Passive Partitioning of Polychlorinated Biphenyls between Seawater

pathway for PCBs in the Gullmar Fjord zooplankton. The field- observed partition coefficients were also suggestive of passive partitioning, as all fie...
1 downloads 0 Views 177KB Size
Environ. Sci. Technol. 2006, 40, 6703-6708

Passive Partitioning of Polychlorinated Biphenyls between Seawater and Zooplankton, a Study Comparing Observed Field Distributions to Equilibrium Sorption Experiments A N N A S O B E K , * ,†,‡ G E R A R D C O R N E L I S S E N , †,§ PETER TISELIUS,| AND O ¨ RJAN GUSTAFSSON† Department of Applied Environmental Science (ITM), Stockholm University, SE-106 91 Stockholm, Sweden, Agroscope Reckenholz-Ta¨nikon, Research Station ART, Reckenholzstrasse 191, CH-8046 Zurich, Switzerland, Norwegian Geotechnical Institute, P.O. Box 3930, Ullevål Stadium, N-0806 Oslo, Norway, and Department of Marine Ecology, Kristineberg Marine Research Station, Go¨teborg University, SE-450 34 Fiskeba¨ckskil, Sweden

From previous studies, it remains unclear whether polychlorinated biphenyls (PCBs) are biomagnified in zooplankton or if concentrations are simply governed by passive partitioning. In this study, in the Gullmar Fjord on the Swedish west coast, field-determined lipid-normalized partition coefficients (log Klip) were compared to equilibrium partition coefficients from laboratory sorption experiments with dead and preserved zooplankton. There was no significant difference between the linear regressions of log Klip-log Kow (analysis of covariance [ANCOVA], p < 0.05) for field and laboratory-determined partition coefficients, supporting passive partitioning being the dominant uptake pathway for PCBs in the Gullmar Fjord zooplankton. The fieldobserved partition coefficients were also suggestive of passive partitioning, as all field-log Klip-log Kow regressions were significant (p < 0.05, r2 ) 0.74-0.95) and apparently linear. Further, there was generally no positive correlation between apparent biomagnification factors (BMF; concentration in zooplankton [pg/kgoc]/concentration in phytoplankton [pg/kgoc]) and trophic level (on the basis of δ15N). The in-situ organic carbon (-oc)-normalized concentrations in zooplankton (>200 µm) were not statistically different from oc-normalized concentrations in phytoplankton (0.7-50 µm), which supports the absence of significant biomagnification.

Introduction Hydrophobic organic contaminants (HOCs), such as polychlorinated biphenyls (PCBs), are persistent, toxic to biota, * Corresponding author phone: +41 44 377 7596; fax: +41 44 377 72 01; e-mail: [email protected]. † Stockholm University. ‡ Research Station ART. § Norwegian Geotechnical Institute. | Go ¨ teborg University. 10.1021/es061248c CCC: $33.50 Published on Web 09/28/2006

 2006 American Chemical Society

and bioaccumulate in food webs. In the transfer of HOCs through aquatic food webs, the first and important concentration step from water to phytoplankton organic matter is considered to be a passive partitioning process driven by the fugacity gradient between the two phases (1). Sorption of PCBs to phytoplankton has been studied both in field (2-4) and in laboratory sorption experiments (2, 5, 6). A few studies in the previous decade suggested concentrations of PCBs in phytoplankton to be kinetically limited by high phytoplankton growth rates (2) or large cell sizes (3). However, more recent studies, including larger field data sets covering different cell sizes and growth rates (4) or experimental sorption techniques avoiding known artifacts of PCBs sorbing to dissolved organic carbon (DOC) (5, 6), have generated data consistent with phytoplankton being in near-equilibrium with surrounding water. In zooplankton, several parallel processes may influence the final PCB body burden, such as exchange with water through passive partitioning, ingestion of contaminated food, and production of fecal pellets and eggs. Numerous studies have assessed concentrations of PCBs and other HOCs in zooplankton to elucidate the relative importance of passive partitioning versus biomagnification. A study of PCB concentrations in zooplankton in the Arctic marginal ice zone revealed that diet accounted for most of the observed variance in lipid-normalized PCB concentrations (7), suggesting that biomagnification may be significant. Further, one study based on biomagnification factors between phytoplankton and zooplankton observed support for biomagnification in zooplankton (8). In contrast, other studies comparing lipidnormalized concentrations of PCBs (9) or brominated diphenyl ethers (10) in phytoplankton and zooplankton observed no difference or not even a concentration decrease in zooplankton compared to phytoplankton. Additionally, several studies have used the linearity concept to diagnose the extent of equilibrium partitioning in zooplankton and concluded PCB concentrations to be dominated by passive partitioning (11-13). The linearity concept uses the thermodynamic basis of sorption, which states that a linear relationship in a log-log plot of the observed partition coefficient (K) versus the octanol-water partition coefficient (log Kow), for a series of compounds having the same kind of molecular interactions with the interacting phases, is consistent with equilibrium partitioning (14, 15) but does not provide solid evidence of equilibrium partitioning. The conclusions from the zooplankton studies mentioned above were all based on field observations. Another powerful, yet unexplored, way to distinguish among the dominating pathways for assimilation of PCBs in zooplankton is to combine laboratory sorption experiments and field observations. The objective of the present study was to contribute to a better understanding of what governs PCB concentrations in zooplankton. Equilibrium partition coefficients were determined in laboratory sorption experiments for dead, preserved tissues of two zooplankton size fractions collected from the Gullmar Fjord, Sweden. The in-situ partition coefficients of zooplankton were determined and compared with the laboratory-derived equilibrium partition coefficients. Additionally, sorption to fecal pellets was determined to further investigate the potential importance of this elimination pathway for the PCB load in zooplankton. A better understanding of the processes that determine PCB concentrations in zooplankton may contribute to refined models of HOC trophic transfer in aquatic food webs. VOL. 40, NO. 21, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6703

Materials and Methods Sampling of Zooplankton. Field sampling was conducted from the research vessel Arne Tiselius on September 1112th, 2003 at a station in the central Gullmar Fjord (58°16.4N, 11°29.25E), on the Swedish west coast. Prior to sampling, a conductivity-temperature depth (CTD) cast was done to identify the depth of the mixed layer. Three operationally defined size fractions of plankton (50-90 µm, 90-200 µm, and >200 µm) were sampled by serially placing precleaned (rinsed with hexane and acetone) nylon nets (Sefar Nitex; Bigman AB, Ha¨sselby, Sweden) with increasing pore size in a flow-through system consisting of a glass cylinder covered by aluminum foil. This way of operationally defining plankton is a common and established way of studying concentrations of HOCs at the base of aquatic food webs (2, 3, 8-12), although these bulk samples may also contain (unknown) amounts of other particles. In this study, each size fraction was sampled in triplicates. Water from a depth of 5 m was pumped with an in-situ centrifugal pump via silicon tubing into the glass cylinder. The average flow through the nets was 20 L/min, and 2000-4000 L was filtered for each sample. Each sample was placed in a preburned glass bottle and was stored at -18 °C until analysis. To collect plankton for sorption experiments and species determination, vertical tows were taken with 90- or 200-µm closing WP-2 nets from 48-m depth to the surface. The vast majority (usually >90%) of the zooplankton biomass reside in the upper 10 m, and the in-situ samples from 5 m would therefore be representative of this stratum. Samples were preserved immediately after collection in 4% buffered formalin. A subfraction of each sample was used for species analysis, and size determinations of the zooplankton community and the remaining sample was used for sorption experiments. All animals were determined to species or family and 20-50 individuals in each fraction were measured at 400×. Using length-weight regressions, the carbon content of each species was determined (16-18). The siphonophore Chelophyes appendiculata was abundant, and to estimate their biomass, we calculated the volume per individual by assuming spheroid shape (length equaling 2 × width), dry weight equaling 5% of wet weight and carbon content being 10% of dry weight (19). Zooplankton in the 50-90 µm size range was only sampled in the flow-through system, which yielded samples that were somewhat deformed and not suitable for species determinations. Sampling of Phytoplankton and the Dissolved Water Phase for PCB Analysis. Phytoplankton and other particulates (0.7-50 µm) were also collected in triplicate by sampling water from a depth of 5 m through silicone tubing using an in-situ centrifugal pump. Sampled water volume was 400600 L. Phytoplankton was collected on precombusted borosilicate filters (GF/F 293 mm, nominal pore size 0.7 µm; Whatman International, Maidstone, England), preceded by a 50-µm nylon net (precleaned; Sefar Nitex; Bigman AB, Ha¨sselby, Sweden). The GF/F filter was followed by preextracted and cleaned polyurethane foam plugs (PUF; diameter 37 mm, length 100 mm) to collect the total dissolved aqueous concentration of PCBs. The average flow rate through filter and adsorbent was 2 L/min. Collected samples were placed in aluminum envelopes and were stored in double-sealed plastic bags at -18 °C until analysis. Sampling and Analysis of Ancillary Biogeochemical and Ecological Parameters. Water for the analysis of total organic carbon (TOC) and determination of phytoplankton species was collected from 5-m depth in Niskin bottles connected to the CTD rosette. Subsamples for particulate organic carbon (POC; >0.7 µm) analysis were collected from the GF/F filters prior to analysis of PCB content. The analytical protocols for TOC, POC, and the stable isotope ratios of carbon and 6704

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 21, 2006

nitrogen (δ13C reported relative to the PDB standard and δ15N reported relative to atmospheric N2) have been described in detail (20). Water collected for phytoplankton species identification (without any size fractionation) was preserved in acetic Lugol’s solution and was stored in brown glass bottles. In the laboratory, subsamples of 10-50 mL were allowed to settle in Uthermo¨hl chambers and were analyzed at 400-1000× magnification with an inverted microscope. Published values on size and carbon content of single cells were used to calculate the carbon biomass of each species (21). Collection of Fecal Pellets. Fecal pellets for sorption experiments were collected from a laboratory culture of the small copepod Acartia clausi, which was a dominant zooplankton species at the time of sampling in the Gullmar fjord. Copepods were reared through several generations in the laboratory at 18 °C and were fed with an ad libitum mixture of the diatom Thalassiosira weissflogii and the prymnesiophyte Rhodomonas baltica. For pellet collection, a cohort of adult animals (approximately 10 000 animals) was allowed to produce fecal pellets for 24 h. The pellets were collected on a 20-µm sieve, were rinsed thoroughly with filtered (200 µm to ensure that equilibrium was reached and 103 days for zooplankton >90 µm and fecal pellets. The POM strips were extracted by horizontal shaking (180 rpm, 24 ( 0.5 °C, 48 h) with 10 mL of hexane and an internal standard solution of seven 13C-labeled PCBs (500-1000 pg/congener). Zooplankton and pellets were extracted using a hexaneacetone reflux extraction method, which has proven efficient in extracting Polycyclic aromatic hydrocarbons (PAHs) and PCBs from sediments (29). Zooplankton and pellet extracts were eluted on a modified silica column (24). Amounts of PCBs were quanitified on a Fisons 8060 gas chromatograph (GC) equipped with a PTE-5 capillary column (30 × 0.25 mm i.d., 0.25-µm film thickness; Supelco Inc, Bellefonte, PA) with a Fisons MD800 mass spectrometer (MS) operated in EI mode. Partition coefficients between POM and water (KPOM) were taken from ref 27 or were interpolated from the relationship between KPOM (27) and Kow (30). All partition coefficients were corrected for the salting-out effect using the Setschenow equation (14). Quality Assurance. The GF/F-PUF sampling technique has been evaluated in detail, demonstrating that this fieldsampling system is suitable for assessing surface seawater PCBs, with negligible risk of adsorbent breakthrough for amounts at least up to the ng level of single congeners and at these flow rates and volumes (4, 24). Blanks were analyzed in parallel with the samples to control any contamination originating from the described sampling and analytical steps. The amounts of PCB 52 in the PUF and GF/F-field blanks were 8 pg and 3 pg, respectively, and 2 pg (n ) 3) in the laboratory blanks for the zooplankton extraction method. Obtained PCB concentrations that exceeded their respective blank levels by less than a factor of 3 (signal-to-noise ratio < 3) were excluded. Average recoveries of the 13C-labeled internal standards were 76 ( 12% (mean ( stdev) for GF/F and PUF samples and 81 ( 19% for zooplankton samples. The relative standard deviation (RSD %) of the zooplankton PCB analysis was 15% (nine replicates). Estimation of Trophic Level and Biomagnification Factors. Trophic levels (TL) were calculated according to Fisk et al. (31), relative to phytoplankton (0.7-50 µm), which, as primary producer, was assumed to represent the first trophic level of the aquatic food web. An enrichment factor between trophic levels of 3.8 ‰ was used, in accordance with the value for an Arctic marine food web (32). This enrichment factor is within the range of estimates of 3-4 ‰, derived for isotopic models of marine food web structures also elsewhere (32). Biomagnification factors (BMF) were

TABLE 1. Stable Isotope Composition of Carbon and Nitrogen (Mean ( Stdev) and Estimated Trophic Level of the Sampled Plankton Size Fractions sampled size fraction 0.7-50 µm

50-90 µm

90-200 µm

>200 µm

δ15Na (‰) 4.2 ( 0.8 8.5 ( 0.6 9.3 ( 0.7 13 ( 1 δ13Ca (‰) -23 ( 0.4 -20 ( 0.8 -20 ( 0.3 -18 ( 0.9 TLb (‰) 1 2.1 2.3 3.2 a Analytical precision: (0.2 ‰ for δ15N and (0.1 ‰ for δ13C. b Trophic level. Size fraction 0.7-50 µm was assigned TL 1.

calculated as the PCB concentration in each zooplankton sample relative to the PCB concentration in phytoplankton, with PCB concentrations expressed on a common unit basis, as pg PCB/kg organic carbon (oc).

Results and Discussion Characterization of Field-Collected Plankton Samples. The sampled plankton fractions appeared to represent three trophic levels (TL) on the basis of δ15N (Table 1). There was a clear increase in δ15N from phytoplankton (0.7-50 µm; 4.2 ( 0.8 ‰ [mean ( stdev]) to microplankton (50-90 µm; 8.5 ( 0.6 ‰) and another shift between zooplankton 90-200 µm (9.3 ( 0.7 ‰) to zooplankton > 200 µm (13 ( 1 ‰). The 50-90 µm and 90-200 µm size fractions displayed less than 1 ‰ difference and belonged roughly to the same TL (Table 1). Hence, zooplankton size and trophic position were linked. Although there usually is a broad correlation between δ15N and δ13C in food webs, the unique δ13C signature of a consumer reflects more its general diet and carbon source than trophic level (33). The low δ13C signal in the smallest fraction supports this fraction being of different origin (i.e., phytoplankton primary production) than the larger heterotrophic size fractions. Accordingly, autotrophs dominated the smallest (phytoplankton) size fraction (Table 2). Zooplankton biomass in >90 µm was dominated by different copepod species (>50% of zooplankton biomass; predominantly Oithona sp.) and Siphonophores (35%). The larger zooplankton (>200 µm) was dominated by Siphonophores (68%) and had only a small contribution of the copepod Oithona sp. (1%). The individual carbon content of the dominating zooplankton was very different (Table 3), hence the two size fractions were different also in respect of number of individuals per carbon unit. PCB Concentrations. Lipid-normalized PCB concentrations, of for instance PCB 52 and PCB 153, in the three different microplankton and zooplankton size fractions were 3.3-14 ng/g lipid (median 5.2 ng/g lipid) and 10-26 ng/g lipid (median 19 ng/g lipid), respectively. These concentrations are comparable to PCB concentrations in Baltic Sea zooplankton from the same latitude (34) and are a factor >1000 above detection limits. The oc-normalized PCB 52 and PCB 153 concentrations in phytoplankton (0.7-50 µm) were 0.33-0.54 ng/g oc and 2.1-2.7 ng/g oc, respectively, while the dissolved concentrations of PCB 52 and PCB 153 were 0.57-0.81 pg/L and 0.39-0.53 pg/L. These concentrations (phytoplankton + dissolved) agree within a factor of 2 with previous observations at similar latitudes in the North Sea [47°North (35), 62°North (36)]. Evaluation of Equilibrium Partitioning in Laboratory Sorption Experiments. The laboratory-derived lipidnormalized equilibrium partition coefficients (lab-log Klip) for dead zooplankton >200 µm were not significantly different for samples being equilibrated for 55, 75, and 103 days (ttest, R ) 0.05). Hence, equilibrium was apparently reached after 55 days, in accordance with the linear lab-log Klip-log Kow regressions for all replicates and with previous sorption experiments using the POM-SPE approach (27, 28). There VOL. 40, NO. 21, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6705

TABLE 2. Phytoplankton Composition and Relative Contribution of Autotrophs, Mixotrophs, and Heterotrophs in Unfractionated Water POC/cell (pg)

species/group

biomass (µg C/L)

Diatoms

Chaetoceros contortus C. curvisetus C. spp. other

400 185 100

5.6 1.2 1.5 2.9

Dinoflagellates Ceratium furca 4300 Gyrodinium cf. flagellare 70 Protoperidinium cf. divergens 11 700 Protoperidinium spp. 10 000 other

1.7 9.2 1.2 1.0 2.5

Naked Dinoflagellates 15-20 µm other

300

3.9 0.67

20 80

3.0 4.6 1.1

20

1.1 3.2 3.2

Cryptophyce´ e Plagioselmis prolonga Teleaulax sp. other Heterotrophic Flagellates

Biflagellate sp. ciliates other Total Biomass fraction autotrophs (%) fraction heterotrophs (%) fraction mixotrophs (%)

63 29 8

TABLE 3. Average Carbon Content (µg C/Individual) and Biomass Contribution (Contr; %) for Each Species or Family of Zooplankton in Two Size Fractions size fraction >90 µm species/group

>200 µm

POC/individual

POC/individual

(µg)

Contr (%)

(µg)

Contr (%) 8 6 1

Acartia clausi Paracalanus parvus Microcalanus pusillus Pseudocalanus sp. Centropages hamatus Oithona sp. Microsetella norvegica Copepod nauplii

Copepods 2.7 0.74 1.3 3.6 1.5 0.18 0.31 0.13

4 11 1 3 1 22 1 11

1.7 1.6 1.3 3.8 3.0 0.34

3 1

2.9

1

Oikopleura dioica

Tunicates 1.2

7

2.8

1

Chaetognaths 1.5

5

1.7

5

Siphonophores Chelophyes appendiculata 4.5 35

7.6

68

1.0

6

Sagitta setosa

Cladocerans

Penilia avirostris

was a good consistency within the lab-log Klip data, with a mean relative standard deviation for all congeners in both size fractions of 1.0-1.6%. The slope was 1.0 ((0.1) for both zooplankton >90 µm and >200 µm. There was no significant difference for any congener (t-test, R ) 0.05) in lab-log Klip between the two size fractions. Hence, the difference in zooplankton species composition (Table 3) was not reflected in sorbent quality of the organic matter. 6706

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 21, 2006

FIGURE 1. Linear regressions, with 95% confidence intervals, between (i) field-log Klip and (ii) laboratory-log Klip and log Kow, for zooplankton >200 µm. Log Kow values are from ref 30. Comparison of Field and Laboratory-Derived Partition Coefficients. The equilibrium partition coefficients from the sorption experiments provide unique and powerful reference data for evaluating the state of equilibrium observed in field samples. Figure 1 shows the field-lipid-normalized partition coefficient (field-log Klip) and the laboratory-measured equilibrium partition coefficient (lab-log Klip) for zooplankton >200 µm as a function of log Kow. The slopes and intercepts of the log Klip-log Kow regressions for field and laboratory data, respectively, were not significantly different (analysis of covariance [ANCOVA], R ) 0.05). This comparison of equilibrium partition coefficients, produced under controlled laboratory conditions, with field-determined partition coefficients, strongly suggests that passive partitioning is the dominant uptake pathway for PCBs in the Gullmar Fjord zooplankton. The higher field-log Klip for some of the PCBs was not related to their log Kow, suggesting that biomagnification does not explain this observation. The preliminary results from the single sorption experiment of copepod fecal pellets (one replicate) showed that the sorption characteristics of these particular pellets are similar to those of zooplankton lipid. The linear regression between lab-log Klip of the fecal pellets and log Kow had a slope of 1.1 (p < 0.05, r2 ) 0.87). This surprisingly high affinity of PCBs for fecal pellets may partly explain the apparent absence of biomagnification. A similar sorption behavior of zooplankton lipids and fecal pellets toward PCBs could imply that to create a net flux of PCBs from the gut to the zooplankton body (37), the ingested particles (i.e., phytoplankton) need to have a higher PCB concentration than zooplankton lipid (to cause a higher fugacity in the gut than in the body). To our knowledge, this is the first study using this approach of comparing laboratory-determined equilibrium partition coefficients with field-determined partition coefficients to establish the extent of equilibrium partitioning in zooplankton. More studies comparing laboratory and field data are needed to better constrain any variability in zooplanktonwater partitioning of PCBs. Evaluation of Field-Determined Partition Coefficients. The field-log K-log Kow regressions for each plankton size fraction were all significant and apparently linear (Figure 2). This is consistent with equilibrium or near-equilibrium partitioning between water and both phytoplankton and zooplankton (14, 15), thus in line with the comparison of lab and field data in Figure 1. However, considering the amount of scatter often present in environmental data, it may be questioned how sensitive log-log plots alone are in diag-

FIGURE 3. Comparison of the field-observed organic-carbon normalized partition coefficients for four different sizes of plankton (three trophic levels, TL). Log Kow values are from ref 30.

FIGURE 2. Field-observed partition coefficients for PCBs between water and four different sizes of plankton: (A) phytoplankton, (B) microplankton, and (C-D) zooplankton. The plankton in A were normalized to organic carbon (field-log Koc) while B-D were normalized to lipids (field-log Klip). Log Kow values are from ref 30. nosing biomagnification in zooplankton. The slopes varied between 1.06 and 1.32 ((confidence interval). The slightly higher than one slopes suggest that the plankton organic matter had higher solubility capacity compared to octanol (14, 15). Slopes of linear log K-log Kow relationships around one have previously been reported for both phytoplankton and zooplankton in field studies (4, 11-13). The partition coefficients for all size fractions were also normalized to organic carbon to enable a direct comparison of the PCB levels in the different size fractions (Figure 3). T-tests between the intercepts of the regressions of the four size classes (three TLs) showed no significant difference

between any of the groups (R ) 0.05). This further supports the absence of biomagnification in these zooplankton. The slopes of the 50-90 µm and 90-200 µm plankton (both TL 2) were significantly higher than the slope of the phytoplankton (R ) 0.05), which simply suggests that the organic matter of these zooplankton to dissolve PCBs is better than the organic matter of phytoplankton (14, 15). There was no significant difference between the slopes or intercepts of zooplankton >200 µm (TL 3) and phytoplankton (TL 1). The organic matter of Siphonophores, which constituted 68% of zooplankton >200 µm, may thus sorb PCBs to a similar extent as phytoplankton and to some less extent compared to other zooplankton species such as copepods. The many different physiological characteristics of zooplankton species may potentially influence both sorption properties and the possibility of biomagnifying organic pollutants. Thus, although the results from this study support equilibrium partitioning to be the dominating uptake pathway for PCBs in the investigated Gullmar Fjord zooplankton, these results may not apply to all ecosystems. Therefore, this study also highlights the needs for more related studies aiming at understanding the impact of zooplankton physiology on the organic pollutant load in zooplankton. The δ15N (or TL) together with the apparent biomagnification factor (BMF; concentration in zooplankton [pg/kgoc]/ concentration in phytoplankton [pg/kgoc]) form a useful tool in assessing trophic transfer of PCBs and other HOCs in aquatic food webs. In this study, the BMF had no correlation with TL for 11 out of the 13 congeners (p ) 0.06-0.96 and r2 ) 0.00-0.38), further supporting the absence of apparent biomagnification in the sampled zooplankton. A significant correlation with a positive slope between BMF and TL was only detected for PCB 18 (slope ) 2.2, r2 ) 0.80, R ) 0.05) and PCB 28 (slope ) 1.6, r2 ) 0.60, R ) 0.05), that is, the least hydrophobic PCBs tested, for which we cannot provide any direct explanation. Biomagnification of a substance in zooplankton will occur if the total uptake (through diet and water) exceeds elimination (through feces and water). Elimination to water is more efficient for less hydrophobic PCB congeners, implying that biomagnification would be expected to increase with hydrophobicity (14, 37, 38). It therefore seems unlikely that only the least hydrophobic (of the investigated) PCBs would biomagnify. On the other hand, it may be argued that the less hydrophobic congeners are more easily taken up through the zooplankton gut and therefore to a certain extent are more easily biomagnified than more slowly diffusing congeners. The present study supports the hypothesis of PCB concentrations in zooplankton being dominated by nearVOL. 40, NO. 21, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6707

equilibrium partitioning with surrounding water, mainly on the basis of comparisons between laboratory sorption experiments and field data. These results may have implications for future aquatic food-web modeling, in which the complexity may be decreased by parametrizing HOC uptake in zooplankton by an equilibrium partition model instead of kinetic parametrizations of biodilution or biomagnification.

Acknowledgments Kerstin Grunder and Hanna Gustavsson are gratefully acknowledged for help with PCB analyses and Ralf Dahlqvist for help during field work. This study was financially supported by the Swedish Research Council (FORMAS 21.0/ 2002-0629). Stockholm Marine Research Center (SMF) is acknowledged for a Ph.D stipend to A.S. and the Swedish Research Council for a senior research fellowship to O ¨ .G. (VR Grant 629-2002-2309).

Literature Cited (1) Mackay, D. Correlation of bioconcentration factors. Environ. Sci. Technol. 1982, 16, 274-278. (2) Swackhamer, D. L.; Skoglund, R. S. Bioaccumulation of PCBs by algae: kinetics versus equilibrium. Environ. Toxicol. Chem. 1993, 12, 831-838. (3) Axelman, J.; Broman, D.; Na¨f, C. Field measurements of PCB partitioning between water and planktonic organisms: influence of growth, particle size and solute-solvent interactions. Environ. Sci. Technol. 1997, 31, 665-669. (4) Sobek, A.; Gustafsson, O ¨ .; Hajdu, S.; Larsson, U. On the particlewater partitioning of PCBs in the photic zone: a 25-month study in the open Baltic Sea. Environ. Sci. Technol. 2004, 38, 13751382. (5) Koelmans, A. A.; Van der Woude, H.; Hattink, J.; Niesten, D. J. M. Long-term bioconcentration kinetics of hydrophobic chemicals in Selenastrum capricornutum and Microsystis aeruginosa. Environ. Toxicol. Chem. 1999, 18, 1164-1172. (6) Gerofke, A.; Ko¨mp, P.; McLachlan, M. S. Bioconcentration of persistent organic pollutants in four species of marine phytoplankton. Environ. Toxicol. Chem. 2005, 24, 114-123. (7) Borgå, K.; Gabrielsen, G. W.; Skaare, J. U. Differences in contamination load between pelagic and sympagic invertebrates in the Arctic marginal ice zone: influence of habitat, diet and geography. Mar. Ecol. Prog. Ser. 2002, 235, 157-169. (8) Trowbridge, A. G.; Swackhamer, D. L. Preferential biomagnification of aryl hydrocarbon hydroxylase-inducing polychlorinated congeners in the Lake Michigan, USA, lower food web. Environ. Toxicol. Chem. 2002, 21, 334-341. (9) Larsson, P.; Okla, L.; Cronberg, G. Turnover of polychlorinated biphenyls in an oligotrophic and an eutrophic lake in relation to internal lake processes and atmospheric fallout. Can. J. Fish. Aquat. Sci. 1998, 55, 1926-1937. (10) Chiuchiolo, A. L.; Dickhut, R. M.; Cochran, M. A.; Ducklow, H. W. Persistent organic pollutants at the base of the Antarctic marine food web. Environ. Sci. Technol. 2004, 38, 3551-3557. (11) Hargrave, B. T.; Phillips, G. A.; Vass, W. P.; Bruecker, P.; Welch, H. E.; Siferd, T. D. Seasonality in bioaccumulation of organochlorines in lower trophic level Arctic marine biota. Environ. Sci. Technol. 2000, 34, 980-987. (12) Berglund, O.; Larsson, P.; Ewald, G.; Okla, L. Bioaccumulation and differential partitioning of polychlorinated biphenyls in freshwater, planktonic food webs. Can. J. Fish. Aquat. Sci. 2000, 57, 1160-1167. (13) Sobek, A.; Reigstad, M.; Gustafsson, O ¨ . Partitioning of polychlorinated biphenyls between Arctic seawater and sizefractionated zooplankton. Environ. Toxicol. Chem. 2006, 25, 1720-1728. (14) Schwarzenbach, R. P.; Gschwend, P. M.; Imboden, D. M. Environmental organic chemistry, 2nd ed.; John Wiley & Sons: Hoboken, NJ, 2003. (15) Goss, K.-U.; Schwarzenbach, R. P. Linear free energy relationships used to evaluate equilibrium partitioning of organic compounds. Environ. Sci. Technol. 2001, 35, 1-9. (16) Mauchline, J.; Blaxter, J. H. S.; Southward, A. J.; Tyler, P. A. The biology of Calanoid copepods. Adv. Mar. Biol. 1998, 33, 1-710. (17) Paffenho¨fer, G. A. On the biology of Appendicularia of the southern North Sea. In 10th European Symposium of Marine Biology; Persoone, G., Jaspers, E., Eds.; Universal Press: Wetteren, Belgium, 1976. 6708

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 21, 2006

(18) Conway, D. V. P.; Robins, D. B. Collection and chemical analysis of chaetognaths and changes due to preservation. In The biology of chaetognaths; Bone, Q., Kapp, H., Pierrot-Bults, A. C., Eds.; Oxford University Press: New York, 1991. (19) Parsons, T. R.; Maita, Y.; Lalli, C. M. A manual of chemical and biological methods for seawater analysis; Pergamon: Oxford, U.K., 1992. (20) Gustafsson, O ¨ .; Nilsson, N.; Bucheli, T. D. Dynamic colloidwater partitioning of pyrene through a coastal Baltic spring bloom. Environ. Sci. Technol. 2001, 35, 4001-4006. (21) Edler, L. Phytoplankton and the primary production in the Sound. PhD Thesis, Department of Marine Ecology, Gothenburg University, 1977. (22) Bandh, C.; Ishaq, R.; Broman, D.; Na¨f, C.; Ronquist-Nii, Y.; Zebu ¨ hr, Y. Separation for subsequent analysis of PCBs, PCDD/ Fs, and PAHs according to aromaticity and planarity using a two dimensional HPLC system. Environ. Sci. Technol. 1996, 30, 214-219. (23) Zebu ¨ hr, Y.; Na¨f, C.; Bandh, C.; Broman, D.; Ishaq, R.; Pettersen, H. An automated HPLC separation method with 2 coupled columns for the analysis of PCDD/Fs, PCBs and PACs. Chemosphere 1993, 27, 1211-1219. (24) Sobek, A.; Gustafsson, O ¨ .; Axelman, J. An evaluation of the importance of the sampling step to the total analytical variance - a four system field-based sampling intercomparison study for hydrophobic organic contaminants in the surface waters of the Baltic Sea. Int. J. Environ. Anal. Chem. 2003, 83, 177-187. (25) Gschwend, P. M.; Wu, S.-C. On the constancy of sedimentwater partition coefficients of hydrophobic organic pollutants. Environ. Sci. Technol. 1985, 19, 90-96. (26) Burkhard, L. P. Estimating dissolved organic carbon partition coefficients for nonionic organic chemicals. Environ. Sci. Technol. 2000, 34, 4663-4668. (27) Jonker, M. T. O.; Koelmans, A. A. Polyoxymethylene solid phase extraction as a partitioning method for hydrophobic organic chemicals in sediment and soot. Environ. Sci. Technol. 2001, 35, 3742-3748. (28) Cornelissen, G.; Gustafsson, O ¨ . Sorption of phenanthrene to environmental black carbon in sediment with and without organic matter and native sorbates. Environ. Sci. Technol. 2004, 38, 148-155. (29) European Commission. BCR information on reference materials, the certification of the contents and mass fractions of chlorobiphenyls and PAHs in industrial soil; Rep. No. EUR 16215 EN; Ispra, Italy, 1991. (30) Hawker, D. W.; Connell, D. W. Octanol-water partition coefficients of polychlorinated biphenyl congeners. Environ. Sci. Technol. 1988, 22, 382-387. (31) Fisk, A. T.; Hobson, K. A.; Norstrom, R. J. Influence of chemical and biological factors on trophic transfer of persistent organic pollutants in the Northwater Polynya Marine food web. Environ. Sci. Technol. 2001, 35, 732-738. (32) Hobson, K. A.; Welch, H. E. Determination of trophic relationships within a high Arctic marine food web using delta 13C and delta 15N analysis. Mar. Ecol. Prog. Ser. 1992, 84, 9-18. (33) Kiriluk, R. M.; Servos, M. R.; Whittle, M. D.; Cabana, G.; Rasmussen, J. B. Using ratios of stable nitrogen and carbon isotopes to characterize the biomagnification of DDE, mirex, and PCB in a Lake Ontario pelagic food web. Can. J. Fish. Aquat. Sci. 1995, 52, 2660-2674. (34) Burreau, S. On the uptake and biomagnification of PCBs and PBDEs in fish and aquatic food chains. PhD Thesis, Department of zoology, Stockholm University, 2001. (35) Schulz, D. E.; Petrick, G.; Duinker, J. C. Chlorinated biphenyls in North atlantic surface and deep water. Mar. Pollut. Bull. 1988, 19, 526-531. (36) Sobek, A.; Gustafsson, O ¨ . Latitudinal fractionation of polychlorinated biphenyls in surface seawater along a 62° N-89° N transect from the southern Norwegian Sea to the North Pole area. Environ. Sci. Technol. 2004, 38, 2746-2751. (37) Gobas, F. A. P. C.; Wilcockson, J. B.; Russel, R. W.; Haffner, D. G. Mechanism of biomagnification in fish under laboratory and field conditions. Environ. Sci. Technol. 1999, 33, 133-141. (38) Kelly, B. C.; Gobas, F. A. P. C.; McLachlan, M. S. Intestinal absorption and biomagnification of organic contaminants in fish, wildlife, and humans. Environ. Toxicol. Chem. 2004, 23, 2324-2336.

Received for review May 24, 2006. Revised manuscript received August 14, 2006. Accepted August 24, 2006. ES061248C