The distribution and effects on Norwegian fjord and coastal ecosystems of polycyclic aromatic hydrocarbons (PAHs) generated by the production of primary aluminium and manganese alloys Kristoffer Na?s Norwt'xiun Institute for Water Research - NIVA Southern Branch Televeien 1 N-4890 Grimstad Norway June 1998 Thesis for the degree of Doctor Philosophiae University of Oslo, Norway © Kristofjer N<Es Design: Fetter Wang Print: Reclarno ISBN 82-577-3464-0 Abstract Discharges of polycyclic aromatic hydrocarbons (P AHs) generated by the production of primary aluminium and manganese alloys have been a serious environmental problem in Norwegian fjords and coastal waters. Scien­ tists have been performing environmental assessments in the receiving waters from these production plants for more than twenty-five years, detecting high P AH con­ centrations in sediments and biota. Most of the assess­ ments have been limited to individual recipient bodies of water. No previous projects have evaluated the results from a holistic point of view, that is, as an integrated analysis within the same ecosystem compartment (for example, inter-fjord sediment to sediment comparison) or between different compartments (e.g. sediments to organisms within and between fjords). Over the past decade, changes in production processes, the installation of scrubbers etc. have drawn attention to the need for a broad-based evaluation of the ecological fate and effects of present and past P AH discharges. Accordingly, the main objective of this study has been to provide a comprehensive environmental risk assess­ ment of discharges to marine waters of P AHs generated by the production of primary aluminium and manganese alloys. This has been accomplished by studying PAHs in dissolved, colloidal and particulate states in waste and recipient waters, the accumulation of P AHs in bottom sediments and littoral indicator organisms, and the effects of P AHs on benthic fauna at the community and cellular levels. Most of the data have been collected in the course of 25 years of monitoring, although new data have been collected to fill in knowledge gaps. To a large ex­ tent, numerical treatment has employed multivariate techniques such as principal component analysis, redun­ dancy analysis and canonical correspondence analysis. 3 Acknowledgements This project could not have been realised without the interest and enthusiasm of many people. First of all, I am indebted to the institutions that financed it: Elkem Aluminium Lista, Elkem Aluminium Mosje~en, Elkem Mangan Sauda, Elk em Mangan PEA, Hydro Aluminium Helyanger; Hydro Aluminium KarmeJy, Hydro Aluminium Sunndal, Hydro Aluminium Ardal, Tmfos JemverkA/S, the Norwegian Pollution ControlAuthority, the Research Council of Norway (Programme FORFOR- Research Programme for Sustainable Development) and the Norwegian Institute for Water Research. Further, I would thank the advisory committee for the project, consisting of Eirik N ordheim and partly Anders Bruus­ gaard (The NordicAluminium Industry's Secretariat for Health, Environment and Safety), Magne Leinum (Hydro Aluminium Sunndal), Ame Kjellsen (Elkem Mangan PEA), Per Erik Iversen (Norwegian Pollution Control Authority), Jan Rob (Elkem Mangan Sauda, now at Elkem Aluminium Lista) and Jens Skei (Norwegian Institute for Water Research), for their constructive participation and comments on the manuscripts. I am especially indebted to Eirik Nordheim (now at the European Aluminiwn Association) and Cato Isene StoU (th n at Norsk Hydro) who were the "midwives" for th.is project.l would also express my gratitude to aJI the peopl in the industy who followed the project with interest. A special thanks to Svein Han:ySamuelsen(Elkem Aluminium Lista) and Aslak Schanke (then Elkem Aluminium Lista) for their enthusiasm. My appreciation also to my colleagues at the University of Stockholm: Johan Axelman, Dag Broman and Carina Niif for co-operation and fruitful meetings at their laboratory. Last, but not least, I would express my sincere gratitude to my colleague Eivind Oug who introduced me to the world of multivariate statistics. I want to thank him for the many con­ structive discussions, all the scientific enthusiasm and his "painful" precision inwritingthemanuscripts. Without his co­ operation, thi work would never have come to fruition. Finally, thanks to Linda Sivesind for her linguistic correc­ tions on the manuscripts. 4 List of papers This thesis is based on the papers listed below. They are referred to in the text by their Roman numerals. I. N~s, K., }. Knutzen and L. 13erglind, 1995. Occurence of PAl-l in marine orgatlisms and sediments from smelter discharge in Norway. Sci. Total. Environ., 163:93-106 11. N~s, K., T. Bakke ru1d R Konieczny, 1995. Mobilization of poly.cydi <:~romatic hydrocarbons (PAHs) from polluted seabed and uptake in the blue mussel (Mytilus edulis) . f. Mar. Freshwater Re ., 46: 275-285. Ill. Naas, K. and E. Oug, 1997. Multivariate pproad1 to distribution patterns and fate of polycyclic ar matic hydrocarbons in sediments from smelter-affected Norwegian fjords and coastal waters. Environ. Sci. Technol ., 31: 1253-1258. IV. N~s, K. and E. Oug, 1998. The distribution and environmental relationships of polycyclic aromatic hydrocarbons (PAHs) in sediments from Norwe­ gian smelter-affected fjords. CHEMOSPHERE, 36: 561-576. V. N~s, K., E. Oug and J. Knutzen, 1998. Source and species-dependent accumulation of polycyclic aromatic hydrocarbons (PAHs) in littoral indicator organisms from Norwegian smelter-affected marine waters. Mar. Environ. Res., 45: 193-207. VI. Oug, E., K. N~s and B. Rygg, in prep. Relation­ ships between soft bottom macrofauna and polycy­ clic aromatic hydrocarbons (PAHs) from sm.elter discharge in Norwegian fjords and coastal waters. Submitted Mar. Ecol. Prog.Ser. VII. N~s, K., K. Hylland, E. Oug, L. Forlin and G. Eric­ san, in prep. Accumulation and effects of aluminium-smelter generated P AHs on soft-bottom invertebrates and fish. Submitted Environ. Toxicol . Chem. VIII. N~s, K., J. Axelman, D. Broman and C. Naf, 1998. Role of soot carbon and other carbon matrices in the distribution of P AHs among particles, DOC and the dissolved phase in the effluent and recipient waters of an aluminium reduction plant. Environ. Sci. Technol., 32: 1786-1792. IX. Axelman, }., K. N~s, D. Broman and C. Naf, in prep. Accumulation of P AHs by SPMDs and caged blue mussels (Mytilus edulis L.) and comparison with the PAH distribution in the dissolved, colloi­ dal and particulate size fractions in the water column. Submitted Environ. Toxicol. Chem. 5 Table of contents 1. Introduction 7 I . J Background 7 J .2. Objective ~ 2. PAH generation and the route of entry into the marine environment 9 2. J. Aluminium production 9 2.2. Manganese alloy production 9 2.3. Smelter location and production capacity I 0 2.4. Cleaning equipment and discharge arrangements J 0 2.5. Recipient waters 11 3. Data base 12 3.1. Exi sting and new data 12 3.2. Sampling, analysis and quality assurance 13 3.3. Numerical treatment 14 4. Quantitative and qualitative aspects of PAH discharges 14 4. J. The general physico-chemical properties of PAHs 15 4.2. Discharge volumes 15 4.3. PAH speciation in effluent and recipient waters 16 5. Transport and sediment incorporation 18 5.1. Implications of a significant non-particulate phase I 8 5.2. Transformations of the particulate phase 18 5.3. Explanatory variables for the distribution ~f PAHs in sediments 19 5.4. Sediments as a secondary source of PAHs 20 6. Accumulation and effects on fauna 21 7. Implications on monitoring strategies 24 8. Conclusions and implications 26 9. References 28 6 Introduction 1.1. Background 1995), although these properties vary from one Towards the end of the 1900s, Norway developed PAH compound to the next. its hydroelectric power for industrial purposes. The Environmental assessments of the marine country's topography offered an abundant supply receiving waters have been carried out for the past of waterfalls, making hydropower cheap and easily 25 years, and high PAH concentrations have been accessible. This paved the way for a flourishing detected in sediments and organisms alike (Paper metallurgical industry based on Norwegian mines I). In some areas, this has resulted in restrictions on as well as imported ore. The invention of the commercial fishing and dietary warnings to con­ Selderberg electrode (Grjotheim and K vande 1993) sumers. Currently, this situation applies to five in 1917 to 1919 was a milestone for manufactur­ fjords. Environmental programmes have mainly ing. The extensive industrial development earlier been based on concentrations in bottom sediments in the century has given rise to many of today's and indicator organisms (mussels and snails). In local and regional environmental issues. Most many of the fjords, the analyses of the soft-bottom manufacturers are situated along fjords or coastal community structure have been included as a waters, meaning that environmental problems can measure of effects. Most programmes have been affect marine ecosystems, fjords in particular. limited to individual recipient bodies of water. No Norwegian fjord ecosystems are of particular analyses have evaluated the results from a holistic importance to the environment for two reasons: point of view, that is, as an integrated analysis first, because they serve as sedimentation basins performed within the same ecosystem compart­ that trap discharged contaminants, and second, ment (for example, inter-fjord sediment to sedi­ because most of the population of Norway lives ment comparison) or between different compart­ along the coast. ments (for example, sediments to organisms within Norway currently has about 30 metal works in and between fjords). Apart from the obvious operation, of which seven are primary aluminium benefits for summarising and comparing data, a (Al) reduction plants. Three produce manganese holistic approach may also reveal significant (Mn) alloys. For all Al and Mn works, the sea is patterns impossible to discern from the original the main recipient body of water.