CSIRO PUBLISHING

Marine and Freshwater Research, 2010, 61, 170–183 www.publish.csiro.au/journals/mfr

Spatial variability of cadmium, copper, manganese, nickeland zinc in the Port Curtis Estuary, Queensland, Australia

Brad M. AngelA,B,D, Leigh T. HalesA, Stuart L. SimpsonA, Simon C. ApteA,Anthony A. CharitonA, Damon A. ShearerC and Dianne F. JolleyB

ACentre for Environmental Contaminants Research, CSIRO Land and Water, Private Mailbag 7,Bangor, NSW 2234, Australia.

BGeoQuest, Department of Chemistry, University of Wollongong, NSW 2522, Australia.CCRC for Coastal Zone, Estuary and Waterway Management, Centre for Environmental Management,

Central Queensland University, Gladstone, Qld 4680, Australia.DCorresponding author. Email: brad.angel@csiro.au

Abstract. Port Curtis is a rapidly growing industrialised and urbanised harbour in Central Queensland, Australia. Thespatial variability of trace metals in waters and suspended particulates was determined along axial transects withinthe harbour and in oceanic waters adjacent to Port Curtis. Dissolved metal concentrations were significantly elevatedin Port Curtis compared with the concentrations measured in the adjacent coastal waters. Dissolved copper and zincconcentrations ranged from <19 to 800 and <31 to 580 ng L−1, respectively, and maxima were observed in inner harbourwaters adjacent to the southern entrance to the Narrows and in close proximity to anthropogenic sources. Dissolved nickelconcentrations were measured in the range of 110 to 900 ng L−1, and exhibited a maximum concentration in the central tonorthern Narrows, in an area that was not adjacent to anthropogenic sources. Dissolved manganese concentration maximawere measured in close proximity to the dissolved nickel maxima. It appears likely that the elevated dissolved metalconcentrations in Port Curtis and the Narrows were not caused by a single point source, and are the result of severalfactors, including industrial activity around the foreshore, fluxes from sediment-porewater, low flushing durations, lowerwater pH and possibly episodic inputs from adjacent rivers.

Additional keywords: baseline study, spatial trends.

Introduction

Port Curtis estuary is situated on the eastern coast of cen-tral Queensland, Australia, adjacent to the city of Gladstoneand the World Heritage-listed Great Barrier Reef Marine Park(GBRMP) (Fig. 1). Substantial industrialisation has occurred inthe region over the last century, which has resulted in increasedurbanisation, industrialisation and shipping activities in the area.Port Curtis is one of Australia’s largest ports for the export ofraw materials such as coal. Its foreshore is lined with severalheavy industries, including the world’s largest alumina refinery,aluminium smelters, a 1680-MW coal-fired power station andseveral chemical plants (Jones et al. 2005).

A recent screening-level risk assessment of metals in thePort Curtis estuary indicated that dissolved concentrations ofmetals in the waters were in the low- or sub-µg L−1 rangeand generally below levels of regulatory concern in Australia(ANZECC/ARMCANZ 2000; Jones et al. 2005). However, theconcentrations of many metals were elevated above backgroundlevels, and local seagrass, oyster and mud whelk species wereenriched in one or more of the metals cadmium, copper, nickeland zinc (Andersen and Norton 2001; Andersen 2004; Joneset al. 2005). A shell disease in local mud crabs has also been

shown to be associated with copper enrichment (Andersen andNorton 2001). The pathways for metal enrichment were unclear,and further investigation of metal sources was recommended toimprove the information available for environmental planningand decision-making processes. Surprisingly, there are few dataon trace metal concentrations in the adjacent GBRMP, despitethe acknowledged ecological significance of the location and theincreased risk of contamination from anthropogenic activities inthe region.

The aim of this study was to investigate trace metal behaviourand sources in Port Curtis and surrounding waters by characteris-ing the spatial distributions of the cadmium, copper, nickel andzinc. Two sampling surveys were undertaken, first along axialtransects extending away from possible point sources within theharbour and through the Narrows, and then, following analy-sis of results, targeted sampling at selected sites (Fig. 1). Thisincluded a repeat axial transect through Port Curtis, and tem-poral sampling at two selected locations having elevated metalconcentrations. Laboratory experiments using sediment fromthe Narrows were conducted to investigate the effects of pHon metal partitioning between dissolved and particulate phasesin sediment re-suspension and digestion tests.

© CSIRO 2010 10.1071/MF09046 1323-1650/10/020170

Dissolved metals in Port Curtis Estuary, Australia Marine and Freshwater Research 171

Curtis Island

Great KeppelIsland

Keppel Bay

Rockhampton

Ramsay’s Crossing

Fisherman’sLanding

Boat Creek

0 70

Calliope River Gladstone

Facing Island

Kilometres

Rodd’sBay

Survey 1

Surveys 1 and 2

Survey 2

RB

FI

Studyarea

TN

KB

PC

N

FitzroyRiver

Fig. 1. Map of Port Curtis estuary and surrounding waters showing sampling sites for Survey 1 (©), Survey 2 (♦) and both surveys (�).The different areas that were compared statistically were PC = Port Curtis, TN =The Narrows, KB = Keppel Bay offshore area, FI = FacingIsland offshore area and RB = Rodd’s Bay offshore area.

Materials and methodsStudy areaIn Port Curtis, the major component estuaries, the Calliope andBoyne Rivers, the Narrows and Auckland Creek merge into anaturally sheltered 30-km long deepwater harbour protected byCurtis Island and Facing Island (Fig. 1). The main entrances thatthe estuary has with the ocean are located either side of FacingIsland, although the estuary is also connected to the ocean atKeppel Bay in the north-west via the Narrows. Most samplestaken during this study were divided into one of either five zones,in order to make a statistical comparison of metal concentrations.These included Port Curtis (PC) and the Narrows (TN), andthe Keppel Bay (KB), Facing Island (FI) and Rodd’s Bay (RB)offshore sampling zones. The KB, FI and RB offshore zonesincluded sites from each respective transect, which were greaterthan 10 km from the mainland.

Water sample collection and treatmentTwo water sampling surveys were conducted during December2003 (axial transect survey) and 2004 (targeted and temporalsampling survey). The axial transect sampling was conductedin waters through the Port Curtis estuary and extending to sur-rounding coastal waters to identify trends in metal concentrationsthat may indicate possible metal sources and also characterisebackground trace metal concentrations in the adjacent coastalwaters (Fig. 1). Water samples were collected from 49 sites∼4 km apart, along four transects: (i) Keppel Bay through the

Narrows and Port Curtis towards Rodd’s Bay (samples A1.1 toA1.21), with most sites along this transect included in eitherthe PC or TN sample zones, (ii) continuing transect in point(i) east into Rodd’s Bay, and then north-east into offshore waters(A2.1 to A2.8), with all sites included in the RB offshore area,(iii) north-east from the seaward side of Facing Island into off-shore waters (A3.1 to A3.10), with all but one site included inthe FI offshore area, (iv) north-east from the Keppel Bay intowaters beyond Great Keppel Island (A2.1 to A2.10), with all butone site included in the KB offshore area. The transects extendedto ∼45 km from the shore.

Targeted, transect and temporal sampling was conducted dur-ing December 2004 in waters of Port Curtis, the Narrows andthe Fitzroy Estuary to aid the identification of metal sources.Water samples were collected from 51 sites and included: (i) anapproximate repeat of the Survey 1 transect (A1 series) throughPC and TN (B1.1 to B1.15), (ii) targeted sampling in selectedinlets and creeks of Port Curtis and the Narrows (B2.1 to B2.9),(iii) targeted sampling of larger inlets in Keppel Bay betweenthe entrance of the Fitzroy estuary and the northern entrance ofthe Narrows (B3.1 to B3.5), and the Fitzroy estuary and River(B4.1 to B4.8). Temporal sampling was also undertaken at alocation ∼3 km south of Ramsay’s Crossing (7–8 December2004, site B1.6) and near Fisherman’s Landing (8–9 December2004, site B2.4–5) at approximately hourly intervals for 7 h, andafter 24 h.

To minimise sample contamination, all water samples weretaken from the front of the boat while the boat was gently cruis-ing forward into the current. On the vessel, care was taken to

172 Marine and Freshwater Research B. M. Angel et al.

ensure that contamination was minimised at all times by ensur-ing that staff handling the samplers and sample bottles worepowder-free disposable vinyl gloves. Samples were taken fromsurface water at a single depth of 0.5 m because the waters ofPort Curtis are considered well mixed (Herzfeld et al. 2004).Low-density polyethylene (Nalgene, Thermo Fisher Scientific,Rochester, NY) bottles were used for water sampling in thefield and for holding filtrates in the laboratory. The bottles werecleaned rigorously before use using a three-stage sequentialwashing procedure, involving sequential soaking in 2% (v/v)Extran 300 (EMD Chemicals Inc., Gibbstown, NJ) detergent,10% (v/v) high purity nitric acid (Merck Tracepur, Darmstadt,Germany), and 1% high purity nitric acid (Merck Tracepur) andfive rinses with deionised water (18 M�·cm, Milli-Q, Millipore,Academic Water System, Sydney) between different cleaningagents (Apte and Day 1998). The bottles were placed in twozip-lock bags during transportation to the field. At each site,the samples were collected by placing bottles into a holderinstalled on a purpose-built Perspex pole-sampler and usingultratrace sampling techniques (Ahlers et al. 1990; Apte andDay 1998). During water sampling, for purposes of quality assur-ance and quality control (QA/QC), at least 10% site duplicatesand 10% field blanks (air blanks from site filled with deionisedwater in the laboratory) were collected (USEPA 1996; Apte andDay 1998).

Following collection, samples were stored in ice-filledcooler boxes until return to the laboratory and were gener-ally filtered within 8 h, and always filtered within 24 h ofcollection. Filtrations were undertaken using polycarbonate fil-tration units (Sartorius, Goettingen, Germany), which wereacid-washed (10% HNO3, Merck Tracepur, 24 h), and fittedwith pre-weighed 0.45-µm membrane filters (50 mm, typeHA, Millipore) that were rinsed with 50 mL of 10% (v/v)HNO3 (Merck, Tracepur) followed by two 100-mL volumes ofdeionised water before use (Hatje et al. 2001, 2003; Teasdaleet al. 2003).

The initial 50 mL of filtrate was used to rinse the 0.5-Lacid-washed bottles (low-density polyethylene, Nalgene) anddiscarded, and the remaining filtrate was then transferred intothese bottles. The filtrates were preserved by addition of concen-trated HNO3 (Merck Suprapur, 2 mL L−1 final concentration)in a class-100 clean laboratory. The filtration rig and filter wererinsed with deionised water and the suspended particulate matter(SPM) concentration was determined by drying and re-weighingthe material retained on the filters. The total particulate metalconcentrations were determined by digesting the dry materialand filter in 120-mL polycarbonate vials according to the aqua-regia digestion method described by Simpson et al. (2004), anddiluting with deionised water (18 M�·cm, Milli-Q, Millipore)to a 10% HNO3 matrix.

The salinity of waters collected in Survey 1 was determinedusing the micro-scale chlorinity titration method of Grasshoffet al. (1983). A water quality analyser multi-probe (YSI 6600,Springs, OH) was used in Survey 2 to determine water tempera-ture, salinity, pH and dissolved oxygen concentrations, whichwas calibrated immediately before use following the manu-facturer’s instructions. Both methods used to measure salinitywere crosschecked using a laboratory seawater sample of knownsalinity.

Sediment manipulations used to investigatemetal partitioningAt low tide, sediment samples (∼1 kg) were collected fromthe upper 10 cm of sediment at intertidal sites in the mid- andsouthern Narrows to investigate metal partitioning in the labora-tory. The sediment was collected with an acid-washed Teflonspatula and placed into double zip-lock bags with the airsqueezed out, and transferred to the laboratory in ice boxes. In thelaboratory, the sediment was stored in a refrigerator (2–5◦C) forno more than 2 weeks before use to minimise physico-chemicalchanges. An aqua-regia digestion of these sediments was per-formed to measure particulate-bound metals at the differentlocations (Simpson et al. 2004).

The relationship between water pH and metal mobilisationfrom sediment-bound phases was investigated using sedimentscollected from the mid-Narrows and waters of pH 4 to 9.The sed-iment was homogenised with a clean plastic spatula in a nitrogen-filled glove box. Volumes of 0.5 L of filtered seawater (StanwellPark beach, New South Wales) were adjusted to pre-determinedpHs, by spiking with 1–10% NaOH or 1–10% HNO3, so that theoverlying water was at the desired pH 24 h after mixing the sed-iment with the water. The seawater was added to wet sediments10 g L−1 (dry weight), shaken vigorously for 30 s, and allowedto stand at room temperature for 24 h. After 24 h, a sub-sampleof water was taken to measure pH, and the remaining water wasfiltered and acidified to 2 mL HNO3 L−1 (Merck, Tracepur), andretained for measurement of dissolved metals. A large water pHrange and high sediment-to-water ratio were chosen to investi-gate metal mobilisation to reflect the conditions that may occurin areas such as sediment porewaters, organism burrows, saltmarshes and areas with hotspots of acid-sulfate soils.

Chemical analysesThe metal concentrations in filtered samples and particulatedigests were measured using inductively coupled plasma-atomicemission spectrometry (ICP-AES) (Spectroflame EOP, SpectroAnalytical Instruments, GmbH, Kleve, Germany) calibratedwith matrix-matched standards (QCD Analysts, Eaglewood, FL,USA). As concentrations of dissolved cadmium, copper, nickeland zinc were below the detection limit of the ICP-AES in the fil-tered seawater samples, a dithiocarbamate complexation/solventextraction method (Hatje et al. 2003) was used to concentratemetals and remove salt before analysis. The extraction solu-tion (modified from Magnusson andWesterlund 1981) combinedsodium bicarbonate buffer, ammonium pyrrolidine dithiocarba-mate (APDC) and diethyldithiocarbamic acid (DDC) complex-ing reagents, which were extracted into 1,1,1-trichloroethane inplace of Freon (1,1,2-trichlorotrifluoroethane) (Apte and Gunn1987; Hatje et al. 2003). The metals were back-extracted into1 mL of concentrated nitric acid (Merck Tracepur), diluted toa final volume of 10 mL with deionised water (18 M�·cm,Milli-Q, Millipore), and analysed by graphite furnace-atomicabsorption spectrometry (GF-AAS) (Perkin Elmer, Melbourne,4100ZL) using Zeeman effect background correction and oper-ating conditions recommended by the manufacturer.

For purposes of QA/QC, method blanks, method duplicatesand spike recoveries were performed on at least 10% of the fil-tered samples. The certified reference material (CRM) seawater

Dissolved metals in Port Curtis Estuary, Australia Marine and Freshwater Research 173

Survey 1

Salinity (%)33.6–34.534.6–35.235.3–3636.1–36.836.7–37.537.6–39

Survey 2

Fig. 2. Measured salinity of Port Curtis and surrounding waters.

(CASS-4) and particulate (PACS-2) (National Research Councilof Canada) were processed as part of the routine quality controlprocedures.

Data analysisDifferences in the mean concentrations of dissolved and sus-pended solid-bound metals, pH, salinity and suspended solidsmeasured in the Port Curtis, the Narrows and the Keppel Bay,Facing Island and Rodd’s Bay offshore sampling zones wereexamined using a one-way ANOVA. For calculation purposes,samples that had metal concentrations below the detection limitwere assigned a value of half the detection limit. An analysisof variance in the mean concentrations of dissolved and par-ticulate metals in Port Curtis and the Narrows between Survey1 (December 2003) and Survey 2 (December 2004) was alsoperformed. Student–Newman–Keuls test (SNK) was used as apost-hoc procedure to test for multiple comparisons upon means.

ResultsGeneral water quality parametersThe Port Curtis system was dominated by highly saline waters(Fig. 2). The salinities were significantly (P < 0.05) higher inthe TN (mean ± s.d. = 38 ± 1 PSU) and PC (37 ± 1 PSU) sam-pling zones than in the KB (36 ± 1 PSU), RB (35 ± 1 PSU) andFI (35 ± 1 PSU) offshore sampling zones in Survey 1. The onlysalinities that were lower than typical oceanic levels were mea-sured in the Calliope River (22.8 PSU) and the Fitzroy Riveradjacent to the city of Rockhampton (<0.2–11.5 PSU).

During the targeted sampling (Survey 2), the pH of waters inPort Curtis, the Narrows, and adjacent inlets and Rivers rangedfrom 7.73 to 8.24 (Fig. 3). The pH exhibited broad minima inthe range pH 7.95–8.05 between theYarwun trade waste outlet atFisherman’s Landing in the inner harbour and Ramsay’s Crossingin the Narrows. The lowest water pHs measured in this sur-vey occurred in the Upper Fitzroy (pH 7.73–8.12) and CalliopeRivers (pH 7.95), that were the sites with the lowest salinities.

The concentrations of SPM in this study were in the range<0.5–89 mg L−1 (data not shown). They were highest in theFitzroy River near the entrance to Keppel Bay, and decreased inthe order: Keppel Bay > the Narrows > Port Curtis Harbour.The higher SPM measured in waters of the lower Fitzroy estuary

7.9

8.0

8.1

8.2

8.3

0 20 40 60 80

Distance from harbour entrance (km)

pH

FLSE

RC

NE

Fig. 3. Measured pH of Port Curtis and surrounding waters duringSurvey 2, where the symbols FL, SE, RC and NE refer to the locations ofFisherman’s Landing, the southern entrance to the Narrows, Ramsay’s Cross-ing and the northern entrance to the Narrows respectively. The entrance tothe harbour was defined as site B1.15 (Survey 2).

and Keppel Bay (15–89 mg L−1) were the result of the outflowsof the Fitzroy River, which is known to have high and variableSPM concentrations (20–1000 mg L−1) (Ford et al. 2005).

Dissolved metal concentrationsThe analytical detection limits of dissolved cadmium, copper,nickel and zinc were 1, 19, 48 and 31 ng L−1 (parts per trillion),respectively, and were similar to other ultratrace metal studies(Apte and Day 1998; Mackey et al. 2002). Concentrations ofmetals measured in the reference material CASS-4 were within86 to 99% of the certified value for each metal, and spike recov-eries were in the range 88–102%. The mean percentage relativestandard deviations (RSDs) of site and method duplicates weretypically 2–15%.

The concentrations of dissolved metals measured in the twosurveys are shown in Tables 1 and 2, and also in Figs 4 and5. The metal concentrations were generally lowest at offshoresites, particularly north-east of Facing Island, where the concen-trations ranged from <1 to 1.5, <19 to 68, 120 to 160 and <31to 130 ng L−1, for Cd, Cu, Ni and Zn, respectively. There weredistinct gradients in trace metal concentrations with the highest

174 Marine and Freshwater Research B. M. Angel et al.

Table 1. Dissolved (<0.45 µm) and particulate metal concentrations in Port Curtis Estuary and surrounding waters in Survey 1

Site Area Dissolved metals (ng L−1) Suspended particulate metals (µg g−1)

Cd Cu Mn Ni Zn Al Mn Fe Cu Zn

The Narrows and Port Curtis (Keppel Bay → Rodd’s Bay)A1.1 3.0 380 <200 460 61 7000 420 12 000 13 26A1.2 3.5 430 <200 500 57 7000 400 12 000 11 22A1.3 TN Keppel Bay 6.5 520 <200 610 150 4300 240 7000 7.7 19A1.4 TN Northern Narrows 3.5 540 <200 700 55 6800 360 9500 8.8 26A1.5 TN 5.8 580 1400 880 100 7800 460 9700 9.2 32A1.6 TN 5.8 560 3400 910 76 8400 560 11 000 5.8 44A1.7 TN Ramsay’s Crossing 4.0 550 7000 840 130 8500 640 14 000 9.4 43A1.8 TN 3.5 510 5200 780 92 9600 670 16 000 14 55A1.9 TN 3.0 610 2900 620 91 8400 670 13 000 17 57A1.10 TN Southern Narrows 3.0 640 1600 520 140 9300 940 15 000 19 55A1.11 PC 5.0 620 1500 470 210 8400 950 14 000 14 61A1.12 PC Fisherman’s Landing 2.0 550 740 330 130 7800 750 12 000 16 30A1.13 PC Mid-harbour 15 610 1100 360 200 8700 880 15 000 22 56A1.14 PC 4.5 520 620 330 130 9200 920 15 000 24 46A1.15 PC 4.5 500 <200 350 160 6300 600 11 000 15 29A1.16 PC Outer-harbour 15 460 <200 310 230 4500 480 8200 10 26A1.17 PC 6.3 410 <200 280 130 8800 770 14 000 20 61A1.18 15 300 <200 230 190 2200 190 3600 46 27A1.19 2.0 250 <200 200 70 6300 470 8600 14 88A1.20 3.8 190 360 200 64 2400 170 2900 3.0 34A1.21 Rodd’s Bay 3.0 220 490 190 57 NA NA NA NA NA

Rodd’s Bay → offshoreA2.1 RB Rodd’s Bay 10 66 – 170 140 NA NA NA NA NAA2.2 RB 1.8 84 – 140 61 6400 350 7500 <3 71A2.3 RB 3.0 70 – 120 92 2600 130 2500 16 19A2.4 RB <1.5 69 – 150 42 500 90 600 4.2 53A2.5 RB 3.0 60 – 110 95 600 90 730 <3 44A2.6 RB <1.5 41 – 120 68 370 50 510 <3 51A2.7 RB Offshore 2.0 39 – 160 140 NA NA NA NA NAA2.8 RB <1.5 63 – 160 41 NA NA NA NA NA

Facing Island → offshoreA3.1 Facing Island <1.5 120 – 170 61 NA NA NA NA NAA3.2 FI 1.5 68 – 150 <31 NA NA NA NA NAA3.3 FI 1.5 38 – 150 41 NA NA NA NA NAA3.4 FI <1.5 51 – 120 130 230 160 400 <3 <3A3.5 FI <1.5 19 – 140 <31 420 280 640 <3 <3A3.6 FI <1.5 <19 – 130 37 NA NA NA NA NAA3.7 FI <1.5 30 – 140 67 NA NA NA NA NAA3.8 FI 1.5 41 – 140 43 NA NA NA NA NAA3.9 FI Offshore <1.5 22 – 140 <31 NA NA NA NA NAA3.10 FI <1.5 35 – 160 <31 NA NA NA NA NA

Keppel Bay → offshoreA4.1 Keppel Bay 3.0 430 – 540 48 9300 410 14 000 14 31A4.2 KB 3.0 370 – 440 64 4600 230 7100 7.5 26A4.3 KB 2.0 270 – 340 46 8000 510 12 000 10 68A4.4 KB <1.5 190 – 250 82 8300 460 11 000 10 82A4.5 KB <1.5 170 – 260 <31 7300 400 7600 7.9 80A4.6 KB <1.5 160 – 260 <31 4200 460 4500 6.6 110A4.7 KB <1.5 120 – 190 <31 1700 360 2300 <3 100A4.8 KB Offshore beyond 4.0 90 – 170 100 420 80 510 <3 18A4.9 KB Great Keppel <1.5 60 – 160 <31 1700 260 2000 <3 69A4.10 KB Island <1.5 70 – 180 33 NA NA NA NA NA

Dissolved metals in Port Curtis Estuary, Australia Marine and Freshwater Research 175

Tab

le2.

Dis

solv

ed(<

0.45

µm

)an

dpa

rtic

ulat

em

etal

conc

entr

atio

nsin

Por

tC

urti

ses

tuar

yan

dsu

rrou

ndin

gw

ater

sin

Surv

ey2

PC

,Por

tCur

tis;

TN

,The

Nar

row

s;–,

data

notm

easu

red

Sit

eA

rea

Dis

solv

edm

etal

s(n

gL

−1)

Sus

pend

edpa

rtic

ulat

em

etal

s(µ

gg−

1)

Sal

init

y(‰

)pH

Cd

Cu

Mn

Ni

Zn

Al

Mn

FeC

uZ

n

The

Nar

row

san

dP

ort

Cur

tis

(Kep

pelB

ay→

the

mid

-har

bour

)B

1.1

TN

Kep

pelB

ay9.

633

049

038

076

5200

250

7800

8.8

6935

.98.

20B

1.2

TN

Nor

ther

nN

arro

ws

3.1

350

<20

039

079

––

––

–35

.9–

↓B

1.3

TN

4.1

400

1700

530

6144

0028

062

007.

375

36.3

8.15

B1.

4T

N38

450

3100

640

89–

––

––

36.7

8.06

B1.

5T

NR

amsa

y’s

Cro

ssin

g4.

847

039

0070

087

4700

290

6900

8.1

7537

.18.

09↓

B1.

6T

N15

560

6400

800

83–

––

––

37.5

8.01

B1.

7T

N11

610

4600

800

180

5800

410

9700

1168

37.9

8.04

B1.

8T

NS

outh

ern

Nar

row

s12

600

2800

700

100

––

––

–37

.77.

95↓

B1.

9P

C5.

464

021

0053

017

042

0033

068

0011

7737

.38.

01B

1.10

PC

Fish

erm

an’s

Lan

ding

1364

014

0048

015

0–

––

––

37.1

7.99

B1.

11P

CM

id-h

arbo

ur6.

465

056

037

026

049

0039

086

0013

7836

.7–

↓B

1.12

PC

1258

040

030

031

046

0030

076

0011

7436

.58.

10B

1.13

PC

4.2

520

<20

028

015

039

0028

063

0012

7836

.38.

16B

1.14

PC

Out

er-h

arbo

ur2.

327

0<

200

200

94–

––

––

35.8

8.16

B1.

15P

C2.

318

0<

200

150

62–

––

––

34.7

8.16

Cre

eks

and

inle

tsin

dire

ctio

n:P

ort

Cur

tis→

the

Nar

row

sB

2.1

Boa

tCre

ek5.

677

0–

480

140

––

––

–36

.28.

01B

2.2

Cal

iope

15.

667

0–

330

340

5300

370

1000

017

7836

.18.

08B

2.3

Cal

iope

29.

273

069

0043

050

0–

––

––

22.8

7.95

B2.

4Fi

sher

man

sL

andi

ng1

4.6

740

–41

022

0–

––

––

–8.

12B

2.5

Fish

erm

ans

Lan

ding

24.

171

0–

450

170

4800

470

8800

1472

37.1

8.12

B2.

6G

raha

m’s

Cre

ek1

3.9

630

–51

094

––

––

–37

.48.

11B

2.7

Gra

ham

’sC

reek

24.

263

0–

600

140

––

––

–38

.48.

00B

2.8

Targ

inie

Cre

ek5.

561

019

000

640

120

––

––

–38

.17.

99B

2.9

Bla

cksw

anC

reek

4.1

560

2500

790

84–

––

––

37.8

8.11

Del

taof

inle

tsin

dire

ctio

n:no

rthe

rnen

dof

the

Nar

row

s→m

outh

ofF

itzr

oyes

tuar

yB

3.1

Div

isio

nPo

int

5.2

430

–67

012

045

0030

066

008.

174

36.8

8.17

B3.

2C

onno

rsC

reek

1644

0–

660

93–

––

––

36.4

8.12

B3.

3Po

rtA

lma

6.8

530

–76

090

5600

260

8100

1267

36.3

8.24

B3.

4C

asua

rina

17.

054

0–

710

260

7400

430

1300

015

5536

.28.

18B

3.5

Cas

uari

na2

2057

0–

770

120

––

––

–36

.48.

12B

3.5

Cat

tle

Poin

t8.

841

0–

470

120

6100

260

9000

1071

35.9

8.18

Fit

zroy

Riv

er/e

stua

ryan

dD

eeR

iver

sam

ples

:in

dire

ctio

nri

ver→

estu

ary

B4.

1Fi

tzro

yR

iver

14.

112

00–

1600

360

8500

2500

1500

022

160

11.9

7.73

B4.

2Fi

tzro

yR

iver

24.

313

00–

1800

580

––

––

–11

.07.

80B

4.3

Fitz

roy

Riv

er3

2.0

1200

–14

0014

010

000

2700

1800

029

230

0.2

8.12

B4.

4Fi

tzro

yes

tuar

y1

7.2

690

–11

0096

6700

440

1100

013

5536

.58.

11B

4.5

Fitz

roy

estu

ary

28.

965

0–

980

140

––

––

–36

.68.

12B

4.6

Dee

Riv

er22

000

598

000

0–

6500

02

992

000

––

––

––

1.7

176 Marine and Freshwater Research B. M. Angel et al.

0

200

400

600

800(a) (b)

(c) (d )

0 20 40 60 80

Distance from harbour entrance (km)

0 20 40 60 80

Distance from harbour entrance (km)

0 20 40 60 80

Distance from harbour entrance (km)

0 20 40 60 80

Distance from harbour entrance (km)

Dis

solv

ed C

u (n

g L�

1 )

FL SE

RCNE

0

2000

4000

6000

8000

Dis

solv

ed M

n (n

g L�

1 )

FL

SEN

RC NEN

0

200

400

600

800

1000

Dis

solv

ed N

i (ng

L�

1 )

FL

SERC

NE

0

200

400

Dis

solv

ed Z

n (n

g L�

1 )

FL

SE RC NE

Survey 1

Survey 2

Survey 1

Survey 2

Survey 1

Survey 2

Survey 1

Survey 2

Fig. 4. The concentrations of dissolved copper (��), manganese (•◦), nickel (♦�), and zinc (��) measuredalong the transect direct from the outer harbour of Port Curtis into the Narrows during Survey 1 (�•��) andSurvey 2 (�©♦�). The entrance to the harbour was defined as site A1.17 (Survey 1) and site B1.15 (Survey2). The symbols FL, SEN, RC and NEN refer to the locations of Fisherman’s Landing, the southern entrance tothe Narrows, Ramsay’s Crossing and the northern entrance to the Narrows, respectively.

concentrations being measured closest to Port Curtis and theNarrows.

The dissolved cadmium concentrations measured in thewaters of Port Curtis, the Narrows and the multiple inletsbetween the Fitzroy River and the northern mouth of the Narrowswere in the range <1–38 ng L−1, and did not exhibit an observ-able spatial trend (Table 1). In Survey 1, the mean dissolvedcadmium concentrations measured in Port Curtis (7 ± 5 ng L−1)and the Narrows (4 ± 1 ng L−1) were significantly (P < 0.05)higher than those measured in the offshore sampling zones(Fig. 1) of Keppel Bay (2 ± 1 ng L−1), Rodd’s Bay (3 ± 3 ng L−1)and Facing Island (1 ± 0 ng L−1) (Table 2).

Dissolved copper and nickel concentrations ranged between270–740 and 120–910 ng L−1, respectively (Tables 1, 2;Figs 5, 6). In Survey 2, the concentrations of dissolved cop-per and nickel measured in inlets adjacent to the Narrowswere similar to those in the Narrows, which may indicate therewas not a single point-source of copper and nickel in theseinlets at the time of this survey. Similar trends in dissolvedcopper and nickel were measured in Port Curtis Harbour andthe Narrows in Surveys 1 and 2. The mean dissolved cop-per concentrations measured in Port Curtis in Surveys 1 and2 (530 ± 80 ng L−1 and 500 ± 200 ng L−1, respectively) and theNarrows (560 ± 40 ng L−1 and 500 ± 100 ng L−1, respectively)were not significantly (P < 0.05) different. However, the meandissolved nickel concentrations measured in the Narrows in

Surveys 1 and 2 (730 ± 140 ng L−1 and 600 ± 200 ng L−1)were nearly twice the concentrations found in Port Curtis(350 ± 60 ng L−1 and 300 ± 100 ng L−1).

In Survey 1, the dissolved copper concentrations measured inthe waters of Port Curtis and the Narrows were above 500 ng L−1

over a broad area, and exhibited a maximum of 640 ng L−1

in the southern Narrows. In Survey 2, the dissolved copperconcentration measured along this transect exhibited a more pro-nounced trend, with a maximum concentration of 650 ng L−1, inthe area of the inner harbour between Fisherman’s Landing andthe southern entrance to the Narrows (Figs 4, 5).

Dissolved nickel and manganese measured along the tran-sect through Port Curtis and the Narrows exhibited similarpronounced concentration maxima in both surveys (Figs 4, 5).The peak dissolved nickel concentrations measured along thistransect were 910 ng L−1 (situated ∼6 km north of Ramsay’sCrossing) and 800 ng L−1 (situated ∼2 km south of Ramsay’sCrossing) during Surveys 1 and 2, respectively. Correspond-ingly, the peak dissolved manganese concentrations in Surveys 1and 2 were 7000 (situated ∼2 km north of Ramsay’s Cross-ing) and 6400 ng L−1 (situated ∼2 km south of Ramsay’sCrossing), respectively. The dissolved manganese concentrationmeasured at one location in the Targinie inlet in the southernNarrows was 19 000 ng L−1, which was nearly three times higherthan the maximum concentration measured in the Narrows(6400 ng L−1).

Dissolved metals in Port Curtis Estuary, Australia Marine and Freshwater Research 177

0–20

Survey 1 Survey 2

Survey 1 Survey 2

Survey 1 Survey 2

Copper (ng L�1)

21–4041–7071–100101–150151–200201–300

401–600

601–800

801–1000

1001–1600

301–400

0–20

Nickel (ng L�1)

21–4041–7071–100101–150151–200201–300

401–600

601–800

801–1000

1001–1600

301–400

0–20

Zinc (ng L�1)

21–4041 –7071–100101–150151–200201–300

401–600601–800

801–1000

1001–1600

301–400

Fig. 5. Dissolved copper, nickel and zinc concentrations (ng L−1) in Port Curtis estuary and surrounding waters.

178 Marine and Freshwater Research B. M. Angel et al.

0

3000

6000

9000

12 000

15 000(a)

(b)3 4 5 6 7 8 9 10

Dis

solv

ed N

i and

Zn

(ng

L�1 )

0

400

800

1200

1600

Dis

solv

ed C

u (n

g L�

1 )

Ni

Zn

Cu

0

300

600

900

1200

6.5 7.5 8.5 9.5

pH of sediment/water mixture

Dis

solv

ed C

u, N

i and

Zn

(ng

L�1 ) Ni

Zn

Cu

Fig. 6. Dissolved (<0.45 µm) metals measured following re-suspensionand 24-h storage of central Narrows (sampling site B1.5) sediment(10 000 mg L−1) for water pH in the range (a) pH 3.0–10.0 and (b) pH6.5–9.5.

Dissolved zinc concentrations in Port Curtis Harbour and theNarrows ranged from 60 to 310 ng L−1 (Tables 1, 2; Figs 4, 5).The dissolved zinc concentration was broadly elevated in watersbetween the outer harbour of Port Curtis and the southernentrance to the Narrows in Surveys 1 and 2 (Figs 4, 5). Therewas no significant difference in the mean concentrations ofdissolved zinc measured in Port Curtis (170 ± 40 ng L−1) andthe Narrows (110 ± 30 ng L−1) during Survey 1. However, themean dissolved zinc concentration measured in Port Curtis(170 ± 90 ng L−1) was significantly (P < 0.05) higher than theNarrows (100 ± 40 ng L−1) during the Survey 2. In Survey 2,the concentrations of dissolved zinc measured in the Narrowsand in adjacent inlets were similar, likely indicating there wasnot a single point-source of zinc in these inlets.

The dissolved copper and nickel measured in Survey 1 exhib-ited similar spatial trends between offshore sampling zones, withthe KB zone having significantly higher (P < 0.05) dissolvedconcentrations than the RB zone, and both of these zones had sig-nificantly higher (P < 0.05) dissolved concentrations than the FIoffshore zone.The higher mean dissolved copper and nickel con-centrations measured in the KB offshore sampling zone is prob-ably due to water inputs from the Fitzroy River. The higher meandissolved copper, nickel, and zinc measured in the RB offshore

sampling zone than the FI offshore sampling zone is likely dueto tidal movement of waters in Port Curtis towards Rodd’s Bay.

The site sampled in Boat Creek was in close proximity to theClinton Coal Wharf and Fisherman’s Landing in the inner har-bour, and had marginally higher dissolved copper (770 ng L−1)and the Calliope River had marginally higher dissolved copper(670–730 ng L−1) and zinc (340–500 ng L−1) than the maximumconcentration of these metals measured in the transect throughwaters of Port Curtis and the Narrows (Table 2).The highest con-centrations of dissolved copper, nickel and zinc were measuredin the Fitzroy River adjacent to the city of Rockhampton dur-ing Survey 2, and ranged from 1200 to 1410, 1430 to 1760 and140 to 580 ng L−1, respectively (Table 2; Fig. 5), possibly dueto anthropogenic sources from the city and surrounding agri-culture and geological sources (such as derelict mines in theregion) (Moss and Costanzo 1998; Jones et al. 2000; Rolfe et al.2004). The concentrations of dissolved copper, nickel and zincin the lower estuary near the entrance to Keppel Bay were lowerthan upstream and ranged from 650 to 690, 980 to 1010 and100 to 140 ng L−1, respectively, probably due to dilution fromtidal mixing (Baeyens et al. 2005), and scavenging of dissolvedmetals by SPM in the lower estuary (Sañudo-Wilhelmy et al.1996; Teasdale et al. 2003; Ford et al. 2005).

Metals associated with SPMThe concentrations of Al, Fe and Mn bound to SPM mea-sured in waters along the transect through Port Curtis and theNarrows ranged from 4400 to 9600, 7050 to 16200 and 190to 950 µg g−1, respectively (Tables 1 and 2). The concentra-tions of suspended aluminium and iron (SPM-Al and SPM-Fe)did not exhibit a strong trend in Port Curtis and the Narrows,although SPM-Mn exhibited a broad concentration maximumin the inner harbour and the southern Narrows (Fig. 4). TheSPM-Cu and SPM-Zn concentrations were in the range of<3–50 and <3–230 µg g−1, and did not exhibit strong trendsin either survey. The SPM-Cd and SPM-Ni were below the ana-lytical detection limit. The concentrations of SPM-bound metalswere generally lower in the offshore sampling zones and oftenbelow detection limits.

The release of sediment-bound metals with changesin seawater pHResults from aqua-regia metal digests of benthic sediments col-lected in Port Curtis and the Narrows indicated the centralNarrows had significantly (P < 0.05) higher concentrations ofparticulate copper and nickel than the mid-harbour. Concentra-tions in the central Narrows were approximately double thosein the mid-harbour (data not shown). The concentrations ofdissolved copper, nickel and zinc released from the sedimentsre-suspended in seawater of varying pH are shown in Fig. 6. Asexpected, the highest concentrations of dissolved copper, nickeland zinc were measured following resuspension at pH 4 (Fig. 6a).The release of metals in resuspension tests in the pH 7.75–8.25 range (Fig. 6b) were more representative of waters in PortCurtis and the Narrows (pH 7.95–8.20), although lower pHs mayoccur in the mud flats adjacent to the Narrows. Dissolved nickeland zinc increased by 58% and 34%, respectively, as the pHdecreased from pH 8.25 to pH 7.75. In contrast, dissolved copper

Dissolved metals in Port Curtis Estuary, Australia Marine and Freshwater Research 179

decreased as pH decreased from pH 8.25 to pH 7.75, and the cop-per concentration only increased significantly (P < 0.05) abovethat measured for pH 8.25 waters in re-suspension treatmentsthat employed waters with a pH ≤5.

Discussion

Port Curtis is different to many estuaries, as it is connected to theocean at multiple locations (the Narrows and openings betweenCurtis and Facing Islands), and has highly variable, spasmodicinputs of freshwater due to the wet season/dry season climatedynamics of the catchment area (Ford et al. 2005; Jones et al.2005). The hydrodynamics of water movement and the generalwater quality parameters (pH, salinity, SPM) in Port Curtis, theNarrows and the Fitzroy River may have a significant influenceon the concentrations of metals. A recent hydrodynamic modelpredicted that water in Port Curtis undergoes tidal movementover large distances, but little net displacement occurs, as thewater returns close to its original location (Herzfeld et al. 2004).The model proposed a flushing duration in Port Curtis of 22–26days, although poor tidal mixing in some protected inlets andbays of Port Curtis and the Narrows suggests that these watershave longer flushing durations. The model did not include allof the Narrows and it was not entirely clear whether hydro-dynamic forces predominantly direct waters from Port Curtisnorth through the Narrows into Keppel Bay or from Keppel Baysouth through the Narrows into Port Curtis. The direction ofwater flow in the Narrows fluctuates (north-west or south-east)through the tidal cycle making it difficult to predict metal dis-persion from potential sources, such as the point sources fromtheYarwun trade waste outlet and the Calliope Power Station, ordiffuse sources from urban run-off and leaching from sediments.

The broad pH minimum measured between the inner har-bour and Ramsay’s Crossing in the Narrows was in agreementwith past pH measurements (Apte et al. 2006). The positionof the pH minimum was in close proximity to the entrance ofthe Calliope River and the Yarwun trade waste outlet, and boththese areas may potentially contribute water of lower pH intoPort Curtis. The lack of a decrease in salinity indicates thatfreshwater inputs were minimal during the sampling period. Thedischarges from theYarwunTrade Waste outlet were not believedto be acidic (L. Anderson, Central Queensland University, pers.comm.), so the relatively small outflows (110 kg Cu year−1,410 kg Zn year−1) of these waters were unlikely to have a signif-icant influence on the water pH in Port Curtis Harbour (NationalPollution Inventory 2008).

The foreshore of the Narrows is predominantly lined withmangrove mud flats and salt marshes. The sediment surface areathat comes into contact with water is relatively high because ofthe large tidal range (4–5 m, WBM Oceanics 1992) and flatnessof the adjacent sediments. The water depth is also quite shallowin the Narrows and there is a relatively high ratio between thesurface area of sediment that comes into contact with water andthe volume of water. Several studies have reported lower pHs inwaters lined with mangrove foreshores due to factors that resultin the release of H+ into waters including mangrove systemshaving faster rates of diagenesis and organism-facilitated aero-bic oxidation of organic matter because of the larger volumesof organic carbon (Van Cappellen and Wang 1996; Clark et al.

1998; Kristensen 2000; Otero and Macias 2002). The lower pHof waters in the Narrows may also be generated by adjacentacid-sulfate soils (ASS), which are common along the east coastof Queensland and northern New South Wales, with hotspots ofASS activity identified in the vicinity of the Narrows (Ross 2002;Powell and Martens 2005). Mud flats and salt marshes often havehigher water temperatures than the adjacent water column thatmay allow greater rates of chemically driven oxidation (e.g. airoxidation of sulfides), resulting in lower water pH (Otero andMacias 2002).

The salinity of the waters of Port Curtis and the Narrowswas in the range 33 to 39 PSU, and did not decrease at sitesin the inner harbour and the Narrows where several creeks andrivers may potentially input fresh water into the system. On thecontrary, the salinity was higher in the estuary near the out-flows of these inlets than the offshore waters. Typical estuarinewaters have salinities that increase as the estuary approachesthe sea and are ∼35 PSU at the seawater end-member (Sañudo-Wilhelmy et al. 1996; Hatje et al. 2003). The chemical analysisof major ions that may contribute to salinity indicated that thespatial trends in dissolved Ca, Mg, K and Na were the same assalinity (data not shown), indicating no particular species wereresponsible for the increased salinity.

The mangrove mud flats and salt marshes along the foreshoreof the Narrows and inner harbour are subject to evaporation(Saenger 1995). It is likely that the hypersaline conditions mea-sured in the Narrows and the inner harbour were the resultof evaporative water losses. Average daily rainfall measure-ments in the catchment of Port Curtis (Calliope, Mount Larconand Gladstone Airport stations) in the 30 days before and dur-ing the two surveys were <5 mm, and most days had <1 mm,indicating low freshwater inputs from rainfall into the estu-ary (http://www.bom.gov.au/climate/cdo/about/sitedata.shtml,accessed 14 February 2008). Hypersaline conditions may be typ-ical of many tropical Australian estuaries and may be assisted byevapotranspiration occurring in saline tolerant vegetation (waterremoved by plants leading to concentration of salt in estuarinesystem) (Wolanski 1986; Largier et al. 1997; Ridd and Stieglitz2002).

Considering the ecological importance of theWorld Heritage-listed GBRMP, there was surprisingly few data on the concen-trations of trace metals in these waters. The increased industrialand shipping activity in this region of Australia also meansthat it is important to have knowledge of natural (baseline)concentrations for future comparison. A study by Denton andBurdon-Jones (1986) that used Chelex-100 extraction and analy-sis of cadmium, copper, nickel and zinc in waters of the GBRMPreported concentrations of <10–60, 110–240, 60–160 and30–350 ng L−1, respectively, but the results had high replicatevariability and the method had poor extraction efficiencies.

The lowest concentrations of trace metals were measuredin the FI offshore sampling zone. This zone was consideredthe most representative of regional coastal water concentra-tions because Facing Island restricts water flow and thus, metaltransfer from the harbour into these waters. The mean offshoreconcentrations of trace metals measured in this study weresimilar to those measured in offshore sites in Northern andEastern Australia and the Western Pacific, and may be consid-ered metal depleted relative to waters in the northern Atlantic,

180 Marine and Freshwater Research B. M. Angel et al.

which receive significantly greater inputs from atmospheric andriverine sources (Kremling and Pohl 1989; Bruland et al. 1994;Jickells 1995; Apte and Day 1998; Williams et al. 1998; Mackeyet al. 2002) (Table 3). The similar concentrations of trace metalconcentrations measured in offshore waters around Northern andEastern Australia indicate that the offshore metal concentrationsmeasured in the current study are likely to be representative ofthose found in the GBRMP.The current study achieved relativelylow detection limits and excellent quality control results, and toour knowledge, is the most recent and accurate dataset of tracemetal concentrations representative of those found in waters ofthe GBRMP.

The dissolved metal concentrations in Port Curtis didnot exceed the current Australia and New Zealand waterquality guideline trigger values for ecosystem protection(ANZECC/ARMCANZ 2000) (Table 3). The highest concen-trations of dissolved Cd, Cu, Ni and Zn measured in Port Curtisand the Narrows were 40, 740, 910 and 310 ng L−1, respectively.These concentrations are at the lower end of the concentrationrange of these metals measured in industrialised harbours aroundthe world, indicating that Port Curtis was not contaminated tothe same degree (Table 3). The level of metal enrichment in PortCurtis relative to regional concentrations was investigated bycomparing mean metal concentrations from PC to those fromthe FI offshore sampling zone. The ratios of Cd, Cu, Ni and Znmeasured in PC compared with the FI offshore sampling zonewere 2.4, 6.9, 3.0 and 1.7, respectively. The ratio of dissolvedcopper in Port Curtis compared with offshore sampling zoneswas larger than the other metals, indicating greater enrichmentin the harbour. The copper concentrations were also higher thanthose measured in previous studies of pristine estuaries, such asBathurst Harbour, Tasmania or sea lochs in Scotland (Hall et al.1996; Mackey et al. 1996).

The presence of a strong spatial separation of concentrationmaxima for dissolved copper, manganese, nickel and zinc in thisstudy is likely to indicate that there were multiple metal sourcesor parameters influencing metal concentrations. The dissolvedcopper and zinc concentration maxima were located in the innerharbour and southern Narrows. The foreshore of these waterswas lined with urban, commercial and industrial centres, whichmeans there were likely to be multiple sources of metals (Comberet al. 1995; Laslett and Balls 1995; Owens and Balls 1997; Hatjeet al. 2003). For instance, the National Pollution Inventory web-site of the Commonwealth of Australia reports that copper andzinc are discharged into waters of the Calliope River and theYarwun trade waste outlet at Fisherman’s Landing by industriessuch as the Gladstone power station and sewage works (NationalPollution Inventory 2008). The leachate from reclaimed landcontaining fly ash from the nearby power station and antifoulingpaints from boats may also contribute to the higher concentra-tions of dissolved copper and zinc in the inner harbour (Warnkenet al. 2004; Jones et al. 2005). The dissolved concentrations ofCu, Ni and Zn exhibited low temporal variation at the Ramsay’sCrossing and Fisherman’s Landing sites over the 24-h durationof the sampling period (data not shown).

In contrast, the dissolved nickel concentration maximum waslocated in the central to northern Narrows and is not in closeproximity to anthropogenic sources. The dissolved nickel con-centration maximum of each survey was located within close

proximity (2–4 km) to the dissolved manganese maxima, andstrong correlations of 0.67 (P < 0.001) and 0.92 (P < 0.0001)were measured in Surveys 1 and 2, respectively (Fig. 7). Theclose association of nickel and manganese has been shownin several previous studies, indicating that the same processmay be responsible for the release of dissolved manganeseand nickel from soils and sediments (Laslett and Balls 1995;Sañudo-Wilhelmy and Gill 1999; Hatje et al. 2003).

Many trace metals are associated with manganese and iron(hydr)oxides in sediments. The chemically or bacterially drivenreduction of manganese and iron (hydr)oxides may result in therelease of manganese, iron and trace metals into the dissolvedphase, although iron would rapidly re-oxidise and be removedfrom the dissolved phase upon entering the overlying water(Laslett and Balls 1995; Sañudo-Wilhelmy and Gill 1999; Zhanget al. 2002; Simpson and Batley 2003). The reduction of man-ganese and iron hydr(oxides) is a natural process that has beenreported to be enhanced in warmer temperatures, when biologi-cal activity is increased and oxygen concentrations are generallylower (Laslett and Balls 1995; Otero and Macias 2002). Naturalgeochemical weathering and leaching processes may also releasetrace metals including nickel from particulate-bound forms intothe dissolved phase, and these processes are likely to be facil-itated in the Narrows because of ASS hotspots. Processes thatcause the release of nickel from sediments into the dissolvedphase are likely to have a greater influence on sources of nickelin the Narrows than Port Curtis because of the higher particu-late nickel in the sediments of the Narrows than the mid-harbourmeasured previously and in the current study (Vincente-Beckettand Shearer 2005).

The water pH may also influence the behaviour of nickelin Port Curtis and the Narrows, as shown by the increase indissolved nickel with decreasing water pH in re-suspensionexperiments (Fig. 6b). However, the results from these exper-iments were indicative of nickel release from SPM in highlyturbid conditions or in the benthic water layer, as the SPM con-centrations were ∼3 orders of magnitude higher than the SPMmeasured in the Narrows. Therefore, the results are expectedto greatly overestimate the nickel released from SPM due to thedecreased pH of water in the overlying water of the Narrows, butmay be representative of metal release in sediment porewater andthe benthic water layer of mudflats in this area.

The Narrows was unlikely to act as a significant source ofmetals to Port Curtis Harbour because the difference in con-centrations was not particularly great, and the relatively smallwater volume in the Narrows compared with the harbour. TheCalliope River may also contribute natural and/or anthropogenicmetals to the waters of this area of the harbour, although rainfallwas low before and during both surveys and it is unlikely thatoutflows from this river had a significant influence on the con-centrations during this study. The sites sampled in rivers, creeksand inlets adjacent to Port Curtis and the Narrows had similarmetal concentrations to those measured along adjacent transects,indicating that these waters were probably not acting as substan-tial metal sources at the time of the survey. It is unlikely thatatmospheric deposition contributes significant inputs of met-als to the waters sampled, as this would be expected to causehigher concentrations over a broad area (Kremling and Pohl1989; Jickells 1995).

Dissolved metals in Port Curtis Estuary, Australia Marine and Freshwater Research 181

Tab

le3.

Com

pari

son

ofdi

ssol

ved

trac

em

etal

conc

entr

atio

nsin

Por

tC

urti

sw

ith

publ

ishe

dda

tafo

rot

her

estu

arin

ean

dm

arin

ew

ater

s

Loc

atio

nM

etal

conc

entr

atio

n(n

gL

−1)

Ref

eren

ce

Cd

Cu

Ni

Zn

Port

Cur

tis,

offs

hore

<1.

5(<

1.5–

4.0)

40(<

19–8

5)15

0(1

1–19

0)40

(<30

–140

)T

his

stud

yPo

rtC

urti

sH

arbo

ur7.

0(2

.0–1

5.0)

510

(410

–620

)34

0(2

80–4

70)

170

(130

–230

)T

his

stud

yT

heN

arro

ws

8.0

(3.0

–6.5

)53

0(5

20–6

40)

650

(470

–910

)11

0(6

0–21

0)T

his

stud

yN

SW

coas

t2.

530

180

<22

Apt

ean

dD

ay(1

998)

Wes

tern

Paci

fic

Oce

an1–

101

40–2

8012

0–52

2–

Mac

key

etal

.(20

02)

Nor

thPa

cifi

cO

cean

0.3–

112

––

15–5

20B

rula

ndet

al.(

1994

)G

reat

Bar

rier

Ree

f<

10–6

011

0–24

060

–160

30–3

50D

ento

nan

dB

urdo

n-Jo

nes

(198

6)B

athu

rstH

arbo

ur,T

asm

ania

2.2

140

140

392

Mac

key

etal

.(19

96)

Nor

thA

tlan

tic

0.7

68–

–K

rem

ling

and

Pohl

(198

9)S

cott

ish

sea

loch

<10

180

20–

Hal

leta

l.(1

996)

Port

Jack

son,

Aus

tral

ia6–

104

932–

2550

175–

1610

3270

–966

0H

atje

etal

.(20

03)

Torr

esS

trai

tand

Gul

fof

Papu

a<

1–29

36–9

8694

0–46

00–

Apt

ean

dD

ay(1

998)

Port

Phi

llip

Bay

,Aus

tral

ia<

5–70

400–

630

540–

1100

250–

1050

Fabr

isan

dM

onah

an(1

995)

Nin

ees

tuar

ies,

nort

hern

Aus

tral

ia1.

4–72

150–

5500

120–

4250

<10

–11

300

Mun

ksga

ard

and

Parr

y(2

001)

Twee

des

tuar

y,U

K7–

3349

0–47

00–

430–

1900

Las

lett

and

Bal

ls(1

995)

Hum

ber

estu

ary,

UK

80–4

5018

0–10

100

2500

–12

000

3000

–20

500

Com

ber

etal

.(19

95)

Mer

sey

estu

ary,

UK

10–1

1080

0–49

5020

00–1

050

065

00–2

800

0C

ombe

ret

al.(

1995

)Ta

yes

tuar

y,S

cotl

and

<3–

8525

0–15

5025

0–90

0<

100–

3200

Ow

ens

and

Bal

ls(1

997)

Fort

hes

tuar

y,S

cotl

and

––

290–

1470

460–

1013

0L

asle

ttan

dB

alls

(199

5)S

chel

dtes

tuar

y,N

ethe

rlan

ds15

–100

750–

1800

1000

–680

010

00–1

000

0B

aeye

nset

al.(

2005

)S

anFr

anci

sco

Bay

estu

ary,

US

A22

–123

315–

2230

140–

2410

160–

1960

Sañ

udo-

Wil

helm

yet

al.(

1996

)S

ixes

tuar

ies,

Texa

s,U

SA

–10

0–32

00–

300–

1800

0B

enoi

teta

l.(1

994)

Aus

tral

ian

Gui

deli

neva

lues

(95%

prot

ecti

on)

700

1300

7000

1500

0A

NZ

EC

C/A

RM

CA

NZ

(200

0)

182 Marine and Freshwater Research B. M. Angel et al.

0

250

500

750

1000

0 2000 4000 6000 8000

Dissolved manganese (ng L�1)

Dis

solv

ed n

icke

l (ng

L�

1 )

Survey 1

Survey 2

Fig. 7. The relationship between dissolved manganese and nickel mea-sured in Port Curtis and the Narrows in Surveys 1 and 2.

Conclusions

Overall, the waters of Port Curtis and the Narrows had amarginally lower water pH and higher salinity, and had elevatedmetal concentrations compared with the surrounding coastalwaters. The concentrations measured in the GBRMP and adja-cent waters were similar to concentrations measured in regionalwaters on the north-east coast of Australia. Strong spatial sepa-rations of maxima for dissolved copper, manganese, nickel andzinc were measured, indicating that there was not one particularsource responsible for the elevated metal concentrations. Copperand zinc are likely to originate from anthropogenic sources,whereas nickel and manganese are likely to be released by nat-ural processes such as reduction of manganese (hydr)oxidesand leaching by the lower sediment and water pH. All metalconcentrations measured in waters of Port Curtis were belowregulatory guideline values. However, dissolved copper was sig-nificantly (P < 0.05) elevated above the concentrations in theadjacent coastal waters and pristine estuarine systems, and maybe of potential concern if the concentrations in Port Curtis andthe Narrows continue to increase.

AcknowledgementsThe authors acknowledge the financial support from the University ofWollongong, CSIRO and the Cooperative Research Centre for Coastal ZoneEstuary and Waterway Management. Leonie Anderson and Andrew Davis(Central Queensland University) are thanked for logistical support and coor-dination of field work, and David Spadaro (CSIRO) for assistance in thelaboratory

ReferencesAhlers, W. W., Reid, M. R., Kim, J. P., and Hunter, K. A. (1990).

Contamination-free sample collection and handling protocols for traceelements in natural waters. Marine and Freshwater Research 41,713–720. doi:10.1071/MF9900713

Andersen, L. E. (2004). Imposex: a biological effect of TBT contamina-tion in Port Curtis, Queensland. Australian Journal of Ecotoxicology 10,105–113.

Andersen, L. E., and Norton, J. H. (2001). Port Curtis mud crab shell disease.Final report. Central Qld University, Department of Primary Industry,Fisheries Research and Development Corporation.

ANZECC/ARMCANZ (2000). ‘Australian and New Zealand Guidelines forFresh and Marine Water Quality.’ (Australia and New Zealand Environ-ment and Conservation Council/Agricultural and Resource ManagementCouncil of Australia and New Zealand: Canberra, Australia.)

Apte, S. C., and Day, G. M. (1998). Dissolved metal concentrations in theTorres Strait and Gulf of Papua. Marine Pollution Bulletin 36, 298–304.doi:10.1016/S0025-326X(98)00188-X

Apte, S. C., and Gunn, A. M. (1987). Rapid determination of copper, nickel,lead and cadmium in small samples of estuarine waters by liquid/liquidextraction and electrothermal atomic absorption spectrometry. AnalyticaChimica Acta 193, 147–156. doi:10.1016/S0003-2670(00)86147-4

Apte, S., Anderson, L., Andrewartha, J., Angel, B., Shearer, D., Simpson, S.,Stauber, J., and Vicente-Beckett, V. (2006). Contaminant pathways inPort Curtis. Cooperative Research Centre for Coastal Zone, Estuary andWaterway Management. Technical Report No. 75. CRC Coastal ZoneEstuary and Waterway Management, Indooroopilly, Qld.

Apte, S. C., Batley, G. E., Szymczak, R., Rendell, P. S., Lee, R., and Waite,T. D. (1998). Baseline trace metal concentrations in New South Walescoastal waters. Marine and Freshwater Research 49, 203–214.

Australian Government Department of the Environment, Water, Heritageand the Arts (2008). National Pollution Inventory. Tracking emissionsacross Australia. Australian Government Department of the Environ-ment, Water, Heritage and the Arts. Available at http://www.npi.gov.au[accessed 22 March 2008].

Baeyens, W., Leermakers, M., De Gieter, M., Nguyen, H. L., Parmentier, K.,Panutrakul, S., and Elskens, M. (2005). Overview of trace metal con-tamination in the Scheldt estuary and effect of regulatory measures.Hydrobiologia 540, 141–154. doi:10.1007/S10750-004-7129-4

Benoit, G., Oktay-Marshall, S. D., Cantu, H.A., Hood, E. M., Coleman, C. H.,Corapcioglu, M. G., and Santschi, P. H. (1994). Partitioning of Cu,Pb, Ag, Zn, Fe, Al, and Mn between filter-retained particles, colloidsand solution in six Texas estuaries. Marine Chemistry 45, 307–336.doi:10.1016/0304-4203(94)90076-0

Bruland, K. W., Orians, K. J., and Cowen, J. P. (1994). Reactive trace metalsin the stratified central North Pacific. Geochimica et Cosmochimica Acta58, 3171–3182. doi:10.1016/0016-7037(94)90044-2

Clark, M. W., McConchie, D., Lewis, D. W., and Saenger, P. (1998).Redox stratification and heavy metal partitioning inAvicennia-dominatedmangrove sediments: a geochemical model. Chemical Geology 149,147–171. doi:10.1016/S0009-2541(98)00034-5

Comber, S. D. W., Gunn, A. M., and Whalley, C. (1995). Comparison of thepartitioning of trace metals in the Humber and Mersey Estuaries. MarinePollution Bulletin 30, 851–860. doi:10.1016/0025-326X(95)00092-2

Denton, G. R. W., and Burdon-Jones, C. (1986). Trace metals in surfacewaters from the Great Barrier Reef. Marine Pollution Bulletin 17, 96–98.doi:10.1016/0025-326X(86)90412-1

Fabris, G. L., and Monahan, C. A. (1995). Characterisation of toxicants inwaters from Port Phillip Bay: metals. CSIRO INRE Port Phillip BayEnvironmental Study Technical Report No. 18, Melbourne.

Ford, P., Tillman, P., Robson, B., and Webster, I. T. (2005). Organic carbondeliveries and their flow related dynamics in the Fitzroy estuary. MarinePollution Bulletin 51, 119–127. doi:10.1016/J.MARPOLBUL.2004.10.019

Grasshoff, K., Ehrhardt, M., and Kremling, K. (1983). ‘Methods of SeawaterAnalysis. 2nd edn.’ (Verlag Chemie: Weinheim.)

Hall, I. R., Hydes, D. J., Statham, P. J., and Overnell, J. (1996). Dissolvedand particulate trace metals in a Scottish sea loch: an example of a pris-tine environment. Marine Pollution Bulletin 32, 846–854. doi:10.1016/S0025-326X(96)00030-6

Hatje, V., Apte, S. C., Hales, L. T., and Birch, G. F. (2003). Dissolvedtrace metal distributions in Port Jackson estuary (Sydney Harbour),Australia. Marine Pollution Bulletin 46, 719–730. doi:10.1016/S0025-326X(03)00061-4

Hatje, V., Birch, G. F., and Hill, D. M. (2001). Spatial and temporal vari-ability of particulate trace metals in Port Jackson Estuary, Australia.

Dissolved metals in Port Curtis Estuary, Australia Marine and Freshwater Research 183

Estuarine, Coastal and Shelf Science 53, 63–77. doi:10.1006/ECSS.2001.0792

Herzfeld, M., Parslow, J.,Andrewartha, J., Sakov, P., and Webster, I.T. (2004).Hydrodynamic modeling of the Port Curtis region. CRC for CoastalZone, Estuary and Waterway Management, Technical Report No. 7,Brisbane.

Jickells, T. (1995). Atmospheric inputs of metals and nutrients to theoceans: their magnitude and effects. Marine Chemistry 48, 199–214.doi:10.1016/0304-4203(95)92784-P

Jones, M.-A., Stauber, J. L.,Apte, S. C., Simpson, S. L.,Vincente-Beckett,V.,Johnson, R., and Duivenvoorden, L. (2005). A risk assessment approachto contaminants in Port Curtis, Queensland, Australia. Marine PollutionBulletin 51, 448–458. doi:10.1016/J.MARPOLBUL.2004.10.021

Jones, M., Duivenvoorden, L., Choy, S., and Moss, A. (2000). Nationalland and water resources report 3, theme 7 – catchment health, Fitzroyimplementation project. Queensland Department of Natural Resources,Brisbane.

Kremling, K., and Pohl, C. (1989). Studies on the spatial and seasonalvariability of dissolved cadmium, copper, and nickel in north-eastAtlantic surface waters. Marine Chemistry 27, 43–60. doi:10.1016/0304-4203(89)90027-3

Kristensen, E. (2000). Organic matter diagenesis at the oxic/anoxic interfacein coastal marine sediments with emphasis on the role of burrowinganimals. Hydrobiologia 426, 1–24. doi:10.1023/A:1003980226194

Largier, J. L., Hollibaugh, J. T., and Smith, S. V. (1997). Seasonally hyper-saline estuaries in Mediterranean-climate regions. Estuarine, Coastaland Shelf Science 45, 789–797. doi:10.1006/ECSS.1997.0279

Laslett, R. E., and Balls, P. W. (1995). The behaviour of dissolved Mn, Niand Zn in the Forth an industrialized partially mixed estuary. MarineChemistry 48, 311–328. doi:10.1016/0304-4203(94)00050-N

Mackey, D. J., O’Sullivan, R. J., Watson, G., and Pont, D. (2002).Trace metalsin the Western Pacific: temporal and spatial variability in the concentra-tions of Cd, Cu, Mn, and Ni. Deep-Sea Research. Part I, OceanographicResearch Papers 49, 2241–2259. doi:10.1016/S0967-0637(02)00124-3

Mackey, D. J., Butler, C. V., Carpenter, P. D., Higgins, H. W., O’Sullivan,R. J., and Paschke, R. B. (1996). Trace elements and organic matterin a pristine environment: Bathurst Harbour, South western Tasmania.The Science of the Total Environment 191, 137–151. doi:10.1016/0048-9697(96)05255-2

Magnusson, B., and Westerlund, S. (1981). Solvent extraction proce-dures combined with back-extraction for trace metal determinations byatomic absorption spectrometry. Analytica Chimica Acta 131, 63–72.doi:10.1016/S0003-2670(01)93534-2

Moss, A., and Costanzo, S. (1998). Levels of heavy metals in sediments ofQueensland rivers, estuaries, and coastal waters, Environmental Tech-nical Report No. 20. Queensland Environment Protection Authority,Brisbane.

Munksgaard, N. C., and Parry, D. L. (2001). Trace metals, arsenic and leadisotopes in dissolved and particulate phases of North Australian coastaland estuarine seawater. Marine Chemistry 75, 165–184. doi:10.1016/S0304-4203(01)00033-0

Otero, X. L., and Macias, F. (2002). Spatial and seasonal variation in heavymetals in interstitial water of salt marsh soils. Environmental Pollution120, 183–190. doi:10.1016/S0269-7491(02)00159-8

Owens, R. E., and Balls, P. W. (1997). Dissolved trace metals in the TayEstuary. Estuarine, Coastal and Shelf Science 44, 421–434. doi:10.1006/ECSS.1996.0133

Powell, B., and Martens, M. (2005). A review of acid sulfate soil impacts,actions and policies that impact on water quality in Great Barrier Reefcatchments, including a case study on remediation at East Trinity.Marine Pollution Bulletin 51, 149–164. doi:10.1016/J.MARPOLBUL.2004.10.047

Ridd, P. V., and Stieglitz, T. (2002). Dry season salinity changes in aridestuaries fringed by mangroves and saltflats. Estuarine, Coastal and ShelfScience 54, 1039–1049. doi:10.1006/ECSS.2001.0876

http://www.publish.csiro.au/journals/mfr

Rolfe, J., Alam, K., and Windle, J. (2004). Overview of the Fitzroy basinand opportunities for offset trading: Establishing the potential for offsettrading in the lower Fitzroy Rivers, research report. Central QueenslandUniversity, Emerald.

Ross, D. J. (2002). Acid Sulfate Soils, Tannum Sands to St. Lawrence, Cen-tral Queensland Coast. Department of National Resources and Mines,Rockhampton.

Saenger, P. (1995). ‘Ecology of Mangroves of Port Curtis: Regional Biogeo-graphy, Productivity and Demography. Mangroves – A Resource UnderThreat:An Issue for the Central Queensland Coast.’ (Central QueenslandUniversity: Gladstone.)

Sañudo-Wilhelmy, S. A., Rivera-Duarte, I., and Flegal, A. R. (1996). Dis-tribution of colloidal trace metals in the San Francisco Bay estuary.Geochimica et Cosmochimica Acta 60, 4933–4944. doi:10.1016/S0016-7037(96)00284-0

Sañudo-Wilhelmy, S. A., and Gill, G. A. (1999). Impact of the Clean WaterAct on the levels of toxic metals in urban estuaries: the Hudson Riverestuary revised. Environmental Science & Technology 33, 3477–3481.doi:10.1021/ES981130Z

Simpson, S. L., Angel, B. M., and Jolley, D. F. (2004). Metal equilibrationin laboratory-contaminated (spiked) sediments used for the develop-ment of whole-sediment toxicity tests. Chemosphere 54, 597–609.doi:10.1016/J.CHEMOSPHERE.2003.08.007

Simpson, S. L., and Batley, G. E. (2003). Disturbances to metal partition-ing during toxicity testing of iron (II)-rich estuarine pore waters andwhole sediments. EnvironmentalToxicology and Chemistry 22, 424–432.doi:10.1897/1551-5028(2003)022<0424:DTMPDT>2.0.CO;2

Teasdale, P. R., Apte, S. C., Ford, P. W., Batley, G. E., and Khoenken, L.(2003). Geochemical cycling and speciation of copper in waters andsediments of Macquarie Harbour, Western Tasmania. Estuarine, Coastaland Shelf Science 57, 475–487. doi:10.1016/S0272-7714(02)00381-5

Van Cappellen, P., and Wang, Y. (1996). Cycling of iron and manganese insurface sediments: A general theory for the coupled transport and reac-tion of carbon, oxygen, nitrogen, sulfur, iron, and manganese. AmericanJournal of Science 296, 197–243.

Vincente-Beckett,V., and Shearer, D. (2005). Contaminant sources in benthicsediments of Port Curtis. Milestone report IC10.1. CRC for Coastal ZoneEstuary and Waterway Management, Brisbane.

USEPA (1996). Sampling ambient water for trace metals at EPA water qual-ity criteria levels. Method 1669. US Environmental Protection Agency,Office of Water Engineering and Analysis Division, Washington, DC.

Warnken, J., Dunn, R. J. K., and Teasdale, P. R. (2004). Investigation ofrecreational boats as a source of copper at anchorage sites using time-integrated diffusive gradients in thin films and sediment measurements.Marine Pollution Bulletin 49, 833–843. doi:10.1016/J.MARPOLBUL.2004.06.012

WBM Oceanics (1992). Water Quality Investigations. Interim Report.Stage II. in Gladstone Industrial Land Study Consortium. GladstoneIndustrial Land Project studies. Stage 2 of 4. Data analysis report workingpapersVol. 2 of 3. Report for Gladstone Industrial Land Use Consortium.WBM Oceanics Australia, Brisbane.

Williams, M. R., Millward, G. E., Nimmo, M., and Fones, G. (1998). Fluxesof Cu, Pb and Mn to the North-Eastern Irish Sea: the importanceof sedimental and atmospheric inputs. Marine Pollution Bulletin 36,366–375. doi:10.1016/S0025-326X(98)00199-4

Wolanski, E. (1986). An evaporation-driven salinity maximum zone inAustralian tropical estuaries. Estuarine, Coastal and Shelf Science 22,415–424. doi:10.1016/0272-7714(86)90065-X

Zhang, H., Davison, W., Mortimer, R. J. G., Krom, M. D., Hayes, P. J.,and Davies, I. M. (2002). Localised remobilization of metals in amarine sediment. The Science of the Total Environment 296, 175–187.doi:10.1016/S0048-9697(02)00078-5

Manuscript received 6 March 2009, accepted 27 July 2009