Dioxin concentrations in sediments of the Baltic Sea – A survey of existing data

www.elsevier.com/locate/chemosphere

Chemosphere 67 (2007) 1762–1775

Dioxin concentrations in sediments of the Baltic Sea – A surveyof existing data

Matti Verta a,*, Simo Salo a, Markku Korhonen a, Timo Assmuth a, Hannu Kiviranta b,Jaana Koistinen b, Paivi Ruokojarvi b, Pirjo Isosaari b,1, Per-Anders Bergqvist c,

Mats Tysklind c, Ingemar Cato d, Jørgen Vikelsøe e, Martin M. Larsen e

a Finnish Environment Institute (SYKE), P.O. Box 140, FI-00251 Helsinki, Finlandb National Public Health Institute, P.O. Box 95, FI-70701 Kuopio, Finland

c Environmental Chemistry, Department of Chemistry, Umea University, SE-901 87 Umea, Swedend Geological Survey of Sweden, P.O. Box 670, SE-751 28 Uppsala, Sweden

e National Environmental Research Institute, P.O. Box 358, DK-4000 Roskilde, Denmark

Accepted 26 May 2006Available online 16 January 2007

Abstract

Recent survey results for polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs, dioxins) in Baltic Sea sediments from Fin-land, Sweden and Denmark were merged with previously published Baltic Sea data. Regional distribution of concentration levels, dif-ferences in congener patterns, and temporal changes in sediment profiles were examined. One of the main objectives was to study if anymajor point sources for different PCDD/F congeners could be identified on a regional scale, based on sediment records. The survey con-firmed the impact of chlorophenol production derived highly chlorinated PCDF-congeners on the total toxicity in sediments in the Gulfof Finland near the Kymijoki river estuary. Signatures of other point sources or combined point sources pertinent to specific industrybranches or particular production processes (such as pulp bleaching, vinyl chloride production, thermal processes) may be discerned.However, the findings did not support any of the known point sources significantly influencing those congeners that are most abundantin Baltic herring and salmon. Instead, regional distributions in the Baltic Sea indicate that atmospheric deposition may act as a majorsource for those congeners and especially for 2,3,4,7,8-PeCDF. There were clear indications of declines in levels in sediment in someareas, but generally the levels of highly chlorinated PCDD/Fs on the northern coast of the Gulf of Finland were still high when comparedwith other areas of the Baltic Sea. Major areas with data gaps cover the south-eastern and eastern coastal regions of the Baltic Properand the southern Gulf of Finland.� 2006 Elsevier Ltd. All rights reserved.

Keywords: Polychlorinated dibenzo-p-dioxins; Dibenzofurans; Baltic Sea; Sediments; Regional distribution

1. Introduction

The Baltic Sea region is one of the areas most contami-nated by persistent organic pollutants (POPs) includingpolychlorinated dibenzo-p-dioxins and dibenzofurans(PCDD/Fs). The high load of dioxins in fish has lead to

0045-6535/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.chemosphere.2006.05.125

* Corresponding author. Tel.: +358 20 4902407; fax: +358 20 4902490.E-mail address: [email protected] (M. Verta).

1 Present address: VTT Technical Research Centre of Finland, P.O. Box1000, FI-02044 VTT.

recommendations to restrict the use of contaminated fishfrom the Baltic for human consumption. Uncertaintiesabout sources, the geographical distribution of these con-taminants, their cycling and fate in the sea, the pathwaysof bioaccumulation and possible ecotoxicological andhuman health effects are of concern.

Basic knowledge of the sources of PCDD/Fs found in theBaltic Sea is scarce. The inventories and estimates of emis-sions of PCDD/Fs are severely limited and the comparisonof different approaches and their results is difficult. Theemission factors used are in many cases based on few

mailto:[email protected]

Table 1The source and the age of the surface sediment data under consideration

Region Samplingyear

Sampling depth(cm)

Reference

Gulf of Finland 1993 0–2 Koistinen et al.(1997)

1997–1998 0–1 Isosaari et al.(2002a)

1996–2003 0–2 SYKE

Gulf of Bothnia <1989 Rappe et al. (1989)1991 0–2 Naf et al. (1997)1993 0–2 Koistinen et al.

(1995)2002–2003 0–2 SYKE2002–2003 0–1 (–3) UMEA University

Baltic Proper 1988 0–2 Kjeller and Rappe(1995)

1993 0–2 Koistinen et al.(1997)

1998–2002 0–1 (–3) UMEA University

Baltic Propersouth west

1994 0–2 Witt et al. (1997)

Belt Sea 1998–2003 0–2 NERIKattegat 1998–2003 0–2 NERI

2000–2001 0–1 (–3) UMEA University

For regions of the Baltic Sea, see Fig. 3.

M. Verta et al. / Chemosphere 67 (2007) 1762–1775 1763

samples and probably not representative of fluxes. In addi-tion, information on activity rates and emission profiles isscarce (Wenborn et al., 1999) although for some processesincreasingly representative and reliable data have been pub-lished (see discussion below). Furthermore, more impor-tance should be attached to cumulative emissions overtime and, as part of them, secondary emissions from earlierfluxes and pools, e.g. soils, solids and products and wastes(see e.g. Bergqvist et al., 2005; Assmuth and Jalonen, 2005).

In general, PCDD/Fs enter the Baltic Sea via atmo-spheric deposition, river input and point sources along thecoast. The ultimate sink for the majority of the compoundsis the open sea and coastal sediments, although some frac-tion enters the food chain. Importantly, vertical profilesof sediments reveal past changes in PCDD/F accumula-tion and inputs, although modifying factors confoundsuch reconstructions. Therefore, analysis of sediment hasbeen widely used to study regional and temporal trends ofdioxin pollution in freshwater, estuaries and oceans (e.g.,Koistinen et al., 1995; Vartiainen et al., 1995, 1997; Bruck-meier et al., 1997; Witt et al., 1997; Sakurai et al., 2000; Frig-nani et al., 2001; Green et al., 2001; Wu et al., 2001; Isosaariet al., 2002a,b; Marvin et al., 2002; Gotz and Lauer, 2003;Barabas et al., 2004). The Baltic Sea has been one of themost studied sea areas for dioxin-like compounds as well.Only a fraction of the analytical results have been publishedin scientific papers, however (Rappe and Kjeller, 1987;Rappe et al., 1989; Jonsson et al., 1993; Kjeller and Rappe,1995; Dannenberger et al., 1997; Koistinen et al., 1997; Wittet al., 1997; Isosaari et al., 2000, 2002b, see also Ishaq et al.,2003). Here we summarize and evaluate information onregional distributions of certain congeners of dioxins in sur-face sediments and in six sediment cores from the Baltic Sea.New survey data from Finland, Sweden and Denmark ismerged and compared with data of earlier sediment surveysand case studies.

The purpose of this paper is to

1. Provide a ‘‘draft’’ picture of the regional distribution ofkey PCDD/Fs (possible hot spots, major regional differ-ences) in surface sediments along the Finnish–Swedish–Danish–German coastal and open sea sites in the Baltic.

2. Analyze differences in congener distributions (sourceidentification).

3. Examine temporal changes reflected in sedimentprofiles.

4. Identify major areas with data gaps.

2. Materials and methods

2.1. Sampling

2.1.1. Finland

Surface sediment samples along the Finnish coast, onesediment profile from the Gulf of Finland (AKL) andone sediment profile from the previously poorly studied

Gulf of Bothnia (SR5) were taken by the Finnish Environ-ment Institute (SYKE) and the Finnish Marine ResearchInstitute in 1997–2003 (Table 1). The samples were col-lected from known sedimentation areas both near the coastand offshore using pistonless corers with an inner diameterfrom 9.0 to 13 cm. All the sediment cores were divided ver-tically in the field to 2 cm slices. The surface (0–2 cm,2–4 cm) and deep (35–50 cm) samples as well as the twoprofiles were analyzed for dioxin concentrations.

2.1.2. DenmarkThe Danish samples were collected within NOVA, the

Danish Aquatic Environment Monitoring and AssessmentProgramme, 1998–2003. The programme includes a marinesubprogramme involving surface sediment in coastal areasand fjords, Kattegat, and occasionally the Baltic Sea.

2.1.3. Sweden

The Swedish samples were mainly collected by the Geo-logical Survey of Sweden between 1998 and 2003 as a partof the marine surface sediment screening program which isongoing since the end of the 1980s. By 2004 more than 500samples had been taken from Swedish waters for chemicalanalysis.

All surveys have tended to focus sampling on eitherknown accumulation sites of recent sediments (Isosaariet al., 2002a,b; SYKE 2000–2003 sampling) and/or havemade detailed sampling protocols and sediment control(Sweden: Geological Survey of Sweden; Denmark: NationalEnvironmental Research Institute, Kaas and Markager,1998). In general, acoustic sounding was used in site

1764 M. Verta et al. / Chemosphere 67 (2007) 1762–1775

selection, except for Denmark. For PCDD/Fs, the focusingof sampling on sedimentation areas with abundant organiccarbon (OC) was particularly important (see below).

For reference, data from published scientific literature isalso given (Table 1). When comparing the results, one shouldnote that the sediments were collected over a long period(1988–2003) and that the concentrations at surface (and con-centration changes in profiles) reflect the concentrations atthe time of sampling. For example, some of the Swedishand all the German sediments were taken earlier (late1980s–early 1990s) than Finnish/Danish/Swedish samples(late 1990s–2003).

2.2. Analytical methods

The concentrations of PCDD/Fs in Finnish sedimentsamples in studies by Isosaari et al. (2002a) and SYKE(Table 1) were analyzed in the Laboratory of Chemistryat the National Public Health Institute. The analyticalmethod has been described previously in the study by Isosa-ari et al. (2002a). In short, 1.5 g samples of lyophilized andhomogenized sediment were Soxhlet-extracted with tolu-ene. The extracts were fractionated and purified by elutingthem through three columns consisting of sodium sulfateand silica gel, activated carbon and Celite, and aluminumoxide. Sulfur was precipitated from the sediments withactivated copper powder. The quantification of PCDD/Fswas by gas chromatography-high-resolution mass spec-trometry (GC-HRMS). The PCDD/Fs determined includedall the 17 toxic (2,3,7,8-substituted) PCDD/F congeners.The laboratory is an accredited testing laboratory (No.T077) in Finland (current standard: EN ISO/IEC 17025)and the scope of accreditation includes PCDD/Fs fromenvironmental samples.

The Danish analytical method for PCDD/F and c-PCBsadapted from the European standard EN-1948 2-3. The sed-iment sample was air-dried, soxhlet-extracted in toluene,cleaned-up by column chromatography on silica gel/NaOH,silica gel/H2SO4, acidic Al2O3, active carbon AX-21/Celite.Analysis was carried out by GC/MS at resolution 10000.

Swedish samples were analyzed according to a methodbased on EPA. Sediments were air-dried and extracted usingMAE (microwave assisted extraction) in mixture of tolu-ene:methanol 9:1. Extracts were cleaned and fractionatedby column chromatography (combined-silica column/alu-mina column/activated carbon/Celite column – Grabicet al., 2000). Final fractions were analyzed by GC/HRMSat resolution 10000–12000. DB5 ms column (30 m length ·0.25 mm ID · 0.25 lm film) was used for chromatographicseparation of PCDD/F. DB-17 (30 m length · 0.25 mmID · 0.25 lm film) was used as column with complementarychromatographic separation of PCDD/F.

2.3. Statistical analysis

Pearson bivariate correlation coefficients were calculatedfor seven congeners studied in this paper. The analysis was

performed by the SPSS (Version 11.5.1) statistical software(SPSS Inc.). All the surface sediment samples were includedin the analysis, except the sample from Skoldvik due to thehigh PCDD/F level and its dominative impact on the cor-relation coefficient of some congeners.

3. Results and discussion

The main interest was to study the spatial distribution ofdioxins and the contribution of identified point sources inFinland and Sweden to sediment dioxin levels. For thispurpose, we focused on certain point source-identifiedcongeners. The selected congeners reflecting specific pointsources were 2,3,7,8-TCDD (TCDD), 2,3,7,8-TCDF(TCDF), and OCDF as indicators of pulp bleaching andvinyl chloride production (Rappe et al., 1989; Koistinenet al., 1992; Isosaari et al., 2000), and 1,2,3,4,6,7,8-HpCDFas an indicator of chlorophenol production (Verta et al.,1999, 2003; Isosaari et al., 2002a). OCDD is the main con-gener in most air samples (e.g. Lohmann and Jones, 1998and references therein), an impurity in technical pentachlo-rophenol and produced photochemically and in thermalreactions from pentachlorophenol (e.g., Vikelsøe andJohansen, 2000; Masunaga et al., 2001; Stevens et al.,2003, see also Schramm et al., 1995; Eduljee, 1999; Bakerand Hites, 2000). 2,3,4,7,8-PeCDF followed by 1,2,3,7,8-PeCDD makes the main contribution to the sum I-TEQin air and deposition (Lohmann and Jones, 1998; Vulykhand Shatalov, 2001; Korhonen et al., 2006) and are themain contributors to TEQ in Baltic fish (Baltic herring, sal-mon) as well (Kiviranta et al., 2003). The former is themain contributor from PCDD/Fs, along with coplanarPCBs, to WHO-TEQs in fish-consuming mammals, includ-ing humans (Kiviranta et al., 1999, 2001, 2002; Wingforset al., 2000; Glynn et al., 2001) and harbour porpoises(Berggren et al., 1999) in the Baltic Sea region. One ofthe main objectives was to find indications of whetherany major point source for these two congeners can beidentified or discounted on the basis of sediment records.

3.1. Spatial distribution in surface sediments

The spatial distribution of PCDD/F congeners in BalticSea surface sediments has been compiled in Table 2 and isshown for five of the congeners together with I-TEQ ofPCDD/Fs in Fig. 1. Clearly highest PCDD/F levels asI-TEQ have been reported along the Finnish coast of theGulf of Finland (Isosaari et al., 2000, 2002a). Two identi-fied point sources are the vinyl chloride (VC) productionplant at Skoldvik, and the chlorophenol mixture (Ky-5)production along the Kymijoki River. The predominatingchlorophenols in Ky-5 were 2,3,4,6-tetrachlorophenol,2,4,6-trichlorophenol and pentachlorophenol. The contentof PCDD/Fs, which occurred as impurities, was as highas 860 lg I-TEQ kg�1 (Vartiainen et al., 1995). Both sourceshave been closed but a continuous transport of old contam-inated sediments from the Kymijoki river (44 g I-TEQ in

Table 2The concentration of selected PCDD/F congeners, sum of toxic congeners and the total PCDD/F sum (pg g�1 dw) as well as I-TEQ (pg I-TEQ g�1 dw) in the sediment surface sampled from differentsites at each region of the Baltic Sea

Site I-TEQ 2,3,4,7,8-PeCDF

1,2,3,4,6,7,8-HpCDF

2,3,7,8-TCDD

2,3,7,8-TCDF

1,2,3,7,8-PeCDD

OCDD OCDF Sum Toxic sum Source of data

Gulf of Bothnia (GUB)

Oulu 3.07 1.31 43.00 0.17 1.84 0.56 16.80 43.1 250 125 SYKEKokkola 11.60 8.12 99.10 <0.1 2.52 0.75 34.90 212 714 442 SYKEVaasa 3.15 2.09 28.10 <0.1 1.89 0.74 49.90 33.3 216 142 SYKEKaskinen 5.49 2.63 65.40 0.42 3.36 1.39 47.40 73.0 369 227 SYKEPori 1.61 0.69 64.30 <0.1 0.68 0.19 10.10 69.7 229 153 SYKEBO3 14.40 12.40 61.00 0.30 5.39 2.91 108.00 91.2 727 376 SYKEUS5B 7.66 5.46 80.40 0.40 4.73 1.54 63.30 115 498 315 SYKESR5 8.37 6.59 38.10 0.47 6.01 1.79 72.60 58.9 422 237 SYKES3 64.00 <10 3200.00 8.00 25.00 4.00 490.00 2100.00 Koistinen et al. (1995)S8 30.00 <10 310.00 3.00 20.00 8.00 280.00 200.00 Koistinen et al. (1995)HF 6.32 3.58 36.04 0.27 3.79 1.62 50.42 76.28 Naf et al. (1997)UM 0.97 0.55 8.69 0.09 0.26 0.18 15.58 8.03 Naf et al. (1997)HL 4.65 2.79 62.41 0.20 3.21 1.00 115.09 114.76 Naf et al. (1997)GB 7.81 3.28 42.76 0.55 11.45 1.00 142.48 163.03 Naf et al. (1997)SN 5.53 3.07 27.07 0.32 4.20 1.33 134.82 58.96 Naf et al. (1997)Iggesund 30 km 14.74 5.70 1.90 11.00 3.60 96.00 5.1 661 145 Rappe et al. (1989)Iggesund 16 km 21.33 6.20 3.30 20.00 5.30 100.00 <1.5 826 173 Rappe et al. (1989)Iggesund 8 km 86.70 6.30 7.30 48.00 16.00 250.00 <6.6 2775 713 Rappe et al. (1989)Iggesund 4 km 116.51 7.40 26.00 210.00 29.00 330.00 15 2027 706 Rappe et al. (1989)UmU084 18.41 9.07 172.00 1.75 29.40 1.32 278.00 821.00 2324 1439 Umea UniversityUmU085 15.76 8.88 211.00 1.47 17.80 <2.8 127.00 338.00 1650 808 Umea UniversityUmU086 58.38 6.88 112.00 4.48 70.50 8.22 265.00 178.00 2541 1135 Umea UniversityUmU087 4.78 3.23 48.10 <1.0 9.94 <1.1 81.50 68.00 498 249 Umea UniversityUmU088 1.09 1.27 11.10 <0.48 1.45 <0.58 25.50 10.70 112 55 Umea University

Baltic Proper (BAP)

P18 22.23 16.40 87.00 1.00 13.10 3.47 273.00 73.00 1454 Kjeller and Rappe (1995)BY15 50.00 19.00 118.00 <2 42.00 5.00 265.00 57.50 Koistinen et al. (1997)UmU078 5.56 4.06 47.10 <0.51 6.15 <0.61 404.00 223.00 1043 750 Umea UniversityUmU079 19.06 13.10 33.10 2.86 13.90 6.49 220.00 48.00 988 436 Umea UniversityUmU080 28.34 20.50 225.00 1.10 21.50 3.62 1617.00 430.00 3884 2596 Umea UniversityUmU081 20.72 23.60 104.00 <0.94 21.70 <1.1 676.00 146.00 1870 1130 Umea UniversityUmU082 45.87 28.40 373.00 1.81 42.70 6.61 2373.00 638.00 5876 3912 Umea UniversityUmU083 7.10 7.62 37.00 <0.73 7.48 <0.93 107.00 43.30 524 248 Umea University

Gulf of Finland (GUF)

AKL 242.22 7.44 17094.67 1.74 13.16 7.31 393.66 17492.02 52891 35487 SYKEUYK17 71.52 15.21 3360.00 3.02 13.05 9.97 139.60 3460.00 11800 7270 SYKEKAS5 110.10 15.18 5540.00 4.20 21.42 13.15 157.80 6250.00 20200 12400 SYKEKAS8 190.10 23.66 9040.00 6.84 32.80 27.56 93.19 7450.00 30000 17500 SYKEUYK15 20.62 8.80 510.80 3.97 6.56 2.32 51.32 571.80 2230 1240 SYKEKYVY8 23.17 12.31 457.20 2.35 13.01 5.72 64.46 563.00 1990 1200 SYKESNP3 14.01 7.53 293.10 1.89 3.47 1.19 56.16 426.70 1350 864 SYKE

(continued on next page)

M.

Verta

eta

l./

Ch

emo

sph

ere6

7(

20

07

)1

76

2–

17

75

1765

Table 2 (continued)

Site I-TEQ 2,3,4,7,8-PeCDF

1,2,3,4,6,7,8-HpCDF

2,3,7,8-TCDD

2,3,7,8-TCDF

1,2,3,7,8-PeCDD

OCDD OCDF Sum Toxic sum Source of data

UYK10 13.28 7.05 165.30 2.95 4.75 3.47 82.59 256.50 1030 577 SYKEUYK9 10.10 7.02 104.00 0.52 4.69 2.85 7.28 98.90 556 278 SYKEUYK7 7.96 5.67 108.00 0.22 4.35 1.96 10.00 83.90 496 258 SYKEKAS2 55.00 16.60 2170.00 1.19 15.50 10.30 70.00 1840.00 7580 4350 SYKESNP17 8.93 <2 570.00 <2 <2 <2 127.00 630.00 1920 1380 SYKESkoldvik 9617 20.98 6.13 424.36 1.07 6.13 4.42 66.47 1465.00 2656 2103 Isosaari et al. (2000)Skoldvik 648.00 131.22 2928.00 322 108.40 112.02 1593.00 22144.00 38200 29553 Isosaari et al. (2000)K15 116.00 9.40 7570.00 1.41 12.16 9.94 155.40 6 450.0 24165 14529 Isosaari et al. (2002a)LL3a 22.80 10.95 732.80 0.36 7.85 3.47 102.30 777.8 2900 1740 Isosaari et al. (2002a)JML1b 9.73 7.10 104.90 0.49 6.44 1.52 91.75 94.96 606 365 Isosaari et al. (2002a)K19 52.72 13.83 2820.00 0.67 10.73 6.26 132.20 2880.00 9740 6020 Isosaari et al. (2002a)K20 39.37 11.47 1830.00 0.45 10.46 5.94 101.10 1670.00 6174 3770 Isosaari et al. (2002a)KAII 35.26 8.73 1910.00 0.22 7.26 4.02 87.49 1420.00 5740 3550 Isosaari et al. (2002a)XV-1 31.72 12.85 1240.00 0.51 9.94 3.93 107.40 1240.00 4290 2740 Isosaari et al. (2002a)F41 13.37 9.43 183.00 0.53 10.32 2.83 63.86 140.80 810 470 Isosaari et al. (2002a)F40 10.01 8.41 75.13 0.43 10.98 2.05 48.57 61.53 425 247 Isosaari et al. (2002a)SL2s 9.67 8.41 83.54 0.27 10.47 2.10 68.37 112.50 538 323 Isosaari et al. (2002a)SP2 14.00 11.00 171.00 0.73 13.70 2.56 105.00 129.00 863 490 Isosaari et al. (2002a)SP4 13.00 9.68 169.00 0.69 11.40 2.27 88.00 125.00 795 464 Isosaari et al. (2002a)SP7 27.20 14.30 740.00 0.78 14.20 4.92 134.00 516.00 2590 1540 Isosaari et al. (2002a)SP10 26.30 13.50 707.00 0.56 12.10 4.39 129.00 720.00 2860 1710 Isosaari et al. (2002a)SP19 24.10 13.40 616.00 0.59 12.30 4.09 131.00 581.00 2410 1470 Isosaari et al. (2002a)LL7 26.00 10.50 575.00 <1 15.50 3.50 125.00 480.00 Koistinen et al. (1997)

Baltic Proper, South-Western (BAPsw)

109 8.18 7.01 0.13 759 Witt et al. (1997)69 11.74 10.19 0.15 0.54 918 Witt et al. (1997)KB90 1.17 0.22 0.10 527 Witt et al. (1997)RB10 6.05 6.21 0.07 450 Witt et al. (1997)GB19 1.87 1.40 0.06 225 Witt et al. (1997)131 6.98 6.57 0.10 658 Witt et al. (1997)P48 7.11 4.10 n.n. 1264 Witt et al. (1997)KHM 9.26 7.63 0.08 1649 Witt et al. (1997)KHJ 7.39 5.54 0.16 1332 Witt et al. (1997)

Kattegat (KAT)

Oresund, St431 13.73 2.99 30.52 1.40 1.41 65.39 34.6 NERIKattegat, St905 20.10 9.05 55.47 8.64 1.76 198.90 82.8 NERICopenhagen harbour 43.38 26.45 272.81 1.80 27.67 5.40 1588.88 378.5 NERIOresund 4.16 1.05 9.57 0.10 1.66 45.02 9.8 NERIUmU074 13.54 10.00 80.40 <0.58 7.24 1.62 321.00 183.00 1275 727 Umea UniversityUmU075 12.28 8.91 70.00 <0.79 6.37 1.94 278.00 164.00 1093 641 Umea UniversityUmU076 6.51 5.89 39.40 <0.62 4.78 <0.80 173.00 109.00 663 392 Umea UniversityUmU077 39.68 41.20 32.20 <0.47 52.70 <0.72 382.00 78.40 1208 801 Umea University

1766M

.V

ertaet

al.

/C

hem

osp

here

67

(2

00

7)

17

62

–1

77

5

Bel

tS

ea(

BE

S)

2215

.96

11.8

70.

020.

3822

06W

itt

etal

.(1

997)

WB

110

.04

0.62

0.01

2750

Wit

tet

al.

(199

7)W

B3

3.98

2.31

0.04

0.32

427

Wit

tet

al.

(199

7)U

W2

16.8

70.

350.

030.

0746

25W

itt

etal

.(1

997)

UW

319

.77

1.58

n.n

.57

94W

itt

etal

.(1

997)

UW

48.

021.

910.

0914

79W

itt

etal

.(1

997)

DB

192.

060.

340.

0446

0W

itt

etal

.(1

997)

Mec

kle

nb

urg

bay

35.8

014

.55

90.7

30.

5014

.90

2.87

177.

6915

0.7

NE

RI

Od

ense

fjo

rd11

.34

1.72

25.2

50.

402.

6847

.90

8.6

NE

RI

Ho

rsen

s,h

arb

ou

r8.

560.

6410

.47

0.33

6.50

NE

RI

Ho

rsen

sfj

ord

7.32

7.20

15.1

34.

681.

4123

1.03

21.3

NE

RI


2001) is still a major source of dioxins to the Gulf of Fin-land (Verta et al., 2003). Both these point sources haveemitted in particular highly chlorinated dibenzofurans(OCDF, 1,2,3,4,6,7,8-HpCDF, and to smaller extentHxCDFs and PeCDFs), and one of the sources (the VCplant at Skoldvik) emitted 2,3,7,8-TCDF and 2,3,4,7,8-PeCDF as well (Isosaari et al., 2000). For Skoldvik, peakconcentrations occur in near vicinity (few hundred metersfrom sewage outlet) of the source only, and in two sam-pling points near the source a difference by the factor of7 to over 300 depending on the congener is seen (see sitesSkoldvik and Skoldvik 9617 in Table 2). OCDF contami-nation has also been linked with VC production in the Ven-ice Lagoon (Frignani et al., 2001). It may be noted thatpolyvinyl chloride production along with chloralkali pro-duction has caused PCDD/F emissions also to Kattegat(Bergqvist et al., 2005).

Other identified point sources are located along the Gulfof Bothnia in the vicinity of pulp mills at Iggesund, Sweden(Rappe et al., 1989) and Pietarsaari (Finland) (Sites S3 andS8, Koistinen et al., 1995) and are notable only slightly asI-TEQ, but especially in elevated levels of TCDF (Igges-und). Also chloralkali industries and steel mills in bothSweden and Finland are among the point sources thatmay have emitted significant amounts of PCDD/Fs tothe Baltic, especially Bothnian Sea (Rappe et al., 1991;Rappe, 1992; Bergqvist et al., 2005). The congener patternfrom steel mills is very similar to those from copper smelt-ers and municipal solid waste incineration (Buekens et al.,2000, 2001) with PCDFs dominating over PCDDs. Thesame is true for sintering and melting furnaces in the ironand steel foundries and non-ferrous metal smelters inJapan (Kim et al., 2003a,b).

The ‘chloralkali pattern’ in emissions contains particu-larly large amounts of PCDFs including higher chlorinatedcongeners, along with TCDF that dominates the ‘bleachingpattern’ in pulp and paper industry (Rappe et al., 1990, cf.Wu et al., 2001). Two of the sites along the Finnish Gulf ofBothnia coast were situated near potential chloralkali emis-sion sources (Oulu, Pori). Neither of these showed anysigns of increased contamination at surface (Table 2). Itmay be that most PCDD/Fs, if emitted, have been trans-ported in wastes. Besides contamination from these sourceshave been drastically reduced since discontinuation of theprocess late 1960s.

The survey revealed indications of point sources inthe Stockholm archipelago (2,3,4,7,8-PeCDF, stationsUm080-082) and Kattegat area (2,3,4,7,8-PeCDF andTCDF, station UmU077 near Gothenburg and Copenha-gen). PCDD/Fs in Danish fjord and sea sediments variesfrom 4 to 36 pg I-TEQ g�1 dw (Øresund and Mecklen-burger Bay, respectively). In harbour sediment values, upto 43 pg I-TEQ g�1 dw were found. There is no indicationof any major dioxin load from the St. Petersburg area(Fig. 1). No data was available from the eastern coast ofthe Baltic Sea (Poland, Kaliningrad, Lithuania, Latvia,Estonia). Note also that data of some congeners are

Fig. 1. The levels of selected congeners (pg g�1 dw) and I-TEQ (pg I-TEQ g�1 dw) in sediment surface layer at sampling sites in the Baltic Sea (samplestaken before and after 1995 presented as open and bold columns).


missing in German surveys (Dannenberger et al., 1997;Witt et al., 1997).

For 2,3,4,7,8-PeCDF which is the most abundant conge-ner in Baltic fish (e.g. Rappe et al., 1985; Bergqvist et al.,1989; Vuorinen et al., 2002; Kiviranta et al., 2003), thereis only one anomalously high local sediment concentra-tion at Skoldvik sewage outlet (131 pg g�1 dw or 65 pgI-TEQ g�1 dw), but no other findings of highly elevatedlevels (Table 2). Interestingly, many of the lowest concen-trations for both 2,3,4,7,8-PeCDF and 1,2,3,7,8-PeCDDin surface sediments were found in the Gulf of Bothnia,i.e. within the area of highest concentrations in Baltic her-ring of certain weight or age (Kiviranta et al., 2003;SCALE, 2004). The mean concentration (±standard devia-tion) of 2,3,4,7,8-PeCDF was 10.7 ± 2.7 pg g�1 dw in 22surface sediment samples that were collected from the Gulfof Finland in 1996–1997 (Isosaari, 2004). Other identified

possible point sources show only slightly elevated concen-trations compared for instance with sediments northwestand south-east of Gotland (Fig. 1). From all five congenerspresented in Fig. 1, that indicated to derive from atmo-spheric emissions and deposition, namely 2,3,4,7,8-PeCDF,clearly shows the most even distribution between differentstudy sites (except for Skoldvik).

The comparison of spatial levels has been made on thebasis of concentrations in dry sediment used in most stud-ies. Therefore, the results may be biased to some extent bydifferent organic matter concentrations in different surveys.Most POPs including PCDD/Fs are bound to organicmatter (and black soot particles, see e.g. Persson et al.,2002) in sediments. Sediments with high organic mattercontent will have higher PCDD/F concentrations per dryweight. Normalization with organic carbon should there-fore be made in order to improve spatial comparability


of the data, as shown for PCBs in the Baltic by Axelmanet al. (2001). Organic matter content is, however, only sel-dom reported for individual samples and thus the datadoes not allow normalization. Since the surveys have beendirected to known sites of sedimentation, the organic mat-ter content is supposed to vary much less than with ran-dom sampling and, consequently, bias due to organicmatter should not have a major effect on spatial distribu-tion in Fig. 1.

Table 3Pearson bivariate correlation coefficients between concentrations of selected c

I-TEQ 2,3,4,7,8-PeCDF

2,3,4,7,8-PeCDF Correlation 0.366**

Sig. (2-tailed) 0.000N 88

1,2,3,4,6,7,8-HpCDF Correlation 0.948** 0.141Sig. (2-tailed) 0.000 0.252N 68 68

2,3,7,8-TCDD Correlation 0.476** 0.067Sig. (2-tailed) 0.000 0.577N 72 72

1,2,3,7,8-PeCDD Correlation 0.764** 0.332**

Sig. (2-tailed) 0.000 0.002N 83 83

2,3,7,8-TCDF Correlation 0.419** 0.270*

Sig. (2-tailed) 0.000 0.022N 72 72

OCDD Correlation 0.159 0.566**

Sig. (2-tailed) 0.182 0.000N 72 72

OCDF Correlation 0.870** 0.136Sig. (2-tailed) 0.000 0.259N 71 71

The whole data under consideration excluding the sample originated from Sk* Correlation is significant at the 0.05 level (2-tailed).

** Correlation is significant at the 0.01 level (2-tailed).

12378-PeCDD (pg g-1)

2347

8-P

eCD

F (

pg

g-1

)

GUB

GUF

BAP

BAPs

KAT

BES

0 10 20 30

0

10

20

30

40

AKL

UYK17KAS5

KAS8

KAS2

K15

S8

Kbh, Havn

Iggesund 8km

Iggesund 4km

UmU077

UmU081

UmU082

UmU086

a

Fig. 2. Scatter plot of 2,3,4,7,8-PeCDF concentration versus 1,2,3,7,8-PeCDconcentration (b) sorted by subregion of the Baltic Sea (for abbreviations seeUmU082) near Stockholm, Skoldvik and Copenhagen harbour are excluded i

3.2. Congener pattern as an indicator of source

As shown in Table 3, only some of the selected congenerscorrelated positively with each other in surface sediments.Accordingly, high concentration of one congener at onesite would not be an indication of high concentration ofall PCDD/Fs but only for certain other congenerswith the same probable origin. The most significant cor-relation was between 1,2,3,4,6,7,8-HpCDF and OCDF

ongeners in surface sediment in the Baltic Sea

1,2,3,4,6,7,8-HpCDF

2,3,7,8-TCDD

1,2,3,7,8-PeCDD

2,3,7,8-TCDF

OCDD

0.384**

0.00264

0.617** 0.785**

0.000 0.00065 70

0.134 0.896** 0.700**

0.275 0.000 0.00068 68 69

0.019 0.094 0.097 0.249*

0.877 0.448 0.426 0.03568 68 69 72

0.993** 0.094 0.392** 0.010 0.0350.000 0.447 0.001 0.931 0.769

67 68 69 71 71

oldvik.

2347

8-P

eCD

F (

pg

g-1

)

w

OCDD (pg g-1)

0 100 200 300 400

0

10

20

30

40

AKL

KAS8

KYVY8

UYK9

UYK7

KAS2

SNP17Kokkola

S3

Iggesund 4km

UmU074

UmU077

UmU078

b

D concentration (a) and 2,3,4,7,8-PeCDF concentration versus OCDDTable 2). Note that four sites with high OCDD concentration (UmU080,n (b).


(r = 0.993), obviously because of the same source (Ky-5) inalmost all of the samples in the Gulf of Finland and the lackof this source in most other regions. Also 2,3,7,8-TCDD

Table 4Concentrations of selected PCDD/F congeners in sediment profiles (pg g�1, d

Site Sedim.layer

1,2,3,4,6,7,8-HpCDF

1,2,3,7,8-PeCDD

2,3,4,7,8-PeCDF

K15 S 7570 9.94 9.40M 28500 53.95 27.27D 763 1.17 4.34

AKL S 17095 7.31 7.44M 23265 5.29 6.99D 103 0.00 0.00

LL3a S 733 3.47 10.95M 2660 9.14 17.72D 442 1.41 6.32

JML1b S 105 1.52 7.10M 274 5.01 16.25D 8 0.90 1.38

SR5 S 38 1.79 6.59M 51 2.58 9.76D 2 0.15 0.61

P18 S 98 3.47 16.40M 106 5.23 19.90D 3 0.19 0.86

S = surface sediment, M = layer with maximum concentration of I-TEQ, D =

0 100

2˚0'0"E

12˚0'0"E

12˚0'0"E

53˚

0'0

"N5

8˚0

'0"N

63˚

0'0

"N

Poland

Denmark

Baltic ProperSouth-West

Katteg

at

Belt Sea

P18

Iggesund

Sweden

Bot

Fig. 3. Location of identified local PCDD/F sources and sediment

and 2,3,7,8-TCDF correlated strongly (r = 0.896) possiblyindicating several common sources (pulp bleaching, munici-pal/industrial sources) for these congeners.

w) and I-TEQ of PCDD/Fs (pg I-TEQ g�1, dw)

2,3,7,8-TCDD

2,3,7,8-TCDF

OCDF OCDD Toxicsum I-TEQ

1.41 12.16 6450 155 14529 1163.61 18.66 23100 699 53958 4730.28 3.36 2040 17 2870 16.1

1.74 13.16 17492 394 35487 2421.17 5.11 32979 253 56965 3112.45 0.74 30 47 210 6.3

0.36 7.85 778 102 1740 22.80.61 13.07 2470 134 5540 61.20.19 4.91 430 24 959 12.9

0.49 6.44 95 92 365 9.70.79 13.20 231 149 824 23.50.23 1.75 19 13 57 2.4

0.47 6.01 59 73 237 8.40.67 8.21 88 94 329 12.20.22 0.46 5 4 17.5 1.0

1.00 13.10 73 273 656 22.20.55 15.70 190 273 876 27.90.06 0.78 9 12 38 1.3

deep sediment.

200 Km

22˚0'0"E

22˚0'0"E 32˚0'0"E

58˚

0'0

"N6

3˚0

'0"N

Finland

Estonia

Lithuania

Kymijoki

Gulf of Finland

Bal

ticP

rope

r

SR5

AKL

LL3aJML1b

Sköldvik

Pietarsaari

K15

Latvia

hnian

sea

profile sampling sites in each study (for references, see Fig. 4).


Of specific interest is the variation of 2,3,4,7,8 -PeCDFbecause of its importance as the most important toxicPCDD/F congener in fish and fish consumers. 2,3,4,7,8-PeCDF correlated significantly (p < 0.01) only with OCDD(r = 0.566) and 1,2,3,7,8-PeCDD (r = 0.332), both indica-tors of airborne origin (Table 3). Note that the Skoldvikoutlet data was not included in correlation analysis becauseof its high level of PCDD/Fs and major impact on most ofthe correlation coefficients.

A more detailed study of the 2,3,4,7,8-PeCDF/1,2,3,7,8-PeCDD ratio shows clear regional differences in the corre-

1

500

0

8

16

24

32

40

0 100 200 300 400

PCDD/F (pg I-TEQ g -1 )

1962 - 1965

K151)

Sed

imen

t dep

th (

cm)

0

8

16

24

32

0 100 200 300 400 500

1965 - 1973

AKLPCDD/F (pg I-TEQ g-1 )

2)

40

16

24

32

0

8

0 20 40 60 80

LL3a

1970 - 1973

1)

PCDD/F (pg I-TEQ g-1 )

)

0

8

16

24

32

40

0 10000 20000 30000

1234678-HpCDF (pg g-1

Sed

imen

tdep

th (

cm)

16

24

32

0

8

0 10000 20000 30000

1234678-HpCDF (pg g-1 )

0

8

16

24

32

40

0 1000 2000 3000

1234678-HpCDF (pg g )-1

1

0

8

16

24

32

40

0 20 40 6012378-PeCDD (pg g-1 )

Sed

imen

t dep

th (c

m)

0

8

16

24

32

0 5 10 1512378-PeCDD (pg g-1 )

0

8

16

24

32

40

0 5 10 1512378-PeCDD (pg g-1 )

1

1

0

8

16

24

32

40

0 10 20 30

23478-PeCDF (pg g-1 )

0

8

16

24

32

0 2 4 6 8 10

23478-PeCDF (pg g-1 )

0

8

16

24

32

40

0 10 20 30 40

23478-PeCDF (pg g-1 )

Sed

imen

t dep

th (

cm)

0

8

16

24

32

40

0 2 4 62378-TCDD (pg g-1 )

Sed

imen

t dep

th (

cm)

0

8

16

24

32

0 1 2 32378-TCDD (pg g-1 )

0

8

16

24

32

0 0.5 1 1.52378-TCDD (pg g-1 )

1

40

20

0

8

16

24

32

40

0 5 10 152378-TCDF (pg g-1)

0

8

16

24

32

0 10 20 30

2378-TCDF (pg g-1)

0

8

16

24

32

40

0 20 40 60 80

Sed

imen

t dep

th (

cm)

2378-TCDF (pg g-1)

Fig. 4. Concentration distribution of selected PCDD/F congeners in sediment pRappe, 1995.

lation pattern (Fig. 2a). For 2,3,4,7,8-PeCDF and 1,2,3,7,8-PeCDD, the increase of one of these congeners in the Gulfof Finland and in most of the Gulf of Bothnia is followedby a corresponding increase of the other. This may be a sig-nal of a same origin or source of these two congeners, asdiscussed above (atmosphere, and in the Gulf of Finlandprobably Ky-5 as well). The pattern is clearly different inthe vicinity of an old pulp mill point source (Iggesund) withno increase of 2,3,4,7,8-PeCDF and on the other hand inKattegat/Danish Sound area and southern Baltic Properwith no or small increase of 1,2,3,7,8-PeCDD. This finding

40

)

6

0

8

0 10 20 30

JML1b

1976 - 1978

PCDD/F (pg I-TEQ g-1

1)

0

8

16

24

32

0 10 20 30

23478-PeCDF (pg g-1 )

0

8

16

24

32

0 5 10 15 20

2378-TCDF (pg g-1 )

0

8

16

24

32

0 5 10 15

SR5PCDD/F (pg I-TEQ g -1 )

2)

PCDD/F (pg I-TEQ g

0

8

16

24

32

0 10 20 30 40

1974 - 1981

P18PCDD/F (pg I-TEQ g-1 )

3)

PCDD/F (pg I-TEQ g

6

0

8

0 100 200 300

1234678-HpCDF (pg g-1 )

0

8

16

24

32

0 10 20 30 40 50 60 70

1234678-HpCDF (pg g-1 )

0

8

16

24

32

0 50 100 150

1234678-HpCDF (pg g-1)

6

0

8

0 2 4 612378-PeCDD (pg g-1 )

0

8

16

24

32

0 1 2 312378-PeCDD (pg g-1 )

0

8

16

24

32

0 2 4 612378-PeCDD (pg g-1 )

0

8

16

24

32

0 5 10 15

23478-PeCDF (pg g-1 )

0

8

6

0 10 20

23478-PeCDF (pg g-1 )

0

8

16

0 0.5 12378-TCDD (pg g-1 )

0

8

16

24

32

0 0.5 12378-TCDD (pg g-1 )

0

8

16

24

32

0.0 0.5 1.0 1.52378-TCDD (pg g-1 )

0 2 4 6 8 10

2378-TCDF (pg g-1 )

0

8

6

0 5 10 15

2378-TCDF (pg g-1)

rofiles. (1) Isosaari et al., 2002a, (2) present paper (SYKE), (3) Kjeller and

0

20

40

60

80

100

S M D S M D S M D S M D S M D S M D

K15 AKL LL3a JML1b SR5 P18

%fr

om

I-T

EQ

2378-TCDF 2378-TCDD23478-PeCDF 12378-PeCDD1234678-HpCDF Others

Fig. 5. Congener patterns at different sediment profiles. (See Table 4 andFig. 3 for explanations.)


supports other than atmospheric sources for 2,3,4,7,8,-PeCDF in these Southern regions (or a different profile ofPCDD/F congeners in deposition).

Regarding deposition, a comprehensive study has beenmade in Denmark, involving Bornholm in the Baltic(Vikelsøe et al., 2005). The estimated deposition in theBaltic near Bornholm is 3.9 pg I-TEQ m�2 day�1. Preli-minary results from 2002 to 2003 at Uto (northern BalticProper) show deposition less than 1 pg I-TEQ m�2 day�1

(Korhonen et al., 2006) which is somewhat lower than atBornholm.

The variance between site or region in 2,3,4,7,8-PeCDF/OCDD ratio is less than in the case of 2,3,4,7,8-PeCDF/1,2,3,7,8-PeCDD (Fig. 2b). This figure also indicates thatthe 2,3,4,7,8-PeCDF/OCDD ratio is somewhat higher inmost of the Gulf of Finland sites than in other regions.However, only 13% of the 2,3,4,7,8-PeCDF load in the sed-iments that had been deposited in the bottom of the Gulf ofFinland in 1986–1997 was attributed to the Ky-5 source(mainly from River Kymijoki that discharges to the Gulfof Finland) (Isosaari, 2004). Iggesund area and other sitesnear pulp mills (UmU078, S3) but also Kymijoki estuary(AKL) emerge with a low 2,3,4,7,8-PeCDF/OCDD ratio.The Gothenburg region (UmU077) is seen as a probablesource of both of these congeners.

3.3. Temporal changes in sediment profiles

Table 4 shows the concentrations of selected PCDD/PCDF congeners in the surface layer, at concentrationmaximum and at deeper sediments (14–40 cm) of six sed-iment cores from the Gulf of Finland, the Gulf of Bothniaand the Baltic Proper (for locations see Fig. 3). The con-centration profiles are presented in detail in Fig. 4. Fourof the cores have been dated and published earlier (Kjellerand Rappe, 1995; Isosaari et al., 2002a). One from theGulf of Finland (AKL) and one from the Gulf of Bothnia(SR5) are from the present Finnish survey. In all casesexcept one (TCDD at P18), congener maximum leveloccurs deeper than at surface. Site P18, collected in late1980s and with the lowest gross accumulation rate, exhib-its the lowest decline. This indicates a general and cleardecrease in input of dioxins to the Baltic Sea and stillmore clearly in the Gulf of Finland and in the Gulf ofBothnia regions. Data from the southern part of the Bal-tic is missing but studies at the Kattegat-Danish Soundindicate a clear decrease as well (Vikelsøe and Hovmand,unpublished). The highest concentrations and the greatestdrops in I-TEQ occur in the Gulf of Finland (sites K15,LL3a, JML1b, 58–75% reduction) with the exception ofAKL (22%) that is located in the river Kymijoki estuary.In most of the profiles the historical background (before1900s) was not found, as judged from the deep sedimentconcentrations and sediment dating and accumulationrate (cf. evaluation of PCDD/F emission and immis-sion history based on sediment data by Green et al.,2001).

In the Gulf of Finland sediments still exhibit high con-centration of 1,2,3,4,6,7,8-HpCDF and OCDF at the sur-face, at maximum concentration layer and at ‘‘deep’’sediment layers even at station LL3a some 40 km fromKymijoki estuary (AKL), indicating a wide dispersion ofKy-5 and associated PCDD/Fs. The profiles of sites K15and AKL, having been contaminated most by Ky-5, differclearly from other sites, with one order of magnitudehigher concentrations of these congeners. The effect ofKy-5 is also seen in the middle of the Gulf of Finland atLL3a and probably also at site JML1b (Table 4, Fig. 4).The profile pattern of other congeners at this site is almostidentical with that of Ky-5 indicator HpCDF, which stillexhibits higher maximum levels than at other open sea deepsites in the Gulf of Bothnia (SR5) and near Gotland (P18)(Fig. 4).

The concentrations of two of those congeners contrib-uting significantly to PCDD/F-TEQs in fish (2,3,4,7,8-PeCDF, TCDF) do not differ either at surface or at maxi-mum in the Gulf of Finland as compared to other profiles,regardless of the possible impact of Ky-5 (Table 4). Thismay be at least partly explained by higher gross sedimenta-tion rate in the eutrophic Gulf of Finland leading to dilu-tion of anthropogenic pollutants (see also Isosaari et al.,2002a). For 1,2,3,7,8-PeCDD, important in fish also, theeffect of Kymijoki is more clear (Table 4).

At all sites far from point sources or coasts (JML1b, SR5and P18), 2,3,4,7,8-PeCDF is the congener with the highestpercentage share of the total toxicity at all depths (28–40%of I-TEQ). Although OCDF and OCDD form the majorityof total dioxins in open sea sites they do not contribute muchto the overall toxicity. It should be noted however thatOCDD and OCDF undergo transformation to lowerchlorinated PCDD/Fs also in anaerobic sediments (Bar-kovskii and Adriaens, 1996). The five other selected con-geners (2,3,7,8-TCDF, 2,3,7,8-TCDD, 2,3,4,7,8-PeCDF,1,2,3,7,8-PeCDD and 1,2,3,4,6,7,8-HpCDF) account formore than 50% of the total toxicity in all cases (Fig. 5).


4. Conclusions

The survey confirms the impact of Kymijoki-derived,highly chlorinated, PCDF-congeners to the total toxicityin sediments in the Gulf of Finland. However, the findingsdo not support that any of the known point sources wouldsignificantly influence those congeners that are mostabundant in Baltic herring and salmon. Instead, regionaldistributions in the Baltic Sea indicate that atmosphericdeposition may act as a major source for those congeners.On the other hand, some signatures of point sources orcombined point sources within specific industry branchesor particular production processes (such as pulp bleaching,VC production, thermal processes) may be discerned intheir vicinity. There are indications of declines in sedimentlevels in some areas, but generally the levels of PCDD/Fson the northern coast of the Gulf of Finland are still highas compared with other areas of the Baltic Sea, and re-sus-pension and other mobilization will cause the sediments toact as a partial source leaking PCDD/Fs for a considerabletime still. Major areas with data gaps cover the south-east-ern and eastern coastal regions of the Baltic Proper and thesouthern Gulf of Finland.

Acknowledgements

We thank all technical personnel involved in the sam-pling and laboratory analysis. The work was partiallyfunded by the Ministry of Environment, Finland, theSwedish EPA, the Geological Survey of Sweden and theDanish Environment Protection Agency.

References

Assmuth, T., Jalonen, P., 2005. Risks and management of dioxin-likecompounds in Baltic Sea fish: an integrated assessment. TemaNord2005:568. Nordic Council of Ministers, Copenhagen.

Axelman, J., Naf, C., Bandh, C., Ishaq, R., Pettersen, H., Zebuhr, Y.,Broman, D., 2001. Dynamics and distribution of hydrophobic organiccompounds in the Baltic Sea. In: Wulff, F., Rahm, L., Larsson, P.(Eds.), A Systems Analysis of the Baltic Sea, . In: Ecological Studies,vol. 148. Springer Verlag, Berlin and Heidelberg, pp. 257–287.

Baker, J.I., Hites, R.A., 2000. Is combustion the major source ofpolychlorinated dibenzo-p-dioxins and dibenzofurans to the environ-ment? A mass balance investigation. Environmental Science andTechnology 34 (14), 2879–2886.

Barabas, N., Goovaerts, P., Adriaens, P., 2004. Modified polytopic vectoranalysis to identify and quantify a dioxin dechlorination signature insediments 2. Application to the Passaic River. Environmental Scienceand Technology 38 (6), 1821–1827.

Barkovskii, AL., Adriaens, P., 1996. Microbial dechlorination of histor-ically present and freshly spiked chlorinated dioxins and diversity ofdioxin-dechlorinating populations. Applied and EnvironmentalMicrobiology 62 (12), 4556–4562.

Berggren, P., Ishaq, R., Zebuhr, Y., Naf, C., Bandh, C., Broman, D.,1999. Patterns and levels of organochlorines (DDTs, PCBs, non-orthoPCBs and PCDD/Fs) in male harbour porpoises (Phocoena phocoena)from the Baltic Sea, the Kattegat-Skagerrak Seas and the West Coastof Norway. Marine Pollution Bulletin 38 (12), 1070–1084.

Bergqvist, P.-A., Bergek, S., Hallback, H., Rappe, C., Slorach, S.A., 1989.Dioxins in cod and herring from the seas around Sweden. Chemo-sphere 19 (1–6), 513–516.

Bergqvist, P.-A., Tysklind, M., Marklund, S., Aberg, A., Sundqvist, K.,Naslund, M., Rosen, I., Tsytsik, P., Malmstrom, H., Cato, I., 2005.Kartlaggning av utslappskallor for oavsiktligt bildade amnen: PCDD/F, PCB och HCB (Inventory of emission sources for inadvertentlyformed compounds: PCDD/F, PCB and HCB, in Swedish), Environ-mental Chemistry, Umea University, Umea, Sweden, MK 2005:01.

Bruckmeier, B.F.A., Juttner, I., Schramm, K.-W., Winkler, R., Steinberg,C.E.W., Kettrup, A., 1997. PCBs and PCDD/Fs in lake sediments ofGroßer Arbersee, Bavarian Forest, South Germany. EnvironmentalPollution 95 (1), 19–25.

Buekens, A., Cornelis, E., Huang, H., Dewettinck, T., 2000. Fingerprintsof dioxin from thermal industrial processes. Chemosphere 40 (9–11),1021–1024.

Buekens, A., Stieglitz, L., Hell, K., Huang, H., Segers, P., 2001. Dioxinsfrom thermal and metallurgical processes: recent studies for the ironand steel industry. Chemosphere 42 (5–7), 729–735.

Dannenberger, D., Andersson, R., Rappe, C., 1997. Levels and patterns ofpolychlorinated dibenzo-p-dioxins, dibenzofurans and biphenyls insurface sediments from the Western Baltic Sea (Arkona Basin) and theOder River Estuarine system. Marine Pollution Bulletin 34 (12), 1016–1024.

Eduljee, G., 1999. Secondary exposure to dioxins through exposure toPCP and its derivatives. Science of the Total Environment 232 (3),193–214.

Frignani, M., Bellucci, L.G., Carraro, C., Favotto, M., 2001. Accumu-lation of polychlorinated dibenzo-p-dioxins and dibenzofurans insediments of the Venice Lagoon and the industrial area of PortoMarghera. Marine Pollution Bulletin 42 (7), 544–553.

Glynn, A.W., Atuma, S., Aune, M., Darnerud, P.-O., Cnattingius, S.,2001. Polychlorinated biphenyl congeners as markers of toxic equiv-alents of polychlorinated biphenyls, dibenzo-p-dioxins and dibenzofu-rans in breast milk. Environmental Research 86 (3), 217–228.

Gotz, R., Lauer, R., 2003. Analysis of sources of dioxin contamination insediments and soils using multivariate statistical methods and neuralnetworks. Environmental Science and Technology 37 (24), 5559–5565.

Grabic, R., Novak, J., Pacakova, V., 2000. Optimization of a GC-MS/MSmethod for the analysis of PCDDs and PCDFs in human and fishtissue. Journal of High Resolution Chromatography 23, 595–599.

Green, N.J.L., Jones, J.J., Jones, K.C., 2001. PCDD/F deposition timetrend to Esthwaite Water, U.K., its relevance to sources. Environ-mental Science and Technology 35, 2882–2888.

Ishaq, R., Naf, C., Zebuhr, Y., Broman, D., Jarnberg, U., 2003. PCBs,PCNs, PCDD/Fs, PAHs and Cl-PAHs in air and water particu-late samples—patterns and variations. Chemosphere 50 (9), 1131–1150.

Isosaari, P., 2004. Polychlorinated dibenzo-p-dioxin and dibenzofurancontamination of sediments and photochemical decontamination ofsoils. Academic Dissertation, Publications of the National PublicHealth Institute A 11/2004, Kuopio, Finland.

Isosaari, P., Kohonen, T., Kiviranta, H., Tuomisto, J., Vartiainen, T.,2000. Assessment of levels, distribution, and risks of polychlorinateddibenzo-p-dioxins and dibenzofurans in the vicinity of a vinyl chloridemonomer production plant. Environmental Science and Technology34, 2684–2689.

Isosaari, P., Kankaanpaa, H., Mattila, J., Kiviranta, H., Verta, M., Salo,S., Vartiainen, T., 2002a. Spatial distribution and temporal accumu-lation of polychlorinated dibenzo-p-dioxins, dibenzofurans and biphe-nyls in the Gulf of Finland. Environmental Science and Technology36, 2560–2565.

Isosaari, P., Pajunen, H., Vartiainen, T., 2002b. PCDD/F and PCBhistory in dated sediments of a rural lake. Chemosphere 47 (6), 575–583.

Jonsson, P., Rappe, C., Kjeller, L.-O., Kierkegaard, A., Hakanson, L.,Jonsson, B., 1993. Pulp mill related polychlorinated organic com-pounds in baltic sea sediments. Ambio 22, 37–43.


Kaas, H., Markager, S., 1998. NOVA Teknisk anvisning for marinovervagning. Technical guidance document on sediment sampling, m-fdc.dmu.dk (in Danish).

Kim, S.-C., Na, J.-G., Choe, S.-H., Lee, J.-H., Kim, Y.-H., Hwang, S.-R.,Joo, C.-H., Moon, D.-H., You, J.-C., Jung, E.-K., Lee, C.-J., Choi,KS., 2003a. PCDDs/PCDFs emission from ferrous metal industry.Organohalogen Compounds 63, 77–80.

Kim, S.-C., Na, J.-G., Choe, S.-H., Lee, J.-H., Kim, Y.-H., Hwang, S.-R.,Joo Jung, D.-H., You, J.-C., Lee, S.-W., Jeon, S.-E., 2003b. PCDDs/PCDFs emission from nonferrous metal industry. OrganohalogenCompounds 63, 81–85.

Kiviranta, H., Purkunen, R., Vartiainen, T., 1999. Levels and trends ofPCDD/Fs and PCBs in human milk in Finland. Chemosphere 38 (2),311–323.

Kiviranta, H., Hallikainen, A., Ovaskainen, M.L., Kumpulainen, J.,Vartiainen, T., 2001. Dietary intakes of polychlorinated dibenzo-p-dioxins, dibenzofurans and polychlorinated biphenyls in Finland.Food Additives and Contaminants 18 (11), 945–953.

Kiviranta, H., Vartiainen, T., Tuomisto, J., 2002. Polychlorinateddibenzo-p-dioxins, dibenzofurans, and biphenyls in fishermen inFinland. Environmental Health Perspectives 10 (4), 355–361.

Kiviranta, H., Vartiainen, T., Parmanne, R., Hallikainen, A., Koistinen,J., 2003. PCDD/Fs and PCB in Baltic herring during the 1990s.Chemosphere 50, 1201–1216.

Kjeller, L.-O., Rappe, C., 1995. Time trends in levels, patterns, andprofiles for polyhlorinated dibenzo-p-dioxins, dibenzofurans, andbiphenyls in a sediment core from the Baltic Proper. EnvironmentalScience and Technology 29, 346–355.

Koistinen, J., Nevalainen, T., Tarhanen, J., 1992. Identification and levelestimation of aromatic coeluates of polychlorinated dibenzo-p-dioxinsand dibenzofurans in pulp mill products and wastes. EnvironmentalScience and Technology 26, 2499–2507.

Koistinen, J., Paasivirta, J., Suonpera, M., Hyvarinen, H., 1995.Contamination of pike and sediment from the Kymijoki River byPCDEs, PCDDs, and PCDFs: content and patterns compared to pikeand sediment from the Bothnian Bay and Seals from Lake Saimaa.Environmental Science and Technology 29, 2541–2547.

Koistinen, J., Stenman, O., Haahti, H., Suonpera, M., Paasivirta, J., 1997.Polychlorinated diphenyl ethers, dibenzo-p-dioxins, dibenzofurans andbiphenyls in Seals and sediment from the Gulf of Finland. Chemo-sphere 35 (6), 1249–1269.

Korhonen, M., Verta, M., Salo, S., Vuorenmaa, J., 2006. The depositionof PCDD/Fs in Southern Finland (Manuscript).

Lohmann, R., Jones, K.C., 1998. Dioxins and furans in air anddeposition: a review of levels, behaviour and processes. The Scienceof the Total Environment 219, 53–81.

Marvin, C., Alaee, M., Painter, S., Charlton, M., Kauss, P., Kolic, T.,MacPherson, K., Takeuchi, D., Reiner, E., 2002. Persistent organicpollutants in Detroit River suspended sediments: polychlorinateddibenzo-p-dioxins and dibenzofurans, dioxin-like polychlorinatedbiphenyls and polychlorinated naphthalenes. Chemosphere 49 (2),111–120.

Masunaga, S., Yao, Y., Ogura, I., Nakai, S., Kanai, Y., Yamamuro, M.,Nakanishi, J., 2001. Identifying sources and mass balance of dioxinpollution in Lake Shinji Basin, Japan. Environmental Science andTechnology 35 (10), 1967–1973.

Naf, C., Broman, D., Papakosta, O., Rolff, C., Zeburh, Y., Axelman, J.,Bandh, C., Ishag, R., Pettersen, H., Bergqvist, P.-A., Bavel, B.,Lundgren, K., Strandberg, B., Zook, D., Rappe, C., 1997. Ackumula-tion och fordelning av miljogifter i Bottniska Viken –Naringsvavsprojektet. Stockholms Universitet, Stockholm (in Swedish).

Persson, N.J., Gustafsson, O., Bucheli, T.D., Ishaq, R., Naes, K.,Broman, D., 2002. Soot-carbon influenced distribution of PCDD/Fsin the marine environment of the Grenlandsfjords, Norway. Environ-mental Science and Technology 36 (23), 4968–4974.

Rappe, C., 1992. Sources of PCDDs and PCDFs. Introduction. Reactions,levels, patterns, profiles and trends. Chemosphere 25 (1–2), 41–44.

Rappe, C., Kjeller, L.-O., 1987. PCDDs and PCDFs in environmentalsamples air, particulates, sediments and soil. Chemosphere 16 (8–9),1775–1780.

Rappe, C., Bergqvist, P.-A., Marklund, S., 1985. Analysis of polychlo-rinated dibenzofurans and dioxins in ecological samples. In: Keith,L.H., Rappe, C., Choudhary, G. (Eds.), Chlorinated Dioxins andDibenzofurans in the Total Environment II. Butterworth Publishers,Stoneham, MA, USA.

Rappe, C., Bergqvist, P.-A., Kjeller, L.-O., 1989. Levels, trends andpatterns of PCDDs and PCDFs in scandinavian environmentalsamples. Chemosphere 18 (1–6), 651–658.

Rappe, C., Glas, B., Kjeller, L.-O., Kulp, S.E., de Wit, C., Melin, A., 1990.Levels of PCDDs and PCDFs in products and effluent from theSwedish pulp and paper industry and chloralkali process. Chemo-sphere 20 (10–12), 1701–1706.

Rappe, C., Kjeller, L.-O., Kulp, S.-E., de Wit, C., Hasselsten, I., Palm, O.,1991. Levels, profile and pattern of PCDDs and PCDFs in samplesrelated to the production and use of chlorine. Chemosphere 23 (11–12),1629–1636.

Sakurai, T., Kim, J.-G., Suzuki, N., Matsuo, T., Li, D.-Q., Yao, Y.,Masunaga, S., Nakanishi, J., 2000. Polychlorinated dibenzo-p-dioxinsand dibenzofurans in sediment, soil, fish, shellfish and crab samplesfrom Tokyo Bay area, Japan. Chemosphere 40 (6), 627–640.

SCALE 2004. Baseline report on ‘‘Integrated Monitoring of dioxins &PCBs in the Baltic Region, in the Framework of the EuropeanEnvironment and Health Strategy (COM(2003)338 final). Produced bythe Technical Working Group on Integrated Monitoring of dioxins &PCBs in the Baltic Region, 09 January 2004. <http://www.environ-mentandhealth.org/>.

Schramm, K.-W., Henkelman, B., Kettrup, A., 1995. PCDD/F sourcesand levels in River Elbe sediments. Water Research 29 (9), 2160–2166.

Stevens, J.L., Green, N.J.L., Jones, K.C., 2003. Fate of 1,2,3,4,6,7,8-heptachlorodibenzofuran and pentachlorophenol during laboratory-scale anaerobic mesophilic sewage sludge digestion. Chemosphere 50(9), 1227–1233.

Vartiainen, T., Lampi, P., Tolonen, K., Tuomisto, J., 1995. Poly-chlorodibenzo-p-dioxin and polychlorodibenzofuran concentrationsin lake sediments and fish after ground water pollution withchlorophenols. Chemosphere 30, 1439–1451.

Vartiainen, T., Mannio, J., Korhonen, M., Kinnunen, K., Strandman, T.,1997. Levels of PCDD, PCDF and PCB in dated lake sediments inSubarctic Finland. Chemosphere 34 (5–7), 1341–1350.

Verta, M., Lehtoranta, J., Salo, S., Korhonen, M., Kiviranta, H., 1999.High concentrations of PCDD’s and PCDF’s in river Kymijokisediments, South-Eastern Finland, caused by wood preservative Ky-5.Organohalogen Compounds 43, 261–264.

Verta, M., Salo, S., Malve, O., Kiviranta, H., 2003. Continued transportof PCDD/F contaminated sediments from River Kymijoki to the Gulfof Finland, the Baltic Sea. Organohalogen Compounds 61, 405–408.

Vikelsøe, J., Johansen, E., 2000. Estimation of dioxin emission from firesin chemicals. Chemosphere 40 (2), 165–175.

Vikelsøe, J., Andersen, H.V., Hovmand, M.F., 2005. Role of PCDD/F indeposition for soils, percolate and sediments. Organohalogen Com-pounds 67, 1170–1173.

Vulykh, N., Shatalov, V., 2001. Investication of dioxin/furan compositionin emissions and in environmental media. Selection of congeners formodeling, Technical Note 6/2001, Meteorological Synthesizing Centre– EAST.

Vuorinen, P., Parmanne, R., Vartiainen, T., Keinanen, M., Kotovuori, O.,Halling, F., 2002. PCDD, PCDF, PCB and thiamine in Baltic herring(Clupea harengus L.) and sprat (Sprattus sprattus L.) as a backgroundto the M74 syndrome of Baltic salmon (Salmo salar L.). ICES Journalof Marine Science 59, 480–496.

Wenborn, M., King, K., Buckley-Golder, D., Gascon, J.A., 1999. Releasesof dioxins and furans to land and water in Europe. Final Report Issue2, Report produced for Landesumweltamt Nordrhein-Westfalen,Germany on behalf of European Commission DG Environment,

http://www.environmentandhealth.org/

http://www.environmentandhealth.org/


September 1999, Report AEAT-4703, 149 pp. <http://europa.eu.int/comm/environment/dioxin/download.htm>.

Wingfors, H., Lindstrom, G., van Bavel, B., Schuhmacher, M., Hardell,L., 2000. Multivariate data evaluation of PCB and dioxin profiles inthe general population in Sweden and Spain. Chemosphere 40 (9–11),1083–1088.

Witt, G., Schramm, K.W., Henkelman, B., 1997. Occurrence and distri-bution of polychlorinated dibenzo-p-dioxins and dibenzofurans insediments of the Western Baltic Sea. Chemosphere 35 (7), 1465–1473.

Wu, W.Z., Schramm, K.-W., Xu, Y., Kettrup, A., 2001. Mobility andprofiles of polychlorinated dibenzo-p-dioxins and dibenzofurans insediment of Ya-Er Lake, China. Water Research 35 (12), 3025–3033.

http://europa.eu.int/comm/environment/dioxin/download.htm

http://europa.eu.int/comm/environment/dioxin/download.htm

Dioxin concentrations in sediments of the Baltic Sea – A survey of existing data

Documents