Trace metals associated with deep-sea tailings placement at the Batu Hijau copper–gold mine, Sumbawa, Indonesia

Marine Pollution Bulletin 73 (2013) 306–313

Contents lists available at SciVerse ScienceDirect

Marine Pollution Bulletin

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

Trace metals associated with deep-sea tailings placement at the Batu Hijaucopper–gold mine, Sumbawa, Indonesia

Brad M. Angel a,⇑, Stuart L. Simpson a, Chad V. Jarolimek a, Rob Jung a, Jorina Waworuntu b,Grant Batterham b

a Centre for Environmental Contaminants Research, CSIRO Land and Water, Locked Mailbag 2007, Kirrawee, NSW 2232, Australiab PT Newmont Nusa Tenggara, Batu Hijau Project, Sumbawa Barat, Nusa Tenggara Barat, Indonesia

a r t i c l e i n f o

Keywords:CopperArsenicMercuryDeep sea tailings placement

0025-326X/$ - see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.marpolbul.2013.04.013

⇑ Corresponding author. Tel.: +61 297106851; fax:E-mail address: [email protected] (B.M. Angel).

a b s t r a c t

The Batu Hijau copper–gold mine on the island of Sumbawa, Indonesia operates a deep-sea tailings place-ment (DSTP) facility to dispose of the tailings within the offshore Senunu Canyon. The concentrations oftrace metals in tailings, waters, and sediments from locations in the vicinity of the DSTP were determinedduring surveys in 2004 and 2009. In coastal and deep seawater samples from Alas Strait and the SouthCoast of Sumbawa, the dissolved concentrations of Ag, As, Cd, Cr, Hg, Pb and Zn were in the sub lg/Lrange. Dissolved copper concentrations ranged from 0.05 to 0.65 lg/L for all depths at these sites. Dis-solved copper concentrations were the highest in the bottom-water from within the tailings plume insideSenunu Canyon, with up to 6.5 lg Cu/L measured in close proximity to the tailings discharge. In general,the concentrations of dissolved and particulate metals were similar in 2004 and 2009.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

The Batu Hijau copper–gold mine is located on the southwestcorner of the island of Sumbawa in the province of West NusaTenggara, Indonesia (Fig. 1). The mine is operated by PT NewmontNusa Tenggara (PTNNT) and processes ore at a rate of approxi-mately 130,000 t/d. The open-pit mine is at an elevation of450 m above sea level. Crushed ore is processed through a grindingcircuit and flotation-based concentrator for copper and goldextraction. The tailings slurry from the operation consists of finelyground rock that is disposed of by a deep-sea tailings placement(DSTP) facility. This method of disposal was determined to haveseveral advantages over on-land disposal and was approved bythe Government of Indonesia (ANDAL, 1996). The tailings, whichtypically comprise approximately 60% solids, are conveyed bygravity in a pipeline from the process plant to the outfall dischargepoint at the head of a steep submarine canyon (the Senunu Can-yon) located 3.2 km off shore at 110 m depth. From the point ofdischarge, the tailings, flow down the steep off-shore canyon as abottom-attached density current and are ultimately deposited inthe deep oceans at depths in excess of 3000 m.

The use of DSTP for mining operations is not uncommon inIndo-Pacific archipelagic nations, and has also been used bymines in the northern hemisphere (Apte and Kwong, 2004; Jonesand Ellis 1995; Poling et al., 2002; SAMS, 2010) (Supplementary

All rights reserved.

+61 297106800.

Table S1). The existing DSTP facilities of PTNNT deposit tailingsat depths of between 1000 and 3000 m, which are considerablydeeper than a number of past marine disposal operations thathave deposited at depths of between 80 and 500 m (e.g. IslandCopper, Kitsault – Canada; Atlas Copper – Philippines; Black An-gel – Greenland; Minahasa Raya – Indonesia). The justificationfor selecting DSTP for tailings management is usually based onthe challenges in maintaining on-land tailings storage facilitiesin high rainfall, mountainous and seismically active terrain andthe proximity to deep oceanic canyons. The DSTPs are engineeredto ensure that tailings density, seabed gradients and oceano-graphic currents and upwelling characteristics at the pipeline exitpoint minimise material rising into the surface waters. The BatuHijau copper/gold mine is one of the larger examples of DSTP inthe world and deposits tailings at the greatest depth.

The discharged tailings slurry contains elevated concentra-tions of a range of metals that have the potential to impact onthe marine environment. However, the dilution in the receivingenvironment is predicted to result in concentrations being belowthose predicted to cause effects to aquatic biota (ANZECC/ARM-CANZ, 2000). The present study investigated the water-borneand benthic sediment metal concentrations in Senunu Canyon,the South Coast of Sumbawa Island, and Alas Strait, which sepa-rates Sumbawa and Lombok Islands. Depth profile sampling wasperformed at a range of locations to allow comparison of dis-solved metals near the discharge point with those in the sur-rounding environment and with internationally accepted waterquality guidelines.

Fig. 1. Location of the sampling sites in the water off Sumbawa Island, Indonesia.

B.M. Angel et al. / Marine Pollution Bulletin 73 (2013) 306–313 307

2. Methods

2.1. Field site

This study involved two water sampling surveys undertaken inOctober 2004 and April 2009 in the area shown in Fig. 1. The studyarea included five reference sites in Alas Strait, between the islandsof Sumbawa and Lombok, that were sufficiently distant (20–40 km) to not be impacted by the DSTP. There were twelve siteson the south-western side of Sumbawa, within 4–8 km of the DSTP,and six sites in the Senunu Canyon (sites with code ‘‘S’’) within1.5 km of the DSTP.

2.2. Water sample collection and treatment

High purity deionised water (18 MX/cm, Milli-Q, Millipore) wasused in all equipment washing procedures and for the preparationof metal standards. One-litre low-density polyethylene (LDPE)(Nalgene) bottles were used for sampling waters for all metalsother than mercury and 1-L fluorinated ethylene propylene (FEP)(Nalgene) bottles were used for sampling waters for mercury anal-yses. The LDPE bottles were cleaned rigorously before use, using athree-stage sequential washing procedure. This involved soakingbottles and caps in 2% detergent solution (Extran) for at least 2 h,then soaking in 10% HNO3 (Merck, Analytical Reagent grade) forat least 24 h, and then soaking in 1% high purity nitric acid (MerckTracepur) for at least 48 h, with at least five rinses with Milli-Qwater between each step. The 1-L FEP bottles were cleaned by

soaking in 50% analytical grade (AR) nitric acid for 48 h, and thensoaking in 10% ultra-pure grade hydrochloric acid (Merck Trace-pur) for at least 72 h, and then soaked in Milli-Q water for at least48 h, with at least five rinses with Milli-Q water, between eachstep. Each bottle was stored inside two polyethylene sealable bagsfor transport to and from the field.

Samples were collected from between 2 and 4 depths at eachsite using acid-washed (10% HNO3 v/v) Niskin� samplers. The Ni-skin� was suspended at the sample depths of 3, 50, and 120 mbelow the surface and approximately 10 m above the sea floor(varied depending on site) for approximately 10 min to rinsethe sampler with sample before triggering the device to obtainthe sample. Ultratrace sampling techniques were employed inthe field similar to those described by Ahlers et al. (1990). Twopeople were involved in the sampling operation and both worepowderless vinyl gloves (Bioclean 100, Nitritex Ltd). One personoperated the Niskin sampler by releasing the pressure valve andreleasing the tap to dispense the sample. The second person han-dled the LDPE and FEP bottles, which involved removing bottlesfrom zip-lock bags, removing the lids and filling with the samplethat was dispensed from the Niskin sampler, closing the lid, andreplacing the bottle back into two zip-lock bags. For purposesof quality assurance/quality control (QA/QC) during sampling ofwaters, at least 10% site duplicates and 10% field blanks (airblanks) were collected.

Following collection, samples were stored in ice-filled coolerboxes until return to the laboratory. The samples were then filtered(Angel et al., 2010a; Hatje et al., 2003) within 8 h of collectionusing acid-washed (10% HNO3, Tracepur, 24 h) polycarbonate filter

308 B.M. Angel et al. / Marine Pollution Bulletin 73 (2013) 306–313

rigs (Sartorius) fitted with membrane filters (50 mm, type HA, Mil-lipore). Filtrates were transferred into 1 L LDPE and FEP bottles(Nalgene) following a rinse with the initial filtrate. The filtrateswere preserved by addition of 2 mL/L concentrated HNO3 (MerckTracepur).

2.3. Physicochemical measurements

A CTD probe (STD-12 Plus profiler with dissolved oxygen (DO)sensor and transmissometer) was deployed at every water sam-pling site and measurements of water density, light transmission,dissolved oxygen, salinity, and temperature profiles were under-taken prior to sample collection.

2.4. Chemical analyses

The total dissolved arsenic concentrations were determined byhydride-generation atomic absorption spectrometry (AAS), usingprocedures based on the standard methods described by APHA(2012). Samples were first digested by addition of potassium per-sulfate (1% m/v final concentration) and heating to 120 �C for30 min in an autoclave followed by the addition of concentratedhydrochloric (3 M final concentration). Arsenic (V) was then re-duced to arsenic (III) by addition of potassium iodide (1% m/v finalconcentration) and ascorbic acid (0.2% m/v final concentration)and left standing for at least 30 min at room temperature. Arsenicconcentrations were then measured by hydride-generation AASoperating under standard conditions recommended by the manu-facturer (Varian VGA). Arsenic (III) in solution was reduced to ar-sine by borohydride, which was stripped with nitrogen gas into asilica tube and electrically heated at 925 �C. Heating converted ar-sine into arsenic vapour whose atomic adsorption was measured.

Dissolved cadmium, copper, lead, and zinc were analysed usinga dithiocarbamate complexation/solvent based on a modificationof the procedure described by Magnusson and Westerlund(1981). The chelation solution combined sodium bicarbonate buf-fer, ammonium pyrrolidine dithiocarbamate (APDC) and diet-hyldithiocarbamic acid (DDC) complexing reagents, which wereextracted into 1,1,1-trichloroethane in place of Freon (1,1,2-tri-chlorotrifluoroethane) (Angel et al., 2010a; Apte and Gunn 1987).The chelation solution was cleaned before use by extracting withtwo 10 mL aliquots of 1,1,1-trichloroethane to remove any tracemetals remaining after rinses with deionised water. Sample vol-umes (250 mL) were buffered to approximately pH 5 by additionof the chelation solution, and extracted with two 10 mL portionsof triple-distilled trichloroethane in acid-washed (50% nitric acid,48 h) Teflon separation funnels (Oak Ridge Teflon� FEP). The ex-tracts were combined in Teflon centrifuge tubes (Oak Ridge Teflon�

FEP) and the metals back-extracted into 1 mL of concentrated ni-tric acid (Merck, Tracepur). The back extracts were diluted to a fi-nal volume of 10 mL with Milli-Q water, and analysed byinductively coupled plasma – mass spectrometry (ICP-MS, Agilent7500CE) using operating conditions recommended by the manu-facturer. The method resulted in a pre-concentration factor ofapproximately 25 for the samples.

The total dissolved chromium was analysed by graphite furnaceAAS (Perkin Elmer, AAnalyst 600) using standard addition andinductively coupled plasma atomic emission spectrometry (ICP-AES) (Varian 730 ES) was used for manganese. Matrix-matchedstandards (QCD Analysts) were used and operating conditions rec-ommended by the manufacturer.

Total dissolved mercury was determined by cold vapour atomicfluorescence spectrometry (Liang and Bloom 1993; Liang et al.,1994). A sample volume of 80 mL was dispensed into a Pyrex-glasspurging vessel and a 0.4 mL aliquot of bromine monochloride(0.2 M) in hydrochloric acid added to allow oxidation of any

organic mercury present to inorganic mercury. After at least90 min a 50 lL volume of hydroxylamine solution (3 M) was addedto destroy any residual BrCl. The vessel was connected to a custom-built purge trap system and 0.5 mL stannous chloride solution(20% m/V) was added to reduce the inorganic mercury to elementalmercury. The elemental mercury was purged from solution in anitrogen stream (20 min purge time) and trapped on a gold-coatedsand trap. The trap was transferred to a thermal desorption unitinterfaced to a Brooks Rand AFS. The trap was connected to a mer-cury-free helium gas stream and rapidly heated to 320 �C. The re-leased mercury was quantified by the AFS.

As part of QA/QC for each method, 10% of samples had spikerecoveries and analytical duplicates determined, and at least onecertified reference seawater was analysed per batch of samples.The certified reference seawaters included CASS-4 obtained fromthe National Research Council Canada and BCR-579 obtained fromthe Institute for Reference Materials and Measurements at theEuropean Commission Joint Research Centre.

3. Results and discussion

3.1. Water quality

Dissolved oxygen was 5.7–6.5 mg/L at the surface for all sitesduring the 2004 and 2009 studies and decreased with depth to lessthan 2 mg/L in the depths of Senunu Canyon. Surface temperatureswere between 25 and 27 �C during the 2004 study and between 28and 30 �C during the 2009 study. The thermocline in each studywas measured between 20 and 200 m depth in the Senunu Canyon,with temperatures decreasing to less than 10 �C at 300–400 m depthand reported to be 1–2 �C at depths of 2000–4000 m (LIPI, 2004).Thermoclines were detected at depths between 75–100 and 130–140 m in 2004, and between 40–60 and 90–110 m in 2009.

From the transmissivity profiles, the tailings plume was onlydetected at sites within the Senunu Canyon in 2004 (S28, S15and S16) and 2009 (S28, S16, and CI8) studies. During the 2004study, the top of the plume was encountered at a depth of140 m, and during the 2009 study the plume was detected atapproximately 190 m, both plumes being below the dischargedepth (�112 m in 2004 and 125 m in 2009). Low transmissivityvalues (<60%) were only detected at depths greater than 250 and220 m during the 2004 and 2009 studies, respectively, which oc-curred close to the canyon floor and tailings flow from the dis-charge point. Light transmissivity at other South Coast and AlasStrait sites showed no evidence of suspended tailings particles,being consistent with the southward currents through Alas Straitand to the south of Sumbawa Island that transport suspended par-ticles offshore (LIPI, 2004).

3.2. Dissolved metal concentrations

The QC data for the dissolved metals analyses were excellent.The analytical detection limits achieved for the two surveys werein the range 0.04–0.1 lg As/L, 1.2–2.0 ng Cd/L, 4–10 ng Cu/L, 0.2–0.13 ng Hg/L, 0.3–1 lg Mn/L, 17–20 ng Pb/L, and 6–33 ng Zn/L,which were similar to those reported for previous ultratrace metalstudies (Apte et al., 1998; Mackey et al., 2002; Angel et al., 2010a).The analytical recovery of mercury in the reference seawater BCR-579 was 101% in 2009 and for CASS-4, recoveries were between 85and 104% for As, Cd, Cu, Hg, Mn, Pb and Zn in both surveys. Spikerecoveries were in the range 87–112%. Site and method duplicatecoefficient of variance was generally less than 10% except for sam-ples with metal concentrations which were close to the analyticaldetection limit (mean ± SD of 10 ± 9% for Cd, 11 ± 9% for Cr,17 ± 11% for Hg, 28 ± 18% for Mn, and 12 ± 9% for Zn).

B.M. Angel et al. / Marine Pollution Bulletin 73 (2013) 306–313 309

Dissolved copper concentrations were in the low or sub-lg/Lrange and fairly uniform in the surface water sites in the 2004and 2009 surveys (Supplementary Fig. 1 and Tables 1–3). The dis-solved copper was similar at most depths sampled in the AlasStrait and South Coast zones and did not generally exhibit a trend

Table 1The concentration of dissolved metals measured at sites sampled in Senunu Canyon.

Site Depth (m) Arsenic (lg/L) Cadmium (lg/L) Chromium (lg/L) Copper

2004 2009 2004 2009 2004 2009 2004

S09 3 1.20 – 0.010 – <1 – 0.11S09 Bottom 1.15 – 0.011 – <1 – 0.09S12 3 1.13 1.25 0.009 0.007 <1 0.1 0.07S12 50 1.23 1.28 0.016 0.008 <1 0.1 0.09S12 Bottom 1.22 1.50 0.028 0.019 2 0.1 0.12S15 3 1.24 – 0.013 – <1 – 0.10S15 50 1.13 – 0.036 – <1 – 0.39S15 Bottom 1.27 – 0.083 – 2 – 6.5S16 3 1.28 1.30 0.013 0.006 <1 0.2 0.08S16 50 1.34 1.43 0.031 0.01 – 0.2 0.13S16 120 – 1.62 – 0.035 <1 0.2 –S16 Bottom 1.40 1.76 0.085 0.063 <1 0.2 4.3S23 3 1.19 1.29 0.008 0.006 <1 0.1 0.07S23 50 – – – – – – –S23 Bottom 1.12 1.43 0.010 0.009 <1 0.1 0.08S28 3 1.26 1.23 0.013 0.004 <1 0.1 0.11S28 50 1.31 1.32 0.034 0.007 <1 0.1 0.14S28 120 – 1.57 – 0.032 – 0.2 –S28 Bottom 1.42 1.66 0.059 0.051 <1 0.2 1.5

Table 2The concentration of dissolved metals measured at sites sampled in waters off the South

Site Depth (m) Arsenic (lg/L) Cadmium (lg/L) Chromium (lg/L) Coppe

2004 2009 2004 2009 2004 2009 2004

SC1 3 1.27 1.30 0.012 0.004 <1 0.1 0.07SC1 50 1.38 1.40 0.031 0.010 <1 0.1 0.10SC1 120 – 1.65 – 0.034 – 0.2 –SC1 Bottom 1.40 1.71 0.060 0.052 <1 0.3 0.16SC2 3 1.32 1.25 0.008 0.003 <1 0.1 0.07SC2 50 1.23 1.36 0.018 0.009 <1 0.2 0.08SC2 Bottom 1.35 1.47 0.020 0.008 <1 0.2 0.09CI8 50 – 1.48 – 0.017 – 0.1 –CI8 Bottom – 1.56 – 0.032 – 0.2 –CI11 50 – 1.40 – 0.007 – 0.2 –CI11 120 – 1.59 – 0.027 – 0.2 –CO2 3 – 1.24 – 0.005 – 0.1 –CO2 50 – 1.32 – 0.006 – 0.1 –CO2 120 – 1.72 – 0.063 – 0.1 –CO2 Bottom – 1.57 – 0.033 – 0.1 –CO3 3 – 1.26 – 0.004 – 0.1 –CO3 50 – 1.45 – 0.014 – 0.2 –CO3 120 – 1.59 – 0.033 – 0.2 –CO3 Bottom – 1.68 – 0.053 – 0.2 –

Table 3The concentration of dissolved metals measured at sites sampled in Alas Strait.

Site Depth (m) Arsenic (lg/L) Cadmium (lg/L) Chromium (lg/L) Coppe

2004 2009 2004 2009 2004 2009 2004

AS1 3 – 1.22 – 0.004 – 0.1 –AS1 Bottom – 1.55 – 0.014 – 0.2 –B05 3 1.23 1.26 0.007 0.005 <1 0.1 0.07B05 Bottom 1.11 1.38 0.010 0.008 <1 0.1 0.08BCM 3 1.07 1.22 0.008 0.004 <1 0.2 0.05BCM Bottom 1.14 1.35 0.008 0.006 <1 0.1 0.06KR1 3 1.23 1.25 0.006 0.005 <1 0.1 0.05KR1 Bottom 1.26 1.45 0.011 0.011 <1 0.2 0.06TJ1 3 1.23 1.61 0.005 0.025 <1 0.2 0.05TJ1 Bottom 1.45 1.25 0.039 0.004 <1 0.1 0.10

with depth. However, some of the deeper samples at sites in closeproximity to the point of discharge were much higher than theafore-mentioned sites and in the low lg/L) range. In the 2004 sur-vey, the highest concentrations were 1.5, 6.5 and 4.3 lg/L mea-sured in the bottom waters of sites S28, S15 and S16,

(lg/L) Manganese (lg/L) Mercury (ng/L) Lead (lg/L) Zinc (lg/L)

2009 2004 2009 2004 2009 2004 2009 2004 2009

– <1 – – – 0.06 – 0.11 –– <1 – – – <0.02 – 0.09 –0.10 <1 0.5 1.0 0.5 <0.02 <0.02 0.05 0.071.17 <1 0.7 0.4 0.9 0.02 0.63 0.14 2.610.11 <1 0.3 0.5 1.8 0.02 0.02 0.11 0.14– <1 – 0.4 – 0.03 – 0.20 –– <1 – 0.4 – 0.12 – 1.0 –– 6 – 0.5 – 0.06 – 0.84 –0.09 <1 1.1 0.4 1.3 0.03 <0.02 0.14 0.050.10 <1 0.6 0.4 0.5 – 0.02 – 0.080.10 – 1.3 – 0.6 0.03 0.12 0.15 0.204.02 3 10 0.5 0.7 0.05 0.16 0.50 1.010.11 <1 0.3 0.4 1.4 0.02 <0.02 0.06 0.08– – – – – – – – –0.10 <1 <0.3 0.6 0.3 0.02 <0.02 0.05 0.080.10 <1 0.6 0.5 0.4 0.04 0.01 0.26 0.060.08 <1 0.5 0.5 0.6 <0.02 0.02 0.17 0.060.12 – <0.3 – 0.4 – <0.02 – 0.141.76 <1 3.6 0.5 0.5 0.04 0.11 0.36 0.71

Coast of Sumbawa Island.

r (lg/L) Manganese (lg/L) Mercury (ng/L) Lead (lg/L) Zinc (lg/L)

2009 2004 2009 2004 2009 2004 2009 2004 2009

0.08 <1 0.3 – 0.6 0.05 <0.02 0.12 0.070.13 <1 0.4 – 0.6 0.04 0.03 0.17 0.420.09 – 0.5 – 1.9 – <0.02 – 0.130.18 <1 0.3 – 0.4 0.06 <0.02 0.35 0.230.08 <1 0.4 – 1.6 0.03 <0.02 0.08 0.110.09 <1 0.4 – 0.3 0.04 <0.02 0.12 0.120.16 1 0.5 – 1.0 0.04 <0.02 0.11 0.160.07 – 0.4 – 1.1 – <0.02 – 0.100.09 – 0.3 – 0.8 – <0.02 – 0.110.09 – 0.6 – 0.8 – 0.06 – 0.130.07 – 0.4 – 0.7 – 0.02 – 0.110.09 – 0.5 – 0.5 – <0.02 – 0.390.08 – 0.3 – 0.9 – 0.03 – 0.120.19 – 0.5 – 0.8 – 0.02 – 0.210.10 – 0.3 – 0.4 – <0.02 – 0.140.08 – 0.4 – 0.6 – <0.02 – 0.070.09 – 0.4 – 0.4 – <0.02 – 0.080.10 – <0.3 – 0.9 – <0.02 – 0.120.65 – 1 – 0.7 – 0.04 – 0.27

r (lg/L) Manganese (lg/L) Mercury (ng/L) Lead (lg/L) Zinc (lg/L)

2009 2004 2009 2004 2009 2004 2009 2004 2009

0.11 – 0.5 – 0.5 – <0.02 – 0.080.05 – 0.7 – 0.7 – <0.02 – 0.110.30 <1 0.8 – 0.4 0.02 0.09 0.06 0.310.08 <1 0.5 – 0.5 0.03 0.02 0.10 0.060.29 <1 0.5 – 1.0 0.03 0.12 0.07 0.440.09 <1 0.7 – 0.6 <0.02 0.10 0.10 0.080.26 <1 1 – 0.5 0.02 0.04 0.08 0.240.07 <1 0.4 – 0.4 <0.02 0.04 0.07 0.100.07 <1 <0.3 – 0.5 0.03 0.04 0.09 0.150.20 <1 0.5 – 0.7 0.02 0.04 0.17 0.34

Table 4Dissolved trace metal concentrations in oceans and industrialised waters around the world.

Location Arsenic(lg/L)

Cadmium(lg/L)

Copper(lg/L)

Mercury (lg/L) Lead (lg/L) Manganese(lg/L)

Nickel(lg/L)

Zinc (lg/L) References

Sumbawa, Indonesia (this study)Alas Strait 1.28 ± 0.15 0.010 ± 0.008 0.11 ± 0.08 0.0006 ± 0.0002 0.034 ± 0.032 0.54 ± 0.17 0.18 ± 0.02 0.14 ± 0.11 This studyEast sumbawa 1.44 ± 0.16 0.023 ± 0.019 0.12 ± 0.12 0.0008 ± 0.0004 0.024 ± 0.017 0.46 ± 0.19 0.22 ± 0.05 0.16 ± 0.10 This studySenunu Canyon 1.33 ± 0.16 0.025 ± 0.023 0.75 ± 1.55 0.0006 ± 0.0004 0.259 ± 0.117 1.23 ± 2.09 0.23 ± 0.08 0.33 ± 0.52 This studyOffshore and ocean waters (worldwide)Torres straight & Gulf

of Papua– <0.001–

0.0290.036–0.986

– – – 0.94–4.60 – Apte and Day(1998)

Pacific ocean – 0.0003–0.11 – – – – – 0.015–0.52

Bruland et al.(1994)

Tasman sea – 0.0025 0.03 <0.0014 0.01 – 0.18 <0.022 Apte et al. (1998)North pacific – 0.0011 0.04 – – – 0.12 – Mackey et al.

(2002)North atlantic – 0.0007 0.068 – 0.136 – – – Kremling and Pohl

(1989)Southern GBRMPa – <0.0015–

0.004<0.019–0.085

– <0.01–0.12 – 0.11–0.19 <0.03–0.14

Angel et al. (2010a)

North pacific ocean – – – 0.0004 - – – – Gill and Fitzgerald(1988)

Atlantic ocean 0.94–1.39 – – – – – – – Cutter and Cutter(1995)

Chatham rise, pacificocean

1.20–1.50 – – – – – – – Ellwood andMaher (2002)

Guideline valuesb NVc 5.5 1.3 0.4d 44 NVc 70 15 ANZECC/ARMCANZ(2000)

a Great barrier reef marine park.b ANZECC/ARMCANZ, 2000 trigger value for 95% species protection in marine waters.c NV = Insufficient toxicological data available for derivation of guideline trigger value.d Trigger value for inorganic mercury. There is insufficient toxicological data for the derivation of methyl mercury guideline trigger value.

310 B.M. Angel et al. / Marine Pollution Bulletin 73 (2013) 306–313

respectively. In the 2009 survey, the highest concentrations were1.8, 4.0, and 1.2 lg/L measured in the bottom waters of S28 andS16 in the Senunu Canyon and the middle depth of S12 (site near-est to the point of tailings discharge), respectively. These weremuch higher than the dissolved copper concentrations measuredat any other sites (Tables 1–3), and are indicative of a tailings sig-nature. The dissolved copper concentrations in the tailings liquidranged from 0.18 to 5.3 lg/L (Table S2) and, given the likely dilu-tion from mixing with ocean water, would indicate that some re-lease of copper must be occurring from tailings particulates.

During the operation of the DSTP, the area in the vicinity of theturbulent tailings flow is likely devoid of most life-forms due to thephysical impacts of the tailings. Inputs of copper to nearby seawa-ter are likely to occur because of a combination of dissolved copperin the tailings discharge and rapid desorption of copper from thetailings sediments as they mix with seawater. As the Senunu Can-yon zone adjacent to the discharge is physically impacted by thetailings flow, it is not expected that many aquatic organisms wouldreside in this zone to receive chronic impacts from dissolved met-als. Many organisms also have the ability to avoid contaminants(Weber, 1997), and short, pulsed exposures to dissolved coppermay have significantly less effect than constant chronic exposures(Bearr et al., 2006; Angel et al., 2010b). While the dissolved copperconcentrations were below the Indonesian receiving water stan-dard for the protection of marine life (8 lg/L; Kepmen LH 51/2004) at all sites, the concentration at the bottom sites of S15and S16 was above the guideline concentration of 1.3 lg/L appliedto marine waters in Australia (ANZECC/ARMCANZ, 2000).

At the time of sampling mostly fresh ore was being processedand the current study represents a snap-shot under these condi-tions. When the mill is processing stockpiled ore, that may be par-tially oxidised, dissolved copper concentrations have beenobserved to exceed the guidelines in the bottom Senunu Canyonwaters (Jorina Waworuntu, personal communication).

Dissolved zinc concentrations were in the sub-lg/L range andfairly uniform at most sites during the 2004 and 2009 surveys

(Supplementary Fig. 1 and Tables 1–3). The mean concentration(mean ± SD) for all marine sites was 0.25 ± 0.46 lg/L, which iswell below the guideline concentration of 15 lg/L applied formarine waters in Australia (ANZECC/ARMCANZ, 2000). The dis-solved zinc did not generally exhibit a trend with depth, althoughsome sites had higher dissolved zinc concentrations at greaterdepth. This was most evident at deeper samples in close proxim-ity to the point of discharge, which had concentrations in the lowlg/L range. The highest concentration in 2004 was 1.0 lg/L mea-sured at 50 m depth at site S15. The concentrations of dissolvedzinc measured in 2009 at the bottom sites S28 and S16 in the Se-nunu Canyon and the middle depth of S12 were 0.71, 1.01, and2.61 lg/L, respectively. These were much higher than any otherdissolved zinc concentrations measured, and are likely to indicatea signature of the tailings plume. However, even the highest con-centration measured at the middle depth (50 m) at site S12 wasmore than an order of magnitude below guideline concentrations.Sample sites in the Senunu Canyon and South Coast that werefurther from the point of tailings discharge had similar zinc con-centrations to those measured in Alas Strait, indicating that thetailings plume only had a localised effect. The dissolved zinc con-centrations in the tailings liquid ranged from 0.08 to 0.8 lg/L(Table S2) and were lower than or similar to some samples closeto the tailings discharge, indicating zinc is released from the tail-ings particulates.

The concentrations of dissolved arsenic were fairly uniform dur-ing the 2004 study, but exhibited a marginal increase with depth inthe 2009 study (Supplementary Fig. 1 and Tables 1–3). The averageconcentration for all sites sampled in 2004 was 1.13 ± 0.38 lg/L,while the average concentration of all sites sampled in 2009 was1.41 ± 0.16 lg/L. There did not appear to be a seawater signaturefrom the tailings discharge and the dissolved arsenic concentra-tions were typical of open ocean waters (Cutter and Cutter, 1995;Duan et al., 2010; Ellwood and Maher, 2002). Dissolved arsenicwas also lower than the conservative low reliability guideline con-centrations of 2.3 and 4.5 lg/L for As (III) and As (V), respectively,

B.M. Angel et al. / Marine Pollution Bulletin 73 (2013) 306–313 311

indicating a low risk of organism toxicity from arsenic exposure(ANZECC/ARMCANZ, 2000).

The concentrations of dissolved cadmium at the Alas Strait,South Coast and Senunu Canyon sites were in the ng/L range andrelatively low in both the 2004 and 2009 studies (SupplementaryFig. 1). A similar trend in the depth profile of concentrations wasobserved in both studies with dissolved cadmium exhibiting aclear increase with depth at sites within the Senunu Canyon andat some of the South Coast locations. These depth profiles are con-sistent with typical nutrient-like depth profile distributions of dis-solved cadmium in the open ocean, but may also indicate a smallcontribution from the tailings plume in the Senunu Canyon. Inputsof cadmium into nearby seawater are likely to occur because of acombination of dissolved cadmium in the tailings discharge anddesorption of cadmium from the tailings sediments as they mixwith entrained seawater. Cadmium is likely to be one of the moremobile metals in the tailings plume, as chloride complexation facil-itates the partitioning of cadmium into the dissolved phase. Thehighest cadmium concentrations were 0.085 lg/L in the bottomwater at site S16 during the 2004 study, and 0.063 lg/L in the bot-tom water at site S16 during the 2009 study. The dissolved cad-mium concentrations in the tailings liquid ranged from 0.10 to0.66 lg/L (Table S2) and given the dilution with seawater, wouldindicate that release from the tailings particulates influence thedissolved cadmium concentration of sites in close proximity tothe discharge. All dissolved cadmium concentrations were muchlower than the guideline concentration of 5.5 lg/L applied to mar-ine waters in Australia (ANZECC/ARMCANZ, 2000).

The concentrations of dissolved chromium were below thedetection limit of 1 lg/L at all sites except the S12 and S15 watersin the 2004 study, where 2 lg/L was found (Tables 1–3). A lowerdetection limit for dissolved chromium was achieved in the 2009study, which resulted in concentrations being measured abovethe detection limit in all samples. The concentration of dissolvedchromium was fairly uniform and relatively low at all locationsand depths and the mean concentration (mean ± SD) for all marinesites was 0.14 ± 0.06 lg/L. There did not appear to be a seawatersignature of dissolved chromium from the tailings discharge.

The concentrations of dissolved mercury were in the low orsub-ng/L range and uniformly low in waters at all locations anddepths sampled in the 2004 and 2009 studies (SupplementaryFig. 1 and Tables 1–3). The mean concentration (mean ± SD) forall marine sites was 0.7 ± 0.4 ng/L, which is typical of open oceanwaters (Apte et al., 1998; Gill and Fitzgerald, 1988). The highestconcentration measured was 1.9 ng/L, at site SC1, which wasapproximately 500 times below concentrations of Indonesian reg-ulatory concern. The concentration of dissolved mercury in thetailings supernatant ranged from 0.3 to 4.5 ng/L, and there didnot appear to be a seawater signature of dissolved mercury associ-ated with the tailings discharge.

The concentrations of dissolved lead were in the sub-lg/L rangeand similarly low at most sample sites in Alas Strait, the South Coastand Senunu Canyon in 2004 and 2009 (Supplementary Fig. 1 andTables 1–3). The mean concentration (mean ± SD) for all marinesites was 0.09 ± 0.16 lg/L, and many sites were below the limit ofdetection (<0.02 lg/L). Dissolved lead was marginally higher atsome sites in close proximity to the tailings plume, but most siteshad concentrations near or below the detection limit, indicatingthat the tailings plume had a low or negligible seawater signature.

The concentrations of dissolved manganese were mostly in thesub-lg/L range and fairly uniform at nearly all sites in the 2009survey and were mostly below detection in the 2004 survey (Ta-bles 1–3). The mean concentration (mean ± SD) for all marine sitessampled during 2009 was 0.49 ± 0.20 lg/L. Concentrations in thelow lg/L range were measured at two locations in the Senunu Can-yon in close proximity to the point of discharge in both surveys.

The highest concentrations of dissolved manganese measured in2004 were 3 and 6 lg/L at bottom depths of sites S16 and S15,respectively, and during 2009 were 3.6 and 10 lg/L at the bottomdepth at sites S28 and S16, respectively. The sites with the highestconcentrations of dissolved manganese also contained some of thehighest concentrations of dissolved copper and zinc, which is fur-ther evidence of a localised signature from the tailings in these bot-tom waters of Senunu Canyon.

3.3. Sediment metal concentrations

The concentrations of particulate metals in benthic sedimentsdetermined by aqua-regia digestion are shown in SupplementaryTable S3. A similar range of concentrations were measured in the2004 and 2009 surveys, and all particulate metals except copperwere less than the sediment quality guideline values (SQG) appliedin Australia (ANZECC/ARMCANZ, 2000). Similar concentrations ofAs, Cd, Cr, and Hg were measured at all sediment collection loca-tions and in the tailings composite sediments, indicating therewas no influence of the tailings discharge on these particulate met-als. The concentrations of Mn, Pb and Zn at the S01 and S03 sites inthe Senunu Canyon close to the tailings discharge were marginallyhigher than the other South Coast and Alas Strait sites and similarto the tailings composite sediments, indicating the discharge onlyhad a marginal influence on the concentrations of these metals atsites in close proximity to the discharge.

Total particulate copper concentrations at sites S01, S03 andS28 exceeded the SQG high value of 270 mg/kg. The copper con-centrations at these sites were similar to those in the compositetailings sediments (Table S3) and much higher than the otherSouth Coast and Alas Strait sites indicating sediments in closeproximity to the discharge may be predominantly sourced fromthe tailings discharge causing localised elevation of particulatecopper. At all other sites the total copper concentrations were be-low the lower sediment quality guideline (SQG) value of 65 mg/kg.

Total metal concentrations provide little indication of the bio-availability of metals in sediments (Simpson and Batley, 2007).The bioavailability of sediment-bound metals can be measuredthrough their ability to be assimilated by an organism and may beinfluenced by the solid-phase chemical speciation, i.e. metal bindingwith particulate sulfide, organic carbon, and iron and manganese(oxy)hydroxide phases. Naturally occurring metals are often presentin highly mineralised forms (insoluble in dilute acids such as 1 MHCl) and will be less bioavailable to benthic organisms comparedto anthropogenic metals (Simpson and Spadaro, 2011). For analysesof eight different tailing materials (unpublished study that includedpartially oxidised ore feed materials), the mean (± standard devia-tion) total copper concentrations were 1250 ± 770 mg/kg, comparedto 150 ± 120 mg/kg for the concentration extracted in 1 M HCl (10 to20% of the total). For the sediments at S01, S03 and S28, the 1-M HCl-extractable copper concentrations were below SQG value of 65 mg/kg and should not cause direct toxicity to benthic organisms.

The behaviour of the organisms is considered a significant factordetermining the extent of exposure of an organism and thereforethe magnitude of the hazard and the overall risk the sediment posesto ecosystem health. Many organisms will avoid sediments withundesirable properties and limit impacts to their populations(Lopes et al., 2004; Simpson et al., 2012). During the operation ofthe DSTP, the tailings deposition zone within Senunu Canyon islikely to remain devoid of most benthic invertebrates due to theongoing deposition of solids. The ability of meiofauna to recolonisa-tion the two mine tailings (400–1000 mg Cu/kg) from Batu Hijauwas investigated by Gwyther et al. (2009) using meso- and micro-cosms deployed in marine waters at a depth of 15 m. They observedthat the period of time required for abundances of meiofauna to in-crease from zero to levels statistically indistinguishable from

312 B.M. Angel et al. / Marine Pollution Bulletin 73 (2013) 306–313

natural unaffected controls were twice as long for tailings(203 days) than for defaunated controls (97 days). Furthermore,colonisation was well established in tailing from freshly minedore after 40 days, and in tailing from stockpiled, partially oxidised,ore after 65 days. That study provided further evidence of the rela-tively low bioavailability of the copper within the tailing materials.

3.4. Dissolved trace metal distributions

The depth profiles of dissolved metals can be divided into threecategories; those that were fairly uniform (conservative) withdepth (mercury, lead, and nickel), those that were fairly uniformexcept at a few locations in the Senunu Canyon in close proximityto the tailings discharge (manganese and zinc), and those thatexhibited an increase with depth (arsenic, cadmium and copper).The mercury, lead, and nickel that exhibited conservative behav-iour are largely biologically and chemically non-reactive (Millero,1996). Metals such as cadmium and arsenic exhibit natural vari-ability with depth and often correlate with nutrients such as sili-cate and phosphate and exhibit depletion in the euphotic zonedue to biological activity and increased concentrations at greaterdepths (Cutter and Cutter, 1995; Cutter et al., 2001; De Baaret al., 1994; Ellwood and Maher, 2002). An increase in dissolvedcopper at depth was only pronounced in the Senunu Canyon at afew sites in close proximity to the tailings discharge, indicatinglocalised mobilisation from sediments and tailings.

The localised elevation of dissolved copper, manganese and zincabove background concentrations close to the discharge correlatedwith the localised elevation of these metals in the sediments closeto the discharge in the Senunu Canyon. The increase in dissolvedcopper, manganese and zinc may be caused by a combination of di-rect contribution of dissolved metals in the tailings liquid (Supple-mentary information Table S2) and also from desorption of metalsfrom solids (Supplementary information Table S3). Higher metalconcentrations at the bottom-water sites of Senunu Canyon mayalso be the result of local diagenetic processes occurring in settledtailings. Deep-ocean seawater is oxygenated (typical dissolvedoxygen concentrations range from 2 to 5 mg/L) resulting in oxida-tion of the surface layer of sulfidic material and release of dissolvedmetals depending on the oxygen concentrations and the varioussulfide oxidation kinetics (Kwong and Hynes, 2003). Bacterialactivity may also contribute to the release of dissolved metalsalthough there is often low biological productivity in deep oceans.Continuous tailings deposition will bury the previous depositedmaterials, leading to a change in the redox to more reducing con-ditions in the underlying sediments, and stabilisation of the sulfidematerial from oxidation. The development of reducing conditionsis likely to lead to the release of metals associated with iron andmanganese oxides (especially arsenic, if present) (Schulz and Zabel,2000; Sundby, 2006).

When compared to the concentrations previously measured inopen-ocean surface waters in the Pacific and North Atlantic Oceans(Table 4), the concentrations of dissolved cadmium, copper andzinc measured in surface waters in the current study were margin-ally elevated. The concentrations of As, Ni and Hg in the surfacewaters of the current study were similar to concentrations mea-sured in the surface waters in previous studies (Table 4).

Acknowledgements

The authors thank the staff of PT Newmont Nusa Tenggara, andin particular Tonny Gultom, Windy Prayogo, Rissa Anungstri, Joh-ana Muljadi and Tonny Bachtiar, for logistical assistance duringthe field work. Loren Bardwell, James Charlesworth, Leigh Hales,Josh King, Chris Vardenga and Samuel Swain (CSIRO) are thankedfor technical assistance with sample analyses.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.marpolbul.2013.04.013.

References

Ahlers, W.W., Reid, M.R., Kim, J.P., Hunter, K.A., 1990. Contamination-free samplecollection and handling protocols for trace elements in natural waters. Mar.Freshw. Res. 41, 713–720.

ANDAL, 1996. Environmental Impact Assessment, Batu Hijau Copper–Gold Project,Jereweh District, Sumbawa Regency, Nusa Tenggara Province, Indonesia.

Angel, B., Hales, L.T., Simpson, S.L., Apte, S.C., Chariton, A., Shearer, D., Jolley, D.F.,2010a. Spatial variability of cadmium, copper, manganese, nickel and zinc in theport curtis estuary, Queensland, Australia. Mar. Freshw. Res. 61, 170–183.

Angel, B.M., Simpson, S.L., Jolley, D.F., 2010b. Toxicity to Melita plumulosa fromintermittent and continuous exposures to dissolved copper. Environ. Toxicol.Chem. 29, 2823–2830.

ANZECC/ARMCANZ, 2000. Australian and New Zealand guidelines for fresh andmarine water quality. The Guidelines, vol. 1. Australian and New ZealandEnvironment and Conservation Council and Agriculture and ResourceManagement Council of Australia and New Zealand, Canberra, ACT, Australia.

APHA, 2012. Standard Methods for the Examination Water and Wastewater, 22nded. American Public Health Association, Water Environment Federation,American Water Works Association, Washington, USA.

Apte, S.C., Day, G.M., 1998. Dissolved metal concentrations in the torres strait andgulf of Papua. Mar. Pollut. Bull. 30, 298–304.

Apte, S.C., Gunn, A.M., 1987. Rapid determination of copper, nickel, lead andcadmium in small samples of estuarine waters by liquid/liquid extraction andelectrothermal atomic absorption spectrometry. Anal. Chim. Acta 193, 147–156.

Apte, S.C., Kwong, Y.T.J., 2004. Deep Sea Tailings Placement: Critical Review ofEnvironmental Issues. Review Undertaken by CSIRO Australia and CANMETCanada, 94 pp.

Apte, S.C., Batley, G.E., Szymczak, R., Rendell, P.S., Lee, R., Waite, T.D., 1998. Baselinetrace metal concentrations in New South Wales coastal waters. Mar. Freshw.Res. 49, 203–214.

Bearr, J., Diamond, J., Latimer, H., Bowersox, M., 2006. Effects of pulsed copperexposure on early life-stage Pimephales promelas. Environ. Toxicol. Chem. 25,1376–1382.

Bruland, K.W., Orians, K.J., Cowen, J.P., 1994. Reactive trace metals in the stratifiedcentral North Pacific. Geochim. Cosmochim. Acta 58, 3171–3182.

Cutter, G.A., Cutter, L.S., Featherstone, A.M., Lohrenz, S.E., 2001. Antinomy andarsenic biogeochemistry in the western Atlantic Ocean. Deep-Sea Res. II –Topical Stud. Oceanography 48, 2895–2915.

Cutter, G.A., Cutter, L.S., 1995. Behavior of dissolved antinomy, arsenic, andselenium in the Atlantic Ocean. Mar. Chem. 49, 295–306.

De Baar, H.J.W., Saager, P.M., Nolting, R.F., van der Meer, J., 1994. Cadmium versusphosphate in the world ocean. Mar. Chem. 46, 261–281.

Duan, L.-Q., Song, J.-M., Li, X.-G., Yuan, H-M., 2010. The behaviours and sources ofdissolved arsenic and antinomy in Bohai Bay. Cont. Shelf Res. 30, 1522–1534.

Ellwood, M.J., Maher, W.A., 2002. Arsenic and antinomy species in surface transectsand depth profiles across a frontal zone: the chatham rise, New Zealand. Deep-Sea Res. Part 1 – Oceanographic Res. Pap. 49, 1971–1981.

Gill, G.A., Fitzgerald, W.F., 1988. Vertical mercury distributions in the oceans.Geochim. Cosmochim. Acta 52, 1719–1728.

Gwyther, D., Batterham, G.J., Waworuntu, J., Gultom, T.H., Prayogo, W., Susetiono,Karnan, 2009. Recolonisation of mine tailing by meiofauna in mesocosm andmicrocosm experiments. Mar. Pollut. Bull. 58, 841–850.

Hatje, V., Apte, S.C., Hales, L.T., Birch, G.F., 2003. Dissolved trace metal distributionsin Port Jackson estuary (Sydney Harbour), Australia. Mar. Pollut. Bull. 46, 719–730.

Jones, S.G., Ellis, D.V., 1995. Deep water STD at the Misima gold and silver mine,Papua New Guinea. Marine Georesources Geotechnology 13, 183–200.

Kremling, K., Pohl, C., 1989. Studies on the spatial and seasonal variability ofdissolved cadmium, copper, and nickel in north-east Atlantic surface waters.Mar. Chem. 27, 43–60.

Kwong, Y.T.J., Hynes, T.P., 2003. Benefits and risks of submarine tailings disposal œLessons learnt from two historic mine sites in Newfoundland and otherCanadian case studies. In: Proceedings of the 6th International Conference onAcid Rock Drainage, 14–17 July, 2003. The Australian Institute of Mining andMetallurgy, Carlton South, pp. 719–724.

Liang, L., Bloom, N.S., 1993. Determination of total mercury by single-stage goldamalgamation with cold vapour atomic spectrometric detection. J. Anal. At.Spectrom. 8, 591–594.

Liang, L., Horvat, M., Bloom, N.S., 1994. An improved speciation method for mercuryby GC/CVAFS after aqueous phase-ethylation and room temperatureprecollection. Talanta 41, 371–379.

LIPI, 2004. Deep Sea Research at Senunu Canyon, Lombok Basin, South WestSumbawa, Indian Ocean, May, June and October 2003. Joint Research by theResearch Centre of Oceanography – LIPI (P2O-LIPI), Jakarta, and P.T. NewmontNusa Tenggara Barat, Sumbawa.

Lopes, I., Baird, D.J., Ribeiro, R., 2004. Avoidance of copper contamination by fieldpopulations of daphnia longispina. Environ. Toxicol. Chem. 23, 1702–1708.

B.M. Angel et al. / Marine Pollution Bulletin 73 (2013) 306–313 313

Mackey, D.J., O’Sullivan, R.J., Watson, G., Pont, D., 2002. Trace metals in the WesternPacific: temporal and spatial variability in the concentrations of Cd, Cu, Mn, andNi. Deep-Sea Res. Part 1 – Oceanographic Res. Pap. 49, 2241–2259.

Magnusson, B., Westerlund, S., 1981. Solvent extraction procedures combined withback extraction for trace metal determinations by atomic absorptionspectrometry. Anal. Chim. Acta 131, 63–72.

Millero, F.J., 1996. Chemical Oceanography. second ed. CRC Press, Boca Raton, p.469.

Poling, G.W., Ellis, D.V., Murray, J.W., Parsons, T.R., Pelletier, C.A., 2002. UnderwaterTailing Placement at Island Copper Mine. A Success Story. Society of Mining,Metallurgy and Exploration, Inc., Colorado, USA.

SAMS, 2010. Independent Evaluation of Deep-Sea Mine Tailings Placement (DSTP)in PNG, Final Report. <http://ramumine.files.wordpress.com/2010/09/complete-dstp-final-report_master-080610.pdf>.

Schulz, H.D., Zabel, M., 2000. Marine Geochemistry. Springer-Verlag, Berlin, 455 pp.Simpson, S.L., Batley, G.E., 2007. Predicting metal toxicity in sediments: a critique of

current approaches. Integrated Environ. Assess. Manage. 3, 18–31.Simpson, S.L., Spadaro, D.A., 2011. Performance and sensitivity of rapid sublethal

sediment toxicity tests with the amphipod Melita plumulosa and copepod.Environ. Toxicol. Chem. 30, 2326–2334.

Simpson, S.L., Ward, D., Strom, D., Jolley, D.F., 2012. Oxidation of acid-volatilesulfide in surface sediments increases the release and toxicity of copper to thebenthic amphipod Melita plumulosa. Chemosphere 88, 953–961.

Sundby, B., 2006. Transient state diagenesis in continental margin muds. Mar.Chem. 102, 2–12.

Weber, D.N., 1997. Mechanisms of behavioral toxicology: an integrated approach.Am. Zool. 37, 343–345.