THE ROLE OF SEDIMENT BIOTURBATION BY

THE ESTÜARINE AMPHIPOD COROPHIVM VOLUTATOR

IN PAH BIOAVAILABILITY IN WATER

Report nr. D99016

Available from:

Department of Ecology and Ecotoxicology

Vrije Universiteit

De Boelelaan 1087

1081 HV Amsterdam

Tel.: +31-20-4447004

THE ROLE OF SEDIMENT BIOTURBATION BY

THE ESTUARINE AMPHIPOD COROPHWM VOLUTATOR

IN PAH BIOAVAILABILITY IN WATER

Silvana CiarelliRijkswaterstaat

& Rijksinstituut voor Kust en Zee/RIKZBibliotheek (Middelburg)

Nico M. van Straalen

Dept. of Ecology and EcotoxicologyVrije Universiteit

De Boelelaan 10871081 HV Amsterdam

Commissioned by the National Institutefor Coastal and Marine Management (RIKZ)

Grenadierweg314338 PG Middelburg

Contract: RKZ-510

February 1999

CONTENTS

Preface

Chapter 1 1General Introduction 2

Chapter 2 8Effects of sediment bioturbation by the estuarine amphipod Corophium volutator 9on fluoranthene resuspension and transfer into the mussel, Mytilus edulis L.(paper published in Environmental Toxicology and Chemistry in February 1999)

Chapter 3 20Effects of sediment bioturbation by the estuarine amphipod Corophium volutator 21on fluoranthene resuspension and transfer into mussel, Mytilus edulis L.(poster presented in Charlotte -North Carolina in November 1998)

Chapter 4 23The influence of bioturbation by the amphipod, Corophium volutator 24on fluoranthene uptake in the marine polychaete, Nereis virens(paper submitted to Environmental Toxicology and Chemistry)

Chapter 5 40Resuspension of PAH in water column from spiked and unspiked 41sediments induced by Corophium volutator bioturbation

Chapter 6 55Summary 56

Preface

This final report prepared for the National Institute for Coastal and Marine Management(RIKZ-contract RKZ-510) summarizes the results of the past half year-study on theecotoxicological effects of bioturbation by the amphipod Corophium volutator.

The problem, the objectives of the study, and a short review of principal bioturbationstudies with the related references, are outlined in chapter 1.

In chapter 2, a copy of a paper published in the journal, Environmental Toxicology &Chemistry (February 1999) concerning the effects of sediment bioturbation by the marineamphipod C. volutator on partitioning and bioavailability of fluoranthene to the mussel,Mytilus edulis, is included.

Chapter 3 contains a copy of a poster presented at the SETAC-meeting in Charlotte (North-Carolina) in November 1998, describing in short the rnain results published in the papermentioned above.

Chapter 4 reports the results of a study which aimed to investigate the influence ofsediment bioturbation by the amphipod, C. volutator on toxicokinetics of fluoranthene in themarine polychaete, Nereis virens. This manuscript will be submitted to the journal,Environmental Toxicology & Chemistry.

Chapter 5, describes part of the results obtained from a study performed in cooperationwith Drs. R. Kraaij of the "AGING" project which aimed to investigate the influence of agïngand bioturbation on desorption kinetics and bioaccumulation in the amphipod, C. volutator ofdifferent hydrophobic organic compounds. In this report, only the results concerning theresupension of different PAH in overlying water by sediment bioturbation are described.

Chapter 6 summarizes shortly all the results obtained in the past half-year research pointingout the most relevant aspects of this study.

We acknowledge Drs. BJ. Kater, Dr. A.D. Vethaak, Dr. K. Legierse, Dr. A. Belfroid andDr. C.A.M, van Gestel for their valuable comments on the manuscript of chapter 4.

Chapter 1

General Introduction

General Introduction

Theproblem

Current sediment quality criteria are based on the assumption that sediments are merely asink for contaminants and that risk assessments can be performed by the use of theEquilibrium Partitioning theory (EqP) approach. In this approach, sediment-waterdistributions of contaminants are predicted assuming that sorption of compounds, atequilibrium, is mainly regulated by hydrophobicity of contaminants and the organic mattercontent of the sediment. Another assumption is that bioaccumulation in the organisms issolely determined by the freely dissolved concentration of a compound in the water phase andby the lipid content of the species (DiToro et al., 1991). Deviations from the expectedconcentrations of compounds in the pore water, predicted by the EqP model on the basis ofsediment concentrations, have, however, been found by several authors (Kukkonen &Landrum (1994), Landrum et al. (1994) and Belfroid et al. (1995). The recognition thatbioavailability of compounds depends also on biological processes such as biodegradation,biodeposition and bioturbation has only recently been made (Forbes & Forbes, 1994). Theseprocesses may also act as confounding factors in the interpretation of results of Standardsediment toxicity tests with burrowing amphipods and in the evaluation of sediments toxicity.

Previous studies describing and modeling sediment-water exchange processes have dealtexclusively with physical and geochemical characteristics such as sedimentation rate, mixinglayer depth, resuspension, adsorption and desorption of sediment particles to predict the fateof pollutants in the aquatic environments (Kramer et al., 1989a,b; Van Veen, 1989). Thebiological processes mentioned above and the interactions between benthic invertebrates andsediment materials are poorly understood and not well examined in existing sedimenttransport models (Thoms et al., 1995). Bioturbation, which is the the net effect of locomotion,feeding, irrigation and burrowing activities and, the role that it plays in chemical, physical andbiological changes in sediment-water exhange processes is a biological, not well understoodprocess and also poorly represented in models. Therefore, integrated approaches wheresediment, biota and pollutants interact dynamically are needed to better understand the cyclingand the effects of pollutants in sedimentary systems (Forbes & Forbes, 1994; Forbes & Kure,1997).

Effects of bioturbation

Deposit feeders -those that ingest organic and inorganic particles obtained either from thesediment surface or within the sediments, are among the organisms which play the mostimportant role in sediment bioturbation. Although some species of benthic fish are known toinfluence resuspension of sediments, the process of bioturbation itself is particularlydetermined by burrowing activity of infaunal invertebrates. These organisms, depending onsediment reworking mechanisms, size, abundance and density of species can exert their effecton the sediment itself and on sediment-water interface exchanges in various ways (Cadée,1979; Sharon et al., 1995). One may distinguish between two types of bioturbation that havedifferent consequences.

Rapid mixing of sediment particles and pollutants within the sediment, is one aspect ofbioturbation described for polychaetes such as Arenicola marina (Everaarts & Devi, 1996;Rasmussen et al., 1998) and Heteromastus filiformis (Cadée, 1979; Mulsow & Landrum,1995) and for oligochaetes worms such as Limnodrilus hoffineisteri and Stylodrilus

heringianus (Keilty et al., 1998a; b) and for Tubifex spp. (Fisher et al., 1980; Karickoff &Morris, 1985). The interesting aspect of these worms is that they ingest subsurface bottomsediments and convey them to the sediment-water interface as faecal pellets in a conveyor beltfashion. Rates at which buried sediments are egested and subsequently reburied ( = reworkingrate) have been studied under a variety of laboratory conditions (Ciarelli & Van Straalen,1997). The most important effects produced by sediment reworking activity and sedimentmixing by worms are: redistribution of contaminants to greater sediment depth (Everaarts &Devi, 1996; Rasmussen et al., 1998), mixing of solutes as overlying water is pumped from tailto head for oxygenation of gill fïlaments and increase of sediment contaminants to thesediment^water interface through faecal pellets egestion (Karickoff & Morris, 1985; Mulsow& Landrum, 1995). An important consequence of burrow irrigation by worms is thestimulation of bacterial growth and biodegradation processess (Bauer et al, 1998; Kure &Forbes, 1997).

The second interesting aspect of bioturbation is the ability of various subsurface deposit-feeders to increase sediment-water surface by their feeding and continuous burrowing activityand to increase fluxes of oxygen, nutrients and toxicants in pore water and at the sediment-water interface (Aller 1980, 1988; Meadows & Meadows, 1991; Davey, 1994; 1995). Thisphenomenon has been mostly described for crustaceans (amphipods and copepods), insects(midge larvae), and polychaetes (Nereididae species). Irrigation activity of the burrows and theoxygen penetration in the sediment, enhances the oxidized surface layer to spread more deeplyin the sediment favoring nitrification and denitrification processes. The increase of oxygenand denitrification processess stimulates bacterial production (Van de Bund, 1994) and alsobiodegradation of contaminants (McElroy, 1990). In addition, the irrigation activity of theburrows flushes the burrows maintaining a concentration gradiënt favoring diffusion betweenwater in the burrow and the sediment (Gilbert et al., 1998). An important consequence ofbioturbation of subsurface deposit-feeders is sediment partiële resuspension in overlying waterwhich in turn produces an effect on the sediment shear strength, on the sediment water contentand on the sediment erosion (Meadows & Tait, 1989; Paterson, 1997).

From an ecotoxicological point of view the main consequences of bioturbation bysubsurface deposit-feeders is similar to that described above for "conveyor-belt" depositfeeders, i.e. changes in the fate and availability of sediment-bound contaminants. Thedïfference, however, is that the contaminants move from the sediment to the pore water andoverlying water rather than being distributed into deeper layers of the sediment. Significantincreases of sediment-associated compounds from sediment to pore water and overlying waterare described by various authors (Riedel et al., 1989; McElroy et al., 1990; Clements et al.,1994; Green & Chandler, 1994).

The role of bioturbation by the amphipod, Corophiwn volutator is from an ecological pointof view, quite well documented. Studies on the influence of burrowing on the modification ofsediment significant properties, showed that C. volutator burrows can increase the shearstrength of the sediment and significantly decrease the water content at the surface of thesediment and sediment permeability (Meadows & Tait, 1989; Grant & Daborn, 1994). Theincrease in shear strength is probably caused by particle-binding secretions from the burrowwall and by the reduction in water content which result in compaction and stabilization of thesediment. Lower permeability at higher densities of C. volutator is probably caused by thephysical barrier of the U-shaped burrows in the top few centimetres of the sedimentarycolumn. Other long-term (field and laboratory) studies focusing on the effect of C. volutatoron the abundance of benthic diatoms, bacteria and sediment stability showed, however,contrasting results, i.e. bioturbation decreased compaction of the sediment and increased watercontent and sediment permeability (Daborn et al., 1993; Gerdol & Hughes, 1994). The authors

of these studies in concomitant with more recent studies, concluded that bioturbation by C.volutator can cause sediment erosion due to grazing activity of large numbers of benthicdiatoms and consuming organic material such as bacteria attached to the sediment particles.The decrease of microalgal secretions and extracellular polymeric substances (EPS) likelyproduced by diatoms and bacteria, which are responsible for, sediment cohesity and sedimentstability, would increase sediment erodibility (Paterson, 1997; E. de Deckere, pers. comm.).

Other studies have demonstrated the importance of C.volutator in stimulating sediment-water fluxes such as oxygen uptake, increase of N H / fluxes by excretion and transport to thewater column (Henriksen et al., 1983), nitrifïcation and transport of NC«3' from the watercolumn into the sediment and denitrification of NO3" in the sediment by ventilation andirrigation activities (Pelegrï et al., 1994; Pelegn' & Blackburn, 1994).

Ecotoxicological consequences of bioturbation of C. volutator are not described inliterature. This purpose in addition to others mentioned here below explain the reasons whywe focused on this species in our study (Ciarelli & Van Straalen, 1997):

1. bioturbation of C. volutator plays a key role in various ecological aspects in the naturalenvironment due to active burrow irrigation and sediment resuspension (as mentioned above);

2. is one of the most abundant species in muddy sediments of the Dutch coasts andintertidal areas;

3. plays an important role in the food chain as it represents food source for crabs, shrimps,fish and birds;

4. is a widely used test organism in European standardized marine sediment toxicity tests.

The objectives of the study

The objectives of this study were threefold:1. to study the influence of bioturbation on partitioning of fluoranthene in sediment, pore

water and overlying water;2. to investigate the role of amphipod bioturbation on fluoranthene bioavailability in

overlying water and the accumulation in the suspension-feeder, Mytilus edulis and in thepolychaete, Nereis virens;

3. to compare the effects of bioturbation by C. volutator on the behaviour of different PAHin overlying water of a field-contaminated sediment with those of a spiked sediment.

Literature

Aller RC (1980) Quantifying solute distribution in the bioturbated zone of marine sedimentsby defining an average microenvrionment. GeochimCosmochim 44:1955-1965

Aller RC (1988) Benthic fauna and biogeochemical processes in marine sediments: the role ofburrow structures. In: Blackburn TH, Sorensen J (eds) Nitrogen Cyclïng in coastal marineenvironments. Wiley, J & Sons Ltd, pp. 302-338

Bauer JE, Kerr RP, Bautista MF, Decker CJ, Capone DG (1988) Stimulation of microbialactivities and polycyclic aromatic hydrocarbon degradation in marine sediments inhabited byCapitella capitata. Mar Environ Res 25:63-84

Belfroid A, Seinen W, Van Gestel K, Hermens J, Van Leeuwen K (1995) Modelling theaccumulation of hydrophobic organic chemicals in earthworms: application of the equilibriumpartitioning theory. Environ Sci Pollut Res 2:5-15

Cadée GC (1978) Sediment-omwerking door depositfeeders. Meded Werkgr Tert Kwart Geol15:85-100

Cadée GC (1979) Sediment reworking by the polychaete Heteromastus filiformis on a tidalflat in the dutch Wadden sea. Neth J Sea Res 13:441-456

Ciarelli S, Van Straalen NM (1997) The role of bioturbation by the estuarine amphipodCorophium volutator in the fate of sediment-bound contaminants. Report nr. D97011, Dept.of Ecology and ecotoxicology, Vrije Universiteit (Amsterdam) commissioned by the NationalInstitute for Coastal and Marine Management (RIKZ)

Clements WH, Oris JT, Wissing TE (1994) Accumulation and food chain transfer offluoranthene and benzo(a)pyrene in Chironomus riparius and Lepomis macrochirus. ArchEnvironm Contam Toxicol 26:261-266

Daborn GR, Amos CL, Brylinsky M, Christian H, Drapeau G, Grant J, Long B, Paterson DM,Perillo GME, Piccolo MC (1993) An ecological cascade effect: Migratory birds affect stabilityof intertidal sediments. Limnol Oceanogr 38:225-231

Davey JT (1994) The architecture of the burrow of Nereis diversicolor and its quantificationin relation to sediment-water exchange. J Exp Mar Biol Ecol 179:115-129

Davey JT, Watson PG (1995) The activity of Nereis diversicolor (Polychaeta) and its impacton nutriënt fluxes in estuarine waters. Ophelia 41:57-70

DiToro DM, Zarba CS, Hansen DJ, Berry WJ, Swartz RC, Cowan CE, Pavlou SP, Allen HE,Thomas NA, Paquin PR (1991) Technical basis for establishing sediment quality criteria fornonionic organic chemicals using equilibrium partitioning. Environ Toxicol Chem 10:1541-1583

Everaarts JM, SaralaDevi K (1996) Cadmium distribution in sediment and the lugwormArenicola marina in a low concentration exposure experiment. Buil Environ Contam Toxicol57:771-778

Fisher JB, Lick WJ, McCall PL, Robbins JA (1980) Vertical mixing of lake sediments bytubificid oligochaetes. JGeophRes 85:3997-4006

Forbes TL, Kure LK (1997) Linking structure and funciton in marine sedimentary andterrestrial soil ecosystems: implications for extrapolation form the laboratory to the field. In;Ecological Risk Assessment of Contaminants in Soil, Chapman & Hall, London, pp. 127-156

Forbes VE, Forbes TL (1994) Integrated ecotoxicology: linking fate and effect within abiological hierarchy. In: Ecotoxicology in Theory and Practice. Ecotoxicology series 2.Chapman & Hall, London, pp. 149-183

Gerdol V, Hughes RG (1994) Effect of Corophium volutator on the abundance of benthicdiatoms, bacteria and sediment stability in two estuaries in southeastern England. Mar EcolProgSer 114:109-115

Gilbert F, Bonin P, Stora G (1995) Effect of bioturbation on denitrification in a marinesediment from the West Mediterranean littoral. Hydrobiol 304:49-58

Grant J, Daborn G (1994) The effects of bioturbation on sediment transport on an intertidalmudflat. Neth J Sea Res 32:3-72

Green AS, Chandler GT (1994) Meiofaunal bioturbation effects on the partitioning ofsediment-associated cadmium. J ExpMar Biol Ecol 180:59-70

Henriksen K, Rasmussen MB, Jensen A (1983) Effect of bioturbation on microbial nitrogentransformations in the sediment and fluxes of ammonium and nitrate to the overlaying water.Environmental Biogeochemistry 35:193-205

Karickoff SW, Morris KR (1985) Impact of tubificid oligochaetes on pollutant transport inbottom sediments. Environ Sci Technol 19:51-56

Keilty TJ, White DS, Landrum PF (1988a) Sublethal responses to endrin in sediment byLimnodrilus hoffmeisteri (Tubificidae), and in mixed-culure with Stylodrilus heringianus(Lumbriculidae). AquatToxicol 13:227-250

Keilty TJ, White DS, Landrum PF (1988b) Sublethal responses to endrin in sediment byStylodrilus heringianus (Lumbriculidae) as measured by a mcesium marker layer technique.AquatToxicol 13:251-270

Kramer KJM (1989) Toepassing van het WADSEDI model: gevoeligheidsstudie envoorspelling van de concentratie van contaminanten in Waddenzee sediment. Laboratory forApplied Marine Research (MT-TNO) Report R89/325

Kukkonen J, Landrum PF (1994) Toxicokinetics and toxicity of sediment-associated pyrene toLumbriculus variegatus (Oligochaeta). Environ Toxicol Chem 13:1457-1468

Kure LK, Forbes TL (1997) Impact of bioturbation by Arenicola marina on the fate ofparticle-bound fluoranthene. Mar Ecol Prog Ser 156:157-166

Landrum PF, Dupuis WS, Kukkonen J (1994a) Toxicokinetics and toxicity of sediment-associated pyrene and phenanthrene in Diporeia spp.: examination of equilibrium-partitioningtheory and residue-based effects for assessing hazard. Environ Toxicol Chem 13:1769-1780

McElroy AE, Farrington JW, Teal JM (1990) Influence of mode of exposure and the presenceof a tubiculous polychaete on the fate of benzo(a)anthracene in the benthos. Environ SciTechnol 24:1648-1655

Meadows PS, Meadows A (1991) The geotechnical and geochemical implications ofbioturbation in marine sedimentary ecosystems. Symp zool Soc Lond 63:157-181

Meadows PS, Tait J (1989) Modification of sediment permeability and shear strength by twoburrowing invertebrates. Mar Biol 101:75-82

Mulsow SG, Landrum PF (1995) Bioaccumulation of DDT in a marine polychaete, theconveyor-belt deposit feeder Heteromastus filiformis (Claparede). ChemEcoI 31:3141-3152

Paterson DM (1997) Biological mediation of sediment erodibility: ecology and pbysicaldynamics. In; Burt N, Parker R, Watts J (eds) Cohesive Sediments. John Wiley & Sons Ltd.,UK, pp. 215-229

Pelegrï SP, Blackburn TH (1994) Bioturbation effects of the amphipod Corophium volutatoron microbial nitrogen transformations in marine sediments. Mar Biol 121:253-258

Pelegrï SP, Nielson LP, Blackburn TH (1994) Denitrification in estuarine sediment stimulatedby the irrigation activity of the amphipod Corophium volutator. Mar Ecol Prog Ser 105:285-290

Rasmussen AD, Banta GT, Andersen O (1998) Effects of bioturbation by the lugwormArenicola marina on cadmium uptake and distribution in sandy sediments. Mar Ecol Prog Ser164:179-188

Riedel GF, Sanders JG, Osman W (1989) The role of three species of benthic invertebrates inthe transport of arsenic from contaminated estuarine sediment. J Exp Mar Biol Ecol 134:143-155

Thoms SR, Matisoff G, McCall PL, Wang X (1995) Models for alteration of sediments bybenthic organisms. Executive summary (Project 92-NPS-2) for Water Environment ResearchFoundation

Van de Bund W (1994) The effects of deposit feeders activity on bacterial production andabundance in profundal lake sediments, J North Amer Bent Soc 13:532-539

Veen MP, Warmenhoven HP, Kramer KJM (1989) WADSEDI, a mathematical model toestimate the distribution of pollutants in Wadden Sea sediments. Laboratory for AppliedMarine Research (MT-TNO), Report R89/255

Chapter 2

Effects of sediment bioturbation by the estuarine amphipodCorophium volutator on fluoranthene resuspension and transfer

into the mussel, Mytilus edulis

(paper published in Environmental Toxicology and Chemistry in February 1999)

Environmemai Toxiculogy and ChemiMry, Vol. 18. No. 2. pp. 318-328, 1999e 1999 SETAC

Primed in ihe USA0730-7268/99 $9.00 + ,00

EFFECTS OF SEDIMENT BIOTURBATION BY THE ESTUARINE AMPHIPODCOROPHIUM VOLUTATOR ON FLUORANTHENE RESUSPENSION

AND TRANSFER INTO THE MUSSEL (MYTÏLUS EDÜL1S)

SILVANA CIARELLI,*! Nico M. VAN STRAALEN,^ VINCENT A. KLAP,§ and ANNEMARIE P. VAN WEZEL||tNatiottaJ Institute for Coastal and Marine Management, P.O. Box 8039, 4330 EA Middelburg, The Netherlands

^Department of Ecology and Ecotoxicology, Vrije Universiteit, De Boelelaan 1087, 1081 HV Amsterdam, The Netherlands§FOM Institute for Atomic and Molecular Physics, Kruislaan 407, !O98 SJ Amsierdam, The Netherlands

IINational Institute for Public Health and the Environment. P.O. Box 1, 3720 BA Bilthoven. The Netherlands

(Received 17 March 1998: Accepted 1 June 1998)

Abstract—To betier understand the effects of bioturbation on partitioning and availability of sediment-bound contarmnants toinfaunal amphipods and mussels, experiments were carried out with fluoranthcne-spiked sediment. Treatments ineiuded differentdensities of the esiuarine amphipod. Corophium vntutator. Total suspended solids (TSS). particulaie organic carbon/particulateorganic matter (POC/POM) in overlying water, fluoranthene concentrations in sediment, pore water, overlying water, amphipods.and mussels were measured. Bioturbation significantly increased TSS and POC/POM concentrations in overlying water, and thiseffect became greater at higher animal density and longer exposure time. Mean total aqueous fluoranthene concentrations increasedfrom 2.40 to 4.1 and 5.45 |j.g/L in the controi, low-density, and high-density treatments. respectively, after 10 d of exposure. Theparticle-bound fraction of fluoranthene in the overlying water from the high-density treatment was two times higher than that fromthe low-density treatment. Bioturbation did not affect the partitioning of fluoranthene over suspended solids and water, nor did itaffect the concentrations in sediment and porc water. This was iliustrated by the constancy of sediment-intersmial water partitioninacoefficients (log £„,„,), sediment-overlying water partitioning coefficients (log KXUMI). and normaiized POC-water partitioningcoefficients (log Kfai). Uptake of fluoranthene by filter-feeding mussels (Myntus edulm) increased linearly with (he density ofbioturbating amphipods and with exposure time. The difference in concentrations of fluoranthene in mussels between the lovvesiand highest density of amphipods was more than a factor of two. Our resuks showed that bioturbation sigmficantly increases TSSconcentration in (he overlying water and consequenily the total aqueous concentration of sediment-bound contaniinants. which. aresubsequently accumulated by filter-feeders. The increased accumulation in mussels, at a more or less constant concentration in thewater, demonstrates the importance of bioturbation as a flux phenomenon and its role in the transport of resuspended sediment-bound contaminants to organisms in the aquatic food chain.

Keywords—Sediment bioturbationedulis

Corophium volutator Sediment-water partitioning Fluoranthene Myiilus

tNTRODUCTION

Polycyclic aromatic hydrocarbons (PAHs) are largely dis-persed in the aquatic environments, and they partition amongwater, sediments, interstitial water, and organisms. The greatestfraction of these contaminants is associated with sediments,benthic organisms, and pore waters [1,2]; concentrations inthese compartments are several orders of magnitude greaterthan those in overlying water [ 1], Sediment-bound PAHs, how-ever, can also be resuspended from bottom sediments into thewater column by physical (water current and wind induced)or biological (i.e., bioturbation) events [3], Direct uptake ofPAHs from water by fish and other pelagic organisms is gen-erally considered to be the main route of exposure [4). Indynamically bioturbated environments, however, where a sed-iment-water equilibrium does not occur and/or where uptakeefficiencies are variable, PAH accumulation from food andfrom resuspended sediment may be substantial [1,5].

Bioturbation involves a mixing of sediment particles, re-distribution wühin the sediment column, and sediment resus-pension in overlying water as a consequence of burrowing,feeding, defecation, and tube-building activities by benthic

* To whom correspondence may be addressed(s.ciarelli@rikz.rws.minvenw.nl).

organisms. These behaviors are known to produce physicalmodifications of sediment properties, such as changes in theparticle size [6], sediment porosity, and sediment shearstrength [7,8]. With their burrow structures, benthic inveite-brates are able to increase the total surf ace area for diffusiveexchange of oxygen, nutrients, and redox potential at the sed-iment-water interface [9,10],

Macroinvertebrate bioturbation is also known to change thepartitioning of sediment-bound contaminants in different com-partments such as sed̂ nent profiles [11], pore water [12], andthe water column [2,13,14], Physicochemical changes also oc-cur within the pore water due to oxidation of sulfide [15].

The importance of bioturbation to the transport and avail-ability of sediment-bound contaminants is still poorly under-stood and not quantified [11,16]. ïf bioturbation affects sorp-tion coefficients and accumulation levels, as hypothesized byBelfroid et al. [16], the equilibrium partitioning theory (EqP)may over- or underestimate ecological risk assessment. Sed-iment-water distributions of contaminants are often predictedusing the EqP model and assuming that sorption of compounds,at equilibrium, is mainly regulated by their hydrophobicity andby the organic matter in the sediment. Another assumption isthat bioaccumulation in organisms is solely determined by thefreely dissolved concentration of a compound in the water

Effects of sediment bioturbation by Corophium volutator Environ. Toxicol. Chem, 18, 1999 319

phase and by the Upid content of the species [17]. When theEqP model has been used to predict the concentrations ofcompounds in pore water on the basis of sediment' concentra-tions, deviations from the expected results have been describedby several authors. Observations that the bioavailability ofcompounds also depends on animal behaviors (feeding strat-egy, animal density, and burrowing activity), and that thesemay also account for an under- or overestimation nf sorptioncoefficients and accumulation levels, are reported by Landrumet al. [18] and Belfroid et al. [19),

In this study, the effects of sediment bioturbation activityby the estuanne amphipod Corophium volutator on the en-vironmental fate and availability of Suoranthene, a commonnonpolar organic compound, were investigated. This specieslives in muddy sediments, constructs U-shaped burrows in theupper 5 cm of the sediment, and plays an important role inphysical-chemical exchange processes at the sediment-waterinterface. Corophium volutator is known to cause upward anddownward transport of sediment particles during tube con-struction [20], to change sediment permeability, water content,and shear-strength [8,21], to stimulate fluxes and exchangesof oxygen, ammonium, nitrate, and phosphate among sedi-ment, pore water, and overlying water [10], and to increasemineralization processes [22]. The ecotoxicological conse-quences of C. volutator bioturbation in partitioning and reg-ulating bioavailability of sediment-bound contaminants havenot yet been investigated.

To better understand the consequences of bioturbation forsediment-bound fluoranthene resuspension in the water columnand its availability to aquatic organisms, the blue mussel, My-Hlus edulis, was also used in this study. The blue mussel isfrequently used as an indicator species for environmental pol-lution in marine ecosystem monitormg programs [23,24].Through their sessile and filter-feeding behavior, mussels areable to accumulate large amounts of contaminants and are thusvery useful indicators of environmental stress [25].

The specific objectives of this study were twofold. (1) Thefirst was to investigate the infJuence of bioturbation on par-titioning of fluoranthene in sediment, pore water, overlyingwater, and amphipods as a function of C. volutator densityand exposure time. Sediment-water, sediment-pore water, andparticulate organic carbon (POC)-water partitioning coeffi-cients were estimated in different treatments, with and withoutbioturbation. (2) The second objective was to study the roleof C, volutator bioturbation on the trophic transfer of fluor-anthene from sediment to mussels in the overlying water atfour different densities and at five different tiroes of exposure.Accumulation of fluoranthene in the amphipods over time wasalso determined to examine rate processes.

MATERIALS AND METHODS

Collection af experimental organisms and control sediment

Corophium volutator (Crustacea: Amphipoda) was col-lected from a relatively unpolluted intertidal mudflat (Oester-put) located in the Oosterschelde in the southwest of The Neth-erlands (51°36'N, 3°48'E). Sediment was taken as grab sam-ples and wet-sieved through a 500-(j.m-mesh sieve; the am-phipods were rinsed into polyethylene buckets containingseawater. Amphipods were then transported to the laboratoryand transferred to 10-L jars filled with natural filtered seawater(i.e., sandbed-filtered seawater containing particles <1Q (xm)and containing a 3-crn layer of sediment from the sanv lo-

cation. Organisms were acclimatized to the same salinity, tem-perature, and light conditions as used in the experiments.

Mussels (Mytilus edulis) were collected from a relativelyunpolluted subtidal area located in the Oosterschelde in- thesouthwest of The Netherlands. In the laboratory, mussels withshell lengths between 4 and 4.5 cm were selected and cleansedof epibionts. The mussels were acclimatized for 1 week inrunning seawater prior to the start of the experiment.

Spiking procedure and sediment preparation

Sediments for the experiments were collected from the samesite as the amphipods. The sediment was wet-sieved througha 500-jxm mesh to remove indigenous macroinvertebrates,transported to the laboratory, and stored at 4°C. Samples ofsediment were taken for analyses of water content and organicmatter. The sediment was dried to a constant weight at 70°Cfor 24 h. Percentage organic matter was determined after com-bustion at 450°C for 3 h. Physicochemical characteristics andbackground concentrations of PAHs, polychlorinated biphen-yls (PCBs), metals, and organochlorine pesticides of the samesediment are described in Ciarelli et al. [26]; in this sedimentthe background concentration of fluoranthene was <0.25|i.g/g dry wt. (unpublished data).

The spiking technique described here is based on a modifiedversion of the method described by Ditsworth and Schults [27].Fluoranthene (log Kav. - 5.23) [28] was purchased from Al-drich Chemical (Steinheim, Germany; purity 98%) and dis-solved in acetone. A calculated volume of this stock solutionwas injected into each vessel (1-L amber bottles) containing-450 ml of wet sediment to achieve the required nominalconcentration. Each vessel was filled with —450 ml filteredseawater and placed onto a rolling machine for —48 h to allowgood mixture and partitioning of the compound into the sed-iment. The vessels were then stored at 4°C for 10 d to allowequilibration prior the start of the experiment. The fluorantheneconcentrations were chosen on the basis of an unpublishedstudy in which a 10-d 50% lethal concentration (LC50) valueof 50 |j.g/g dry wt. was found for C. volutator and the sublethalconcentration (nommaUy 15 u.g/g dry wt.) used in the presentexperiment was shown to cause no effect, Before starting theexperiments, the water overlying the sediment was decanted,the sediment from each vessel was pooled into a larger poly-ethylene container, and the pooled sediment was placed ontoa rolling machine again for 2 h to obtain a homogeneousmixture. Two samples were taken to measure the fluorantheneconcentration in the sediment and in the pore water prior tothe beginning of the experiment.

Analytical procedures

Fluoranthene concentrations were determined in sediment,overiying water, pore water, and the organisms.

Fluoranthene from sediment samples was Soxhlet extractedwith an acetone/hexane mixture (1:1). The extracts werewashed with deionized water to remove hexane, dried withNa2SO4, evaporated, and the remaining hexane was blown offunder a gentle stream of nitrogen.

The residues were dissolved in 4 ml methanol/ethanediol.For analysis of fluoranthene, the extracts were automatkallyinjected into a high-performance liquid chromatography(HPLC) system with an LC-PAH C,B reverse-phase column,and concentrations were quantified by ultraviolet (UV) absor-bance and fluorescence detection (FD). The filters used to sep-arate suspended solids from water were also analyzed for fluo-

10

320 Environ. Toxicol. Chem. 18, 1999 S. CiareIJi et al.

ranthene using a method of extraction and cleanup comparableto that described for the sediment. The concentrations wereexpressed as ng/L based on the amount of water that camethrough each filter.

Sediments were also analyzed for total organic carbon(TOC) using a carbon-hydrogen-nitrogen (CHN; Carlo Erba,Milan, Italy) elemental analyzer. Sediment samples were firstdried at 100'C and then combusted at l,380°C to produce CO3,which was measured using an infrared detector. Inorganic car-bon was removed by acidiflcation with H3PO41 and the CO2

produced was subtracted from the total.Fluoranthene concentrations in the overlying water were

determined in the filtrate of water subsamples after filtrationthrough preweighed and pre-ashed (500"C) glass fiber filters(type GF/C, Whatman, l-u,m nominal pore size). Pluoranthenewater samples were extracted twice with 100 ml of hexane.Extracts were dried and evaporated under nitrogen. Residueswere dissolved in methanol, and fluoranthene concentrationswere quantified by HPLC-UV/FD.

Pore water samples were collected at the end of the ex-periment, after the removal of organisms and overlying water.Sediment from each replicate was homogenized, centrifuged,and then dried for 30 min at 1,200 g to obtain pore watersamples. Analysis of fluoranthene in pore water followed aque-ous sample procedures.

Amphipod samples were Soxhlet exiracted (boiling time 1h, rinsing time 0.5 h) with 70 ml of hexane:acetone (1:1). Theextracts were rinsed with deionized water to remove acetone.The hexane extract was dried at 25°C and concentrated to ~1.5ml under nitrogen in a Turbo-Vap evaporator. The residueswere dissolved in —5 ml of methanol and further concentratedto ~ 1 ml, Fluoranthene analyses were perforreed by HPLC-UV/FD, as above.

Musset tissues were dried with Na2SO4 after removal fromtheir shells. The two mussels from each treatment were pooledand homogenized in a stainless steel blender. The homogenatewas again dried with NajSO4 and placed into a desiccator for1 night to remove excess water. The dried samples were ex-tracted in a Tecator® Soxtec apparatus (boiling time I h, rinsingtime 2 h) in 70 ml hexane. The extract was then rinsed withdeionized water to remove the hexane. The extracts were driedand evaporated under nitrogen, and residues were dissolvedin 1 ml methanol, Fluoranthene in mussel tissues was measuredusing the same method as for amphipods.

Total suspended solids (TSS) and POC in the overlyingwater were determined in the filtrate of water subsamples afterfiltration through preweighed and pre-ashed (500°C) glass fiberfilters (type GF/C, 1-M-m nominal pore size). Total suspendedsolids were determined gravimetrically in the low- and high-density treatments after filtration of overlying water samplesof 250 and 150 ml, respectively. Filters were dried at 50°C for48 h and weighed. The concentration was expressed as mg/Lby calculating the difference between the total and initialweight and dividing by the volume of water filtered. The filterswere then analyzed for POC. The method involved the high-temperature oxidation of organic material to CO2 by flash com-bustion at 1,000-1,800°C using a CHN Carlo-Erba elementalanalyzer. The CO2 produced was analyzed by a thermal con-ductivity detector, Concentrations were expressed as mg/L andrelated to the volume of water filtered for the TSS determi-nations. Total suspended solids and POC were not measuredin the control containers because the overlying water was com-pletely clear.

All chemical analyses were carried out by OMEGAM, ac-credited as a testing laboratory by the Laboratory Accredita-tion Board of The Netherlands. Each set of samples was an-alyzed under quality assurance (QA) protocols, which includedprocedural blanks, replicate analyses, and control materials.Identification and quantifkation of fluoranthene were per-formed by comparing retention times and peak areas with thoseof certified standards (Standard Reference Material andSchmidt, Amsterdam, The Netherlands). Recoveries of fiuo-ranthene in water and sediment ranged between 85 and 110%,and Standard deviations were <20%.

Estimation of partitioning coefficients

The apparent partitioning coefficients of fluoranthene be-tween the organic carbon from the sediment and overlyingwater (log K^^)) and between the sediment and interstitialwater (log Kcc(m) were calculated from the dissolved concen-tration of fluoranthene and the normalized organic carbon con-centration in the sediment on a dry weight basis. Thus

and

" o' *<x

where #<*,<,„,) and Kxim, are the partitioning coefficients (L/kgorganic carbon) for sediment/overlying water and interstitialwater, respectively, Cs is the fluoranthene concentration in thesediment on a dry weight basis ((Jig/kg dry wt.), CTO is thefluoranthene concentration in the overlying water ((ig/L), C,„is the fluoranthene concentration in interstitial water (fig/L),and fx is the fraction organic carbon in the sediment. Theapparent POC water partitioning coëfficiënt (K^; L/kg) wascalculated as follows:

-•TSS

where CTSS is the TSS-bound or particle-bound fluoranthene(u-g/L), Cow is the dissolved fluoranthene concentration in theoverlying water (|u,g/L), and poe is the particulate organic car-bon (POC) concentration (kg/L, dry wt.). The dissolved frac-tion of fluoranthene in the interstitial and overlying water isbased on the sum of the freely dissoJved fluoranthene and thefractions bound to dissolved organic matter. The freely dis-solved fraction was not measured in this study.

Experimental design

In all experiments, 5-L glass aquaria were filled with ~900ml of spiked sediment and 3,500 ml of overJying seawater.Sediment and water were aliowed to equilibrate for 24 h priorthe addition of amphipods in the different treatments,

The first experiment was carried out in March 1997 andinvolved the following three treatments: (1) control, withoutorganisms; (2) low density, i.e„ 100 amphipods per aquarium;and (3) high density, i.e., 300 amphipods per aquarium. Low-and high-density treatments corresponded to 3,300 and 9,900organisms per m2, respectively. Three replicates were used foreach treatment and exposure time. The experiment lasted 10d (control, low density, and high density) and 30 d (controland low density).

In the second experiment, carried out in August 1997, fivetreatments were used: (1) control, without amphipods and with

11

Effects of sediment bioturbation by Corophium voluiator Environ. Toxicol. Chem. 18. 1999 321

two mussels alone; (2) very low density, L,e„ 40 amphipodsper aquarium; (3) low density, i.e., 80 amphipods per aquar-ium; (4) high density, i.e.( 160 amphipods per aquarium; and(5) very high density, i.e., 320 amphipods per aquarium. Thenumbers of amphipods used in this experiment correspondedto 1,334, 2,668, 5,335, and 10,670, orgaiusms per m2, respec-tively (treatments 2-5). Two mussels were added to all aquaria48 h af ter amphipods had been introduced. Mussels were col-lected from each aquarium foliowing 2, 4, 8, 13, and 18 daysof exposure and analyzed for fluoranthene residues. At the endof the 20-d experiment, overlying water samples (1 L each)were taken for TSS and fluoranthene analyses.

Quality parameters (temperature, dissolved oxygen, pH,and salinity) for the overlying water were monitored at thebeginning, after 5 d, and at the end of the experiment. Aerationin each vessel was provided by Pasteur pipettes connected tothe pressurized air system of the laboratory. The experimentswere carried out under static conditions at a constant watertemperature of 15°C; dissolved oxygen in all vessels rangedbetween 70 and 100% saturation and was checked daily. Am-phipods and mussels were not fed during the course of theexperiments. At the end of the experimental period, the sed-iment from each test chamber was passed through a 500-M-m-mesh sieve, and the survivors of each treatment were counted.Survivors of both density treatments in experiment 1 and thelow (n = 80) and very high (rt = 320) density treatments inexperiment 2 were rinsed in deionized water, placed in poly-ethylene vials, freeze-dried for 24 h, and then analyzed forfluoranthene. Dry weight measurements were also performedon the survivors from experiment 1.

Data analyses

Differences in fluoranthene concentrations in all compart-ments (i.e., sediment, pore water, overlying water, and testorganisms) and differences in POC and TSS concentrationsamong treatments and between 10 and 30 d were subjected toone-way analysis of variance (ANOVA), folio wed by pairwisecomparisons of means. The data were log-transformed first toobtain normal distributions. Differences in dry weight anddifferences in POC and TSS concentrations among treatmentsand among different experimental times were analyzed by Tu-key's honestly significant difference test. Differences fromcontrols in fluoranthene concentration and comparisons withcontrols for sediment-water partition coefficients during thetwo different experimental time periods were analyzed usingDunnett's test. Relationships between animal density and TSSconcentration and between animal density and fluorantheneconcentration in the mussels over time were determinedthrough regression analyses using the multivariate general lin-ear hypothesis module of SYSTAT 5.0 (M.A. HilL Evanston,IL, USA). Differences were considered to be significant at p^ 0.05.

Because it was assumed that bioturbation increases the fluxof fluoranthene into the water in proportion to the number ofbioturbating amphipods, the increased fluoranthene concen-tration in mussels was described by two linear components—one indepeadent of bioturbation and the other depending onthe number of bioturbating amphipods—as follows:

Cm(t) = C0 + a X t + n X b X t

where Cm(t) is the concentration of fluoranthene in mussels(lAg/g) at different times (t = days), Co is the backgroundconcentration, a is the bioturbation-independcnt uptake rate

control low highdensity density

control lowdensity

Fig. 1. Mean (±SD) Total Suspended Soiids (TSS) in overlying water(mg/L) after 10 and 30 d of exposure in the bioturbation treatments.* = higheT than iow density treatment (p •£ 0.05); •* = higher thanlow and high density treatments (T = IOd;ps 0.05).

(jj.g/g/d), b is the bioturbation-dependent uptake rate (jxg/g/amphipod/d), and n is the number of amphipods. The param-eters a and b were estimated from the experimental data by aleast-squares regression routine, using the Nonlin model fromthe SYSTAT 5.0 software package.

RESULTS

Overlying water

In the first experiment, the TSS concentration in the bio-turbation treatments ranged between 49.6 (low density) and144.6 (high density) mg/L after 10 d. After 30 d, TSS valuesin the low density treatment were almost a factor of 4 higherthan those found after 10 d (Fig. 1). The burrowing activityof C. volutator significantly increased the TSS concentrationwith higher density and over time (Fig. 1). Particulate organiccarbon concentrations also increased significantly with higherdensity, from 2.7 (low density) to 5.3 mg/L (high density).After 30 d, POC had increased significantly compared to thehigh- and low-density treatments at the 10-d (t = 10) values(1.4- and 2.6-fold, respectively). The percentage of POC (POC/TSS) after 30 d, however, dïd not change substantially in thehigh-density treatment compared to the t = 10 values (Table1). A positive correlation was found, however, between POC

Table 1. Particulate organic carbon (POC) in overlying water after10 and 30 d in the low- and high-density treatments (mean ± Standarddeviation). Values for POC relative to total suspended soiids (TSS)

are given in percentages

Duration Experiment 1 POC (mg/L) (POC/TSS) X 100

10 d

30 d

ControlLow density

(n = 100)High density

(n = 300)ControlLow density

(n = 100)

BD»2.77

5.31

BD7.29

(±0.34)Ab

(±0.90)B

(±0.90)C

BD5,68

3.68

BD3.78

<2l.20)A

(ï0.09)B

(ï0.13)BC

a BD - below the limits of detection (practically zero).b Means that share a common letter are not significantly different from

eacii other at p •& 0.05.

12

322 Environ. ToxtcoL Chem. 18, 1999

EZ 2

10 d T = 30 d

# *

• I I n Icontrol low high control low

density density density• • • > dissolved I I particulate

Fij». 2. Mean total fluoranthene (FluJ concentration in overiying waterafter 10 and 30 d of exposure in controls and bioturbation treatments.The dissolved and particle-bound fractions are given in microgram*per liter by different shades in the bars. In the controls, only the mean(±SD) of the dissolved fraction is given, as the particle-bound fractionwas not possible to detect. * = higherthan control (p •& 0.05): ** =higher than control and low density treatment (p s, 0.0S).

and TSS values for both exposure times and for both biotur-bation treatments (f- = 0.962; p ~ 0.0001). Particulate organiccarbon and TSS values in the control treatments at 10 and 30d of exposure were below the limits of detection, and theoverlying water was completely clear.

S. Ciarelü et al.

Mean total overJying water concentrations of fluoranthene(i.e., dissolved plus particle-bound) in the bioturbation treat-ments ranged between 4.1 (low density) and 5.4 (high density)|j,g/L and were significantly higher than the control value (2.4u.g/L). A significant difference was also found between thetwo density treatments. After 30 d, the mean dissolved fractionof fiuoranthene in the overlying water decreased 3 to 15 timesfrom the low-density treatment (1.46 jJ,g/L) to the control (0.16)xg/L), respectively. Also in this case, the total fluorantheneconcentration in the low-density treatment was significantlyhigher than the control value. In contrast to the t - 10 treat-

' ments, however, dissolved fractions of fluoranthene were notstatistically different from each other due to a high variabilitybetween replicates in the control (Fig. 2).

The particle-bound fraction of fluoranthene (i.e,, bound toTSS) ranged between 0.91 (low density) and 2,2 (j.g/L (highdensity) after 10 d. After 30 d (t = 30), the partiële fractionof fluoranthene in the low-density treatment was higher thanthe value measured at t = 10 (1.4 |i.g/L; Standard deviation[SD] = 0.3 u.g/L). This means that 22, 40, and 95% (low- andhigh-density treatments at t = 10 and low-density treatmentat t = 30. respectively) of the total fiuoranthene concentrationwas particte-bound (Fig. 2).

In the second experiment with mussels, TSS and particulateorganic matter (POM) values in the overlying water rangedfrom 8.4 to 97 and from 2.8 to 16 mg/L, respectively. Thedifferences in TSS and POM between the very low and thevery high density treatments were on an average 6- and 3.5-fold, respectively (Table 2). The regression lines in Figure 3

Table 2. Total suspended solids (TSS), particulate organic mauer (POM), percentage of POM, andfluoranthene concentration in overlying water from experiment 2 after different exposure times. The

treatments consisting of different densities are given by the numbers of amphipods used at t = 0

Days ofexposure

2

4

8

13

18

Densities ofamphipods

04080

160320

04080

160320

04080

160320

04080

160320

04080

160320

TSS3

(mg/L)

9.3A"13A21B60C

8.4A12A29B43B

10A25B42C95D

9.IA30B97C35B

I6A43B73C86D

POM(mg/L)

4A4AB5.4AB9.3C

2.8A3.5 AB5.7AB6.4B

3.4A5.2A7.3A

16B

3.7A6.9AB

15B9 0AB

4.0A7.8A

12A13A

POM(%)

42332715

34281915

32201717

40231526

25181615

Fluorantheneconcn. inoverlying

water (fj,g/L)

0.063.33.2ND'61.72,63.6ND4.4i.42.44ND6.81.73.04.16.14.91.1NDND4.22.9

n AH results are based on duplicates with the exception of fluoranthene concentrations, which are basedon single values.

b Means that share a common letter are not significantly different from each other at p £ 0.05.c ND = not determined.

13

Effecls of sediment bioturbation by Corophium valutator

u

s?ti,

eriy

in

o.2

is

uu -

75-

50 •

25-

0-

T=2d

r - - " | 1 H

^ ^^ - ^ ^

TSS =

R : =

1 1

0 192n

0 934

1

Sesc

C/l

100 T

75--

50 ••

25--

100 T

100 T

50 100 150 200 250

Numbers of amphipods

300

50 100 150 200 250

Numbers of amphipods

300

50 100 150 200 250Numbers of amphipods

300

0 50 100 150 200 250 300

Numbers of amphipods

Fig. 3. Relationship between the amphipod density in sediment andthe Total Suspended Solids (TSS) in overlying water at different ex-posure times. Constants. x coefficients (represented by the number ofamphipods [«]), and squared regression coefficients (/f3) are given foreach regression line. The total number of samples for each time is 8.

show that the quantity of resuspended particles in the overlyingwater increased linearly with animal density and over time, asin the first experiment without mussels. Concentrations offluoranthene in the overlying water generally increased withanimal density and over time, although not consistently (Table2). The high variability among the data was probably caused

Environ. Toxicol. Chem. 18. 1999 323

Table 3. Organtc carbon (OC)-novmaliied sediment concentrations offluoranthene. given in mg/kg OC dry wt.. after 10 and 30 d (mean ±

standard deviation)

Experiment 1 t 10 d t = 30 d

ControiLow densityHigh density

615 (±75)'624 ( i l 19)661 (±32)

640 (±13)537 (±97)

NDb

'The sediment concentrauon prior the start of r1" experiment was532 (±11).

h ND = nol deiermined.

by the fikration activity, uptake, and elimination by the mus-sels.

Sediment

The mean percentage of dry matter for the fluoranthene-spiked sediment in the first experiment was 57.4 <SD = 1,7).The mean percentage of organic carbon in all treatments was1.95 (SD = 10.1) and did not change throughout the exposuretime of 30 d.

The mean measured fluoranthene concentrauon at t = 0,prior to the start of the experiment, was 13 (±0.28) (Jtg/g drywt. Mean measured sediment concentrations of fluoranthenein the three treatments ranged between 12.1 and 12.6 andbetween 10.8 and 13 |xg/g dry wt, after 10 and 30 d of ex-posure, respectively. Sediment concentrations for al] treat-ments and both experimental times were normalized to organiccarbon and were not statistically different from each other orfrom the values measured at t = 0. This suggests that biotur-bation did not affect the total concentrations of fluoranthenedunng the period of exposure (Table 3).

The sediment used in the second experiment was charac-terized by a mean percentage dry matter of 61.4 % (SD =0.08) and mean percentage organic carbon of 2.6% (SD =0.21). The fluoranthene concentration in the sediment (i.e,, 10p.g/g dry wt.) was measured only bef ore starting the experi-ment,

Pore water

Pore water concentrations of fluoranthene after 10 d rangedbetween 18 and 22 \i$lh and were not significantly differentamong the different treatments. Concentrations in the controi,low-, and high-density treatment were 12, 16, and 19% higher,respectively, than the pore water concentration at t = 0 (16.50± 0.50 |xg/L). After 30 d, concentrations were 2 to 2.5 timeslower than the values obtained after 10 d. Pore water concen-trations at both exposure times in the bioturbation treatmentswere not statisücaHy different from the controls.

Sediment-wa. r and particulate matter-water panitioncoefficients

Mean organic carbon normalized sediment-water partition-ing coefficients for the overlying water (log Kwwl) and inter-stitial water {log Ü ,̂,*,) and particulate matter-water coeffi-cients (log K^) were calculated for each replicate experi-mental unit, and the means per treatment are given in Table4. Sediment-overlying water and sediment-interstitial waterpartitioning coefficients (log KKmm and log K„.lM), respective-ly) of the bioturbation treatments were not statistically differ-ent from controi values. No difference in log Xpoc values wasfound among bioturbation treatments after 10 d. After 30 d,significantly higher (log K^,^, [approximately two orders of

14

324 Environ. Toxicol. Chem. 18, 1999 S. Ciarelli et al.

Table 4. Partitióning coefficients for the distribation of fluorunthene over sediment artd overlying water(log Kxmt), sediment and pore (intefstitial) water (log ^ „ „ J , and over particulate organic carbon and

overlying water (log A^), normalized to organic carbon (OC: mean i SD) after 10 and 30 d

Time Experiment 1log *«,„,(L/kg OC)

log Ar„„„,(L/kg OC) <L/kg OC)

l O d '

30 <

ControlLow densityHigh densityControlLow density

5.41 (+0.06) A5.28 (±0.11)A5.28 (±0,02)A8.11B7.95 (±0.08)B

4.51 (zO.Q2)A4.50 (±0.01 )A4.51 (ïO.O2)A4.99 (±0.22)A4.79 (±0.04)A

ND"5.04 (±0,03>A5.13 (±0.02)AND6.49 (±0,03)B

Means that share a common letter do not differ signiflcandy from each other at p < 0.05.a Log KKI0W and log A'K|11VI values are .based on three replicates and K^- values are based on iwo replicates.b ND, not determined, as particuJate organic carbon was practically zero.c Number of replicates is two for both treatments except for the control, which is based on a single

value.

magnitude] and log ATpw. [more than one order of magnitude])values were found compared to the t = 10 values; no effectof bioturbation, however, was found compared to the control(Table 4), Higher log K^^ a»d l°g *M>« values at t = 30 weremainly due to the decline of dissolved concentrations of fluo-ranthene in the overlying water over the course of the exper-iment. Values of log ATOC(W) for both treatments were compa-rable to those found at t = 10.

Recovery, dry weight ofC. volutator, and fluorantheneavailability

The resutts concerning recovery and dry weight of C. vol-utator exposed to fluoranthene-spiked sediment for 10 and 30d showed that the concentration used did not affect the survivaland growth of the organisms. Percentage recovery ranged be-tween 88.3% (t = 30) and 90.7% (t = 10). Dry weight valuesincreased with exposure time in both bioturbation treatmentsand were significantly higher than the value measured at r = 0

(0.31 ± 0.05), indicating growth during the experimental pe-riod of 30 d (Table 5).

Fluoranthene concentrations in C, volutator ranged be-tween 86.5 (high density) and 97.8 (low density) jxg/g dry wt.after 10 d and were sigm'ficantly different between the twodensity treatments. Concentrations of fluoranthene decreaseddramaticalJy, to 4.7 u,g/g dry wt. after 30 d (Table 5).

In the second experiment, the percentage amphipod recov-ery ranged from 64 to 94% and from 59 to 90% in the lowand in the very high density treatments, respectively. MortaJityincreased with animal density and was higher than that ob-served in experiment 1 (Table 5). This was probably due to ahigher sensitivity of the summer population of amphipods, aswas shown in Ciareili et al. [26]. The number of bioturbatingamphipods at a certain exposure time was estimated from theinitial and final numbers by linear interpolation in time. Fluo-ranthene concentrations in the amphipods ranged from 33 to92 (Ag/g dry wt. and from 51 to 120 (ig/g dry wt. in the low-

Table 5. Percentage recovery, dry weight (milligrams per individuai), and concentration of fluoranthene(micrograms per gram of dry wt.) in C, volutatar after 10 and 30 A'

Time (d)

Experiment10

30

Experiment2

4

8

13

18

Density

1

Low (n •= 100)

High (n = 300)

Low

2Low (n » 80)Very high (n = 320)LowVery highLowVery highLowVery highLowVery high

Recovery (%)

90.7 (2.5)(r = 3)

89.4(1.9)( r= 3)

85.3 (3.9)(r = 3)

94908880857864738159

Fluorantheneconcn.

(|tg/g dry wt.)

97.81 (4.0)A(r = 3)

86.55 (1.6)B( r - 3)4.7 (l.l)C(r = 2)

921208488

1007475593351

Dry weight(mg/ind.)0

0.47 (0.06)A(r = 6)

0.52 (0.09)A(r = 9)

0.63 (0.09)B(r = 6)

ND

ND

ND

ND

ND

a Values are expressed as means ± standard deviation. Means that share a common letter do not differsignificamly from each other at p •£ 0.05. In experiment 2, the results shown are based on single values.

b ind. = individuai; n - number of amphipods used in each density treatment; r = number of replicates(r = 1 in experiment 2); ND = not determined. In all treatments, 20 animals were used for eachreplicate. Dry weight at t = 0 was 0.31 (±0.05). All dry weight values are significantly higher thanthe value of t = 0 at p s 0,05.

15

Effects of sediment bioturbation by Corophium volutator Environ. Toxicol. Chem, IS, 1999 325

density and in the very high density treatments, respectively.Concentrations decreased by a factor of ~-2 after 20 days.Accumulation did not differ significantly among the differenttreatments (Table 5).

Fluoranthene uptake by mussels

For all exposure times, the results showed a Ünear increaseof fluoranthene in mussels with increasing amphipod density(Fig. 4). With the exception of t = 13 (data not shown), thedifference in concentrations between the very low density andthe very high density was more than a factor of 2. Relationshipsbetween fluoranthene concentrations in the mussels and timeof exposure are shown in Figure 5. The overall experimentaldata were fitted to a linear model to estimate the total uptakeof fluoranthene depending on the background concentration(Co) and the uptake rates <a [bioturbation-independent] and b[bioturbation-dependent]). Estimates (plus 95% confidence in-tervals) for Co, a, and b are given in Table 6. Concentrationsof fluoranthene in mussels (Cm) and the resulting regressionUnes for each density treatment were calculated on the basisof the above estimated values of Co and the uptake rate con-stants.

DISCUSSION

Fluoranthene partitioning

The bioturbation activity of C. volutator significantly in-creased TSS and POC/POM concentrations in the water col-umn. A linear increase was observed with amphipod densityand with time. The total Öuoranthene concentration in the over-lying water also increased significantly with bioturbation in-tensity. This finding is comparable to those of several authorswho also described a positive relationship between animal den-sity and sediment-bound cc-^minant resuspension in the wa-ter column. Riedel et al. [13] found that the transport of arsenicout of contaminated sediment in the water column increasedsignificantly with animal density. They found increases byfactors of 4 to 10 using the marine invertebrates Macomabalthica (clam), Nereis succinea (polychaete worm), and Mi-crura leidyi (nemertean worm). They also suggested thatsmaller burrowing organisms, Iike the nemertean worm, aremore efficiënt than larger ones m causing the release of sub-stances from sediment due to their greater surface area tovolume ratio [13]. Clements et al. [2] showed that bioturbationby chironomids at high density was directiy responsible forbenzo[a]pyrene resuspension by a factor of 2.5 in the overlyingwater. A significant increase of TSS (by a factor of 4) andcadmium concentrations (by a factor of 3) in the water columnas a consequence of koi carp bioturbation was also found byWall et al. [14]. The authors, however, found a weak correlationbetween large koi carp size and TSS concentrations. AlthoughPetr [29] suggested that the impact of bioturbation should bea function of the body size of the animal, its activity, and itsdepth penetration, the studies mentioned above demonstratedthat higher density and smaller body size were more effectivethan biomass per se.

In our experiments, bioturbation did not influence fluo-ranthene concentrations in the sediment during the experi-mental period of 30 d in the different treatments. This indicatesthat no substantial depletion had occurred due to uptake bythe amphipods or stimulated degradation by bacteria. The porewater concentration of fluoranthene was also unaffected bybioturbation after 10 and 30 d. Although pore water concen-

60 T

50 100 150 200

Numbers of amphipods

250 300

60 r

50 100 150 200

Numbers of amphipods

250 300

au •

B * 4 0 -

'S 'S 20 -

0-

T = 8d

u*--*-—1 1—

Cm = 0.075n + 7.998R' = 0.907

• - ~~~~~*

1 1 1 1

50 100 150 200

Numbers of amphipods

250 300

* 4 0 - •

2 0

m-16.19!

- h50 100 150 aoo

Numbers of amphipods

250 300

Fig. 4. Relationship between the amphipod density in sediment andthe fluoranthene (Flu) concentration in mussel at different exposuretimes. Constants. x coefflcients (represented by the number of am-phipods [«]), and squared regression coefflcients (Z?2) are given foreach regression line. The number of samples for each exposure timeis 5.

trations were a factor of 2 lower after 30 d, the sediment-interstitial water coefficients remained constant among treat-ments and over time; they were within the range of laboratory-derived values (i.e., 4.92 to 5) reported in the literature for

16

326 Environ, Toxicol. Ckem. 18. 1999 S. Ciarelli et at.

I

DU -

50-

40-

30-

20-

10-

1

0-

X

X

< ^•zz%~- •

(

•

1

J> n=320

^ ^ ^ n=160

— • ^ ^ ' ^ ^ ^ ^ ~ n=80

-*~ . n=40

n=Q

10

Time (days)

15 £0

Fig, 5. Relationship between the fluoranthene (Flu) body burdens in mussel (Cmj and the exposure time foreachdensity treatment. The experimentaldata for each density treatment (n) are given by different symbols. The regression lines are calculated on the basis of the estimated values ofCo, a and b (see Materials and Methods).

fluoranthene [28,30]. This was in agreement with the reportby Landrum et al. [31], who found no difference in sediment-interstitial water partition coefficients for pyrene and phenan-threne between sediments with and without organisms after28 d exposure and no difference between different aging times,

Sediment-overlying water partition coefficients (log /£„.,„„>)and particulate matter-water coëfficiënt distributions (logKfK), calculated after 10 d of exposure, did not differ signif-icantly among control and bioturbation treatments and amongdensity treatments, respectively. Due to a decline in the dis-solved fluoranthene concentration for the overlying water, logKpiK and log KBelaw) values calculated after 30 d were more thanone to two orders of magnitude higher, respectively, than thosecalculated for t = 10. As commonly stated, numerous exper-imental conditions, such as exposure time, the presence orabsence of animals [31,32], sediment aging [31], the use ofdissolved versus freely dissolved compounds, and the approx-imation of steady state are factors that should be taken intoaccount when partitioning coefficients are calculated [17,33].ït is generally believed, however, that the organic carbon nor-malized sediment-water partition coëfficiënt for nonionic or-

ganic compounds is essentially equivalent to the log KM, andthat the log Km is approximately equivalent to its log AT0C [17].Our log ^oclow) values from the bioturbation treatments werevery close to the log Kow of fluoranthene and to the log KrKC

values calculated at t = 10. Despite the higher organic carbonnormaüzed sediment-water partitioning coefficients found af-ter 30 d, the range of log KKlmi) described here (5,3-8.1) over-lapped with the range (i.e., 5.9-7.2) reported by Van Hattum[32] for fluoranthene in field-contaminated sediments. The re-duced aqueous concentration of fluoranthene jn the pore waterand the dissolved fraction in overlying water after 30 d wasprobably caused by microbial degradation. De Maagd [33],for example, concluded from sorption studies of PAHs thatbiodegradation is probably a f aster process than desorption.The particle-bound fraction of fluoranthene after 30 d, in con-trast, was almost comparable to that after 10 d, indicating thatwhen fluoranthene is sorbed to organic particles, its availabilityto bacteria is probably reduced. This was in agreement withthe finding of Weissenfels et al. [34], who reported lower deg-radation of fluoranthene by bacteria and fungi when fluoran-thene was sorbed to the organic carbon in soil particles.

Table 6. Estimates for Co, a, and b (plus 95% confldence intervals) obtained by fitting fluorantheneconcentrations in musseis to the model Cm (t) = Co + a.t. + n.b.t

Parameters Symbol" Estimate95% Confidence

interval Units

Background concentration Ca 7.297 5.250-9.344Bioturbation-independent rate a 0.478 0.230-0.705Bioturbation-dependent rate b 0.0109 0.0092-0.0127

|xg/g/dng/g/amphipod/d

a Co = background concentration; a = bioturbation-independent uptake rate; b = bioturbation-dependentuptake rate.

17

Effects of sediment bioturbation by Corophium volutator Environ. Toxicol. Chem. 18. 1999 327

Fluoranthene bioavailability to C. volutator and M. edulis

In the present study, the densities of C. volutator used werewithin the range of natural field densities [35] and comparableto the chironomid densities used in the bioturbation experi-ments described by Clements et al. [2). Although authors suchas Riedel et al. [13] and Clements et al. [2] reported signifi-cantly higher body burdens in higher density treatments, thefluoranthene concentrations in C volutator from the differenttreatments in our study were very close to each other in bothexperiments, The decrease in body burdens for C. volutatorafter 18 d (experiment 2) and after 30 d (in experiment 1) maydepend on a combination of different factors. Driscoll et al.[36] and Harkey et al. [37] also reported a rapid initial uptakeduring the first period of exposure and then a decline of fluo-ranthene in Hyalella azteca after 10 d. This was explained byrapid depletion of the more readiiy bioavaüable fraction ofSuoranthene (i.e., the dissolved fluoranthene) from the porewater until a steady state was reached between pore water andbody burden. Because desorption from sediment particles maynot be rapid enough to maintain the initial pore water con-centration, bioavailability declines [37].

Another explanation for lower body burdens after 30 d maybe the ability of C. volutator to metabolize and eliminatefluoranthene. The capability of PAH biotransformation by am-phipods has been shown by Driscoll et al. [36] for H. aztecaand Diporeia after water-only exposures to various concen-trations of fluoranthene (elimination half-lives of 3-6 h and7-25 d, respectively). In addition, biotransformation by Eo-haustorius washingtonianus and Rhepoxynius abronius after1 week of exposure to benzo[a]pyrene and other aromatichydrocarbons was reported by Varanasi et al- [38],

Fluoranthene body burdens in the mussels were positivelycorrelated with animal density (with the exception of the valuesobtained for t = 13). A rapid and linear Lncrease in PAHaccumulation was also found in studies by Pruell et al. [4] andEertman et al. [25], who exposed marine mussels to a sus-pension of contaminated sediments for 40 and 30 d, respec-tively. Resuspension of PAHs in the water column wasachieved in both studies by mechanically stirring the sediment.Pruell et al. [4] found that concentrations of fluoranthene inmussels were more than a factor of 100 lower after 20 d ofexposure than those obtained in our study. Concentrations insediment and in overlying water were also lower, by a factorof 10 and at least a factor of 100, respectively, compared toour study [4], Eertman et al. [25] also found a lower accu-mulation of fluoranthene in mussels (by a factor of ~40) after30 d of exposure, although the concentrations of fluoranthenein the water column were within the same range as that usedin this study (i.e., 1-6 u,g/L). The differences among theseresults are not surprising, however, considering that variousfactors, such as the sediment-water ratio, the number of mus-sels in the water column, the concentrations of contaminantsin the sediment, and the way in which these are bound (i.e.,spiked versus field-contaminated sediments) can infiuence theoutcomes of these experiments.

The time to reach steady state and the elimination rates mayvary among studies [1]. Pruell et al. [4] found that musselsaccumulated a large portion of the PAHs after 10 d; between10 and 20 d the tissue burdens increased slightly, and between30 and 40 d concentrations started to decline. The authorsreported a half-Iife for fluoranthene of 30 d. The reduced con-centrations of the parent compounds were explained as pos-

sibly resul ting from metabolic activation or from a decline inthe pumping and uptake rate due to the toxic effect of PAHaccumulation. In our study, because body burdens were stillincreasing over time in the nonbioturbated treatment, and es-peciafly in the bioturbated treatment with the highest numberof amphipods, steady-state conditions did not seem to havebeen achieved after 18 d. Although fluoranthene concentrationsin the overlying water were more or less constant, musselscontinued to accumulate fluoranthene over time. This suggeststhat mussels received their body burden not only from thedissolved phase of fluoranthene .., a consequence of diffusiveilux across the sediment-water interface but also from theparticle-bound fraction as a consequence of sediment biotur-bation, which increases the diffusive exchange process be-tween the two phases. This was in contrast with the reportsof Meador et al. (1] and Pruell et al. [4], who stated that thedissoived phase is the main source of PAH for species thatfilter large quantities of water, even though a very smal! per-centage of PAHs are dissolved.

The sediment mixing and resuspension efficiencies of bio-turbation activity by the different densities of amphipods usedin our study demonstrated that C. volutator can simulate thedegree of sediment resuspension that occurs naturally (inducedby wind or currents) or may be produced by mechanical stir-ring. The ranges of resuspended particles and fluorantheneconcentrations in the water column obtained in our study wereindeed comparable to those reported from a field study byEertman et al. [39]. In their study, mussels were exposed tocontaminated sediments of the Westerschelde estuary (south-western Holland) for 6 weeks (39).

The implications of sediment bioturbation with regard tothe transfer of contaminants through the food chain, however,still seem to be ambiguous and poorly understood. Clementset al. [2] studied the influence of midge larvae bioturbationon the accumulation of fluoranthene and benzo[ajpyrene inbluegill, and Wall et al. [14] analyzed the influence of benthickoi carp bioturbation on cadmium transfer to Daphnids, butneither group reported significant effects due to the resuspen-sion of contaminants from the sediment.

Our work demonstrated a significant effect on mussels dueto the resuspension of sediment-bound fluoranthene, It alsoshowed that mussels are very suitable organisms for studiesconcerning sediment-water exchange processes because oftheir ability to filter large amounts of water.

CONCLUSIONS

The results of this study showed that bioturbation by C.volutator significantly affected the concentration of suspendedsolids in the overlying water and consequently increased thetotal aqueous concentration of fluoranthene. Sediment biotur-bation had greau: consequences for the water column anduptake by filter-feeders than it did for the sediment, pore water,and the amphipods themselves.

Sediment-water partitioning coefficients calculated fortimes of 10 or more days in this study were in good agreementwith log Kw values reported in literature. Bioturbation doesnot contribute to the discrepancies between predicted and mea-sured chemical concentrations that are often found when porewater concentrations are predicted from the EqP model.

As a result of C. volutator bioturbation activity, fluoran-thene body burdens in the mussels increased linearly withamphipod density and over time of exposure. This observationdemonstrated the importance of bioturbation in sediment-

18

328 Environ. ToxicoL Chem. 18. 1999

bound contaminant resuspension and trophic transfer to or- 18ganisms in the aquatic food chain, Due to their limitated ca-pacity for elimination and/or metabolization of PAHs, theirsessile behavior, and their ability to filter large amounts ofwater, mussels may be among the organisms most affected by 19.resuspension of sediment-bound contaminants.

Acknowledgement—We thank the National Institute for Coastal and 20.Marine Management for the financial support of this study. J. Pietersfor the helpful preparation of figures, and J. Jol for technical assis- 21.tance.

REFERENCES 22.1. Meador JP, Stein JE, Reichert WL. Varanasi U. 1995. Bioaccu-

mulaiion of polycyclic aromatic hydrocarbons by marine organ-isms. Rev Environ Contam Toxicol 143:79-165. 23.

2. Clements WH, Oris JT, Wissing TE. 1994. Accumulation andfood chain transfer of fluoranthene and benzo(a)pyrene in CM- 24.ronomus riparius and Lepomis macrochirus. Arch Environ Con-tam Toxicol 26:261-266.

3. NSf C. Axelman J. Broman D. 1996. Orgamc contaminants insediments of the Baltic Sea: Distribution, behavior and fate. InMunawar M, Dave G. eds, Developmenr and Progress in Sedi- 25.ment Quahly Assessment: Rationale. Challenges, Techniquesand Strategies. SPB-Academic. Amsterdam. The Netherlands. pp15-25.

4. Pruell RJ, Lake JL. Davis WR. Quinn JG. 1986. Uptake and 26.depuration of organic contaminants by blue mussels Mytilus ed-ulis exposed to environmentatly contammased sediment. MarBial91:497-507. 27

5. McEIroy AE. Farrington JW. Teal JM. 1990. Influence of modeof exposure and the presence of a tubiculous polychaete on thefate of benzo(a)anthracene in the benthos Environ Sa Technol 28.24: [648-1655

6. Krantzberg G. 1985, The influence of bioturbation on physical.chemical and biological parameters in aquatic environments: Areview, Environ Pollui 39:99-122.

7. Meadows PS, Tait J. 1989. Modification of sediment permeability 29.and shear strength by two burrowmg invertebrates. Mar Biol 10!:75-82.

8. Meadows PS, Meadoi.. A. 1991. The geotechnical and geo- 30.chemical implicaiions of bioturbatioti in marine sedimentary eco-systems, Symp Zool Soc Lond 63:157-181.

9. Aller RC. 1988. Benthic fauna and biogeochemical processes in 31.marine sediments: The role of burrow structures. In BlackburnTH, Sorensen J, eds, Nitrogen Cychng in Coastal Marine En-vironments, John Wiley & Sons, New York, NY. USA, pp 302-338. 32.

10. Pelegri SP, Nieison LP, Blackburn TH. 1994. Denitrification inestuarine sediment sumulated by the irrigation activity of theamphipod Corophium volutator. Mar Ecol Prog $er 105:285- 33.290.

11. Forbes VE, Forbes TL. 1994. Integrated ecotoxicology: Linkingfate and effect within a biological hierarchy. In Depledge MH, 34.Sanders B, eds, Ecotoxicology in Theory and Practice, Series 2,Ecotoxicology. Chapman & Hall, London, UK, pp 149-183.

12. Green AS, Chandler GT. 1994. Meiofaunal bioturbation effectson the partitioning of sediment-bound cadmium. J Exp Mar Biol 35.Ecol 180:59-70.

13. Riedel GF, Sanders JG, Osman W. 1989. The role of three speciesof benthic invertebrates in the transport of arsenic from contam- 36.inated estuarine sediment. J Exp Mar Biol Ecol 134:143-155,

14. Wall SB, Isely JJ, La Point TW. 1996. Fish bioturbation of cad-mium-contaminated sediments: Factors affecting Cd availability 37.to Daphnia magna, Environ Toxicol Chem 15:294-298

15. Peterson GS, Ankley GT, Leonard EN. 1996. Effect of biotur-bation On metal-sulfide oxidation in surfïcial freshwater sedi- 38.ments, Environ Toxicol Chem 15:2147-2155.

16. Belfroid AC, Sijm DTHM, Van Gestel CAM. 1996. Bioavail-ability and toxicokinetics of hydrophobic aromatic compounds inbenthic and terrestrial invertebrates. Environ Rev 4:276-299. ' 39.

17. Di Toro DM, et al. 1991. Technical basis forestablishing sedimentquality criteria for nonionic organic Chemicals using equihbriumpartitioning. Environ Toxicol Chem 10:1541-1583.

S. Ciarelli et al.

. Landrum PF, Dupuis WS, Kukkonen J. 1994. Toxicokinetics andtoxicity of sediment-bound pyrene and phenantrene in Diporeiaspp.: Examination of equilibrium-partitioning theory and residue-based effects for assessing hazard. Environ Toxicol Chem 13'1769-1780.Belfroid A. Seinen W, Van Gestel K. Hermens J, Van LeeuwenK. 1995. Modeling the accumulation of hydrophobic organicChemicals in earthworms: Application of the equilibrium parti-tioning theory. Environ Sci Pollul Res 2:5-15.Grant J, Daborn G. 1994. The effects of bioturbation on sedimenttransport on an intertidal mudflat. Net/i J Sea Res 32:3-72.Jones SE. Jago CE 1993, In situ assessmem of modification ofsediment propemes by burrowtng invertebrates. Mar Biol 115'133-142.Pelegri' SP. Blackburn TH 1994. Bioturbation effects of the am-phipod Corophium volutator on microbial niirogen transforma-tions in marine sediments Mar Biol 121:253-258.Widdows J. Johnson D. 1988. Physiological energetics of Mytiluseduhs: Scope for growth. Mar Ecol Prog Ser 46:113-121.Smaal AC, Knoester M. Nienhuis PH, Meire PM. 1991. Changesin the Oosterschelde ecosystem induced by the Delta works. InElliott M. Ducrotory JP. eds. Estuaries and Coasts: Spatial andTemporal intercomparisons. Olsen & Olsen. Fredensborg, Den-mark, pp 375-384.Eertman RHM. Groenmk CLFMG. Sandee B. Hummel H. SmaalAC. 1995. Response of the blue mussel Mynlus editlis L, fol-lowtng exposure to PAHs or contaminated sediment. Mar EnvironRes 39:169-173.Ciarelli S. Vonck APMA. Van Straaten NM. 1995. Reproduci-bility of marine spiked-sediment toxicity (ests using the benthicamphipod, Corophium volutator. Mar Environ Res 4:329-343.Ditsworth GR. Schulis DW. 1990. Preparation of benthic sub-strates for sediment toxicity testing. Environ Toxicol Chem 9'1523-1529De Maagd PGJ, Ten Hulscher DThEM. Van den Heuvel H. Op-perhuiïen A. Sijm DTHM. 1998. Physicochemical propemes ofpolycyclic aromatic hydrocarbons: Aqueous solubitites. H-octan-ol/water parntion coefficients.. and Henry's law constants, EnvironToxicol Chem 17:251-257.Petr T. 1977. Bioturbation and exchange in the mud-water in-terface. In Golterman HL, ed. Interface Between Sediment andFreshwater. Junk, The Hague, The Netherlands.Swam RC. ei al. 1995. Total PAH: Model to predict the toxicityof polynuclear aromatic hydrocarbon mixtures in field-collectedsediments. Environ Toxicol Chem 14:1977-1987.Landrum PF, Eadie BJ, Faust WR. 1992. Variation in the bio-availability of polyclic aromatic hydrocarbons to the amphipodDiporeia (spp.) with sediment aging, Environ Tnxicol Chem 11-1197-1208.Van Hattum AGM. 1995. Bioaecumulation of sediment-boundcontaminants by the freshwater isopod Asellus aquaticus. PhDthesis. University of Amsterdam, Amsterdam, The Netherlands.De Maagd PGJ. 1996. Polycyclic aromatic hydrocarbons: Fateand effects in the aquatic environment. PhD thesis. Universi.y ofUtrecht, Utrecht, The Netherlands.Weissenfels WD, Klewer HJ, Langhoff J. 1992. Adsorption onpolycylic aromatic hydrocarbons (PAHs) by soil particles: Influ-ence on biodegradability and biotoxicity Appl Microbial Bh-technol 36:689-696.Flach EC. 1992. The influence of four macrozoobenthic specieson the abundance of the amphipod Corophium volutator on tidalflats of the Waddeti sea. Neth J Sea Res 29:379-394.Driscoll SK, Landrum PF, Tigue E. 1997. Accumulation and tox-icokinetics of fluoranthene in water-only exposures with fresh-water amphipods. Environ Toxicol Chem 16:754-761.Harkey GA, Driscoll SK, Landrum PF. 1997. Effect of feedingin 30-day bioaecumulation assays using Hyalella azteca in fluo-ranthene-dosed sediment. Environ Toxicol Chem 16:762-769.Varanasi U, Reichert WL, Stein JE, Brown DW, Sanborn HR.1985. Bioavailability and biotransformation of aromatic hydro-carbons in benthic organisms to sediment from an urban estuary.Environ Toxicol Chem 19:836-841.

Eertman RHM, Wagenvoort AJ, Hummel H, Smaal AC. 1993."Survival in air" of the blue mussel Mytilus edulis L. as a sen-sitive response to pollution-induced environmental stress. Envi-ron Toxicol Chem 170:179-195.

19

Chapter 3

Effects of sediment bioturbation by the estuarine amphipodCorophium volutator on fluoranthene resuspension

and transfer into mussel, Mytilus edulis L.(poster presented in Charlotte -North Carolina in November 1998)

20

Effects of sediment bioturbation by the estuarineamphipod Corophium volutator on fluorantheneresuspension and transfer into the mussel Mytilus edulis

Mlnlstry cl Ttamport and Water Management

Directorato-Ccnoral for Public Works and Water Management

National Inssmte (orCoaiUI and Marine Manajtmenl <R!KZ)

Srfvana Oarellt1, Nico M. van Straaten1, and Belinds 1. Kater2

1 EJepartmiint et EM'ogy and etotomcalagv. Urllt UnLerslleil. D< BotWaan 1087, 1C81MV AmUeidam, The NdhsKin

JfJaliornl mirliile for Coostai anil M»mi Management «tK2>. FO Bm 3039,43J0 (f, Middelburg. The Ncttrcrlinds

AbstractThe effects of sediment bioturbation by the amphipod Corophiumvolutator on fluoranthene resuspension and uptake in thesuspension feeder Mytilus edulis were investigated. Five densitiesof C. volutator were used and amphipods were exposed to a no-effect level of fluoranthene (10 mg/kg d.w.}. Concentrations offluoranthene in overlying water, mussels and amphipods and totalsuspended solids in water were measured after 2, 4, 8 and 18days.

Results showed that amphipods bioturbation increased significantly totalsuspended solids in overiying water (2 to 4 times between the towest andhighest density) and concentration of fiuoranthene in the suspensionfeeding mussels (more than 2 times). Uptake of fluoranthene by musselsincreased linearly with the density of bioturbating amphipods and withexposure t/me. The increased accumufation in mussels demonstrafes theimportance of bioturbation as a flux phenomenon and its role in thetransport of contaminants to pelagic orgatiisms.

Results from comparable experiments as those here described incombination with literature data, may be used for modeling the dynamicbehaviour of PAHs in sediment-water systems that are not in equilibrium,Model simulations will provide a better insight into the importance ofbioturbation on the fate of sediment-bound contaminants.

JntroductlooBioturbation involves a mixing of sediment particles and redistributionwithin the sediment column and sediment resuspension in overlying wateras a consequence of tube-building, feeding, defecation activities bybenthic organisms. These behaviours are known to produce physical andchemical changes in sediments, pore water and overlying water.

Corophium volutator lives in " I T - shaped burrows that are continuouslyirrigated. By burrow construction and irrigation, total surface area fordiffusive exchange of oxygen and nutrients at the sediment-waterinterface is increased and mineralization processes are stimulated (Pelegriet af., 1994; Pelegri & Blackburn, 1$94). C. volutator bioturbation is ateoknown to change sediment permeability, water content and sedimentshear-strength (Crant & Daborn, 1994; Gerdol & Hughes, 1994).

Macroinvertebrate bioturbation also affects the fate and partitioning ofsediment-bound contaminants in sediment profiles (Kure & Forbes, 1997;Rasmussen et al., 1998), pore water (Green & Chandler, 1994) and watercolumn (Clements et al., 1994). Bioturbation is hypothesized to increasethe rate of important physicochemical processes that occur into thesediment and at the sediment-water interface as schematized in the modelhere below:

sorptlon

1. BiotwrbatiQrt made!

ellmlnaUon

TSS-boundcontaminaitts

Dissolved 'contaralnants

s-fo

' ' • .

AquatJcerganlsns '

degradation

sorptionM • tfmlww» Benthic^ M - f r - f<mwaia ^ ' ~ " ' > l organlsffls

I 1 *w^1 «iwnauon u^te

Hypotheticalmodel illustratingthe mostimportantphysicochemical

;—* sssss msediment-watersystems uponwhichbioturbation hasan effect

t

Objectives of the studyto investigate the effect of bioturbation on Total Suspended Soiids(TSS) in the overlying water as a function of C. volutator densityand exposure time

to study the role of C. volutator bioturbation on the trophictransfer of ftuoranthene from sediment to mussels in theoverlying water at five different densities and at five differenttimes of exposure

" Experimental design:- testchambers: 5-t aquaria filledwith 1-1 of spiked sediment

and 3,5-L seawater- Fluoranthene (Flu) concentration: 10 mg/kg d.w.- 5 densities: 0, 40, 80, 160, 320 amphipods per chamber (0

1.334, 2,668, 5,335, 10,670 m2}- exposure time: 18 (mussels) and 19 (amphipods) days- sampling times: 2, 4, 8 and 18 days after additon of mussels

" Test conditions: static system, temperature cpntrolled room: 15 ±2° C, 24 h light, salmity: 327<>°

* Test organisms:• Corophium volutator (Pallas) (Crustacea: Amphipoda)- Mytilus edulis (Linneus) (Mollusca: eivalvia)

* Parameters:- TSS concentration in water column- Flu concentration in mussels, amphipods and overlying water

upiake M =effeet of bioturbation

Corophiumvotutator

Picturesshöwing theexperimentalset-up

with mussels

21

ResultsTSS concentation in the overlying water increased linearly with increasingamphipod density and exposure time (Figure 2)

For all exposure times, Flu concentration in mussels increased linearlywith increasing amphipod density {Figure 3)

Assuming a linear relation between the number of amphipods and theflux ofFlu to the overlying water, the increase of fluoranthene in mussels overtime was described by fitting the following linear model to the data:

C m (t) = Co + a.t + n.b.t

where, C m (t) is the concentration of fluoranthene in the mussels

(mg/kg) at different times (t), Co is the background concentration, a is

the non-bioturbation-dependent uptake rate (mg/kg/d ), b is the

bioturbation-dependent uptake rate (mg/kg/amphipod/d) and n is the

number of amphipods (Figure 4).

* Bioturbation by C. volutator significantly increases TSS concentrations inoverlying water. The increase is positively related with the number ofbioturbating amphipods and with time

* Concentration of fluoranthene in overlying water remains more or lessconstant in all densities due to the filtration activity of the mussels (datanot shown). Concentrations of fluoranthene in mussels, in contrast,increase linearly with the number of amphipods in sediment and withtime of exposure due to bioturbation. Tissue concentrations in C.volutator were not significantly different among density treatments (datanot shown).

* This study shows the ability of C. volutator bioturbation to remobilizesediment-bound contaminants from sediment to organisms of theaquatic food chain and to stimulate fluxes at the sediment-water waterinterface

* Results of the present study may be used for modeling the dynamicbehaviour of PAHs in sediment-water systems that are not in equilibrium.Model simulations in the compartment system as showed above in the'Introduction' will provide insight into the effects of bioturbation in therate of various processes dealing with sediment-water exchanges and inthe fate of sediment-bound contaminants.

2, Effect «f bioturbation on TSS

1 Efféqt of frfot«rt>alion on fJuorafithene uptaic«

O 3ÜO:

BÜ 1OD t BO 30O 2SD 3OO-

:?;? RelatiöFtship between amphipod density; in sediment andy^liM flüDranthene{Fiü)conceBtra,Öonir!musse[5(pg/gwet ;.>£y$S wt> at differeat eKpostire tïme-s, The numbei; of samples.:^

•s,;.<

§^40H

'•:••«'•:.?

5 10Time (<fays|

20-

p t«twe$n amphipód dertsity In sediment at5dTota* Suspended Solfdï (TSS) i*i overiyiog water at dlffe^ntexposure ttfiaes- Ttie number of samples for eaeh time Is 8. •

p fluömnthene (Ftt)) body byr<fens iny lis (Cmï ati^ exposüfe-tlme for each density.

The expeümertial data for eacN density (n) m ^ *en bydiffererttsymbols. The'regression'iïnes of Cm are calculatedon the basis of the estimated values of CO, a and b <see'Resut'te')

RaferencesClements et al., 1994. Aren. Environ. Contam, Toxicol. 26:261-266.Gerdol & Hughes, 1994. Mar. Ecol. Prog, Ser, 114: 109-115.Grant & Daborn, 1994. Neth. J. Sea. Res. 32: 63-72.Green & Chandler, 1994. J. Exp. Mar. Biol, Ecol. 180: 59-70.Kure & Forbes, 1997. Mar. Ecol. Prog. Ser. 156:157-166.Pelegrf & Blackburn, 1994. Mar.eiol, 121: 253-2S8.Pelegri et at., 1994. Mar. Ecol. Prog. Ser. 105: 285-290.Rasmussen et al., 1998. Maf- Ecol. Prog. Ser. 164:179-188.

22

Chapter 4

The influence of bioturbation by the amphipod, Corophium volutatoron fluoranthene uptake in the marine polychaete, Nereis virens

(paper subraitted to Environmental Toxicology and Cheraistry)

23

The influence of bioturbation by the amphipod, Corophium volutatoron fluoranthene uptake in the marine polychaete, Nereis virens

(paper submitted to Environmental Toxicology and Chemistry)

Silvana Ciarelli,*Beiinda J.Kater,* and Nico M. van Straalen*

*Department of Ecology and Ecotoxicology, Vrije Universiteit, De Boelelaan 1087,1081 HV Amsterdam, The Netherlands

fNational Institute for Coastal and Marine Management, P.O. Box 8039,4330 EA Middelburg, The Netherlands

ABSTRACT

The uptake kinetics of fluoranthene in the polychaete worm, Nereis virens were investigatedin the presence and in the absence of amphipods to examine the effects of sedimentbioturbation by the benthic amphipod Corophium volutator on the uptake in worms. Wormsonly and worms together with two different densities of amphipods were exposed tofluoranthene-spiked sediment for 12 days. Worms and overlying water samples forfluoranthene analyses were taken and total suspended solids in water column were measuredafter 1, 2, 5, 8 and 12 days. Results showed that in all treatments fluoranthene was rapidlyaccumulated by N. virens during the first two days and a steady state was reached within fivedays of exposure. Biota to sediment accumulation factors normalized to lipid concentrationand to sediment organic carbon (BAFi0C) of worms exposed with the highest number ofamphipods were significantly higher (two to three times) compared to worms exposed withfewer or without amphipods after 1 and 2 days of exposure. After 12 days, due to amphipodmortality, accumulation in worms was only 35% higher compared to the treatment withoutamphipods. Bioconcentration factors (BCF), calculated as the ratio between the uptake (ki)and elimination (k2> rate constants were not significantly different among treatments. Whenbioconcentration factors were calculated on the basis of dissolved fluorantheneconcentrations, (BCFdiSS), values of the treatments where worms were exposed with 100 and300 amphipods, were 24 and 43% higher, respectiveiy, than those calculated on the basis oftotal (dissolved + particle-bound) aqueous fluoranthene (BCFtot). Total suspended solids inwater column of the treatment with the highest density of amphipods were generallysignificantly higher compared to that with the lowest number of amphipods. Also the totalfluoranthene concentration in water column was significantly higher in the treatments withworms and amphipods exposed together compared to the treatment without amphipods, after1 day. The results suggest that bioturbation by amphipods affected the concentration offluoranthene in the worms not by changing the worm/water partitioning (kt/k2) but bychanging the worm/sediment partitioning (B AF). The B AF values might be underestimated bythe fact that the presence of worms had a negative impact on the amphipods. In the treatmentswith worms a higher mortality of amphipods was found compared to those without worms.

Keywords - Corophium volutator Bioturbation Fluoranthene Toxicokinetics Nereis virens

24

INTRODUCTION

Soft-bottom marine environments in general have been shown to be strongly affected bybioturbation of benthic invertebrates [1]. Effects of benthic bioturbation on physical [2] andchemical characteristics of sediments [3,4], pore water and overlying water [5,6] have beenwell documented.

Bioturbation by the amphipod Corophium volutator is known to change sedimentpermeability, water content and sediment shear-strength [7,8]. Corophium volutator lives in"U"- shaped extending 3 to 4 cm into the sediment and which are continuously irrigated. Byburrow construction and irrigation, total surface area for diffusive exchange of oxygen andnutrients at the sediment-water interface is increased and mineralization processes arestimulated [9,10]. Macroinvertebrate bioturbation also affects the fate and partitioning ofsediment-bound contaminants in sediment profiles [11,12], pore water [13] and water column[14], Bioturbation is assumed to increase the rate of important physicochemical processes thatoccur at the sediment-water interface such as diffusion, desorption, degradation andresuspension of organic and inorganic compounds [2,15]. Studies quantifying and explainingthe effects of bioturbation on these processes through dynamic models are, however, scarce.Also studies on the role of bioturbation in contaminant transport from sediment to watercolumn and transfer to other trophic levels are lacking. A previous study showed thatsediment bioturbation by C. volutator affected the total concentration of Polycyclic AromaticHydrocarbons (PAH) in overlying water and increased the concentration of fluoranthene insuspension-feeding mussels [16]. The increase was linearly related to the number ofbioturbating amphipods in the sediment,

Based on the results of the previous investigation, the purpose of this study was toexamine whether bioturbation by C. volutator would exert an influence in fluoranthene uptakealso in the marine polychaete worm, Nereis virens. N, virens constructs deep (8 to 10 cm),vertical well-irrigated, semipermanent burrows lined with mucus and is known as anomnivorous nonselective deposit feeder [17] which can also filter feed [18]. Previous studieson bioaccumulation suggest that the water phase is an important route of uptake for theaccumulation of sediment contaminants in N. diversicolor [18,19]. AT. virens and C. volutatorare among the most common and abundant species in intertidal and shallow mudflats innorthwestern Europe. The species coexist in the same habitat, reach high population densitiesduring summer and construct burrows at the sediment-water interface [20,21],

The aim of this study was to compare the uptake kinetics of fluoranthene in N. virens in theabsence of amphipods with that obtained in the presence of bioturbating amphipods. Wetested the hypothesis that in a bioturbated system, desorption kinetics of sediment-boundcontaminants in water would be increased and that accumulation levels of fluoranthene inworms would be consequently higher than in non-bioturbated system. The role of amphipodbioturbation in fluoranthene bioavailability and the relationship between amphipod density,total suspended solids in overlying water and fluoranthene body burdens in worms were alsoinvestigated.

MATERIALS & METHODS

Field sampling and animals

Sediment samples from the top 3 cm were collected from an intertidal (relatively clean)mudflat (Oesterput) located in the Oosterschelde in the South-Eastern part of The Netheriands(51°36'N, 3°48'E). Sediment was wet-sieved through a 500-iim mesh sieve to removeindigeneous macroinvertebrates, transported to the laboratory, and stored at 4 °C.

25

Amphipods were collected from a (relatively unpolluted) intertidal mudflat(Biezelingseham) located in the Westerschelde in the South-western part of The Netherlands(51° 27'N, 3°55'E). Samples of the surface sediment (top 3 cm) were taken and wet-sievedthrough a 500-jam mesh sieve; the amphipods were rinsed into polyethylene bucketscontaining freshly collected sea water. Amphipods were then transported to the laboratory andtransferred to 10-L jars containing a 3 cm layer of (Oosterschelde) sediment filled with naturalfiltered seawater (i.e. sandbed-filtered seawater containing particles <10 jam) with a salinity ofapprox. 32 mg/L. Organisms were acclimated to the same salinity, temperature and lightconditions as used in the experiments.

Polychaete worms, Nereis virens (Sars, 1835; Polychaeta: Nereidae) were obtained from aseabait farm, Topsy Baits (Wilhelminadorp, The Netherlands). Worms were acclimated andheld in running seawater in tanks containing 2 to 3 cm of Oosterschelde sediment at atemperature of 15° ± 2 °C for 1 week before commencement of the experiment.

Sediment spiking procedure

Sediment subsamples were taken for analyses of water content and organic matter. Thewater content was determined by drying the sediment at 70 °C for 24 h to constant weight.The percentage organic matter was determined after combustion at 450 °C for 3 h.Physicochemical characteristics and background concentrations of total PAHs, PCBs, metalsand organochlorine pesticides of the sediment used are described in CiarelH et al. [22]; in thissediment the background concentration of fluoranthene was < 0.25 ug/g dry wt (unpubl.data).

Fluoranthene (Log Kov/ = 5.23) [23] was purchased from Aldrich Chemical Company(purity 98%) and dissolved in acetone (8 g/L). A calculated volume (66,375 mL) of the stocksolution was added dropwise with a rate of 60 mL/h to 60 kg wet sediment slurry to achievethe required nominal concentration (10 ng/g dry wt), while the mixture was stirred for approx.4 h. The fluoranthene concentration was chosen on the basis of a previous (unpublished) studywhich showed no effect to the amphipods at this exposure concentration. After spiking,sediment was kept in the refrigerator at 4°C for 10 days to allow equilibration and partitioningof fluoranthene into the sediment prior the start of the experiment. Before starting theexperiment, sediment was again mixed homogeneously for approx. 2 h and two subsampleswere taken for fluoranthene analysis.

Experimental set-up

3-L beakerglasses were filled with 850 mL of spiked sediment corresponding with a 4 cmthick sediment layer and 2000 mL of overlying seawater. Sediment and water were allowed toequilibrate for 24 h prior to the addition of the organisms, The experiment involved thefollowing three treatments: a) control without amphipods; b) low density of amphipods (i.e.,N=100); c) high density of amphipods (i.e., N=300). Low and high density treatmentscorresponded with 3300 and 9900 organisms m'2, respectively. The number of amphipodsused in this experiment was based on realistic environmental densities occurring in the fieldduring winter time [24]. Two worms ranging from 2.5 to 3.5 g wet weight were added to eachvessel 24 h after addition of the amphipods. Three replicates were used for each treatment andexposure time. In addition to the three treatments with worms, a low (i.e. 100 amphipods) andhigh density (i.e. 300 amphipods) treatment without worms (three replicates each) were usedto compare amphipod survival in the absence of worms.

Worms and amphipods were exposed to fluoranthene-spiked sediment for 12 and 13 days,respectively. The experiment was conducted at 15 °C and the overlying water was notreplaced during the exposure; temperature and dissolved oxygen were checked at thebeginning, after 5 days and at the end of the experiment, in the overlying water. The

26

organisms were not fed during the exposure because it was considered that the addition oforganic material would alter the partitioning and availability of fluoranthene.

Water samples for Total Suspended Solids (TSS) were taken after 2, 3, 6, 9 and 13 daysafter the addition of the amphipods. Water samples (1 L) for total fluoranthene (dissolved +partiële bound) analysis were taken after 1, 5 and 12 days after the addition of worms. Tissuesamples for fluoranthene analysis were taken from unexposed amphipods and worms beforethe experiment and from exposed organisms after 1, 2, 5, 8 and 12 days. At the end of eachexposure time worms were placed in clean seawater for 4 h to allow gut depuration. The waterwas changed twice during depuration time and the mucous-sediment mass excreted by theworms and sticking to their bodies was discarded. Tissue samples were then immediatelyfrozen in liquid nitrogen and stored at - 20° C . Amphipods were placed in clean seawater andrinsed with deionized water and frozen at - 20° C. Worms and amphipods were freeze-driedprior to fluoranthene analysis

Analytical Meihods

Concentrations of fluoranthene were determined in sediment, overlying water, in worms andamphipods.

Approximately 10 (± 0.01) grams of sediment were mixed with 10 mL of acetone for 10min on a rotary-mixer. Subsequently 10 mL of hexane was added and the sediment was mixedfor another 10 min. The acetone-hexane extract was centrifuged for 3 min at 500 rpm. Afteraddition of 20 mL of a sodiumsulphate solution the extract was shaken and centrifugatedagain. The hexane fraction was removed, dried over an anhydrous sodiumsulphate column andevaporated under nitrogen in a turbovap-apparatus. After addition of 3.5 mL of a 2%ethanediol solution in methanol, the extract was further reduced in volume to 1 mL usingnitrogen blow-down.

For fluoranthene determination in overlying water, water samples were adjusted to pH = 2using 33% nitric acid prior to extraction. After addition of 50 mL of hexane the sample wasshaken for 1 min in a separation funnel. The water fraction was collected into the samplebottle and the pH was adjusted to 9 using 5 M sodiumhydroxide. Subsequently hexane (100mL) was added in the water fraction and the mixture was shaken for 1 min in a separationfunnel. Both hexane fractions were combined and subsequently dried over an anhydroussodiumsulphate column. The dried hexane extract was evaporated under nitrogen in aturbovap-apparatus and further reduced to 1 mL after addition of 3.5 mL of a 2% ethanediolsolution in methanol.

Total Suspended Solids (TSS) in overlying water were determined gravimetrically in thelow and high density treatments after filtration of 200 and 100 mL of water, respectively. Pre-weighed and pre-ashed (500 X ) glass fiber filters (Type GF/C Whatman, 1 urn nominal poresize) were dried at 50 °C and weighed after 24 h. The concentration was expressed as mg/L bycalculating the difference between the total and initial weight and dividing by the volume ofwater filtered.

Amphipods and worms samples were Soxtec®-extracted for 1.5 hours with 70 mL of ahexane:acetone mixture (1:1). Amphipods and worms extracts were evaporated under nitrogenin a block heater (40EC) and residues were dissolved in 15 mL of hexane. The extracts wereconcentrated and reduced to approx. 1 mL in a turbovap-apparatus (30 EC) and after additionof 3.5 mL of a 2% ethanediol solution in methanol, extracts were further reduced to 1 mL.

For lipid content analyses, the worm extracts were cleaned-up over a 15-g aluminumoxide (6% H2O) column and eluted with 200 mL of hexane to remove lipids. The lipid weightof the samples was determined by evaporating a rïxed amount (10%) of the total Soxtec®extract and weighing the residue.

Analyses of the PAH compounds were performed by injecting 20 ui of methanol extractsinto a high performance liquid chromatograph (Separations HPLC-system). The PAH

27

compounds were separated over a Qg column, using a 56-100% acetonitrile gradiënt (0.7mL/min). PAHs were detected by compound-specific fluorescence-detection (Jasko 785A).All the PAH analyses were carried out by OMEGAM (Amsterdam), accredited as a testinglaboratory by the Laboratory Accreditation Board of The Netherlands

Uptake kinetics and Bioaccumulation Factors

The kinetics of the accumulation of fluoranthene in the worms were determined by fittingthe data to a first-order one-compartment toxicokinetic model assuming that the overlyingwater was the main route of uptake of fluoranthene for the worms and assuming the internalconcentration to be zero at t= 0:

whereCa (t) = fluoranthene concentration in worms at time t (ug/g)ki = uptake rate constant defined as the amount of fluoranthene accumulated per volume ofwater, per mass of organism per unit of time (mL g 1 d"1)k2 = elimination rate constant of fluoranthene (d'1)Cw = fluoranthene concentration in water column (mg/L)t = time (days)

The uptake (ki) and elimination (k2) rate constants were estimated by fitting the models to thedata using nonlinear curve fitting with Systat 5.0 (1990). The uptake clearance (ki) wasestimated for each treatment while the elimination rate constant (k2) was estimated for alltreatments together. Both constants were estimated twice, once, using the dissolved fraction offluoranthene only in water column and, the second time, using the total aqueous fluoranthene(i.e. dissolved + TSS-bound). The dissolved fraction was not measured in this study and wascalculated on the basis of results of a previous study [16] where analagous treatments (i.e.with 100 and 300 amphipods) and sediment concentration were used and where comparableconcentrations of TSS in overlying water were found. Results of this former study where thetwo fractions (dissolved and particle-bound) in overlying water were quantified, showed that78% and 60% of total fluoranthene in water column was present as dissolved in the low andhigh density treatments, respectively, af ter 10 d of exposure. For all treatments, thefluoranthene concentrations at each sampling time were pooled and the averages were used forthe estimation of the uptake (ki) and elimination (k2) rate constants.The following equations were used to determine the bioconcentration factors (BCFdiss andBCFtot), and the lipid (up) and sediment organic carbon (oc) normalized bioaccumulation factor(BAF]0C):BCFdiSS = ki /k2 (based on dissolved fluoranthene)BCFtot = ki/k2 (based on total aqueous fluoranthene)

= [(tissue/fiiP)/(sedoc)]

where, ki (mL.g'Vd'1) and k2(d-1) are the uptake and elimination rate constants as defined aboveand, tissue/fiip is the lipid normalized fluoranthene body burden (g g'1 dry wt) and sedoC, is thesediment organic carbon normalized fluoranthene (g g'1 dry wt), respectively.

28

Statistical analyses

Significant differences between means were tested with one-way analysis of variance(ANOVA) followed by pairwise comparisons among treatments (Tukey's HSD test). TSS andBAFioc data were checked for normality (Shapiro-Wilks test) and for homogeneity ofvariances (Bartlett's test) before performing analyses of variances. Differences wereconsidered significant when p < 0.05.

RESULTSSediment

Mean percentage dry matter of the fluoranthene-spiked sediment was 58.7 (± 3.3). Themean organic carbon percentage was 2.53 (± 0.5) and was calculated on the basis of thefollowing formula: f00 = 0.6 x fom [25]. Mean measured fluoranthene concentration insediment at t=0 was 8.3 (± 2.26) ug/g d. wt.

Percentage recovery of amphipods and fluoranthene uptake

Recovery of amphipods gradually decreased during the exposure time of 13 days. Meanpercentage recovery of amphipods after 1 day ranged from 90.7 (± 4.0) and 81.1 (± 6.9) in thelow and high density treatment, respectively. After 13 days, mean percentage recoverydeclined to 54 (± 2.9) and to 34.8 (±6.7) in the low and high density, respectively. In thetreatments without worms, mean recovery remained high with 92.3 % after 13 days ofexposure. This suggests a negative impact of worms on the survival of amphipods (Table 1).

Accumulation of fluoranthene in C. volutator was higher in the treatment with 300amphipods compared to the treatment with 100 amphipods. The internal concentrationsranged from 58 (± 33) to 90 (± 56) and from 93 (± 5) to 135 (± 50) ug/g, in the amphipods ofthe low and high density treatments, respectively. Due to the few data and the high SD, thedifferences between the two treatments were, however, not statistically significant. Based onthese results, steady-state seemed to be reached already after 2 days of exposure in bothdensities and after 13 days, fluoranthene concentration in the amphipods of the high densitytreatment, seemed to decline (Table 1).

Overlying water

Two days after addition of the amphipods to the test chambers, Total Suspended Solids(TSS) concentration in the low and high density treatments with worms were 38.5 (± 2) and106 (± 16) mg/L, respectively. After 13 days, TSS values declined to 34,8 (± 7) and to 39 (±4) mg/L in the low and high density treatments, respectively. A substantial decrease in time(almost a factor 3) was found in the high density treatment which coincided with a lowrecovery of amphipods. Nevertheless, TSS concentrations in the high density treatments weresignificantly higher than in the low density treatment for each sampling time with theexception of t = 13 days. In the treatments without worms, TSS values were almost a factor 2to 4.5 higher 70 (± 4.6) and 177 (± 7) mg/L in the low and high density, respectively), after 13days (Table 1). This was due to a better survival of the amphipods exposed without wormscompared to those exposed with worms.

At t - 1 day, total average fluoranthene concentrations in the water column ranged from 5.8\xgfL in the treatment without amphipods to 15 ug/L, in the high density treatment. After 13days, mean concentrations of fluoranthene decreased to 9.9 and 9.2 fig/L in the low and highdensity treatments, respectively, and were not significantly different among treatments exceptfor the values at t = 1 day.

29

Fluoranthene uptake in Nereis virens

Fluoranthene was rapidly accumulated by N. virens during the first two days of exposure in alltreatments. Steady-state body burdens were reached after 5 days in all treatments.Fluoranthene concentrations in the worms exposed with 300 amphipods were significantlyhigher (p< 0.05) than those in worms exposed without amphipods at t= 1, 2 and 12 days (Fig.1, 2 and 3). After 1 and 2 days, accumulation was also significantly higher compared toworms of the treatment with 100 amphipods. Although not significantly, worms that wereexposed with 100 amphipods accumulated more fluoranthene than those exposed withoutamphipods. Bioconcentration Factors calculated as the ratio between the uptake (ki) and theelimination (k^) rate constants obtained using the dissolved fluoranthene concentrations(BCFdiss) ranged from 4.68 to 4.87 (Log basis) in the treatment without and with 300amphipods, respectively (Table 2). BCFdiSS values of the treatment with 100 and 300amphipods were 24% and 43% higher compared to bioconcentration factors based on totalfluoranthene concentrations (BCFtot). As 95% confïdence limits of the uptake (ki) rateconstants estimated with both methods overlapped, BCFdiSS and BCFtot were probably notsignificantly different from each other and were also not significantly different amongtreatments. Although total concentrations in water and worms were significantly affected bybioturbation, the uptake kinetics and bioconcentrations factors were, however, not influenced.

BAFioc-biota to sediment concentrations normalized to total tissue lipids and to totalsediment organic carbon were calculated and reported in Table 3. BAFi0C values in wormsexposed with amphipods ranged from 0.1 to 1.6 during the exposure and were significantlydifferent (p<0.05) from those obtained for worms exposed without amphipods after 1,2 and 12days. Results support our hypothesis that amphipod bioturbation increased fluoranthenebioavailability for worms and enhanced uptake of fluoranthene from the water phase.

DISCUSSION

Toxicokinetics of fluoranthene in Nereis virens

Polychaete worms have been extensively used in several sediment toxicity studies [26,27]and in bioaccumulation studies [19,20,28,29] to assess effects of contaminated sediments.Most of these studies suggested that feeding habits of infaunal invertebrates andhydrophobicity of the compounds are important factors that influence the route of uptake ofsediment-bound contaminants. Under non-equilibrium conditions the route of contaminantuptake for the organisms can have a major influence on tissue concentration and can influenceuptake kinetics [30]. This study, however, is one of the few in which toxicokinetics in wormswere investigated in two different modes of exposure (i.e. in the presence and in the absenceof sediment resuspension by bioturbating amphipods) to assess the importance of the waterphase as a major route of uptake for the worm, Nereis virens. The results confirmed thehypothesis tested, i.e. that the worm's body burdens would be higher in the presence ofbioturbating amphipods than in their absence. The uptake (ki) and elimination ((k2) rateconstants and bioconcentration factors (BCF) were, however, not affected by bioturbatingamphipods. Fluoranthene resuspension in the water column, due to bioturbation, increased itsbioavailability for the worms suggesting that aqueous fluoranthene (either bound to TSS ordissolved) is an important source for uptake for N. virens. This fmding was consistent withresults of other studies that implied that N. virens is able to accumulate contaminants fromsediment via the water phase. Ray et al. [20] for example, found that uptake rates of cadmiumfrom water were 16 to 39 times higher than the uptake rates of cadmium from sediment. Theauthor stated that most of the cadmium accumulated by the worms exposed in sediment wasfrom release of cadmium to the water phase by activities of bottom-dwelling animals.

30

McElroy et ai. [19] also showed that benzo(a) anthracene (BA) added to the water column wasmore bioavailable for uptake and metabolism by worms than BA added to the sedimentreservoir. The authors also noted the ability of worms to remobilize BA in the water columnemphasizing the importance of mode of introduction of BA in the system for the fate andavailability of sediment-bound contaminants. Fowler et al. [21] found that uptake of PCBfrom water in the closely related polychaete, Nereis diversicolor, was more rapid compared tosediment and that concentration factors were two to three orders of magnitude higher thanthose based on uptake from sediment. Under normal environmentally conditions, however, theauthor stated that PCB from sediment is the principal source by which worms acccumulatedthese compounds.

In our experiment, the overlying water of treatments without amphipods was very clearsuggesting that sediment remobilization by worms might not have occurred during the totalexposure time. The presence of fluoranthene in the water column of the treatments withoutamphipods was probably due to diffusive exchanges between sediment and water stimulatedby the presence of burrows which can significantly increase the total sediment-water surfaceas was shown in studies of Gerino [31] and Davey [6,32].

The hypothesis that accumulation of PAH from overlying water can play an important rolein the uptake of fluoranthene is also based on the fact that most of the members of the familyNereididae are omnivorous, burrowing in the sediment in search of food and do notnecessarily obtain their food by ingesting sediment. Studies of Fauchald & Jumars [33] andSmith et al. [34] for example showed that Nereis diversicolor may be scavenger, suspensionfeeder or surface deposit feeder depending on the tide conditions and the availability of food.Since it has been shown that most of the burrowing organisms ventilate the burrows forrespiratory purposes by pumping oxygenated overlying water in their burrows [35] it is Ukelythat overlying water enriched with resuspended fluoranthene by bioturbating amphipods mighthave led to higher body burdens in the worms exposed with the amphipods.

The BCF values calculated in our study, even those based on dissolved fluorantheneconcentrations, may be underestimated, for two different reasons. The first reason is that onlythe freely dissolved fraction of fluoranthene, is probably mostly available for uptake in theworms. This, on the basis of a study carried out by Meador et al. [28] who reported higherBCF values (approx. a factor 5), for the amphipod, Rhepoxynius abronius when based on freeinterstitial PAH concentrations (BCFfoe) than on total ones (BCFtot). BCFfree were also closerto the predicted ones than BCFtot. The second reason for underestimated BCF values is thedeclining number of amphipods and TSS and Ukely of total fluoranthene in overlying water(see further in discussion).

Calculated BAF]0C values at steady state, varied from 0.8 to 1.4 in all treatments. Thesevalues are in accordance with those found by Meador et al. [28] for Armanda brevis (a non-selective deposit feeding polychaete) and Rhepoxynius abronius (a non-deposït feedingamphipod). In this study, where bioaccumulation of PAH by the two different infaunalinvertebrates was compared, the authors found that tissue body burdens of high molecularPAH were significantly higher in the polychaete compared to the amphipod. Sedimentingestion was suggested to be the dominant route of exposure for the polychaete, A. brevis.Fluoranthene and pyrene, however, seemed to behave more üke low moiecuiar PAH as thetwo species acquired nearly identical body burdens. The authors concluded that, for the lowmolecular PAH and, for fluoranthene and pyrene, the contribution of tissue burden bysediment ingestion was insïgnificant for the polychaete and that the interstitial water was themain route of uptake for both organisms. This observation in conjunction with the BAF|0C datawhich were comparable to our results, support the hypothesis of aqueous fluoranthene uptakeby N. virens.

31

Biotic interactions between N. virens and C. volutator

Results on percentage recovery of amphipods indicated that mortality was higher in thepresence than in the absence of worms suggesting that N. virens had a negative effect on C.volutator. Percentage recovery in the treatment with 100 amphipods was approximately 10%higher than the treatment with 300 amphipods. What type of interaction occurred between theamphipods and the worms in our experiment is not clear. The relationships between C.volutator and the closely related polychaete, N. diversicolor have been extensively studied.Some authors found that the two species may coexist without having negative effects on eachother [24,36] while others have recorded negative correlations between their densities andreported that N. diversicolor may cause disturbance in amphipod recruitment and induce theirmigration after destroying Corophium burrows and forcing them to construct new burrows[18]. Smith et al. [34] observed interspecific competition for food between the two specieswhen they coexist under natural conditions and feed on the same epipelic diatoms species inlaboratory experiments. Some authors [33,34] also reported the ability ofNereis species to actas predators. In our experimental system, we think that competition for food (diatoms) and thepresence of mucus secretions produced by the worms might have been the most importantfactors that caused stress and mortality of the amphipods rather than predation. If theamphipods were a real component in the diet of the worms, we would not have been able tosee the dead amphipods in the sediment at the beginning of the exposure (after 1, 2 and 5days) as we did. Due to the negative impact of the worms on the amphipods, TSSconcentration, total fiuoranthene concentration in overlying water and BCF and BAF valuesmay probably be underestimated.

CONCLUSIONS

The results of this study showed that the presence of bioturbating amphipods increasedbioavailability of aqueous fiuoranthene and consequently accumulation by the worm, N.virens. This was shown by the BAF^ values calculated at different times. This observationsuggested the importance of the overlying water as one of the most important routes of uptakefor tube-dwellers deposit feeders such as Nereis spp. The uptake and elimination rateconstants of fiuoranthene in the worms and bioconcentration factors (BCF) were however, notsignificantly affected by bioturbating amphipods. BCF and BAFi0C values may beunderestimated due the fact that the presence of worms in the sediment caused stress and rapidmortality of the amphipods.

Acknowledgement - This work was funded by the National Institute for Coastal and MarineManagement (RIKZ). We are also grateful to A. Hannewijk and P. Schout for technical assistance.We acknowledge Drs. BJ. Kater, Dr. A.D, Vethaak, Dr. K. Legierse, Dr. A. Belfroid and Dr. C.A.M,van Gestel for their valuable comments on the manuscript

REFERENCES

1. Townsend EC, Fonseca MS. 1998. Bioturbation as a potential mechanism influencingspatial heterogeneity of North Carolina seagrass beds. Mar Ecol Prog Ser 169:123-132.

2. Davis WR. 1993. The role of bioturbation in sediment resuspension and its interactionwith physical shearing. J Exp Mar Biol Ecol 171:187-200.

32

3. Meadows PS, Meadows A. 1991. The geotechnical and geocheraical implications ofbioturbation in marine sedimentary ecosystems. Sympzool Soc Lond 63:157-181.

4. Aller RC. 1988. Benthic fauna and biogeochemical processes in marine sediments: therole of burrow structures. In TH Blackburn, J Sorensen eds, Nitrogen Cycling in coastalmarine environments. Wiley, J & Sons Ltd, pp 302-338

5. Riedel GF, Sanders JG, Osman W. 1989. The role of three species of benthic invertebratesin the transport of arsenic from contaminated estuarine sediment. J Exp Mar Biol Ecol134:143-155.

6. Davey JT, Watson PG. 1995. The activity of Nereis diversicolor (Polychaeta) and itsimpact on nutriënt fluxes in estuarine waters. Ophelia 41:57-70.

7. Grant J, Daborn G. 1994. The effects of bioturbation on sediment transport on an intertidalmudflat. NethJSea Res 32:3-72.

8. Gerdol V, Hughes RG. 1994. Effect of Corophium volutator on the abundance of benthicdiatoms, bacteria and sediment stabüity in two estuaries in southeastern England. MarEcol Prog Ser 114:109-115.

9. Pelegrf SP, Blackburn TH. 1994. Bioturbation effects of the amphipod Corophiumvolutator on microbial nitrogen transformations in marine sediments. Mar Biol 121:253-258.

10. Pelegrf SP, Nielson LP, Bladcburn TH. 1994. DenitrificatJon in estuarine sedimentstimulated by the irrigation activity of the amphipod Corophium volutator. Mar Ecol ProgSer 105:285-290.

11. Kure LK, Forbes TL. 1997. Impact of bioturbation by Arenicola marina on the fate ofparticle-bound fluoranthene. Mar Ecol Prog Ser 156:157-166.

12. Rasmussen AD, Banta GT, Andersen O. 1998. Effects of bioturbation by the IugwormArenicola marina on cadmium uptake and distribution in sandy sediments. Mar Ecol ProgSer 164:179-188.

13. Green AS, Chandler GT. 1994. Meiofaunal bioturbation effects on the partitioning ofsediment-associated cadmium. J'Exp Mar Biol Ecol 180:59-70.

14. Clements WH, Oris JT, Wissing TE. 1994. Accumulation and food chain transfer offluoranthene and benzo(a)pyrene in Chironomus riparius and Lepomis macrochirus. ArchEnvironm Contam Toxicol 26:261-266.

15. Thomann RV, Merklin W, Wright B. 1993. Modeling cadmium fate at Superfund site:impact of bioturbation. J Environ Eng 119:424-442.

16. Ciarelli S, Van Straalen NM, Klap VA, Van Wezel AP. 1999. Effects of sedimentbioturbation by the estuarine amphipod Corophium volutator on fluorantheneresuspension and transfer into the mussel Mytilus edulis. Environ Toxicol Chem 18:318-328.

33

17. McElroy AE, Farrington JW, Teal JM, 1990. Influence of mode of exposure and thepresence of a tubiculous polychaete on the fate of benzo(a)anthracene in the benthos,Environ Sci Technol 24:1648-1655.

18. Ray S, McLeese D, Pezzack D. 1980. Accumulation of cadmium by Nereis virens. ArchEnvironm Contam Toxicol 9:1-8.

19. Fowler SW, Polikarpov GG, Elder DL, Parsi P, Villeneuve JP. 1978. Polychlorinatedbipheoyls: accumulation from contaminated sediraents and water by the polychaete Nereisdiversicolor. Mar Biol 48:303-309.

20. Newell RC. 1970. Mechanisms of feeding. In Elek P. (Scientific Books) Ltd, Biology ofintertidal animals. London, United Kingdom, pp 167-238.

21. Olafsson EB, Persson L-E. 1986. The interaction between Nereis diversicolor O.F.Mullerand Corophium volutator Pallas as a structuring force in a shallow brackish sediment, JExp Mar Biol Ecol 103:103-117

22. Ciarelli S, Vonck W, Van Straalen NM. 1997. Reproducibility of spiked-sedimentbioassays using the marine benthic amphipod, Corophium volutator. Mar Environ Res329:329-343.

23. De Maagd PJ, Ten Hulscher DThEM, Van den Heuvel H, Opperhuizen A, Sijm DTHM.1998. Physichochemical properties of polycyclic aromatic hydrocarbons:aqueoussolubilities, n-octanol/water partition coefficients, and Henry's law constants. EnvironToxicol Chem 17:251-257.

24. Flach EC. 1992. The influence of four macrozoobenthic species on the abundance of theamphipod Corophium volutator on tidal flats of the Wadden sea. Neth J Sea Res 29:379-394.

25. Van Leeuwen, CJ. 1995. Sediment Toxicity. In CJ Van Leeuwen, JLM Hermens, eds, RiskAssessment of chemicals: an introduction. Kluwer Academie Publishers, Dordrecht, TheNetherlands, pp 204-210.

26. Pesch CE, Muuns W, Gutjahr-Gobell R. 1991. Effects of a contaminated sediment on lifehistory traits and population growth rate of Neanthes arenaceodentata(Polychaeta:Nereidae) in the laboratory. Environ Toxicol Chem 10:805-815.

27. Dillon TM, Moore DW, Gibson AB. 1993. Development of a chronic sublethal bioassayfor evaluating contaminated sediment with the marine polychaete worm Nereis (Neanthes)arenaceodentata. Environ Toxicol C7ieml2:589-605.

28. Meador JP, Casillas E, Sloan CA, Varanasi U. 1995. Comparative bioaccumulation ofpolycylic aromatic hydrocarbons from sediment by two infaunal invertebrates. Mar EcolProgSer 123:107-124.

29. Meador JP, Adams NG, Casillas E, Bolton JL. 1997. Comparative bioaccumulation ofchlorinated hydrocarbons from sediment by two infaunal invertebrates. Arch EnvironmContam Toxicol 33:388-400.

34

30. Meador JP, Stein JE, Reichert WL, Varanasi U. 1995. Bioaccumulation of polycyclicaromatic hydrocarbons by marine organisms. Rev Environ Contam Toxicol 143:79-165.

31. Gerino M. 1990. The effects of bioturbation on partiële redistribution in Mediterraneancostal sediment. Prelimmary results. Hydrobiol 207:251-258.

32. Davey JT. 1994. The architecture of the burrow of Nereis diversicolor and itsquantification in relation to sediment-water exchange. JExp MarBiol Ecol 179:115-129.

33. Fauchald K, Jumars PA. 1979. The diet of worms: a study of polychaete feeding guilds.Oceanogr Mar BiolAnw Revl7:l93-2&4.

34. Smith D, Hughes RG, Cox EJ. 1996, Predation of epipelic diatoms by the amhipodCorophium volutator and the polychaete Nereis diversicolor. Mar Ecol Prog Ser 145:53-61.

35. Gilbert F, Rivet L, Bertrand JC. 1994. The in vitro infïuence of the burrowing polychaeteNereis diversicolor on the fate of petroleum hydrocarbons in marine sediments.Chemosphere 29:1-12.

36. Hughes RG, Gerdol V. 1997. Factors affecting the distribution of the amphipodCorophium volutator in two estuaries in south-east England. Estuarine, Coastal and ShelfScience 44:621-627.

35

Table 1. Percentage recovery of C. volutator (%), concentration of fluoranthene in amphipodsand Total Suspended Solids (TSS) in water column at different densities and at different timesof exposure (Means ± SD)

Time(days)

2

3

6

9

13

Amphipoddensity

100300100300100300100300100300100*

300*

Amphipodrecovery (%)

91 (±4.0)81 (±7.9)78 (±2.9)75 (±1.7)73 (± 1.5)64 (± 0.9)64 (±3.1)51 (± 1.8)54 (±2.9)35 (± 6.7)92 (±4.0)92 (±1.2)

Fluoranthene conc. inC. volutator (|ig/g dwt)

58.5 ( ± 33)135 (±49)90 (± 56)130 (±14)63 (± 5.7)

93.5 (± 4.9)n.d.n.dn.d58+

TSS (mg/L)#

38 (± 2.0)a

100 (± 17)b

29 (± 6.6)a

71 (± 1.4^34 (± 3.6)a

64 (± 8.6)b

26 (± 2.3)a

66 (± 6.4)b

35 (± 6.7)a

39 (± 4.0)a

70 (± 4.6)a

177 (± 7.3)b

* = treatments without worms+ = based on one valuen.d. = not determined# = raeans that do not share the same letter are significantly different from each other at p<0.05

Table 2. Measured fluoranthene concentrations in the water colum (total and dissolved),calculated uptake and elimination rate constants (ki and k2) and (Log basis) bioconcentrationfactors (BCF) in the different amphipod densities. Standard deviations (± SD) and 95%confidence limits are given in parentheses

Parametersa

Cwtot(ug/L)Cwdiss (wg/L)

k2tot ( d " )

k2diss (d"1)

kltot(mL.g-1.d-1)kidiSS(mL.g"1.d'1)

LogBCFto,(mL.g"1

LogBCFdisstraL-g'1t

Estimates at different araphipod0

5.85 (± 0.5)5.85 (± 0.5)

0.60 (0.37-0.82)0.62 (0.38-0.86)

28,900 (18,400-39,400)29,700 (18,700-41,000)

4.684.68

10011 (±1.3)

8.57 (±1.0)0.60 (0.37-0.82)0.62 (0.38-0.86)

23,000(14,800-31,000)27,500(18,100-37,000)

4.594.65

densities300

12.5 (± 2.9)7.49 (±1.7)

0.60 (0.37-0.82)0.62 (0.38-0.86)

25,000(16,000-34,000)45,500 (29,000-62,000)

4.624.87

= parameters are estimated on the basis of dissolved (CWdiss) and total (Cwtot) (particle-bound+ dissolved) fractions of fluoranthene in the water colum; the values here reported are basedon the average of concentrations obtained at each time (i.e. n-2 and n=6, for the treatmentswithout and with amphipods, respectively);b = BCFtot, calculated as the ratio between k]tot and k2tot, obtained on the basis of totalfluoranthenec = BCFtfss, calculated as the ratio between kidiss and k2diSS» obtained on the basis of dissolvedfluoranthene

36

Table 3. Lipid content and BAFi0C - Biota to sediment accumulation factorsnormalized to lipid concentration (|ig/g lipid) and to organic carbon concentration(Hg/| fluoranthene/g OC) in N. virens

Time (days)1

2

5

8

12

in different treatments at different timesNumber of amphipods BAF ioc (SD) *

0100300

0100300

0100300

0100300

0100300

0.14(0.02)a

0.15(0.05)a

0.48 (0.06)b

0.80 (0.08)a

0.91 (0.07)a

1.63 (0.25)b

0.88 (0.23)a

l.ll(0.32)a

1.42(0.23)a

0.93(O.15)a

1.45(0.49)a

1.60(0.36)a

0.97 (0.06)a

U 6 (0.22)a

1.48 (0.09)a>b

* = means thatdo not share the same letter are significantly different from each otherat p<0.05 (Tukey test)

37

LEGENDS

Fig. 1. Uptake kinetics curve of fluoranthene in the worm Nereis virens as a function of time

in the absence of amphipods

Fig. 2. Uptake kinetics curve of fluoranthene in worm Nereis virens as a function of time in

the presence of 100 amphipods

Fig. 3. Uptake kinetics curve of fluoranthene in worm Nereis virens as a function of time in

the presence of 300 amphipods

38

Fig.1

4 6 i

Time (days)

10 12

Fig.2700

4 6 ï

Time (days)

10 12

Fig.3700

Time (days)

10 12

39

Chapter 5

Resuspension of PAH in water column from spiked and unspiked

sediments induced by Corophium volutator bioturbation

40

Resuspension of polycyclic aromatic hydrocarbons in water columnfrom spiked and unspiked sediments induced by bioturbation

by the amphipod Corophium volutator

ABSTRACT

We examined the transport in overlying water of different polycyclic aromatic hydrocarbons(PAH) from a spiked and an unspiked sediment by the burrowing amphipod, Corophiumvolutator, C. volutator was exposed to a lightly contaminated sediment (unspiked) from theWesterschelde (located in the South-western part of The Netherlands), and to a sediment fromthe same origin onto which six different PAHs (phenanthrene, anthracene, fluoranthene,pyrene, benzo(b)fluoranthene and benzo(k) fluoranthene) with a logKow ranging from 4.6 to6.5, were spiked. The difference in PAH loading in the sediments between the two treatmentswas 13 fold on an average. The exposure in both treatments lasted for 25 days and watersamples for Total Suspended Solids (TSS) and PAH analyses and amphipod samples for dryweights determination, were taken after 1, 4, 7, 12, 19 and 25 days. Sediment samples wereanaysed for PAH at the beginning and the end of the exposure. Results for TSSconcentrations, amphipod survival and dry weights, showed that sediment resuspension by C.volutator occurred in a similar way in both treatments and that bioturbation activity was notaltered by the spike of extra PAH compounds onto the sediment. In overïying water the meanratio PAH/POM (normalized to Particulate Organic Matter) of the spiked to the PAH/POM ofunspiked treatment was approximately 25% lower than the sediment ratio PAH/OM(normalized to Organic Matter) of the spiked to the PAH/OM of the unspiked treatment. Thisfinding suggests that the binding of PAH to the bulk sediment particles might be differentfrom that to the particles resuspended in the water column. This was explained by the fact thatonly the smallest partiële size-fractions from the bulk sediment are resuspended in overlyingwater during bioturbation. When PAH/POM values of the overlying water were divided by thePAH/OM values of the sediment, the ratio PAH/POM to PAH/OM showed significantdifferences among PAH compounds. The ratio calculated for the high molecular PAH (i.e.B(b)Fla and B(k)Fla) were signifïcantly lower compared to those of the low molecular PAH.This finding suggests that C. volutator mobilizes a selected fraction only of the sedimentparticles that contains more of the low molecular weight PAHs than the high molecularweight PAHs, which bind probably more tightly to the sediment organic carbon and are lessresuspended and available in water column .

INTRODUCTION

Polycyclic Aromatic Hydrocarbons (PAH) have a strong tendency to bind to organiccarbon particles and to accumulate in sediments which act as a sink for these contaminants(Meador et al., 1995). Due to physical (i.e. current and wind induced) and biological (i.e.bioturbation) events which are known to cause desorption and resuspension of sedimentbound contaminants, sediments may also become sources of contaminants (Krantzberg, 1985;Reynoldson, 1987), The difficulty in measuring in situ the extent and frequency ofremobilization of resuspended particles in the water column, hampers the quantification and

41

the potential risk assessment for benthic and non-benthic organisms (Naf et al., 1996). Alsothe interaction between toxicants and benthic organisms are relatively poorly studied.Ecological studies of marine deposit feeders have suggested that these organisms play animportant role in removal and/or burial of organic matter because of their intimate associationwith sediments and because of their bïoturbation (Reynoldson, 1987; Aller, 1988). Differentfunctions regulated by bioturbation of benthic organisms such as vertical distribution,sediment mixing, ventilation and irrigation activities, deposition of faecal pellets andresuspension can cause changes in the fate and bioavailability of sediment-boundcontaminants within ecosystems (Lee & Swartz, 1980; Diaz & Rosenberg, 1996). Increases ofsediment contaminants to the sediment-water interface through faecal pellets production, forexample, have been reported by Karickhoff & Morris (1985) and Mulsow & Landrum (1995).In addition to these indirect effects, bioturbation may also have direct effects by releasing,sediment contaminants into overlying water causing changes in the partitioning of sediment-bound contaminants in the water column (McElroy et al., 1990) and increasing thebioavailability to aquatic organims (Davis, 1993; Clements et al., 1994).

Bioturbation by the amphipod Corophium volutator is known to affect the totalconcentration of PAH in overlying water and to linearly increase the concentration offluoranthene in suspension-feeding mussels with the number of bioturbating amphipods in thesediment (Ciarelli et al., 1999).

In this study we investigated the resuspension ability, due to bioturbation by C. volutator,of different PAH from a lightly contaminated field sediment (unspiked) and from a similarsediment, spiked with different compounds of PAH. Specific objectives were to compare theeffects of resuspension in both type of treatments, in terms of total supended solids and PAHconcentrations in overlying water. PAH concentrations in overlying water resuspended fromboth type of sediments were related to the bulk concentrations of PAH in their respectivesediments. Also differences in behaviour between the low and high molecular PAH inoverlying water of both treatments were assessed.

MATERIALS & METHODS

Field sampling and animals

Sediment samples were collected from an intertidal mudflat (Kappellebank) located in theWesterschelde in the South-West of The Netherlands (51° 27'N, 3°58'E). Samples of the top 3cm of the sediment were taken and transported to the laboratory in polyethylene buckets andstored at 4°C for approx. 2 weeks before utilisation for the experiment.

Amphipods were collected from a (relatively unpolluted) intertidal mudflat (Oesterput)located in the Oosterschelde in the South-western part of The Netherlands (51°36'N, 3°48'E).Sediment samples of the top 3 cm was wet-sieved through a 500-um mesh sieve to removeindigeneous macroinvertebrates, The amphipods were poured into polyethylene bucketscontaining sea water and transported to the laboratory where transferred to 10-L jarscontaining a 3 cm layer of (Oosterschelde) sediment filled with natural filtered seawater (i.e.sandbed filtered seawater containing particles <10 |im). Organisms were acclimated to thesame salinity (approx. 32 mg/L), temperature (15 °C ) and light conditions (24 h light) as usedin the experiments.

42

Sediment spiking procedure

Sediment subsamples were taken for analyses of water content and organic matter.Sediment drying occurred at 70°C for 24 h to constant weight. Percentage organic matter wasdetermined after combustion at 450 °C for 3 h.

The following six PAH were spiked onto the sediment: Phenanthrene (Phe), Anthracene(Antra), Fluoranthene (Ha), Pyrene (Pyr), BenzoCb)fluoranthene (B(b)Fla) andBenzo(k)fluoranthene (B(k)Fla). PAH were purchased from Aldrich Chemical Company(purity 98%) and dissolved jointly in acetone. A calculated volume (approx. 200 mL) of thestock solution was added dropwise (120 mL/h) to a slurry of 30 kg of wet sediment and water(v/v: 2:1) to achieve the required nominal concentration (i.e. 10 times the backgroundconcentrations of the sediment of concern). The concentrations were several times lower thanthe expected effect concentrations based on no effect body burdens for narcotic chemicals foramphipods descrbed in the literature. The sediment/water slurry was mixed with a stirrer forapprox. 5 h and then kept in the refrigerator at 4°C for 8 days to allow equilibration andpartitioning of PAH into the sediment. Sediment was mixed homogeneously for approx. 2 hbefore the beginning of the experiment and two subsamples were taken for PAH analyses.

Experimental set-up

The experiment consisted of two different treatments: an unspiked sediment and a spikedsediment. 3-L beakerglasses filled with 900 mL of sediment and 2000 mL of overlyingseawater were used. Sediment and water were allowed to equilibrate for 24 h prior theaddition of the organisms. In both treatments 300 amphipods (corresponding approx. with9900 organisms m"2) were placed in each test chamber and two replicates per treatment andper sampling time were used. The density used was in the range of environmentally realisticdensities occurring in the field during winter time (Flach, 1992). The exposure lasted 25 daysin total and, overlying water samples for Total Suspended Solids (TSS), Particular OrganicMatter (POM) and PAH concentrations and, amphipods for dry weight and PAHconcentrations of both treatments were sampled after 1, 3, 7, 12, 19, 25 days. Amphipodswere sieved out of the sediment without the addition of water, rinsed in deionized water andthen immediately frozen at - 20° C prior to analysis, or, dried for dry weight determinations.Results on PAH body burdens are not reported in this paper. Samples of the unspiked andspiked sediments for PAH analyses were taken at the beginning and the end of the exposureonly. The (static) experiment was conducted at 15 °C and the overlying water was notreplaced during the exposure; temperature and dissolved oxygen were checked almost daily.

Analytical methods

Sediment samples of the unspiked and the spiked treatments were taken at the beginningand at the end of the experiment and analysed for PAH. Sediment samples were soxhletextracted with hexane/acetone mixture (90:10) for 6 hours. The extracts were washed withdeionized water to remove hexane, dried with Na2SO4 and then evaporated and concentratedto 10 mL with a Kuderna-Danish apparatus. The concentrated extracts were eluted withpetroleumether through a 3 g deactivated alumina column (previously washed with 15%water). The eluates were further concentratrated to 1 mL under nitrogen and taken up inacetonitirile. Analyses of PAH were performed with a reversed-phase column fitted in an

43

HPLC system (Mistral 9000, 9100, 9012-Varian). The HPLC was equipped with aprogrammable fluorescence detector (Jasco FP-920) and a programmable diode-array detector(Varian 9065) for the separation and detection of 16 PAH according to the list of U.S.-EPA.Quality assurance of the analytical procedure included procedural blanks, repücate analysesand control materials, Recoveries of PAH were > 95% and results were corrected forprocedural blanks. Sediment analyses were carried out by the Istitute of EnvironmentalStudies (IVM), (Vrije Universiteit, Amsterdam -The Netherlands).

Water column samples were taken af ter 1, 4, 7, 12, 19 and 25 days and analysed for TotalSuspended Solids (TSS) and for PAH. In most of the water samples, analyses of PAH wereperformed by using total water samples and by measuring total PAH concentration (i.e. thedissolved + particle-bound fractions). In some water samples taken at t = 7, t = 12 (spikedtreatment only), and t = 25 day (spiked and unspiked treatments), the dissolved and theparticle-bound fractions were separated by centrifugation at 2700 g and the PAHconcentrations were determined in both fractions and expressed as ng/L and |jg/g (wet),respectively. The total aqueous PAH concentration (expressed as ng/L) was calculated as thesum of the PAH measured in the dissolved fraction + the PAH measured in the resuspendedparticles (|ig/g).

The procedure used for the PAH analysis in resuspended particles was analogous to thatused for the sediment (described above). Water samples ( IL) were adjusted to pH = 2 using33% nitric acid prior to extraction. Af ter addition of 50 ml of hexane the sample was shakenfor 1 min in a separation funnel. The water fraction was collected into the sample bottle andthe pH was adjusted to 9 using 5 M sodiumhydroxide. 100 ml of hexane was added and themixture was shaken for 1 min in a separation funnel. The two hexane fractions were combinedand subsequently dried over an anhydrous sodiumsulphate column. The dried hexane extractwas evaporated under nitrogen in a turbovap-apparatus and further reduced to 1 ml afteraddition of 3.5 ml of a 2% ethanediol solution in methanol. The extracts were automaticallyinjected into an HPLC system with an LC-PAH Cis reversed-phase column andconcentrations were quantified by UV absorbance and fluorescence. Water samples analyseswere carried out by OMEGAM, accredited as an analytical Iaboratory by the LaboratoryAccreditation Board of The Netherlands.

Total Suspended Solids (TSS) in overlying water were determined gravimetrically in thelow and high density treatments after filtration of 200 and 100 mL, respectively. Pre-weighedand pre-ashed (500 °C) glass fiber filters (Type GF/C Whatman, 1 um nominal pore size)were dried at 50 °C and weighed after 24 h. POM concentrations were determined after ashingthe filters at 450 CC and weighed after 24 h. Concentration was expressed as mg/L bycalculating the difference between the total and initial weight and dividing by the volume ofwater filtered.

Statistical analyses

Significant differences between means were tested with one-way analysis of variance(ANOVA) followed by pairwise comparisons among treatments or among PAH compounds(Tukey's HSD test). Differences were considered significant at p<0.05. Analyses wereconducted using the statistical program SYSTAT 5.0 (1990).

44

RESULTS

Sediment

Mean percentage dry matter of the sediment used was 36.7 (± 1.3) and mean percentageorganic matter (OM) was 9.5 (± 0.9). Mean measured concentrations of Phe, Antra, Pyr,B(b)Fla and B(k)Fla of the spiked and the unspiked treatments are given in Table 1. Bothsediments are dominated by Pyr and Fla (sediment data are incomplete and therefore notshown) which, among the range of PAH, are present in the highest concentrations. As nosignificant differences were found between the concentrations measured at the beginning andthe end of the experiment, all the data were pooled. The ratio of the concentrations in the twosediments ranges from 11 (B(b)Fla) to 17 (Pyr).

Percentage recovery ofamphipods and dry weights

Mean recovery ranged from 58 (t = 25 day) to 91 % (t = 1 day) and from 66 (t - 25 day) to 89% (t = 1 day) in the unspiked and spiked treatments, respectively. Results of the twotreatments were very similar indicating that the organisms were as tolerant to the unspiked asto the spiked sediment. Mean dry weight values ranged from 1.24 to 1.67 and from 1.19 to1.52 mg/org in the unspiked and the spiked treatments, respectively. The values of bothtreatments varied within a narrow range and were not statistically different from each othersuggesting that the spiking of extra PAH in the sediment probably did not cause any stress tothe amphipods (Table 2).

Overlying water

Total Suspended Solids (TSS) in overlying water ranged from 269 (± 19.8) to 728 (± 71)mg/L and from 190 (±) to 940 (± 47.5) mg/L in the unspiked and spiked sediments,respectively. The values of the two treatments overlapped and were not significantly differentfrom each other. Almost the same trend was seen in both treatments during the the course ofthe experiment (Fig. la & lb). As the percentage recovery of amphipods was comparable, thesmall differences found for t=12 and t=25 days were likely due to a biological factor (i.e.different bioturbation activity) and were probably not related to the different concentrations ofPAHs.

PAH concentrations in overlying water followed approximately a comparable trend in thetwo treatments and, as was seen for the sediments, concentrations of Fla and Pyr were presentin the highest concentrations compared to others species of PAHs.For all PAH spiked, more than 80% of the compound was bound to suspended solids, (datanot shown), and so total concentrations (dissolved + particulate-bound) of the PAH of concernwere normalized to Particular Organic Matter (POM) in both treatments. The differences inloading between the spiked and the unspiked treatments were calculated and compared withthose found for the sediments. The differences in PAH concentrations in overlying waterbetween the spiked and the unspiked treatments ranged between 7.5 (Phen) and 16.5 (Fla)(Table 3). The ratio PAHs of spiked to PAHs of unspiked treatments in overlying water, were15 to 35% smaller than the sediment ratios of PAHs found between the two treatments.

To investigate the difference in binding capacity between the different PAH in overlyingwater, the ratio PAH normalized to POM (i.e. the average of values at each sampling time) toPAH normalized to OM in sediment (i.e. the average of values at t=0) for the spiked and the

45

unspiked treatments, was calculated. In both treatments, this ratio was significantly higher forthe low molecular PAH (Phe, Antra) and Pyrene than the high molecular PAH (B(b)Fla andB(k)Fla) (Fig. 2).

DISCUSSION

Results of this study showed that sediment resuspension due to bioturbation by Corophiumvolutator in the spiked and the unspiked sediments was comparable. Burrowing activity wasnot infiuenced by the extra load of contaminants. This was primarily shown by the increasedwater column turbidity in all test chambers during the course of the experiment. This was adirect result of considerable reworking of the upper 2 cm of the sediment which occurred withsimilar intensity in the spiked and unspiked treatments. Also TSS concentrations in overlyingwater of the two treatments overlapped and similar results on the number of survingamphipods and on dry weight values were obtained for the two treatments.

The difference in PAH concentrations in overlying water between the spiked and theunspiked treatment reflected the difference in PAH loading between spiked and unspikedsediment. The ratio of aqueous PAHs/POM of spiked to the PAHs/POM of the unspikedtreatment was approx. 25% lower on an average, compared to the sediment ratio PAHs/OM ofspiked to the PAHs/OM of the unspiked treatments. This difference was probably due to adifferent binding of PAH to the organic fraction of suspended solids in overlying watercompared to that of the organic fraction in the sediment. Studies on bioavailability andaccumulation of PAH have demonstrated that some PAH can distribute differently among theparticle-size fractions of the sediment and can bind more efficiently to the smaller fractions(i.e. < 10 urn) of the sediment than to particles larger than 30 um (Harkey et al., 1994;Landrum et al., 1994). Results of Landrum et al. (1994) on pyrene toxicity to the amphipod,Diporeia spp., showed also that the binding of this compound was dependent on the differentorganic carbon content of the sediment fractions and that the particle-size distribution variedwith dose (i.e. at higher dose pyrene concentration on the finest partices was 10 times higherthan in the bulk sediment). In this study, although the sediment partiële size fractions of thesediment were not measured, it is very likely that Corophium volutator resuspendedpreferentially the smallest fractions of the sediment in the water column and that the PAHconcentrations associated with these particles were different than those measured in the bulksediment. Additionally, if the binding to the smallest size fractions is also dose dependent, it isreasonable to think that the ratio aqueous PAH/POM of spiked to unspiked treatments doesnot necessarily have to be similar to the ratio PAH/OM of spiked to unspiked treatmentsfound for the sediment.

When (total) PAH concentrations in the water column of the two treatments werenormalized to Particulate Organic Matter (POM) and divided by the OM-normalized PAHconcentrations in sediment, a difference in behaviour among the low and high molecular PAHwas revealed. The ratio PAH/POM of the overlying water to PAH/OM of the sediment wassignificantly lower for the high molecular PAHs (B(b)Fla and B(k)Fla) compared to the lowmolecular PAHs (Phe and Antra) and to pyrene, in both spiked and unspiked treatments. Thehigh molecular PAH, pyrene, behaved thus more like a low molecular PAH. These datasuggest that as hydrophobicity of PAH increases, the compounds will be more tightly boundto the sediment, will be less resuspended in the water colum and so their concentration in thedissolved phase will be low. This was also confirmed by the ratio bound to dissolved fractionsmeasured for some water samples where the dissolved and particle-bound fractions were

46

separated by centrifugation and analysed individually for PAH. In these samples results,showed that B(b)Fla and B(k)Fla were particle-bound for more than 90 % and that Phe, Antra,Pyr and Fla were particle-bound for about 80%. (As these results were based on few replicatesthe data were not shown). This finding on the different binding capacity of PAH and thus thedifferent bioavailablity in water is also supported by the study of Meador et al. (1995) whofound that uptake of low molecular PAH and of Fla and Pyr in the polychaete, Armandabrevis (feeding on sediment) and in the amphipod, Rhepoxynius abronius (feeding on detritusand diatoms) was similar, since the interstitial water was the main route of uptake for bothorganisms. But accumulation of high molecular PAH in the polychaete was significantlyhigher than in the amphipod due to the different feeding habit. The authors also concludedthat the most hydrophobic PAH were more tightly sorbed to the organic carbon of thesediment than the less hydrophobic PAH and were therefore less available for uptake throughpore water for the amphipod R. abronius. Also in this study, the high molecular PAH, pyreneand fluoranthene, behaved more like low molecular PAH, in that the organisms accumulatedsimilar Pyr and Fla tissue concentrations because the interstitial water was probably the mainroute of uptake for both species.

This study was designed to investigate the abiüty of C. volutator bioturbation to resuspendsediment-bound PAH as a function of two different treatments (spiked and unspikedsediments) and of two different degrees of PAH hydrophobicity. The ability of Corophiumvolutator bioturbation to resuspend sediment-bound contaminants to the overlying water in adensity dependent way and the consequences of this proces in the bioavailability and transferof compounds to aquatic organisms has already been assessed in a previous study (Ciarelli etal,, 1999). The ability to resuspend and to remobilize sediment-bound contaminants to theoverlying water by deposit-feeders and the consequences of this proces in the increase of thebioavailability has also been reported by several other authors (McElroy et al., 1990, Clementset al., 1994; Riedel et al., 1994).

Another intersting aspect which should receive more attention in the future whenimplementing remediation procedures for lightly contaminated sediments, is the greateravailability in water column of sediment-assodated PAH to bacteria for biodegradationprocesses, induced by bioturbation of deposit feeders. Finally, the results on the effects of C.volutator bioturbation can be used to generate rate constants needed to effectively model thefate of PAH in water column and transfer to other organisms of the aquatic food chain inmarine sediment water systems.

REFERENCES

Aller, RC (1988): Benthic fauna and biogeochemical processes in marine sediments: the roleof burrow structures. In: Nitrogen Cycling in coastal marine environments. (Eds:Blackburn,TH; Sorensen, J) WileyJ & Sons Ltd„ 302-338.

Belfroid AC, Hopman G, Kaim P(1998): Desorptie van organische microverontreinigingenvanuit mariene sedimenten. Onderzoek in het kader van aging/bioturbatieonderzoek. IVMrapport R-98/09

Ciarelli, S; Van Straalen, NM; Klap, VA; Van Wezel, AP (1999): Effects of sedimentbioturbation by the estuarine amphipod Corophium volutator on fluoranthene resuspensionand transfer into the mussel Mytilus edulis. Environ.ToxicoI.Chem. 18, 318-328.

47

Clements.WH; Oris, JT; Wissing, TE (1994): Accumulation and food chain transfer offluoranthene and benzo(a)pyrene in Chironomus riparius and Lepomis macrochirus. Arch.Environm. Contam. Toxicol. 26,261-266.

Davis, WR (1993): The role of bioturbation in sediment resuspension and its interaction withphysical shearing. J.Exp.Mar.Biol.Ecol. 171, 187-200.

De Maagd, PJ; Ten Hulscher, DThEM; Van den Heuvel, H; Opperhuizen, A; Sijm, DTHM(1998): Physichochemical properties of polycyclic aromatic hydrocarbonsraqueoussolubilities, n-octanol/water partition coefficients, and Henry's law constants.Environ.Toxicol.Chem. 17,251-257.

Diaz, RJ; Rosenberg, R (1996): The influence of sediment quality on functional aspects ofmarine benthic communities. In: Development and Progress in Sediment Quality Assessment:Rationale, Challenges, Techniques & Strategeies. (Eds: Munawar.M; Dave.G) AcademiePublishing, Amsterdam (The Netherlands), 57-68.

Flach, EC (1992): The influence of four macrozoobenthic species on the abundance of theamphipod Corophium volutator on tidal flats of the Wadden sea. Neth.J.Sea Res. 29(4), 379-394.

Green, AS; Chandler, GT (1994): Meiofaunal bioturbation effects on the partitioning ofsediment-associated cadmium. J.Exp.Mar.Biol.Ecol. 180,59-70.

Harkey, GA; Lydy, MJ; Kukkonen, J; Landrum, PF (199): Feeding selectivity andassimilation of PAH and PCB in Diporeia spp. Environ.Toxicol.Chem. 13(9), 1445-1455.

Karickhoff, SW; Morris, KR (1985): Impact of tubificid oligochaetes on poUutant transport inbottom sediments. Environ.Sci.Technol. 19, 51-56.

Krantzberg, G (1985): The influence of bioturbation on physical, chemical and biologicalparameters in aquatic environments: a review. Environ.Pollut. 39,99-122.

Landrum, PF; Dupuis, WS; Kukkonen, J (1994): Toxicokinetics and toxicity of sediment-associated pyrene and phenanthrene in Diporeia spp.: examination of equilibrium-partitioningtheory and residue-based effects for assessing hazard. Environ.Toxicol.Chem, 13, 1769-1780.

Lee, H,ü; Swartz, RC (1980): Biological processes affecting the distribution of pollutants inmarine sediments. Part II. Biodeposition and Bioturbation. In: Contaminants and Sediments.Analysis, Chemistry, Biology. Vol. 2. (Ed: Baker,RA) Ann Arbor Science Publishers inc. AnnArbor, Ann Arbor, MI (USA), 555-606.

McElroy, AE; Farrington, JW; Teal, JM (1990): Influence of mode of exposure and thepresence of a tubiculous polychaete on the fate of benzo(a)anthracene in the benthos.Environ.Sci.Technol. 24, 1648-1655.

48

Meador, JfP; Casillas, E; Sloan, CA; Varanasi.U (1995): Comparative bioaccumulation ofpolycylic aromatic hydrocarbons from sediment by two infaunal invertebrates. Mar. Ecol.Prog. Ser. 123, 107-124.

Meador, JP; Stein, JE; Reichert,WL; Varanasi.U (1995): Bioaccumulation of polycyclicaromatic hydrocarbons by marine organisms. Rev. Environ. Contam. Toxicol. 143,79-165.

Mulsow, SG; Landrum, PF (1995): Bioaccumulation of DDT in a marine polychaete, theconveyor-belt deposit feeder Heteromastusfiliformis (Claparede). Chem.Ecol. 31, 3141-3152.

Naf, C; Axelman, J; Broman, D (1996): Organic contaminants in sediments of the BalticSea:distribution, behaviour and fate. In: Development and progress in sediment qualityassessment: rationale, challenges, techniqes & strategies. (Eds: Munawa,M; Dave,G) SPBAcademie Publishing, Amsterdam, 15-25.

Reynoldson, TB (1987): Interactions between sediment contaminants and benthic organisms.Hydrobiol. 149,53-66.

Riedel, GF; Sanders, JG; Osman.W (1989): The role of three species of benthic invertebratesin the transport of arsenic from contaminated estuarine sediment. J.Exp.Mar.Biol.Ecol. 134,143-155.

49

Table 1. Log KoW of the PAH spiked onto the sediment and means (± SD) concentrations ofPAH normalized to sediment organic matter (OM) in spiked and unspiked treatments and theratio PAH spiked to unspiked sediment.

PAH

PheAntraFlaPyr

B(b)FlaB(k)FIa

LogKow

4.57 $

4.68 $

5.23 s

5.07#

6.52#

6.11S

PAH concentrations (ng/g OM dry)Spiked sedimenta

82 (12)33(5)

n.d215 (34)108(16)62(9)

Unspiked sedimentb

7 (0.5)2.5 (0.2)

n.d12 (0.6)10 (0.6)5 (0.3)

Ratio

11.513n.d1711

12.5* = De Maagd et al , 1998 ; Means (and 95% confidence limits)# = Meadoretal., 1995a= number of replicate samples is 5b = number of replicate samples is 2n.d - not detemined

Table 2. Mean (± SD) percentage recovery and mean dry weigh (± SD) of C volutator in thespiked and unspiked treatments. Dry weight values are expressed in mg/organism.

Days ofexposure

147121925

Percentage

Unspiked sediment91 (7.5)84 (4.7)86(1.4)76(15)75(1.4)

58+

recovery

Spiked sediment89(10.6)83 (5.6)84(1.6)86 (1.6)76 (6.8)66 (8.2)

Dry weight2

Unspiked sedimentn.d.

1.25(0.3)1.24(0.32)1.67(0.10)1.47(0.15)1.59(0.42)

Süiked sedimentn.d.

1.42(0.1)1.19(0.07)1.52(0.44)

n.d.1.22(0.16)

= number of replicates is 22 - number of replicates is 4 (2 for each test chamber)n.d. = not determined+ = based on one replicate

50

Table 3. Mean (± SD) concentrations of PAH in overlying water normalized to Parti-culate Organic Matter (POM) in spiked and unspiked treatments. The ratio between PAHïn spiked and unspiked treatment is also give.

PAHPhe

AntraFlaPyr

B(b)FlaB(k)Fla

PAH Concentrations (ng/g POM)Spiked treatmenta Unsmked treatmentb

5000 (2370)1815(800)

11120(4610)14440 (3330)3620(1830)1530 (395)

665 (156)184 (41)

869 (214)838(189)389(138)177 (66)

Ratio

7.59

16.513.59.58.5

a = is the average of every single PAH measured at each sampling time in spikedsediment; the number of replicates is 6 to 8;

b = is the average of every single PAH measured at each sampling time in unspiked sediment;the number of replicates is 9.

51

LEGENDS

Fig la/lb. Concentrations of Phenanthrene (Phen), Anthracene (Antra), Fluoranthene (Fla),Pyrene (Pyr), Benzo(b)fluoranthene (B(b)Fla) and Benzo(k)fluoranthene (B(k)Fla) at differenttimes (Mean ±SD) on the left axis and, Total Suspended Solids (TSS) on the right axis (Mean±SD) in the unspiked treatment (la) and the spiked treatment (lb). Mean total concentrationsof PAH (bars) are mostly based on two replicates. Bars without SD are based on single values.Mean concentrations of TSS (line) are based on four replicates (two per testchamber).

Fig 2. Ratio of aqueous PAH (Phen, Antra, Pyr, B(b)Fla and B(k)Fla), noraialized toParticulate Organic Matter (POM), to sediment PAH, normalized to OM (Organic Matter) ofthe spiked and unspiked treatments (Mean ± SD). Mean values of both treatments are basedon the sum of PAH concentrations measured at different sampling times. For the spikedtreatment, means are based on 5 (B(k)Fla), 6 (B(b)F) and 8 (Phen, Antra and Pyr) values. Forthe unspiked treatment, all PAH means are based on 9 values. Asteriks indicate ratios of PAHwhich are significantly different from other PAH ratios at p<0.005.

52

Fig. l a

400

PAH and Total Suspended Solids (TSS) concentrations inoverlying water in the unspiked treatments (Mean ± SD)

7 12

Time (days)

19

iPhen iFla ÏPyr lB(b)Ha !B(k)Fla -TSS

Fig. lb

25

B(k)Ha -TSS

53

Fig.2

•ê Dunspiked

• spiked

Phen Antra Pyr B(b)F B(k)F)

54

Chapter 6

SUMMARY

55

SUMMARY

Theproblem and objectives of the study

In 2002 sediment quality evaluations in the Netheriands will be established either on thebasis of chemical analyses or on the basis of Standard toxicity tests wich are carried out with anumber of selected species of benthic organisms. The responses of different test organims tosediment pollutants will contribute to a sediment quality ranking system which range from"no toxic" to "very toxic". The use of biological responses is a considerable step forward andis a useful tooi for deriving sediment toxicity evaluations as it will give additional informationon the bioavailability of sediment associated compounds. This approach, however, does nottake into account some important biological processes that may influence the bioavailabilityand the toxicity of chemicals, such as the mode of exposure of the benthic test organisms ofconcern and their interactions with the sediment and the overlying water. Also factors such asfeeding strategy and burrowing activities, known as bioturbation, animal size and density andthe role that these play in the bioavailability of compounds are poorly understood. Theequilibrium partitioning model (EqP) is one of the most accepted model used to assessbioavailability and ecotoxicological risks of sediment contaminants. Moreover, the use ofcurrent models to predict the fate and transport of organic pollutants in aquatic environmentsincorporate mainly chemical and physical processes such as adsorption and desorptionprocesses, sedimentation, resuspension and degradation processes. The role that benthicorganisms play in these processes are not well understood and poorly represented in suchmodels.

To have a better insight in the role of biological processes on the bioavailability ofsediment-associated contaminants and to better understand the effects of bioturbation,sediment-water partitioning coefficients were calculated and the accumulation of fluoranthenein amphipods of different densities were measured. This was done by using the estuarineamphipod, Corophium volutator, which is an important organism in the benthic food chainand a widely accepted test organism in European standardized marine sediment toxicity tests,Also the role of amphipod bioturbation in the trophic transfer of fluoranthene into mussels andin the increased availability of fluoranthene to the polychaete worm, Nereis virens wereinvestigated. Finally, effects of bioturbation by C.volutator in a field-contaminated sedimenton the behaviour of different PAH in the overlying water were compared to those of a spikedsediment.

Effects of sediment bioturbation by the estuarine amphipod C. volutator on fluorantheneresuspension and transfer into the musset, Mytilus edulis

The results of this study showed that bioturbation by C. volutator significantly affected totalsuspended solids concentration in the overlying water and consequently the total aqueousconcentration of fluoranthene. This effect was greater at higher animal density and longerexposure time.

Sediment-water partitioning coefficients calculated in this study after 10-d were in goodagreement with Log/sfOw values reported in the literature. Bioturbation did not affect thepartitioning of fluoranthene over suspended solids and water, nor did it affect theconcentrations in sediment and pore water. Biotuibation will probably not contributesignificantly to the discrepancies between predicted and measured chemical concentrationswhen pore water concentrations are predicted from the EqP.

Through C. volutator bioturbation activity, fluoranthene body burdens in the musselsincreased linearly with amphipod density and over time of exposure. Sediment bioturbation

56

had greater consequences for the water column and uptake by filter-feeders than for thesediment, pore water and the amphipods themselves. This observation demonstrated theimportance of bioturbation in sediment-bound containinants resuspension and trophic transferto organisms of the aquatic food chain. Due to their limitated capacity of elimination andmetabolization of PAHs, their sessile behaviour and their ability to filter large amounts ofwater, mussels may be among the organisms affected most by resuspension of sediment-bound contaminants. The increased accumulation in mussels demonstrated the importance ofbioturbation as a flux phenomenon and its role in the transport of resupended sediment-boundcontaminants to aquatic organims.

The influence bioturbation by the estuarine amphipod Corophium valutator on fluorantheneuptake in the marine polychaete, Nereis virens

The results of this study showed that bioturbation by C. volutator increased total suspendedsolids and bioavailability of aqueous fluoranthene and consequently accumulation by theworm, N. virens. Biota to sediment accumulation factors normalized to lipid concentratïonand to sediment organic carbon (BAFi0C) of worms exposed with the highest number ofamphipods were significantly higher (two to three times) compared to worms exposed withfewer or without amphipods afterl and 2 days of exposure. After 12 days, due to amphipodmortality, however, accumulation in worms was only 35% higher compared to the treatmentwithout amphipods. The rate constants of fluoranthene in the worms and bioconcentrationfactors (BCF) were, however, not significantly affected by bioturbating amphipods. This studysuggested that the enhanced uptake of fluoranthene by worms is due to amphipod-mediatedincrease of the dissolved concentration.

Due to the negative impact of N. virens on C. volutator, which caused stress and rapidmortality of the amphipods, the effects of bioturbating amphipods on fluorantheneaccumulation in worms may likely be underestimated.

Resuspension ofPAH in water column front spiked and unspiked sedimentsinduced by Corophium volutator bioturbation

The transport in overlying water of PAH of different hydrophobicity (phenanthrene,anthracene, fluoranthene, pyrene, benzo(b)fluoranthene and benzo(k) fluoranthene), from aspiked and an unspiked sediment by C. volutator bioturbation was examined. Results showedthat sediment resuspension by C. volutator occurred in a similar way in both treatments andthat bioturbation activity was not altered by the spike of extra PAH compounds onto thesediment. The mean ratio of PAH concentrations (normalized to POM-Particulate OrganicMatter) of the spiked to the unspiked treatment in overlying water was approximately 25%lower than the mean ratio of sediment PAH (normalized to OM-organic matter) of the spikedto the unspiked treatment. This suggests that the binding of PAH to the bulk sedimentparticles might be different from that to particles resuspended in the water column due to thedifferent resuspended partiële size-fractions from the bulk sediment. The ratio of PAH/POMin overlying water to PAH/OM in sediment showed that resuspension in the water column ofthe high molecular PAH (i.e. B(b)Fla and B(k)Fla) was significantly lower compared to therest of PAH. This trend was seen in both the spiked and the unspiked treatments. This wasprobably due to the fact that the more hydrophobic compounds, bind more tightly to thesediment organic carbon, are less soluble and less resuspended by in the water column. Thisobservation confirms the general idea that high molecular PAH are less availably in watercolumn than low molecular PAH.

57

General conclusions

The following conclusions can be drawn on the basis of the results described in this report:

• bioturbation by Corophium volutator causes resuspension of total suspended solids andconcomitant transport of sediment-bound contaminants in overlying water as a fluxphenomenon

• the increase of fluoranthene concentration in the overlying water and in the uptake inmussels is linearly related to the number of amphipods present in the sediment and thetime of exposure

• bioturbation of Corophium volutator increases bioavailabilty and uptake of fluorantheneto the polychaete worm, Nereis virens when amphipods and worms are exposed together.This process gains importance with increasing density of amphipods. The uptake andelimination rate constants and BCF values are, however, not affected by bioturbation

• bioturbation of Corophium volutator does not affect sediment-water partitioningcoefficients and does not increase the uptake of contaminants to themselves in higherdensity treatments

• resuspension of total suspended solids and PAH concentrations in overlying water of aspiked sediments is comparable to that of non-spiked sediments. In both sedimentsresuspension of the less hydrophobic PAH is greater than the most hydrophobic PAHcompounds

58