Testing common sediment-porewater distribution models for their ability to predict dissolved concentrations of POPs in The Grenlandsfjords, Norway

PCDD/Fs, dioxin-like, and non-dioxin like PCBs in the sediments of high Arctic fjords, Svalbard

Polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDD/Fs) are highly toxic organic compounds, and very few studies on their presence in polar environments have been conducted. This study assessed the concentration and distribution of PCDD/Fs, dioxin-like polychlorinated biphenyls (DL-PCBs), and non-dioxin-like polychlorinated biphenyls in selected fjords of the Svalbard archipelago in Norway. The ∑PCDD/Fs observed for Raudfjorden, Smeerenburgfjorden, Magdalenefjorden, and Kongsfjorden were 22.80 pg/g, 25.65 pg/g, 18.27 pg/g, 33.50 pg/g, and 21.69 pg/g, respectively. The WHO's toxic equivalents values of both ∑PCDD/Fs and ∑DL-PCBs were comparatively higher than those reported in other polar regions. Of the four fjords studied, the sediments from Kongsfjorden exhibited the presence of the most toxic materials, including PCB-126 and PCB-169, of DL-PCBs. More than 80% of the total analysed PCDD/Fs were comprised of highly chlorinated congeners (hexa-to-octa forms). More studies are required to understand the destination and transport of these hazardous pollutants in high Arctic sediments.

Stochastic Framework for Addressing Chemical Partitioning and Bioavailability in Contaminated Sediment Assessment and Management

Passive sampling to quantify net partitioning of hydrophobic organic contaminants between the porewater and solid phase has advanced risk management for contaminated sediments. Direct porewater (Cfree) measures represent the best way to predict adverse effects to biota. However, when the need arises to convert between solid-phase concentration (Ctotal) and Cfree, a wide variation in observed sediment-porewater partition coefficients (KTOC) is observed due to intractable complexities in binding phases. We propose a stochastic framework in which a given Ctotal is mapped to an estimated range of Cfree through variability in passive sampling-derived KTOC relationships. This mapping can be used to pair estimated Cfree with biological effects data or inversely to translate a measured or assumed Cfree to an estimated Ctotal. We apply the framework to both an effects threshold for polycyclic aromatic hydrocarbon (PAH) toxicity and an aggregate adverse impact on an assemblage of species. The stochastic framework is based on a "bioavailability ratio" (BR), which reflects the extent to which potency-weighted, aggregate PAH partitioning to the solid-phase is greater than that predicted by default, KOW-based KTOC values. Along a continuum of Ctotal, we use the BR to derive an estimate for the probability that Cfree will exceed a threshold. By explicitly describing the variability of KTOC and BR, estimates of risk posed by sediment-associated contaminants can be more transparent and nuanced.

Impact of passive sampler protection apparatus on sediment porewater profiles of hydrophobic organic compounds

Passive sampling techniques have been widely used to determine the dissolved concentration profiles of hydrophobic organic compounds (HOCs) in sediment porewater. However, the effects of having a protection for the passive sampler on profiling HOCs concentrations in sediment porewater, especially in deep sediment, have remained unclear. To address this issue, low density polyethylene passive samplers with and without protectors, which consisted of glass fiber filter and porous stainless steel shield, were simultaneously deployed in sediment of the Dongjiang River, South China. The results showed that the protectors retarded the dissipation of performance reference compounds (PRCs) from the sampler by a factor of 2-9. The protectors seemed to exert a negligible effect on the measured concentrations of PAHs, BDE-47, and BDE-99 in surficial sediment porewater (0-14 cm depth) from both samplers. However, the sediment porewater concentration profiles of PAHs and BDE-47 from the sampler with protectors were in agreement with those normalized by dry weight in deep sediment (16-34 cm depth), indicating that a diffusion layer established by the protectors may minimize the probability of local depletion of the target analytes in deep sediment. In addition, the log K_oc values of PAHs, BDE-47, and BDE-99 exhibited a slight increasing trend with sediment depth. This finding suggested that in situ passive sampling techniques could be a feasible tool in determining the site-specific log K_oc values of HOCs at different sediment depths.

Distribution, sources, and behavior of PAHs in estuarine water systems exemplified by Salt River, Taiwan

Water, suspended particulate matter (SPM), and sediment samples were collected from Salt River in Taiwan and analyzed the concentrations of 16 types of polycyclic aromatic hydrocarbons (PAHs). The analysis results were used to examine the distribution, source, partition behavior, and potential ecological risks of PAHs in the estuarine water systems. The mean concentration of total PAHs in water, SPM, and sediment samples was 0.485-10.2 μg/L, 26.7-169 mg/kg dw, and 0.343-29.4 mg/kg dw, respectively. The highest concentration was found at the river mouth and decreased toward the river and sea with the tide. The distribution of the diagnostic ratios of PAHs showed that the combustion of coal and petroleum products are the main sources of PAHs in Salt River. The in site organic carbon normalized partition coefficients for SPM-water (K'_oc(SPMW)) and sediment-water (K'_oc(SedW)) were 2.8-4.5 and 4.6-6.0 (log units), respectively, increasing with the number of rings in PAHs. The values log K'_oc(SedW) and log K'_oc(SPM-W) showed a significant linear correlation with their octanol-water partition coefficients (p < 0.01), and their slopes were 0.427 and 0.316, respectively. The fugacity fraction was used to evaluate the exchange of PAHs in water-SPM-sediment systems. Results showed that in SPM, 2-4-ring PAHs tend to be released into water, whereas 5-6-ring PAHs in water tend to be adsorbed onto SPM. The exchange of PAHs between water and sediment occurs in the direction of adsorption onto sediment from water. The assessment of the mean risk quotient, total toxicity equivalence, and mean effect range-median quotient of PAHs showed that the PAHs in the water and SPM of Salt River may have moderate to high ecological risk. In sediment, PAHs in the lower reaches and estuary may pose moderate to high ecological risk, whereas PAHs in the middle and upper reaches show low to moderate ecological risk.

Occurrence and partitioning of antifouling booster biocides in sediments and porewaters from Brazilian Northeast

Fouling organisms attach and grow on submerged surfaces causing several economic losses. Thus, biocides have been introduced in antifouling paints in order to avoid this phenomenon, but their widespread use became a global problem, mainly in ports, leisure and fishing boat harbors, since these substances can be highly toxic to non-target organisms. The occurrence and environmental behavior of antifouling biocides are especially unknown in some peculiar regions, such as Amazon areas. Thus, the aim of this work was to evaluate, for the first time, levels and the partitioning behavior of the antifouling organic biocides irgarol, diuron and also stable degradation products of dichlofluanid and diuron (DMSA and DCPMU, respectively) in sediments and porewaters from a high boat traffic area located in the Northeast of Brazil, a pre-Amazon region. Our results showed high concentrations of irgarol (<1.0-89.7 μg kg^-1) and diuron (<5.0-55.2 μg kg^-1) in sediments. In porewater, DCPMU (<0.03-0.67 μg L^-1) and DMSA (<0.008-0.263 μg L^-1) were the mainly substances detected. High K_d and K_oc obtained for both irgarol and diuron showed a partitioning preference in the solid phase. This work represents one of the few registers of contamination by antifouling substances in Amazonian areas, despite their environmental relevance.

The utility of solid-phase microextraction in evaluating polycyclic aromatic hydrocarbon bioavailability during habitat restoration with dredged material at moderately contaminated sites

The over- or underprediction of risk in moderately contaminated sediments can have a large impact on the nature of applied management strategies given that concentrations border on being toxic or not toxic. Project managers should give significant consideration as to how moderate levels of contaminants in native sediments and dredged material used for restoration will impact recovery of habitat. Total solid-phase (C_total ) and porewater (C_free ) polycyclic aromatic hydrocarbons (PAHs) were quantified in native sediments and dredged material to determine if the predictions of risk from C_total are consistent with those based on C_free . The sediment matrix phase in which PAHs were quantified resulted in disparate conclusions regarding the predicted reduction in contamination following restoration. Total solid-phase PAH concentrations suggested a significant decrease following restoration, whereas little to no change was observed in measured C_free . Risk metrics based on C_total gave inconclusive estimates for toxicity, whereas measured C_free suggested toxicity is unlikely, a conclusion consistent with toxicity testing. The incorporation of black carbon (BC) into model estimates for C_free gave predictions more consistent with measured C_free , suggesting that geochemical conditions (especially BC) play an important part in predicting toxicity at moderately contaminated sites. In addition to the use of C_free in toxicity evaluation, in-situ C_free measurements provided a constraint on diffusive PAH loads from sediment relative to ongoing stream loads. If passive sampling had been employed during the sampling designs and site evaluations, the costs of toxicity testing would not have been incurred, given that C_free suggested little to no toxicity. The results from the project highlight the benefits to be gained by moving beyond inconclusive, screening-level C_total metrics and implementing more sensitive and accurate C_free metrics in assessments of risk in moderately contaminated sediments. Integr Environ Assess Manag 2018;14:212-223. © 2017 SETAC.

The partitioning behavior of persistent toxicant organic contaminants in eutrophic sediments: Coefficients and effects of fluorescent organic matter and particle size

In the shallow lakes, the partitioning of organic contaminants into the water phase from the solid phase might pose a potential hazard to both benthic and planktonic organisms, which would further damage aquatic ecosystems. This study determined the concentrations of polycyclic aromatic hydrocarbons (PAHs), organochlorine pesticides (OCPs), and phthalate esters (PAEs) in both the sediment and the pore water from Lake Chaohu and calculated the sediment - pore water partition coefficient (K_D) and the organic carbon normalized sediment - pore water partition coefficient (K_OC), and explored the effects of particle size, organic matter content, and parallel factor fluorescent organic matter (PARAFAC-FOM) on K_D. The results showed that log K_D values of PAHs (2.61-3.94) and OCPs (1.75-3.05) were significantly lower than that of PAEs (4.13-5.05) (p < 0.05). The chemicals were ranked by log K_OC as follows: PAEs (6.05-6.94) > PAHs (4.61-5.86) > OCPs (3.62-4.97). A modified MCI model can predict K_OC values in a range of log 1.5 at a higher frequency, especially for PAEs. The significantly positive correlation between K_OC and the octanol - water partition coefficient (K_OW) were observed for PAHs and OCPs. However, significant correlation was found for PAEs only when excluding PAEs with lower K_OW. Sediments with smaller particle sizes (clay and silt) and their organic matter would affect distributions of PAHs and OCPs between the sediment and the pore water. Protein-like fluorescent organic matter (C2) was associated with the K_D of PAEs. Furthermore, the partitioning of PARAFAC-FOM between the sediment and the pore water could potentially affect the distribution of organic pollutants. The partitioning mechanism of PAEs between the sediment and the pore water might be different from that of PAHs and OCPs, as indicated by their associations with influencing factors and K_OW.

Occurrence and significance of polychlorinated biphenyls in water, sediment pore water and surface sediments of Umgeni River, KwaZulu-Natal, South Africa

The Umgeni River is one of the main sources of water in KwaZulu-Natal, South Africa; however; there is currently a lack of information on the presence and distribution of polychlorinated biphenyls (PCBs) in its sediment, sediment pore water and surface water. This study aims to determine the occurrence and significance of selected PCBs in the surface water, sediment pore water and surface sediment samples from the Umgeni River. Liquid-liquid and soxhlet extractions were used for water or pore water, and sediments, respectively. Extracts were cleaned up using a florisil column and analysed by gas chromatography-mass spectrometry. The total concentrations of eight polychlorinated biphenyls were 6.91-21.69 ng/mL, 40.67-252.30 ng/mL and 102.60-427.80 ng/g (dry weight), in unfiltered surface water, unfiltered sediment pore water and surface sediments, respectively. The percentage contributions of various matrices were 4, 36 and 60 % for unfiltered surface water, unfiltered pore water and sediment, respectively. The highest concentrations of PCBs were found in water, pore water and sediment collected from sampling sites close to the Northern Wastewater Treatment Works. The highest chlorinated biphenyl, PCB 180, was the most abundant at almost all sampling sites. To our knowledge, this is the first report on occurrence of polychlorinated biphenyls in the Umgeni River water, pore water and sediment system and our results provide valuable information regarding the partitioning of the PCBs between the water and sediment systems as well as the organic chemical quality of the water.

PCDD/F release during benthic trawler-induced sediment resuspension

Benthic trawling can cause the resuspension of large amounts of sediments. Such regular practice in the Grenland fjord system in the south of Norway has the potential to affect the fate, movement, and bioavailability of sediment-associated polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs). A novel mode of exposing passive sampling devices consisting of towing semipermeable membrane devices attached to the trawl net was used to gauge in situ changes in the freely dissolved concentration of PCDD/Fs on benthic trawler-induced sediment resuspension. Significant accumulation of a number of PCDD/F congeners was observed despite the short (5 h) sampler exposure times. On average, a one order of magnitude increase in freely dissolved PCCD/F concentrations was seen within minutes of the sediment being resuspended. This observation was supported by similar changes in filtered PCDD/F concentrations measured by high-volume sampling prior to resuspension and in the sediment plume.

Bottom trawling resuspends sediment and releases bioavailable contaminants in a polluted fjord

Sediments are sinks for contaminants in the world's oceans. At the same time, commercial bottom trawling is estimated to affect around 15 million km(2) of the world's seafloor every year. However, few studies have investigated whether this disturbance remobilises sediment-associated contaminants and, if so, whether these are bioavailable to aquatic organisms. This field study in a trawled contaminated Norwegian fjord showed that a single 1.8 km long trawl pass created a 3-5 million m(3) sediment plume containing around 9 t contaminated sediment; ie. 200 g dw m(-2) trawled, equivalent to c. 10% of the annual gross sedimentation rate. Substantial amounts of PCDD/Fs and non-ortho PCBs were released from the sediments, likely causing a semi-permanent contaminated sediment suspension in the bottom waters. PCDD/Fs from the sediments were also taken up by mussels which, during one month, accumulated them to levels above the EU maximum advised concentration for human consumption.

A quantitative structure-property relationship analysis of soot-water partition coefficients for persistent organic pollutants

Geometrical optimization and electrostatic potential calculations have been performed at the HF/6-31G level of theory for investigated persistent organic pollutants (POPs). A number of statistically based parameters have been obtained. Relationship between soot-water partition coefficients (logK(SC)) of POPs and the structural descriptors has been established by the multiple linear regression method. The result shows that the quantities derived from electrostatic potential V(s)(-)¯ and V(s,max), together with molecular surface area (A(S)) and the energy of the highest occupied molecular orbital (E(HOMO)) can be well used to express the quantitative relationship between structure and logK(SC) (QSPR) of POPs. Predictive capability of the model has been demonstrated by leave-one-out cross-validation with the cross-validated correlation coefficient of 0.9797. Furthermore, the predictive power of this model was further examined for the external test set with the correlation coefficient of 0.9811 between observed and predicted logK(SC), validating the robustness and good predictive ability of our model. Furthermore, in order to further investigate the applicability of these parameters derived from electrostatic potential in prediction of soot-water partition coefficient for organic pollutants, eleven polycyclic aromatic hydrocarbons (PAHs), eleven polychlorinated biphenyls (PCBs) and nine phenyl urea herbicides (PUHs) from other source have also been studied. The QSPR models established may provide a new powerful method for predicting soot-water partition coefficients (logK(SC)) of organic pollutants.

Role of black carbon in the sorption of polychlorinated dibenzo-p-dioxins and dibenzofurans at the Diamond Alkali superfund site, Newark Bay, New Jersey

The sorption of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) to organic carbon (OC) and black carbon (BC) was measured in two sediment cores taken near the Diamond Alkali superfund site (DA) in the Passaic River and Newark Bay, New Jersey (U.S.A.). An OC partitioning model and a BC-inclusive, Freundlich distribution model were used to interpret measurements of freely dissolved PCDD/Fs using passive samplers in sediment incubations, together with measured sedimentary concentrations of OC, BC, and PCDD/Fs. Samples were also analyzed for polycyclic aromatic hydrocarbons (PAHs) as controls on the two distribution models. The OC partitioning model underpredicted the distribution of PAHs and PCDD/Fs by 10-100-fold. The Freundlich model predicted the distribution of PAHs at the DA to within a factor of 2-3 of observations. Black carbon-water partition coefficients (K(iBC)) for PCDD/Fs, derived from literature results of both field and laboratory studies differed up to 1000-fold from values derived from this study. Contrary to expectations, PCDDs displayed stronger sorption than either PCDFs or PAHs relative to their subcooled liquid aqueous solubilities. Even though the presence of BC in the sediments reduced the overall bioavailability of PCDD/Fs by >90%, the sediments at 2 m depth continue to display the highest pore water activities of PCDD/Fs.

Benthic macroinvertebrate assemblages in remediated wetlands around Sydney, Australia

To investigate potential high organisational level impacts of persistent organic pollution in the wetlands in the Sydney Olympic Park (SOP) remediated site, the benthic macroinvertebrate assemblages of seven wetlands within SOP and two off-site reference wetlands were examined. Sediment cores were collected, stained and preserved from each study site and the macroinvertebrates identified to the appropriate taxonomic level (Class, Order, Family, Subfamily). Data were analysed for taxon richness and macroinvertebrate abundance and multivariate techniques were used to identify chemical/physical characteristics of the sediment, which were important influences on the differences in the assemblage between study sites. Macroinvertebrate abundance was highly variable between study sites and taxon richness was low across all sites. Oligochaetes, nematodes, ostracods and chironomids were the most common taxa found and were the most important in influencing differences between the macroinvertebrate assemblages among the study sites. Sediment grain size and chemical characteristics of the sediments (ΣPAH, ΣPCB, TCDDeq and heavy metal concentrations) were important in separating the study sites based on taxon richness and abundance. Canonical correspondence analysis separated the macroinvertebrate assemblages at newly two created wetlands from those at other study sites including the urban reference sites. Increased sediment POP contamination (particularly as measured TCDDeq and ΣDDT concentrations) is a likely contributor in excluding pollution sensitive taxa and, therefore, alterations to benthic macroinvertebrate assemblages. Further, the influence of TOC suggests the significance of catchment inputs in contributing to changes in macroinvertebrate assemblage. The SOP remediation led to the establishment of wetlands with benthic communities representative of those expected in urban wetlands.

QSPR studies on soot-water partition coefficients of persistent organic pollutants by using artificial neural network

Two quantitative structure property relationship (QSPR) models for predicting soot-water partition coefficients (K(sc)) of 25 persistent organic pollutants (POPs) were developed. One model was established with linear artificial neural network (L-ANN), the other model was developed by using back propagation artificial neural network (BP-ANN). Leave one out cross validation was adopted to assess the predictive ability of the developed models. For the L-ANN model, the square of correlation coefficient (R(2)) between the predicted and experimental log K(SC) is 0.8358 and the RMS%RE is 6.32 for all the compounds. For the BP-ANN model, R(2) is 0.9628 and the RMS%RE is 4.12 for all the compounds. The result of leave one out cross validation demonstrates that both L-ANN and BP-ANN are practicable for developing the QSPR model for K(SC) of the investigated POPs. However, the model established with BP-ANN is better than the model established with L-ANN in prediction accuracy. It is shown that BP-ANN is a promising method for developing QSPR models for K(SC) of POPs.

Estimating the in situ sediment-porewater distribution of PAHs and chlorinated aromatic hydrocarbons in anthropogenic impacted sediments

It has become increasingly apparent that the in situ sediment-porewater distribution behavior of organic compounds within anthropogenic impacted sediments is quite diverse, and challenging to generalize. Traditional models based on octanol-water partitioning generally overestimate native porewater concentrations, and modern approaches accounting for multiple carbon fractions, including black carbon, appear sediment specific. To assess the diversity of this sorption behavior, we collected all peer-reviewed total organic carbon (TOC)-normalized in situ sediment-porewater distribution coefficients, K(TOC), for impacted sediments. This entailed several hundreds of data for PAHs, PCBs, PCDD/Fs, and chlorinated benzenes, covering a large variety of sediments, locations, and experimental methods. Compound-specific K(TOC) could range up to over 3 orders of magnitude. Output from various predictive models for individual carbonaceous phases found in impacted sediments, based on peer-reviewed polyparameter linear free energy relationships (PP-LFERs), Raoult's Law, and the SPARC online-calculator, were tested to see if any of the models could consistently predict literature K(TOC) values within a factor of 30 (i.e., approximately 1.5 orders of magnitude, or half the range of K(TOC) values). The Raoults Law model and coal tar PP-LFER achieved the sought-after accuracy for all tested compound classes, and are recommended for general, regional-scale modeling purposes. As impacted sediment-porewater distribution models are unlikely to get more accurate than this, this review underpins that the only way to accurately obtain accurate porewater concentrations is to measure them directly, and not infer them from sediment concentrations.

A model assessment of polychlorinated dibenzo-p-dioxin and dibenzofuran sources and fate in the Baltic Sea

The contamination of the Baltic Sea with polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) has resulted in restrictions on the marketing and consumption of Baltic Sea fish, making this a priority environmental issue in the European Union. To date there is no consensus on the relative importance of different sources of PCDD/Fs to the Baltic Sea, and hence no consensus on how to address this issue. In this work we synthesized the available information to create a PCDD/F budget for the Baltic Sea, focusing on the two largest basins, the Bothnian Sea and the Baltic Proper. The non-steady state multimedia fate and transport model POPCYCLING-Baltic was employed, using recent data for PCDD/F concentrations in air and sediment as boundary conditions. The PCDD/F concentrations in water predicted by the model were in good agreement with recent measurements. The budget demonstrated that atmospheric deposition was the dominant source of PCDD/Fs to the basins as a whole. This conclusion was supported by a statistical comparison of the PCDD/F congener patterns in surface sediments from accumulation bottoms with the patterns in ambient air, bulk atmospheric deposition, and a range of potential industrial sources. Prospective model simulations indicated that the PCDD/F concentrations in the water column will continue to decrease in the coming years due to the slow response of the Baltic Sea system to falling PCDD/F inputs in the last decades, but that the decrease would be more pronounced if ambient air concentrations were to drop further in the future, for instance as a result of reduced emissions. The study illustrates the usefulness of using monitoring data and multimedia models in an integrated fashion to address complex organic contaminant issues.

On the use of the partitioning approach to derive Environmental Quality Standards (EQS) for persistent organic pollutants (POPs) in sediments: a review of existing data

A review of experimental data has been performed to study the relationships between the concentration in water, pore water and sediments for different families of organic contaminants. The objective was to determine whether it is possible to set EQS for sediments from EQS defined for surface waters in the Daughter Directive of the European Parliament (COM (2006) 397). The analysis of experimental data showed that even though in some specific cases there is a coupling between water column and sediments, this coupling is rather the exception. Therefore it is not recommendable to use water column data to assess the chemical quality status of sediments and it is necessary to measure in both media. At the moment EQS have been defined for the water column and will assess only the compliance with good chemical status of surface waters. Since the sediment toxicity depends on the dissolved pore water concentration, the EQS developed for water could be applied to pore water (interstitial water); hence, there would be no need of developing another set of EQS. The partitioning approach has been proposed as a solution to calculate sediment EQS from water EQS, but the partitioning coefficient strongly depends on sediment characteristics and its use introduces an important uncertainty in the definition of sediment EQS. Therefore, the direct measurement of pore water concentration is regarded as a better option.

Modelling the long-term fate of polychlorinated dibenzo-p-dioxins and furans (PCDD/Fs) in the Grenland Fjords, Norway

The development and application of a predictive fate model (DIG--Dioxins in Grenland) for polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/F) in the marine environment of the Norwegian Grenland Fjords are described. The objective of the modelling study was to predict long-term future changes in PCDD/F concentrations in the fjord following the cessation of point source emissions. To assess the reliability of the model, the model performance was evaluated by comparing model results to field measurements collected between 1989 and 2001. Model bias (defined as the ratio of median predicted concentration and median observed concentration) for prediction of concentrations for three different PCDD/F congeners (2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), 1,2,3,6,7,8-hexadibenzofuran (HxCDF) and octachlorodibenzofuran (OCDF)) in the fjord was between 0.53 and 24, which was deemed a satisfactory result for models of this type. The model was run to determine the dynamic change in concentrations between 1950 and 2050 and to examine the key fluxes of PCDD/Fs in the fjord. Between 1990 and 2050 sediment concentrations of TCDD, HxCDF and OCDF were predicted to fall at fairly constant but gradually slowing rates to concentrations 52, 98 and 88 times lower, respectively, of their 1990 values. Losses of PCDD/Fs from the bottom sediments in the Frierfjord were predicted to be a combination of sediment burial and net resuspension to the water column. Sediment burial was shown to be relatively more important in the fjord's deep-water sediments, whereas resuspension was relatively more important in the shallow sediments. For the shallower sediments, a net water-to-sediment flux was predicted for all three congeners up until the mid-1970s, when emission reductions were initiated, and thereafter a net sediment-to-water flux was predicted. The shallow sediments acted as net sources to the deeper sediments and to the fishing areas in the outer fjord.