You are on page 1of 8

Chemosphere xxx (2013) xxx–xxx

Contents lists available at ScienceDirect

Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere

Incorporating bioavailability into management limits for copper


in sediments contaminated by antifouling paint used in aquaculture
Stuart L. Simpson a,⇑, David A. Spadaro a, Dom O’Brien b
a
Centre for Environmental Contaminants Research, CSIRO Land and Water, Locked Bag 2007, Kirrawee, NSW 2232, Australia
b
Huon Aquaculture Group, PO Box 42, Dover, Tasmania 7117, Australia

h i g h l i g h t s

 Low bioavailability of copper in antifouling paints.


1
 Probability of chronic toxicity increases when bioavailable copper >65 mg kg .
 Guidelines based on dilute acid-extractable copper most effective.
 Acid-volatile sulfide was ineffective as a SQG-modifying factor.
 Zinc from fish feed is more bioavailable, but much less toxic, than the copper.

a r t i c l e i n f o a b s t r a c t

Article history: Although now well embedded within many risk-based sediment quality guideline (SQG) frameworks,
Received 19 June 2013 contaminant bioavailability is still often overlooked in assessment and management of contaminated
Received in revised form 13 August 2013 sediments. To optimise management limits for metal contaminated sediments, we assess the appropri-
Accepted 22 August 2013
ateness of a range methods for modifying SQGs based on bioavailability considerations. The impairment
Available online xxxx
of reproduction of the amphipod, Melita plumulosa, and harpacticoid copepod, Nitocra spinipes, was
assessed for sediments contaminated with copper from antifouling paint, located below aquaculture
Keywords:
cages. The measurement of dilute acid-extractable copper (AE-Cu) was found to provide the most useful
Sediment quality guidelines
Chronic toxicity
means for monitoring the risks posed by sediment copper and setting management limits. Acid-volatile
Risk assessment sulfide was found to be ineffective as a SQG-modifying factor as these organisms live mostly at the more
Antifoulant oxidised sediment water interface. SQGs normalised to %-silt/organic carbon were effective, but the ben-
Fish efits gained were too small to justify this approach. The effectiveness of SQGs based on AE-Cu was attrib-
uted to a small portion of the total copper being present in potentially bioavailable forms (typically <10%
of the total). Much of the non-bioavailable form of copper was likely present as paint flakes in the form of
copper (I) oxide, the active ingredient of the antifoulant formulation. While the concentrations of paint-
associated copper are very high in some sediments, as the transformation of this form of copper to AE-Cu
appears slow, monitoring and management limits should assess the more bioavailable AE-Cu forms, and
further efforts be made to limit the release of paint particles into the environment.
Crown Copyright Ó 2013 Published by Elsevier Ltd. All rights reserved.

1. Introduction as zinc oxide or pyrithione when used in antifouling paint


formulations.
Antifouling paints are used on the hulls of boats and fish nets in While the benefits to industries from reducing biofouling are
the aquaculture industry to prevent biofouling, i.e. the growth and considerable, elevated concentrations of copper frequently occur
colonization of marine microorganisms (Yebra et al., 2004; Brai- in confined waters such as marinas (Schiff et al., 2007), and in sed-
thwaite et al., 2007). Since the phasing out of the use of tributyltin iments due to the abrasion of painted surfaces (Dean et al., 2007;
(TBT)-based paint products due to well recognised environmental Turner, 2010; Takahashi et al., 2012; Simpson et al., 2012a). Fish
harm, paints that utilise copper as the biocidal component, mainly aquaculture also introduces metals into the aquatic environment
in the form of copper (I) oxide, now dominate the market. Zinc is through input of fish feed (Burridge et al., 2010). Zinc is an impor-
also a component of many paint formulations, and usually present tant nutrient for fish and the major source of the zinc in the
aquaculture sediments will originate from fish food formulations,
rather than the paint. While the organic waste components of
⇑ Corresponding author. Tel.: +61 2 9710 6807.
unconsumed feed and fish faeces that settle beneath aquaculture
E-mail address: stuart.simpson@csiro.au (S.L. Simpson).

0045-6535/$ - see front matter Crown Copyright Ó 2013 Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.chemosphere.2013.08.100

Please cite this article in press as: Simpson, S.L., et al. Incorporating bioavailability into management limits for copper in sediments contaminated by anti-
fouling paint used in aquaculture. Chemosphere (2013), http://dx.doi.org/10.1016/j.chemosphere.2013.08.100
2 S.L. Simpson et al. / Chemosphere xxx (2013) xxx–xxx

cages are degraded by microbes, the associated metals tend to agent grade) for a minimum of 24 h, followed by thorough rinsing
accumulate. The enrichment of organic matter surrounding aqua- with deionized water (Milli-Q, 18 MX cm). All chemicals were ana-
culture cages may also enhance biofouling of nets (Dean et al., lytical reagent grade or equivalent analytical purity. Water pH,
2007). Frequent cleaning of aquaculture nets is often undertaken salinity, temperature and dissolved oxygen measurements were
to remove biofouling, but less frequently for nets with antifoulants. made with probes, each calibrated according to manufacturer
This may occur in situ, but frequently the empty nets are lifted out instructions. Methods for sediment particle size (by wet sieving
of the water and dried before encrusted biofouling is removed by through 63 lm nylon sieves followed by gravimetry), total organic
washing. carbon (TOC, by high temperature TOC analyzer), porewater (PW)
As the copper antifouling paints are intended to be toxic, the extraction under nitrogen (centrifugation at 800g for 5 min), and
accumulation in aquatic environments has the potential to exert dissolved ammonia analyses are described in Spadaro et al.
toxic effects on a range of non-target organisms (Gammon et al., (2008). Total recoverable metals (TRM, by microwave assisted
2009; Turner et al., 2009; Simpson et al., 2012a; Ward et al., aqua regia digestion), acid-volatile sulfide (AVS) and the simulta-
2013). In the sediments beneath aquaculture cages, the concentra- neously extracted metal (SEM) measurements were made on
tions of copper and zinc are often in the 100–1000 mg kg 1 range subsamples of the same homogenized sediment according to the
(Dean et al., 2007) and this results in considerable areas which ex- method described previously (Simpson, 2001; Simpson and
ceed recommended sediment quality guidelines (SQGs) (ANZECC/ Spadaro, 2011). The dilute acid-extractable metal (AEM) concen-
ARMCANZ, 2000). Based on projected increases in aquaculture tration is the same as the SEM concentration, i.e., determined by
worldwide (Brugère and Ridler, 2004), improved management of extraction of wet sediment sample in 1 M HCl for 60 min, but not
water and sediment quality within and surrounding aquaculture necessarily extracted simultaneously with AVS. Dissolved metal
leases is an ongoing challenge for the aquaculture industry. While concentrations in acid-digests of water and sediment samples
there has been an industry trend towards using large and heavy were determined by inductively coupled plasma-optical emission
circular cages that has increased use of antifouling paints due to spectrometry (ICP–OES, Varian 730-ES) and inductively coupled
the large effort required for on-shore cleaning, the industry also plasma-mass spectrometry (ICP-MS, Agilent 7500ce). As part of
recognises the need for decreased use of antifoulants, particularly the quality assurance, analyses of filter and acid-digest blanks, rep-
those containing non-biodegradable components such as copper. licates for 20% of samples, analyte sample-spikes and the certified
The environmental risk posed of paint-derived metals in waters reference materials (CRMs) were made. Replicates were within 20%
is strongly linked to the degree of flushing that can provide uncon- and recoveries for spikes and CRMs, and PACS-2 for sediment (Na-
taminated supplies of water (Schiff et al., 2007). Copper-based tional Research Council Canada, NRCC), were within 85–115% of
antifouling paints are intended to result in a slow, but continual re- expected values. The limits of reporting for the particulate metals
lease of the active copper (Yebra et al., 2004). However, once bur- analyses (AEM and TRM) were 0.1 mg kg 1.
ied in sediments, the risk to the environment is influenced by a
broad range of factors that alter the bioavailability and bioaccessi- 2.2. Test media
bility of the metals (Simpson et al., 2002; Rainbow, 2007; Simpson
and Batley, 2007; Jones and Turner, 2010; Maruya et al., 2010). Clean seawater was collected from Port Hacking, Sydney,
Within most assessment frameworks, the use of SQGs for Australia, membrane filtered (0.45 lm), and acclimated to a room
common contaminants are a fundamental initial step in evaluating temperature of 21 ± 1 °C. Where necessary, the salinity of the fil-
risk (Batley et al., 2005; Ahlf et al., 2009; Choueri et al., 2010). The tered seawater was adjusted to the test salinity of 30 PSU using
exceedance of a SQG usually triggers additional assessment steps Milli-Q water. Clean sediments for use in culturing and controls
to determine whether there is indeed a risk posed by the were collected (0–3 cm depth) from a range of intertidal estuarine
contaminant. The toxicity of metals is strongly influenced by locations near Sydney, Australia. These sediments had previously
sediment properties, and increased amounts of silt, organic carbon, been characterised and found to have low concentrations of
acid-volatile sulfide (AVS) will reduce the bioavailability and toxic- contaminants of concern (Spadaro et al., 2008; Strom et al., 2011).
ity of copper and zinc (Di Toro et al., 2005; Strom et al., 2011). Sediments contaminated with copper-based antifouling paint were
Therefore successive assessment steps usually involve assessing collected from in and around fish farms within the Huon River Estu-
factors that influence contaminant bioavailability, and toxicity ary, Tasmania, Australia. Based on existing survey data (unpub-
testing (USEPA, 2005; Mayor et al., 2008; Ahlf et al., 2009; Simpson lished), sediment sites were chosen to have copper concentrations
et al., 2011). towards the higher end of the contamination range, but to not have
This study evaluates approaches for incorporating metal bio- high concentrations of ammonia that is often high in unfallowed
availability when setting SQGs for managing copper contamination aquaculture sediments. A consequence of this process was that
of sediments arising from the use of antifouling paint in aquacul- the sample range was not representative of the typical, and lesser,
ture. For a range of sediments from salmon aquaculture leases, degree of copper contamination within the majority of the Tasma-
the sublethal effects to reproduction of the amphipod, Melita plu- nian aquaculture leases. While some sediments exist within the
mulosa, and harpacticoid copepod, Nitocra spinipes, was assessed. leases with higher copper concentrations, typically most have con-
The results were interpreted in relation to a range of possible SQGs, siderably lower concentrations. All samples were stored refriger-
including those based on total or dilute acid-extractable metal con- ated until use. The copper-based antifouling paint used on the
centrations, mean SQG quotients (Long et al., 2006), and the use of aquaculture nets contained 13% copper (I) oxide and 4% zinc oxide
AVS (USEPA, 2005) or organic carbon and silt (Simpson et al., 2011) Hempanet 7150A (Hempel, 2009). For all sediments, analyses were
as metal bioavailability modifying factors. made of porewater metals, ammonia and sulphide, and AVS, TRM,
AEM and TOC on the unmodified sediment, and TRM, AEM and
TOC analyses of the <63 lm sediment fraction.
2. Material and methods
2.3. Organisms and bioassays
2.1. General methods
Sub-lethal whole-sediment toxicity tests for Melita plumulosa
All glass and plasticware for analyses were usually new and and Nitocra spinipes were based on rapid chronic (reproduction)
were cleaned by soaking in 10% (v/v) HNO3 (BDH, Analytical Re- tests initially described by Mann et al. (2009) and Perez-Landa

Please cite this article in press as: Simpson, S.L., et al. Incorporating bioavailability into management limits for copper in sediments contaminated by anti-
fouling paint used in aquaculture. Chemosphere (2013), http://dx.doi.org/10.1016/j.chemosphere.2013.08.100
S.L. Simpson et al. / Chemosphere xxx (2013) xxx–xxx 3

and Simpson (2011) but modified as described previously (Simp- SEM–AVS are provided for each of the test sediments in Table 1.
son and Spadaro, 2011; Campana et al., 2012). Both test methods The concentrations of As, Cd, Cr, Mn, Ni, and Pb were relatively
are amenable to a broad spectrum of sediment types (sand to silt, low and are provided in Table S1 of the Supplementary Informa-
high to low organic carbon). Both species were obtained from con- tion. Within the aquaculture sediments used in the study, the cop-
tinuous laboratory-stock cultures as described previously (Simp- per concentrations ranged from 16 to 3100 mg kg 1 and zinc
son and Spadaro, 2011). All test containers were washed concentrations from 58 to 560 mg kg 1.
sequentially with phosphate-free detergent, 1% HNO3 and Milli-Q The concentrations of copper and zinc in the aquaculture sedi-
water before use. All tests were undertaken in an environmental ments were strongly correlated (r = 0.967), potentially indicating a
chamber on a 12 h light/12 h dark cycle (21 ± 1 °C; 3.5 lmol pho- common source. The composition of the antifouling paints applied
tons s 1 m 2). The amphipod tests were undertaken in 250 mL to the nets within the aquaculture leases have typically ranged
beakers with 35 g of sediment and 200 mL of seawater. For the from 15% to 20% copper and 3% to 5% zinc (Hempel, 2009; Simpson
copepod tests, 10 mL polycarbonate (15 mm cylindrical diameter, et al., 2012a), and therefore paint particles are expected to be a
10 cm height) vials were filled with 0.5 g of sediment and 8 mL common source of both these metals. Fish feed used by the Tasma-
of seawater and capped. The dissolved oxygen concentration in nia aquaculture industry typically contain 100–120 mg kg 1 of zinc
the overlying water was maintained in the amphipod tests by bub- and less than 10–15 mg kg 1 copper (Skretting Australia). The
bling air with care not to disturb the surface of sediment, but aer- company producing the feed (Skretting Australia) estimates that
ation was not necessary for the copepod tests. The amphipod tests that at least 70 mg kg 1 of the zinc is undigestible (bound in raw
comprised five replicates and the copepod tests comprised six rep- materials), and this is likely to be deposited beneath fish cages.
licates. In the case of the amphipod tests, three more replicates per At sites with high water currents, much of the low density organic
treatment were used only for sediment chemistry analysis. Both matter from fish food and faeces may be resuspended and trans-
test procedures included addition of small quantities of food (pow- port away from the aquaculture sites, thus diluting any impacts.
ered fish food and algae) and water changes were made only at day When deposited as dense layers of sediment at sites with lower
5 for the amphipod test. water currents, there is potential for increases in metal concentra-
Physicochemical parameters, including dissolved oxygen (>85% tions to occur as faecal matter and other organic debris are
saturation was achieved by algal photosynthesis), pH (8.0 ± 0.2), consumed by microbes and sediment fauna (Kristensen, 2000).
salinity (31 ± 2‰), temperature (21 ± 1 °C) and ammonia (<3 mg Dean et al. (2007) observed that the majority of the zinc from fish
NH3-N/L) were monitored at the beginning, periodically through- food is deposited in the sediments beneath cages and reported zinc
out the test, and at test termination to ensure that they remained concentrations of 196 mg kg 1 in feed and 364 mg kg 1 in faeces,
within acceptable limits. The test endpoints of survival and total compared to 9 and 13 mg kg 1 for copper, respectively. Such pro-
offspring (embryos and juveniles for the amphipod; nauplii plus cesses may contribute significantly to the zinc concentrations of
copepodites for the copepod) were expressed as a percentage of 58–560 mg kg 1 (mean ± SD = 250 ± 130 mg kg 1) in the sedi-
the control. For quality assurance purposes, 10–18 offspring per fe- ments (Table 1). The co-occurrence of net deployment and feed
male were required in all controls for tests to be considered accept- addition over aquaculture leases would therefore also result in
able (Simpson and Spadaro, 2011). In the present study, the range the strong correlation between copper and zinc concentrations in
was 10–15 offspring/female amphipod. For the copepod test, the the sediments. Thus, while highly elevated copper concentrations
number of nauplii and copepodites in each vial was counted by can be attributed predominantly to the antifouling paint, the zinc
microscopy and total offspring expressed as a percentage of the has two significant sources and the fish feed is likely to be the ma-
control. For quality assurance purposes, a total reproductive output jor one.
of 20–60 offspring (nauplii plus copepodites) per female was
required in all controls for tests to be considered acceptable (Simp-
son and Spadaro, 2011). In the present study, the range was 23–49 3.2. Sediments quality guidelines and bioavailability modifying factors
offspring/female copepod.
The assessment of sediment quality usually involves the initial
2.4. Statistical analyses comparison of total concentrations of contaminants with sediment
quality guideline (SQG) values (Batley et al., 2005). In Australia, the
Unless otherwise stated, p = 0.05 was the level of significance. lower SQGs (termed trigger values, TVs) for copper and zinc are
All statistical analyses were carried out using the software Toxcalc 65 mg kg 1 and 200 mg kg 1, respectively (ANZECC/ARMCANZ,
for Microsoft Excel (Version 5.0.23, TidePool Scientific Software, 2000). To represent higher probabilities for biological effects,
California, USA). Effect concentrations causing x% of inhibitory ef- SQG-high values of 270 mg kg 1 for copper and 410 mg kg 1 for
fect (ECx) were estimated applying a logistic sigmoidal model zinc are used. Of the 18 sediments used in this study (Table 1),
(maximum likelihood). To evaluate the potential contributions 16 exceeded the TV for copper and 11 exceeded the TV for zinc.
considered of copper and zinc, mean sediment quality guideline Fourteen sediments exceeded the SQG-high for copper and 4 for
quotients (SQGQ) were calculated (Long et al., 2006; Simpson zinc.
and Spadaro, 2011). The concentrations of other metals and organ- SQGs are not intended to be used on a pass/fail basis, and the
ic contaminants were considered negligible and not considered. lower and higher guidelines represent low and high probabilities
The AVS–SEM and (SEM–AVS)/fOC models were applied as de- respectively for biological effects (Batley et al., 2005). Where con-
scribed in USEPA, 2005, where SEM equals the molar sum of Cd, taminant concentrations exceed the TVs, further investigations are
Cu, Ni, Pb, and Zn, and fOC equals the fraction of TOC in the usually initiated to determine whether there is indeed an environ-
sediment. mental risk associated with the exceedances. It is well recognised
that the bioavailability of metals in sediments and the effects
3. Results and discussion thresholds for benthic invertebrates are also strongly influenced
by sediment properties (USEPA, 2005; Costello et al., 2011; Strom
3.1. Sediment metal concentrations et al., 2011; Campana et al., 2013). Within this study, we firstly
consider factors influencing contaminant bioavailability and then
The concentrations of total recoverable and dilute acid-extract- undertake toxicity testing to demonstrate the presence or absence
able copper and zinc, sediment particle size, TOC, AVS and of an unacceptable impact.

Please cite this article in press as: Simpson, S.L., et al. Incorporating bioavailability into management limits for copper in sediments contaminated by anti-
fouling paint used in aquaculture. Chemosphere (2013), http://dx.doi.org/10.1016/j.chemosphere.2013.08.100
4 S.L. Simpson et al. / Chemosphere xxx (2013) xxx–xxx

Table 1
Chemical parameters of the test sediments.
1 1 1 1
Size% < 63 lm TOC (%) AVS (lmol g ) TRM (mg kg ) AEM (mg kg ) SEM–AVS (lmol g )
Cu Zn Cu Zn
Control sediments
C1 98 4.5 5±1 25 210 13 140 3.3
C2 65 3.0 3±1 17 140 9 94 2.3
C3 50 2.7 0.5 ± 0.5 95 350 15 175 2.0
C4 49 2.4 2±1 13 110 6.8 70 1.2
C5 22 3.5 2±1 4 50 4.0 31 1.6
C6 20 1.0 1±1 6 43 3.0 28 0.7
C7 10 0.5 0.6 ± 0.4 3 22 1.8 15 0.4
C8 8 0.8 15 ± 3 33 61 8.8 51 14.4
C9 <0.1 0.1 <0.2 1.0 1.2 0.5 0.6 0.1

Aquaculture sediments
A1 96 3.3 16 ± 4 16 58 5.2 40 16
A2 94 4.4 15 ± 2 143 120 37 110 13
A3 88 5.3 13 ± 4 980 270 14 170 10
A4 85 4.0 24 ± 9 970 230 54 120 21
A5 84 4.0 35 ± 5 3100 560 96 230 30
A6 79 4.6 58 ± 9 400 150 170 140 53
A7 77 4.3 10 ± 4 980 270 98 200 6
A8 68 4.3 23 ± 3 500 210 57 170 20
A9 65 4.8 42 ± 4 810 310 200 250 34
A10 65 4.1 13 ± 3 2300 420 48 160 9
A11 53 4.0 2±1 2100 410 140 260 4
A12 53 5.2 17 ± 3 310 200 31 140 14
A13 50 3.9 9±3 790 230 86 110 6
A14 46 3.0 9±3 370 160 49 150 6
A15 15 1.9 <1 1760 410 140 190 5
A16 12 1.4 40 ± 4 64 120 17 100 38
A17 8.3 0.93 41 ± 4 720 170 140 81 38
A18 5.4 0.78 5±1 110 150 35 96 3

TRM = total recoverable metals. <63 lm = <63 lm sediment fraction.


AEM = 1 M HCl-extractable metals (equivalent to SEM). SEM = simultaneously extracted metals (RCd, Cu, Ni, Pb, Zn).
TOC = total organic carbon. AVS = acid-volatile sulfide. SEM–AVS = molar difference between SEM and AVS.

The measurement of dilute acid-extractable metal (AEM) con- sediments, the porewater copper concentrations were below the
centrations frequently provides better representation of potentially limit of reporting (<1 lg L 1) and the porewater zinc concentra-
bioavailable metal pool than total recoverable metal concentrations tions ranged from <3 to 12 lg L 1. This indicated that some of
(TRM) (Simpson et al., 2011), and forms part of the recommended the zinc was either associated with phases other than AVS or easily
hierarchy of measurements for assessing compliance with SQGs released from AVS during the porewater extraction. Consistent
in Australia and New Zealand (ANZECC/ARMCANZ, 2000). Of the with the higher AE-Zn concentrations (Table 1), the porewater
18 sediments, 8 exceeded the TV based on AE-Cu and 4 exceeded measurements indicated that the zinc was more bioavailable than
the TV based on AE-Zn (Table 1). For these sediments, the ratio of the copper, however the relatively low porewater metal concentra-
AE-Cu/TR-Cu was 0.16 ± 0.14 (mean ± SD; with a range of 0.01– tions were not expected to cause to toxic effects for either species
0.43) and AE-Zn/TR-Zn was 0.67 ± 0.18 (0.38–0.94) (Table S2 of (Simpson et al., 2011; Campana et al., 2012).
the Supplementary Information). For the dry paint used within For many metals, and particularly copper, the metal binding
the aquaculture leases (Hempel, 2009), the ratios of AE-Cu/TR-Cu strength and capacity is greater in sediments that have higher con-
and AE-Zn/TR-Zn were determined to be 0.17 and <0.001, respec- centrations of particulate organic carbon (OC) and fine particles
tively. This indicates that the Cu(I)2O within the paint is not readily (e.g. <63 lm) (Besser et al., 2003; Strom et al., 2011; Campana
extractable and the zinc is essentially non-extractable in 1 M HCl. et al., 2012). The TRM concentrations of copper and zinc in
While weathering and ageing processes may increase the extract- the <63 lm sediment fraction were generally greater than in the
ability of copper and zinc from the paint materials, the AE-Cu and bulk sediment (1.5 ± 1.0  greater for Cu; 1.8 ± 1.6  for Zn;
AE-Zn concentrations measured in the sediments were consistent Table S2). For the AEM concentrations, these factors were
with paint particles being the major source of the copper and the 3.4 ± 3.4 for copper and 1.7 ± 1.5 for zinc. This indicates that the
fish feed being the major source of the zinc. bioavailability of the copper in the fine sediments is likely to be pro-
For sediments that contain a molar excess of AVS over SEM the portionally greater than that in the larger fractions. Based on the
porewater concentrations of these metals are predicted to be neg- low solubility of paint-associated copper in dilute acid (AE-Cu),
ligible and chronic effects should not result from these metals (Di the results also indicate that a smaller portion of the copper may
Toro et al., 2005; USEPA, 2005). The majority of the aquaculture be associated with paint materials in the <63 lm sediment fraction
sediments had a molar excess of AVS to SEM (Tables 1 and S1). than in the bulk sediment. The similar factors for zinc in the <63 lm
Consequently, copper or zinc released from paint particles within and bulk sediment (based on TR-Zn and AE-Zn) were consistent
the sediments is expected to exchange with the AVS phase (pre- with fish food being a major contributor to the zinc concentrations.
dominantly iron(II) sulfide) to form insoluble copper and zinc Recent studies have demonstrated that for sediments in which
sulfides (Casas and Crecelius, 1994; Simpson et al., 2000). The con- AVS is not the dominant metal binding phase, e.g. surface sedi-
sideration of partitioning of these metals to both AVS and particu- ments, that a SQG based on the OC-normalized copper concentra-
late organic carbon (Di Toro et al., 2005) also predicted the absence tion of the <63 lm sediment fraction provide better predictions of
of toxicity for the majority of the sediments (Table S1). For the 18 acute and chronic toxicity of copper to benthic invertebrates

Please cite this article in press as: Simpson, S.L., et al. Incorporating bioavailability into management limits for copper in sediments contaminated by anti-
fouling paint used in aquaculture. Chemosphere (2013), http://dx.doi.org/10.1016/j.chemosphere.2013.08.100
S.L. Simpson et al. / Chemosphere xxx (2013) xxx–xxx 5

(Strom et al., 2011; Campana et al., 2012, 2013). Those studies indi- Lethality was not assessed as an endpoint for the copepod. The
cated that 3.5 mg Cu g 1 OC in the <63 lm sediment fraction could numbers of juveniles produced by the female copepods in the con-
be applied as a SQG that provides adequate protection for copper trols were within test acceptability limits of 20–54 juveniles per fe-
exposure in a broad range of sediment types. The aquaculture sed- male. Based on the variability in N. spinipes reproduction in the
iments had OC concentrations ranging from 0.8 to 5.3 (3.6 ± 1.4)% control sediments (Simpson and Spadaro, 2011), a sediment was
and most had >50% silt (<63 lm; Table 1). Of the 18 sediments, classified as toxic to the copepods when the mean reproductive
15 exceeded the OC-%silt normalised SQG based on TR-Cu mea- output of the test sediment was less than the control by 25% and
surements and 7 exceeded based on AE-Cu measurements was statistically significantly less than the controls (t-test,
(Table S2). This indicates that for many of the sediments the cop- p < 0.05). Toxic effects on the reproductive output of N. spinipes oc-
per-binding capacity will be sufficiently modified by OC and fine curred for 7 of the 18 aquaculture sediments (Fig. 1; Table S3).
particles to prevent toxicity within the more oxidised sediments The relationships between the reproduction of each species and
layers. TR-Cu and AE-Cu concentrations in the sediments are shown in
Fig. 1. For both species, toxic effects were not observed in sedi-
ments having TR-Cu concentrations below the ISQG-high value of
3.3. Toxicity to amphipod and copepod 270 mg kg 1 (Fig. 1a). There was a high degree of variability in
the effects observed for sediments with similar copper concentra-
For the majority of the sediments, the pore water con- tions. This was attributed mostly to differences in the bioavailabil-
tained <5 mg total ammonia-N/L and <0.03 mg sulphide L 1 ity of copper, as a result of differences in the properties of the
(Table S3). For one sediment, the pore water contained 210 mg to- sediments (e.g. particle size, amount of organic matter) (Campana
tal ammonia-N/L and 28 mg sulphide L 1 and the reproduction of et al., 2012). While this variability prevented the calculation of
M. plumulosa and N. spinipes was reduced to 46% and 25% (of con- accurate effects thresholds, EC20 values of 350 and 660 mg TR-
trols), respectively. Accurate thresholds for ammonia on the repro- Cu kg 1 were estimated for M. plumulosa and N. spinipes.
duction of these two species are not available, however, no toxicity It is well recognised that total metal concentrations are often
to reproduction of either species was observed for a sediment with poor predictors of exposure and toxicity (Chapman et al., 1998;
24 mg total ammonia-N/L and <0.01 mg sulphide L 1. The 10 d Besser et al., 2003; Borgmann, 2003; Simpson and Batley, 2007;
LC50 for juvenile M. plumulosa is approximately 30 mg total Luoma and Rainbow, 2008; Casado-Martinez et al., 2010), and
ammonia-N/L (pH 8) (Spadaro et al., 2008). For N. spinipes, no ef- analyses of the AE-Cu was used to provide a better representation
fects of ammonia are typically observed in seawater with <20 mg of the copper in potentially bioavailable forms (Fig. 1b). When the
total ammonia-N/L (Ward et al., 2011). On the basis of these obser- AE-Cu concentrations exceeded the ISQG-low value of 65 mg kg 1,
vations, porewater ammonia and sulfide were not believed to be there was a greater probability of correctly predicting effects to
contributing to effects in the sediments other than the sediment reproduction of both species than when TR-Cu was used. The var-
with the highest concentrations of these two stressors. iability was high for these effects relationships also, however the
Despite the high total copper and zinc concentrations, and the EC20s estimated by applying logistic sigmoidal models were
AEM measurements indicating the presence of potentially bio- 46 mg AE-Cu kg 1 for both species. EC10 values are not relevant
available metals, none of the sediments resulted in acute toxicity for either species because such small effects cannot be distin-
to the amphipods (10 d lethality) (Table S3). The reproduction end- guished due to the variability in the reproduction in all tests.
point for M. plumulosa is usually more sensitive than the lethality The concentrations of both copper and zinc exceeded SQGs in
endpoint, but variability in the results can also increase (Mann many of the sediments. For most aquatic organisms, copper is con-
et al., 2010; Simpson and Spadaro, 2011). For the nine different siderably more toxic than zinc (AQUIRE Database: http://cfpub.e-
control sediments, the number of embryos produced by the female pa.gov/ecotox/). For juvenile M. plumulosa, the LC50 values are
amphipod was within test acceptability limits of 13–18 embryos 76 lg L 1 for copper (Strom et al., 2011) and 220 lg L 1 for zinc
per female. Based on the variability observed for M. plumulosa (Spadaro et al., 2008). For adult N. spinipes, survival is reduced
reproduction in these controls, a sediment was classified as toxic when dissolved copper is >250 lg L 1 (Perez-Landa and Simpson,
to the reproduction of the amphipods when the mean reproductive 2011) and dissolved zinc is >500 lg Zn L 1 (Ward et al., 2011).
output of the test sediment was less than the control by 15% and Accurate effect thresholds for dissolved copper alone on the repro-
was statistically significantly less than the controls (t-test, duction of these two benthic species have been difficult to deter-
p < 0.05) (Simpson and Spadaro, 2011). Ten of the 18 aquaculture mine, but copper reduces reproduction of M. plumulosa at
sediments caused toxicity to the reproduction of the amphipod 36 lg L 1 (Strom et al., 2011; Campana et al., 2012) and N. spinipes
(Fig. 1; Table S3). at 67 lg L 1 (Perez-Landa and Simpson, 2011).

(a) (b)

Fig. 1. Toxic effects on the reproduction of the amphipod and copepod in relation to the copper concentrations in the sediments of (a) total sediment copper (TR-Cu) and (b)
dilute acid-extractable copper (AE-Cu). The vertical lines represent SQGs of 65 and 270 mg kg 1, respectively (ANZECC/ARMCANZ, 2000). The 15% and 25% effects limits
(considered to be no effects limits) are shown as horizontal lines for M. plumulosa and N. spinipes, respectively.

Please cite this article in press as: Simpson, S.L., et al. Incorporating bioavailability into management limits for copper in sediments contaminated by anti-
fouling paint used in aquaculture. Chemosphere (2013), http://dx.doi.org/10.1016/j.chemosphere.2013.08.100
6 S.L. Simpson et al. / Chemosphere xxx (2013) xxx–xxx

To assist in evaluating whether zinc was likely to be contribut- sediments (Besser et al., 2003; Strom et al., 2011; Campana et al.,
ing significantly to the toxicity, mean sediment quality guideline 2012). No obvious improvement was observed in the relationships
quotients (SQGQ) were calculated based on SQGs of 65 mg kg 1 between the reproduction of each species and copper concentra-
for copper and 200 mg kg 1 for zinc (Long et al., 2006; Simpson tions when TR-Cu and AE-Cu concentrations (Fig. 1) were replaced
and Spadaro, 2011). These calculations indicated firstly that TR- with the TR-Cu and AE-Cu concentrations of the <63 lm sediment
Cu and AE-Cu concentrations provided a better description of the fractions (Fig. S2) or the TR-Cu and AE-Cu concentrations norma-
effects than TR-Zn or AE-Zn (Fig. S1a–d of the Supplementary infor- lised to OC (Fig. S3). However, when the toxicity was considered
mation). Secondly, the consideration of zinc in addition to copper in relation to the%silt/OC-normalised copper concentration
did not improve the description of the effects, compared to consid- (Fig. 3), the chronic SQG of 3.5 mg Cu g 1 proposed by Simpson
ering solely copper (Fig. S1e and f). However, while these observa- et al. (2011) was adequate for protecting against chronic effects
tions indicated that the zinc was probably not contributing to these species based on TR-Cu and may also be suitable when
substantially to observed toxicity, this conclusion was influenced based on AE-Cu. However, the degree of variability in the effects
by the choice of sediments tested (i.e. we selected those with high observed for a given concentration was similar or greater than
copper concentrations). Due to the strong correlation between the variability observed for relationships based on solely TR-Cu
copper and zinc in the sediments, any contribution of zinc to the and AE-Cu (Fig. 1).
observed toxicity could not be isolated from the effects of copper.
Although the risk posed by the sediment zinc is expected to be low, 3.5. Management limits for antifouling paint contaminated sediments
in order to improve the risk assessment, it would be necessary to
assess effects in sediments where AE-Zn concentrations were For the purposes of setting management limits, guidelines
greater than the ISQG-Low value 200 mg kg 1, but AE-Cu concen- should ideally be based on easily measurable parameters. The
trations are well below the ISQG-Low value of 65 mg kg 1. use of AVS as a modifying factor does not appear appropriate for
developing management limits for copper in aquaculture sedi-
3.4. Bioavailability modifying factors for ecotoxicology ments (Fig. 2). The use of%silt/OC-normalised guidelines appeared
effective, but the application requires analyses of both copper and
A major factor impeding the establishment of robust SQGs is the TOC in the <63 lm sediment fraction (Fig. 3). This added complex-
influence that the variability in sediment properties has on metal ity would be appropriate if the benefits over other guidelines were
bioavailability (Borgmann, 2003; Di Toro et al., 2005; Costello large. The present study indicates that a large portion of the total
et al., 2011; Simpson et al., 2011; Campana et al., 2013). To deter- copper was present in forms that were not extractable in 1 M
mine whether the variability could be accounted for by considering HCl, and consequently have a low bioavailability and toxicity to
certain sediment properties, we considered SEM–AVS and (SEM– benthic organisms (Table 1). Based on these results and past stud-
AVS)/fOC relationships (USEPA, 2005) and also copper concentra- ies (Simpson and Spadaro, 2011), we consider SQGs based on total
tions normalised to%-silt and OC (Simpson et al., 2011; Campana copper concentrations to be inappropriate for managing the risk
et al., 2012). Neither of the AVS-based approaches provided useful posed by copper in sediments contaminated with antifouling
relationships with the observed toxicity (Fig. 2). Toxic effects were paints. We believe the measurements of AE-Cu in the bulk sedi-
observed to be occurring in sediments for which an excess of AVS ment provided the most practical and effective means for setting
over SEM was present in the bulk sediments. For copper, it can be management limits (Fig. 1b). The measurement of AE-Cu is simple
difficult to interpret AVS–SEM relationships because copper sul- and excludes many forms of copper that are not bioavailable (e.g.
fides are not readily extracted in 1 M HCl (Simpson et al., 1998). forms of TR-Cu that are measured only using concentrated acid
Another important factor to consider is that the measurements digestions).
may not adequately reflect the AVS–SEM concentrations in the sur- The sediment contamination comprised elevated concentra-
face layers which are exposed to higher oxygen concentrations. tions of both copper and zinc. However the analyses indicated that
Any oxidation of AVS within the surface sediments, in which the the major source of the zinc in the sediments is likely to have been
two test organisms mostly reside, may make copper and zinc more from fish feed, rather than from the paint. While the total concen-
bioavailable (Simpson et al., 2012b), while burial of the sediments, trations of copper were usually much higher than zinc, a much
due to sediment movement or deposition from the water column greater portion of the zinc was present in forms extractable in
may be expected to result in greater formation of metal sulphide 1 M HCl (Tables 1 and S2 of the Supplementary information). How-
phases and decreased metal bioavailability. ever, we found no evidence that this zinc contributed to the ob-
Increases in the portion of fine particles or the concentration of served toxicity (Fig. S1 of the Supplementary information). For
OC have been demonstrated to increase the binding of copper in this reason, the present study indicates that little benefit will be

(a) (b)

Fig. 2. Toxic effects on the reproduction of the amphipod and copepod in relation to the (a) AVS–SEM and (b) (SEM–AVS)/fOC models of metal toxicity. The vertical dashed
lines are the corresponding model thresholds. The horizontal lines are described in Fig. 1.

Please cite this article in press as: Simpson, S.L., et al. Incorporating bioavailability into management limits for copper in sediments contaminated by anti-
fouling paint used in aquaculture. Chemosphere (2013), http://dx.doi.org/10.1016/j.chemosphere.2013.08.100
S.L. Simpson et al. / Chemosphere xxx (2013) xxx–xxx 7

(a) (b)

Fig. 3. Toxic effects on the reproduction of the amphipod and copepod in relation to the %-silt/organic carbon-normalised copper concentrations in the sediments of (a) total
sediment copper as mg <63 lm TR-Cu g 1 OC, and (b) dilute acid-extractable copper as mg <63 lm AE-Cu g 1 OC. The vertical line is the guideline of 3.5 mg <63 lm Cu g 1
OC from Simpson et al. (2011). The horizontal lines are described in Fig. 1.

gained by specifying an additional management limit for zinc. As labile and potentially more bioavailable AE-Cu forms. Transforma-
part of aquaculture lease management, monitoring of the zinc con- tions of this type are of particular concern as the pool of TR-Cu is
centrations in the sediments should continue at sites that exceed very large for a number of sediments (Table 1), and if
the SQG and the risks re-evaluated if zinc concentrations increase. transformation rates are significant then the current ecotoxicology
assessment may only be valid for the present situation of the
3.6. Level of protection achieved from ecotoxicology-derived guidelines sediments.
The transformation of the copper (I) oxide in the paint to more
The SQGs that are largely applied throughout Australia (e.g. bioavailable forms will be strongly influenced by the rate of leach-
TV = 65 mg kg 1 for copper) were intended to protect against acute ing from the surface of the paint, which on antifouled ships is facil-
effects (ANZECC/ARMCANZ, 2000). The present study used chronic itated by water movement with designed leaching rates near
toxicity tests to assess the impairment of reproduction of two ben- 5 lg Cu cm 2 d 1. Buried in a sediment, in the absence of mixing
thic invertebrates. When compared to the tests utilised to develop currents, the release rate of copper will be governed by much low-
the existing SQGs, both tests are particularly sensitive to metals, er diffusional fluxes to pore waters and eventually overlying
including copper (Simpson et al., 2011). Consequently, the SQGs waters. In buried sediments, any such releases of copper are likely
or management limits based on effects to these species are ex- to be further inhibited in anoxic sediments by AVS formation. The
pected to provide greater ecosystem protection than the existing availability of paint flakes for ingestion by organisms has been can-
SQGs. vassed but is expected to be restricted to finer particles only
While the application of a TV of 65 mg kg 1 for AE-Cu appears (Strom et al., 2011; Campana et al., 2012, 2013). Thus, while total
to provide a suitable no effect threshold for most of the sediments copper concentrations at more than twice the SQG-High value of
tested (Figs. 1, S1b), the use of the EC20 of 46 mg kg 1 of AE-Cu 270 mg kg 1 might normally be considered as unacceptable, the
estimated for both species would appear more suitable based on indications are that the majority of the copper is currently rela-
the laboratory toxicity test results (Fig. S1g). However, labora- tively inert in the sediments under fish farms and will be subject
tory-based chronic ecotoxicology tests are frequently criticised to burial and redistribution with time. To facilitate this process,
for providing overly conservative assessments of potential toxicity allowing fish pens sites within leases to fallow, in relation to the
when compared to the same sediments in situ (Mann et al., 2010). use of copper-antifouled nets, for several years is a desirable man-
Laboratory tests can often create an excessive exposure to dis- agement option. At the same time, continuing to limit the release
solved contaminants in the water column compared to what oc- of paint particles into the environment and investigate other net
curs for the same sediment in the field due to the high sediment biofouling management options is recommended.
to water ratio and non-continuous renewal of overlying water (Lib-
er et al., 2007). In the present study, the exposure of the organisms
to dissolved copper, in particular, may have been far greater than Acknowledgements
for the same sediments in field locations. This creates additional
conservatism when deriving SQGs using laboratory-based toxicity This research was funded by the CSIRO Wealth from Oceans
tests. On this basis, while a management limit for copper of Flagship and a grant from the Fisheries Research and Development
46 mg kg 1 may provide greater protection than the SQG of Corporation (FRDC 2011-041 IMAS Study). Ian Hamilton is thanked
65 mg kg 1, the existing data do not justify applying the lower va- for assisting with the analyses and bioassays. Dom O’Brien and
lue as the management limit. We believe the SQG value of Cameron Dalgleish on behalf of Huon Aquaculture and Tassal
65 mg kg 1 AE-Cu is sufficiently conservative and will provide pro- Group coordinated collection and transport of the sediments to
tection for most, if not all, benthic invertebrates. CSIRO. Ian Hamilton and Chad Jarolimek (CSIRO Land and Water)
is thanked for contributions to the sample analyses and tests. We
3.7. Transformation of copper into more bioavailable forms thank Graeme Batley (CSIRO), Lisa Golding (CSIRO), Marie Jesus
Belzunce Segarra (AZTI – Tecnalia, Spain), Catriona McLeod and
It is likely that a large portion of the non-bioavailable forms of Ruth Erikson (Institute for Marine & Antarctic Studies – Fisheries,
copper in the sediments is as copper (I) oxide, as a component of Aquaculture & Coasts Division (IMAS-FAC), University of Tasma-
antifoulant formulations in paint flakes. Uncertainty remains about nia), and Graham Woods and Erik Brain (Department of Primary
the rate at which this paint-associated copper, that is mostly Industries, Parks, Water and Environment, Tasmania) for reviewing
present in the less labile TR-Cu forms, may transform into the more of this study.

Please cite this article in press as: Simpson, S.L., et al. Incorporating bioavailability into management limits for copper in sediments contaminated by anti-
fouling paint used in aquaculture. Chemosphere (2013), http://dx.doi.org/10.1016/j.chemosphere.2013.08.100
8 S.L. Simpson et al. / Chemosphere xxx (2013) xxx–xxx

Appendix A. Supplementary material for improving methods and approaches. Integr. Environ. Assess. Manag. 3, 246–
257.
Long, E.R., Ingersoll, C.G., Macdonald, D.D., 2006. Calculation and uses of mean
Supplementary data associated with this article can be found, in sediment quality guideline quotients: a critical review. Environ. Sci. Technol. 40,
the online version, at http://dx.doi.org/10.1016/j.chemosphere. 1726–1736.
Luoma, S.N., Rainbow, P., 2008. Metal Accumulation in Aquatic Environments,
2013.08.100.
Science and Lateral Management. Cambridge University Press, Cambridge, UK.
Mann, R.M., Hyne, R.V., Spadaro, D.A., Simpson, S.L., 2009. Development and
References application of a rapid amphipod reproduction test for sediment-quality
assessment. Environ. Toxicol. Chem. 28, 1244–1254.
Mann, R.M., Hyne, R.V., Simandjuntak, D.L., Simpson, SL., 2010. A rapid amphipod
Ahlf, W., Drost, W., Heise, S., 2009. Incorporation of metal bioavailability into reproduction test for sediment quality assessment: in situ bioassays do not
regulatory frameworks-metal exposure in water and sediment. J. Soils Sed. 9, replicate laboratory bioassays. Environ. Toxicol. Chem. 29, 2566–2574.
411–419. Maruya, K.A., Landrum, P.F., Burgess, R.M., Shine, J.P., 2010. Incorporating
ANZECC/ARMCANZ, 2000. Australian and New Zealand Guidelines for Fresh and contaminant bioavailability into sediment quality assessment frameworks.
Marine Water QualityWater Quality Monitoring and Reporting. Australian and Integ. Environ. Assess. Manage. 8, 659–673.
New Zealand Environment and Conservation Council and Agriculture and Mayor, D.J., Solan, M., Martinez, I., Murray, L., McMillan, H., Paton, G.I., Killham, K.,
Resource Management Council of Australia and New Zealand, Canberra, 2008. Acute toxicity of some treatments commonly used by the salmonid
Australia. aquaculture industry to Corophium volutator and Hediste diversicolor: whole
Batley, G.E., Stahl, R.G., Babut, M.P., Bott, T.L., Clark, J.R., Field, L.J., Ho, K., Mount, sediment bioassay tests. Aquaculture 285, 102–108.
D.R., Swartz, R.C., Tessier, A., 2005. The scientific underpinnings of sediment Perez-Landa, V., Simpson, S.L., 2011. A short life-cycle test with the epibenthic
quality guidelines. In: Wenning, R.J., Batley, G.E., Ingersoll, C.G., Moore, D.W. copepod Nitocra spinipes for sediment toxicity assessment. Environ. Toxicol.
(Eds.), Use of Sediment Quality Guidelines and Related Tools for the Assessment Chem. 30, 1430–1439.
of Contaminated Sediments. SETAC (Society of Environmental Toxicology and Rainbow, P.S., 2007. Trace metal bioaccumulation: models, metabolic availability
Chemistry) Press, Pensacola, FL, pp. 39–119. and toxicity. Environ. Int. 33, 576–582.
Besser, J.M., Brumbaugh, W.G., May, T.W., Ingersoll, C.G., 2003. Effects of organic Schiff, K., Brown, J., Diehl, D., Greenstein, D., 2007. Extent and magnitude of copper
amendments on the toxicity and bioavailability of cadmium and copper in contamination in marinas of the San Diego region, California. USA. Mar. Pollut.
spiked formulated sediments. Environ. Toxicol. Chem. 22, 805–815. Bull. 54, 322–328.
Borgmann, U., 2003. Derivation of cause-effect based sediment quality guidelines. Simpson, S.L., 2001. A rapid screening method for acid-volatile sulfide in sediments.
Can. J. Fish. Aquat. Sci. 60, 352–360. Environ. Toxicol. Chem. 20, 657–2661.
Braithwaite, R.A., Cadavid Carrascosa, M.C., McEvoy, L.A., 2007. Biofouling of salmon Simpson, S.L., Batley, G.E., 2007. Predicting metal toxicity in sediments: a critique of
cage netting and the efficacy of a typical copper-based antifoulant. Aquaculture current approaches. Integ. Environ. Assess. Manage. 3, 18–31.
262, 219–226. Simpson, S.L., Spadaro, D.A., 2011. Performance and sensitivity of rapid sublethal
Brugère, C., Ridler, N., 2004. Global aquaculture outlook in the next decades: an sediment toxicity tests with the amphipod Melita plumulosa and copepod
analysis production forecasts to 2030. FAO Fisheries Circular. No. 1001. FAO, Nitocra spinipes. Environ. Toxicol. Chem. 30, 2326–2334.
Rome. pp. 47. Simpson, S.L., Apte, S.C., Batley, G.E., 1998. Effect of short term resuspension events
Burridge, L., Weis, J.S., Cabello, F., Pizarro, J., Bostick, K., 2010. Chemical use in of trace metals speciation in polluted anoxic sediments. Environ. Sci. Technol.
salmon aquaculture: a review of current practices and possible environmental 32, 620–625.
effects. Aquaculture 306, 7–23. Simpson, S.L., Rosner, J., Ellis, J., 2000. Competitive displacement reactions of
Campana, O., Spadaro, D.A., Blasco, J., Simpson, S.L., 2012. Sublethal effects of copper cadmium, copper, and zinc added to a polluted, sulfidic estuarine sediment.
to benthic invertebrates explained by changes in sediment properties and Environ. Toxicol. Chem. 19, 1992–1999.
dietary exposure. Environ. Sci. Technol. 46, 6835–6842. Simpson, S.L., Pryor, I.D., Mewburn, B., Batley, G.E., Jolley, D.F., 2002. Considerations
Campana, O., Blasco, J., Simpson, S.L., 2013. Demonstrating the appropriateness of for capping metal-contaminated sediments in dynamic estuarine
developing sediment quality guidelines based on sediment geochemical environments. Environ. Sci. Technol. 36, 3772–3778.
properties. Environ. Sci. Technol. 47, 7483–7489. Simpson, S.L., Batley, G.E., Hamilton, I., Spadaro, D.A., 2011. Guidelines for copper in
Casado-Martinez, C., Smith, B.D., Luoma, S.N., Rainbow, P.S., 2010. Metal toxicity in sediments with varying properties. Chemosphere 85, 1487–1495.
a sediment-dwelling polychaete: threshold body concentrations or Simpson, S.L., Yverneau, H., Cremazy, A., Jarolimek, C., Price, H.L., Jolley, D.F., 2012a.
overwhelming accumulation rates? Environ. Pollut. 158, 3071–3076. DGT-induced copper flux predicts bioaccumulation and toxicity to bivalves in
Casas, A.M., Crecelius, E.A., 1994. Relationship between acid volatile sulfide and the sediments with varying properties. Environ. Sci. Technol. 46, 9038–9046.
toxicity of zinc, lead and copper in marine sediments. Environ. Toxicol. Chem. Simpson, S.L., Ward, D., Strom, D., Jolley, D.F., 2012b. Oxidation of acid-volatile
13, 529–536. sulfide in surface sediments increases the release and toxicity of copper to the
Chapman, P.M., Wang, F., Janssen, C., Persoone, G., Allen, H.E., 1998. Ecotoxicology of benthic amphipod Melita plumulosa. Chemosphere 88, 953–961.
metals in aquatic sediments: binding and release, bioavailability, risk Spadaro, D.A., Micevska, T., Simpson, S.L., 2008. Effect of nutrition on toxicity of
assessment, and remediation. Can. J. Fish. Aquat. Sci. 55, 2221–2243. contaminants to the epibenthic amphipod, Melita plumulosa. Arch. Environ.
Choueri, R.B., Cesar, A., Abessa, D.M.S., Torres, R.J., Riba, I., Pereira, C.D.S., Contam. Toxicol. 55, 593–602.
Nascimento, M.R.L., Morais, R.D., Mozeto, A.A., Delvalls, T.A., 2010. Strom, D., Simpson, S.L., Jolley, D.F., Batley, G.E., 2011. Accounting for the influence
Harmonised framework for ecological risk assessment of sediments from of sediment particle size and organic carbon on toxicity of copper to benthic
ports and estuarine zones of North and South Atlantic. Ecotoxicology 19, 678– invertebrates in oxic/sub-oxic surface sediments. Environ. Toxicol. Chem. 30,
696. 1599–1610.
Costello, D.M., Burton, G.A., Hammerschmidt, C.R., Rogevich, E.C., Schlekat, C.E., Takahashi, K., Turner, A., Millward, G.E., Glegg, G.A., 2012. Persistence and metallic
2011. Nickel phase partitioning and toxicity in field-deployed sediments. composition of paint particles in sediments from a tidal inlet. Mar. Pollut. Bull.
Environ. Sci. Technol. 45, 5798–5805. 64, 133–137.
Dean, R.J., Shimmield, T.M., Black, K.D., 2007. Copper, zinc and cadmium in marine Turner, A., 2010. Marine pollution from antifouling paint particles. Mar. Pollut. Bull.
cage fish farm sediments: an extensive survey. Environ. Pollut. 145, 84–95. 60, 159–171.
Di Toro, D.M., McGrath, J.A., Hansen, D.J., Berry, W.J., Paquin, P.R., Mathew, R., Wu, Turner, A., Barrett, M., Brown, M.T., 2009. Processing of antifouling paint particles
K.B., Santore, R.C., 2005. Predicting sediment metal toxicity using a sediment by Mytilus edulis. Environ. Poll. 157, 215–220.
biotic ligand model: methodology and initial application. Environ. Toxicol. USEPA, 2005. Procedures for the Derivation of Equilibrium Partitioning Sediment
Chem. 24, 2410–2427. benchmarks (ESBs) for the Protection of Benthic Organisms: Metal Mixtures
Gammon, M., Turner, A., Brown, M.T., 2009. Accumulation of Cu and Zn in discarded (cadmium, copper, lead, nickel, silver and zinc). US Environmental Protection
antifouling paint particles by the marine gastropod, Littorina littorea. Estuar. Agency, Office of Research and Development Report EPA-600-R-02-011,
Coast. Shelf Sci. 84, 447–452. Washington, DC, USA.
Hempel. Safety Data Sheet for Hempanet 7150A. Hempel Pty. Ltd. (Australia): Ward, D., Perez-Landa, V., Simpson, S.L., Jolley, D.F., 2011. An assessment of three
Laverton North, Australia, 2009. harpacticoid copepod species for use in rapid life-cycle tests. Arch. Environ.
Jones, D.E., Turner, A., 2010. Bioaccessibility and mobilisation of copper and zinc in Contam. Toxicol. 61, 414–425.
estuarine sediment contaminated by antifouling paint particles. Estuar. Coast. Ward, D., Simpson, S.L., Jolley, D.F., 2013. Slow avoidance response to contaminated
Shelf Sci. 87, 399–404. sediments elicits sub-lethal toxicity to benthic invertebrates. Environ. Sci.
Kristensen, E., 2000. Organic matter diagenesis at the oxic/anoxic interface in Technol. 47, 5947–5953.
coastal marine sediments, with emphasis on the role of burrowing animals. Yebra, D.M., Kiil, S., Dam-Johansen, K., 2004. Antifouling technology – past, present
Hydrobiologia 426, 1–24. and future steps towards efficient and environmentally. Progr. Org. Coat. 50,
Liber, K., Goodfellow, W., Green, A., Clements, W.H., den Besten, P.J., Galloway, T.S., 75–104.
Gerhardt, A., Simpson, S.L., 2007. In situ-based effects measures: considerations

Please cite this article in press as: Simpson, S.L., et al. Incorporating bioavailability into management limits for copper in sediments contaminated by anti-
fouling paint used in aquaculture. Chemosphere (2013), http://dx.doi.org/10.1016/j.chemosphere.2013.08.100

You might also like