Estuarine sediment acute toxicity testing with the European amphipod Corophium multisetosum Stock, 1952 Ana Ré, Rosa Freitas, Leandro Sampaio, Ana Maria Rodrigues, Victor Quintino * CESAM (Centro de Estudos do Ambiente e Mar), Departmento de Biologia, Universidade de Aveiro, 3810 – 193 Aveiro, Portugal article info Article history: Received 27 January 2009 Received in revised form 4 June 2009 Accepted 15 June 2009 Keywords: Sediment toxicity Acute bioassay Amphipod Corophium multisetosum abstract This study assessed the use of the European amphipod Corophium multisetosum Stock [Stock, J.H., 1952. Some notes on the taxonomy, the distribution and the ecology of four species of the genus Corophium (Crustacea, Malacostraca). Beaufortia 21, 1–10] in estuarine sediment acute toxicity testing. The sensitiv- ity of adults to the reference toxicant CdCl 2 was determined in water-only 96 h exposures in salinity 2. LC 50 values ranged from 0.33 mgCd 2+ L 1 at 22 °C to 0.57 mgCd 2+ L 1 at 15 °C. Adult survival was studied in control sediment with water salinity from 0 to 36 and with fine particles content (<63 lm) from 2% to 97% of total sediment, dry weight. Experiments were conducted at 15, 18 and 22 °C and the results indi- cate that the species can be used under the full salinity range although higher mortality was observed at the lower salinity in the higher water temperature, and at the higher salinity in the lower water temper- ature. The species also tolerated the studied range of sediment fines content and showed the highest sen- sitivity at intermediate values of fines, especially at the higher temperature, thus advising that tests which have to accommodate sediments with a wide range in fines content should preferably be con- ducted at 15 °C rather than at 22 °C. The response in natural sediments was studied in samples collected yearly from 1997 to 2006, at a site located off the Tagus Estuary, western Portugal. A major flood event in winter 2000–2001 induced detectable alterations in sediment baseline descriptors (grain-size, redox potential and total volatile solids), organic contaminants (PAHs, PCBs, DDT metabolites and c-HCH) and the macrofauna benthic community. Mortality of the amphipod diminished significantly from the before to the after flood period, in close agreement with diminishing sediment contamination and increasing benthic fauna diversity, in the same time period. C. multisetosum is suitable to conduct acute sediment toxicity tests and presents good potential for the development of a full life-cycle sediment test, due to its amenability to laboratory culture and high survival in the control sediment. Ó 2009 Elsevier Ltd. All rights reserved. 1. Introduction Sediment bioassays are a powerful tool for the study of sedi- ment toxicity and are recommended along with others methodol- ogies to classify and prioritise areas of contaminated sediments (Chapman and Long, 1983; Luoma and Carter, 1993; Riba et al., 2004; Kirkpatrick et al., 2006; Allen et al., 2007; Scarlett et al., 2007; Morales-Caselles et al., 2008). Amphipods are an abundant component of the marine and estuarine soft bottom and are amongst the principal prey of many fish, birds and larger inverte- brate species. Many species are detritus feeders and ingest sedi- ment, which may directly expose them to sediment bound contaminants. Amphipods have been shown to be sensitive to con- taminated sediments and to disappear from benthic communities impacted by pollution and reappear under recovery conditions (Swartz et al., 1986). The negative effects of toxicants on amphipod populations may alter the structure and functioning of ecosystems by affecting the availability of food for higher trophic levels. This led to the development of benthic quality indices based upon the relationship between amphipods and other fauna, namely oppor- tunistic polychaetes (Dauvin and Ruellet, 2007), and also to a worldwide development of sediment toxicity protocols using amphipods as test-organisms (Lamberson and Swartz, 1988, 1989; DeWitt et al., 1988, 1992; Lamberson et al., 1992; Bat and Raffaelli, 1998; Bat et al., 1998; Woodworth et al., 1999; Bat, 2005; ISO, 2005; Prato et al., 2006; McCready et al., 2006; Allen et al., 2007; ASTM, 2008; Picone et al., 2008; Prato et al., 2008). Amphipods have been successfully used in sediment toxicity testing namely due to their sensitivity to a wide variety of contam- inants, easy collection and handling in the laboratory, but also be- cause their protection ensures the protection of the whole benthic community. Amphipod toxicity tests have been successfully used in the United States and Canada to assess coastal sediment toxicity 0045-6535/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2009.06.041 * Corresponding author. Address: CESAM (Centro de Estudos do Ambiente e do Mar), Departamento de Biologia, Universidade de Aveiro, Campus Universitario de Santiago, 3810-193 Aveiro, Portugal. Tel.: +351 234370769; fax: +351 234372587. E-mail address: [email protected](V. Quintino). Chemosphere 76 (2009) 1323–1333 Contents lists available at ScienceDirect Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
0045-6535/$ - see front matter � 2009 Elsevier Ltd. Adoi:10.1016/j.chemosphere.2009.06.041
* Corresponding author. Address: CESAM (Centro dMar), Departamento de Biologia, Universidade de AveSantiago, 3810-193 Aveiro, Portugal. Tel.: +351 23437
This study assessed the use of the European amphipod Corophium multisetosum Stock [Stock, J.H., 1952.Some notes on the taxonomy, the distribution and the ecology of four species of the genus Corophium(Crustacea, Malacostraca). Beaufortia 21, 1–10] in estuarine sediment acute toxicity testing. The sensitiv-ity of adults to the reference toxicant CdCl2 was determined in water-only 96 h exposures in salinity 2.LC50 values ranged from 0.33 mgCd2+ L�1 at 22 �C to 0.57 mgCd2+ L�1 at 15 �C. Adult survival was studiedin control sediment with water salinity from 0 to 36 and with fine particles content (<63 lm) from 2% to97% of total sediment, dry weight. Experiments were conducted at 15, 18 and 22 �C and the results indi-cate that the species can be used under the full salinity range although higher mortality was observed atthe lower salinity in the higher water temperature, and at the higher salinity in the lower water temper-ature. The species also tolerated the studied range of sediment fines content and showed the highest sen-sitivity at intermediate values of fines, especially at the higher temperature, thus advising that testswhich have to accommodate sediments with a wide range in fines content should preferably be con-ducted at 15 �C rather than at 22 �C. The response in natural sediments was studied in samples collectedyearly from 1997 to 2006, at a site located off the Tagus Estuary, western Portugal. A major flood event inwinter 2000–2001 induced detectable alterations in sediment baseline descriptors (grain-size, redoxpotential and total volatile solids), organic contaminants (PAHs, PCBs, DDT metabolites and c-HCH)and the macrofauna benthic community. Mortality of the amphipod diminished significantly from thebefore to the after flood period, in close agreement with diminishing sediment contamination andincreasing benthic fauna diversity, in the same time period. C. multisetosum is suitable to conduct acutesediment toxicity tests and presents good potential for the development of a full life-cycle sediment test,due to its amenability to laboratory culture and high survival in the control sediment.
� 2009 Elsevier Ltd. All rights reserved.
1. Introduction
Sediment bioassays are a powerful tool for the study of sedi-ment toxicity and are recommended along with others methodol-ogies to classify and prioritise areas of contaminated sediments(Chapman and Long, 1983; Luoma and Carter, 1993; Riba et al.,2004; Kirkpatrick et al., 2006; Allen et al., 2007; Scarlett et al.,2007; Morales-Caselles et al., 2008). Amphipods are an abundantcomponent of the marine and estuarine soft bottom and areamongst the principal prey of many fish, birds and larger inverte-brate species. Many species are detritus feeders and ingest sedi-ment, which may directly expose them to sediment boundcontaminants. Amphipods have been shown to be sensitive to con-taminated sediments and to disappear from benthic communities
ll rights reserved.
e Estudos do Ambiente e doiro, Campus Universitario de0769; fax: +351 234372587.).
impacted by pollution and reappear under recovery conditions(Swartz et al., 1986). The negative effects of toxicants on amphipodpopulations may alter the structure and functioning of ecosystemsby affecting the availability of food for higher trophic levels. Thisled to the development of benthic quality indices based upon therelationship between amphipods and other fauna, namely oppor-tunistic polychaetes (Dauvin and Ruellet, 2007), and also to aworldwide development of sediment toxicity protocols usingamphipods as test-organisms (Lamberson and Swartz, 1988,1989; DeWitt et al., 1988, 1992; Lamberson et al., 1992; Bat andRaffaelli, 1998; Bat et al., 1998; Woodworth et al., 1999; Bat,2005; ISO, 2005; Prato et al., 2006; McCready et al., 2006; Allenet al., 2007; ASTM, 2008; Picone et al., 2008; Prato et al., 2008).
Amphipods have been successfully used in sediment toxicitytesting namely due to their sensitivity to a wide variety of contam-inants, easy collection and handling in the laboratory, but also be-cause their protection ensures the protection of the whole benthiccommunity. Amphipod toxicity tests have been successfully usedin the United States and Canada to assess coastal sediment toxicity
1324 A. Ré et al. / Chemosphere 76 (2009) 1323–1333
since the mid eighties (Chapman and Long, 1983; Swartz et al.,1985; DeWitt et al., 1992; USEPA, 1994; ASTM, 2008), and were la-ter developed with European species (Ciarelli et al., 1997; Bat et al.,1998; Bat and Raffaelli, 1998; Costa et al., 1998; 2005; Brown et al.,1999; Casado-Martinez et al., 2007; Guerra et al., 2007; van denHeuvel-Greve et al., 2007; Prato et al., 2008). Also, most sedimentassessment methods developed to date are primarily devoted tofresh or marine waters, whereas there is a relative paucity of appli-cations in the estuarine environment (Chapman and Wang, 2001).Corophium multisetosum has been previously used as a test organ-ism in sediment acute and chronic toxicity tests (Castro et al.,2006) as well as in whole effluent testing (Ré et al., 2007). Resultsindicated that this could be a suitable test species for SouthernEuropean transitional waters, where it is more abundant than theNorthern counterpart Corophium volutator, also reported as a suit-able test species (Ciarelli, 1994; Ciarelli et al., 1997; Peters and Ahlf,2005; van den Heuvel-Greve et al., 2007). With a growing numberof species reported as suitable for sediment toxicity testing, it isimperative to assure that the results obtained at the measured end-point reflect a true negative effect due to contamination and not thesensitivity of the test species to natural factors, namely tempera-ture, salinity, sediment grain size or organic matter, factors whichmay confound the toxicity evaluation of sediment samples is sometest protocols (Benton et al., 1995; Quintino et al., 1995).
This work presents a study of baseline conditions for the devel-opment of an acute sediment toxicity test with the Southern Euro-pean estuarine amphipod C. multisetosum. It includes the setting upof laboratory cultures and the experimental study of adult amphi-pods response when exposed to natural factors, namely a range oftemperature, of interstitial and overlying water salinity and of sed-iment grain size, as well as to the reference toxicant cadmiumchloride. Using the proposed test procedure, the species was alsoexposed to sediments collected yearly from 1997 to 2006, at a sitelocated off the Tagus estuary, Western coast of Portugal. Withinthis period, a major flood event occurred in the winter 2000–2001, the effects of which were analysed in the benthic macrofa-una community, on sediment baseline descriptors and organiccompounds as well as on the survival of C. multisetosum in acutesediment toxicity exposures.
2. Materials and methods
2.1. Species characteristics and sampling site
Species of the genus Corophium Latreille are common in estuar-ies worldwide. The European species C. multisetosum has beenfound in the upper, less saline, parts of estuaries (Stock, 1952)and in Portugal is known from the Sado and Mondego estuaries(Marques and Bellan-Santini, 1985), Ria Formosa (Marques andBellan-Santini, 1990) and Ria de Aveiro (Queiroga, 1992). In Riade Aveiro, C. multisetosum is distributed over most of the lengthof the Mira Channel, in a soft-bottom and shallow water habitat,with densities up to hundreds of individuals per square meter.The species breeds throughout the year, but in May, July and Au-gust only a few incubating females are present in the population.An intense recruitment peak occurs in autumn and a smaller peakin spring (Cunha et al., 2000).
In Ria de Aveiro, C. multisetosum is mainly found on clean med-ium sand and tends to avoid sediments with high levels of grainsize below 125 lm and rich in organic matter (Queiroga, 1992).In order to set up laboratory cultures, the species was sampled ina meso/oligohaline site in Ria de Aveiro where salinity remains be-low 5 during most of the year, the water temperature varied from20–24 �C in summer to 7–15 �C in winter/spring, and the speciescan be extremely abundant (>80 000 ind m�2) (Queiroga, 1992).
2.2. Laboratory cultures
Specimens of C. multisetosum were collected in an intertidalsand bank during low tide, separated from most of the sedimentby gentle sieving in the field through 0.5 mm mesh screen andwere transported to the laboratory in water collected in the sam-pling site. In the laboratory, they were kept in cultures in climaticchambers at 15, 18 and 22 �C. The cultures were set up in plasticcontainers (0.38 m length � 0.24 m width � 0.12 m height), hold-ing about 6 L of filtered and UV sterilised water (6–7 cm) and anapproximately 2 cm thick sediment layer from the field samplinglocation. The sediment layer needs to be adequate because theorganisms check its depth by probing the substrate with their elon-gated second antennae. If the conditions are suitable they burrowbut if not they move off to investigate other sites. Each culture con-tainer holds several hundred mature animals. The cultures weremaintained with natural seawater diluted to salinity 18 in deion-ised water. The culture water was filtered, UV sterilized and con-stantly aerated. In the field, the water salinity in the amphipodsampling area ranges from 0, in winter and low tide to 22 in sum-mer and high tide.
In the present work salinity was measured with a handheldrefractometer and expressed using the Practical Salinity Scale thatdefines salinity as a pure ratio, with no dimensions. By decision ofthe Joint Panel of Oceanographic Tables and Standards, salinityshould be reported as a number with no symbol, or indicator ofproportion after it, such as ppt or ‰, and it is not correct to addthe letters PSU, implying Practical Salinity Units, after the number.
The sediment used to set up the cultures was collected at theamphipods collecting site and was sieved through a 0.5 mm meshscreen to remove larger animals and debris. Unused sieved sedi-ment was stored in the dark at 4 �C for later use. The culture con-tainers were kept under constant aeration, in walk-in climaticchambers, with a photoperiod of 14 h light: 10 h dark. About 30%of the culture water was renewed three times a week. At this timewater added to the culture was enriched with algal food (106 cellsper mL – Coast Seafood Algae, Diet C). Feeding was also combinedwith water changes, consisting of dry food components (i.e. 48%Tetramin�, 24% dried alfalfa, 24% dried wheat leaves and 4% Neo-Novum� (DeWitt et al., 1992)), combined and ground to a finepowder and sprinkled on the water surface.
2.3. Test design and procedure: negative control and response to areference toxicant
The acute sediment toxicity test used adult specimens only, iso-lated from the cultures. The static exposures were conducted in1000 mL beakers containing 200 mL of sediment and 800 mL ofoverlying water. The exposure beakers, each containing 20 adultamphipods, were placed in walk-in chambers with controlled tem-perature and photoperiod, 14 h light: 10 h dark. Up to five replicatesamples per treatment were considered in the experiments. Theexposure containers were permanent aerated, without disturbingthe sediment, and monitored daily for temperature and aeration.Acute sediment toxicity duration was set to 10 d, following thestandard procedure with other species (USEPA, 1994; Bat andRaffaelli, 1998; Bat et al., 1998; ISO, 2005; Casado-Martinezet al., 2007; Guerra et al., 2007; ASTM, 2008; Prato et al., 2008).During the 10 d period, the amphipods were not feed and the waterwas not renewed. At the end of the exposure, the sediment fromeach beaker was sieved through a 0.5 mm mesh screen to collectand count the surviving amphipods.
The response criterion in the acute test is survival. Dead animalsare recognized by their discoloration, absence of pleopod move-ments or lack of response to an external mechanical stimulation.Missing animals are assumed to have died during the exposure.
A. Ré et al. / Chemosphere 76 (2009) 1323–1333 1325
All tests were accompanied by a control sediment (a negativecontrol), which measures the response of the amphipods in the ab-sence of contaminant stress and under the best possible exposureconditions. The control sediment furnishes an acceptability mea-sure of the experiment, by providing evidence of the test organ-ism’s health, and the suitability of the overlying water, testconditions and handling procedures (ASTM, 1992). It can also beused for statistical comparison with the test sediments unless ref-erence sediment is used for that purpose. The control sedimentwas prepared with the sediment from the collection site.
The acute 10 d sediment toxicity test was considered non validwhenever mean survival in the control sediment was below 85%, orsurvival in any individual replicate from the control sediment wasbelow 80% (ISO, 2005).
Besides the negative control, the acute tests were sometimesaccompanied by a positive control, which determines the sensitiv-ity of the animals when exposed to a single reference toxicant un-der repeatable conditions and can be employed to verify if thesensitivity of the adult animals is consistent among experiments.The positive control consisted of a 96 h, water-only exposure tocadmium chloride. The animals for the controls were selected fromthe same population as the test animals. The cadmium chloridetest concentration series was prepared from a solution of10 mgCdCl2 L�1 and corresponded to 0.056, 0.10, 0.18, 0.32, 0.56,1.0, 1.8 and 3.2 mgCdCl2 L�1 (corresponding to 0.034, 0.061,0.110, 0.196, 0.343, 0.613, 1.104 and 1.962 mgCd2+ L�1). The posi-tive control was run twice at 15, 18 and 22 �C, at salinity 2, usingtwo replicates at each concentration and once at 15 �C and salinity18. It was also run at 22 �C and salinity 2 with amphipods collectedin the field and from a 2 year long continuous laboratory culture.cadmium chloride LC50 values, corresponding to the toxicant con-centration causing mortality in 50% of the exposed population,were calculated by linear interpolation (Norberg-King, 1993).
2.4. Response to natural variables
The study of the relationship between survival and naturalenvironmental variables is important in the development of eco-toxicological tests and should be conducted before negative re-sponses can be ascribed to contaminant effects (DeWitt et al.,1988; ASTM, 1992; Prato and Biandolino, 2006). In this respect,the tolerance of C. multisetosum to a full salinity and sediment fineparticles content gradient was investigated in a range of exposuretemperatures.
2.4.1. Interstitial and overlying water salinity in a range of exposuretemperature
C. multisetosum was exposed in the control sediment to a fullgradient of overlying salinity, at 15, 18 and 22 �C, in a 10 d test.The test solutions were prepared by mixing filtered and UV steril-ised natural seawater (salinity 36) with deionised water to obtainthe seawater concentrations 0.0 (deionised water only), 0.2%, 0.4%,0.8%, 1.6%, 3.2%, 6.25%, 12.5%, 25.0%, 50.0% and 100.0% (seawateronly). The test specimens were directly transposed from the cul-tures, set at salinity 18 (corresponding to a concentration of50.0% natural seawater in deionised water), to the various treat-ments. Prior to the inoculation of the amphipods, the control sed-iment was sieved with water at the various seawaterconcentrations, in order to equilibrate the interstitial water salinityto that of the overlying water and acclimatised to the various testtemperatures. The test vessels were left to rest for 24 h, before theinoculation of the amphipods. A negative control was produced,using control sediment exposed at salinity 18 without sieving onwater at the same salinity as described before. This control wascompared to a procedure control, corresponding to the same sedi-ment and salinity but also including the sieving procedure. The
procedure control corresponds to the 50% seawater concentrationlevel. Temperature included three levels, salinity twelve levels,and the experiment was run in three occasions, using three repli-cates per salinity and temperature. The survival data, expressedas a percentage of the exposed population, was arcsine trans-formed (Sokal and Rohlf, 1995) and analysed using 3-way ANOVA,with temperature and salinity as fixed, orthogonal, factors, and theexperiment run nested in both temperature and salinity (randomfactor), under the null hypothesis of no significant interaction be-tween temperature and salinity, no significant survival differenceamong salinity values and no significant survival difference amongtemperature values.
2.4.2. Sediment fine particles content in a range of exposuretemperature
In this experiment, C. multisetosum was exposed to control sed-iment to which a range of silt and clay was added. The exposureswere conducted at three temperatures, 15, 18 and 22 �C, for a10 d period, at salinity 18. The test sediments were prepared bymixing control sediment from which the fines were removed (par-ticles < 63 lm), with portions of a natural sediment with 97% finescontent. The fine particles were removed from the control sedi-ment by washing it through a 63 lm mesh sieve and recoveringthe sieve residue.
The sediments grain-size analysis was performed by wet anddry sieving, following Quintino et al. (1989). The silt and clay frac-tion was wet sieved through a 63 lm mesh screen. The sand (par-ticles with diameter from 63 lm to 2 mm) and the gravel fractions(particles with diameter above 2 mm) were dry sieved. All grainsize classes were expressed as percentage of the total sediment,dry weight. In this experiment, temperature included three levels,sediment fine particles content included nine levels (control sedi-ment, with 1.8% fines and eight prepared sediments, with 2.1%,3.8%, 7.5%, 14.5%, 27.3%, 56.5%, 77.8% and 97.0% fines) and theexperiment was run twice, using three replicates per sedimentfines content and temperature. The survival data, expressed as apercentage of the exposed population, was arcsine transformed(Sokal and Rohlf, 1995) and analysed using 3-way ANOVA, withtemperature and sediment fines content as fixed, orthogonal, fac-tors, and the experiment run nested (random factor) in both previ-ous factors, under the null hypothesis of no significant interactionbetween temperature and sediment fines content, no significantsurvival difference among sediment fines content values and nosignificant survival difference among temperature values.
2.5. Response in natural sediments
Sediment from a site located at approximately 50 m depth onthe mud facies off the Tagus estuary, Western Portugal (site 19–38�39.527 N and 9�26.106 W), was collected yearly with a 0.1 m2
Smith-McIntyre grab between 1997 and 2006 (except in 2005) tosynoptically monitor benthic macrofauna species compositionand abundance, sediment baseline descriptors and organic com-pounds and sediment acute toxicity. Site 19 is influenced by histor-ical contamination originated from the Tagus estuary, in particularpolycyclic aromatic hydrocarbons, PAHs (Quintino et al., 2001; Sil-va et al., 2004). During the monitoring period, heavy rainfall oc-curred during most of the winter 2000–2001, causing one of thelargest floods in the 20th century in Portugal. This is characterizedhere by presenting runoff measured at Almourol (Fig. 1), a site lo-cated in the Tagus River. The data were obtained from the Portu-guese National Information System on Water Resources (SNIRH)and concerns the total yearly runoff from 1990/91 to 2006/07and the total monthly runoff from October 2000 to September2001 (http://snirh.pt/ National Information System on WaterResources).
OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEPMonth (2000/01)
Run
off (
m x
10
)3
6
Fig. 1. Runoff measured at Almourol, on the River Tagus. Total yearly data collected from 1990 to 2007 and detail of the total monthly data in the period October 2000–September 2001.
1326 A. Ré et al. / Chemosphere 76 (2009) 1323–1333
Baseline sediment descriptors included grain-size, redox poten-tial and total volatile solids. Grain-size analysis was performed asdescribed before. The total volatile solids content was determinedby loss on ignition at 450 �C (Byers et al., 1978; Kristensen andAndersen, 1987). Redox potential and temperature were measuredon board, introducing specific probes 4 cm into the sediment sam-ple before emptying the grab (Pearson and Stanley, 1979). The or-ganic compounds analyzed included the chlorinated pesticides,namely lindane and DDT metabolites, PCB congeners (28, 52,101, 118, 153, 138 and 180) and PAH compounds (naphthalene,methyl-naphthalene, dimethyl-naphthalene, trimethyl-naphtha-lene, phenanthrene, anthracene, methyl-phenanthrene, dimethyl-phenanthrene, fluoranthene, pyrene, benzo [e] pyrene, benzo [a]pyrene, benzo [b] fluoranthene, benzo [k] fluoranthene, benzo [a]anthracene, chrysene + triphenylene, indene [1,2,3-cd] pyrene andbenzo [ghi] perylene), following the analytical and QA/QC proce-dures given in Silva et al. (2004).
The benthic data were obtained by standard methods: Smith-McIntyre 0.1 m2 grab samples, sieved through 1 mm mesh, andsorted, identified and enumerated to the highest possible taxo-nomic resolution, usually the species level. Only the samples withsediment in both grab buckets and with a minimum of 5 cm of sed-iment were retained. All the faunal sorting was rechecked by a sec-ond person through a second sorting of the material. The data wereanalyzed at the abundance per taxa level, but also to produce avariety of synthetic indices: S (total species richness), A (totalabundance), d (Margalef diversity index), J0 (Pielou evenness in-dex), H0 (Shannon-Wiener diversity index, loge), 1 � k0 (Simpson’sdiversity index), ES50 (number of expected species in a sample with50 specimens) (Clarke and Warwick, 2001).
The groups of samples representing the before and after theflood time periods were tested for statistical differences, underthe null hypothesis of no significant difference between the twoperiods. The tests were conducted separately for the sedimentbaseline data, the organic compounds data, the benthic macrofa-una abundance and synthetic indices data and the amphipod sed-iment toxicity survival data, and were performed with the ANOSIMprocedure (Clarke, 1993). Years were nested in the before and afterthe flood time periods whenever there were replicate samples peryear (the macrofauna data), allowing a two way nested model withtime periods as fixed factor and the yearly samples as random fac-tor, otherwise a single factor analysis was conducted (all the otherdata). Analysis of Similarities (ANOSIM) produces the statistic R,which relates the within to the between group distances, in a trian-gular resemblance matrix between samples. Depending on thedata, the triangular [sites � sites] resemblance matrices were ob-tained either by calculating the Euclidean distance (ED) followinga log(x + 1) transformation, the normalized Euclidean distance(NED) when variables are expressed in different units, or theBray–Curtis similarity (B–C) for the benthic abundance data, fol-lowing a log(x + 1) transformation. The R statistic varies from �1to +1, approaches the value 0 when the null hypothesis is trueand +1, rejecting the null hypothesis, when all the distance valuesbetween groups are larger than all the within group distances. TheR statistic is accompanied by a significance value obtained by cal-culating the probability of the observed R within a series of R val-ues obtained after a permutation procedure (Clarke and Warwick,2001). The results obtained are summarized in tables and pre-sented in 2D non-metric multidimensional scaling ordination dia-grams. All data analysis was performed with PRIMER 6 (Clarke and
A. Ré et al. / Chemosphere 76 (2009) 1323–1333 1327
Gorley, 2006), with the add-on PERMANOVA+ (Anderson et al.,2008).
3. Results
3.1. Laboratory cultures
The amphipod culture conditions reported by DeWitt et al.(1992) set the basis for those used in this study. Filtered and UVsterilized natural seawater collected locally during high tide, di-luted in distilled water to final salinity 18, was used in all cultures,instead of the natural water obtained from the amphipods collec-tion site. To avoid overcrowding, cultures were thinned approxi-mately twice a month by sieving through a 1 mm mesh sieve,allowing the young to pass and remain in the sediment. Only about500 healthy adults were returned to the culture. The rest wereused to start new cultures. Densities were maintained below1000 individuals per culture to prevent overcrowding (in cultureboxes with a surface area of about 0.09 m2, filled with an approx-imately 2 cm layer of culture sediment and topped by 6–7 cm ofculture water). As overcrowding occurred (more than 2000 indi-vidual per culture), it was noticed that smaller animals and fewgravid females with very small number of eggs dominated the cul-tures. If kept under constant and favourable conditions C. multi-setosum cultures provide animals in sufficient numbers fortoxicity tests year-round.
3.2. Test design and procedure: negative control and response to areference toxicant
The mean percent survival obtained in the control sediments(=negative control), was always higher than 85% and in each indi-vidual replicate it was also always higher than 80%, meeting theacceptability criteria established for this type of sediment tests(ISO, 2005).
Concerning the response to the reference contaminant cad-mium chloride, Table 1 presents the LC50 values obtained for C.
Table 1Cadmium chloride LC50 values (mgCd2+ L�1) in 96 h water-only exposures for a variety of
a Salinity values, temperature or both, unavailable.b Tests conducted with laboratory cultured amphipods.c Test conducted with amphipods at the end of a 2 year long uninterrupted laboratord Test conducted at the same time as Footnote c, with field collected amphipods.e Winter.f Summer.
multisetosum in this study and for other amphipod species usedin sediment toxicity tests, showing that the results here obtainedare in the lower range of those reported for other species. The re-sults also revealed a general tendency to obtain lower LC50 valueswith increasing temperature, which may be due to an increasingabsorption of the contaminant related to higher metabolic ratesat higher temperature. The results also indicate that the amphi-pods raised in the laboratory were neither more sensitive nor moreresistant to the reference toxicant, given the closeness of the LC50
values obtained with amphipods from a 2 year consecutive cultureand a field population, 0.34 and 0.31 mgCd2+ L�1, respectively (cf.Table 1).
3.3. Response to natural variables
3.3.1. Interstitial and overlying water salinity in a range of exposuretemperatures
C. multisetosum survival data when exposed for 10 d in the con-trol sediment to a full salinity range at various temperatures areshown in Table 2. Mean survival was above 90% at seawater con-centrations between 6.25% and 50.0% and tended to diminish to-wards the extreme salinity values, in all three temperatures. Thevariability around the mean was also consistently lower betweenthose two salinity concentrations, irrespective of temperature. Sur-vival was below 80% within a combination of the highest watertemperature with the lowest seawater concentrations and the low-est water temperature with the highest seawater concentration.Mean percent survival was lowest in full deionised water at thehighest temperature (cf. Table 2). The described survival patternof C. multisetosum, accounts for a significant interaction betweensalinity and temperature, as shown in the ANOVA results, Table3. As such, differences between salinity levels were checked indi-vidually at each temperature, and differences between tempera-ture levels were checked separately within salinity. Thesepairwise comparisons indicated that survival in the negative con-trol (salinity 18) was different from that at 50.0% seawater concen-tration (salinity 18) only at 15 �C (p < 0.01), suggesting that the
20 ± 2; 32 ± 2 Lamberson et al. (1992)16; 36–37 Prato and Biandolino (2006)16 ± 2; 36 Prato et al. (2006)16 ± 2�C; a Annicchiarico et al. (2007)10, 15, 20 and 25; 35.9 ± 0.2 Prato et al. (2008)15 ± 1; 18 This studyb
15 ± 1; 2 This studyb
18 ± 1; 2 This studyb
22 ± 1; 2 This studyb
22 ± 1; 2 This studyc
22 ± 1; 2 This studyd
16 ± 2; 35 Picone et al. (2008)10, 15, 20, 25 and 30; 35 Bigongiari et al. (2004)16 ± 2; 36 ± 1 Lera et al. (2008)
14.5–16.5; 29.7–32 Ciarelli (1994)11 ± 1; 32 ± 1 Bat et al. (1998)15; 33 Costa et al. (1998)16 ± 2; 36 Prato et al. (2006)20 ± 2; 32 ± 2 Lamberson et al. (1992)15; 28 DeWitt et al. (1992)
y culture.
Table 2C. multisetosum mean percent survival (±standard deviation) in 10 d control sedimentacute exposure in a range of interstitial and overlying water salinity and temperatures(n = 9 for all the treatments). The negative control corresponds to a seawaterconcentration of 50% (see Section 2).
Seawater concentration (%) Mean amphipod survival (%±sd)
Table 3Analysis of variance. Comparison of C. multisetosum survival data when exposed to arange of temperature values to a full interstitial and overlying water salinity gradient.Temperature and salinity are crossed fixed factors and the experiment run is nested inboth previous factors.
Table 4C. multisetosum mean percent survival (±standard deviation) in acute exposure tosediments with a prepared range of fine particles content at various temperatures(n = 6 for all treatments). Fine particles content expressed as a % of total sediment, dryweight. The control sediment presents 1.8% fines content.
Sediment fines content (%) Mean amphipod survival (%±sd)
1328 A. Ré et al. / Chemosphere 76 (2009) 1323–1333
adjustment of the interstitial salinity of the sediment in the prep-aration phase could nevertheless play some role in the results,which is in agreement with the fact that survival was almost al-ways higher in the negative control than in the treatments (cf. Ta-ble 2). The comparison between the temperature levels withineach salinity showed that there were no differences between thethree temperatures with the exception of the seawater concentra-tion 0.0% and 6.25% and only between the temperatures 15 �C and22 �C (p < 0.05). In the comparison between seawater concentra-tion values within each temperature, only the treatment seawaterconcentration 50.0% (salinity 18) was compared to the remaining,given that this was the salinity of the negative control. Survivalin this treatment was significantly different from that obtained inseawater concentration 100.0% (seawater only) at temperature15 �C (p < 0.01) and from that in seawater concentration 0.0%(deionised water only), at temperature 22 �C (p < 0.05). At temper-ature 18 �C no differences were obtained between the 50.0% sea-water concentration level and any of the other seawaterconcentration treatment levels.
Overall, the high survival obtained in these experiments indi-cates that C. multisetosum has high tolerance to salinity and thatacute toxicity tests can be performed within wide range of salinity.Nevertheless, the variability around the mean advises to avoidtesting in the extreme situations, such as below salinity 2 at thehigher temperature. In fact, the significant interaction betweentemperature and salinity indicates that testing of sediments fromthe upper estuary (lower ambient salinity) should take place atlower temperature whereas tests with sediments from the lowerestuary (higher ambient salinity) should be performed at highertemperature, in a strategy to avoid false positives (toxicity re-sponse due to natural factors and not to degraded sediments).
In this experiment, the control sediment was handled in a wayto ensure equilibrium between the test sediment interstitial water
and the overlying test water salinity. This manipulation is never-theless artificial when analysing natural sediments. Another exper-iment was then conducted, in which marine sublittoral sedimentfrom an exposed sandy beach was offered as test sediment. Thissediment was tested using as overlying water full seawater (salin-ity 36) and seawater diluted in deionised water to a final salinity of2 (seawater concentration close to 6.25%). The experiment was runonly with these two salinity concentrations and at temperatures 15and 22 �C. In this experiment, no attempt was made to equilibratesalinity in the sediment interstitial water and the overlying water.All remaining testing conditions were similar to the previousexperiment. Mean percent survival (± standard deviation) reached93.3% (±2.89) and 90.0% (±5.00), at 15 �C and 22 �C, respectively, inwater with salinity 2, and 78.3% (±10.41) and 83.3% (±5.77), at15 �C and 22 �C, respectively, in seawater with salinity 36. Theseresults are in close agreement with those reported in the previousexperiment.
3.3.2. Sediment fine particles content in a range of exposuretemperature
Table 4 summarises C. multisetosum survival data, obtained inthe 10 d exposure experiments at various temperatures to sedi-ments with a range of fines content (particles < 63 lm). Mean sur-vival was always higher than 85% in the control sediment and inthe majority of the tested sediments. Within each temperature, atendency was observed to obtain lower survival at the higher finescontent, especially at the higher temperatures. The lowest survivalat each temperature was nevertheless obtained at intermediatefines contents (56.5%), namely at 15 and especially at 22 �C butwith the exception of 18 �C (cf. Table 4). These patterns explainthe significant interaction term (Temperature � Fines content) inthe ANOVA results, Table 5. Differences in survival between finescontent levels were thus checked for each temperature, and differ-ences between temperature levels were checked separately withineach fines content level. Significant survival differences betweentemperature levels were mainly noticed at the 56.5% fines contentlevel, with significantly different survival (p < 0.001) at all temper-ature levels. Apart from these, differences were only observed at7.5% fines, between 18 and 22 �C (p < 0.05) and at 14.5% fines, be-tween 15 and 22 �C (p < 0.001). Survival differences between thefines content levels, at each exposure temperature, were checkedbetween the control sediment and the treatment levels. At 15 �Cthere were no significant differences, at 18 �C differences were no-ticed between the control and the fines content levels 7.5%(p < 0.05), 56.6% (p = 0.01) and 97% (p < 0.01). At 22 �C, the majordifferences were observed between the control and the 56.5% finescontent level (p < 0.01), but marginal differences (0.05 > p > 0.04)were also noticed between the control and the fines content levels14.5%, 27.3%, 77.8% and 97.0%.
Table 5Analysis of variance. Comparison of C. multisetosum survival data when exposed in arange of temperature values to control sediment prepared with a range of finescontent (particles with diameter <0.063 mm). Fine particles content expressed as a %of total sediment, dry weight. Temperature and sediment fines content are crossedfixed factors and the experiment run is nested in both previous factors.
Table 6Benthic macrofauna indices, sediment baseline, sediment organic compounds andacute toxicity descriptors studied at site 19 in the period 1997–2006. Mean valueswith standard deviation for the time periods before and after the flood. Macrofaunadata: n = 12, in both time periods; all other data: n = 4 before and n = 5 after. S –number of species; A – abundance; d – Margalef diversity; J0 – Pielou eveness; ES(50) –number of species expected in a sample with 50 specimens; H0 – Shannon-Wienerdiversity; 1 � k0 – Simpson diversity; >2 mm till <0.063 mm – grain-size fractions;TVS – total volatile solids; Eh – sediment redox potential.
A. Ré et al. / Chemosphere 76 (2009) 1323–1333 1329
These results indicate an overall tendency for increasing sensi-tivity of the adult population to fines content of the sediment withincreasing temperature, thus suggesting that tests which have toaccommodate sediment with a wide range of fines content shouldpreferably be performed at a lower temperature, namely 15 �C.These results also point out the highest sensitivity of the speciesto intermediate values of fines, especially clear at higher tempera-ture. It was noticed that at the higher fines particles content(97.0%), the amphipods do not attempt to build a tube, whereasthey will devote energy for that at lower fines content. This at-tempt however was never successful in sediment with fine parti-cles close to 50%, which may eventually justify the highersurvival in sediments with fines content above the threshold atwhich the amphipods no longer try to build the tube.
3.4. Response in natural sediments
The acute toxicity of natural sediments to C. multisetosum wasstudied by exposing the species to sediment samples collected ata site located on the mud facies off the Tagus estuary (site 19, cf.Quintino et al., 2001; Silva et al., 2004). Either due to runoff, tothe coastal storm episodes which occurred during winter 2000–2001 or to a mixture of both factors, the superficial sediment at site19 experienced alterations at the grain-size level as well as interms of the organic compounds concentrations, especially thePAHs, which presented much lower values following that winter.The sediment grain-size alterations consisted of a relative increasein the particle size fraction 0.063–0.125 mm and a relative dimin-ish in the other fractions. Table 6 presents the mean values forthese and other sediment variables in the time periods beforeand after the flood observed in winter 2000–2001, as well as forthe biological and the ecotoxicological descriptors studied. The dif-ferences between the two time periods are very clear and concernan increase in species richness (S) as well as in all the measuresassociated with diversity (d, H0, 1 � k0, ES(50)), while lowering dom-inance (J) and maintaining species abundance (A). This biologicalimprovement is accompanied by diminishing concentrations ofthe organic compounds in the superficial sediment, especially totalPAHs and by diminished mortality of C. multisetosum in the acutetoxicity exposures (cf. Table 6). The high statistical significance ofthese alterations is shown in Table 7, which summarises the R sta-tistic values and associated probability for the one way and twoway nested ANOSIM tests conducted. Fig. 2 shows the ordinationdiagrams for the benthic macrofauna indices (A), the sedimentbaseline descriptors (B), the sediment organic compounds (C)and the mortality in the sediment toxicity tests (D), illustrating,for each set of data, the distinction between the two sampling peri-ods. Fig. 3, shows the ordination diagrams obtained with the taxaabundance data during the study period, individualizing all thereplicate samples (A) and their centroids (B). The figure shows avery clear separation between the two time periods (before andafter the flood) and also the larger spread of the data in the after
flood period, indicating that the benthic community was alteredby the flood event and is showing larger variability than beforethe flood.
4. Discussion and conclusions
C. multisetosum was successfully kept in laboratory culturesduring 2 years, supplying newborns, juveniles, and adults in suffi-cient number to run the tests described in the present work. Thisresult is in favour of the viability of C. multisetosum as a test spe-cies. DeWitt et al. (1992) for Leptocheirus plumulosus, Nair and An-ger (1979), Lamberson et al. (1992) for Corophium insidiosum andCorreia et al. (1995) for Gamarus locusta give examples of regularmaintenance of laboratory amphipod cultures.
The sensitivity of C. multisetosum to the reference toxicant cad-mium chloride (CdCl2) was studied in 96 h water-only exposure,adapting the methodologies used by DeWitt et al. (1992), ASTM(1992, 1993), Ciarelli (1994), USEPA (1994) and Gully et al.(1995). The LC50 values reported in this study for C. multisetosumare generally lower than those reported by other authors for otherCorophium species (cf. Table 1). The data here obtained also indi-cates that the toxic effects associated with the exposure to cad-mium chloride increased with temperature, presumably due toan increase in the metabolic rate, so the LC50 values decreased from0.574 mgCd2+ L�1 at 15 �C to 0.327 mgCd2+ L�1 at 22 �C. Changes inthe toxicity of chemicals due to temperature have been attributedby some authors to changes in respiration rate, chemical absorp-tion, excretion and detoxification (Knezovich, 1992). It is thus nec-essary to determine testing conditions, namely exposuretemperature, because this factor may interfere with the results(Bigongiari et al., 2004; Prato et al., 2008). Physicochemical factorsmay not affect the acute toxicity per se, but rather the bioavailabil-ity or the chemical form of contaminants. According to DeWittet al. (1992), temperature may modify a species’ sensitivity byaltering its metabolic rates; these authors reported that sensitivityof L. plumulosus to cadmium is strongly temperature dependent.
Table 7R statistic and associated probability values obtained in the one way and two way nested ANOSIM tests performed for the various study components monitored at site 19 in theperiod 1997–2006. For all one way tests and for the test for differences between the flood groups in the two way nested designs, the probability value is based on a maximumpossible number of 126 permutations, giving a minimum possible probability value of 0.008. In the test for differences between the year groups, in the two way nested designs, amuch higher number of permutations are possible and so the significance (p-value) was set for 999 permutations. See Table 6 for the meaning of abbreviations given in thesediment baseline descriptors and the benthic macrofauna indices.
Study component R-value p-Value
One way ANOSIMBenthic macrofauna abundance per taxa – year centroids 0.77 0.008Benthic macrofauna indices – year centroids 0.83 0.008Sediment baseline descriptors (grain size, TVS and Eh) 0.67 0.008Sediment organic compounds (PAH, PCB, DDT and lindane) 0.47 0.04Sediment acute toxicity (acute mortality) 0.65 0.008
Two way nested ANOSIMBenthic macrofauna abundance per taxa
Test for differences between year groups 0.71 <0.001Test for differences between flood groups 0.79 0.008
Benthic macrofauna indices (S, A, d, J0 , H0(loge), 1 � k0 and ES50)Test for differences between year groups 0.70 <0.001Test for differences between flood groups 0.78 0.008
1997
1998
1999 2000
2001
2002
2003
2004
2006
2D Stress: 0.01A
1997
1998
1999
2000
2001
2002
2003
2004
2006
2D Stress: 0.06
B
1997
1998
1999
2000
2001
2002
2003
2004
2006
2D Stress: 0.01C
1997 1998
19992000
2001
2002
20032004
2006
2D Stress: 0D
Flood: before after
Fig. 2. Non-metric multidimensional scaling diagrams of the data collected at site 19 from 1997 to 2006, individualizing the periods before and after the flood that occurredin the winter 2000–2001. A: centroids of the replicate samples taken yearly for the benthic macrofauna indices; B: sediment baseline data; C: sediment organic compounds;and D: acute toxicity mortality. See Tables 6 and 7 for the mean values of the various descriptors studied in the two time periods and the ANOSIM tests results.
1330 A. Ré et al. / Chemosphere 76 (2009) 1323–1333
The importance of maintaining historical and consistent recordswas thus reported by Gully et al. (1995), namely the quality assess-ment of testing protocols, inter-laboratorial experiments or inter-bioassays variability.
In the course of this study, the adult amphipods obtained fromlaboratory cultures did not show increased resistant to the refer-ence toxicant, nor the opposite. However, several authors (Ciarelliet al., 1997; McGee et al., 1998; Kater et al., 2000) compared thesensitivity of laboratory to field collected animals and concludedthat the field populations were typically more sensitive to refer-
ence contaminant, namely cadmium chloride, and appeared to ex-hibit seasonal variability possibly relate to natural variations infood availability. Sosnowski et al. (1979), examined the toxicityof copper to field populations of the copepod Acarcia tonsa and con-cluded that the LC50 values obtained in toxicity tests with field-col-lected organisms were more variable than those obtained withlaboratory organisms, and they attribute the variability to seasonalchanges in food supply and population density. Similarly, Tatemet al. (1976) reported enhanced sensitivity of grass shrimp to dode-cyl sodium sulphate during winter, presumably in response to food
1997(1)
1997(2)
1997(3)1998(1)
1998(2)
1998(3)
1999(1) 1999(2)1999(3)
2000(1)
2000(2)
2000(3)
2001(1)
2001(2)
2001(3)
2002(1)
2003(1)
2003(2)
2004(1)2004(2)
2004(3)
2006(1)
2006(2)2006(3)
2D Stress: 0.14A
Floodbeforeafter
1997
1998 1999
2000
2001
2002
2003
2004
2006
2D Stress: 0.05B
Fig. 3. Non-metric multidimensional scaling diagrams of the benthic macrofaunaabundance data collected at site 19 from 1997 to 2006, individualizing the periodsbefore and after the flood that occurred in the winter 2000–2001. A: individualreplicates and B: year centroids. See Table 7 for the associated two way nestedANOSIM test results.
A. Ré et al. / Chemosphere 76 (2009) 1323–1333 1331
limitations. Lera et al. (2008) also reported higher sensitivity ofCorophium orientale to cadmium chloride in winter, when com-pared to summer (cf. Table 1). Our results however, showed a sim-ilar sensitivity response between field collected and laboratoryamphipods at the end of an uninterrupted 2 year long culture (cf.Table 1). The controlled C. multisetosum laboratory culture condi-tions with an adequate feeding plan seems to be a good way to ob-tain test organisms with decreased variability in their physiologicalstatus and consequently in the variability of their toxicologicalresponses.
When exposed in the control sediment to a range of interstitialand overlying water salinity values, C. multisetosum survivaltended to be lower at extreme salinities, namely below 0.4% (salin-ity < 1) and in full seawater (salinity 36). The results also indicatedincreasing sensitivity with increasing temperature, especially atthe lower salinity values. Although C. multisetosum could be usedto test sediments with interstitial and overlying water salinityfrom full freshwater to full seawater, testing at the most extremesituations should be avoided. According to the data obtained in thisstudy, the recommended temperature for the 10 d acute sedimenttoxicity test with C. multisetosum is 15 �C for lower salinity and22 �C for higher salinity. These results agree with the conditionsto which the species is exposed in nature: the field populationsduring summer, with increasing temperature and lower freshwaterinputs, are exposed to higher water salinity, whereas in winter, theopposite situation prevails. Fully marine sediments can thus be as-sessed at higher temperature (�22 �C) and upstream sedimentsshould be assessed at lower temperatures (�15 �C). Survival higherthan 90% was found at the three tested temperatures (15, 18 and22 �C) for salinity values from 2 to 20. The results obtained also
indicate that marine sediments may be tested with survival above90% if the salinity of the overlying water is lowered. However, thisprocedure should be considered carefully due to the higher bio-availability of some contaminants namely metals, with decreasingsalinity (USEPA, 1994).
The sediment content in fine particles (below 0.063 mm) had asmall effect on the survival of C. multisetosum. Although mean sur-vival was almost always higher than 85% in the tested fines con-centrations, ranging from 2% to 97% of total sediment, dryweight, in nature the species exhibits preference for sedimentswith fewer amounts of fines (Queiroga, 1992). Survival also tendedto be lower at intermediate fine contents (�50%), and this wasespecially observed at the higher temperature. In this amount offines, the amphipod could not build a tube, but spent energy tryingto construct it. In sediment with much more than 50% of fine con-tents, the amphipod did not try to build the tube, just burrowed,thus saving energy which may explain their higher survival inthe sediments with higher fines content when compared to sedi-ment with �50% fines content. A direct relationship between finescontent and total volatile solids content of sediment has been doc-umented, indicating that organic contaminants will also usually behigher in sediments with higher fines content (Silva et al., 2004;Gray and Elliott, 2009). It is thus fundamental to be certain thatthe species is not responding to fines but to other factors, whentesting sediments with a high content in fines particles. The resultspresented in this study confirm the finding of Castro et al. (2006),who also verified that the exposure of C. multisetosum to finer sed-iment did not result in diminished survival or development andthat the survival results they reported could not be related to base-line sediment variables (fines content, total volatile solids contentand redox potential). Since estuarine sediments are heterogeneousin terms of grain size, these results indicate a potential wide use ofthis amphipod in sediment acute toxicity tests. The successful useof C. multisetosum in a full range of sediment fine contents isimportant because sediments with different grain size have signif-icantly different capacities for collecting contaminants. Accordingto Chapman and Wang (2001), sediment grain size is probablythe most important factor controlling sediment metalconcentration.
The acute tests with adult organisms have the advantage ofallowing a faster acquisition of results and are a good choice to ob-tain an overall image of an area (Castro et al., 2006), and so they aremuch more common as a screening tool (Cole et al., 2000; Longet al., 2001). The present study indicates that C. multisetosum per-forms well as such a screening tool, given that it shows a strongeracute response when exposed to higher levels of contaminationand is in close agreement with the response from other sedimentdisturbance assessment tools, such as a modification of the resi-dent benthic community. This was clearly confirmed in this study,in a synoptic survey from 1997 to 2006 conducted at a site locatedoff the Tagus Estuary, Western Portugal, including macrofaunabenthic communities’ descriptors, sediment baseline descriptorsand concentrations in organic compounds and sediment acute tox-icity data. The results reported in Ré et al. (2007) also corroboratethis. In that study, the C. multisetosum lower survival and the Para-centrotus lividus lower fertilization success were mostly occurringat the same sites, which were also highlighted through the studyof the macrofauna benthic community showing a strong reductionin species richness, abundance and biomass. Nevertheless, underlower contaminant concentrations, the acute response of C. multi-setosum may be less ambiguous and it should be expected that asublethal descriptor will be more efficient in discriminating sites.
Estuarine sediments vary considerably with regard to intersti-tial and overlying salinity, sediment grain-size, temperature andtotal volatile solids content. In some cases, these differences havebeen ignored and sediment toxicity has been evaluated using
1332 A. Ré et al. / Chemosphere 76 (2009) 1323–1333
marine or freshwater species exposed to interstitial water or elutri-ates or to sediments whose salinity has been artificially increasedor decreased. In all these cases, potential alterations in contami-nant bioavailability and toxicity may occur. The only reasonableway to determine the toxicity and bioavailability of contaminantsin estuarine sediments is to test the sediments as they are, usingestuarine organisms capable of tolerating the full range of estua-rine conditions, in particular salinity (Chapman and Wang, 2001)and sediment fines content. In this concern, C. multisetosum isunquestionably a good species to use in the bioassessment of estu-arine sediments acute toxicity, with a test specifically designed forestuarine conditions. Also, the species presents a high potential forthe development of full life-cycle chronic essays, giving its amenityto laboratory culture and consistent high survival under sedimentcontrol conditions.
Acknowledgments
This work was co-financed by Direcção Geral do Ambiente(DGA), under the project ‘‘Development and comparative study ofecotoxicological descriptors for the bioassessment of sedimentsand industrial residues (DGA, PEAM/NMA/685/95), and by SANEST,Saneamento da Costa do Estoril, S. A., under the project ‘‘Programade Monitorização Ambiental do Emissário Submarino e da ETAR daGuia do Sistema Multimunicipal de Saneamento da Costa do Esto-ril”. Authors acknowledge the comments from two anonymousreviewers.
References
Allen, Y.T., Thain, J.E., Haworth, S., Barry, J., 2007. Development and application oflong-term sublethal whole sediment tests with Arenicola marina and Corophiumvolutator using Ivermectin as the test compound. Environ. Pollut. 146, 92–99.
Anderson, M.J., Gorley, R.N., Clarke K.R., 2008. PERMANOVA+ for PRIMER: Guide toSoftware and Statistical Methods. PRIMER-E: Plymouth, UK.
Annicchiarico, C., Biandolino, F., Cardellicchio, N., Di Leo, A., Giandomenico, S., Prato,E., 2007. Predicting toxicity in marine sediment in Taranto Gulf (ionian Sea,Southern Italy) using sediment quality guidelines and a battery bioassay.Ecotoxicology 16, 239–246.
ASTM (American Society for Testing and Materials), 1992. Standard guide forconducting 10-days static sediment toxicity tests with marine and estuarineamphipods. In: Annual Book of ASTM Standards, Water and EnvironmentalTechnology, vol. 11.04, E1367-90. American Society for Testing and Materials,Philadelphia, PA, USA.
ASTM (American Society for Testing and Materials), 1993. Standard guide forconducting sediment toxicity tests with freshwater invertebrates. In: AnnualBook of ASTM Standards, Water and Environmental Technology, vol. 11.04,E1383-93. American Society for Testing and Materials, Philadelphia, PA, USA.
ASTM (American Society for Testing and Materials), 2008. Standard test method formeasuring the toxicity of sediment-associated contaminants with estuarineand marine invertebrates. Standard E1367-03 (2008). ASTM International, WestConshohocken, PA, USA.
Bat, L., 2005. A review of sediment toxicity bioassays using the amphipods andpolychaetes. Turk. J. Fish. Aquat. Sci. 5, 119–139.
Bat, L., Raffaelli, D., 1998. Sediment toxicity testing: a bioassay approach using theamphipod Corophium volutator and the polychaete Arenicola marina. J. Exp. Mar.Biol. Ecol. 226, 217–239.
Bat, L., Raffaelli, D., Marr, I.L., 1998. The accumulation of copper, zinc and cadmiumby the amphipod Corophium volutator (Pallas). J. Exp. Mar. Biol. Ecol. 223, 167–184.
Benton, M.J., Malott, M.L., Knight, S.S., Cooper, C.M., Benson, W.H., 1995. Influence ofsediment composition on apparent toxicity in a solid-phase test usingbioluminescent bacteria. Environ. Toxicol. Chem. 14, 411–414.
Bigongiari, N., Braida, T., Carretti, F., Pellegrini, D., 2004. Influence of temperature onthe mortality and sensitivity of Corophium orientale. Bull. Environ. Contam.Toxicol. 72, 881–887.
Brown, R.J., Conradi, M., Depledge, M.H., 1999. Long-term exposure to 4-nonylphenol affects sexual differentiation and growth of the amphipodCorophium volutator (Pallas, 1766). Sci. Total Environ. 233, 77–88.
Byers, C., Mills, E.L., Stewart, P.L., 1978. A comparison of methods of determiningorganic carbon in marine sediments with suggestions for a standard method.Hydrobiologia 58, 43–47.
Casado-Martinez, M.C., Forja, J.M., DelValls, T.A., 2007. Direct comparison ofamphipod sensitivities to dredged sediments from Spanish ports.Chemosphere 68, 677–685.
Castro, H., Ramalheira, F., Quintino, V., Rodrigues, A.M., 2006. Amphipod acute andchronic sediment toxicity assessment in estuarine environmental monitoring:an example from Ria de Aveiro, NW Portugal. Mar. Pollut. Bull. 53, 91–99.
Chapman, P.M., Long, E.R., 1983. The use of bioassays as part of a comprehensiveapproach to marine pollution assessment. Mar. Pollut. Bull. 14, 81–84.
Ciarelli, S., 1994. Guideline for conducting 10 days static sediment toxicity testsusing marine or estuarine amphipods. Report RIKZ – 94.031 – National Institutefor Coastal and marine Management (RIKZ).
Ciarelli, S., Vonck, W.A.P.M.A., van Straalen, N.M., 1997. Reproducibility of spiked-sediment bioassays using the marine benthic amphipod, Corophium volutator.Mar. Environ. Res. 43 (4), 329–343.
Clarke, K.R., 1993. Non-parametric multivariate analysis of changes in communitystructure. Aust. J. Ecol. 18, 117–143.
Clarke, K.R., Gorley, R.N., 2006. PRIMER v6: User Manual/Tutorial. PRIMER-E:Plymouth, p. 190.
Clarke, K.R., Warwick, R.M., 2001. Change in Marine Communities: An Approach toStatistical Analysis and Interpretation, second ed. PRIMER-E, Plymouth, UK.
Cole, F.A., Boese, B.L., Swartz, R.C., Lamberson, J.O., DeWitt, T.H., 2000. Effects ofstorage of sediments spiked with fluoranthene to the amphipod Rhepoxoniusabronius. Environ. Toxicol. Chem. 19, 744–748.
Correia, A.D., Costa, F.O., Costa, M.H., 1995. Gammarus locusta (Amphipoda) comoindicador biológico. Dinâmica da população e metodologias de manutenção daespécie em laboratório. 1� Congresso Ibérico sobre Contaminação e ToxicologiaAmbiental – Coimbra Março 95 (Livro de Resumos P40).
Costa, F.O., Correia, A.D., Costa, M.H., 1998. Acute marine sediment toxicity: apotential new test with the amphipod Gammarus locusta. Ecotoxicol. Environ.Saf. 40 (1–2), 81–87.
Costa, F.O., Neuparth, T., Correia, A.D., Costa, M.H., 2005. Multi-level assessment ofchronic toxicity of estuarine sediments with the amphipod Gammarus locusta:II. Organism and population-level endpoints. Mar. Environ. Res. 60, 93–110.
Cunha, M.R., Moreira, M.H., Sorbe, J.C., 2000. The amphipod Corophium multisetosum(Corophiidae) in Ria de Aveiro (NW Portugal). I. Life history and aspects ofreproductive biology. Mar. Biol. 137, 637–650.
Dauvin, J.-C., Ruellet, T., 2007. Polychaete/amphipod ratio revisited. Mar. Pollut.Bull. 55, 215–224.
DeWitt, T.H., Distworth, G.R., Swartz, R.C., 1988. Effects of natural sediment featureson survival of the phoxocephalid amphipod, Rhepoxymius abronius. Mar.Environ. Res. 25, 99–124.
DeWitt, T.H., Redmond, M.S., Sewall, J.E., Swartz, R.C., 1992. Development of achronic sediment toxicity test for marine benthic amphipods. Chesapeake BayProgram – Contribution no. N-24 from EPA’s Environmental ResearchLaboratory – Narragansett.
Gray, J.S., Elliott, M., 2009. Ecology of Marine Sediments. Oxford University Press,Hull University, UK. 225pp.
Guerra, R., Pasteris, A., Ponti, M., Fabbri, D., Bruzzi, L., 2007. Impact of dredging in ashallow coastal lagoon: Microtox� Basic Solid-Phase Test, trace metals andCorophium bioassay. Environ. Int. 33, 469–473.
Gully J.R., Baird, R.B., Markle, P.J., Bottomley, J.P., 1995. Evaluation of submittedreference toxicant data and preliminary recommendations of toxicant andconcentration series for the EPA freshwater chronic protocols. SouthernCalifornia Toxicity Assessment Group (SCTAG), Quality Assurance Committee(QAC), Second World SETAC Congress – Vancouver 5–9 November 1995. Canada(Abstract Book pp. 315–PH 161).
ISO (International Standard Organization), 2005. ISO 16712:2005 (E): Water quality– determination of acute toxicity of marine and estuarine sediment toamphipods.
Kater, B.J., Hannewijk, A., Postma, J.F., Dubbeldam, M., 2000. Seasonal changes inacute toxicity of cadmium to amphipod Corophium volutator. Environ. Toxicol.Chem. 19 (12), 3032–3035.
Kirkpatrick, A.J., Gerhardt, A., Dick, J.T.A., McKenna, M., Berges, J.A., 2006. Use ofmultispecies freshwater biomonitor to assess behavioral changes of Corophiumvolutator (Pallas, 1766) (Crustacea, Amphipod) in response to toxicant exposurein sediment. Ecotoxicol. Environ. Saf. 64, 298–303.
Knezovich, J.P., 1992. Chemical and biological factors affecting bioavailability ofcontaminants in seawater. In: Bioavailability, Physical, Chemical and BiologicalInteractions. SETAC Special Publications Series, Proceedings of the ThirteenthPellston Workshop, Pellston, Michigan, 239pp.
Kristensen, E., Andersen, F.Ø., 1987. Determination of organic carbon in marinesediments: a comparison of two CHN-analyzer methods. J. Exp. Mar. Biol. Ecol.109, 15–23.
Lamberson, J.O., Swartz, R.C., 1988. Use of bioassays in determining the toxicity ofsediment to benthic organisms. In: Evans, M.S. (Ed.), Toxicology Contaminationand Ecosystem Health; A Great Lakes Focus. John Wiley and Sons, New York, pp.257–259.
Lamberson, J.O., Swartz, R.C., 1989. Spiked-sediment toxicity test approach. In:Sediment Classification Methods Compendium. USEPA, Criteria and StandardsDivision, Washington, DC.
Lamberson, J.O., Redmond, M.S., Jones, J.K.P., Sewall, J.E., Chapman, J.W., Cole, F.A.,Schults D.W., Swartz, R.C., 1992. Development of an amphipod sedimenttoxicity test for Hawaii. U.S.EPA, ERL-N ASCI, August 1992.
Lera, S., Macchia, S., Dentone, L., Pellegrini, D., 2008. Variations in sensitivity of twopopulations of Corophium orientale (Crustacea: Amphipoda) towards cadmiumand sodium laurylsulphate. Environ. Monit. Assess. 136, 121–127.
A. Ré et al. / Chemosphere 76 (2009) 1323–1333 1333
Long, E.R., Hong, C.B., Severn, C.G., 2001. Relationships between acute sedimenttoxicity in laboratory tests and abundance and diversity of benthic infauna inmarine sediments: a review. Environ. Toxicol. Chem. 20, 46–60.
Luoma, S.M., Carter, J.L., 1993. Understanding the toxicity of contaminants insediments: beyond the bioassay-based paradigm. Environ. Toxicol. Chem. 12,793–796.
Marques, J.C., Bellan-Santini, D., 1985. Contribution a l’étude systématique etécologie des amphipodes (Crustacea-Amphipoda) des côtes du Portugal.Premier inventaire des espèces Gammariens et Caprelliens. Cien. Biol. Ecol.Sist. 5, 299–353.
Marques, J.C., Bellan-Santini, D., 1990. Faune d’amphipodes des milieux lagunairesdu Portugal: actualisation des donnés et comparaison avec d’autres systèmesatlantiques et méditerranées. Bull. Mus. Hist. Nat. Marseilles 50, 53–61.
Morales-Caselles, C., Kalman, J., Micaelo, C., Ferreira, A.M., Vale, C., Riba, I., DelValls,T.A., 2008. Sediment contamination, bioavailability and toxicity of sedimentsaffected by an acute oil spill: four years after the sinking of the tanker Prestige(2002). Chemosphere 71, 1207–1213.
Nair, K.K.C., Anger, K., 1979. Life cycle of Corophium insidiosum (Crustacea,Amphipoda) in laboratory culture. Helgol. Meeresunters 32, 279–294.
Norberg-King, T.J., 1993. A linear interpolation method for subletal toxicity. TheInhibition concentration (Icp) approach. Version 2.0. National Effluent ToxicityAssessment Centre Technical Report 03-93, Environmental Research LaboratoryDuluth, Minnesota, MN 55804, June 1993.
Pearson, T.H., Stanley, S.O., 1979. The comparative measurement of the redoxpotential of marine sediments as a rapid means of assessing the effect of organicpollution. Mar. Biol. 53, 371–379.
Peters, C., Ahlf, W., 2005. Reproduction of the estuarine and marine amphipodCorophium volutator (Pallas) in laboratory for toxicity testing. Chemosphere 59,525–536.
Picone, M., Bergamin, M., Alessandra, A.N., Noventa, S., Delaney, E., Barbanti, A.,Ghirardini, A.V., 2008. Evaluation of Corophium orientale as bioindicator forVenice Lagoon: sensitivity assessment and toxicity-score proposal. Ecotoxicol.Environ. Saf. 70, 174–184.
Prato, E., Biandolino, F., 2006. Monocorophium insidiosum (Crustacea, Amphipoda) asa candidate species in sediment toxicity testing. Bull. Environ. Contam. Toxicol.77, 1–8.
Prato, E., Di Leo, A., Biandolino, F., Cardellicchio, N., 2006. Sediment toxicity testsusing two species of marine amphipods: Gammarus aequicauda and Corophiuminsidiosum. Bull. Environ. Contam. Toxicol. 76, 629–639.
Prato, E., Scardicchio, C., Biandolino, F., 2008. Effects of temperature on the acutetoxicity of cadmium to Corophium insidiosum. Environ. Monit. Assess. 136, 161–166.
Queiroga, H., 1992. Corophium multisetosum (Amphipoda: Corophiidae) in Canal deMira, Portugal: some factors that affect its distribution. Mar. Biol. 104, 397–402.
Quintino, V., Rodrigues, A.M., Ré, A., Pestana, M.P., Silva, S., Castro, H., 2001.Sediment alterations in response to marine outfall operation off Lisbon,
Portugal: a sediment quality Triad study. J. Coast. Res. 34, 535–549 (specialissue).
Quintino, V., Picado, A.M., Rodrigues, A.M., Mendonça, E., Costa, M.H., Costa, M.B.,Lindgaard-Jorgensen, P., Pearson, T.H., 1995. Sediment chemistry–infaunalcommunity structure in a Southern European Estuary related to solid-phaseMicrotox� toxicity testing. Neth. J. Aquat. Ecol. 29 (3–4), 1–10.
Quintino, V., Rodrigues, A.M., Gentil, F., 1989. Assessment of macrozoobenthiccommunities in the lagoon of Óbidos, western coast of Portugal. Sci. Mar. 53,645–654.
Ré, A., Rodrigues, A., Quintino, V., 2007. Acute toxicity testing with the Europeanestuarine amphipod Corophium multisetosum. Hydrobiologia 587, 89–99.
Riba, I., Conradi, M., Forja, J.M., DelValls, T.A., 2004. Sediment quality in theGaudalquivir estuary: lethal effects associated with the Aznalcóllar mining spill.Mar. Pollut. Bull. 48, 144–152.
Scarlett, A., Rowland, S.J., Canty, M., Smith, E.L., Galloway, T.S., 2007. Method forassessing the chronic toxicity of marine and estuarine sediment-associatedcontaminants using the amphipod Corophium volutator. Mar. Environ. Res. 63,457–470.
Silva, S., Ré, A., Pestana, P., Rodrigues, A.M., Quintino, V., 2004. Sediment disturbanceoff the Tagus estuary, western Portugal: chronic contamination, sewage outfalland runoff events. Mar. Pollut. Bull. 49, 154–162.
Sokal, R., Rohlf, F.J., 1995. Biometry, third ed. W.H. Freeman and Company, NewYork. 887pp.
Sosnowski, S.L., Germond, D.J., Gentile, J.H., 1979. The effect of nutrition on theresponse of field populations of the calanoid copepod Acartia tonsa to copper.Water Res. 13, 449–452.
Stock, J.H., 1952. Some notes on the taxonomy, the distribution and the ecology offour species of the genus Corophium (Crustacea, Malacostraca). Beaufortia 21,1–10.
Swartz, R.C., DeBen, W.A., Philips, J.K., Lamberson, J.D., Cole, F.A., 1985.Phoxocephalid amphipod bioassay for marine sediment toxicity. Pp. 284-307.In: Aquatic toxicology and hazard assessment: Proceedings of the SeventhAnnual Symposium. R.D. Cardwell, R. Purdy, R.C. Bahner (Eds.). ASTM STP 854.American Society for Testing and Materials, Philadelphia, PA, USA.
Swartz, R.C., Cole, F.A., Schults, DeBen, W.A., 1986. Ecological changes in theSouthern California Bight near a large sewage outfall: benthic conditions in1980 and 1983. Mar. Ecol. Prog. Ser. 31, 1–13.
Tatem, H.E., Anderson, J.W., Neff, J.M., 1976. Seasonal and laboratory variations inthe health of grass shrimp Palaemonetes pugio: dodecyl sodium sulphatebioassay. Bull. Environ. Contam. Toxicol. 16, 368–375.
U.S.EPA, 1994. Methods for assessing the toxicity of sediment-associatedcontaminants with estuarine and marine amphipods. U.S. EnvironmentProtection Agency, Report EPA 600/R – 94/025. (Narragansett, Rhode Island02882, USA, June 1994).
van den Heuvel-Greve, M., Postma, J., Jol, J., Kooman, H., Dubbeldam, M., Schipper,C., Kater, B., 2007. A chronic bioassay with the estuarine amphipod Corophiumvolutator: test method description and confounding factors. Chemosphere 66,1301–1309.
Woodworth, J.G., King, C., Miskiewicz, A.G., Laginestra, E., Simons, J., 1999.Assessment of the comparative toxicity of sewage effluent from 10 sewagetreatment plants in the area of Sidney, Australia using an amphipod and two seaurchin bioassays. Mar. Pollut. Bull. 39 (1–12), 174–178.
