Development of acute and chronic sediment bioassays with the harpacticoid copepod Quinquelaophonte sp

Development of acute and chronic sediment bioassays with theharpacticoid copepod Quinquelaophonte sp

Tristan J. Stringer a,b, Chris N. Glover a, Vaughan Keesing c, Grant L. Northcott d,Sally Gaw e, Louis A. Tremblay f,g,n

a School of Biological Sciences, University of Canterbury, Private Bag 4800, Christchurch 8140, New Zealandb Landcare Research, PO Box 40, Lincoln 7640, New Zealandc Boffa Miskell Ltd, PO Box 13340, Wellington 6142, New Zealandd Northcott Research Consultants Limited, 20 River Oaks Place, Hamilton 3200, New Zealande Chemistry Department, University of Canterbury, Private Bag 4800, Christchurch 8140, New Zealandf Cawthron Research, Private Bag 2, Nelson 7042, New Zealandg School of Biological Sciences, The University of Auckland, Private Bag 92019, Auckland 1142, New Zealand

a r t i c l e i n f o

Article history:Received 10 July 2013Received in revised form27 September 2013Accepted 2 October 2013Available online 28 October 2013

Keywords:EstuaryPollutionReproductionToxicityNew Zealand

a b s t r a c t

Reliable environmentally realistic bioassay methodologies are increasingly needed to assess the effects ofenvironmental pollution. This study describes two estuarine sediment bioassays, one acute (96 h) andone chronic (14 d), with the New Zealand harpacticoid copepod Quinquelaophonte sp. utilisingbehavioural and reproductive endpoints. Spiked sediments were used to expose Quinquelaophonte sp.to three reference compounds representing important categories of estuarine chemical stressors: zinc(a metal), atrazine (a pesticide), and phenanthrene (a polycyclic aromatic hydrocarbon). Acute-to-chronicratios (ACR) were used to further characterise species responses. Acute sediment (sandy and lowtotal organic content) 96 h EC50 values for the sublethal inhibition of mobility for zinc, atrazine andphenanthrene were 137, 5.4, and 2.6 mg/g, respectively. The chronic EC50 values for inhibition ofreproduction (total offspring) were 54.5, 0.0083, and 0.067 mg/g for zinc, atrazine, and phenanthrene,respectively. For phenanthrene, a potentially novel mode of action was identified on reproduction.Quinquelaophonte sp. was found to be more sensitive than several other estuarine species indicatingchoice of test organism is important to characterising the effects of environmentally relevant levels ofcontamination. The bioassay sediment results demonstrate the sensitivity and suitability of Quinquelao-phonte sp. as a tool for the assessment use of estuarine health.

& 2013 Elsevier Inc. All rights reserved.

1. Introduction

Sediments may contain complex mixtures of chemical con-taminants, each of which may impact in-dwelling fauna in distinctways. For this reason bioassays utilising relevant endpoints, suchas mobility and reproduction, are required for determining sedi-ment toxicity. The major advantage of bioassays is that they assessthe toxicity on the basis of biological parameters rather thanby chemical characterisation and trigger values (as per ANZECC/

ARMCANZ, 2000 recommendations). Although trigger values areoften based on those generated from bioassays, the species used toestablish the values may not be from the particular habitat ofinterest. Additionally, when sediment quality assessments arebased solely on abiotic parameters (i.e. trigger values), they maybe poorly predictive of toxicity (Davoren et al., 2005). In part this isdue to the inability of such approaches to account for any additive,synergistic, or antagonistic interactions of contaminants, whichcan alter the toxicological impact of the sediment. In addition,a chemistry-based approach is limited by the selected contami-nants measured (Davoren et al., 2005). Estuarine sediments aretypically complex, with a large variation in chemical compositionand substrate type even within the same estuary. As these twofactors contribute significantly to toxicity, this makes it even morechallenging to predict potential adverse effects on the ecosystem(Simpson et al., 2011).

The choice of bioassay endpoints is important for assessing thetoxicity of complex chemical mixtures. For example, the use of acute

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/ecoenv

Ecotoxicology and Environmental Safety

0147-6513/$ - see front matter & 2013 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.ecoenv.2013.10.002

n Corresponding author at: Cawthron Research, Private Bag 2, Nelson 7042,New Zealand. Fax: þ64 3 546 9464.

E-mail addresses: [email protected] (T.J. Stringer)[email protected] (C.N. Glover)[email protected] (V. Keesing)[email protected] (G.L. Northcott)[email protected] (S. Gaw), [email protected](L.A. Tremblay).

Ecotoxicology and Environmental Safety 99 (2014) 82–91

lethal endpoints may be inadequate as ecosystems can collapsethrough comparatively subtle sublethal impacts, prior to the deathof individuals due to contaminant exposure. A focus on lethal endpoints may miss food web and community level changes critical tosustaining the local ecosystem. Sublethal endpoints and chronic testsallow for better assessments of toxicity (Brown et al., 2005; Greensteinet al., 2008; Kennedy et al., 2009; Simpson and Spadaro, 2011). Suchendpoints can be based on development (Dahl et al., 2006; Perez-Landa and Simpson, 2011), behaviour (Lotufo, 1997; Silva et al., 2009;Ward et al., 2013), and reproduction (Ingersoll et al., 1998; Bejaranoand Chandler, 2003; Fleeger et al., 2007; Simpson and Spadaro, 2011).Of these, reproductive endpoints are most commonly used inmeiobenthic bioassays, primarily due to the difficulty in observingdevelopment and behaviour in a sediment-dwelling organism. Repro-duction is an important endpoint as it is sensitive and can significantlymodulate population and community (i.e. food web) levels (Bejaranoet al., 2005).

The in situ observation of behavioural patterns can be difficultfor sediment-dwelling organisms, but if it can be tailored to thespecific bioassay species, can provide important insights intoeffects of contaminants (Lotufo, 1997; Ward et al., 2013). Examplesof behavioural endpoints include respiration, locomotion, habitatselection, feeding, and predator avoidance. These endpoints reflecta variety of genetic, neurobiological, physiological, and environ-mental effects (Atchison et al., 1996; Lotufo, 1997; Riddell et al.,2005; Ward et al., 2013). As effects on behavioural patterns arelikely to reflect subtle changes in the biology of an organism,these endpoints provide sensitive estimates of toxicity that mayoccur in natural populations in situ (Ingersoll et al., 1998). Suchinsights might assist in whole ecosystem non-lethal adverse effectassessments.

Before bioassays can be used in toxicity assessment, it isimportant to characterise the responses of the test species todifferent reference toxicants (Chapman et al., 2002). This providesinsight into how the different bioassay endpoints vary withexposure to toxicants with distinct modes of action and bioavail-ability. In addition to characterising bioassay species responsesthis information can also be used in environmental risk manage-ment for the studied chemicals.

There are several species of harpacticoid copepods that arecurrently being used in sediment bioassays around the worldincluding: Amphiascus tenuiremis (Hagopian-Schlekat et al., 2001;Bejarano and Chandler, 2003), Nitocra spinipes (Simpson andSpadaro, 2011; Perez-Landa and Simpson, 2011), Schizopera kna-beni (Fleeger, 2007), Tigriopus brevicornis (Forget et al., 1998), andTisbe battagliai (Hutchinson et al., 1999). The use of local species iscritical in toxicity assessment and the lack of a suitable species forNew Zealand sediments is required to enable the best local speciesprotection possible (ANZECC/ARMCANZ, 2000).

A previous study by Stringer et al. (2012a) examined thesuitability of a variety of harpacticoid copepods as test organisms.This included assessment of favourable bioassay species traits such assensitivity, the ability to be cultured in the lab, high reproductiverates and a short life-cycle. That study identified Quinquelaophontesp. as the most suitable bioassay species for testing New Zealandestuarine and coastal sediments (Stringer et al., 2012a). The objec-tives of the current study were to create two sediment bioassays, oneacute and one chronic, using Quinquelaophonte sp. as a bioindicatorof toxicity. The acute sediment assay examined both sublethal andlethal endpoints, the former of which increased the sensitivity of theassay, helping to enhance the understanding of the potentialmechanisms of toxic effects. The chronic partial life-cycle test wasdeveloped to assess chronic effects of pollutants using mortality andreproductive responses (fecundity and total juvenile production).Three reference contaminants; zinc (a metal), atrazine (a pesticide),and phenanthrene (a polycyclic aromatic hydrocarbon; PAH) were

selected to better understand toxic effects on Quinquelaophonte sp.These contaminants were chosen as they are of environmentalconcern, have diverse chemical structures, different modes of actionand have been used in other studies (Lotofu 1997; Hagopian-Schlekatet al., 2001; Fleeger et al., 2007; Silva et al., 2009), allowing for theresults attained in this study to be compared to literature findings.

2. Materials and methods

2.1. Test species

Quinquelaophonte sp. were cultured in a laboratory system based on therecirculating seawater system designed by Chandler (1986) and modified for usewith Quinquelaophonte sp. (Stringer et al., 2012a). Copepods were mono-cultured ina flow-through plastic aquarium with approximately 2 cm of cleaned sterilisedsediments (size o125 mm), with 5 cm of overlying artificial seawater (ASW;RedSeas sea salts) under a dripping flow of 5 ml/min. The culture system wasmaintained at 20 1C and at 30‰ ASW, with a 12 h:12 h light:dark photoperiod. Thecopepods were fed 40 ml of a concentrated (�2�106 cells/mL) mixed suspendedalgal diet of Dunaliella tertiolecta, Isochrysis galbana, and Chaetoceros muelleri(1:1:1) twice a week (Chandler and Green, 1996).

2.2. Sediment preparation and spiking

Sediments were collected from Akaroa Harbour, Canterbury, New Zealand(43145′29″S 172155′5″E) at low tide by scraping the top 0–2 cm of sediments, andwere processed according to the methods detailed in Chandler and Green (1996).Toxicant concentrations were achieved by spiking sediment with stock solutions ofzinc, atrazine, and phenanthrene in an orbital mixer (Stringer et al., unpublisheddata). Zinc (as ZnSO4 � 7H2O) was dissolved in deionised water. Atrazine andphenanthrene were spiked using acetone (14 d chronic bioassay for both, 96 hacute assay for phenanthrene) or DMSO (96 h acute assay for atrazine) as a carrier.The solubility of atrazine in acetone was inadequate to generate the highersediment concentrations required for the acute tests, hence the use of DMSO as areplacement carrier in these trials. Acetone is the preferable carrier as it is morelikely to volatilise, thus reducing solvent concentrations in the sediment (Northcottand Jones, 2000). Carrier controls were used in all tests where a carrier was used.The carrier/sediment concentration was consistent in all treatments and was below3 ml/g sed. After the addition of the toxicants, sediments were aged for a total of14 d to allow for sediment/pore-water equilibration.

2.3. Acute 96 h toxicity tests

Acute toxicity tests were performed with the three reference chemicals (zinc,atrazine, and phenanthrene). All of the tests used laboratory-cultured copepods,and followed the methods outlined previously (Green et al., 1993) with minormodifications.

For each contaminant, there were five geometrically increasing concentrations,with four replicates for each treatment and control. Three of the replicates wereused for biological enumeration at the termination of the test; the fourth was usedfor water quality and chemical analysis. Each test vessel consisted of a 50 mlborosilicate Erlenmeyer flask with 10 ml clean or spiked control sediments, 20 mlof aerated 30‰ ASW, and 30 adult copepods (15 male, 15 non-gravid female) thatwere loaded at test initiation via Pasteur pipette. The test vessels were looselycovered with aluminium foil to prevent evaporation. The tests were static (no waterexchange) and run over 96 h in an environmental chamber at 2071 1C and a 12:12light:dark cycle. Dissolved oxygen, salinity, and pH were measured at test initiationand at test termination (96 h). Water quality was maintained at salinity 30‰,pH 8.2, dissolved oxygen 46 mg/l, with nitrate, nitrite, and ammonia kept belowdetection limits (2.5, 0.05, and 0.25 mg/L, respectively; RedSeas Marine Lab).

2.4. Chronic 14 d toxicity tests

The chronic partial life-cycle 14 d test was based on a method originallydescribed by Chandler and Green (1996), but modified for Quinquelaophonte sp.In brief, for each of the three reference contaminants, tests included fivegeometrically increasing concentrations based on the results from the acute 96 hexposures described above. An overlap in concentrations was used to compare thedifferences in acute and chronic toxicity. Unspiked sediment was used as a controland carrier controls were implemented when a solvent carrier was used to spikesediments. For each concentration there were four replicates. Test chambersconsisted of a 50 ml borosilicate Erlenmeyer flask with two opposing 1 cmdiameter holes at the base of the neck of the flask which were covered with55 μm nylon mesh. These apertures acted to maintain a constant level of overlyingASW (25 ml). Prior to water addition, 10 ml of sediment that had been aged 14-d

T.J. Stringer et al. / Ecotoxicology and Environmental Safety 99 (2014) 82–91 83

following the addition of contaminant were added to each test chamber viasyringe. After the sediments were added, the test chambers were inoculated with30 copepods (15 male and 15 non-gravid females) via Pasteur pipette from thelaboratory cultures. Test chambers were placed in a purpose built flow-throughASW system, under a dripping flow of approximately 1 ml/min. The tests wereconducted for 14 d at 2072 1C under a 12:12 dark:light photoperiod. Copepodswere fed 1 ml (�2�106 cells) of a concentrated 1:1:1 mixture of D. tertiolecta, I.galbana, and C. muelleri algae twice a week. Water quality parameters such assalinity, dissolved oxygen, nitrate, nitrite, pH and ammonia were monitored at theinitiation of the test, after one week, and at test termination. Water quality wasmaintained at salinity 30‰, pH 8.2, dissolved oxygen 46 mg/l, with nitrate, nitrite,and ammonia kept below detection limits (2.5, 0.05, 0.25 mg/L, respectively;RedSeas Marine Lab).

2.5. Physicochemical sediment characteristics

The sediment was analysed for grain size, total organic content (TOC), redoxpotential, and pH using the same methods as those described by Stringer et al.(2012b). Redox potential of the sediment was determined by using a platinum-tipped Redox probe (Jenway 924 003), and pH was measured using a gel-filled PEIpH combination electrode (Hanna Instruments HI-1230).

2.6. Measuring reference toxicants

Contamination concentrations were measured in pore-water and sediment-bound fractions at test initiation and termination. A subsample (c. 10 ml) ofsediment taken at the beginning of the test, and sediment from one of the fourreplicates at test termination, was frozen at �20 1C until analysis. Sediments werecentrifuged prior to analysis at 2500 RCF for 15 min to separate sediments andpore-waters. Sediment and pore-water concentrations were analysed according tothe brief methods below.

The concentration of zinc in pore-water samples was determined by 0.45 mmfiltering the pore-waters, diluting samples with two percent nitric acid, andanalysing the solutions by Inductively Coupled Plasma Mass Spectrometry (ICP-MS Agilent 7500 cx). Pore-water concentrations of phenanthrene were determinedby extraction with pentane after the addition of anthracene-d10 as a surrogate.Extracts were then concentrated under nitrogen gas and reconstituted in apredetermined volume of toluene. The internal standard phenenthrene-d10 wasthen added, and the final extracts analysed by high resolution gas chromatographywith mass spectrometric detection (GCMS). Atrazine, along with an addedsurrogate (simazine), was extracted from pore-waters with dichloromethane(DCM) before being concentrated to dryness under nitrogen gas. It was thenreconstituted in ethyl acetate and a predetermined quantity of the internalstandard tetrachlorvinphos was added to the reconstituted sample extracts beforebeing analysed by GC with electron capture (ECD) and nitrogen-phosphorousdetection (NPD).

Zinc sediment concentrations were analysed according to the modified USEPA200.8 protocol with ICP-MS (Agilent 7500 cx). QA/QC procedures included theanalysis of a certified reference material (NIST #2702, inorganics in marinesediments). Phenanthrene associated with the sediment phase was extracted usinga modified method based on the US Environmental Protection Agency (USEPA) andthe US National Oceanic and Atmospheric Administration (NOAA) protocols(Holland et al., 1993). Phenanthrene was extracted using a 1:1 (v/v) pentane:acetone extraction solvent. After extraction, approximately 70 ml of Milli-Q waterwas added to the solvent extract to partition the PAHs into the pentane. The upperpentane layer was pipetted into a glass vial and concentrated to dryness undernitrogen gas, before being reconstituted in DCM, and then subjected to furtherpurification by gel permeation chromatography (GPC). The purified sedimentextracts were reconstituted in toluene, and the final extracts were analysed byGCMS. QA/QC procedures included the addition of surrogate (anthracene-d10) andan internal standard (phenenthrene-d10). Sediment-bound atrazine was solvent-extracted after the addition of the simazine surrogate recovery spike usingmethanol before extracts were concentrated to dryness under nitrogen gas andreconstituted with ethyl acetate. Extracts were then purified using GPC beforeresidues of atrazine, its primary metabolites desethyl-atrazine and desisopropyl-atrazine, the surrogate recovery chemical simazine, and internal standard tetra-chlorvinphos were analysed by GC with ECD and NPD.

Test concentrations of contaminants in sediment and pore-water were calcu-lated by averaging the initial and final concentrations to give an average exposedconcentration over the test period. Percent losses were also calculated by dividingthe difference of the initial and final concentrations by the initial concentration,to examine the extent of contaminant losses to pore- and overlaying-water.

2.7. Toxicological analysis

2.7.1. Acute 96 h testsAt test termination, survival was determined by sieving replicates on a 55 mm

mesh sieve. All individuals (alive or dead) were then gently washed into a Petri dishwhere survival was assessed and enumerated, and individuals were sexed.

Copepods were considered dead if there was no response or movement followingbeing gently prodded with a needle. A sublethal endpoint, lethargy (a delayed(410 s) movement away from the stimulus) or lack of mobility (response to thestimulus, but lack of movement) was also measured. Copepods that immediately(o10 s) moved away from stimulus were considered unimpacted by exposure.

2.7.2. Chronic 14 d testsCopepods were enumerated, sexed, and a determination of mortality was

conducted as described above for the acute test. Gravid females were isolated andpreserved in five percent formaldehyde for later fecundity assessment. After adultindividuals were isolated the remaining sediment and juvenile copepods werepreserved in five percent formaldehyde and then stained with Rose Bengal tofacilitate counting of nauplii and copepodites. Numbers of nauplii and eggs perfemale were then used to calculate the reproductive endpoints, defined below:

1. Total offspring¼naupliiþcopepodites.2. Potential offspring per female¼(eggsþnaupliiþcopepodites)/surviving

females.3. Realised offspring per female¼ (naupliiþcopepodites)/surviving females.

2.7.3. Statistical analysis and toxicological metricsMeasured concentrations of the contaminants in both sediment and pore-

water were used to generate dose–response curves and estimate LCx and ECx valuesfor acute and chronic endpoints. The proportion of individuals surviving (“unim-pacted”; i.e. not lethargic) or total juveniles produced (as percent of control) wasregressed against log contaminant concentration using binomial regression. LCx

and ECx (x¼50, 20, 10) values along with 95 percent confidence intervals werecalculated using the FIELLER procedure in the GenStat Statistical Package (12thedition). Means and standard deviations were calculated for the reproductiveendpoints using GraphPad Prism (Version 5) and statistical significance was testedusing one-way ANOVA with a Dunnett's post-hoc analysis to assess differencesfrom controls. From the results of these analyses the following metrics werederived. The no observed effect concentration (NOEC) was the highest testconcentration that did not produce a response that was statistically significantlydifferent from that of the control value. The lowest observed effect concentration(LOEC) was the lowest concentration that showed a statistically significantdifference to the controls. The maximum acceptable threshold concentration(MATC) was then calculated as the geometric mean of the LOEC and NOEC(Mebane et al., 2008). The LOEC, NOEC, and MATC were used in this study forcalculation of acute-to-chronic ratios, to enable comparison to previous research.

Acute-to-chronic ratios (ACRs) were calculated using a variety of methods.In Europe, Australia and New Zealand it is common to calculate ACRs based on theacute LC50 divided by the chronic NOEC (Lange et al., 1998). In the United States,ACRs are commonly calculated as the LC50/MATC. However, recently there has beena shift to acute LC50/chronic EC20 (Mebane et al., 2008). In other parts of the world,acute LC50/chronic EC10 is increasingly used as the NOEC and LOEC are becominglargely obsolete measures (Harrass, 1996). All of these endpoints were calculated inthe current study when possible, along with the acute EC50/chronic EC10 ratio, as afurther ACR metric.

3. Results

3.1. Physicochemical sediment characteristics

Sediments were found to have a total organic content of 1.0370.10 percent, consistent with fine sandy sediment types. Themajority of the sediment was in the 32–62 mm sediment fraction(43.870.60 percent), with the 63–74 mm sediment fraction beingthe second largest (22.870.72 percent). The pH and redoxpotential did not change after spiking and averaged 8.0770.24and 14075.3 mV, respectively.

3.2. Chemical analysis

Measured concentrations in sediment and pore-water for allthree contaminants are listed in Table 1. The mean recovery of zincfrom the certified reference material was 7973 percent, spikedsample recovery was an average of 8574 percent, and the relativepercent difference of the duplicate samples was 1.571.3 percent.For phenanthrene the mean recovery of the surrogate anthracene-d10 spiked into pore-water samples was 9770.05 percent and therecovery of anthracene-d10 spiked into sediments was 8974

T.J. Stringer et al. / Ecotoxicology and Environmental Safety 99 (2014) 82–9184

percent. The recovery of phenanthrene spiked into the controltreatments was 9176 percent. For the atrazine analysis the meanrecovery of simazine spiked into pore-water samples was 8876percent and the recovery of simazine spiked into sediments was88711 percent. The control treatments that were spiked withdesethyl-atrazine and desisopropyl-atrazine, simazine and atrazinehad a mean recovery of 8575 percent, 8576 percent, 9175percent, and 10077 percent, respectively. The two primary meta-bolites of atrazine, desethyl-atrazine and desisopropyl-atrazinedesethyl, were not detected (o0.001 mg/l) in the analysed pore-water or sediment samples.

Measured pore-water concentrations of all three referencecontaminants were proportional to spiked levels. Atrazine showedthe highest pore-water concentrations relative to sediment burdenin both acute and chronic assays, followed by zinc and phenan-threne (Table 2). The losses from the sediment over the testingperiods were low (less than ten percent) in the zinc-spikedsediments, but very high (up to 90 percent in chronic tests) inthe atrazine-spiked sediments. Phenanthrene showed moderatelosses over the test period of 4.8–29 percent in the acute tests and29–65 percent in the chronic tests.

3.3. Acute 96 h tests

The calculated LCx and ECx values, and corresponding 95percent confidence limits, for all three reference contaminantsare given in Table 2. Copepods in acute tests were assessed forinhibition of mobility (the sublethal endpoint, described by theECx data). Individuals displaying inhibited mobility were often

incapacitated by the contaminant and would spasm and convulseafter the stimulus. This effect was most prominent in thephenanthrene-exposed individuals. For zinc the sediment LC50(95 percent confidence interval) value was 196 (184–209) mg/g sed.dry wt., with an EC50 of 137 (131–142) mg/g sed. dry wt. (Table 2).Atrazine had an LC50 of 12.7 (15.1–47.0) mg/g sed. dry wt., and anEC50 of 5.35 (4.24–6.72) mg/g sed. dry wt. (Table 2). Sedimentspiked with phenanthrene had a plateau in mortality at about 40percent for concentrations greater than 12.3 mg/g sed. dry wt.(Table 2). This corresponded with saturation in the pore-waterconcentration (see Table 1). This combination of mortality plateauand solubility limit of phenanthrene (1 mg/l) meant that even ifhigher concentrations of phenanthrene were used it is unlikelythat 100 percent mortality would have been achieved. Conse-quently no phenanthrene treatment generated a sufficiently highmortality rate to facilitate the calculation of LCx values (ASTM,2007). The EC50 was still able to be calculated, yielding a value of2.56 (2.15–3.18) mg/g sed. dry wt. The inhibited mobility endpointshowed a typical dose–response with effects on sensitive indivi-duals seen at low concentrations (2.12 mg/g) and then completespopulation immobility at 12.3 mg/g and above.

LC50 and EC50 values based on pore-water contaminationconcentrations were calculated (Table 3) to provide insight intothe importance of the aqueous exposure. This also facilitatedcomparison to aquatic LCx values for these reference contaminantspreviously calculated in Stringer et al. (2012a). For zinc the pore-water LC50 was 2.05 (1.66–2.47) mg/l, with an EC50 for lethargy of0.85 (0.77–0.94) mg/l. For atrazine the calculated LC50 was 9.11(8.44–9.84) mg/l and the EC50 was 4.10 (3.6–4.61) mg/l for pore-water exposure. Phenanthrene had a LC50 of 0.42 (0.29–0.81) mg/land an EC50 of 0.046 (0.036–0.064) mg/l for pore-water exposure.

3.4. Chronic 14 d tests

Chronic toxicity to the measured sublethal endpoints of totaloffspring, potential offspring per female and realised offspring perfemale for the three reference contaminants is shown in Fig. 1.The calculated ECx and LCx values derived from the chronic assaysare reported in Table 2. Mortality was minimal over the tests, onlyoccurring at the highest tested concentrations of 161, 0.77, and1.85 mg/g sed. dry wt. for zinc, atrazine, and phenanthrene,respectively. Zinc significantly reduced the total number of off-spring at concentrations of 36 mg/g sed. dry wt. and higher(Po0.0001) and reduced potential and realised offspring perfemale at concentrations over 55 mg/g sed. dry wt. (P¼0.0033,Po0.0001, respectively; Fig. 1A–C). Atrazine caused a significantreduction in total offspring at sediment concentrations of0.003 mg/g sed. dry wt. and above (Po0.0001, Fig. 1). There wasalso a significant reduction in realised offspring per female atatrazine concentrations of 0.003 mg/g sed. dry wt. and above(P¼0.015), and although there was a trend towards reducedpotential offspring per female, this effect was not statisticallysignificant due to high inter-replicate variability (P¼0.063 at0.49 mg/g sed. dry wt., Fig. 1E). Phenanthrene caused a significantreduction in total offspring at sediment concentrations of0.037 mg/g sed. dry wt. and above (Po0.0001, Fig. 1G), withsignificant reduction in potential and realised offspring at0.26 mg/g sed. dry wt. (P¼0.0015, Po0.0001, respectively).

The LC50 values for the chronic sediment exposures were onlyable to be calculated in the zinc spiked sediment, due to the lowmortality in the atrazine and phenanthrene exposures. Zinc had achronic LC50 of 128 (126–130) mg/g sed. dry wt.

Chronic endpoints for the three reference contaminants basedon their pore-water concentrations are given in Table 3. Zinc hadan EC50 for total offspring of 0.11 mg/l and a LC50 of 0.62 mg/l.Atrazine had an EC50 for total offspring of 0.10 mg/l. LC50 values

Table 1Measured sediment and pore-water concentrations for acute and chronic tests.Values are an average of the test initiation and termination concentrations.Sediment values are given in mg/g dry wt., and pore-water in mg/l. Percent lossesover the test are also reported for the sediment concentrations of contaminants.Limits of detection (LOD) were 0.01 mg/l for zinc and 0.001 for atrazine andphenanthrene.

Pollutant Testtype

Averageconcentration(lg/g sed.)

Pore-waterconcentration(mg/l)

Percent lost insediments overtest period

Zinc Acute 88.6 0.30 6.7122 0.62 3.4172 1.54 2.6255 4.85 3.3384 19.3 9.8

Chronic 26.6 0.04 0.7236.8 0.05 1.955.2 0.11 0.4594.8 0.27 0.92

162 1.16 9.1

Atrazine Acute 0.122 0.40 630.35 1.25 603.05 2.81 69

21.2 11.5 55175 18.1 42

Chronic 0.00005 0.002 4900.0005 0.01 4900.0033 0.041 800.053 0.11 850.49 0.40 75

Phenanthrene Acute 0.28 0.008 142.12 0.043 11

12.3 0.19 4.825.1 0.3 8.071.3 0.14 29

Chronic 0.0025 oLOD 330.012 oLOD 290.037 0.001 650.26 0.009 571.85 0.038 58

T.J. Stringer et al. / Ecotoxicology and Environmental Safety 99 (2014) 82–91 85

were unable to be calculated for atrazine due to insufficientmortality, and chronic pore-water endpoints were not determinedfor phenanthrene due to concentrations being below detectionlimits.

Of the five endpoints calculated for the three reference con-taminants, total offspring was the most sensitive overall, followedby realised offspring per female, potential offspring per female,EC50 of lethargy, and finally the LC50 value. Potential and realisedoffspring were the most variable endpoints.

3.5. Acute to chronic ratio

The different ACR metrics are listed in Table 4 and producedvalues that ranged from 1.53 to 7.37 for zinc and 1660 to 117,000for atrazine. Due to the inability to calculate an acute LC50 forphenanthrene, only the EC50/EC10 ACR (731) could be determinedfor phenanthrene.

4. Discussion

4.1. Zinc toxicity

Zinc is a common toxicant in estuaries, owing largely to its highconcentration in storm water runoff (Lotze et al., 2006). In thisstudy zinc was found to have an acute LC50 of 196 mg/g and an EC50

for inhibited mobility of 137 mg/g. This characterises Quinquelao-phonte sp. as more acutely sensitive to sediment Zn than severalother common benthic bioassay species. For example, the harpac-ticoid copepod A. tenuiremis exhibits an LC50 of 671 mg/g zinc(Hagopian-Schlekat et al., 2001), and the amphipods Melita plu-mulosa (King et al., 2006) and Hyalella azteca (Borgmann andNorwood, 1997) display LC50 values of 1790 and 3530 mg/g zinc,respectively (Table 5). Only one studied species showed highersensitivity to zinc than Quinquelaophonte sp. and that was theamphipod Corophium volutator with an LC50 of 32 mg/g (Bat andRaffaelli, 1998). Differences in metabolism, uptake, elimination

Table 3Comparison of pore-water test endpoints calculated for the acute and chronic bioassays. Bracketed values represent 95 percent confidence intervals, or P values (for LOECanalyses). All values are given in mg/l. A (�) signifies an incalculable endpoint.

Test endpoint Zinc (mg/l) Atrazine (mg/l) Phenanthrene (mg/l)

Acute bioassayLC10 0.89 (0.59–1.17) 0.87 (0.73–1.02) 0.019 (0.007–0.032)LC20 1.26 (0.93–1.59) 2.08 (1.84–2.31) 0.059 (0.035–0.083)LC50 2.05 (1.66–2.74) 9.11 (8.44–9.84) 0.42 (0.29–0.81)Inhibited mobility EC10 0.45 (0.43–0.47) 0.14 (0.099–0.19) 0.02 (0.010–0.034)Inhibited mobility EC20 0.57 (0.55–0.59) 0.48 (0.38–0.60) 0.032 (0.018–0.040)Inhibited mobility EC50 0.85 (0.77–0.94) 4.10 (3.6–4.61) 0.046 (0.036–0.064)

Chronic bioassayTotal offspring LOEC 0.05 (Po0.0001) 0.041 (Po0.0001) –

Total offspring EC10 0.037 (0.026–0.047) 0.0038 (0.0023–0.0056) –

Total offspring EC20 0.054 (0.042–0.066) 0.013 (0.0091–0.017) –

Total offspring EC50 0.11 (0.091–0.13) 0.10 (0.092–0.12) –

LC10 0.30 (0.24–0.36) – –

LC20 0.39 (0.33–0.45) – –

LC50 0.62 (0.54–0.69) – –

Table 2Comparison of test endpoints calculated for the acute and chronic sediment bioassays. Bracketed values represent 95 percent confidence intervals, or P values (for LOECanalyses). All values are given in mg/g dry wt. A (�) signifies an incalculable endpoint.

Test endpoint Zinc (lg/g) Atrazine (lg/g) Phenanthrene (lg/g)

Acute bioassayLC10 140 (116–157) 0.29 (0.057–0.54) –

LC20 159 (138–175) 1.27 (0.51–2.45) –

LC50 196 (184–209) 25.7 (15.1–47.0) –

Inhibited mobility EC10 106 (100–111) 0.0074 (0.0039–0.013) 1.20 (0.74–1.52)Inhibited mobility EC20 117 (112–121) 0.084 (0.054–0.13) 1.59 (1.15–1.91)Inhibited mobility EC50 137 (131–142) 5.35 (4.24–6.72) 2.56 (2.15–3.18)

Chronic bioassayTotal offspring LOEC 36.8 (Po0.0001) 0.003 (Po0.0001) 0.037 (Po0.0001)Total offspring EC10 29.3 (21.9–35.2) 0.00022 (9.3E�5–0.00046) 0.0035 (0.0006–0.0088)Total offspring EC20 36.9 (29.6–42.7) 0.0020 (0.0011–0.0033) 0.010 (0.003–0.022)Total offspring EC50 54.5 (47.7–61.5) 0.083 (0.061–0.11) 0.067 (0.034–0.13)Potential offspringa LOEC 55.2 (P¼0.0033) – 0.26 (P¼0.0015)Potentiaa offspringa EC10 28.8 (15.1–39.6) 0.0049 (0.0021–0.009) 0.0048 (0.001–0.012)Potential offspringa EC20 38.1 (23.3–46.4) 0.018 (0.010–0.029) 0.016 (0.005–0.033)Potential offspringa EC50 61.7 (47.1–74.8) 0.17 (0.14–0.21) 0.13 (0.073–0.24)Realised offspringaLOEC 55.2 (Po0.0001) 0.003 (Po0.0105) 0.037 (Po0.0001)Realised offspringa EC10 22.4 (14.9–28.5) 0.00096 (0.00038–0.0020) 0.0056 (0.0012–0.013)Realised offspringa EC20 30.8 ( 22.9–37.2) 0.0055 (0.0028–0.0092) 0.014 ( (0.0046–0.027)Realised offspringa EC50 53.0 (45.1–62.1) 0.11 (0.080–0.14) 0.066 (0.035–0.12)LC10 97.7 (95.0–100) – –

LC20 108 (105–110) – –

LC50 128 (126–130) – –

a Per female.

T.J. Stringer et al. / Ecotoxicology and Environmental Safety 99 (2014) 82–9186

(Lotufo et al., 2000; Mc Clellan-Green et al., 2007), and exposurepathway (i.e. dietary, pore-water, or sediment contact; Simpsonand King, 2005) of each of the different species may impactsensitivity. In this study, zinc had the lowest ACR, 6.69, relativeto the other two reference contaminants. ACRs reported for zincare quite variable, ranging from 1 for the trout, Oncorhynchusmykiss (Mebane et al., 2008), 2.9 for the cladoceran Moinamacrocopa (Wong, 1992, 1993) to 19.5 for a mysid shrimp(Gentile et al., 1982).

When the aquatic LC50 values from Stringer et al. (2012a) arecompared to the pore-water zinc LC50 of the present study, thepore-water value is higher (2.1 mg/l versus 0.9 mg/l). In sandysediments it is believed that pore-water is the predominantexposure route for copepods (Strom et al., 2011), and thereforethe sediment-exposure LC50 based on pore-water concentrations

should theoretically be similar to that of the aquatic only expo-sures. One explanation for the difference in the toxicity isdissolved organic matter. Metals can bind to dissolved organicmatter, which can reduce their bioavailability (Santore et al., 2002;Kramer et al., 2004; Di Toro et al., 2005). Sediment pore-watershave higher levels of dissolved organic matter compared to theoverlying waters (Weston et al., 2006), thus making the bioavail-able fraction of zinc much lower, and raising the effect levels. It isalso possible that “in the wild” copepods minimise their exposureto pore-water by staying near the sediment/water interface, wheredilution of pore-water by the overlying water would occur, and/orby moving out of the sediment altogether (Mc Murtry, 1984;Lotufo, 1997).

The inhibition of reproductive output in the chronic bioassayswas found at very low concentrations, and may have resulted

Fig. 1. Reproductive endpoints of 14 d chronic exposures to zinc: (A) total offspring, (B) potential offspring per female, and (C) realised offspring; Atrazine: (D) Totaloffspring, (E) potential offspring per female, and (F) realised offspring; Phenanthrene: (G) total offspring, (H) potential offspring per female, and (I) realised offspring arecalculated as percent of control offspring. One-way ANOVA with a Dunnett's post hoc test was used to determine significant differences from controls (n¼3). All values aregiven in mg/g dry wt. *P40.05, **P40.01, ***P40.001. ‘Acetone’ refers to the carrier controls.

T.J. Stringer et al. / Ecotoxicology and Environmental Safety 99 (2014) 82–91 87

through several distinct mechanisms. There is some evidence thatzinc can accumulate in reproductive tissues (De Schamphelaereet al., 2004), and disrupt vitellogenesis (Hook and Fisher, 2002).Together these reproductive function effects would directly impairreproduction. Additionally, reproductive effects can be generatedby more indirect causes. These may include inhibition of feedingand digestion (De Coen and Janssen, 1998), or by interference withmoulting through chitinase inhibition (Poynton et al., 2007).In fact, moult inhibition has been correlated with reduced reproduc-tion in Daphnia (Poynton et al., 2007). A further possibility is that theeffect on reproduction represents the high mortality of early lifestages, as these are more sensitive than adult copepods (Brown et al.,2005). However, in the current study the low ACR suggests that thereis likely one major mode of action of zinc responsible for theobserved toxicity in the acute and chronic endpoints.

Sediment type determines contaminant bioavailability (Simpsonet al., 2011; Strom et al., 2011). As the sediment organic carbonproportion increases the sediment will bind higher levels of metals(Mahony et al., 1996; Chapman et al., 1998; Machado et al., 2008).This means that the pore-water concentration of metals will belower in sediments with higher organic content than those ofsediments with lower organic content (Simpson et al., 2004; DiToro et al., 2005). This study used very fine sandy sediments with alow organic content of one percent, which have resulted in higherpore-water concentrations of zinc.

The concentrations at which toxic effects of zinc were noted areof concern as they are within the levels of contamination found inNew Zealand estuaries. For example, zinc has been found atconcentrations up to 435 mg/g in Porirua Harbour (Glasby et al.,1990), a level that is ten times higher than the EC10 for reducedoffspring in the chronic bioassay and over twice that of the acuteLC50 of 196 mg/g. It is to be noted that these toxic levels of zinc arebelow the ANZECC/ARMCANZ (2000) interim sediment qualityguideline low value (ISQG-Low) of 210 mg/g. At contamination

levels such as these there is a potential for diminished copepodpopulations, which in turn could alter the structure of estuarinecommunities such as juvenile fish that rely on copepods as afood source (Hicks and Coull, 1983; Le Blanc, 2007). There wererelatively high levels of zinc in the pore-waters in the currentstudy, and thus the comparison to sediment quality guidelinesshould be used with caution. This does further reinforce thenecessity of bioassays over trigger values to assess potentialsediment toxicity.

4.2. Atrazine toxicity

Quinquelaophonte sp. showed a large effect range in response toatrazine exposure, with an acute LC50 of 25.7 mg/g, an inhibitedmobility EC50 of 5.35 mg/g and an EC50 for total offspring of0.083 mg/g. Atrazine showed the highest ACR in this study, witha value of 117,000 recorded for the LC50/EC10 metric. Atrazine haspreviously been shown to be an endocrine disrupting chemicalthrough aromatase modulation (Hayes et al., 2002, Hayes et al.,2003). Studies have seen high ACRs in compounds with specificmodes of action (Roex et al., 2000), and our result is consistentwith this finding.

In another copepod species, A. tenuiremis, inhibition of repro-duction at atrazine concentrations where there was no effect ondevelopment or survival has been described (Bejarano andChandler, 2003). However, the mechanism by which atrazineaffects reproductive functions in invertebrates is unknown. It ispossible that the effect is indirect, acting via impairment ofbehaviour and feeding. This has been observed in mussels, whereatrazine was shown to cause starvation (Tuffnail et al., 2009),and this could explain the reduced reproduction in Quinquelaophontesp. observed in this study. Further research into the effects of atrazineon invertebrate physiology is needed.

Atrazine showed high pore-water concentrations due to sig-nificant desorption from sediments. This strong dissociation fromsediments into pore- and overlying water has been noted pre-viously. For example, other studies have shown that up to 40percent of sediment-bound atrazine can diffuse into the watercolumn (Mersie et al., 1998, 2000). In the chronic exposuresbetween 75 percent and 90 percent of spiked atrazine was lostfrom sediments and resulted in pore-water concentrations as highas 0.4 mg/l. The high pore-water concentrations suggest that thisis the main exposure route for atrazine in sediment exposures.When the acute 96-h pore-water LC50 of 9.1 mg/l is compared to20.8 mg/l reported in Stringer et al. (2012a) for aquatic onlyexposures, it appears that there is a potential for sediments tobe playing a factor in toxicity. However, the two fold difference is

Table 5Comparison of acute and chronic bioassay responses of Quinquelaophonte sp. to zinc, phenanthrene, and atrazine with selected other estuarine species. LC50 and EC50 valuespresented from present to allow for the easiest comparison between studies. All values are based on sediment contaminant concentrations given in mg/g dry wt.

Group Species Test type Zinc lg/g Atrazine lg/g Phenanthrene lg/g Reference

Copepod (Harpaticoid) Schizopera knabeni 96 h LC50 440 Fleeger (2007)10 d Reduced reproduction 22 Lotofu and Fleeger (1997)48 h decreased feeding rate 109 Silva et al. (2009)

Amphiascus tenuiremis 96 h LC50 671 Hagopian-Schlekat et al. (2001)Quinquelaophonte sp. 96 h LC50 196 25.7 – Present study

96 h EC50 –lethargy 137 5.4 2.56 Present study14 d EC50 - total offspring 54.5 0.083 0.067 Present study14 d LC50 128 – – Present study

Amphipod Melita plumulosa 10 d LC50 1790 King et al. (2006)Corophium volutator 48 h LC50 4250 Hellou et al. (2008)

10 d LC50 32 Bat and Raffaelli (1998)Hyalella azteca 10 d LC25 3530 Borgmann and Norwood (1997)

Midge Chironomus tentans 10 d LC50 4 9 Douglas et al. (1993)Bivalve Mercenaria mercenaria 10 d LC50 420 Lawton et al. (2006)

Table 4Acute to chronic ratios (ACRs) resulting from the most sensitive chronic endpoint,total juveniles produced. ACRs were calculated by dividing the 96 h LC50 or EC50 bya variety of calculated parameters from the chronic tests.

Chemicaltested

LC50/NOEC

LC50/MATC

LC50/EC10

EC50/EC10

LC50/EC20

LC50/LC50

Zinc 7.37 6.20 6.69 4.68 5.31 1.53Atrazine 25,700 1660 117,000 23,400 12,800 –

Phenanthrenea – – – 731 – –

a Due to the inability to calculate LC50 values for phenanthrene, only the EC50

was able to be used.

T.J. Stringer et al. / Ecotoxicology and Environmental Safety 99 (2014) 82–9188

likely a result of the high flux of atrazine into the pore-waters atthe start of the test, evident by the greater than 60 percent loss ofatrazine from sediments over the test period.

Inhibition of reproduction can have serious consequences forpopulations, especially at low, environmentally realistic contami-nant concentrations. A study by Bejarano et al. (2005) showedsignificant effects on mesocosm assemblages at atrazine concen-trations at the USEPA seawater criterion (26 mg/l). These couldeventually translate to changes in overall community structure.Atrazine had reproductive effects at very low concentrations in thecurrent study (0.003 mg/g sediment and 0.04 mg/l pore-waterconcentrations), which creates concern that the adverse impactsof atrazine may be seen in wild populations. Atrazine has beenfound at concentrations as high as 49 mg/g in soils in the US(Douglas et al., 1993). In New Zealand there is little information onlevels of atrazine in rivers or estuaries but it is commonly found ingroundwater at levels as high as 37 mg/l (Close and Rosen, 2001;Close, 1993).

There is limited data on sediment exposures of marine inverte-brates to atrazine. This is a large knowledge gap, and is particularlyimportant in countries where atrazine is used extensively, andatrazine and its metabolites are likely to persist in the environ-ment (Rice et al., 2004). There is also a lack of knowledgeregarding the potential effects of sediment-bound atrazine(Douglas et al., 1993; Lawton et al., 2006). However, given theresults of this study, sediment-bound atrazine desorbs veryquickly, at least in sandy sediments, which implies that concen-trations of sediment-bound atrazine are likely to be low withinestuaries. In contrast to the relatively low probability of toxicimpact associated with sediment-bound atrazine, the reproductiveeffects seen at low pore-water concentrations (EC10 total offspring3.8 mg/l) are likely to be of greater concern. This is particularly trueconsidering that in New Zealand concentrations of atrazine anorder of magnitude higher have been reported (Close, 1993).

4.3. Phenanthrene toxicity

Similar to the pattern observed for atrazine, phenanthreneshowed a large difference between levels at which effects onreproduction and acute toxicity were noted. The predominantmode of action of phenanthrene is through non-polar narcosis(Ren, 2002). This was evident in the present study by inhibitedmobility. Non-polar narcosis is a general mode of action shared byphenanthrene with many other organic compounds includingchlorobenzenes, alcohols, esters, ethers, and chlorinated alkanes(Roex et al., 2000). The lethargic effect was very pronounced in theacute test with an EC50 of 2.56 mg/g recorded. While theseindividuals are not dead, their incapacitation and failure torespond to stimuli results in them being ‘environmentally dead’.In other words, due to their inability to move they are likely to bepreyed upon, and if they avoid this fate, they are unlikely torecover (Brooks et al., 2009; Trekels et al., 2011).

An acute sediment LC50 was unable to be calculated forphenanthrene due to the plateau in mortality at concentrationsabove 12.3 mg/g. This can be explained by saturation of the pore-water at around 0.15–0.3 mg/l. Further increases in the sedimentconcentration of phenanthrene led to no further increases in pore-water levels, corresponding with no increase in toxicity. Thissuggests that pore-water concentrations of phenanthrene are thepredominant exposure mechanism for Quinquelaophonte sp. Addi-tionally the acute pore-water LC50 of 0.42 mg/l is very close to theaquatic LC50 in Stringer et al. (2012a) of 0.75 mg/l and the twostudies have similar mortality responses, further giving evidenceand is consistent with other literature showing that the pore-water is the main exposure pathway for hydrophobic compounds

to benthic species (DiToro et al., 1991; USEPA, 1993b) in this shortterm acute bioassays.

The EC10 of total offspring (0.0035 mg/g) was three orders ofmagnitude lower than the acute immobility EC10. This suggeststhat phenanthrene may have a second mode of action, other thannon-polar narcosis, acting specifically on reproductive tissues.Non-polar narcotic compounds often have low ACRs due to theirnon-specific action, whereas chemicals with more defined modesof action (i.e. pesticides and metals) have higher ACRs (Roex et al.,2000; Paumen et al., 2008; Marinkovic et al., 2011). In this study,phenanthrene showed a high ACR (immobility EC50/total offspringEC10) of 731. This ACR is in contrast to previous studies thatexamined phenanthrene and similar non-specific narcotic com-pounds (e.g. Marinkovic et al., 2011).

Phenanthrene may have a direct effect of reproductive func-tions. Lotufo (1998a, 1998b) suggested that sediment exposure offluoranthene (a low molecular weight PAH, similar to phenan-threne) caused reduced reproduction in the copepods S. knabeniand Coullana sp. by a mode of action other than general narcosis.Lotufo (1998b) attributed this to the accumulation of the PAH inthe lipid-rich tissues, especially eggs, which in turn inhibitedhatching success and development of juveniles. The passage ofcontaminant to the egg potentially assists with the detoxificationof phenanthrene in females by reducing their body burdens(Lotufo, 1998b). However, it could affect population growth as itresults in reduced offspring.

The levels at which phenanthrene caused reproductive toxicity(reduction in total offspring EC50 of 0.067 mg/g) were lower thanseveral sediment quality guidelines, including the Effects Range-Median concentration (ERM) of 1.5 mg/g by Long et al. (1998), theProbable Effects Level (PEL) of 0.5 mg/g by MacDonald et al. (1996),and the ANZECC/ARMCANZ (2000) ISQG-low of 0.24 mg/g forsediment normalised to one percent organic carbon. This furtherreinforces the idea that a trigger value is only indicative and notalways a suitable replacement for a bioassay (Davoren et al., 2005).This is especially relevant with PAHs as they are always present inmixtures (Latimer and Zheng, 2003), and can exhibit/ellicit addi-tive toxicity which is not predicted by trigger values (Latimer andZheng, 2003).

While environmental concentrations of total PAH over10,000 mg/g have been recorded (Huntley et al., 1995), in NewZealand values up to 3.7 mg/g have been reported in estuarinesediments (Holland et al., 1993). These local levels of phenan-threne are close to those found to cause mortality in this study(�2 mg/g in the acute and chronic exposures).

One factor not considered in regarding the risk of contaminantlevels is possible behavioural responses that might modify theexposure of organisms. Studies have shown that copepods exhibitavoidance behaviour in response to PAH-contaminated sediments,and actively leave these sediments in search of less contaminatedones (Carman and Todaro, 1996; Lotufo, 1997). While this may bebeneficial in terms of limiting the exposure to the toxicant, it canhave additional negative implications for copepod populations.For example, copepods preferentially inhabiting the water columnface the risk of increased predation by fish (Lotufo, 1997; Brookset al., 2009).

From a New Zealand perspective these harpacticoid copepodbioassays are valuable as they incorporate the use of an indigenousspecies. The incorporation of indigenous bioassays into environ-mental monitoring will provide the best local protection, and willsubsequently reduce the reliance on trigger values derived fromoverseas species. Earlier work (Stringer et al., 2012a) suggestsQuinquelaophonte sp may be a species representative of estuarinehabitats around New Zealand. Future research should focus onfurther developing complementary local bioassay species to beable to create a suite of bioassays to best assess contamination

T.J. Stringer et al. / Ecotoxicology and Environmental Safety 99 (2014) 82–91 89

across taxa (Greenstein et al., 2008). This allows for multiple linesof evidence to assess pollution and also to elucidate causes oftoxicity through patterns in the bioassay responses (USEPA, 1993a;Burton et al., 2002; Chapman and Anderson, 2005). Studies arealso needed to integrate results from specific organismal assaysinto a wider environmental context. For example, determininghow changes in copepod behaviour, reproduction and mortalitywould impact food webs, and in turn what this may mean for theecosystem as a whole, is an important next step.

5. Conclusions

This study has validated two sediment bioassays with multiplelethal and sublethal endpoints that can be applied in a wide rangeof sediment monitoring and risk assessment frameworks. Underthe ANZECC/ARMCANZ (2000) guidelines local species are to beused wherever possible, and this is an element that has beenlacking in many assessment frameworks.

The development of these two bioassays also allows for flex-ibility in the application of a monitoring scheme. Sublethal end-points were valuable in detecting toxic effects at environmentallyrelevant levels of contaminants. The acute test can be utilised as arapid high-throughput assessment of potentially high concentra-tions of pollution. The chronic bioassay is much more sensitive andtherefore would be better utilised in situations where the max-imum environmental protection is required (i.e. near marinereserves or particularly sensitive environments).

References

ANZECC/ARMCANZ, 2000. Australian and New Zealand Guidelines for Fresh andMarine Waters. Australia and New Zealand Environment and ConservationCouncil, Melbourne.

ASTM, 2007. Standard Guide for Behavioral Testing in Aquatic Toxicology. MethodE1604-94. Annual Book of ASTM Standards, American Society for Testing andMaterials, West Conshohocken, Philadelphia, PA, p. 22.

Atchison, G.L., Sandheinrich, M.B., Bryan, M.D., 1996. Effects of environmentalstressors on interspecific interactions in aquatic animals. In: Newman, M.C.,Jagoe, C.H. (Eds.), Ecotoxicology: A Hierarchial Treatment. Lewis Publishers,Boca Raton, Florida, pp. 319–337.

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.

Bejarano, A.C., Chandler, G.T., 2003. Reproductive and developmental effects ofatrazine on the estuarine meiobenthic copepod Amphiascus tenuiremis. Environ.Toxicol. Chem. 22, 3009–3016.

Bejarano, A.C., Pennington, P.L., DeLorenzo, M.E., Chandler, G.T., 2005. Atrazineeffects on meiobenthic assemblages of a modular estuarine mesocosm. Mar.Pollut. Bull. 50, 1398–1404.

Borgmann, U., Norwood, W.P., 1997. Toxicity and accumulation of zinc and copperin Hyalella azteca exposed to metal-spiked sediments. Can. J. Fish. Aquat. Sci. 54,1046–1054.

Brooks, A,C, Gaskell, P.N., Maltby, L.L., 2009. Sublethal effects and predator-preyinteractions: implications for ecological risk assessment. Environ. Toxicol.Chem. 28, 2449–2457.

Brown, R.J., Rundle, S.D., Hutchinson, T.H., Williams, T.D., Jones, M.B., 2005.A microplate freshwater copepod bioassay for evaluating acute and chroniceffects of chemicals. Environ. Toxicol. Chem. 24, 1528–1531.

Burton, G.A., Chapman, P.M., Smith, E.P., 2002. Weight-of-evidence approaches forassessing ecosystem impairment. Hum. Ecol. Risk Assess. 8, 1657–1673.

Carman, K.R., Todaro, M.A., 1996. Influence of polycyclic aromatic hydrocarbons onthe meiobenthic-copepod community of a Louisiana salt marsh. J. Exp. Mar.Biol. Ecol. 198, 37–54.

Chandler, G.T., 1986. High-density culture of meiobenthic harpacticoid copepodswithin a muddy sediment substrate. Can. J. Fish. Aquat. Sci. 43, 53–59.

Chandler, G.T., Green, A., 1996. A 14-day harpacticoid copepod reproductionbioassay for laboratory and field contaminated muddy sediments. In: Ostran-der, G. (Ed.), Techniques in Aquatic Toxicology. Lewis Publishers, Boca Raton,Florida, pp. 23–39.

Chapman, P.M., Anderson, J., 2005. A decision-making framework for sedimentcontamination. Integr. Environ. Assess. Manage. 1, 163–173.

Chapman, P.M, Wang, F.Y., Janssen, C., Persoone, G., Allen, H.E., 1998. Ecotoxicologyof metals in aquatic sediments: binding and release, bioavailability, riskassessment, and remediation. Can. J. Fish. Aquat. Sci. 55, 2221–2243.

Chapman, P.M., Ho, K.T., Munns, W.R., Solomon, W., Weinstein, M.P., 2002. Issues insediment toxicity and ecological risk assessment. Mar. Pollut. Bull. 44, 271–278.

Close, M.E., 1993. Assessment of pesticide contamination of groundwater in NewZealand: 2. Results of groundwater sampling. N. Z. J. Mar. Freshwater Res. 27,267–273.

Close, M.E., Rosen, M.R., 2001. 1998/99 national survey of pesticides in groundwaterusing GCMS and ELISA. N. Z. J. Mar. Freshwater Res. 35, 205–219.

Dahl, U., Gorokhova, E., Breitholtz, M., 2006. Application of growth-relatedsublethal endpoints in ecotoxicological assessments using a harpacticoidcopepod. Aquat. Toxicol. 77, 433–438.

Davoren, M., Shuilleabhain, S.N., O'Halloran, J., Hartl, M.G.J., Sheehan, D., O'Brien, N.M., Van Pelt, F., Mothersill, C., 2005. A test battery approach for the ecotox-icological evaluation of estuarine sediments. Ecotoxicology 14, 741–755.

De Coen, W.M., Janssen, C.R., 1998. The use of biomarkers in Daphnia magna toxicitytesting – I. The digestive physiology of daphnids exposed to toxic stress.Hydrobiologia 367, 199–209.

De Schamphelaere, K.A.C., Canli, M., Van Lierde, V., Forrez, I., Vanhaecke, F., Janssen,C.R., 2004. Reproductive toxicity of dietary zinc to Daphnia magna. Aquat.Toxicol. 70, 233–244.

Di Toro, D.M., McGrath, J.A., Hansen, D.J., Berry, W.J., Paquin, P.R., Mathew, R., Wu, K.B., Santore, R.C., 2005. Predicting sediment metal toxicity using a sedimentbiotic ligand model: methodology and initial application. Environ. Toxicol.Chem. 24, 2410–2427.

DiToro, D.M., Zarba, C.S., Hansen, D.J., Berry, W.J., Swartz, R.C., Cowan, C.E., Pavlou, S.P., Allen, H.E., Thomas, N.A., Paquin, P.R., 1991. Techinical basis for establishingsediment quality criteria for nonionic organic-chemicals using equlibriumpartitioning. Environ. Toxicol. Chem. 10, 1541–1583.

Douglas, W.S., McIntosh, A., Clausen, J.C., 1993. Toxicity of sediments containingatrazine and carbofuran to larvae of the midge Chironomus tentans. Environ.Toxicol. Chem. 12, 847–853.

Fleeger, J.W., Gust, K.A., Marlborough, S.J., Tita, G., 2007. Mixtures of metals andpolynuclear aromatic hydrocarbons elicit complex, nonadditive toxicologicalinteractions in meiobenthic copepods. Environ. Toxicol. Chem. 26, 1677–1685.

Gentile, J.H., Gentile, S.M., Hairston, N.G., Sullivan, B.K., 1982. The use of life-tablesfor evaluating the chronic toxicity of pollutants to Mysidopsis bahia. Hydro-biologia 93 (1-2), 179–187.

Glasby, G.P., Moss, R.L., Stoffers, P., 1990. Heavy metal pollution in Porirua Harbor.N. Z. J. Mar. Freshwater Res. 24, 233–237.

Green, A.S., Chandler, G.T., Blood, E.R., 1993. Aqueous-phase, pore-water andsediment-phase cadmium toxicity relationships for a meiobenthic copepod.Environ. Toxicol. Chem. 12, 1497–1506.

Greenstein, D., Bay, S., Anderson, B., Chandler, G.T., Farrar, J.D., Keppler, C., Phillips,B., Ringwood, A., Young, D., 2008. Comparison of methods for evaluating acuteand chronic toxicity in marine sediments. Environ. Toxicol. Chem. 27, 933–944.

Hagopian-Schlekat, T., Chandler, G.T., Shaw, T.J., 2001. Acute toxicity of fivesediment-associated metals, individually and in a mixture, to the estuarinemeiobenthic harpacticoid copepod Amphiascus tenuiremis. Mar. Environ. Res.51, 247–264.

Harrass, M.C., 1996. Regulatory use of ecotoxicity statistics: A US perspective.Ecotoxicology 5, 145–154.

Hayes, T., Haston, K., Tsui, M., Hoang, A., Haeffele, C., Vonk, A., 2003. Atrazine-induced hermaphroditism at 0.1 ppb in American leopard frogs (Rana pipiens):laboratory and field evidence. Environ. Health Perspect. 111, 568–575.

Hayes, T.B., Collins, A., Lee, M., Mendoza, M., Noriega, N., Stuart, A.A., Vonk, A., 2002.Hermaphroditic, demasculinized frogs after exposure to the herbicide atrazineat low ecologically relevant doses. Proc. Natl. Acad. Sci. U.S.A. 99, 5476–5480.

Hellou, J., Cheeseman, K., Desnoyers, E., Johnston, D., Jouvenelle, M.L., Leonard, J.,Robertson, S., Walker, P., 2008. A non-lethal chemically based approach toinvestigate the quality of harbour sediments. Sci. Total Environ 389, 178–187.

Hicks, G.R.F., Coull, B.C., 1983. The ecology of marine meiobenthic harpacticoidcopepods. Oceanogr. Mar. Biol. 21, 67–175.

Holland, P.T., Hickey, C.W., Roper, D.S., Trower, T.M., 1993. Variability of organiccontaminants in inter-tidal sandflat sediments from Manukau Harbour,New Zealand. Arch. Environ. Contam. Toxicol. 25, 456–463.

Hook, S.E., Fisher, N.S., 2002. Relating the reproductive toxicity of five ingestedmetals in calanoid copepods with sulfur affinity. Mar. Environ. Res. 53, 161–174.

Huntley, S.L., Bonnevie, N.L., Wenning, R.J., 1995. Polycyclic aromatic hydrocarbonsand petroleum contamination in sediment from the Newark Bay estuary, NewJersey. Arch. Environ. Contam. Toxicol. 28, 93–107.

Hutchinson, T.H., Pounds, N.A., Hampel, M., Williams, T.D., 1999. Life-cycle studieswith marine copepods (Tisbe battagliai) exposed to 20-hydroxyecdysone anddiethylstilbestrol. Environ. Toxicol. Chem. 18, 2914–2920.

Ingersoll, C.G., Brunson, E.L., Dwyer, F.J., Hardesty, D.K., Kemble, N.E., 1998. Use ofsublethal endpoints in sediment toxicity tests with the amphipod Hyalellaazteca. Environ. Toxicol. Chem. 17, 1508–1523.

Kennedy, A.J., Steevens, J.A., Lotufo, G.R., Farrar, J.D., Reiss, M.R., Kropp, R.K., Doi, J.,Bridges, T.S., 2009. A comparison of acute and chronic toxicity methods formarine sediments. Mar. Environ. Res. 68, 118–127.

King, C.K., Gale, S.A., Hyne, R.V., Stauber, J.L., Simpson, S.L., Hickey, C.W., 2006.Sensitivities of Australian and New Zealand amphipods to copper and zinc inwaters and metal-spiked sediments. Chemosphere 63, 1466–1476.

Kramer, K.J.M., Jak, R.G., van Hattum, B., Hooftman, R.N., Zwolsman, J.J.G., 2004.Copper toxicity in relation to surface water-dissolved organic matter: Biologicaleffects to Daphnia magna. Environ. Toxicol. Chem. 23, 2971–2980.

T.J. Stringer et al. / Ecotoxicology and Environmental Safety 99 (2014) 82–9190

Lange, R., Hutchinson, T.H., Scholz, N., Solbe, J., 1998. Analysis of the ECETOC aquatictoxicity (EAT) database - II - comparison of acute to chronic ratios for variousaquatic organisms and chemical substances. Chemosphere 36, 115–127.

Latimer, J., Zheng, J., 2003. The sources, transport, and fate of PAHs in the marineenvironment. In: Douben, P. (Ed.), PAHs: An Ecotoxicological Perspective. JohnWiley & Sons, Ltd., West Sussex, England, pp. 7–33.

Lawton, J.C., Pennington, P.L., Chung, K.W., Scott, G.I., 2006. Toxicity of atrazine tothe juvenile hard clam, Mercenaria mercenaria. Ecotoxicol. Environ. Saf. 65,388–394.

Le Blanc, G.A., 2007. Crustacean endocrine toxicology: a review. Ecotoxicology 16,61–81.

Long, E.R., Field, L.J., Mac Donald, D.D., 1998. Predicting toxicity in marinesediments with numerical sediment quality guidelines. Environ. Toxicol. Chem.17, 714–727.

Lotufo, G.R., 1997. Toxicity of sediment-associated PAHs to an estuarine copepod:effects on survival, feeding, reproduction and behavior. Mar. Environ. Res. 44,149–166.

Lotufo, G.R., 1998a. Bioaccumulation of sediment-associated fluoranthene inbenthic copepods: uptake, elimination and biotransformation. Aquat. Toxicol.44, 1–15.

Lotufo, G.R., 1998b. Lethal and sublethal toxicity of sediment-associated fluor-anthene to benthic copepods: application of the critical-body-residueapproach. Aquat. Toxicol 44, 17–30.

Lotufo, G.R., Landrum, P.F., Gedeon, M.L., Tigue, E.A., Herche, L.R., 2000. Comparativetoxicity and toxicokinetics of DDT and its major metabolites in freshwateramphipods. Environ. Toxicol. Chem. 19, 368–379.

Lotze, H.K., Lenihan, H.S., Bourque, B.J., Bradbury, R.H., Cooke, R.G., Kay, M.C.,Kidwell, S.M., Kirby, M.X., Peterson, C.H., Jackson, J.B.C., 2006. Depletion,degradation, and recovery potential of estuaries and coastal seas. Science 312,1806–1809.

MacDonald, D.D., Carr, R.S., Calder, F.D., Long, E.R., Ingersol,l, C.G., 1996. Develop-ment and evaluation of sediment quality guidelines for Florida coastal waters.Ecotoxicology 5, 253–278.

Machado, W., Santelli, R.E., Carvalho, M.F., Molisani, M.M., Barreto, R.C., Lacerda, L.D., 2008. Relation of reactive sulfides with organic carbon, iron, and manganesein anaerobic mangrove sediments: Implications for sediment suitability to traptrace metals. J. Coast. Res. 24, 25–32.

Mahony, J.D., Di Toro, D.M., Gonzalez, A.M., Curto, M., Dilg, M., De Rosa, L.D.,Sparrow, L.A., 1996. Partitioning of metals to sediment organic carbon. Environ.Toxicol. Chem. 15, 2187–2197.

Marinkovic, M., Verweij, R.A., Nummerdor, G.A., Jonker, M.J., Kraak, M.H.S.,Admiraal, W., 2011. Life cycle responses of the midge Chironomus riparius tocompounds with different modes of action. Environ. Sci. Technol. 45,1645–1651.

Mc Clellan-Green, P., Romano, J., Oberdorster, E., 2007. Does gender really matter incontaminant exposure? A case study using invertebrate models. Environ. Res.104, 183–191.

Mc Murtry, M.J., 1984. Avoidance of sublethal doses of copper and zinc by tubificidoligochaetes. J. Great Lakes Res. 10, 267–272.

Mebane, C.A., Hennessy, D.P., Dillon, F.S., 2008. Developing acute-to-chronic toxicityratios for lead, cadmium, and zinc using rainbow trout, a mayfly, and a midge.Water Air Soil Pollut. 188, 41–66.

Mersie, W., Seybold, C., Tierney, D., McNamee, C., 1998. Effect of temperature,disturbance and incubation time on release and degradation of atrazine inwater columns over two types of sediments. Chemosphere 36, 1867–1881.

Mersie, W., McNamee, C., Seybold, C.A., Tierney, D.P., 2000. Diffusion and degrada-tion of atrazine in a water/sediment system. Environ. Toxicol. Chem. 19,2008–2014.

Northcott, G.L., Jones, K.C., 2000. Developing a standard spiking procedure for theintroduction of hydrophobic organic compounds into field-wet soil. Environ.Toxicol. Chem. 19, 2409–2417.

Paumen, M.L., Stol, P., Ter Laak, T.L., Kraak, M.H.S., Van Gestel, C.A.M., Admiraal, W.,2008. Chronic exposure of the oligochaete Lumbriculus variegatus to polycyclicaromatic compounds (PACs): bioavailability and effects on reproduction.Environ. Sci. Technol. 42, 3434–3440.

Perez-Landa, V., Simpson, S.L., 2011. A short life-cycle test with the epibenthiccopepod Nitocra spinipes for sediment toxicity assessment. Environ. Toxicol.Chem. 30, 1430–1439.

Poynton, H.C., Varshavsky, J.R., Chang, B., Cavigiolio, G., Chan, S., Holman, P.S.,Loguinov, A.V., Bauer, D.J., Komachi, K., Theil, E.C., Perkins, E.J., Hughes, O.,Vulpe, C.D., 2007. Daphnia magna ecotoxicogenomics provides mechanisticinsights into metal toxicity. Environ. Sci. Technol. 41, 1044–1050.

Ren, S., 2002. Predicting three narcosis mechanisms of aquatic toxicity. Toxicol. Lett.133, 127–139.

Rice, P.J., Anderson, T.A., Coats, J.R., 2004. Effect of sediment on the fate ofmetolachlor and atrazine in surface water. Environ. Toxicol. Chem. 23,1145–1155.

Riddell, D.J., Culp, J.M., Barid, D.J., 2005. Behavioral responces to sublethal cadniumexposurte with an experenental food web. Environ. Toxicol. Chem. 24, 431–441.

Roex, E.W.M., Van Gestel, C.A., Van Wezel, A.P., Van Straalen, N.M., 2000. Ratiosbetween acute aquatic toxicity and effects on population growth rates inrelation to toxicant mode of action. Environ. Toxicol. Chem. 19, 685–693.

Santore, R.C., Mathew, R., Paquin, P.R., Di Toro, D., 2002. Application of the bioticligand model to predicting zinc toxicity to rainbow trout, fathead minnow, andDaphnia magna. Comp. Biochem. Physiol. C Toxicol. Pharmacol. 133, 271–285.

Silva, S.J., Carman, K.R., Fleeger, J.W., Marshall, T., Marlborough, S.J., 2009. Effects ofphenanthrene- and metal-contaminated sediment on the feeding activity ofthe harpacticoid copepod, Schizopera knabeni. Arch. Environ. Contam. Toxicol.56, 434–441.

Simpson, S.L., King, C.K., 2005. Exposure-pathway models explain causality inwhole-sediment toxicity tests. Environ. Sci. Technol. 39, 837–843.

Simpson, S.L., Spadaro, D.A., 2011. Preformance and sensitivity of a rapid sublethalsediment toxicity test with the amphipod Melita plumulosa and the copepodNitocra spinipes. Environ. Toxicol. Chem 30, 2326–2334.

Simpson, S.L., Angel, B.M., Jolley, D.F., 2004. Metal equilibration in laboratory-contaminated (spiked) sediments used for the development of whole-sedimenttoxicity tests. Chemosphere 54, 597–609.

Simpson, S.L., Batley, G.E., Hamilton, I.L., Spadaro, D.A., 2011. Guidelines for copperin sediments with varying properties. Chemosphere 85, 1487–1495.

Stringer, T.J., Glover, C.N., Keesing, V., Northcott, G.L., Tremblay, L.A., 2012a.Development of a harpacticoid copepod bioassay: Selection of species andrelative sensitivity to zinc, atrazine and phenanthrene. Ecotoxicol. Environ. Saf.80, 363–371.

Stringer, T.J., Korsman, J.C., Peralta, G., Keesing, V., Tremblay, L.A., Glover, C.N.,2012b. Effects of environmental gradients on the distribution of harpacticoidcopepods in an intertidal flat, Portobello Bay, Otago Harbour, New Zealand. N. Z.J. Mar. Freshwater Res. 46, 385–397.

Strom, D., Simpson, S.L., Batley, G.E., Jolley, D.F., 2011. The influence of sedimentparticle size and organic carbon on toxicity of copper to a benthic inverter-brates in oxic/suboxic surface sediments. Environ. Toxicol. Chem. 30,1599–1610.

Trekels, H., Van de Meutter, F., Stoks, R., 2011. Effects of species-specific interactionswith predation risk on the relative species sensitivities to a pesticide in waterboatmen (Corixidae). Oikos 120, 897–905.

Tuffnail, W., Mills, G.A., Cary, P., Greenwood, R., 2009. An environmental 1H NMRmetabolomic study of the exposure of the marine mussel Mytilus edulis toatrazine, lindane, hypoxia and starvation. Metabolomics 5, 33–43.

USEPA, 1993a. Methods for Aquatic Toxicity Identification Evaluations: Phase IIIToxicity Confirmation Procedures for Samples Exhibiting Acute and ChronicToxicity, EPA 600/R-92/081, Duluth, MN.

USEPA, 1993b. Sediment Quality Criteria for the Protection of Benthic Organisms:Phenanthrene, U.S. Environmental Protection Agency, Deparptment of Water,Washington, DC, EPA/822/R-93/014.

Ward, D., Simpson, S.L., Jolley, D.F., 2013. Avoidance responses to contaminatedsediments of an amphipod (Melita plumulosa), harpacticoid copepod (Nitocraspinipes) and snail (Phallomedusa solida). Environ. Toxicol. Chem. 32, 644–652.

Weston, N.B., Porubsky, W.P., Samarkin, V.A., Erickson, M., Macavoy, S.E., Joye, S.B.,2006. Porewater stoichiometry of terminal metabolic products, sulfate, anddissolved organic carbon and nitrogen in estuarine intertidal creek-banksediments. Biogeochemistry 77, 375–408.

Wong, C.K., 1992. Effects of chromium, copper, nickel and zinc on survival andfeeding of the cladoceran Moina macrocopa. Bull. Environ. Contam. Toxicol. 49,593–599.

Wong, C.K., 1993. Effects of chromium, copper, nickel and zinc on longevity andreproduction of the cladoceran Moina macrocopa. Bull. Environ. Contam.Toxicol. 50, 633–639.

T.J. Stringer et al. / Ecotoxicology and Environmental Safety 99 (2014) 82–91 91