lable at ScienceDirect

Environmental Pollution 219 (2016) 883e896

Contents lists avai

Environmental Pollution

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

Current status and historical variations of DDT-related contaminantsin the sediments of Lake Chaohu in China and their influencingfactors*

Lei Kang, Qi-Shuang He, Wei He*, Xiang-Zhen Kong, Wen-Xiu Liu, Wen-Jing Wu,Yi-Long Li, Xin-Yu Lan, Fu-Liu Xu**

MOE Laboratory for Earth Surface Processes, College of Urban & Environmental Sciences, Peking University, Beijing 100871, China

a r t i c l e i n f o

Article history:Received 1 April 2016Received in revised form18 August 2016Accepted 26 August 2016Available online 6 September 2016

Keywords:DDXsSedimentTemporal-spatial distributionSource apportionmentInfluencing factorsAlgal bloomLake Chaohu

* This paper has been recommended for acceptanc* Corresponding author.** Corresponding author.

E-mail addresses: harveymylife@aliyun.com (W.(F.-L. Xu).

http://dx.doi.org/10.1016/j.envpol.2016.08.0720269-7491/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

The temporal-spatial distributions of DDT-related contaminants (DDXs), including DDT (dichlor-odiphenyltrichloroethane), DDE (dichlorodiphenyldichloroethylene) and DDD (dichlorodiphenyldi-chloroethane), in the sediments of Lake Chaohu and their influencing factors were studied. p,p-DDE andp,p-DDD were found to be the two dominant components of DDXs in both surface and core sediments.The parent DDT compounds were still detectable in sediment cores after the late 1930s. Historical usageof technical DDT was identified as the primary source of DDXs in sediments, as indicated by DDT/(DDD þ DDE) ratios of less than one. The residual levels of DDXs were higher in the surface and coresediments in the western lake area than in other lake areas, which might be due to the combined infloweffects of municipal sewage, industrial wastewater and agricultural runoff. The DDX residues in thesediment cores reached peak values in the late 1970s or early 1980s. There were significant positiverelationships between DDX residues in sediment cores with annual DDT production and with fine par-ticulate sizes (<4.5 mm). The relationship between the DDXs and TOC in sediment was complex, asindicated by the significant differences among the surface and core sediments. The algae-derived organicmatter significantly influenced the amount of residue, composition and distribution of DDXs in thesediments. The DDD/DDE ratios responded well to the anaerobic conditions in the sediments that werecaused by algal blooms after the late 1970s in the western lake area. This suggests that the algae-derivedorganic matter was an important factor and served as a biomarker of eutrophication and also affected theDDX residues and lifecycle in the lake ecosystem.

© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

Dichlorodiphenyltrichloroethane (DDT) was first synthesized in1874; it was first used as pesticide in 1937 (Apeti et al., 2010) and forepidemic prevention in the US Army during the Second World War(Kannan et al., 1995). Since the mid-1940s, the production of DDTrapidly increased until the 1960s, when DDT and its metabolites(including DDE (dichlorodiphenyldichloroethylene) and DDD(dichlorodiphenyldichloroethane)) were found to resist

e by Eddy Y. Zeng.

He), xufl@urban.pku.edu.cn

degradation in the environment and to have negative impacts onnon-target organisms (Saxena and Siddiqui, 1980). During the1970e1980s, many countries began to regulate the productions anduses of DDT. DDXs, including o, p0-DDT, p, p0-DDT, o, p0-DDD, p, p0-DDD, o, p0-DDE and p, p0-DDE, have been proven to be seriouslyharmful to ecosystems and to human health (Snedeker, 2001;Wetterauer et al., 2012; Mrema et al., 2013). In 2001, DDT was lis-ted in the Stockholm Convention on Persistent Organic Pollutantsas one of the 12 initial POPs (UNEP, 2001). Meanwhile, some pro-cedures have been undertaken to prohibit the productions and useof DDXs (SCPOPs, 2005). However, although their productions anduse have been globally regulated for years, there are still detectablelevels of DDXs in various media worldwide (e.g., Satcher, 2015; Liet al., 2015; Najam and Alam, 2015; Bouwman et al., 2015; Aamiret al., 2016; Gerber et al., 2016; Booij et al., 2016).

China started to produce technical DDT in the 1950s and became

L. Kang et al. / Environmental Pollution 219 (2016) 883e896884

the second largest DDT manufacturer in the world by the time itwas banned in 1983 (Wang et al., 2013). From the 1950s to 1982,over 0.45 million tons of DDT was produced, accounting for 20% ofthe total world production during that time (Wang et al., 2013). InChina, DDT was not officially banned for agricultural usage until1983, and small scale DDT applications are still allowed for publichealth purposes, as an additive of marine anti-fouling paint, and asthe primary raw material of dicofol, which is a pesticide for con-trolling a variety of mites and annihilating mite eggs in agriculture(He et al., 2015). It is estimated that 10,000 tons of anti-foulingpaints are used in fishing ships in China each year, and approxi-mately 250 tons of DDT are used in anti-fouling paints (He et al.,2012). Although the production and use of DDT have been regu-lated for more than 30 years in China, DDXs are still detectable indifferent environmental media, such as soil, water, dustfall, sedi-ment and ambient air (Li et al., 2006; Zhou et al., 2008; Wu et al.,2005; Liu et al., 2012; Yang et al., 2008). Among the various envi-ronment media, sediments are the primary sinks of DDXs, sincethey are easily adsorbed onto particulate matter due to their highKOW (i.e., octanolewater partition coefficient) or KOC (carbonnormalized partitioning coefficient) and transported into thewater-sediment system in surface runoff and dry-wet deposition(Cao et al., 2007; He et al., 2012, 2016). Lake sediment cores arehigh-resolution information databases that record the history ofsuch contaminants as DDXs (Zhang et al., 2010). However, infor-mation regarding DDXs in sediment cores from shallow lakes is stilllimited (Barakat et al., 2013; Li et al., 2015) compared with studiesof the historical records of DDXs in sediment cores frommarine andestuarine environments (Qiu et al., 2009).

Lake Chaohu, with an average depth of 3 m and a water area ofapproximately 760 km2, is located in the center of Anhui Provincein southeastern China (Fig. 1). As the fifth largest freshwater lake inChina, it was well known for its scenic beauty and rich aquaticbiodiversity before the 1950s (Xu, 1997). However, the lake hassince suffered from serious eutrophication, with frequent algalblooms in spring and summer, and it has been one of the mosteutrophic lakes in China since the early 1980s due to increasingpressures from rapid population growth and economic

Fig. 1. Geographic locations of (a) Lake Chaohu, (b) Lake Chaohu in

development in the drainage area (Xu et al., 1999; Jiang et al., 2014).Our previous studies from recent years have shown that there arestill high DDX residues in the water, sediment, suspended partic-ulate matter, ambient air, dust fall, and aquatic biota (He et al.,2012; Liu et al., 2012, 2013; Ouyang et al., 2012, 2013, 2014; Heet al., 2014, 2015; Liu et al., 2016). This could be attributed to theextensive historical DDT usage in the Lake Chaohu basin, which isone of the primary farming areas of China, before it was banned in1983 (Ouyang et al., 2013). Furthermore, new inputs of technicalDDT in Lake Chaohu have been found from the source apportion-ment of DDXs in the water, dust fall and suspended particulatematter, which may be a result of the recent usage of dicofol andmarine paints (Ouyang et al., 2013, 2014; He et al., 2015). Recently, astudy on the historical deposition behaviors of organochlorinepesticides (OCPs) in Lake Chaohu sediments was conducted (Liet al., 2015). It was determined that DDT contamination derivedfrom the historical use of technical DDTs and that the DDX con-centrations were highly significantly related to the total organiccarbon (TOC) content, sediment grain sizes, nutrient content andheavy metal content (Li et al., 2015). However, the effects of algae-derived organic matter in the sediments on the residual levels andcomposition of DDXs remains unknown. Previous studies havebeen undertaken to determine the relationship between algae-derived organic matter and typical organic pollutants in differentmedia. For example, algae-derived organic matter was found tohave a significant influence on low-molecular polycyclic aromatichydrocarbons (PAHs) (Wu et al., 2012). As for DDXs, their watersolubility could be enhanced by the existence of algae-derivedorganic matter (Ma et al., 2012). However, limited information isavailable about the relationships between sedimentary records ofDDXs and algae-derived organic matter in the sediments of shalloweutrophic lakes. This is very important information for under-standing the historical accumulation of DDXs in the sediments ofLake Chaohu and their association with intensive algal blooms.

The objectives of this study were as follows: (1) to investigatethe residual levels and distributions of DDXs in the surface and coresediments in Lake Chaohu; (2) to identify the composition andpotential sources of DDXs in the surface and core sediments; and

East China and (c) the sediment sampling sites in Lake Chaohu.

L. Kang et al. / Environmental Pollution 219 (2016) 883e896 885

(3) to analyze the effects of certain factors, including DDT pro-duction, granularity, TOC and algae-derived organic matter, on theresidues and distribution of DDXs in the surface and coresediments.

2. Materials and methods

2.1. Sample collection

Surface sediment samples and sediment cores were collected inMay of 2011. The geographic locations of the sampling sites areshown in Fig. 1. Sample sites L1~L5 were located in the western lakearea (WEL), while L6~L10 were sited in the eastern lake area (EAL).Sample sites R1, R3, R4 and R5 were collected from the estuaries ofinflow rivers in the western lake area, and sample sites R6~R9 wereobtained from 100 m upstream in the estuaries of the easterninflow rivers (R10 was an outflow river sampling site, which willnot be discussed in the present study). Surface sediment sampleswere collected by a grab bucket sampler, which obtained sedimentmixtures from a depth of 0~10 cm. Sediment cores were collectedfrom the center of the western lake (L4) and the center of theeastern lake (L8) by a tube sampler at a depth of 0~30 cm. Thesediment cores were then sliced into layers with a thickness of 1 cmeach to distinguish different decades. All of the sediment sampleswere freeze-dried, ground and sieved through a 70-mesh sievebefore the extraction procedure.

2.2. Reagents and materials

Analytical grade n-hexane, acetone and dichloromethane (DCM)(HPLC grade) used for extraction and elution were purchased fromTedia Co. Inc., Fairfield, Ohio, USA. A dilution of a commercialmixture of OCPs standard (Accustandard Inc., New Haven, Con-necticut, USA) by n-hexane was applied for the preparation of thestocked OCPs standard mixture. The internal standard (IS) samplewas also purchased fromAccustandard Inc. All of the glassware wascleaned by an ultrasonic cleaner (KQ-500B, Kunshan UltrasonicInstrument, Kunshan, China) and heated at 400 �C for 6 h.

2.3. Sample extraction and cleanup

After sieving, the sediment samples were extracted with DCMfor 48 h at a rate of 4e6 cycles per hour with a Soxhlet extractorwith the addition of activated copper granules to remove sulfur. Theremaining extracts were then concentrated to approximately 1 mlwith a rotary evaporator. The concentrated extracts were loadedinto a chromatography column filled with neutral silica gel (12 cm),neutral alumina (6 cm) and anhydrous sodium sulfate (1 cm) frombottom to top. The column was first extracted with 20 ml of n-hexane. While the solution collected from the first extract step wasdiscarded, the DDXs were extracted using 70 ml of mixed solvent(DCM and n-hexane, V:V ¼ 3:7). The elution was concentrated to1 ml by a rotary evaporator. Before GC-MS analysis, the sampleswere added with IS and transferred to glass vials.

2.4. Sample analysis and quality insurance

The DDXs in the concentrated extract were analyzed using anAgilent 6890 gas chromatograph equipped with a HP-5 MS column(30 m, 0.250 mm inner diameter, 0.25 mm film thickness) and anAgilent 5973mass spectrometer. Heliumwas used as the carrier gasat a speed of 1 ml/min. The temperature of the injection port wasset at 220 �C. After the injection of the samples, the initial tem-peraturewas held at 50 �C for 2 min, then reached 150 �C at a speedof 10 �C/min. Next, the oven temperaturewas increased by 3 �C/min

to 240 �C and held for 5 min, finally reaching 300 �C at 10 �C/minand was maintained for 5 min. The temperatures of the ion source,quadrupole and the transmission line were 200 �C, 150 �C and250 �C. The quantitation of DDXs was performed using an internalcalibration method with the selected ion-monitoring mode.

Two or three parallel samples were determined for the surfacesediments and sediment cores. Method blanks and procedureblanks were extracted and analyzed using the samemethod as withthe samples. Method recoveries and instrument detection limitsare listed in Table S1, where the recoveries ranged from 76% to132%.

2.5. Other analyses

2.5.1. Isotope dating of sediment coresThe sediment core dating was performed with 210Pb and 137Cs

dating. Compared with 14C dating, 210Pb and 137Cs dating can datemodern events more accurately (Cundy and Stewart, 2004). Mea-surement of the 210Pb and 137Cs radionuclide activity for eachsample of sediment cores was conducted with an Ortec HPGe GWLseries of well-type, coaxial, low background, intrinsic germaniumdetectors (GCW3022 HeP Ge) at the State Key Laboratory of LakeScience and Environment of the Chinese Academy of Science (CAS),in Nanjing, China. The chronologies of the sediment cores weredetermined from the 210Pb and 137Cs radionuclide activity (Kanget al., 2016).

2.5.2. Particle size and total organic carbonThe particle sizes of sediment were analyzed using a FRITSCH

A22 laser particle size analyzer (Fritsch GmbH Milling and SizingIndustries, Idar-Oberstein, Germany). The sediment samples weredivided into two parts to determine the total carbon (TC) andinorganic carbon (IC). A total organic carbon (TOC) analyzer (TOC-5000A; Shimadzu Corp., Japan) with a solid sampler (SSM-5000A)was used to obtain the TOC levels. Analytical grade glucose andsodium carbonate were used to calibrate the TOC for measurementof the TC and IC, respectively.

2.5.3. Lipids and n-alkanesLipids and n-alkanes were extracted and measured according to

a previous study (Lin et al., 2014). 5 g of sediment samples weretreated with an ultrasonic treatment in 200 ml of mixed solvent(DCM and methanol, V:V ¼ 1:1) at 40 �C for 20 min, 3 times. Theextract was then concentrated, dried with nitrogen, and loaded intoa chromatography column filled with 100-mesh activated silica gel.After that, 3 eluents were used sequentially to elute differentcomponents: mixed solvent I (n-hexane and DCM, V:V ¼ 9:1) foralkanes (F1), mixed solvent II (DCM and methanol, V:V ¼ 2:1) forfree lipids (F3), and methanol for alcohols and aromatic hydrocar-bons (F2). The residues from the ultrasonic treatment weresaponified with KOH/methanol (1 mol/L) for 12 h. The neutralcomponents were extracted with a mixed solvent (n-hexane andDCM, V:V ¼ 9:1). Afterwards, the pH of the remaining solution wasadjusted to 1 by 30% HCl, then extracted with n-hexane and DCM(V:V ¼ 9:1) and the bound lipids were obtained (F4). F1, F3 and F4were concentrated and dried with nitrogen. F3 and F4 were methylesterified by BF3/ether and methanol and extracted with n-hexane.Therefore, modified F3, F4 and F1 were concentrated and driedbefore a GC/MS analysis.

The chain-length distributions of n-alkanes and free/bound fattyacids in the sediment samples were determined using GC-MS (VGInstruments Platform II). The GC conditions were as follows: DB-1MS capillary column (60 m � 0.32 mm), with high-purity He ascarrier gas. The inlet temperature was 290 �C. The injection volumewas 1 mL, without split injection. The initial column temperature

L. Kang et al. / Environmental Pollution 219 (2016) 883e896886

began at 80 �C and was retained for 5 min, then it was heated at8 �C/min to 310 �C and was retained for 20 min until the samplewas completely drained. The MSD conditions were as follows:70 eV for an EI ionization source, 50~600 amu for the mass range,2341 V for the multiplier voltage, the ion source temperature at246 �C, and a selected ion monitoring (SIM) mode. The n-alkaneswith an even number of carbon produced standard markings, andthe odd push carbon n-alkanes were deduced according to the evencarbon markings. The internal standard substance was C24D50.Any odd-numbered carbon n-alkane were quantified with theaverage response factor of adjacent n-alkenes because their stan-dards were not available.

3. Results and discussion

3.1. Residual levels and spatial distributions of DDXs in the surfacesediments

3.1.1. Residual levels of DDXs in the surface sedimentsThe statistical features, results of the normal distribution test,

and frequency distribution histogram before/after the logarithmictransformation of DDX residue levels in the surface sediments arepresented in Table 1, Table S2, and Figs. S1 and S2 in the Supple-mentary Materials (SM), respectively. Comparisons of the currentresearch with other studies are tabulated in Table S3 in the SM.

There were 57 surface sediments samples (including parallelsamples) collected from the 19 sampling sites in Lake Chaohu. Allsix components of DDXs, including o,p0-DDE, p,p0-DDE, o,p0-DDD,p,p0-DDD, o,p0-DDT and p,p0-DDT, in the surface sediment weredetected by GC-MS. The most common component of DDXs was p,p0-DDE, with a detection rate of 100%, followed by p, p0-DDD(63.5%), o, p0-DDT (63.5%),o, p0-DDE (51.9%), o, p0-DDD (30.8%) andp, p0-DDT (17.3%). Table 1 shows that the content of total DDXs(P

DDXs) in the surface sediment ranged from 0.262 to 25.412 ng/g dw, with a geometric mean value of 3.093 ng/g dw. The p, p0-DDEhad the highest mean content (1.501 ng/g dw), followed by p, p0-DDD (0.354 ng/g dw), o, p0-DDT (0.241 ng/g dw), o, p0-DDE(0.028 ng/g dw), o, p0-DDD (0.017 ng/g dw) and p, p0-DDT (0.004 ng/g dw). All of the DDX components in the surface sediments fol-lowed log-normal distributions (Table S2, and Figs. S1 and S2).

As shown in Table S3, theP

DDXs level was lower in the presentstudy than in another study in Lake Chaohu (Li et al., 2015), whichmight be due to the different sampling sites. The concentration ofP

DDXs in the surface sediments from Lake Chaohu was higherthan that measured in the Songhua River (Cai et al., 2014), LakeQarun in Egypt (Barakat et al., 2013), Congo River Basin and LakeMaVall�ee in the Democratic Republic of Congo (Mwanamoki et al.,2014) and the Incheon North Harbor in Korea (Kim et al., 2010),but lower than measurements from Taihu Lake (Zhao et al., 2009),Honghu Lake (Yuan et al., 2013), East Lake (Yun et al., 2014), theYangtze River (Tang et al., 2007) and the CauBay River in Vietnam(Vu, 2015).

Table 1Statistical features of DDXs in the surface and core sediments from Lake Chaohu.

Chemicals Surfacea L4

Range Mean SD Range

o, p0-DDE N.D.~0.590 0.028 / N.D.~2.234p, p0-DDE 0.116~4.660 1.501 / 0.110~9.956o, p0-DDD N.D.~2.764 0.017 / 0.008~1.208p, p0-DDD N.D.~11.486 0.354 / 0.010~5.096o, p0-DDT N.D.~7.512 0.241 / 0.004e3.246p, p0-DDT N.D.~0.750 0.004 / 0.010~1.270P

DDXs 0.310~25.411 3.093 / 0.176~16.534

a In the surface sediments, geometric means were used. SD e Standard deviance.

3.1.2. Spatial distributions of DDXs in the surface sedimentsThe spatial distributions of DDX components and

PDDXs in the

surface sediments from Lake Chaohu are presented in Fig. 2. Asignificant difference in DDXs was found between the samplingsites in the western lake area (L1~L5) and in the eastern lake area(L6~L10). The DDX residues in the surface sediments from thewestern lake area were higher than those in the western lake area(Fig. 2a). The average residual level of

PDDXs in the surface sedi-

ments from the western lake was 9.454 ng/g dw (2.424e25.412 ng/g dw), while that from the eastern lake was 1.682 ng/g dw(0.520e5.509 ng/g dw) (Fig. 2b). A similar situation was observedbetween the western (R1, R3eR5) and eastern inflow rivers(R6eR9). The average concentration of

PDDXs in the surface sed-

iments in the western inflow rivers (2.602 ng/g dw(0.310e15.396 ng/g dw)) was higher than that in the eastern inflowrivers (2.108 ng/g dw (0.380e6.156 ng/g dw)) (Fig. 2b). Among thenine inflow river surface sediments, the R1 site in the NanfeiheRiver from Hefei City, a capital in Anhui Province, had the highestP

DDXs content to an extreme.The high-west and low-east spatial distribution pattern of DDXs

in the surface sediments might reflect the impact of differences inurbanization, industrialization and agriculture between the west-ern and eastern lake basin. Farmland is mainly distributed in thesouthwestern area of the Lake Chaohu Basin. The Hangbuhe-Fenglehe River flows through this area and carries approximately60% of the total water volume into Lake Chaohu. The Nanfeihe Riverflows through Hefei City and carries approximately 20% of the totalvolume of water into the lake. The combined effect of surface runofffrom the Hangbuhe-Fenglehe River and municipal sewage and in-dustrial wastewater from the Nanfeihe River would result in ahigher content of DDXs in the surface sediments from the westernlake area.

3.2. Historical variations of DDX residual levels in the sedimentcores

3.2.1. Residual levels of DDXs in the sediment coresThe statistical characteristics and the results of a normal dis-

tribution test of the residual levels of DDX components andP

DDXsin the two sediment cores, L4 and L8, from the western and easternregions of the lake are presented in Table 1 and Table S4, respec-tively. The frequency distribution histogram before/after a loga-rithmic transformation of the residue levels of DDXs in thesediment cores are presented in Fig. S3 e S6.

The DDX detection rates were all greater than 85% in both of theL4 and L8 sediment cores. As shown in Table 1, the

PDDXs residual

levels in the L4 and L8 sediment cores ranged from 0.176 to16.534 ng/g dw and from 0.076 to 3.01 ng/g dw, with a mean valueof 7.107 ± 5.382 ng/g dw and 1.332 ± 0.933 ng/g dw, respectively.For the L4 sediment core, the p, p0-DDE had the highest meanconcentration (3.279 ± 2.89 ng/g dw), followed by p,p0-DDD(2.429 ± 1.864 ng/g dw), o,p0-DDD (0.561 ± 0.455 ng/g dw), p,p0-

L8

Mean SD Range Mean SD

0.229 0.397 0.008~0.190 0.051 0.0463.279 2.89 0.008~1.634 0.777 0.5710.561 0.455 0.014~0.150 0.048 0.0422.429 1.864 0.014~0.660 0.282 0.2310.181 0.584 0.010~0.472 0.064 0.1170.428 0.367 N.D.~0.210 0.11 0.0747.107 5.382 0.076~3.010 1.332 0.933

Fig. 2. Spatial distributions of (a) DDX components and (b)P

DDXs in the surface sediments from Lake Chaohu.

L. Kang et al. / Environmental Pollution 219 (2016) 883e896 887

DDT (0.428 ± 0.367 ng/g dw), o,p0-DDE (0.229 ± 0.397 ng/g dw) ando,p0-DDT (0.181 ± 0.584 ng/g dw). In the L8 sediment core, the p,p0-DDE also had the highest mean concentration (0.777 ± 0.571 ng/g dw), followed by p,p0-DDD (0.282 ± 0.231 ng/g dw), p,p0-DDT(0.11 ± 0.074 ng/g dw), o,p0-DDT (0.064 ± 0.117 ng/g dw), o,p0-DDE(0.051 ± 0.046 ng/g dw) and o,p0-DDD (0.048 ± 0.042 ng/g dw). Thehigher content of DDXs and

PDDXs in the L4 sediment core from

the western lake area is consistent with the surface sediment. Thehigher residual levels of p, p0-DDE and p, p0-DDD in both the L4 andL8 sediment cores are also consistent with the surface sediment.This means that the p,p0-DDT metabolites, the p,p-DDE and p,p-DDD, were predominant in the surface and core sediments from

Lake Chaohu, which is consistent with the literature (Li et al., 2016).All of the DDX components and

PDDXs in both the L4 and L8

sediment cores followed log-normal distributions (Table S4, andFig. S3 e S6).

3.2.2. Historical variations of DDX residual levels in the sedimentcores

The historical variations of residual levels of DDX componentsand

PDDXs in the sediment cores L4 and L8 from Lake Chaohu are

illustrated in Fig. 3. The DDX residues in the L4 sediment core fromthe western lake region were historically higher than those in theL8 sediment core from the eastern lake region. The DDX

Fig. 3. Historic comparisons of residual levels of DDX components andP

DDXs in the L4 and L8 sediment cores from Lake Chaohu.

L. Kang et al. / Environmental Pollution 219 (2016) 883e896 889

contamination in Lake Chaohu underwent three stages from thelate 1930s to 2010. During the first stage, from late 1930s to late1940s, small amounts of DDX residues were detected in the sedi-ment cores. Although DDT was first synthesized in 1874, its insec-ticidal properties were not discovered until the 1937 (Apeti et al.,2010). The second stage, from the late 1940s until the late 1970sor early 1980s, the DDX residues in the sediment cores rapidlyincreased. The peak DDX contents in the sediment cores occurred inthe late 1970s or early 1980s. This is likely caused by the insensitiveusage of a large amount DDXs for agricultural and public healthpurposes. The third stage, starting from the early 1980s, the DDXresidues in the Lake Chaohu sediment cores were determined tohave a fluctuating and decreasing trend. This decrease potentiallyresulted from the government's ban on the usage of DDXs foragriculture due to their adverse health effects on wildlife and hu-man beings. For instance, in the US and China, DDXs were bannedin the 1970s and in, respectively. However, the actual deposition ofDDXs in the lake sediment occurred decades after the DDXs werebanned in the 1970se1980s because of their persistent character-istics (Li et al., 2015). As shown in Fig. 3, an increase in the early tomiddle 1990s and mid-2000s was observed in two sediment cores.The reasons for these increases may be because a new type ofpesticide, such as dicofol containing DDTs, is being used to control avariety of mites and annihilating mite eggs in agriculture andbecause anti-fouling paints containing DDTs have been used for3500 fishing vessels in Lake Chaohu (He et al., 2012, 2015).

3.3. Compositions and sources of DDXs in the surface and coresediments

3.3.1. Spatial variations of DDX compositions in the surfacesediments

The spatial variations of DDX compositions in the surface sedi-ments are presented in Fig. 4 and in Table S5. For all of the lake andriver samples, p, p0-DDE (69.96%) was the major component ofDDXs, while the percentages of p, p0-DDD and o, p0-DDT were16.49% and 11.23%, respectively. Other DDX components made upless than 5% of the total DDXs. There were no significant differencesin the compositions of the DDXs, but there was a predominance ofp, p0-DDE in the surface sediments between thewhole lake samplesand the whole inflow river samples. However, significant differ-ences (p < 0.01) in the compositions of DDXs in the surface

Fig. 4. Spatial variations of DDX composition in the surface sediments from LakeChaohu (WB: whole lake and river samples, WL: whole lake, IR: inflow rivers, WEL:western lake area, EAL: eastern lake area, WIR: western inflow rivers and EIR: easterninflow rivers).

sediment between the western lake samples and the eastern lakesamples were found. In the surface sediments from the westernlake area, the average percentages of p, p0-DDE, p, p0-DDD and o, p0-DDT were 32.50%, 38.76% and 25.36%, respectively, while those inthe surface sediments from the eastern lake area were 88.88%,5.56% and 3.88%, respectively. Furthermore, there were some dif-ferences in the compositions of DDXs in the surface sedimentsbetween the western inflow river samples and the eastern inflowriver samples. In the surface sediments from the western inflowrivers, the average percentages of p, p0-DDE, p, p0-DDD and o, p0-DDT were 65.92%, 17.45% and 12.13%, respectively, while those inthe surface sediment from the eastern inflow rivers were 76.67%,12.87% and 8.73%, respectively. The significant differences in theaverage percentages of p, p0-DDE and p, p0-DDD in the surfacesediment from thewestern lake area and from the eastern lake areaare likely attributed to the different degradation conditions causedby the different intensities of algal blooms in the two lake areas(please see next section for discussion).

3.3.2. Historical variations in DDX composition in the sedimentcores

The historical variations of DDX composition in the L4 and L8sediment cores from Lake Chaohu are presented in Fig. 5. The p, p0-DDT in both the L4 and L8 sediment cores was still detected in all

Fig. 5. Historic variations of DDX composition in the sediment cores (a) L4 and (b) L8from Lake Chaohu.

Fig. 6. Source apportionment of DDXs in the surface sediments of Lake Chaohu.

L. Kang et al. / Environmental Pollution 219 (2016) 883e896890

of the sediment slices after the late 1930s, with an average per-centage of 6.12% and 8.03%, ranging from 3% to 9.61% and from3.04% to 10.68%, respectively. Furthermore, the o, p0-DDT in boththe L4 and L8 sediment cores was still detectable in all of thesediment slices after the late 1930s, with an average percentage of2.01% and 4.2%, ranging from 0.13% to 24.33% and from 0.58% to17.52%, respectively. Higher percentages of p, p0-DDE and p, p0-DDD were found in both the L4 and L8 sediment cores. In the L4sediment core from the western lake area, the average percent-ages after the late 1930s of p, p0-DDE and p, p0-DDD were 46.87%and 35.01%, ranging from 26.82% to 65.32% and from 21.63% to46.72%, respectively; the percentages in the L8 sediment core fromthe eastern lake areawere 59.34% and 21.30%, ranging from 47.81%to 75.39% and from 9.44% to 30.21%, respectively. However,different historical variations were found of p, p0-DDE and p, p0-DDD in the two sediment cores. In the L8 sediment core, therewere no significant historical variation trend of p, p0-DDE and p,p0-DDD; however, in the L4 sediment core, there was a significantdecrease in p, p0-DDE and a significant increase in p, p0-DDD afterthe early 1970s. These different historical variation trends be-tween p, p0-DDE and p, p0-DDD in the L4 sediment core are likelyattributed to the changes in the degradation conditions caused byalgal blooms after the early 1970s (please see the next section for adiscussion).

3.3.3. Source apportionment of DDXs in the surface and coresediments

The primary sources of DDXs in the environment include theproduction and use of technical DDT, pesticides containing DDTsuch as dicofol, marine anti-fouling paints, and other goods con-taining technical DDT (e.g., mosquito repellent incense) (He et al.,2015). The sources of DDX can be determined by their composi-tion (He et al., 2012). The o,p0-/p,p0-DDT ratio is indicative oftechnical DDT or dicofol (He et al., 2015). The o,p0-/p,p0-DDT ratio isapproximately 0.2 for technical DDT and approximately 7 fordicofol, since technical DDT contains approximately 75% p,p0-DDTand 15% o,p0-DDT (WHO,1989), and one kilogram of Chinese dicofolcontains 17 g of p,p0-DDT and 114 g of o,p0-DDT (Qiu et al., 2004).The ratio of parent compounds to their metabolic products (DDT/(DDE þ DDD)) can be used for determining new or historical inputsbecause parent DDTs can be degraded gradually to DDD and DDE. ADDT/(DDE þ DDD) ratio <1 indicates the historical inputs of DDTs,whereas a DDT/(DDE þ DDD) ratio �1 is indicative of the new in-puts of DDTs (Hitch and Day, 1992). The DDD/DDE ratio can be usedto identify the degradation of parent compounds (DDTs). A DDD/DDE ratio <1 suggests aerobic conditions, whereas a DDD/DDEratio > indicates anaerobic conditions because DDTs can be bio-degraded to form DDEs through an oxidative dehydrochlorinationprocess under aerobic conditions, and can form DDDs through areductive dechlorination process under anaerobic conditions(Schmitt et al., 1990).

As shown in Fig. 6, the DDT/(DDE þ DDD) ratios of all of thesurface sediment samples from the lake and from the inflow riverswere less than one, which indicates the sources of DDXs were thehistorical inputs of DDTs. For the surface sediment samples at sitesL1eL4 from the western lake sites, at site L10 from the eastern lake,and at site R1 from the Nanfeihe River, one of the western inflowrivers, the DDD/DDE ratios were greater than one, which indicates adominance of anaerobic conditions in the sediment. The above-mentioned surface sediment sample sites (L1eL4, L10 and R1)were vulnerable to urban DDXs' pollution from the cities of Hefeiand Chaohu and were likely suffering from algal blooms associatedwith eutrophication. In other surface sediment sample sites, theDDD/DDE ratios were less than one, which indicates a dominance

of aerobic conditions. As show in Table S5 in SI, in the surfacesediments at sites L6 and R6, p,p0-DDT was detectable and o,p0-DDTwas not detectable, which indicates that the source of DDX wastechnical DDT. However, for the surface sediments at sites L1eL5,L9 and R5, R7eR9, p,p0-DDT was not detectable and o,p0-DDT wasdetectable, which indicates that the sources of DDXs were fromdicofol. For the surface sediments from sites L10 and R3, R4, bothp,p0-DDT and o,p0-DDT were detectable, the o,p0-/p,p0-DDT ratiowas 3.95, 0.7 and 0.4 for sites L10, R3 and R4, respectively, whichsuggests that the sources of DDX included both dicofol and tech-nical DDT, in different proportions.

As shown in Fig. 7, the DDT/(DDE þ DDD) ratios for all thesediment slices in both the L4 and L8 cores were less than one,which indicates that the sources of DDX were the historical inputsof DDTs (Fig. 7a). With the exception of the two sediment slices inthe L4 core and five sediment slices in the L8 core, the o,p0-/p,p0-DDT ratios of the all of the remaining sediment slices in the L4 andL8 cores were approximately 0.2, which means that the primarysources of DDX were technical DDTs (Fig. 7b). The historical var-iations of DDD/DDE ratios in the L4 and L8 sediment cores (Fig. 7c)indicated different degradation conditions in the western andeastern lake area, especially after the late 1970s. For the L8 sedi-ment core from the eastern lake area, the DDD/DDE ratios wereless than one, which indicates dominance of aerobic conditions.However, for the L4 sediment core, the DDD/DDE ratios were lessthan one before the late 1980s, and greater than one after the late1970s, which indicates changes in the degradation conditionsfrom aerobic to aerobic in the early 1980s. The differences in thedegradation condition changes in the western and eastern lakeareas are likely attributed to their different eutrophic states,showing different intensities of algal blooms. The western lakearea receives far more nutrients from the municipal sewage andindustrial wastewater of Hefei City, the capital of Anhui Province,and from the agricultural runoff in the Hangbuhe-Fenglehe River,which carries 60% of the total inflow water volume into the lake.This causes more severe eutrophication in the western lake area.The intensive occurrences of algal blooms from the early 1980s inthe western lake area would result in the anaerobic conditionsbecause of dead algae biodegradation. However, compared withthe western lake area, the eastern lake area was suffering the ef-fects of algal blooms less. The DDD/DDE ratios >1 in the sedimentcore of the western lake area from the early 1980s could serve asdirect evidence of the anaerobic conditions in the sedimentcaused by the intensive algal blooms.

Fig. 7. Source apportionment of DDXs in the sediment cores of Lake Chaohu. (a) DDT/(DDE þ DDD), (b) o,p0-DDT/p,p0-DDT, (c) DDD/DDE.

Fig. 8. Relationships between annual DDT production in China and the residue level oftotal DDXs (

PDDXs) in the L4 and L8 sediment cores from Lake Chaohu.

L. Kang et al. / Environmental Pollution 219 (2016) 883e896 891

3.4. Influencing factors of DDX residues and distributions in thesurface and core sediments

3.4.1. The effects of DDT production in ChinaA Pearson correlation test was preformed to analyze the effects

of the annual usage or production of DDT in China on the historicalsediment records of DDXs in Lake Chaohu (L4 and L8), and theresults are presented in Fig. 8. A significant positive linear rela-tionship was found between annual DDT production in China and

the residual levels of total DDXs (P

DDXs) in the L4 and L8 sedimentcores, with a determination coefficient (R2) of 0.653 and 0.475, andat a significance level (p) of 0.01 and 0.05, respectively. This sig-nificant linear relationship indicates that the residual levels ofP

DDXs in the sediment cores of Lake Chaohu could sufficientlyrepresent the historical annual production of technical DDT inChina.

3.4.2. The effects of sediment granularityThe results of a correlation test between the sediment granu-

larity and the residual levels of DDXs in the sediment cores aresummarized in Table 2. For the L4 sediment core, significant posi-tive relationships (r ¼ 0.596~0.906, p < 0.01) were found betweenthe percentage contents of clay (< 4.5 mm) and the residual levels ofDDXs, except for o, p0-DDT, and the significant negative relation-ships (r ¼ �0.461~�0.709, p < 0.01 or 0.05) between the percent-age contents of sand (18e190 mm) and the residual levels of p, p0-DDE, p, p0-DDT and total DDXs. This indicates that DDXs are likelyadsorbed onto the fine particles in the sediments, which isconsistent with other studies (e.g., Li et al., 2015). However, for theL8 sediment core, the relationships between particle size and theassociation of DDXs with particles were complex.

3.4.3. The effects of TOC in the sedimentPearson correlation tests of the DDX residues and TOC content in

the surface and core sediments from Lake Chaohu were performed,and the results are summarized in Table 3. The considerable dif-ferences in the relationships of DDX residues and TOC contentswere found between the surface sediments and the core sediments,and between the L4 sediment core and the L8 sediment core, whichmeans that the relationships between the residual levels of DDXand the TOC contents in the sediments are complex. For the surfacesediments, only o,p0-DDE, p,p0-DDE and

PDDXs exhibited signifi-

cant positive correlations with TOC (r ¼ 0.584~0.723, p < 0.01 or0.05). However, for both the L4 and L8 sediment cores, there wereno significant correlations between the TOC contents and the re-sidual levels of all of the DDX components and

PDDXs from the

late 1930s to 2010. The differences in the relationships of the DDXresidues with the TOC between the surface sediments and the coresediments might imply that the biodegradations of DDX and TOC inthe sediment would be different during their burial.

Table 2Pearson correlation between the DDXs' residues and particle sizes in sediment cores from Lake Chaohu.

Chemicals L4 L8

Clay Silt Sand Grit Clay Silt Sand Grit

o, p0-DDE 0.597b �0.212 �0.027 0.154 �0.208 0.039 �0.002 �0.154p, p0-DDE 0.596b 0.517a �0.709b �0.359 0.292 0.218 �0.254 0.111o, p0-DDD 0.798b �0.272 �0.032 0.031 �0.575a �0.647b 0.668b 0.455p, p0-DDD 0.806b �0.249 �0.055 0.005 �0.266 �0.498a 0.491a 0.359o, p0-DDT 0.181 �0.202 0.147 �0.135 �0.319 �0.116 0.113 0.656b

p, p0-DDT 0.683b 0.395 �0.621b �0.341 0.519a 0.256 �0.276 �0.594bP

DDXs 0.772b 0.193 �0.461a �0.247 �0.081 �0.146 0.125 0.423

Clay: <4.5 mm, Silt: 4.5e18 mm, Sand: 18e190 mm, Grit: >190 mm.a The correlation is significant at the 0.05 level (2-tailed).b The correlation is significant at the 0.01 level (2-tailed).

Table 3Pearson correlation between the DDXs' residues and the TOC contents in the surface and core sediments from Lake Chaohu.

Contaminant Sediment core L4 Sediment core L8 Surface sediment

1940se 1940se1970s 1980se 1940se 1940se1970s 1980se

o,p0-DDE 0.271 0.813a �0.350 0.118 0.817a 0.157 0.627b

p,p0-DDE �0.331 0.830a �0.740b �0.124 �0.079 �0.169 0.723b

o,p0-DDD 0.358 0.823a �0.427 0.425 0.623 0.148 0.279p,p0-DDD 0.335 0.867a �0.468 0.377 �0.643 �0.131 0.467o,p0-DDT 0.138 0.835a 0.030 0.049 0.045 �0.005 0.466p,p0-DDT �0.247 0.878b �0.776b �0.170 �0.022 �0.429 �0.318P

DDXs �0.046 0.844a �0.712b 0.133 �0.069 �0.132 0.584a

a The correlation is significant at the 0.05 level (2-tailed).b The correlation is significant at the 0.01 level (2-tailed).

L. Kang et al. / Environmental Pollution 219 (2016) 883e896892

As shown in Table 3, for the L4 sediment core from the westernlake area, there was a significant positive correlation between TOCand DDX (including the residual levels of all DDX components andP

DDXs) from the 1940s to the 1970s (r ¼ 0.813~0.878, p < 0.05),and the significant negative correlations between the TOC contentsand the residual levels of p,p0-DDE, p,p0-DDT and

PDDXs during

the early 1980s to 2010 (r¼�0.712~�0.776, p < 0.01). However, forthe L8 sediment core from the eastern lake area, only o,p0-DDE hada significant positive correlation with TOC from the 1940s to the1970s (r ¼ 0.817, p < 0.05). These significant positive correlationsduring the 1940s to the 1970s and the significant negative corre-lations from the early 1980s to 2010 between the TOC content andthe residual levels of DDX in the L4 sediment core suggest that theresidual levels of DDX in the sediment are determined by both theusage of DDXs and the TOC contents. The combined effects of thecontinuous increase of TOC from the 1940s to 2010 and thecontinuous increase in TOC from the 1940s to the 1970s, along withthe fluctuating decrease of TOC from the early 1940s to 2010, islikely a result of such different relationships between the TOCcontent and the residual levels of DDXs in the L4 sediment corebefore and after the early 1080s or the late 1970s.

3.4.4. The effects of algae-derived organic matter in the sedimentFive indicators, including C17/C31 n-alkanes, the contribution of

algae-derived organic matters (calculated as follows: [C17/(C17 þ C31) þ 2C17/(2C17 þ C23 þ C25)]/2), C16 bound fatty acids,C18 bound fatty acids, and the total bound fatty acids, were selectedas the parameters indicating algae-derived organic matter. Forvarious n-alkane sources, the peaks of n-alkane carbon numberdistributions are different. C17 is the main peak of algae andphotosynthetic bacteria-sourced organic matter (Cranwell et al.,1987). The emergent and floating plants and other large non-exogenous vascular plants are abundant from n-alkanes in C21,C23 and C25. Nevertheless, C27 and C29 represent the input ofwoody terrestrial plants, and C31 indicates the input of herbaceous

terrestrial plants (Meyers, 2003). Therefore, C17/C31 n-alkanes andthe contribution of algae-derived organic matter could wellrepresent the ratio of n-alkanes from algae. For fatty acids, thepeaks of carbon number distribution could indicate sources oforganic matter, too. C16 and C18 bound fatty acids are dominantshort-chain fatty acids, which indicate input of plankton, photo-synthetic bacteria and aquatic plants (Meyers and Ishiwatari, 1993).Hence, C16, C18 and the total bound fatty acids were chosen asindicators of algae-derived organic matter.

The historical variations of the algae-derived organic matterindicators in both the L4 and L8 sediment core from Lake Chaohuare presented in Fig. S7. The Pearson correlation between the DDXresidues and the algae-derived organic matter indicators are pre-sented in Table S6. The relationships between the DDD/DDE ratiosand the algae-derived organic matter indicators are also illustratedin Fig. 9.

As shown in Fig. S7, the five algae-derived organic matter in-dicators significantly increased after the late 1970s. This is associatedwith the mass algal blooms after the late 1970s, especially in thewestern lake area (L4) (Kong et al., 2016). Table S6 shows how the L4and L8 sediment cores had obvious differences in the relationship ofthe algae-derived organic matter indicators with DDX residues andwith the DDD/DDE ratios. In case the algal bloom resulted in a lack ofoxygen in the sediments during the same period, thereby affectingthe value of the DDD/DDE, we explained the differences by thetiming of the algal bloom. This phenomenon might likely be attrib-uted to the different intensities of algal blooms in the western andeastern lake areas after the late 1970s. The large algal blooms havebeen occurring frequently since the early 1980s in the western lakearea (L4), whichmay have resulted in the anaerobic conditions in thesediment. This could explain the significant positive relationships ofthe algae-derived organic matter indicators with the residues of p,p0-DDD and o, p0-DDD (r ¼ 0.507~0.633, p < 0.01 or 0.05) and withthe DDD/DDE ratios (r ¼ 0.796~0.87, p < 0.01) in the western lakearea (L4) after the late 1970s.

Fig. 9. Correlations between algae-derived organic matter indicators and the DDD/DDE ratios in the L4 and L8 sediment cores from Lake Chaohu.

L. Kang et al. / Environmental Pollution 219 (2016) 883e896 893

3.4.5. Association between the DDXs and multiple influencingfactors

A principle component analysis (PCA) and multiple regressionanalysis were used to further investigate the association betweenDDXs in the sediment and the above influencing factors, such asparticle size, TOC, and algae-derived organic matter (Table 4 and

Fig. 10). Two components were extracted using PCA. For both L4and L8, the 1st component (PC 1) was dominant, accounting for>65% of the explained variances. PC 1 was associated with algae-derived organic matter and sand. However, the 2nd component(PC 2) was associated with DDXs and clay in L4 and with DDD/DDEin L8. The DDX residue seemed to be associated with fine size

Table 4The contribution of TOC, algae-derived organic matters, and particle size to theresidues of DDXs and anaerobic/aerobic condition of sediment (DDE/DDD) revealedby the multiple regression analysis in Table S7.

L4 L8

DDXs DDE/DDD DDXs DDE/DDD

Standardized regression coefficienta

TOC 0.13 (1.1%) 0.04 (0.6%) �0.13 (1.1%) �0.13 (1.2%)C17/C31 �0.36 (3.2%) �0.19 (2.9%) �0.79 (6.9%) 0.84 (8.2%)AOM 1.16 (10.2%) 0.38 (6.0%) 0.88 (7.7%) �0.66 (6.4%)C16FA �2.20 (19.4%) �1.56 (24.4%) 1.00 (8.8%) �0.17 (1.6%)C18FA �0.02 (0.2%) 0.19 (3.0%) �0.70 (6.1%) 0.33 (3.2%)TFA 2.58 (22.7%) 1.20 (18.8%) 0.07 (0.6%) �0.41 (4.0%)Clay 1.10 (9.7%) 0.69 (10.8%) 0.53 (4.6%) �0.50 (4.9%)Silt 2.41 (21.2%) 0.65 (10.2%) 3.35 (29.3%) �3.42 (33.3%)Sand 1.39 (12.2%) 1.49 (23.3%) 4.00 (34.9%) �3.81 (37.2%)Model fittingb

R2 0.648 0.915 0.583 0.196ANOVA 0.004 0.000 0.017 0.827

a C17/C31, AOM, C16FA, C18FA, and TFA denoted the C17/C31 n-alkanes, contri-bution of algae-derived organic matters, C16 bound fatty acids, C18 bound fattyacids, and total bound fatty acids, respectively; the standardized regression co-efficients of the nine-parameter linear model were employed to compare thecontribution of each indicator. The values in the bracket, denoting the relativecontribution of each indicator, were calculated by dividing the absolute values ofeach coefficient to the sum of absolute values of all coefficients.

b The performance of the nine-parameter models were checked by the coefficientof determination (R2) and the significance revealed by ANOVA.

DDXs

DDD/DDE

TOC

C17/C31AOM

C16FAC18FATFA

Clay

Silt

Sand

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

-1 -0.5 0 0.5 1

PC 2

(18.

1%)

PC 1 (70.6%)

DDXs

DDD/DDE

TOCC17/C31

AOM

C16FAC18FATFA

ClaySilt

Sand

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

-1 -0.5 0 0.5 1

PC 2

(10.

4%)

PC 1 (68.3%)

(a)

(c)

Fig. 10. Loading (a and c) and score (b and d) plots of the first 2 components extracted by thalgae-derived organic matter, and particle size of core sediments L4 (a and b) and L8 (c an

L. Kang et al. / Environmental Pollution 219 (2016) 883e896894

sediment (clay) and aerobic/anaerobic conditions (DDD/DDE ra-tios), and DDX degradation was affected by algae-derived organicmatter in core sediment from west lake, which is consistent withthe correlation analysis above. The PCA again proved that eutro-phication, causing aquatic hypoxia, would definitely affect thedegradation behavior of DDT in a eutrophic lake. On the contrary,the east part of Lake Chaohu was less affected by eutrophication,and the aerobic microorganisms or other degradation mechanisms(e.g., photo-oxidization) might dominate DDX transformation fromDDT to DDE. Furthermore, the residues of the DDXs were coupledwith algae-derived organic matter probably due to the coupledusage of OCPs and nitrogen-phosphorus fertilizers from agricul-tural activity. In detail, these two activities would increase DDXresidues and nutrient levels in Lake Chaohu from rivers flowinginto the western part of the lake. The algae blooms were wide-spread where high-levels of nutrients occurred in the lakeecosystem and caused large quantity of algae-derived organicmatter. This phenomenonwas more clearly observed in the easternpart than in the western part because aquatic hypoxia in the laststage of algae bloom increased the activity of anaerobic microor-ganisms. The coupling mechanisms between DDXs and algae-derived organic matter were damaged by the bio-degradation ofanaerobic microorganisms. After identifying the PC 1 and PC 2, wereconstructed the historical variation of DDXs and their associationwith algae-derived organic matter in the score plots (Fig. 10 b andd). Interestingly, algae-derived organic matter increased moreseriously from the 1980s to the present than before, in both the

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

-3 -2 -1 0 1 2 3

PC 2

(18.

1%)

PC 1 (70.6%)

1980s-

1940s-1970s

-1940s

-4

-3

-2

-1

0

1

2

3

4

-3.5 -2.5 -1.5 -0.5 0.5 1.5 2.5 3.5

PC 2

(10.

4%)

PC 1 (68.3%)

1980s-

1940s-1970s

-1940s

(b)

(d)

e PCA for DDX residues, anaerobic/aerobic condition of the sediment (DDE/DDD), TOC,d d).

L. Kang et al. / Environmental Pollution 219 (2016) 883e896 895

eastern and western parts of Lake Chaohu. DDXs increased beforethe 1980s and decreased after the 1980s in the western part anddecreased before the 1940s and increased after the 1940s in eastpart. The anaerobic conditions increased gradually after the 1980sin the western lake area and increased before the 1980s anddecreased after the 1980s in the eastern lake area.

The correlation analysis and PCA showed the potentialconnection between DDXs and the influencing factors and betweenDDE/DDD and the influencing factors. Thus, a multiple regressionanalysis was conducted as shown in Table S7 and summarized inTable 4. In the western part of Lake Chaohu, better performance ofthe nine-parameter models was obtained than in the eastern part,indicating that those factors potentially affected the DDX residuesand their transformation. Based on the standardized regressioncoefficient, we found that TOC contributed litter (1.1%) to the DDXsin both L4 and L8. The algae-derived organic matter contributedmore (>50%) than particle size to the DDX residues and theirtransformation in the western part, whereas particle size contrib-uted more in the eastern part. Conclusively, the algae-derivedorganic matter, as a biomarker of eutrophication, was an impor-tant factor that affected the DDX residues and lifecycles in the lakeecosystem.

4. Conclusion

p,p0-DDT metabolites, p,p-DDE and p,p-DDD, were found to bethe two dominant components of DDXs in the both surface and coresediments. The historical usage of technical DDT was identified asthe primary source of DDXs in the sediment, as indicated by DDT/(DDD þ DDE) ratios of less than one. Higher DDX residual levelswere found in surface and core sediments from the western lakearea, whichmight have been caused by the combined inflow effectsof municipal sewage and industrial wastewater from the NanfeiheRiver, carrying approximately 20% of the total water volume intothe lake, and from agricultural runoff from the Hangbuhe-FengleheRiver, contributing approximately 60% of the total water volumeinto the lake. Historical variations of DDX residues in the sedimentcores during the late 1930s to 2010 were divided into three stages:the first stage occurred from the late 1930s to the late 1940s andexhibited small amounts of DDX residue; the second stage occurredfrom the late 1940s to the late 1970s and exhibited a significantincrease in DDXs; and the third stage, from the early 1980s to 2010,had a fluctuating decreasing trend.

DDT production in China, fine particulate sizes (<4.5 mm), andalgae-derived organic matter significantly influenced the amountof residue, composition and distribution of DDXs in the sediments.However, the relationships between DDXs and the TOC in sedimentwere found to be more complex, as indicated by the significantdifferences between surface and core sediments; between the L4and L8 sediment cores from the western and eastern lake areas,respectively; and in the samples before and after the late 1970s orearly 1980s in the L4 sediment cores. The different intensities ofalgal blooms in the western and eastern lake areas after the late1970s caused significant differences in the relationships of thealgae-derived organic matter indicators with the DDX residues andthe DDD/DDE ratios. The historical variations of the DDD/DDE ratiowell-represented the anaerobic conditions in the sediments causedby mass algal blooms after the late 1970s in the western lake area.Therefore, the DDD/DDE ratio could serve as an indicator of theeffects of algal blooms on the redox state of lake sediments.

Acknowledgments

Funding for this study was provided by the National Project forWater Pollution Control (2012ZX07103-002), and the National

Natural Science Foundation of China (NSFC) (41503083, 31570508,41271462, 41030529). This work was also supported by a grantfrom the 111 Project (B14001) and from the undergraduate studentresearch training program (YKM Entrepreneurship EducationFoundation) of the Peking University.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.envpol.2016.08.072.

References

Aamir, M., Khan, S., Nawab, J., Qamar, Z., Khan, A., 2016. Tissue distribution of HCHand DDT congeners and human health risk associated with consumption of fishcollected from Kabul River, Pakistan. Ecotoxicol. Environ. Saf. 125, 128e134.

Apeti, D.A., Lauenstein, G.G., Christensen, J.D., Kimbrough, K., Johnson, W.E.,Kennedy, M., Grant, K.G., 2010. A historical assessment of coastal contaminationin Birch Harbor, maine based on the analysis of mussels collected in the 1940and the mussel watch program. Mar. Pollut. Bull. 60 (5), 732e742.

Barakat, A.O., Khairy, M., Aukaily, I., 2013. Persistent organochlorine pesticide andPCB residues in surface sediments of Lake Qarun, a protected area of Egypt.Chemosphere 90, 2467e2476.

Booij, P., Holoubek, I., Klanova, J., Kohoutek, J., Dvorska, A., Magulova, K., Al-Zadjali, S., Cupr, P., 2016. Current implications of past DDT indoor spraying inOman. Sci. Total Environ. 550, 231e240.

Bouwman, H., Bornman, R., van Dyk, C., Barnhoorn, I., 2015. First report of theconcentrations and implications of DDT residues in chicken eggs from amalaria-controlled area. Chemosphere 137, 174e177.

Cai, S., Sun, K., Dong, S., Wang, Y.M., Wang, S., Jia, L., 2014. Assessment of organo-chlorine pesticide residues in water, sediment, and fish of the Songhua River,China. Environ. Forensics 15, 352e357.

Cao, H., Liang, T., Tao, S., Zhang, C., 2007. Simulating the temporal changes of OCPpollution in Hangzhou, China. Chemosphere 67, 1335e1345.

Cranwell, P.A., Eglinton, G., Robinson, N., 1987. Lipids of aquatic organisms as po-tential contributors to lacustrine sediments .2. Org. Geochem. 11, 513e527.

Cundy, A.B., Stewart, I.S., 2004. Dating recent colluvial sequences with Pb-210 andCs-137 along an active fault scarp, the Eliki Fault, Gulf of Corinth, Greece. Tec-tonophysics 386, 147e156.

Gerber, R., Smit, N.J., Van Vuren, J.H.J., Nakayama, S.M.M., Yohannes, Y.B., Ikenaka, Y.,Ishizuka, M., Wepener, V., 2016. Bioaccumulation and human health riskassessment of DDT and other organochlorine pesticides in an apex aquaticpredator from a premier conservation area. Sci. Total Environ. 550, 522e533.

He, Q., Wang, Q., Wang, Y., He, W., Qin, N., Kong, X., Liu, W., Yang, B., Xu, F.L., 2015.Temporal and spatial variations of organochlorine pesticides in the suspendedparticulate matter from Lake Chaohu, China. Ecol. Eng. 80, 214e222.

He, W., Qin, N., He, Q., Wang, Y., Kong, X., Xu, F., 2012. Characterization, ecologicaland health risks of DDTs and HCHs in water from a large shallow Chinese lake.Ecol. Inf. 12, 77e84.

He, W., Qin, N., Kong, X., Liu, W., He, Q., Wang, Q., Yang, C., Jiang, Y., Yang, B., Wu, W.,Xu, F., 2014. Water quality benchmarking (WQB) and priority control screening(PCS) of persistent toxic substances (PTSs) in China: necessity, method and acase study. Sci. Total Environ. 472, 1108e1120.

He, W., Yang, C., Liu, W., He, Q., Wang, Q., Li, Y., Kong, X., Lan, X., Xu, F., 2016. Thepartitioning behavior of persistent toxicant organic contaminants in eutrophicsediments: coefficients and effects of fluorescent organic matter and particlesize. Environ. Pollut. 216, 724e734. http://dx.doi.org/10.1016/j.envpol.2016.07.014.

Hitch, R.K., Day, H.R., 1992. Unusual persistence of DDT in some western USA soils.Bull. Environ. Contam. Toxicol. 48, 259e264.

Jiang, Y.J., He, W., Liu, W.X., Qin, N., Ouyang, H.L., Wang, Q.M., Kong, X.Z., He, Q.S.,Yang, C., Yang, B., Xu, F.L., 2014. The seasonal and spatial variations of phyto-plankton community and their correlationwith environmental factors in a largeeutrophic Chinese lake (Lake Chaohu). Ecol. Indic. 40, 58e67.

Kong, X.Z., He, Q.S., Yang, B., He, W., Xu, F.L., Janssen, A.B.G., Kuiper, J.J., vanGerven, L.P.A., Qin, N., Jiang, Y.J., Liu, W.X., Yang, C., Bai, Z.L., Zhang, M.,Kong, F.X., Janse, J.H., Mooij, W.M., 2016. Hydrological regulation drives regimeshifts: evidence from paleolimnology and ecosystem modeling of a largeshallow Chinese lake. Glob. Change Biol. http://dx.doi.org/10.1111/gcb.13416.

Kang, L., Wang, Q.M., He, Q.S., He, W., Liu, W.X., Kong, X.Z., Yang, B., Yang, C.,Jiang, Y.J., Xu, F.L., 2016. Current status and historical variations of phthalateester (PAE) contamination in the sediments from a large Chinese lake (LakeChaohu). Environ. Sci. Pollut. Res. 23, 10393e10405.

Kannan, K., Tanabe, S., Tatsukawa, R., 1995. Geographical distribution and accu-mulation features of organochlorine residues in fish of tropical Asia and Oce-ania. Environ. Sci. Technol. 29, 2673e2683.

Kim, S., Kang, D., Kim, K., Lee, D.S., 2010. Distribution of organochlorine pesticides inintertidal and subtidal sediments in coastal wetland with high tidal ranges.Arch. Environ. Contam. Toxicol. 58, 514e522.

Li, C., Huo, S., Xi, B., Yu, Z., Zeng, X., Zhang, J., Wu, F., Liu, H., 2015. Historicaldeposition behaviors of organochlorine pesticides (OCPs) in the sediments of a

L. Kang et al. / Environmental Pollution 219 (2016) 883e896896

shallow eutrophic lake in Eastern China: roles of the sources and sedimento-logical conditions. Ecol. Indic. 53, 1e10.

Li, C., Sun, H.J., Juhasz, A.L., Cui, X.Y., Ma, L.Q., 2016. Predicting the relativebioavailability of DDT and its metabolites in historically contaminated soilsusing a Tenax-Improved Physiologically Based Extraction test (TI-PBET). Envi-ron. Sci. Technol. 50 (3), 1118e1125.

Li, J., Zhang, G., Qi, S., Li, X., Peng, X., 2006. Concentrations, enantiomeric compo-sitions, and sources of HCH, DDT and chlordane in soils from the Pearl RiverDelta, South China. Sci. Total Environ. 372, 215e224.

Lin, T., Wang, L., Chen, Y., Tian, C., Pan, X., Tang, J., Li, J., 2014. Sources and preser-vation of sedimentary organic matter in the Southern Bohai Sea and the YellowSea: evidence from lipid biomarkers. Mar. Pollut. Bull. 86, 210e218.

Liu, W., He, W., Qin, N., Kong, X., He, Q., Ouyang, H., Xu, F., 2013. The residues,distribution, and partition of organochlorine pesticides in the water, suspendedsolids, and sediments from a large Chinese lake (Lake Chaohu) during the highwater level period. Environ. Sci. Pollut. Res. 20, 2033e2045.

Liu, W.X., Wang, Y., He, W., Qin, N., Kong, X.Z., He, Q.S., Yang, B., Yang, C., Jiang, Y.J.,Jorgensen, S.E., Xu, F.L., 2016. Aquatic biota as potential biological indicators ofthe contamination,bioaccumulation and health risks caused by organochlorinepesticides in a large, shallow Chinese lake (Lake Chaohu). Ecol. Indic. 60,335e345.

Liu, Y., Song, C., Li, Y., Liu, Y., Song, J., 2012. The distribution of organochlorinepesticides (OCPs) in surface sediments of Bohai Sea Bay, China. Environ. Monit.Assess. 184, 1921e1927.

Ma, H., Cui, F., Zhao, Z., Liu, Z., Liu, D., 2012. In: Iranpour, R., Zhao, J., Wang, A.,Yang, F.L., Li, X. (Eds.), Water solubility enhancement of DDT by humic acid andextracellular organic matter of aquatic algae. Adv. Mater. Res. 565e568.

Meyers, P.A., Ishiwatari, R., 1993. Lacustrine organic geochemistry - an overview ofindicators of organic-matter sources and diagenesis in lake-sediments. Org.Geochem. 20, 867e900.

Meyers, P.A., 2003. Applications of organic geochemistry to paleolimnological re-constructions: a summary of examples from the Laurentian Great Lakes. Org.Geochem. 34, 261e289.

Mrema, E.J., Rubino, F.M., Brambilla, G., Moretto, A., Tsatsakis, A.M., Colosio, C., 2013.Persistent organochlorinated pesticides and mechanisms of their toxicity.Toxicology 307, 74e88.

Mwanamoki, P.M., Devarajan, N., Thevenon, F., Birane, N., de Alencastro, L.F.,Grandjean, D., Mpiana, P.T., Prabakar, K., Mubedi, J.I., Kabele, C.G., Wildi, W.,Pote, J., 2014. Trace metals and persistent organic pollutants in sediments fromriver-reservoir systems in Democratic Republic of Congo (DRC): spatial distri-bution and potential ecotoxicological effects. Chemosphere 111, 485e492.

Najam, L., Alam, T., 2015. Levels and distribution of OCPs, (specially HCH, Aldrin,Dieldrin, DDT, Endosulfan) in Karhera drain and surface water of Hindon Riverand their adverse effects. Orient. J. Chem. 31 (4), 2025e2030.

Ouyang, H., He, W., Qin, N., Kong, X., Liu, W., He, Q., Wang, Q., Jiang, Y., Yang, C.,Yang, B., Xu, F., 2012. Levels, temporal-spatial variations, and sources oforganochlorine pesticides in ambient air of Lake Chaohu, China. Sci. World J.article number 504576.

Ouyang, H., He, W., Qin, N., Kong, X., Liu, W., He, Q., Yang, C., Jiang, Y., Wang, Q.,Yang, B., Xu, F., 2013. Water-gas exchange of organochlorine pesticides at LakeChaohu, a large Chinese lake. Environ. Sci. Pollut. Res. 20, 2020e2032.

Ouyang, H., Wang, Q., He, W., Qin, N., Kong, X., Liu, W., He, Q., Jiang, Y., Yang, C.,Yang, B., Xu, F., 2014. Organochlorine pesticides in the dust fall around LakeChaohu, the fifth largest lake in China. Environ. Monit. Assess. 186, 383e393.

Qiu, X.H., Zhu, T., Jing, L., Pan, H.S., Li, Q.L., Miao, G.F., Gong, J.C., 2004. Organo-chlorine pesticides in the air around the Taihu Lake. China. Environ. Sci. Tech-nol. 38 (5), 1368e1374.

Qiu, Y.W., Zhang, G., Guo, L.L., Cheng, H.R., Wang, W.X., Li, X.D., Wai, O.W.H., 2009.Current status and historical trends of organochlorine pesticides in the ecosys-tem of Deep Bay, South China. Estuar. Coast. Shelf Sci. 85 (2), 265e272.

Satcher, N., 2015. DDT and the American century: global health, environmental

politics, and the pesticide that changed the world. Southeast. Geogr. 55 (4),497e498.

Saxena, S., Siddiqui, M., 1980. Role of chlorinated hydrocarbon pesticides in abor-tions and premature labour. Toxicology 17, 323e331.

Schmitt, C.J., Zajicek, J.L., Peterman, P.H., 1990. National contaminant biomoni-toringprogram: residues of organochlorine chemicals in U.S. freshwater fish,1976e1984. Arch. Environ. Contam. Toxicol. 19, 748e781.

Snedeker, S.M., 2001. Pesticides and breast cancer risk: a review of DDT, DDE, anddieldrin. Environ. Health Perspect. 109, 35e47.

Stockholm Convention on Persistent Organic Pollutants (SCPOPs), 2005. Evaluationof the Continued Need for DDT for Disease Vector Control. http://chm.pops.int/Implementation/DDT/ProgressReport/tabid/132/ctl/Download/mid/4466/Default.aspx?id¼5&ObjID¼2206.

Tang, Z., Yang, Z., Shen, Z., Niu, J., Liao, R., 2007. Distribution and sources oforganochlorine pesticides in sediments from typical catchment of the YangtzeRiver, China. Arch. Environ. Contam. Toxicol. 53, 303e312.

UNEP, 2001. Final Act of the Plenipotentiaries on the Stockholm Convention onPersistent Organic Pollutants. United Nations Environment Program, Geneva,Switzerland.

Vu, D.T., 2015. Residue of select organochlorine pesticides (OCPs) in sediment fromVietnam's CauBay River and their impact on agricultural soil and human health.Pol. J. Environ. Stud. 24, 301e306.

Wang, Y., He, W., Qin, N., He, Q.S., Kong, X.Z., Tao, S., Xu, F.L., 2013. Distributions,sources, and ecological risks of DDT-related contaminants in water, suspendedparticulate matter, and sediments from Haihe Plain, Northern China. Environ.Monit. Assess. 185, 1777e1790 (2013).

Wetterauer, B., Ricking, M., Otte, J.C., Hallare, A.V., Rastall, A., Erdinger, L.,Schwarzbauer, J., Braunbeck, T., Hollert, H., 2012. Toxicity, dioxin-like activities,and endocrine effects of DDT metabolites-DDA, DDMU, DDMS, and DDCN. En-viron. Sci. Pollut. Res. 19, 403e415.

WHO, 1989. Environmental Health Criteria for DDT and its Derivatives, Environ-mental Aspects. World Health Organization, Geneva, 92-4r-r154283-7.

Wu, F., Xu, L., Sun, Y., Liao, H., Zhao, X., Guo, J., 2012. Exploring the relationshipbetween polycyclic aromatic hydrocarbons and sedimentary organic carbon inthree Chinese lakes. J. Soils Sediments 12, 774e783.

Wu, S.P., Tao, S., Lan, T., Li, B.G., Cao, J., Liu, W.X., Liu, Y., Meng, B.J., 2005. Hexa-chlorocyclohexane and dichlorodiphenyltrichloroethane residues in the dustfallof Tianjin, China. J. Environ. Sci. Health Part A-Toxic/Hazard. Subst. Environ. Eng.40, 1715e1730.

Xu, F.L., 1997. Exergy and structural exergy as ecological indicators for the devel-opment state of the Lake Chao ecosystem. Ecol. Model. 99, 41e49.

Xu, F.L., Tao, S., Xu, Z.R., 1999. The restoration of wetlands and macrophytes in theLake Chao: possibility and effects. Hydrobiologia 405, 169e178.

Yang, Y., Li, D., Mu, D., 2008. Levels, seasonal variations and sources of organo-chlorine pesticides in ambient air of Guangzhou, China. Atmos. Environ. 42,677e687.

Yuan, L., Qi, S., Wu, X., Wu, C., Xing, X., Gong, X., 2013. Spatial and temporal vari-ations of organochlorine pesticides (OCPs) in water and sediments from Hon-ghu Lake, China. J. Geochem. Explor. 132, 181e187.

Yun, X., Yang, Y., Liu, M., Wang, J., 2014. Distribution and ecological risk assessmentof organochlorine pesticides in surface sediments from the East Lake, China.Environ. Sci. Pollut. Res. 21, 10368e10376.

Zhang, E.L., Liu, E.F., Jones, R., Langdon, P., Yang, X.D., Shen, J., 2010. A 150-yearrecord of recent changes in human activity and eutrophication of Lake Wushanfrom the middle reach of the Yangze River, China. J. Limnol. 69, 235e241.

Zhao, Z., Zhang, L., Wu, J., Fan, C., 2009. Distribution and bioaccumulation oforganochlorine pesticides in surface sediments and benthic organisms fromTaihu Lake, China. Chemosphere 77, 1191e1198.

Zhou, R., Zhu, L., Chen, Y., 2008. Levels and source of organochlorine pesticides insurface waters of Qiantang River, China. Environ. Monit. Assess. 136, 277e287.