SEARCH

SEARCH BY CITATION

Keywords:

  • Bioconcentration;
  • Bioaccumulation;
  • Biomagnification;
  • Trophic magnification;
  • Fugacity

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BIOACCUMULATION TERMINOLOGY AND ENDPOINT METRICS
  5. COMPARISON APPROACH
  6. EVALUATION OF BIOACCUMULATION ENDPOINT DATA
  7. METHODS
  8. RESULTS AND DISCUSSION
  9. CONCLUSIONS
  10. SUPPLEMENTAL DATA
  11. EDITOR'S NOTE
  12. Acknowledgements
  13. REFERENCES
  14. Supporting Information

An approach for comparing laboratory and field measures of bioaccumulation is presented to facilitate the interpretation of different sources of bioaccumulation data. Differences in numerical scales and units are eliminated by converting the data to dimensionless fugacity (or concentration-normalized) ratios. The approach expresses bioaccumulation metrics in terms of the equilibrium status of the chemical, with respect to a reference phase. When the fugacity ratios of the bioaccumulation metrics are plotted, the degree of variability within and across metrics is easily visualized for a given chemical because their numerical scales are the same for all endpoints. Fugacity ratios greater than 1 indicate an increase in chemical thermodynamic activity in organisms with respect to a reference phase (e.g., biomagnification). Fugacity ratios less than 1 indicate a decrease in chemical thermodynamic activity in organisms with respect to a reference phase (e.g., biodilution). This method provides a holistic, weight-of-evidence approach for assessing the biomagnification potential of individual chemicals because bioconcentration factors, bioaccumulation factors, biota–sediment accumulation factors, biomagnification factors, biota–suspended solids accumulation factors, and trophic magnification factors can be included in the evaluation. The approach is illustrated using a total 2393 measured data points from 171 reports, for 15 nonionic organic chemicals that were selected based on data availability, a range of physicochemical partitioning properties, and biotransformation rates. Laboratory and field fugacity ratios derived from the various bioaccumulation metrics were generally consistent in categorizing substances with respect to either an increased or decreased thermodynamic status in biota, i.e., biomagnification or biodilution, respectively. The proposed comparative bioaccumulation endpoint assessment method could therefore be considered for decision making in a chemicals management context. Integr Environ Assess Manag 2012;8:17–31. © 2011 SETAC


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BIOACCUMULATION TERMINOLOGY AND ENDPOINT METRICS
  5. COMPARISON APPROACH
  6. EVALUATION OF BIOACCUMULATION ENDPOINT DATA
  7. METHODS
  8. RESULTS AND DISCUSSION
  9. CONCLUSIONS
  10. SUPPLEMENTAL DATA
  11. EDITOR'S NOTE
  12. Acknowledgements
  13. REFERENCES
  14. Supporting Information

National and international screening and assessment of substances (e.g., Stockholm Convention, UNECE-LRTAP, REACH) rely on knowledge of the persistent, bioaccumulative, and toxic (PBT) properties of the chemical and/or chemicals under evaluation. For assessing a chemical's bioaccumulative (B) potential, the n-octanol–water partition coefficient (KOW), bioconcentration factor (BCF), and bioaccumulation factor (BAF) are the metrics traditionally used by regulatory agencies. In general, threshold values of 1000 to 5000 and 100 000 are used for the BCF-BAF and KOW, respectively (Gobas et al. 2009). Additional bioaccumulation metrics are available from field, mesocosm, and laboratory studies, and they include the biota–sediment accumulation factor (BSAF), biomagnification factor (BMF), biota–suspended solids accumulation factor (BSSAF), and trophic magnification factor (TMF). These additional metrics provide valuable insights for characterizing the bioaccumulation potential of nonionic organic chemicals. One of the current difficulties in comparing BCF and BAF data to other bioaccumulation metrics is the difference in numerical scale and reference media to which chemical concentrations in organisms are compared. For example, BCFs and BAFs have values on the order of 100 to 108 (Arnot and Gobas 2006) whereas BMFs, BSAFs, BSSAFs, and TMFs have values on the order of 10−4 to 102. BCFs and BAFs express ratios of chemical concentrations in biota to water, whereas BSAFs and BSSAFs represent ratios of lipid-normalized chemical concentrations in biota to organic carbon-normalized concentrations in sediment or suspended solids. BMFs and TMFs reflect ratios of chemical concentrations in predator–prey relationships.

The objective of the present study is to propose and evaluate a method for direct comparison of all bioaccumulation metrics for assessing the biomagnification potential of a substance. This method allows for more holistic assessments of biomagnification potential by using all available data while avoiding potential biases from evaluations that rely solely on BCF and/or KOW (Arnot and Gobas 2006; Kelly et al. 2007; Gobas et al. 2009). This weight-of-evidence approach, in essence, evaluates the extent to which the available bioaccumulation endpoint data support the hypothesis that a chemical will or will not biomagnify.

Bioaccumulation is the process that causes an increased chemical concentration in an organism compared to that in its ambient environment through all exposure routes. Biomagnification only quantifies the changes in chemical residues between an organism and its prey items due to dietary absorption processes. Bioconcentration and biotransformation are other processes influencing residues in an organism, and the result of biomagnification, bioconcentration, and biotransformation processes is bioaccumulation. The approach proposed and evaluated here only examines biomagnification; 1 of 3 processes involved in the overall bioaccumulation of a chemical in an organism.

In this report, the comparison method is proposed and evaluated using 15 chemicals. The bioaccumulation metrics and data quality issues associated with the metrics used in the proposed approach are briefly reviewed. Merits and limitations of this approach in the broader context of bioaccumulation assessment are then discussed.

BIOACCUMULATION TERMINOLOGY AND ENDPOINT METRICS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BIOACCUMULATION TERMINOLOGY AND ENDPOINT METRICS
  5. COMPARISON APPROACH
  6. EVALUATION OF BIOACCUMULATION ENDPOINT DATA
  7. METHODS
  8. RESULTS AND DISCUSSION
  9. CONCLUSIONS
  10. SUPPLEMENTAL DATA
  11. EDITOR'S NOTE
  12. Acknowledgements
  13. REFERENCES
  14. Supporting Information

In this article, bioaccumulation is defined as the net accumulation of a chemical by an organism as a result of uptake from all routes of exposure (e.g., water, sediment, and food). Bioconcentration is defined as the net accumulation of a chemical by an organism as a result of uptake directly from its surrounding physical environment only through respiratory or dermal surfaces. Biomagnification is defined as the increase in concentration of a chemical in the tissue of organisms along a series of predator-prey associations, primarily through the mechanism of dietary accumulation (USEPA 2009a). In the laboratory, 3 bioaccumulation metrics can be measured: BCF, BSAF, and BMF. In the field, 5 bioaccumulation metrics can be measured: BAF, BSAF, BSSAF, BMF, and TMF. It is typically assumed that steady-state is approximated when determining the aforementioned bioaccumulation endpoints; however, this condition can only be approximate in most situations and may not occur in some situations, particularly for field measurements and for very persistent and very high KOW chemicals. Bioaccumulation measurements for very hydrophobic, persistent chemicals that have not approached steady-state are underestimates of the true values.

The BCF is the ratio of the steady-state chemical concentrations in a water-respiring organism to the water (steady-state method) or the ratio of the uptake rate to depuration rate constants; i.e., k1/k2 (kinetic method) as determined in controlled laboratory experiments where the test organisms are exposed to chemical in the water (but not the diet). The BCF can be determined from wet weight tissue or lipid weight concentrations in the organism and dissolved or total (bulk) chemical concentrations in the water. The BCFWD/LW (L-water/kg-lipid) can be calculated from the lipid-normalized chemical concentration in biota (CB-LW; µg-chemical/kg-lipid) and the truly (or freely) dissolved chemical concentration in the water (CWD; µg-dissolved chemical/L-water). The BCFWT/WW (L-water/kg-wet weight) can be calculated from the wet weight chemical concentration in an organism (CB-WW; µg-chemical/kg-wet weight) and the total (bulk) chemical concentration in water (CWT; µg-total chemical/L-water).

The BAF is the ratio of the steady-state concentration in an organism to the water where it resides including exposure to chemical from its surrounding environment and its diet. The BAF can be determined from wet weight tissue or lipid weight concentrations in the organism and dissolved or total (bulk) chemical concentrations in the water. The BAFWD/LW (L-water/kg-lipid) can be calculated from the lipid-normalized chemical concentration in biota (CB-LW; µg-chemical/kg-lipid) and the truly (or freely) dissolved chemical concentration in the water (CWD; µg-dissolved chemical/L-water). In this study, these BAFs are referred to as either BAFM-WD/LW or BAFF-WD/LW corresponding with mesocosm (M) or field (F) studies, respectively. The BAFWT/WW (L-water/kg-wet weight) is calculated as CB-WW/CWT and can be referred to as either BAFM-WW/WT or BAFF-WW/WT corresponding with mesocosm (M) or field (F) studies, respectively.

The BSAFOC/LW (kg-organic carbon/kg-lipid weight) is the ratio of the steady-state lipid normalized chemical concentration in the organism (CB-LW; µg-chemical/kg-lipid) to the organic carbon-normalized chemical concentration in the bottom sediments (CS-OC; µg-chemical/kg-organic carbon). The BSAF can be calculated from laboratory (BSAFL), mesocosm (BSAFM), and field (BSAFF) measurements.

The BSSAFOC/LW (kg-organic carbon/kg-lipid weight) is the ratio of the steady-state lipid-normalized chemical concentration in the organism (CB-LW; µg-chemical/kg-lipid) to the organic carbon-normalized chemical concentration in the water column suspended sediments (CSS-OC; µg-chemical/kg-organic carbon). The BSSAF can be calculated from mesocosm (BSSAFM) and field (BSSAFF) measurements.

The BMFLW/LW (kg-lipid weight/kg-lipid weight) is the ratio of the steady-state lipid normalized chemical concentration in an organism (CB-LW; µg-chemical/kg-lipid) and the lipid normalized chemical concentration in its diet (CD-LW; µg-chemical/kg-lipid). The BMF can be determined from a controlled laboratory experiment (BMFL), mesocosm study (BMFM), or field site (BMFF).

The TMF is an average factor across an entire food web by which the chemical concentrations in biota change per trophic level within a food web. For nonionic organic chemicals, chemical concentrations are lipid normalized whereas for other chemical classes, chemical concentrations are expressed on a wet weight basis (Fisk et al. 2001; Borgå et al. 2012). The TMF is similar in concept to the BMF; however, instead of a “single predator–prey items relationship” (i.e., the BMF), the TMF is based on all of the organisms sampled in the food web. The TMF is essentially an “average BMF” value that considers all of the trophic interactions and organisms sampled in the food web.

As first presented by Broman et al. (1992) and later refined by Fisk et al. (2001), TMFs are calculated by regressing the measured chemical concentrations in biota against their relative trophic positions (or trophic levels) to quantify the rate of trophic transfer of the chemicals. Because of differences in biomass and contaminant transfer efficiency, contaminant concentrations often increase exponentially through the food web, and thus, regressions are most often carried out on the logarithm of the lipid normalized concentrations in the organisms (CB-LW; µg-chemical/kg-lipid) against their relative trophic positions. The relative trophic positions in the food web can be estimated by the application of a trophic positioning model through gut content analysis or by using stable 15N-to-14N isotope ratio analysis (Gobas et al. 2009). The TMF is then calculated from the slope (m) of the linear regression. If the concentration data were transformed using base-10 logarithms then TMF = 10m. If the concentration data were transformed using natural logarithms then TMF = em. With stable 15N-to-14N isotope ratios, the relative trophic positions in the food web are placed on a continuum, e.g., an organism with a mixed diet might reside at a relative trophic position of 3.62. The stable isotope signal arises from the slight preference for retention of the heavier isotope over the lighter isotope (e.g., enrichment factors of approximately 3.4‰ per trophic position) by an organism from its diet (Fisk et al. 2001).

Table 1 summarizes the bioaccumulation assessment endpoint metrics of interest in this study, their units, and concentration-based calculations. Henceforth in this article, the units for the individual metrics are those shown in Table 1, and for readability, these units (shown in the subscripts) will not be included with each metric except when needed. Furthermore, the subscripts L, M, and F on a bioaccumulation metric mean that the value came from a controlled laboratory experiment (BCFL, BMFL, and BSAFL), mesocosm study (BAFM, BSAFM, BSSAFM, and BMFM), or field site (BAFF, BSAFF, BSSAFF, and BMFF), respectively.

Table 1. Bioaccumulation assessment endpoints, metrics, and standard calculations
Bioaccumulation endpointBioaccumulation metric (units)Calculation
  1. BAF = bioaccumulation factor; BCF = bioconcentration factor; BMF = biomagnification factor; BSAF = biota–sediment accumulation factor; BSSAF = biota–suspended sediment accumulation factor; TMF = trophic magnification factor.

BCFBCFWD/LW (L-water/kg-lipid)CB-LW/CWD
BAFBAFWD/LW (L-water/kg-lipid)CB-LW/CWD
BSAFBSAFOC/LW (kg-organic carbon/kg-lipid)CB-LW/CS-OC
BSSAFBSSAFOC/LW (kg-organic carbon/kg-lipid)CB-LW/CSS-OC
BMFBMFLW/LW (kg-lipid/kg-lipid)CB-LW/CD-LW
TMFTMF (unitless)em or 10m

For bioaccumulation metrics measured in the field, these metrics represent a snapshot of the conditions in the field at the time of the sampling. As a result, these metrics can vary over a year and across years due to changes in the ecosystem conditions, e.g., prey item availability, changes in chemical loadings, storm events, temperatures, and season migrations.

COMPARISON APPROACH

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BIOACCUMULATION TERMINOLOGY AND ENDPOINT METRICS
  5. COMPARISON APPROACH
  6. EVALUATION OF BIOACCUMULATION ENDPOINT DATA
  7. METHODS
  8. RESULTS AND DISCUSSION
  9. CONCLUSIONS
  10. SUPPLEMENTAL DATA
  11. EDITOR'S NOTE
  12. Acknowledgements
  13. REFERENCES
  14. Supporting Information

The approach in the present study for comparing bioaccumulation data is to express the assessment metrics in terms of the thermodynamic status of the chemical with respect to a reference phase using fugacity (or normalized concentration) ratios. Fugacity is an equilibrium criterion and can be used to assess the relative thermodynamic status (chemical activity or chemical potential) of a system comprised of multiple phases or compartments (Mackay 2001). At thermodynamic equilibrium, the chemical fugacities in the different phases are equal. A fugacity ratio between an organism and a reference phase (e.g., water, sediment, or diet) that is greater than 1 indicates that the chemical in the organism is at a higher fugacity (or chemical activity) than the reference phase. For example, Connolly and Pederson (1988) used the fugacity approach to show that certain polychlorinated biphenyl congener (PCB) fugacities increased over multiple trophic levels in an aquatic food web and therefore these PCBs were subject to biomagnification.

The fugacity of a chemical in a specific medium or phase can be calculated from the measured chemical concentration as:

  • equation image(1)

where f is the fugacity (Pa), C is concentration (mol/m3), and Z is the fugacity capacity (mol/(m3 · Pa)). The fugacity capacity expresses the capacity or affinity of a phase for a chemical and depends on the phase composition, temperature, pressure, and the physicochemical properties of the chemical. Table 2 summarizes the Z value calculations for the media of interest in this analysis. The fugacity capacity for water (ZW) is calculated first from the Henry's law constant (H; [m3 · Pa/mol]) of the chemical. As outlined below, all other Z values are calculated from ZW using partition coefficients. Details of the fugacity concept can be found elsewhere (Mackay 2001).

Table 2. Fugacity capacity (Z; [mol/(m3 · Pa)]) calculations
MediaZ valueRelevant parameters
WaterZW = 1/HH = Henry's law constant (m3 · Pa/mol)
Biota (or diet)ZB = KBW × ZWKBW = biota–water partition coefficient (m3/m3

) (see Eqn. 2

)
SedimentZS = KSW × ZWKSW = sediment–water partition coefficient (m3/m3

) (see Eqn. 3

)
Suspended sedimentZSS = KSSW × ZWKSSW = suspended sediment–water partition coefficient (m3/m3

) (see Eqn. 3

)

The fugacity capacity of an organism (biota) is calculated from ZW and the biota–water partition coefficient (KBW; m3-water/m3-biota). KBW can be calculated as:

  • equation image(2)

where mL is the mass fraction of the lipid (kg-lipid/kg-biota) in the analyzed tissue (or whole organism), KLW is the lipid–water partition coefficient (m3-water/m3-lipid), ρB is the density of biota (kg-biota/m3-biota), and ρL is the density of lipid (kg-lipid/m3-lipid). In this analysis, both biota and lipid phases are assumed to have equivalent densities of 1000 kg/m3 (Mackay et al. 1996), and thus, the density values in the equation cancel out. It is also assumed that n-octanol and lipid are equivalent with respect to their capacity to dissolve (store) nonionic organic chemicals, i.e., KLW = KOW and Zlipid = Zoctanol.

The fugacity capacities of solid phases such as sediments and suspended sediments are calculated from ZW and the sediment–water partition coefficient (KSW; m3-water/m3-sediment) and the suspended sediment–water partition coefficient (KSSW; m3-water/m3-suspended sediment), respectively. For example, KSW can be calculated as:

  • equation image(3)

where mOC is the mass fraction of the organic carbon (kg-organic carbon/kg-sediment) in the sediment (or suspended sediment), KOC is the organic carbon–water partition coefficient (L-water/kg-organic carbon), ρS is the density of the sediment (2400 kg-sediment/m3-sediment), and 1000 corresponds to a unit conversion (L-water/m3-water) (Mackay et al. 1996). To calculate KSSW using Equation 3, the density for suspended sediment (ρSS; 1500 kg-suspended sediment/m3-suspended sediment) is used instead of the density for bottom sediment (Mackay et al. 1996). The organic carbon–water partition coefficient KOC is estimated as 0.35 × KOW, where 0.35 is a constant with units of liters per kilogram (Seth et al. 1999). This constitutes a relatively simple, but often applied, approach for relating the chemical solubility in organic carbon to the chemical solubility in octanol. Measured data for neutral organic chemicals suggest that the median value of 0.35 actually ranges from approximately 0.14 to 0.89 (Seth et al. 1999).

Table 3 summarizes the concentration and fugacity calculations for deriving comparable and dimensionless bioaccumulation values from the various bioaccumulation assessment endpoints. The concentration-based analysis requires normalization of the BCFWD/LW, BAFWD/LW, BSAFOC/LW, and BSSAFOC/LW endpoints using partition coefficients related to the reference phase of interest. The BMF and TMF lipid weight concentration-based analyses do not require adjustments because these values are already equivalent to fugacity-based values. The fugacity-based BCF, BAF, BSAF, BSSAF, and BMF directly reflect the thermodynamic equilibrium status of the chemical between the 2 media included in the ratio calculations. The TMF calculated using the fugacity approach is analogous to the TMF calculated using lipid normalized concentrations, except the chemical data are expressed in terms of fugacities rather than concentrations.

Table 3. Calculation of fugacity ratios for bioaccumulation endpoints
Bioaccumulation endpointConcentration calculationsFugacity calculations
  • BAF = bioaccumulation factor; BCF = bioconcentration factor; BMF = biomagnification factor; BSAF = biota–sediment accumulation factor; BSSAF = biota–suspended sediment accumulation factor; TMF = trophic magnification factor.

  • a

    KLW (lipid–water partition coefficient) is assumed equivalent to KOW (n-octanol–water partition coefficient)

  • b

    KOC (organic carbon–water partition coefficient) is estimated as 0.35 × KOW following Seth et al. (1999).

  • c

    y-axis data are lipid normalized concentrations for the concentration-based TMF or fugacities for the fugacity-based TMF.

BCFa(CB-LW/CWD)/KLW = BCFWD/LW/KLWBCFWD/LW/KOWfbiota/fwater
BAF(CB-LW/CWD)/KLW = BAFWD/LW/KLWBAFWD/LW/KOWfbiota/fwater
BSAFb(CB-LW/CS-OC) (KOC/KLW) = BSAFOC/LW (KOC/KLW) ≈ BSAFOC/LW (KOC/KOW) = 0.35 × BSAFOC/LWfbiota/fsediment
BSSAF(CB-LW/CSS-OC) (KOC/KLW) = BSSAFOC/LW (KOC/KLW) ≈ BSAFOC/LW (KOC/KOW) = 0.35 × BSSAFOC/LWfbiota/fsuspended sediment
BMFCB-LW/CD-LW = BMFLW/LWfbiota/fdiet
TMFcem or 10m=TMFem or 10m

Conversions of the BCFWD/LW and BAFWD/LW to fugacity ratios require that the metrics be calculated based on the truly dissolved chemical concentrations in the water. These measurements can be technically challenging for very hydrophobic chemicals (i.e., log KOW > ∼7.5) and model estimates of the dissolved concentration from the measured total water concentration for such chemicals can also be uncertain. This aspect of uncertainty should be considered when interpreting and comparing BCFs and BAFs for very hydrophobic chemicals. Finally, it may be necessary to consider “lipid equivalent” normalization when the lipid content in certain biota are low (e.g., <1%), as can be the case for plankton and certain invertebrates (deBruyn and Gobas 2007). When lipid contents are low, proteins, carbohydrates, and water phases of the organism can become important sorptive phases relative to the lipid phase. Lipid equivalent normalization accounts for the influences of these additional phases when the lipid phase is not the dominant sorptive phase for the chemical in the organism.

EVALUATION OF BIOACCUMULATION ENDPOINT DATA

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BIOACCUMULATION TERMINOLOGY AND ENDPOINT METRICS
  5. COMPARISON APPROACH
  6. EVALUATION OF BIOACCUMULATION ENDPOINT DATA
  7. METHODS
  8. RESULTS AND DISCUSSION
  9. CONCLUSIONS
  10. SUPPLEMENTAL DATA
  11. EDITOR'S NOTE
  12. Acknowledgements
  13. REFERENCES
  14. Supporting Information

When bioaccumulation data are used in support of regulatory decision making, a critical review of measured bioaccumulation data and predicted bioaccumulation estimates is essential before their use. This recommendation, in part, is based on the varying quality of reported values in the literature and the substantial improvements in measurement quality that have occurred over time due to improvements in sampling techniques, analytical measurements, and experimental methods. The goal of any data quality evaluation is to limit and define uncertainties using the available data so that these uncertainties are transparently communicated. Unfortunately, carrying out the evaluation does not necessarily guarantee that all possible errors and biases associated with the data will be identified.

Published methods and regulatory guidelines are available for measuring laboratory-based BCFLs (OECD 1996; USEPA 1996a, 1996b; ASTM 2007), BSAFLs (USEPA/USACE 1998; USEPA 2000), and BMFLs (Woodburn et al. 2008; OECD 2010). Tests that follow these methods and report complete test results provide some assurance of data quality, however, it remains necessary to critically evaluate the data. For example, USEPA guidelines for measuring BCFLs (USEPA 1996a, 1996b) state that the solubility of the chemical should not be exceeded in the test. Although the guidelines allow for the use of solubilizing agents and their use is not recommended, it is not uncommon to see results from test using solubilizing agents where the aqueous exposure concentrations exceed the water solubility limit of the test chemical (Geyer et al. 1994; Yakata et al. 2006; Arnot and Gobas 2006).

Numerous BCFLs have been reported in the literature; however, many of these measurements were not carried out to current standards, because the tests were done before the development of standardized test guidelines or because the bioaccumulation data reported were a secondary objective of the primary study. Arnot and Gobas (2006) developed data quality assessment methods derived from standard test protocols and applied these methods to approximately 5300 BCFL data points. Measured data were either considered to be “acceptable” or “uncertain” based on this evaluation. Approximately 45% of the BCFL data were considered to contain at least one major source of uncertainty. These uncertainties included the lack of measured and/or reported concentrations in the water, exceeding the aqueous solubility of the chemical, insufficient data and/or reporting of data to document steady-state conditions, and lack of lipid content data in the BCFL measurements.

In 2008, Parkerton et al. (2008) derived an evaluation scheme for BCFL data following an approach developed for assessing toxicity data (Klimisch et al. 1997). The approach of Parkerton et al. (2008) assigns measurements into 1 of 4 reliability categories. “Reliable” data were generated from studies that comply with, or are comparable to, published guidelines or were conducted with accepted methods that are described in sufficient detail. If studies include adequate documentation, but lack specific details or deviate from guideline requirements, such data may be deemed “reliable with restrictions,” provided that deviations were judged acceptable, i.e., were unlikely to significantly influence the study results. Data are determined to be “not reliable” when key considerations were clearly omitted or when documentation reveals unacceptable test performance or methodological flaws. Data were defined as “not assignable” when sufficient detail was not provided on critical study aspects to reach an objective decision on data quality.

The aforementioned BCFL data evaluation schemes of Arnot and Gobas (2006) and Parkerton et al. (2008) generally determine the acceptability or reliability, respectively, of data from a study based on: 1) clear specification of the test substance (e.g., purity) and test species (e.g., life stage, sex, weight, lipid content); 2) analysis of both test species and exposure medium for the parent chemical and important ancillary parameters like lipid and organic carbon content; 3) no significant adverse effects on the test organisms in the study; and 4) reported test data that reflect steady-state conditions with unambiguous units. Additionally, for BCFL studies, 2 important exposure conditions used to judge reliability were: 1) exposure concentrations below the chemical's aqueous solubility; and 2) constant exposure concentrations during the uptake phase. For all studies, conditions and parameters like temperature, dissolved oxygen, total organic carbon, and exposure pH for dissociating chemicals need to be reported. Evaluation schemes for data reliability are not available for laboratory-determined BSAFLs and BMFLs. The general guidelines for BCFL data should, however, be applicable to these laboratory metrics as well.

Guidelines for measuring and evaluating field bioaccumulation endpoints (i.e., BAFFs, BMFFs, BSAFFs, and TMFs) are limited. The USEPA (2009a) has recently published guidance for designing field studies to measure BAFFs or BSAFFs for fish and shellfish. The key to measuring meaningful bioaccumulation metrics with accuracy is that the samples for the exposure medium must be representative of the actual exposure of the organisms collected. Furthermore, adequate field collection designs must also account for the variability in exposure concentrations and tissue residues. Because of chemical uptake and depuration kinetics in biota, the spatial and temporal variability of chemical concentrations in the exposure media, variability in the concentrations in the organisms, and biotransformation rates of the chemical in the food web, a universal sampling design (i.e., number of sampling events over time, number of organisms per sample, and total number of individual organisms or composite samples collected and analyzed) cannot be provided for all chemicals and sampling locations (Burkhard 2003). In addition to USEPA guidance for site specific bioaccumulation determinations, guidelines for contaminant monitoring programs are available (UNEP 2003, 2004; de Wit et al. 2004; Swackhamer et al. 2009) and general recommendations for interpreting the reliability of BAFF data have been proposed (Burkhard 2003, Arnot and Gobas 2006). In general, these guidelines stress: 1) the collection of an adequate number of samples to obtain sufficient statistical power to detect temporal and/or spatial changes; 2) consistency in the types of samples collected over time (e.g., species, age class, and sex); 3) consistency in the time and place of the sample collections over time; and 4) QA and/or QC programs for sample collection and chemical analysis. Monitoring programs with these characteristics are typically able to document the presence of a contaminant and the temporal trend in contaminant concentrations.

Guidelines for measuring BMFLs and BMFFs are currently unavailable. With the ongoing update of the Organization of Economic Cooperation and Development (OECD) 305 guideline, a standardized laboratory procedure will soon become available for a dietary bioaccumulation test, and this procedure will enable high quality BMFLs to be determined. For measurement of BMFs in the field, the major difficulty is determining the actual diet of the organism. Often, 1 major prey species is used in the calculation even though the organism has a mixed diet, and clearly, this introduces uncertainty into the measured BMFFs. Collection and analyses of representative samples is critical in the measurement of BMFFs.

At this time, there are no published guidelines for measuring TMFs even though this field metric is increasingly being reported. Use of contaminant monitoring program guidelines for the collection and analysis of samples would provide residue data suitable for TMF calculations. Borgå et al. (2012) provide additional guidance on measuring TMFs, e.g., number of trophic levels, base species for TMF calculation, and number of samples.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BIOACCUMULATION TERMINOLOGY AND ENDPOINT METRICS
  5. COMPARISON APPROACH
  6. EVALUATION OF BIOACCUMULATION ENDPOINT DATA
  7. METHODS
  8. RESULTS AND DISCUSSION
  9. CONCLUSIONS
  10. SUPPLEMENTAL DATA
  11. EDITOR'S NOTE
  12. Acknowledgements
  13. REFERENCES
  14. Supporting Information

Selection of chemicals

To demonstrate the method of comparing bioaccumulation metrics, a set of chemicals was selected with a wide variety of physicochemical properties, from different chemical classes, with different relative biotransformation rates (Table 4). Additionally, chemicals were selected that exhibited different behavior in aquatic and terrestrial food webs. A key consideration in the selection of chemicals was the availability of adequate data for more than just 1 or 2 bioaccumulation metrics from both laboratory and field studies.

Table 4. Chemical properties and fish biotransformation rates
Chemical (abbreviation)Physicochemical propertiesFish biotransformation rate
LogReferenceHalf-lifea (days)Qualitative ratebReference
KOWKOA
  • A = Schenker et al. 2005; B = selected values and Schenker et al. 2005 harmonization methods; C = USEPA 2009b, Mackay et al. 2006, and Schenker et al. 2005 harmonization methods; E = Arnot et al. 2009; F = USEPA 2009b.

  • a

    Median half-life estimate for a 10 g fish.

  • b

    See Arnot et al. (2009) for qualitative categories.

Pyrene5.198.73A2.1ModerateE
Di(2-ethylhexyl) phthalate (DEHP)7.5613.01B2.8ModerateE
Endosulfan sulfate3.709.91B4.3ModerateF
β-Endosulfan4.789.53A10.5SlowF
α-Endosulfan4.938.49A11.7SlowF
Decabromodiphenyl ether (PBDE-209)8.6815.29A14.5SlowE
Perfluorooctanesulfonic acid (PFOS)5.476.47B17.7SlowF
β-Hexachlorocyclohexane (β-HCH)3.918.74A27.6SlowE
γ-Hexachlorocyclohexane (γ-HCH)3.767.72A39.5SlowE
2,2′,4,4′-Tetrabromodiphenyl ether (PBDE-47)6.3910.44A39.5SlowE
2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD)6.8010.05C60.1SlowE
2,2′,4,4′,5,5′-Hexabromodiphenyl ether (PBDE-153)7.0811.89A102.6Very slowE
2,2′,5,5′-Tetrachlorobiphenyl (PCB-52)5.958.22A132.0Very slowF
Hexachlorobenzene (HCB)5.617.12A272.9Very slowE
2,2′,4,4′,5,5′-hexachlorobiphenyl (PCB-153)6.869.45A352.4Very slowE
Decamethylcyclopentasiloxane (D5)7.524.80B545.8Very slowE
Decachlorobiphenyl (PCB-209)8.2710.95C731.1Very slowE

Assembly of bioaccumulation data

The literature was searched for chemicals using a variety of searching tools and databases including ECOTOX, Science Direct, Cambridge Scientific Abstracts, SCOPUS, and Dialog. In addition, data were taken from existing compilations for BCFLs and BAFLs (Arnot and Gobas 2006) and BSAFFs (USEPA 2008).

In this study, measured bioaccumulation data were screened with regard to reliability. The BCFL and BAFL data used were those values considered acceptable from the Arnot and Gobas (2006) data compilation and evaluation. For the purposes of the current evaluation, all values in the BSAFF data set of USEPA (2008) were considered acceptable. For the other bioaccumulation metrics; e.g., BMFLs, BMFFs, and TMFs, individual data were evaluated on a study-by-study basis because of the lack of defined evaluation guidelines. In general, to determine the acceptable nature of the data, the quality of analytical measurements, approach to steady-state conditions, number of samples, and adequacy of the field sampling design; e.g., proper temporal and spatial coordination, were examined. For each chemical, bioaccumulation data used in this analysis are provided in the Supplemental Data, sorted by endpoint. Summary tables are also provided for each chemical in the Supplemental Data, and the figures in this report were directly created from the data in the summary tables. In total, 2393 bioaccumulation endpoints were assembled from 171 studies and/or reports for the evaluated chemicals.

RESULTS AND DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BIOACCUMULATION TERMINOLOGY AND ENDPOINT METRICS
  5. COMPARISON APPROACH
  6. EVALUATION OF BIOACCUMULATION ENDPOINT DATA
  7. METHODS
  8. RESULTS AND DISCUSSION
  9. CONCLUSIONS
  10. SUPPLEMENTAL DATA
  11. EDITOR'S NOTE
  12. Acknowledgements
  13. REFERENCES
  14. Supporting Information

The conversion of all bioaccumulation metrics to fugacity ratios enables direct comparison of both laboratory and field data sets. For example, a laboratory-measured BCFL can be compared to a field-measured BMFF when expressed on a fugacity ratio basis. Additionally, the fugacity ratios above, at, or below unity provide insights regarding biomagnification, equilibrium partitioning, or trophic dilution, respectively, for the chemical of interest. For fugacity ratios greater than unity, chemical residues increase with increasing trophic level; i.e., biomagnification occurs. For fugacity ratios less than unity, chemical residues decrease with increasing trophic level; i.e., biomagnification does not occur. Plausible explanations for the fugacity ratios to be less than unity include biotransformation of the chemical in the organism and/or its prey, kinetic and/or bioavailability limitations, chemical disequilibrium conditions between the sediment and water column (Thomann 1989; USEPA 2003), and non steady-state conditions.

Caution must be exercised in interpreting laboratory and field metrics (after conversion to fugacity ratios) relative to unity due to potential biases and uncertainties associated with the individual measurements. Biases and uncertainties will arise because of inadequate sample collection and analysis with respect to number of samples, actual linkage of the biota to exposure of the reference phase (water, food, or sediment) samples, analytical detection limits, chromatographic resolution in the instrumental analysis, and detector specificity (electron capture detector [ECD] versus mass spectrometry [MS]). However, some generalities can be made. For example, the BCFL fugacity ratio does not include dietary exposure, and thus, this ratio should in principle never exceed unity; i.e., equilibrium conditions. Furthermore, the BMF and TMF fugacity ratios do not include the accumulation step from the sediment and water to the organism but rather only biomagnification processes involved with predator-prey interactions. As a result, across ecosystems, BMFs and TMFs are expected to have smaller variability than BAFFs, BSAFFs, and BSSAFFs metrics because these latter metrics incorporate biomagnification processes plus all other bioaccumulation processes associated with ecosystem conditions; i.e., bioavailability of the chemical, chemical disequilibrium conditions between the sediment and water column, and the dynamics of the system.

PCBs, PBDEs, HCHs, and HCB

To evaluate the proposed fugacity ratio comparison approach, data for 9 recalcitrant chemicals were assembled from the literature: 2,2′,5,5′-tetrachlorobiphenyl (PCB-52), 2,2′,4,4′,5,5′-hexachlorobiphenyl (PCB-153), decachlorobiphenyl (PCB-209), 2,2′,4,4′-tetrabromodiphenylether (PBDE-47), 2,2′,4,4′,5,5′-hexabromodiphenylether (PBDE-153), decabromodiphenylether (PBDE-209), β-hexachlorocyclohexane (β-HCH), γ-hexachlorocyclohexane (γ-HCH), and hexachlorobenzene (HCB) (Figure 1 and Supplemental Data). These chemicals have slow to very slow biotransformation rates and have a wide range of hydrophobicities; e.g., log KOWs ranging from 3.76 to 8.68 (Table 4).

thumbnail image

Figure 1. Bioaccumulation endpoint fugacity ratios for 2,2′,5,5′-tetrachlorobiphenyl (PCB-52), 2,2′,4,4′,5,5′-hexachlorobiphenyl (PCB-153), decachlorobiphenyl (PCB-209), 2,2′,4,4′-tetrabromodiphenylether (PBDE-47), 2,2′,4,4′,5,5′-hexabromodiphenylether (PBDE-153), decabromodiphenylether (PBDE-209), β-hexachlorocyclohexane (β-HCH), γ-hexachlorocyclohexane (γ-HCH), and hexachlorobenzene (HCB). The minimum (blue circle), 25th percentile (red downward triangle), geometric mean (green circle), median (orange diamond), 75th percentile (pink upperward triangle), and maximum (brown circle) values. Vertical line connects minimum and maximum values for an endpoint. For endpoints with 1 or 2 values, the value or the minimum, geometric mean, and maximum values are plotted, respectively. See Supplemental Data for numerical data.

Download figure to PowerPoint

The largest ranges (by visual inspection) in fugacity ratios primarily occur for BSAFF and BMFF endpoints for all 9 chemicals (Figure 1 and Supplemental Data) whereas the smallest ranges in fugacity ratios primarily occur for TMF and BMFL. The exception is the PBDE-153 BMFLs where all studies seem to be reliable; however, the BMFL for carp of 0.013 is 100-fold less than the BMFLs for trout of 9.4 and 19.5, and salmon of 2.9 (see Supplemental Data). As discussed earlier, the greater variance of BSAF is explained by the incorporation of differing ecosystem conditions into the individual measurements and the difficulties in collecting sediments that are truly representative of the organism's actual exposure. Furthermore, reduced chemical bioavailability due to the presence of black carbon phases or other highly sorptive phases in the sediments will increase variability in reported BSAF values (Cornelissen et al. 2005). For BMFFs, uncertainties arise from inexact knowledge of the prey items consumed by the organism. In most cases, this information is approximate and variability in the BMFFs reflects this lack of knowledge, e.g., the BMFF is computed using 1 or more of the major prey items and not all prey items.

The metric with the least variability in the current data compilations is the TMF. TMFs represent the average change in chemical residues between trophic positions across the complete food web. Relative trophic positions are assigned using stable isotopes of nitrogen and the requirement to know precise predator-prey relationships can be eliminated (Fisk et al. 2001; Borgå et al. 2012). The other bioaccumulation fugacity ratio metric with limited variability is the laboratory-measured BMFL. With laboratory feeding studies, the concentration of the chemical and lipid content in the organism's diet are known. With high quality analytical measurements, reliable BMFLs can be determined that contribute to greater precision in this metric.

Comparison of the fugacity ratios for laboratory-measured BCFLs and field-measured BAFFs reveals that the BAFFs are larger than BCFLs for the same chemical. The ratios of median BAFF to median BCFL for fish are 4.2, 10, 113, 5.0, 1.9, and 2.5 for PCB-52, PCB-153, PCB-209, β-HCH, γ-HCH, and HCB, respectively. To our knowledge, there are no reliable quality BCFLs for PBDE congeners currently available based on data quality assessment methods (Arnot and Gobas, 2006; Parkerton et al. 2008). The BAFFs were expected to be larger than the BCFLs because of dietary exposure and subsequent biomagnification that occurs in the field. The fugacity ratios for the laboratory-measured BMFLs for mammals are all less than the field-measured BMFFs. The ratios of the median BMFF to median BMFL are 3.0, 9.2, and 2.3 for PCB-153, PBDE-153, and γ-HCH, respectively. In contrast, comparisons of the BMFs for fish present an inconsistent picture. For PCB-52, PCB-153, PBDE-47, and PBDE-153, corresponding laboratory values are larger than median field values whereas for PBDE-209 and HCB, their median laboratory values are less than the median field values. We attribute this discrepancy to the uncertainties in knowing the actual diets of field-collected organisms. The median BSAFs for mollusks also exhibit a conflicting picture with some chemicals having larger laboratory values in comparison to the field and vice versa. We attribute this behavior, in part, to sampling issues (obtaining sediments representative of the mollusks actual exposures) and the lack of steady-state residues for the organisms in the laboratory and/or field measurements.

One-way analysis of variance for the individual chemicals reveals that the field metrics are significantly greater than the laboratory endpoints for PCB-52, PCB-153, HCB, β-HCH, and γ-HCH (Tukey test, α = 5% using log transformed fugacity ratios). For the PBDEs, field metric averages were greater but not statistically greater (Tukey test, α = 5% using log transformed fugacity ratios) than the average of the laboratory metrics.

All of the chemicals in Figure 1 are generally considered bioaccumulative and have slow to very slow biotransformation rates. Of particular interest is that the fugacity ratios for both the laboratory and field measurements yield similar interpretation relative to equilibrium conditions; i.e., observed fugacity ratios relative to unity. Although there are large ranges in some of the metrics, the median fugacity ratios are consistently greater than 1. Furthermore, the TMFs for all 9 chemicals are greater than 1 or range across 1 consistent with the fact that these chemicals are all considered to biomagnify. In terms of rank order, PCB-153 and PBDE-153 have the highest biomagnification potential whereas γ- and β-HCH's, and HCB appear to have the lowest biomagnification potential among these 9 chemicals. The laboratory and field fugacity ratios provide generally consistent biomagnification assessment information for these chemicals and support the use of this weight of evidence approach in such evaluations.

Pyrene, di-2-ethylhexyl phthalate, and 2,3,7,8-tetrachlorodibenzo-p-dioxin

To complement information compiled for the 9 chemicals discussed above, data were assembled for 3 substances that represent chemical classes known to be susceptible to biotransformation; i.e., polyaromatic hydrocarbons (PAHs), phthalate esters, and polychlorinated dibenzo-p-dioxins. Pyrene and di-2-ethylhexyl phthalate (DEHP) have moderate biotransformation rates whereas 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) has a slower but still significant biotransformation rate (Table 4). The differences in biotransformation rates are anticipated to cause fugacity ratios for pyrene and DEHP to be much lower than those for TCDD.

As expected, laboratory metrics for fish are well below unity for pyrene and DEHP, and for TCDD, they are much closer to and span unity. Laboratory metrics for mussels and oligochaetes (Lumbricidae) are higher and approach or exceed unity for all 3 chemicals (Figure 2 and Supplemental Data). These results are consistent with available data indicating the limited metabolic capability of lower invertebrates and higher capacity for biotransformation in higher invertebrates and fish for these types of chemicals. Field metrics for pyrene exhibit a similar pattern with fish and bird endpoints being lower than endpoints for decapods and mussels. Field metrics for TCDD span unity except for BSAFs where the ratios are approximately 10% for all phyla. In contrast, although fish BAFF fugacity ratios for DEHP are well below unity, fish BSAF ratios vary over 6 orders of magnitude from 0.001 to approximately 100. Whereas disequilibrium between overlying water and sediment can result in elevated BSAFs for metabolized chemicals, a more likely explanation is contamination introduced during extraction and analysis of field tissue samples that contain low levels of DEHP. Background contamination in the trace analysis of DEHP in environmental matrices is a commonly recognized problem that can result in false positives and overestimated concentrations (David and Gans 2003). Thus, difficulty in obtaining reliable analytical measurements likely accounts for the large variability in reported fish BSAFs and confounds interpretation of bioaccumulation metrics derived from such data for DEHP. These results highlight the potential erroneous conclusions that can be drawn if limited data for a single bioaccumulation metric are interpreted without careful consideration of data reliability and in isolation of other available information.

thumbnail image

Figure 2. Bioaccumulation endpoint fugacity ratios for pyrene, di-(2-ethylhexyl)-phthalate (DEHP), and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). See Figure 1 for description of symbols and Supplemental Data for numerical data.

Download figure to PowerPoint

As expected, TMFs for both pyrene and DEHP are below unity, consistent with the concept of trophic dilution due to the mitigating role of substantial biotransformation. The TMFs for TCDD are all slightly less than unity (geometric mean of 0.94 [see Supplemental Data, 2378TCDD.xls]) indicating that residues across aquatic food webs decrease very slightly if at all. TMFs reported for DEHP were generated using detailed procedures that minimized laboratory contamination artifacts (Mackintosh et al. 2004). Visual inspection of Figure 2 suggests that for these 3 chemicals laboratory metrics tend to be less variable than field metrics, except for the TMFs. The TMF is often less variable because normalization is averaged across multiple trophic levels as previously discussed. However, the absolute magnitude depends on the nature and number of species included in the food web investigated (Borgå et al. 2012). Variability in bioaccumulation metrics for pyrene and DEHP also appears to be greater than that for substances not susceptible to biotransformation (see Figure 1 for PCBs). Thus, variability in metrics may be diagnostic of the variable influence biotransformation plays in altering bioaccumulation potential of a substance in laboratory and field data sets. These findings underscore the need for adopting a weight-of-evidence approach in analysis of bioaccumulation metrics. Despite variability, the laboratory and field fugacity ratios provide similar interpretation of biomagnification potential, i.e., pyrene, DEHP, and TCDD do not biomagnify and exhibit a lower biomagnification potential than chemicals shown in Figure 1. The progression from pyrene and DEHP to TCDD, and then to the chemicals shown in Figure 1 demonstrates the importance of the biotransformation rate (i.e., moderate, slow, and very slow rates, respectively) on the biomagnification potential, and potentially, where their individual fugacity ratios should reside relative to unity. Because bioaccumulation is the net result of competing rates of chemical uptake and elimination by an organism, some degree of biotransformation does not automatically suggest that the chemical will be subject to biodilution, i.e., its fugacity ratio will be less than 1, and therefore it will not biomagnify.

One-way analysis of variance of the DEHP, pyrene, and TCDD data reveals that the field metrics are not significantly greater than their laboratory endpoints, respectively (Tukey test, α = 5% using log transformed fugacity ratios). Visual examination of the data in Figure 2 (that contains chemicals that are metabolized at a moderate and slow rates), in comparison to Figure 1 (that contains chemicals that are slowly and very slowly metabolized) suggests that biotransformation processes cause the median values of laboratory and field metrics to be in closer agreement when compared to chemicals that have slow or very slow biotransformation rates.

Endosulfan

The bioaccumulation metric comparison method was further evaluated by assembling and evaluating data for endosulfan (Figure 3 and Supplemental Data). Endosulfan was introduced in 1954 and is a current-use organochlorine pesticide that has been targeted recently for phase-out by the USEPA. Endosulfan is sold as a mixture of the α- and β-isomers with ratios from 2:1 to 7:3 depending on the technical mixture (Weber et al. 2010). Although the biotransformation rates for the α- and β-isomers are considered slow in aquatic organisms (Table 1), the α- and β-isomers are transformed into endosulfan sulfate, and endosulfan sulfate is recalcitrant (Weber et al. 2010). Consequently, risk assessment currently focuses on the total residue of the α-isomer, β-isomer, and endosulfan sulfate (USEPA 2010). Testing of endosulfan has been carried out using the commercial mixture and individual isomers at the laboratory and mesocosm scales. Field measurements have been carried out for some combination of the 3 chemicals, although endosulfan sulfate has the fewest measurements. One of the past difficulties in carrying out low-level chemical analysis of the α-isomer is interferences from chlordanes when using gas chromatography analysis with electron capture detection (Weber et al. 2010).

thumbnail image

Figure 3. Bioaccumulation endpoint fugacity ratios for α-endosulfan, β-endosulfan, endosulfan product (mixture of α-endosulfan and β-endosulfan isomers), and endosulfan sulfate. See Figure 1 for description of symbols and Supplemental Data for numerical data.

Download figure to PowerPoint

The fugacity ratios for α-endosulfan and β-endosulfan bioaccumulation metrics are consistent with those for the halogenated organics in Figure 1. Endosulfan has the greatest variability in fugacity ratios for BSAFs and BMFs and the smallest variability for TMFs. The field fugacity ratios are larger than the laboratory- and mesocosm-derived metrics. Endosulfan sulfate has fugacity ratios greater than 1 for all metrics, consistent with the chemical's recalcitrant nature and formation of endosulfan sulfate in vivo from parent isomers. Figure 3 includes data for studies carried out with and reported for the endosulfan mixture (sum of the α- and β-isomers).

For endosulfan, the fugacity ratio approach allows easy examination of all of the bioaccumulation data, i.e., individual compounds, mixture of the α- and β-isomers, and its biotransformation product, endosulfan sulfate. The approach places all data on the same numerical scale and readily allows detection of the forms of the chemical that biomagnify.

Decamethylcyclopentasiloxane and perfluorooctane sulfonate

To further evaluate the approach, data were assembled for 2 additional chemicals, (decamethylcyclopentasiloxane [D5] and perfluorooctane sulfonate [PFOS]) (Figure 4 and Supplemental Data). These chemicals are not considered to be traditional persistent organic pollutants (POPs), e.g., halogenated nonionic organic chemicals like DDTs or PCBs, due to their unique chemical properties, chemical composition, and behavior.

thumbnail image

Figure 4. Bioaccumulation endpoint fugacity ratios for decamethylpentacyclosiloxane and bioaccumulation endpoint data for perfluorooctane sulfonic acid (PFOS). The PFOS data are expressed on wet weight in tissue, dry weight in sediments and particulates, and total concentration in water phases, and the data encircled in blue are tissue to water concentration ratios. See Figure 1 for description of symbols and Supplemental Data for numerical data.

Download figure to PowerPoint

Decamethylcyclopentasiloxane is hydrophobic (log KOW = approximately 8), has some characteristics of nonpolar organic chemicals in terms of its partitioning behavior, and has unique low surface energy properties (i.e., surface tension). D5 has been used in some household products and in selected personal care products, such as antiperspirants and shampoos, so the compound may be found downstream of wastewater treatment plants. The bioaccumulation and biomagnification behavior of cyclic siloxanes including D5 in aquatic ecosystems and food webs is still undergoing assessment, as noted by Howard and Muir (2010). Unlike many compounds of bioaccumulative interest, the analytical methodology for determining cyclic siloxanes is not well-established (Alaee and Steer 2008). An additional complicating factor regarding D5 and other commonly used cyclic siloxanes is that their wide use, combined with high volatility (e.g., vapor pressures >1 Pa), may make them widespread contaminants in both laboratories and laboratory reagents. Thus, minimizing siloxane contamination during sample collection, storage, and chemical analysis is a major challenge for determination of reliable bioaccumulation metrics. D5 has a high empirical laboratory BCF of greater than 5000 L/kg (wet weight in fish) (Drottar and Merrifield 2005) suggesting significant bioconcentration of D5 via the aqueous exposure route. However, the BCFL fugacity ratio value is less than 0.1 (Figure 4 and Supplemental Data), similar to PCB-209 (Fig. 1). Laboratory BMFL values for dietary uptake into fish range from less than 1 to 4 whereas field BMFF values for fish are considerably lower, ranging from 0.1 to 1, suggesting unknown attenuation mechanisms in the field. Sediment-dwelling oligochaetes and benthic invertebrates show greater consistency between laboratory and field bioaccumulation measurements for D5, with BSAF values between 1 and 10. Finally, TMF values for D5 are approximately 0.4 in both freshwater and marine aquatic ecosystems, indicating biodilution behavior. From a screening perspective, the bioaccumulation metrics expressed as fugacity ratios beyond just the BCFL are helpful in characterizing D5 behavior. Collectively, the data presented in Figure 4 indicate that D5 biomagnification may be highest for sediment-to-benthic invertebrate transfer.

Perfluoroalkyl substances are a class of polar organic chemicals that possess both hydrophilic and hydrophobic properties. This characteristic of PFOS causes it to behave quite differently than other persistent organic pollutants. Nonionic organic pollutants will partition to lipid when accumulated by animals, whereas in the case of PFOS, the unique physicochemical structure of the chemical causes it to preferentially accumulate in or bind to protein-rich tissues in plasma, blood, and liver (Martin et al. 2003a, 2003b). PFOS mimics the structure of fatty acids in biological systems and PFOS undergoes active transport mechanisms (EFSA 2008). Consequently, traditional KOW-based models using lipid-normalized contaminant concentrations do not apply when describing PFOS bioaccumulation (Giesy et al. 2010). As a result, the assembled data were not converted into fugacity ratios but rather left as chemical concentration ratios, e.g., concentration in predator (wet weight) to prey (wet weight) (BMFWW/WWs), concentration in organism (wet weight) to sediment (dry weight) (BSAFDW/WWs), concentration in organism (wet weight) to suspended-solids (dry weight) (BSSAFDW/WWs), and concentration in tissue (wet weight) to water (total) (BCFL-WT/WW and BAFF-WT/WW) (Figure 4 and Supplemental Data).

PFOS bioaccumulation metrics are highly variable depending on the type of tissue selected for analysis. Bioaccumulation metrics derived from whole-body tissue samples are considered the most appropriate for characterizing PFOS behavior, because the tendency for PFOS to preferentially accumulate in liver and blood can result in an overestimation of its bioaccumulative potential. However, values that were derived from liver and blood samples were not excluded from this analysis because of the limited amount of data available. These issues highlight the inherent difficulty in assessing such metrics for perfluorinated compounds such as PFOS.

The majority of PFOS bioaccumulation data available in the literature comes from monitoring programs conducted in the last decade. Comparatively few laboratory tests exist for PFOS, which makes the comparison between laboratory and field-derived values somewhat tenuous. There exists a considerable lack of agreement between laboratory and field bioaccumulation metrics reported in the literature (Giesy et al. 2010; Houde et al. 2006). When comparing laboratory BCFL-WT/WW and field BAFF-WT/WW values for fish, the BAF values derived from field-collected samples are generally higher than the laboratory BCF values (Figure 4 and Supplemental Data). In general, the concentration ratios of BMFs, BSAFs, BSSAFs, and TMFs are greater than 1.0 indicating that PFOS biomagnifies, i.e., increasing chemical transfer to top predators in marine and freshwater environments.

From a screening perspective, D5 and PFOS test the limits of the applicability of the fugacity ratio approach. The approach is best applied to chemicals that partition to and from lipids and organic carbon by diffusion processes and D5 can be evaluated by this approach. However, the fugacity ratio approach is not applicable to chemicals like PFOS that bind to non lipid tissues and/or undergo active transport processes within organisms.

Application of the fugacity ratio approach for biomagnification assessment

Careful examination of the presented bioaccumulation data reveals that the TMF has the least variability and the BMFF and BSAFF have the greatest variability (Table 5). The variability was determined by using a one-way analysis of variance using natural logarithm of fugacity ratios as the response with chemical as the factor. As a result, for the field metrics, their variabilities incorporate differences across ecosystem types (fresh, estuarine, marine), across species within a group, and ecosystem conditions and current chemical loading patterns. For the laboratory metrics, their variabilities incorporate differences across species within a group. The lower variability associated with TMF was expected because it represents the average change in chemical residues between trophic levels across the organisms sampled in the food web. The BMFF and BSAFF have greater variabilities and these variabilities are explainable. The BSAFFs capture all bioaccumulation processes between the sediment and organism for the chemical and clearly, ecosystem differences (e.g., chemical bioavailability, sediment water disequilibria, and dynamics and/or energy of the system) are incorporated into this metric. For both BMFFs and BSAFFs, substantial effort is required to define and collect samples truly representative of their exposures (i.e., the diet of the organism and exposure to the sediment, respectively), and clearly the variability is reduced with increased sample size. The variability of the fugacity ratios for the BMFL endpoint was higher than expected (Table 5). Examination of the BMFL data reveals that removal of 1 data point for PBDE-153 drops the coefficient of variation to 154% for that metric (see discussion above on PBDE-153 and Supplemental Data). With the removal of this data point, the BMFL and BCFL fugacity ratios have similar variances. This outlier highlights the importance of procedures for examining the reliability and comparability of data for bioaccumulation assessment (see section on Evaluating Bioaccumulation Endpoint Metric Data).

Table 5. Variance of bioaccumulation endpoint fugacity ratiosa
Bioaccumulation endpoint Within group mean squared errorCoefficient of variationb (%)ncNumber of chemicals with data
  • a

    One-way ANOVA: response/natural logarithm of fugacity ratio; factors/chemical.

  • b

    Coefficient of variation = (exp(σ2) −1)1/2.

  • c

    Number of individual data points used in the analysis. BCFL = laboratory bioconcentration factor; BMFL = laboratory biomagnification factor; BSAFL = laboratory biota-sediment accumulation factor; BAFF = field bioaccumulation factor; BMFF = field biomagnification factor; BSAFF = field biota-sediment accumulation factor; TMF = trophic magnification factor.

BCFLFish1.251588911
BCFLMollusks0.1338192
BMFLFish1.912405613
BSAFLMollusks2.0025219312
BAFFFish1.7621915815
BAFFMollusks0.94125688
BAFFDecapoda1.42177104
BAFFMammalian0.508193
BMFFFish2.1727914110
BMFFAvian1.60199798
BMFFMammalian2.4732913610
BSAFFFish2.3030046815
BSAFFMollusks2.4532541914
BSAFFDecapoda2.302992098
TMF 0.538317314

Recently, Gobas et al. (2009) proposed a new framework for assessing biomagnification potential for individual chemicals using metrics beyond the BCF or BAF, and proposed the TMF as the “gold standard” for assessing biomagnification potential. The data assembled and subsequent analysis presented here supports this proposal. In the Gobas et al. (2009) biomagnification assessment framework, when reliable TMF data are unavailable, BMF data from both the laboratory and field studies are examined to determine if the BMF is greater than 1. If so, the chemical would be characterized as a chemical that biomagnifies. Given the variability for the laboratory and field measurements, coupled with practical cost and time considerations, laboratory BMF data would be the preferred metric for this next assessment tier. With the on going development of the OECD dietary bioaccumulation test method that is planned as part of the OECD 305 guideline update, standardized laboratory procedures for a dietary bioaccumulation test will be available. Consequently, it is anticipated that variability in future BMFL measurements will decrease relative to the variability observed in compiled data sets such as the data collected in the present study. When reliable TMF and BMF data are unavailable, the biomagnification assessment framework defaults to BCF and BAF metrics. If these measurements are unavailable, models using physicochemical properties (e.g., KOW and KOA) coupled with improved models to estimate biotransformation rates at different trophic levels could be used to provide an improved assessment of biomagnification potential.

CONCLUSIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BIOACCUMULATION TERMINOLOGY AND ENDPOINT METRICS
  5. COMPARISON APPROACH
  6. EVALUATION OF BIOACCUMULATION ENDPOINT DATA
  7. METHODS
  8. RESULTS AND DISCUSSION
  9. CONCLUSIONS
  10. SUPPLEMENTAL DATA
  11. EDITOR'S NOTE
  12. Acknowledgements
  13. REFERENCES
  14. Supporting Information

The approach for comparing and interpreting laboratory and field bioaccumulation metrics on a common thermodynamic basis (i.e., fugacity analysis) represents a straightforward methodology for performing a holistic, substance-specific, weight of evidence evaluation of biomagnification potential. The approach avoids potential biases in traditional assessment procedures that rely solely on the BCF and/or KOW of the chemical (Arnot and Gobas 2006; Kelly et al. 2007; Gobas et al. 2009). Fugacity ratios for all metrics are plotted for a chemical and the differences within and among metrics become readily detectable and easily visualized. Additionally, data gaps as well as data outliers become readily apparent. The identification of data gaps enables direction of future research such as laboratory testing or field work for a specific substance to reduce uncertainty in existing bioaccumulation data. This approach also allows data for different species and food webs to be integrated into the evaluation process. Plotting various bioaccumulation metrics as fugacity ratios also facilitates comparison of bioaccumulation model predictions to laboratory and field bioaccumulation measurements.

Fifteen organic chemicals were examined using the fugacity ratio approach. For chemicals with slow biotransformation rates, field metrics when expressed as fugacity ratios were slightly larger than laboratory metrics. In contrast, for chemicals with moderate biotransformation rates, field metrics (when expressed as fugacity ratios) were equal to and/or slightly less than their associated laboratory metrics. The proposed approach is limited in its applicability to certain ionic organic chemicals. Chemicals like PFOS with active uptake mechanisms and/or that bind to tissues other than lipids cannot be evaluated using this approach.

Another important limitation to this approach is its applicability to the broader perspective of “Will this chemical bioaccumulate?”. The fugacity ratio approach most definitely detects chemicals that biomagnify (i.e., chemicals with fugacity ratios >1) and chemicals that biomagnify do bioaccumulate. However, as illustrated by TCDD, TCDD does not biomagnify (i.e., TMFs are <1, and other fugacity ratios are generally <1) but TCDD does bioaccumulate to unacceptable levels in fish. Based on our current understanding, the biotransformation rate of TCDD is just enough to offset increases in chemical residues due to biomagnification processes within aquatic food webs. Consequently, if one applies the proposed comparison method to assess bioaccumulation potential and not just biomagnification potential, chemicals with fugacity ratios less than unity, indicating trophic dilution, may still require additional evaluations.

Consistent with other efforts that draw on data from the scientific literature, the proposed method is clearly dependent on the availability and quality of bioaccumulation endpoint data. As illustrated by the 15 chemicals examined, some chemicals have endpoints spanning multiple orders of magnitude. Efforts were made to reduce uncertainty in the data sets; however, there will always be uncertainty and it is possible that data quality assumptions used in the present study also contribute to the observed variability. This suggests a need to further develop standardized test methods and data quality assessment methods for all bioaccumulation metrics.

The proposed method for defining, comparing, and interpreting bioaccumulation metrics have several advantages over current practices. These advantages include: 1) the use of all laboratory and field data for which metrics can be derived, 2) the inclusion of field data from ongoing monitoring programs, 3) the detection of questionable data and outliers, 4) the elimination of single metric cutoffs that are difficult to translate or interpret relative to other metrics, and 5) guidance for prioritizing additional bioaccumulation testing for existing or new chemicals.

EDITOR'S NOTE

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BIOACCUMULATION TERMINOLOGY AND ENDPOINT METRICS
  5. COMPARISON APPROACH
  6. EVALUATION OF BIOACCUMULATION ENDPOINT DATA
  7. METHODS
  8. RESULTS AND DISCUSSION
  9. CONCLUSIONS
  10. SUPPLEMENTAL DATA
  11. EDITOR'S NOTE
  12. Acknowledgements
  13. REFERENCES
  14. Supporting Information

This paper is 1 of 5 articles resulting from the “Lab-Field Bioaccumulation Workshop” held in November 2009 in New Orleans, Louisiana, USA. Workshop participants focused on three objectives: 1) compare laboratory and field measurements of bioaccumulation endpoints, 2) evaluate the reasons why laboratory and field bioaccumulation data may not align, and 3) explore the measurement and application of TMFs.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BIOACCUMULATION TERMINOLOGY AND ENDPOINT METRICS
  5. COMPARISON APPROACH
  6. EVALUATION OF BIOACCUMULATION ENDPOINT DATA
  7. METHODS
  8. RESULTS AND DISCUSSION
  9. CONCLUSIONS
  10. SUPPLEMENTAL DATA
  11. EDITOR'S NOTE
  12. Acknowledgements
  13. REFERENCES
  14. Supporting Information

The authors thank ILSI-HESI, USEPA, and SETAC for sponsoring the Laboratory-Field Bioaccumulation Workshop(18–19 November 2009, New Orleans, LA, before the SETAC 30th North America Annual Meeting). The information in this document has been funded in part by the USEPA. It has been subjected to review by the National Health and Environmental Effects Research Laboratory and approved for publication. Approval does not signify that the contents reflect the views of the Agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use. Support for Kevin Farley was in part provided through the NIEHS Superfund Research Program (2P42ES010344-06A2). Jon Arnot acknowledges postdoctoral funding from NSERC.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BIOACCUMULATION TERMINOLOGY AND ENDPOINT METRICS
  5. COMPARISON APPROACH
  6. EVALUATION OF BIOACCUMULATION ENDPOINT DATA
  7. METHODS
  8. RESULTS AND DISCUSSION
  9. CONCLUSIONS
  10. SUPPLEMENTAL DATA
  11. EDITOR'S NOTE
  12. Acknowledgements
  13. REFERENCES
  14. Supporting Information
  • Alaee M, Steer H. 2008. First Annual Workshop on Organosilicon Compounds in the Environment; WSTD contribution 08-041; 2008 March 27–28; Burlington, Ontario, Canada. Burlington (ON): Environment Canada, Water Science and Technology Directorate. p 95.
  • Arnot JA, Gobas FAPC. 2006. A review of bioconcentration factor (BCF) and bioaccumulation factor (BAF) assessments for organic chemicals in aquatic organisms. Environ Rev 14: 257297.
  • Arnot JA, Meylan W, Tunkel J, Howard PH, Mackay D, Bonnell M, Boethling RS. 2009. A quantitative structure-activity relationship for predicting metabolic biotransformation rates for organic chemicals in fish. Environ Toxicol Chem 28: 11681177.
  • [ASTM] American Society for Testing and Materials. 2007. Standard guide for conducting bioconcentration tests with fishes and saltwater bivalve mollusks. In: ASTM annual book of ASTM standards, Volume 11.06. Philadelphia (PA): ASTM. E 1022-94. p 18.
  • Borgå K, Kidd K, Muir DCG, Berglund O, Conder JM, Gobas FAPC, Kucklick J, Malm O, Powell DE. 2012. Trophic magnification factors: Considerations of ecology, ecosystem and study design. Integr Environ Assess Manag 8: 6482.
  • Broman D, Näf C, Rolf C, Zebühr Y, Fry B, Hobbie J. 1992. Using ratios of stable nitrogen isotopes to estimate bioaccumulation and flux of polychlorinated dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs) in two food chains from the Northern Baltic. Environ Toxicol Chem 11: 331345.
  • Burkhard LP. 2003. Factors influencing the design of bioaccumulation factor and biota-sediment accumulation factor field studies. Environ Toxicol Chem 22: 351360.
  • Connolly J, Pedersen C. 1988. A thermodynamic-based evaluation of organic chemical accumulation in aquatic organisms. Environ Sci Technol 22: 99103.
  • Cornelissen G, Gustafsson Ö, Bucheli TD, Jonker MTO, Koelmans AA, van Noort PCM. 2005. Extensive sorption of organic compounds to black carbon, coal, and kerogen in sediments and soils: Mechanisms and consequences for distribution, bioaccumulation, and biodegradation. Environ Sci Technol 39: 68816895.
  • David R, Gans G. 2003. Summary of mammalian toxicology and health effects of phthalate esters. In: The handbook of environmental chemistry. Berlin/Heidelberg (DE): Springer. p 299316.
  • deBruyn MH, Gobas FAPC. 2007. The sorptive capacity of animal protein. Environ Toxicol Chem 26: 18031808.
  • de Wit C, Fisk A, Hobbs K, Muir DCG, Gabrielson GW, Kallenborn R, Krahn MM, Norstrom RJ, Skaare JU. 2004. Arctic monitoring and assessment programme (AMAP) assessment 2002: Persistent organic pollutants in the Arctic. Oslo (NO): AMAP. p xvi.
  • Drottar K, Merrifield J. 2005. 14C-Decamethylcyclopentasiloxane (14C-D5): Bioconcentration in the fathead minnow (Pimephales promelas) under flow-through test conditions. Unpublished report of the Dow Corning Corporation. Report 2005-I0000-55052.
  • [EFSA] European Food Safety Authority. 2008. Perfluorooctane sulfonate (PFOS), perfluorooctanoic acid (PFOA) and their salts. Scientific opinion of the panel on contaminants in the food chain. EFSA J 653: 1131.
  • Fisk AT, Hobson KA, Norstrom RJ. 2001. Influence of chemical and biological factors on trophic transfer of persistent organic pollutants in the north water polynya marine food web. Environ Sci Technol 35: 732738.
  • Geyer HJ, Muir DCG, Scheunert I, Steinberg CEW, Kettrup AAW. 1994. Bioconcentration of superlipophilic persistent chemicals. Environ Sci Pollut Res 1: 7580.
  • Giesy JP, Naile JE, Khim JS, Jones PD, Newsted JL. 2010. Aquatic toxicology of perfluorinated chemicals. In: Reviews of environmental contamination and toxicology. Vol 202. New York (NY): Springer. p 152.
  • Gobas FAPC, de Wolf W, Burkhard LP, Verbruggen E, Plotzke K. 2009. Revisiting bioaccumulation criteria for POPs and PBT assessments. Integr Environ Assess Manag 5: 624637.
  • Houde M, Bujas TAD, Small J, Wells RS, Fair PA, Bossart GD, Solomon KR, Muir DCG. 2006. Biomagnification of perfluoroalkyl compounds in the bottlenose dolphin (Tursiops truncatus) food web. Environ Sci Technol 40: 41384144.
  • Howard PH, Muir DCG. 2010. Identifying new persistent and bioaccumulative organics among chemicals in commerce. Environ Sci Technol 44: 22772285.
  • Kelly BC, Ikonomou MG, Blair JD, Morin AE, Gobas FAPC. 2007. Food web-specific biomagnification of persistent organic pollutants. Science 317: 236239.
  • Klimisch HJ, Andreae M, Tillmann U. 1997. A systematic approach for evaluating the quality of experimental toxicological and ecotoxicological data. Reg Toxicol Pharmacol 25: 15.
  • Mackay D, Di Guardo A, Paterson S, Cowan CE. 1996. Evaluating the environmental fate of a variety of types of chemicals using the EQC model. Environ Toxicol Chem 15: 16271637.
  • Mackay D. 2001. Multimedia environmental models: The fugacity approach. 2nd ed. Boca Raton (FL): Lewis Publishers. p 272.
  • Mackay D, Shiu WY, Ma KC, Lee SC. 2006. Handbook of physical-chemical properties and environmental fate for organic chemicals. 2nd ed. Vol I–IV. Boca Raton (FL): CRC Press. p 4216.
  • Mackintosh CE, Maldonado J, Hongwu J, Hoover N, Chong A, Ikonomou M, Gobas FAPC. 2004. Distribution of phthalate esters in a marine aquatic food web: comparison to polychlorinated biphenyls. Environ Sci Technol 38: 20112020.
  • Martin JW, Mabury SA, Solomon KR, Muir DCG. 2003a. Bioconcentration and tissue distribution of perfluorinated acids in rainbow trout (Oncorhynchus mykiss). Environ Toxicol Chem 22: 196204.
  • Martin JW, Mabury SA, Solomon KR, Muir DCG. 2003b. Dietary accumulation of perfluorinated acids in juvenile rainbow trout (Oncorhynchus mykiss). Environ Toxicol Chem 22: 189195.
  • [OECD] Organisation of Economic Co-operation and Development. 1996. Bioconcentration: Flow-through fish test. OECD Guidelines for the Testing of Chemicals305E. Paris (FR): OECD. p 23.
  • [OECD] Organisation of Economic Co-operation and Development. 2010. Work plan for the test guidelines programme (TGP). Paris (FR): OECD. p 26.
  • Parkerton TF, Arnot JA, Weisbrod AV, Russom C, Hoke RA, Woodburn K, Traas T, Bonnell M, Burkhard LP, Lampi MA. 2008. Guidance for evaluating in vivo fish bioaccumulation data. Integr Environ Assess Manag 4: 139155.
  • Schenker U, MacLeod M, Scheringer M, Hungerbuehler K. 2005. Improving data quality for environmental fate models: a least-squares adjustment procedure for harmonizing physicochemical properties of organic compounds. Environ Sci Technol 39: 84348441.
  • Seth R, Mackay D, Muncke J. 1999. Estimating the organic carbon partition coefficients and its variability for hydrophobic chemicals. Environ Sci Technol 33: 23902394.
  • Swackhamer DL, Needham LL, Powell DE, Muir DCG. 2009. Use of measurement data in evaluating exposure of humans and wildlife to POPs/PBTs. Integr Environ Assess Manag 5: 638661.
  • Thomann R. 1989. Bioaccumulation model of organic chemical distribution in aquatic food chains. Environ Sci Technol 23: 699707.
  • [UNEP] United Nations Environment Program. 2003. Proceedings of the UNEP Workshop to Develop a Global POPs Monitoring Programme to Support the Effectiveness Evaluation of the Stockholm Convention, 2003 March 24–27; Geneva (CH): UNEP. 260 p.
  • [UNEP] United Nations Environment Program. 2004. Guidance for a Global Monitoring Programme for Persistent Organic Pollutants. Geneva (CH): UNEP. 105 p.
  • [USEPA] US Environmental Protection Agency. 1996a. Ecological effects test guidelines, OPPTS 850.1710, Oyster BCF (Public Draft). EPA 712/C-96/127. Washington (DC): USEPA.
  • [USEPA] US Environmental Protection Agency. 1996b. Ecological Effects Test Guidelines, OPPTS 850.1730, Fish BCF (Public Draft). EPA 712/C-96/129. Washington (DC): USEPA.
  • [USEPA/USACE] US Environmental Protection Agency/US Army Corps of Engineers. 1998. Evaluation of Dredged Material Proposed for Discharge in Waters of the U.S.—Testing Manual. EPA/823/B-98/004. Washington (DC): USEPA.
  • [USEPA] US Environmental Protection Agency. 2000. Methods for Measuring the Toxicity and bioaccumulation of Sediment-associated Contaminants with Freshwater Invertebrates,2nd ed.EPA 600/R-99/064. USEPA Office of Research and Development, Mid-Continent Ecology Division, Duluth (MN)/Office of Science and Technology, Washington (DC).
  • [USEPA] US Environmental Protection Agency. 2003. Methodology for Deriving Ambient Water Quality Criteria for the Protection of Human Health (2000), Technical Support Document, Volume 2: Development of National Bioaccumulation Factors. EPA 822/R-03/030. Washington (DC): USEPA Office of Water.
  • [USEPA] US Environmental Protection Agency. 2008. BSAF (Biota-Sediment Accumulation Factor) database. Duluth (MN): USEPA, Office of Research and Development, Mid-Continent Ecology Division. [cited 26 May 2011]. Available from: http://www.epa.gov/med/Prods_Pubs/bsaf.htm
  • [USEPA] US Environmental Protection Agency. 2009a. Methodology for Deriving Ambient Water Quality Criteria for the Protection of Human Health (2000) Technical Support Document. Volume 3: Development of Site-Specific Bioaccumulation Factors. Washington (DC): USEPA. EPA/822/R-09/008.
  • [USEPA] US Environmental Protection Agency. 2009b. Exposure Assessment Tools and Models, Estimation Programs Interface (EPI) Suite, version 4.0. Washington (DC): USEPA, Exposure Assessment Branch.
  • [USEPA] US Environmental Protection Agency. 2010. Endosulfan: 2010 Environmental fate and ecological risk assessment. Washington (DC): USEPA Office of Chemical Safety and Pollution Prevention, Office of Pesticide Programs.
  • Weber J, Halsall CJ, Muir D, Teixeira C, Small J, Solomon K, Hermanson M, Hung H, Bidleman T. 2010. Endosulfan, a global pesticide: A review of its fate in the environment and occurrence in the Arctic. Sci Total Environ 408: 29662984.
  • Woodburn KB, Marino TA, McClymont EL, Rick DL. 2008. Determination of the dietary absorption efficiency of hexachlorobenzene with the channel catfish (Ictalurus punctatus). Ecotoxicol Environ Saf 71: 419425.
  • Yakata N, Sudo Y, Tadokoro H. 2006. Influence of dispersants on bioconcentration factors of seven organic compounds with different lipophilicities and structures. Chemosphere 64: 18851891.

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BIOACCUMULATION TERMINOLOGY AND ENDPOINT METRICS
  5. COMPARISON APPROACH
  6. EVALUATION OF BIOACCUMULATION ENDPOINT DATA
  7. METHODS
  8. RESULTS AND DISCUSSION
  9. CONCLUSIONS
  10. SUPPLEMENTAL DATA
  11. EDITOR'S NOTE
  12. Acknowledgements
  13. REFERENCES
  14. Supporting Information

Additional Supporting information can be found in the online version of this article:

FilenameFormatSizeDescription
IEAM_260_sm_SuppInfo.pdf137KSupplementary Information
IEAM_260_sm_SuppInfo.xls167KSupplementary Information
IEAM_260_sm_SuppInfoD5.xls190KSupplementary Information
IEAM_260_sm_SuppInfoDEHP.xls144KSupplementary Information
IEAM_260_sm_SuppInfoalpha.xls162KSupplementary Information
IEAM_260_sm_SuppInfoalphabeta.xls131KSupplementary Information
IEAM_260_sm_SuppInfobeta.xls143KSupplementary Information
IEAM_260_sm_SuppInfosulfate.xls141KSupplementary Information
IEAM_260_sm_SuppInfoHCB.xls164KSupplementary Information
IEAM_260_sm_SuppInfoHCHbeta.xls142KSupplementary Information
IEAM_260_sm_SuppInfoHCHgama.xls156KSupplementary Information
IEAM_260_sm_SuppInfoPBDE47.xls140KSupplementary Information
IEAM_260_sm_SuppInfoPBDE153.xls134KSupplementary Information
IEAM_260_sm_SuppInfoPBDE209.xls122KSupplementary Information
IEAM_260_sm_SuppInfoPCB52.xls221KSupplementary Information
IEAM_260_sm_SuppInfoPCB153.xls208KSupplementary Information
IEAM_260_sm_SuppInfoPCB209.xls164KSupplementary Information
IEAM_260_sm_SuppInfoPFOS.xls125KSupplementary Information
IEAM_260_sm_SuppInfoPyrene.xls157KSupplementary Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.