phaseduring transport to the coastal ocean.

3.4. Riverine inputs of OCPs and PCBs

The monthly inputs ofP

21OCPs andP

20PCBs from the eightriverine runoff outlets were 0.64–277 and 0.09–19.1 kg/month,respectively. The monthly inputs of DDTs and HCHs, two dominantcomponents of OCPs, were 0.20–71.4 and 0.17–102 kg/month,respectively (Table S1). Individually, five outlets (Humen, Jiaomen,Hongqilimen, Hengmen, and Modaomen) discharged considerablymore OCPs and PCBs than the other three outlets (Jitimen,Hutiaomen, and Yamen), which may be attributed to the enormousvolumes discharged through the five outlets as the averageconcentrations did not show the same variability as the inputs. Theannual inputs of

P21OCPs and

P20PCBs from the PRD to the coastal

ocean were estimated at 3090 and 215 kg/yr, respectively, andthose of DDTs and HCHs were 1020 and 1110 kg/yr, respectively(Table 1).

To place the riverine input levels of OCPs and PCBs in the PRD ina large perspective, Eq. (1) was employed to estimate the annualfluxes of the same contaminants from several rivers of China andaround the world. The results are summarized in Table 4. The

Table 3The values of distribution coefficient (log Kp) between the SPM and dissolvedphases, octanol–water partition coefficient (log Kow), equilibrium partition coeffi-cient (log Koc), the observed organic carbon-normalized in-situ particle/waterpartition coefficient (log Koc

0), and mean water concentrations (Cw, 103 ng/m3) ofa-HCH and p0p-DDT in 96 water samples from eight major runoff outlets of the PearlRiver Delta, China

Kpa log Kow

b log Kocc log Koc

0 Cw

HCHs 0.4 3.9–4.1 3.4–3.6 4.1–5.2DDTs 1.04 5.5–6.2 5.1–5.8 4.8–6.5a-HCH 0.42 3.9 3.4 4.6 0.75p0p-DDT 1.02 6.2 5.8 5.5 2.14

a Kp¼ Cp/Cd, mean for 96 samples.b log Kow values from literature (UNEP Chemicals).c log Koc was predicted as: log Koc¼ 1.03 log Kow� 0.61 (Kumata et al., 2002; Seth

et al., 1999).

Table 4Average concentrations (ng/L) and total inputs (t/yr) of DDTs, HCHs, and PCBs, as well as annual water discharge (Q), for several rivers of China and other countries

Average concentration Qa Total input References

DDTs HCHs PCBs 109 m3 DDTs HCHs PCBs

Pearl River 3.89b 3.69b 0.77b 334 1.30 1.23 0.26 This studyYangtze River 1.37 4.8 2.4 951 1.30 4.56 2.28 (Sun et al., 2002)Liaohe River 2.77 14.0 14.8 0.041 0.21 (Zhang and Dong, 2002)Huaihe Riverc 4.45–78.9 1.11–7.55 62.2 0.28–4.91 0.07–0.47 (Yu et al., 2004)Qiantang River 4.9 27.4 40.4 0.2 1.11 (Zhou et al., 2006)Jiulong River 12.8 71.8 11.7 0.15 0.84 (Zhang et al., 2001)Red River 54.2 23.8 119 6.45 2.83 (Hung and Thiemann, 2002)St. Lawrence River 1.02 0.28 0.28 (Pham et al., 1996)Ebro River 3.1 3.38 76.3 9.7 0.03 0.03 0.74 (Fernandez et al., 1999)S. Dvina Riverd 0.18e

Lena Riverd 3.0f 0.62e

Pechora Riverd 0.38f 1.79e

Pur Riverd 1.13f 5.78e

Taz Riverd 1.28f 5.49e

Yenisei Riverd 11.3e

Ob Riverd 8.07f 28.6e

a From the ‘‘China Statistical Yearbook 2006’’ and ‘‘Global Statistical Yearbook 2006’’.b Mean concentrations obtained from 96 water samples analyzed in this study (Table S1).c No mean values were reported, and the ranges were used for estimation.d Annual fluxes of OCPs in Russian northern rivers during 1990–1996 reported by Alexeeva et al. (2001).e Sum of a-HCH and g-HCH.f Sum of DDT and DDE.

Y.-F. Guan et al. / Environmental Pollution 157 (2009) 618–624622

annual riverine flux of DDTs from the PRD was similar to those fromYangtze River and Huaihe River, but was more than those fromLiaohe River, Qiantang River, and Jiulong River of China. However,the flux of HCHs from the PRD was lower than that from YangtzeRiver, similar to those from Qiantang River, and higher than thosefrom other rivers of China. Globally, the riverine inputs of DDTs andHCHs from the PRD were far lower than those from several Russianrivers (Ob River, Yenisei River, Pur River, and Taz River; Table 4) andRed River of Vietnam, but higher than those from St. Lawrence Riverof Canada, S. Dvina River of Russia, and Ebro River of Spain. The fluxof PCBs from the PRD was much lower than that from Yangtze River,

Water flow (109 m

3)

In

pu

t o

f H

CH

s (kg

)

c

r2 = 0.88

r2 = 0.82

0.1 10

In

pu

t o

f to

tal O

CP

s (kg

)

100

101

102

100

101

102

a

1

0.1 101

Fig. 3. Relationship between the riverine inputs of (a) organochlorine pesticides (OCPs)dichlorodiphenyltrichloroethanes (DDTs) and water flows (discharge amounts) from the ei

but comparable to that from Ebro River. These results suggest thatriverine runoff has been one of the important sources of OCPs andPCBs transported from the PRD to the coastal ocean. Because of theextensive use of OCPs-containing products in the past, largequantities of OCPs have remained in soils of the PRD. For example,a recent study estimated that the inventory of OCPs in soils of thethree urban centers of the PRD, Guangzhou, Foshan, and Dongguan,was approximately 360 metric tons (Ma, 2007). These contami-nants may gradually be leached from soils with occurrence ofrainfalls, and carried to the aquatic systems and eventually to thecoastal ocean off the PRD via surface runoff.

Water flow (109 m

3)

d

In

pu

t o

f D

DT

s (kg

)In

pu

t o

f P

CB

s (kg

)

r2 = 0.80

r2 = 0.81

10-1

100

101

100

101

102

b

0.1 101

0.1 101

; (b) polychlorinated biphenyls (PCBs); (c) hexachlorocyclohexanes (HCHs); and (d)ght runoff outlets of the Pearl River Delta, China.

Fig. 4. Mass balance diagram for input and output pathways ofP

21OCPs in the PearlRiver Estuary.

Y.-F. Guan et al. / Environmental Pollution 157 (2009) 618–624 623

The correlations between the riverine inputs of OCPs, PCBs,HCHs, and DDTs and water discharge amounts from the eightriverine outlets of the PRD were considerably linear, as indicated bythe near-unity values of correlation coefficients (r2), i.e. 0.88 forOCPs, 0.81 for PCBs, 0.82 for HCHs, and 0.80 for DDTs (Fig. 3).Apparently, the riverine fluxes of persistent chlorinated hydrocar-bons can be estimated from discharge amount via Flux¼ S(discharge amount), with S designated as a proportional constantthat can be obtained from the slope of a specific regression line.This conclusion is consistent with those obtained with poly-brominated diphenyl ethers (Guan et al., 2007), polycyclic aromatichydrocarbons (Wang et al., 2007), and total organic carbon(Figure S3). However, the correlations were relatively weak forlinear alkylbenzenes and benzothiazoles (Ni et al., 2008b,c)(Figure S3). These results suggest that the levels of organic pollut-ants in the PRD watershed may have reached a quasi-steady stateand precipitation has become a controlling factor in transport ofthese contaminants to the coastal ocean.

3.5. Water–air exchange and fate of OCPs

Water–air exchange is an important mode dictating the fate ofOCPs in a coastal environment. Such an exchange can be assessedusing a fugacity-based model (Falconer et al., 1995):

fw=fa ¼ HCw=CaRT (6)

where fw and fa are the fugacities in water and air, respectively; H isHenry’s Law constant (Pa m3/mol); Cw and Ca are the analyteconcentrations in water (ng/m3) and air (ng/m3), respectively; R isthe ideal gas constant (8.314 Pa m3/mol K); and T is the absolutetemperature. Because a-HCH and p0p-DDT have been detected withabundance in various environmental compartments of the PRD(Guo et al., 2008; Li et al., 2007; Luo et al., 2004), they are chosenherein to represent HCHs and DDTs and demonstrate theassessment.

The H values of a-HCH and p0p-DDT at 298 K are 0.55 and1.1 Pa m3/mol (Sahsuvar et al., 2003; Shen and Wania, 2005). Themean air concentrations of a-HCH (0.125 and 0.049 ng/m3) and p0p-DDT (0.638 and 0.636 ng/m3) measured in Guangzhou and HongKong (Li et al., 2007), respectively, are used as Ca values. The meanwater concentrations of a-HCH and p,p0-DDT of the PRD are used asCw (Table 3). Therefore, fw/fa values estimated of a-HCH and p0p-DDT are 1.33 and 1.38 for Guangzhou and 3.40 and 1.47 for Hong

Kong, respectively, all greater than 1. These results indicate netvolatilization of OCPs from water to air, and probably establish thatriverine inputs are the dominant mode transporting OCPs from thePRD to the coastal ocean.

To estimate the net outflow of OCPs from the PRD (via the PearlRiver Estuary) to the global ocean, a mass balance concept(Wang et al., 2007) can be applied:

Fr þ Fa ¼ Fs þ Fo þ Fd (7)

where Fr, Fa, Fs, Fo, and Fd are the annual inputs by riverinerunoff, sum of the fluxes of water–air exchange, dry deposition, andwet deposition, sedimentation, outflow to ocean, and degradation,respectively, in kg/yr. The annual input of OCPs from the PRD was3090 kg/yr (Table 1). Based on the estimated flux of water–airexchange (�80.6 ng/m2 day) in the PRD (Supplementary Material),the flux of air-water exchange was �59.2 kg/yr assuming that thearea of the PRE is 2.01�103 km2. It also suggests that the netvolatilization rate in the PRE was 59.2 kg/yr. On the other hand, thedry and wet deposition was approximately 3.85 ng/m2 day in thecoast of Hong Kong (Li, 2005), resulting in a flux of dry and wetdeposition atw2.82 kg/yr. Therefore, Fa was �56.3 kg/yr. Thesedimentary rate of OCPs was estimated at 171 kg/yr (Table S6) inthe PRE (Chen et al., 2006). Degradation rate of OCPs in water was463 kg/yr in this region (Table S6). Hence, Fo of OCPs was estimatedat 2400 kg/yr using Eq. (7) (Fig. 4). Apparently, the majority of OCPsdischarged via riverine runoff from the PRD is further carried intothe coastal ocean.

Acknowledgements

This research was financially supported by the National NaturalScience Foundation of China (40588001, 40532013 and U0633005)and the K.C. Wong Education Foundation, Hong Kong SpecialAdministrative Region, China. The authors thank Mr. T. S. Xiang forassistance in the GC/MS analysis. We are also grateful to thesampling team consisted of mostly graduate students from theGuangzhou Institute of Geochemistry and Sun Yat-sen Universityfor assistance in field work. This is contribution NO. IS-1004 fromGIGCAS.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.envpol.2008.08.011.

References

Alexeeva, L.B., Strachan, W.M.J., Shlychkova, V.V., Nazarova, A.A., Nikanorov, A.M.,Korotova, L.G., Koreneva, V.I., 2001. Organochlorine pesticides and trace metalmonitoring of Russian rivers flowing to the Arctic Ocean: 1990–1996. MarinePollution Bulletin 43, 71–85.

An, T.C., Chen, J.X., Fu, J.M., Sheng, G.Y., Li, G.Y., Hu, Z.Y., Kuang, Y.Q., 2005. Thepollution situation and control strategy of persistent organic pollutants in thePearl River Delta, China. Ecology and Environment 14, 981–986 (in Chinese).

Chen, S.J., Luo, X.J., Mai, B.X., Sheng, G.Y., Fu, J.M., Zeng, E.Y., 2006. Distribution andmass inventories of polycyclic aromatic hydrocarbons and organochlorinepesticides in sediments of the Pearl River Estuary and the northern South ChinaSea. Environmental Science and Technology 40, 709–714.

Falconer, R.L., Bidleman, T.F., Gregor, D.J., 1995. Air–water gas exchange andevidence for metabolism of hexachlorocyclohexanes in Resolute Bay, N.W.T.Science of the Total Environment 160�161, 65–74.

Fernandez, M.A., Alonso, C., Gonzalez, M.J., Hernandez, L.M., 1999. Occurrence oforganochlorine insecticides, PCBs and PCB congeners in waters and sedimentsof the Ebro River (Spain). Chemosphere 38, 33–43.

Fu, J.M., Mai, B.X., Sheng, G.Y., Zhang, G., Wang, X.M., Peng, P.A., Xiao, X.M., Ran, R.,Cheng, F.Z., Peng, X.Z., Wang, Z.S., Tang, U.W., 2003. Persistent organic pollut-ants in environment of the Pearl River Delta, China: an overview. Chemosphere52, 1411–1422.

Guan, Y.F., Wang, J.Z., Ni, H.G., Luo, X.J., Mai, B.X., Zeng, E.Y., 2007. Riverine inputs ofpolybrominated diphenyl ethers from the Pearl River Delta (China) to thecoastal ocean. Environmental Science and Technology 41, 6007–6013.

Y.-F. Guan et al. / Environmental Pollution 157 (2009) 618–624624

Guo, Y., Meng, X.Z., Tang, H.L., Zeng, E.Y., 2008. Tissue distribution of organochlorinepesticides in fish collected from the Pearl River Delta, China: implicationsfor fishery input source and bioaccumulation. Environmental Pollution 155,150–156.

Hung, D.Q., Thiemann, W., 2002. Contamination by selected chlorinated pesticidesin surface waters in Hanoi, Vietnam. Chemosphere 47, 357–367.

Jaward, F.M., Zhang, G., Nam, J.J., Sweetman, A.J., Obbard, J.P., Kobaba, Y., Jones, K.C.,2005. Passive air sampling of polychlorinated biphenyls, organochlorinecompounds, and polybrominated diphenyl ethers across Asia. EnvironmentalScience and Technology 39, 8638–8645.

Kumata, H., Yamada, J., Masuda, K., Takada, H., Sato, Y., Sakurai, T., Fujiwara, K., 2002.Benzothiazolamines as tire-derived molecular markers: sorptive behavior instreet runoff and application to source apportioning. Environmental Scienceand Technology 36, 702–708.

Lee, K.T., Tanabe, S., Koh, C.H., 2001. Distribution of organochlorine pesticides insediments from Kyeonggi Bay and nearby areas, Korea. Environmental Pollution114, 207–213.

Li, J., 2005. Regional Geochemistry of Organochlorine Pesticides in the Pearl RiverDelta. Ph.D. thesis. Beijing: Graduate School of the Chinese Academy ofSciences, 105 p. (in Chinese).

Li, J., Zhang, G., Guo, L.L., Xu, W.H., Li, X.D., Lee, C.S.L., Ding, A.J., Wang, T., 2007.Organochlorine pesticides in the atmosphere of Guangzhou and Hong Kong:regional sources and long-range atmospheric transport. Atmospheric Environ-ment 41, 3889–3903.

Luo, X.J., Mai, B.X., Yang, Q.S., Fu, J.M., Sheng, G.Y., Wang, Z.S., 2004. Polycyclicaromatic hydrocarbons (PAHs) and organochlorine pesticides in water columnsfrom the Pearl River and the Macao harbor in the Pearl River Delta in SouthChina. Marine Pollution Bulletin 48, 1102–1115.

Ma, X.X. 2007. The distribution and environmental behaviors of representativeorganic pollutant in soils and suspended particles in the Pearl River Delta. Ph.D.thesis. Beijing: Graduate School of the Chinese Academy of Sciences. 150 p.(in Chinese).

Mai, B.X., Zeng, E.Y., Luo, X.J., Yang, Q.S., Zhang, G., Li, X.D., Sheng, G.Y., Fu, J.M., 2005.Abundances, depositional fluxes, and homologue patterns of polychlorinatedbiphenyls in dated sediment cores from the Pearl River Delta, China. Environ-mental Science and Technology 39, 49–56.

Meng, X.Z., Zeng, E.Y., Yu, L.P., Mai, B.X., Luo, X.J., Ran, Y., 2007. Persistent haloge-nated hydrocarbons in consumer fish of China: regional and global implicationsfor human exposure. Environmental Science and Technology 41, 1821–1827.

Middeldorp, P.J.M., Jaspers, M., Zehnder, A.J.B., Schraa, G., 1996. Biotransformation ofa-, b-, g-, and d-hexachlorocyclohexane under methanogenic conditions.Environmental Science and Technology 30, 2345–2349.

Ni, H.G., Lu, F.H., Luo, X.L., Tian, H.Y., Wang, J.Z., Guan, Y.F., Chen, S.J., Luo, X.J.,Zeng, E.Y., 2008a. Assessment of sampling designs to measure riverine fluxesfrom the Pearl River Delta, China to the South China Sea. EnvironmentalMonitoring and Assessment 143, 291–301.

Ni, H.G., Lu, F.H., Luo, X.L., Tian, H.Y., Zeng, E.Y., 2008b. Occurrence, phase distri-bution, and mass loadings of benzothiazoles in riverine runoff of the Pearl RiverDelta, China. Environmental Science and Technology 42, 1892–1897.

Ni, H.G., Lu, F.H., Wang, J.Z., Guan, Y.F., Luo, X.L., Zeng, E.Y., 2008c. Linear alkyl-benzenes in riverine runoff of the Pearl River Delta (China) and their applicationas anthropogenic molecular markers in coastal environments. EnvironmentalPollution 154, 348–355.

Pham, T., Lum, K., Lemieux, C., 1996. Seasonal variation of DDT and its metabolites inthe St. Lawrence River (Canada) and four of its tributaries. The Science of theTotal Environment 179, 17–26.

Qiu, X., Zhu, T., Yao, B., Hu, J., Hu, S., 2005. Contribution of dicofol to the current DDTpollution in China. Environmental Science and Technology 39, 4385–4390.

Quintero, J.C., Moreira, M.T., Feijoo, G., Lema, J.M., 2005. Anaerobic degradation ofhexachlorocyclohexane isomers in liquid and soil slurry systems. Chemosphere61, 528–536.

Sahsuvar, L., Helm, P.A., Jantunen, L.M., Bidleman, T.F., 2003. Henry’s law constantsfor a-, b-, and g-hexachlorocyclohexanes (HCHs) as a function of temperatureand revised estimates of gas exchange in Arctic regions. AtmosphericEnvironment 37, 983–992.

Seth, R., Mackay, D., Muncke, J., 1999. Estimating the organic carbon partitioncoefficient and its variability for hydrophobic chemicals. Environmental Scienceand Technology 33, 2390–2394.

Shen, L., Wania, F., 2005. Compilation, evaluation, and selection of physical–chemical property data for organochlorine pesticides. Journal of Chemical andEngineering Data 50, 742–768.

Sun, C., Dong, Y., Xu, S., Yao, S., Dai, J., Han, S., Wang, L., 2002. Trace analysis ofdissolved polychlorinated organic compounds in the water of the Yangtse River(Nanjing, China). Environmental Pollution 117, 9–14.

UNEP Chemicals. Regional reports of the regionally based assessment of persistenttoxic substances programme. http://www.chem.unep.ch/pts/regreports/ (accessedAugust 2008).

Wang, J.-Z., Guan, Y.-F., Ni, H.-G., Luo, X.-L., Zeng, E.Y., 2007. Polycyclic aromatichydrocarbons in riverine runoff of the Pearl River Delta (China): concen-trations, fluxes and fate. Environmental Science and Technology 41,5614–5619.

Xue, N.D., Zhang, D.R., Xu, X.B., 2006. Organochlorinated pesticide multiresiduesin surface sediments from Beijing Guanting reservior. Water Research 40,183–194.

Yang, Q.S., Mai, B.X., Fu, J.M., Sheng, G.Y., Wang, J.X., 2004a. Spatial and temporaldistribution of organochlorine pesticites (OCPs) in surface water from the PearlRiver Artery Estuary. Environmental Science 25, 150–156 (in Chinese).

Yang, Q.S., Mai, B.X., Luo, X.J., Fu, J.M., Sheng, G.Y., Lin, Z., 2004b. Vertical distributionof organochlorine pesticites (OCPs) in Macao coastal waters. Acta ScientiaeCircumstantiae 24, 428–434 (in Chinese).

Yang, Y.H., Sheng, G.Y., Fu, J.M., Min, Y.S., 1997. Organochlorinated compoundsin waters of the Pearl River Delta region. Environmental Monitoring andAssessment 44, 569–575.

Yu, M., Luo, X.J., Chen, S.J., Mai, B.X., Zeng, E.Y., 2008. Organochlorine pesticides insurface water and sediments of the Pearl River Estuary, South China. Environ-mental Toxicology and Chemistry 27, 10–17.

Yu, Y.J., Huang, H., Wang, B., Wang, X.D., Wang, L.S., 2004. Level of organochlorinepesticide pollution in the water of the Huaihe River, Jiangsu Province. Envi-ronmental Chemistry 23, 568–572 (in Chinese).

Zhang, G., Li, J., Cheng, H.R., Li, X.D., Xu, W.H., Jones, K.C., 2007. Distribution oforganochlorine pesticides in the northern South China Sea: implications forland outflow and air–sea exchange. Environmental Science and Technology 41,3884–3890.

Zhang, X.F., Dong, X.L., 2002. Organic chlorinated pesticides in middle and lowerreaches of Liaohe. Journal of Dalian Institute of Light Industry 21, 102–103 (inChinese).

Zhang, Z.L., Chen, W.Q., Khalid, M., Zhou, J., Xu, L., Hong, H.S., 2001. Evaluation andfate of the organic chlorine pesticides at the waters in Jiulong River estuary.Environmental Science 22, 88–92 (in Chinese).

Zhou, R., Zhu, L., Yang, K., Chen, Y., 2006. Distribution of organochlorine pesticidesin surface water and sediments from Qiantang River, East China. Journal ofHazardous Materials 137, 68–75.