A decision framework for possible remediation of contaminated sediments in the River Kymijoki, Finland

Background, aim, and scope  The paper describes the spatial contamination of the River Kymijoki, South-Eastern Finland, and the coastal region of the Gulf of Finland with PCDD/Fs and mercury. The findings of ecotoxicologial and human health studies are also reported, including environmental and human risk assessments. Sediments from the River Kymijoki, draining into the Gulf of Finland, have been heavily polluted by the pulp and paper industry and by chemical industries. A wood preservative, known as Ky-5, was manufactured in the upper reaches of the river between 1940 and 1984 causing severe pollution of river sediments with polychlorinated dibenzo-p-dioxins (PCDD) and dibenzofurans (PCDF). Moreover, the sediments have been polluted with mercury (Hg) from chlor-alkali production and the use of Hg as a slimicide in pulp and paper manufacturing. Materials and methods  An extensive sediment survey was conducted as well as sediment transport modeling, toxicity screening of sediment invertebrates, and a survey of contaminant bioaccumulation in invertebrates and fish. Studies on human exposure to PCDD/Fs and the possible effects on hypermineralization of teeth as well as an epidemiological study to reveal increased cancer risk were also conducted. An assessment of the ecological and human health risks with a null hypothesis (no remediation) was undertaken. Results  The sediment survey revealed severe contamination of river and coastal sediments with PCDD/Fs and Hg. The total volume of contaminated sediments was estimated to reach 5 × 10⁶ m³ and hot spots with extremely high concentrations (max 292,000 ng g⁻¹ or 1,060 ng I-TEQ g⁻¹ d.w.) were located immediately downstream from the pollution source (approximately 90,000 m³). Sediment contamination was accompanied by changes in benthic assemblages, but direct effects were masked by many factors. The fish showed only slightly elevated PCDD/F levels in muscle, but orders of magnitude higher in the liver compared with reference freshwater sites and the Baltic Sea. The concentrations in human fat did not reveal high human exposure in the Kymijoki area in general and was lower than in sea fishermen. The relative risk for total cancer among farmers was marginally higher (RR = 1.13) among those living close to the river, compared with farmers living further away, and the possibility of increased cancer risk cannot be ruled out. A conservative risk assessment revealed that the present probability of exceeding the WHO upper exposure limit of 4 pg WHO-TEQ kg⁻¹ d⁻¹ for PCDD/Fs and DL-PCBs was 6%. The probability of exceeding the WHO limit value of 0.23 μg kg⁻¹ d⁻¹ for methyl mercury was estimated to be notably higher at 62%. Based on these studies and the estimated risks connected with different remediation techniques a general remediation plan with cost benefit analysis was generated for several sub-regions in the river. Dredging, on-site treatment, and a close disposal of the most contaminated sediments (90,000 m³) was suggested as the first phase of the remediation. The decision regarding the start of remediation will be made during autumn 2008. Conclusions  The sediments in the River Kymijoki are heavily polluted with PCDD/Fs and mercury from earlier chlorophenol, chlor-alkali, and pulp and paper manufacturing. A continuous transport of contaminants is taking place to the Gulf of Finland in the Baltic Sea. The highly increased PCDD/F and Hg levels in river sediments pose an ecotoxicological risk to benthic fauna, to fish-eating predators and probably to human health. The risks posed by mercury exceed those from PCDD/Fs and need to be evaluated for (former) chlor-alkali sites and other mercury releasing industries as one basis for remediation decision making. Recommendations and perspectives  The studies form the basis of a risk management strategy and a plan for possible remediation of contaminated sediments currently under consideration in the Southeast Finland Regional Environment Centre. It is recommended that a detailed restoration plan for the most seriously contaminated areas should be undertaken. Based on current knowledge, the restoration of the whole river is not feasible, considering the current risk caused by the contaminated sediment in the river and the costs of an extensive restoration project. The experiences gained in the present case should be utilized in the evaluation of PCDD/F- and mercury-contaminated sites in other countries. The case demonstrates that the historic reservoirs are of contemporary relevance and should be addressed, e.g., in the national implementation plans of the Stockholm Convention. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

A national inventory to estimate release of polychlorinated dibenzo-p-dioxins and dibenzofurans in Portugal

Taking into account current environmental concerns, the main objective of this work focused a national inventory aiming to estimate the amount of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDD/PCDF) released in Portugal in 2006. The methodology used was based on the Standardized Toolkit for Identification and Quantification of Dioxins and Furan Releases, developed by UNEP Chemicals, in 2005. The method allows the assessment of the amount of PCDD/PCDF released into the environment along five vectors involving air, water, land, products and residues. Facing some difficulties mainly regarding to the availability of data for some activities known to produce PCDD/PCDF, three scenarios (Sc1 to Sc3) corresponding to lower, central and upper estimates were established. The Sc1 scenario (lower estimate) includes the situations where in case of doubt or scarce information, reduced or none emission values were assumed, Sc2 refers to a central estimate, which is believed to be the most realistic for the Portuguese situation, while Sc3 corresponds to the worst case (upper estimate). The results obtained pointed out that the total amount of PCDD/PCDF emitted in Portugal during the period under analysis was in the range of 51.2-217.9 g TEQ year⁻¹, with the most likely value of 95.2 g TEQ year⁻¹ achieved under the Sc2 scenario. This study also showed that the methodology developed by UNEP Chemicals is a very simple one, and the main difficulty is the availability of data. The main indicators calculated in this study were 8.98 μg TEQ/(year person) by taking into account the total amount of PCDD/PCDF released, and 3.63 μg TEQ/(year person) when only air emissions were considered.     

Determination of dibenzo-p-dioxin (DBD) and dibenzofuran (DBF) in paper pulp mill defoamer additives and in paper pulp samples

An analytical procedure has been developed for quantitatively measuring dibenzo-p-dioxin (DBD) and dibenzofuran (DBF) present at part-per-billion levels in paper mill defoamers and paper pulp using gas chromatography coupled with either low resolution mass spectrometry (GC-LRMS) or high resolution mass spectrometry (GC-HRMS).     

Prediction of polychlorinated dibenzofuran congener distribution from gas-phase phenol condensation pathways

A model for predicting the distribution of dibenzofuran and polychlorinated dibenzofuran (PCDF) congeners from a distribution of phenols was developed. The model is based on a simplified chemical mechanism. Relative rate constants and reaction order with respect to phenol precursors were derived from experimental results using single phenols and equal molar mixtures of up to four phenols. For validation, experiments were performed at three temperatures using a distribution of phenol and 19 chlorinated phenols as measured in municipal waste incinerator exhaust gas. Comparison of experimental measurements and model predictions for PCDF isomer distributions and homologue pattern shows agreement within measurement uncertainty. The R-squared correlation coefficient exceeds 0.9 for all PCDF isomer distributions and the distribution of PCDF homologues. These results demonstrate that the distribution of dibenzofuran and the 135 PCDF congeners from gas-phase condensation of phenol and chlorinated phenols can be predicted from measurement of the distribution of phenol and the 19 chlorinated phenol congeners.     

Intra- and inter- laboratory variability in paper industry PCDD/PCDF analysis

Studies were conducted to quantify variability for pulp industry wastewater effluent, pulp and sludge analyses. Intralaboratory variability studies indicated that there is a potential for greater between batch variability in results than for within a batch. Intralaboratory relative standard deviations for replicate analyses ranged from 6% to 60%. Interlaboratory studies, using standard reference materials showed relative standard deviations from 4% to 135%. There appeared to be little dependance of variability on concentration, suggesting matrix effects were very important. In general, effluent variability was greater than observed for pulps or sludges. Analysis of reference standards indicated that 13% to 17% variability can be attributed to differences in calibration standards. A limited study of full congener PCDD/PCDF analysis variability showed it to be greater than observed for TCDD/TCDF.     

Human exposure to dioxins through diet in Italy

We have measured the content of polychlorinated dibenzo-p-dioxins, and polychlorinated dibenzofurans (together defined as “dioxins”) in 269 samples of food of animal origin collected through the regional veterinary services, covering the national territory. Quantification of the dioxins was accomplished by isotope dilution method, and toxic equivalents (TEQ) were calculated. The average daily food intake was obtained from two main sources: national data collected by the National Institute of Nutrition, and data from an ongoing cohort study on diet and cancer including 40, 000 Italian subjects.

The mean value of dioxins measured in food of animal origin was 0.144 ± 0.266 pg-TEQ/g (range: 0.003–1.655 pg-TEQ/g). Fish was the item with the highest content. The estimated intake of dioxins with main food items of animal origin is presented. The major contribution to dioxins intake with food comes from cow milk and fish consumption. These results are in agreement with what observed in studies conducted in other countries, such as Germany, Finland, Japan, Spain, and are below the limits set by the European legislation.     

Results from the first year of the Swedish dioxin survey

PCDD/F levels are presented for food baskets, raw and fried herring, wolf muscle and moose tallow collected from the same area, three burbot liver samples, a sediment core from the Baltic Sea and leach water from three garbage dumps. Levels in food baskets were at or below detection limits making the estimation of intake via the diet difficult. Frying herring produces no net change in PCDD/F levels. Moose have lower PCDD/F levels than wolf from the same area. Levels in burbot liver samples are much higher near industrial discharges than in a more undisturbed area. The top 4 cm of the sediment core contain the highest levels of PCDD/F. Thereafter the levels drop and reach a low level. This low level is probably caused by contamination during sampling and is not evidence of PCDD/F presence far back in time. Leach water from city dumps contains low levels of PCDD/F but that from a dump that takes industrial wastes had higher levels and a different congener pattern.     

Long-term sampling method for polychlorinated dibenzofurans (PCDFs) and dibenzo(p)dioxins (PCDDs) in flue gas of combustion facilities

A long-term sampling method for PCDF/D in emissions from combustion facilities is presented. The sampling of PCDF/D was performed over a four week period. The results are compared to results of four one-week measurements.     

Photodegradation of polychlorinated dibenzo-p-dioxins and dibenzofurans in soil with vegetable oil

GOAL, SCOPE AND BACKGROUND: This study was carried out to investigate the effect of olive oil on the photodegradation of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) in contaminated sawmill soil. Previous studies had shown that the solubility of PCDD/Fs in olive oil is high and a rapid photodegradation of PCDD/Fs takes place in olive oil when irradiated with ultraviolet (blacklight) lamps. The efficiency of this treatment method was evaluated under more practical conditions. These included the use of sunlight irradiation and a lower-grade olive oil, without a preliminary extraction of soil with olive oil. METHODS: A 1-cm layer of contaminated sawmill soil was blended with 20 weight-% of olive oil and exposed to sunlight for four weeks. In another experiment, a new dose of olive oil was added at the middle of the exposure period. The PCDD/F concentrations of the soils were monitored periodically. RESULTS AND DISCUSSION: A reduction in the concentration of 2,3,7,8-chlorinated PCDD/Fs by 59% and in WHO-TEQ in contaminated sawmill soil by 48% was attained after blending the soil with two doses (20 + 20%) of olive oil and exposing the mixture to sunlight for four weeks. Photodegradation with only one dose of olive oil was less efficient. This suggests that periodical additions of olive oil would be needed to maintain a proper degradation rate. After the oil additions, the WHO-TEQ content of the soil declined with first order reaction half-lives of 19.2 to 19.7 d. The overall half-life during the four-week treatment, however, was 30 d. CONCLUSION: A significant reduction in the PCDD/F concentration of aged sawmill soil can be achieved with a relatively simple olive oil-sunlight treatment. RECOMMENDATIONS AND OUTLOOK: Some theoretical and technological questions need to be solved before using the investigated soil decontamination method in larger-scale applications. The functions of vegetable oils in photodegradation processes should be studied in more detail. The amount of oil that is needed for a proper solubilisation and photodegradation of PCDD/Fs should be minimised. Moreover, special care should be taken to prevent mobilisation of PCDD/Fs to the surrounding environment and to avoid leaving bioavailable residuals of PCDD/Fs in soil.