Contamination of organochlorine pesticides (OCPs) in the environment has been of great significance for decades, due to their physicochemical properties such as semi-volatility, persistence, hydrophobicity, bioaccumulation and adverse effects on human health and animals. During 1950s–1980s, approximately 4,460,000 t of technical HCHs and 435,200 t of DDTs were produced in the world (Cao et al. 2007; Zhang et al. 2009). The worldwide historical application of OCPs has raised severe environmental problems, although the application of OCPs has been banned in many developed countries since 1980s.

Sediment is one of the main sinks of OCPs in the environment, since OCPs are readily adsorbed onto suspended particulate matter and subsequently precipitate into river, lake and marine sediments due to their high hydrophobicity and low water solubility (Yang et al. 2005; Sun et al. 2010). However, OCPs in sediment are still harmful to benthic organisms due to their possibility to release back to water column. A full understanding of the fate of OCPs requires information on their concentration and distributions in the aquatic environment among different phases.

Taihu Lake is the third largest freshwater lake in China, with an average depth of 1.9 m. With the rapid urbanization around the Taihu region, it receives large amounts of wastes from agricultural non-point sources, municipal sewage, industrial wastewater, aquaculture, etc., causing the pollution of the lake a great concern by both scientific and public communities (Zhang et al. 2011). The contamination of OCPs in water, sediments, biota and surrounding agricultural fields and airs in Taihu region has been extensively investigated by many researchers (Qiu et al. 2004; Zhao et al. 2010), however, information on the distribution of OCPs between sediment-overlying water and sediment-porewater, and their sources is still limited. This study investigated the distribution of OCPs in sediment and corresponding overlying water and porewater samples in Taihu Lake, interpreted the partition of OCPs between sediment-porewater and sediment-overlying water matrices, and identified the possible sources of pollution. Results will be helpful for better understanding the environmental behavior and fate of OCPs in the sediment–water system.

Materials and Methods

A mixture standard solution of OCPs containing o,p′-DDE, p,p′-DDE, o,p′-DDD, p,p′-DDD, o,p′-DDT, p,p′-DDT, α-HCH, β-HCH, γ-HCH, δ-HCH, hexachlorobenzene (HCB), oxy-chlordane, trans-chlordane, cis-chlordane, heptachlor, heptachlor epoxide, aldrin, dieldrin, endrin, methoxychlor, and mirex at a concentration of 100 mg/L was purchased from Chem Service, Inc., USA. The surrogate (13C-p,p′-DDT) and the internal standard (deuterated phenanthrene, deuterated pyrene, and deuterated chrysene) were purchased from AccuStandard, Inc., USA. Organic solvents including acetone, n-hexane and methylene chloride were purchased from Fisher Scientific, USA.

Thirty surface sediment (0–10 cm) and corresponding overlying water samples were collected from Taihu Lake in May 2010, and the sampling locations were shown in Fig. 1. Surface sediments were grab sampled, and placed in glass bottles with Teflon-lined caps. Overlying water was taken 1 m beneath the surface, and stored in 4-L glass amber bottles. Samples were transported to the laboratory on ice. Porewater from each sediment was obtained by high-speed centrifugation. Both overlying water and the isolated porewater were stored in the dark at 4°C, and the remainder of sediment samples was frozen at −20°C before treatment. All samples were analyzed within one week.

Fig. 1
figure 1

The sampling sites in the study area (arrows show the inflows and outflows of the lake)

Full size image

Water samples were passed through 0.45 μm glass fibers to remove suspended particles, and concentrated by solid phase extraction (SPE). The HyperSep C18 cartridges (500 mg/6 mL, Thermo Electron Corporation) were preconditioned by 5 mL of ethyl acetate, 5 mL of methanol and 5 mL of distilled water sequentially. Filtered water samples were loaded onto cartridges at a flow rate of approximate 8 mL min−1. After the sample loading, 10 mL of distilled water were used to wash the cartridges, followed with the vacuum on for about 1 h till the cartridges were all dry. The cartridges were eluted by 4 mL of 1:1 dichloromethane/n-hexanes and 4 mL of n-hexanes. The extract was concentrated to less than 1 mL with a gentle nitrogen stream, reconstituted with 8 mL of n-hexanes, and again concentrated to less than 1 mL. Anhydrous sodium sulfate was used to remove water residue. The extract was finally concentrated to 200 μL for analysis.

Sediments were extracted with an automated Dionex ASE 300 accelerated solvent extractor (Sunnyvale, CA, USA) (Zhang et al. 2009). Surface sediments were freeze-dried and ground in a mortar to pass through a sieve with 0.5 mm openings. Five grams of each sediment sample were weighed accurately, combined with 2 g of copper powder and 2 g of diatomite, and placed into the steel cells. Samples were extracted using 1:1 methylene chloride/acetone as extracting solvent. Extraction temperature was 100°C and extraction pressure was 1,500 psi. Preheating time and static time were both set to 5 min. A total flush volume of 100 % the cell volume and a purge time of 60 s with nitrogen was used. The final extraction volume was approximately 20 mL with two extraction cycles. The extracts were concentrated to approximately 1 mL and further cleaned up with a Florisil SPE cartridge (1 g, 6 mL, Supelco, USA). After sample loading, the cartridge was eluted with 12 mL acetone/hexane (2/98, v/v). The eluate was concentrated to 1 mL under a centrifugal vacuum evaporator system (CVE3100, Eyala, Japan). The internal standards, 20 μL of deuterated phenanthrene, deuterated pyrene, and deuterated chrysene mixture standard solution (10 ng μL−1) was added prior to GC analysis.

OCPs were analyzed on a Shimadzu GCMS-QP2010 equipped with a fused silica capillary DB-5MS column (30 m, 0.25 mm i.d., 0.25 μm film thickness) using electron ionization with selective ion monitoring mode. Helium was used as the carrier gas at a flow rate of 1.0 mL min−1. The injector, transfer line, and ion source temperature were set at 250, 260 and 230°C, respectively. The GC oven temperature was programmed from 70°C (1 min) to 180°C at 20°C min−1, then to 260°C at 4°C min−1, and to 300°C at 15°C min−1 and held for 6 min. Two microliters of samples were injected in the split mode with 1:14.1. Quantification was performed by internal standard.

All analytical procedures were monitored with strict quality assurance and control measures. The instruments were calibrated daily with calibration standards. Procedural blanks were analyzed concurrently with the water and sediment samples and showed no detectable target compounds. The detection limit of OCPs ranged from 0.014 to 0.178 μg kg−1 for sediment, and from 0.228 to 0.567 ng L−1 for water. The recoveries of 13C-p,p′-DDT fell within the range of 75 %–125 % for sediment and 73 %–108 % for water, respectively. For statistical analysis, two-tailed t test was used to compare the average concentrations between samples in the present study.

Results and Discussion

The range and average concentrations of individual OCPs in sediments, porewater and overlying water matrices from Taihu Lake were summarized in Table 1. Altogether 6 OCPs, i.e. HCB, β-HCH, p,p′-DDE, p,p′-DDD, p,p′-DDT, and o,p′-DDT were detected in all samples. In sediments, the total OCPs varied from 7.84 to 32.23 ng g−1 dry weight, with a mean value of 17.59 ng g−1. The sediment concentrations and compositions of OCPs in different regions around the lake differed significantly, with the increasing trend along the direction of water flow (as shown in Fig. 1 with arrows indicating the water flow directions). The highest concentration was discovered at S28 (32.23 ng g−1), situated in the southeast part of Taihu Lake, where water met and was the grass-dominated regions. This result was consistent with that reported by Zhao et al. (2009) and Fan and Wang (2007).

Table 1 OCPs concentrations in different matrices from Taihu Lake (May 2010)
Full size table

For lipophilic organic compounds, their concentrations in sediments/soils were generally closely related to the sediment/soil organic matter contents, because sediment with high organic carbon content is more likely to adsorb hydrophobic compounds by a partition way (Xu et al. 2007, 2009a, b; Yu et al. 2009). However, no correlations were found in this study between OCPs residues and organic matter content (R2 = 0.0119) in the sediment from Taihu Lake. Zhao et al. (2009) also observed this phenomenon and explained it by three reasons: the complex process of OCPs accumulation, the higher microbial activities and biodegradation of OCPs in sediments with high organic matter contents, and the complex organic matter sources. The different OCPs application history in different regions around the lake may also partially contribute to the spatial distribution of OCPs in the sediments.

For aqueous phases, the total OCPs ranged from 136.97 to 2,185.14 ng L−1 in porewater and from 24.27 to 154.07 ng L−1 in overlying water, respectively. The highest concentrations of total OCPs for both porewater and overlying water were found at S2. S2 located in Zhushan Bay, one of the most polluted areas in Taihu Lake, where there were two severely polluted rivers (Taige Canal and Caoqiao River) carrying contaminants directly into the lake (Zhang et al. 2011). Agricultural runoff from adjacent fields entered the rivers and contributed to the total OCPs in the lake water.

To better understand the distribution of OCPs between different matrices, two partition coefficients, sediment-overlying water (KSW) and sediment-porewater (KSP) were calculated:

$$ {\text{K}}_{\text{SW}} = {\text{C}}_{\text{S}} /{\text{C}}_{\text{W}} $$
$$ {\text{K}}_{\text{SP}} = {\text{C}}_{\text{S}} /{\text{C}}_{\text{P}} $$

where CS is the sediment concentration, Cw is the water concentration, Cp is the porewater concentration. The results were shown in Table 2. It was observed that all KSW values of OCPs were higher than KSP values, showing different distribution patterns. The monitoring data showed higher OCPs levels in porewater samples than in overlying water. OCPs belong to hydrophobic organic compounds, which have very low aqueous solubilities. The presence of dissolved organic colloids in porewater may enhance the aqueous concentrations of these hydrophobic compounds beyond their solubilities significantly (Luthy et al. 1997; Nam and Alexander 1998), resulting in the lower KSP values compared to KSW.

Table 2 Distribution coefficients of OCPs between sediment-overlying water (KSW) and sediment-porewater (KSP)
Full size table

p,p′-DDT was the active ingredient in DDTs and typically made up approximately 80 % of the technical formulation, which was degraded to p,p′-DDE under aerobic conditions and to p,p′-DDD under anaerobic conditions (Zhao et al. 2009). Thus, the ratio of DDT/(DDD + DDE) can be used as an indicator of the residence time of p,p′-DDT in the environment (Qiu et al. 2004; Cheng et al. 2008). A ratio over 1 meant the historical usage and a ratio less 1 indicated relatively recent input of the parent DDTs (Jaga and Dharmani 2003). In this study, the ratios of DDT/(DDD + DDE) in sediments were site-specific, fluctuating around 1 (ranging from 0.57 to 2.28), indicating that in some areas around Taihu Lake, technical DDTs were still used. The ratios were 0.93–13.02 in porewater and 0.84-15.98 in overlying water, respectively, suggesting the potential new source of DDTs into the lake.

The characteristics of DDT compositions may be used to distinguish the DDT pollution caused by technical DDTs from that caused by dicofol (Cheng et al. 2008). In China, dicofol has been widely used in agriculture since DDT was banned in 1983 (Sun et al. 2010). Dicofol contained approximately 11.4 % of o,p′-DDT and 1.7 % of p,p′-DDT (Qiu et al. 2005). The ratio of o,p′-DDT/p,p′-DDT ranged from 0.2 to 0.3 in technical DDTs, and from 1.3 to 9.3 or higher in dicofol (Qiu et al. 2005). Results from this study showed that in the sediment samples the ratio varied from 0.04 to 0.66, with 0.18 as the median value, suggesting the important contribution of technical DDTs. The similar ratios were also found in aqueous phases, ranging from 0.22 to 0.44 in porewater and from 0.24 to 0.38 in overlying water, respectively, confirming the contribution from technical DDTs.

This work revealed the distribution of OCPs in sediment, overlying water and corresponding porewater samples from Taihu Lake, which extended our understanding of the OCPs contamination status in this shallow freshwater lake in China. The OCPs levels at each sampling location were site-specific, which was related to the pesticide application history in surrounding areas. Sources of DDTs identified by specific molecular ratios indicated that DDT residues in Taihu Lake were mainly originated from technical DDTs from both historical usage and recent input.