heavy metal contamination in muscle tissue of four key...

Heavy metal contamination in muscle

tissue of four key recreational fish

species from the Derwent Estuary

Jeremy Verdouw

Research Thesis submitted in partial fulfilment of the requirements for

Honours in Aquaculture

National Centre for Marine Conservation and Resource Sustainability

University of Tasmania

2008

2

Declaration

I hereby declare that this thesis contains no material which has been accepted for a degree or

diploma by the University or any other institution and that, to the best of my knowledge this

thesis contains no material previously published or written by another person, except where

due acknowledgment is made.

Jeremy Verdouw

3

Acknowledgements

I would like to thank Leigh Mackenzie, Alister Clarke, Plinio Taurian, Mark Stalker and

Andrew Hunt for providing invaluable field support in the collection of fish. Thanks must

also go to Graeme Ewing for his practical and theoretical advice on fish processing and aging.

I am very thankful to Alison Featherstone, Stuart Black and Damien Norman from Analytical

Services Tasmania for all their hard work, advice and cooperation (using their gear); to

Zinifex (Nystar) and Fishwise for their generous financial support in completing the heavy

metal analyses. Finally, I am very much indebted to my three supervisors, Catriona Macleod,

Jeremy Lyle and Barbara Nowak, firstly for the opportunity to undertake this project, and

secondly for their extensive and invaluable practical and theoretical advice which helped me

greatly in completing this study.

4

Table of Contents

1. ABSTRACT ................................................................................................................................................. 6

2. INTRODUCTION ....................................................................................................................................... 7

3. METHODS................................................................................................................................................. 13

3.1 STUDY SITE ......................................................................................................................................... 13

3.2 STUDY SPECIES ................................................................................................................................... 16

3.2.1 Sand flathead (Platycephalus bassensis)....................................................................................... 17

3.2.2 Black bream (Acanthopagrus butcheri) ........................................................................................ 18

3.2.3 Sea-run trout (Brown trout- Salmo trutta) ................................................................................... 19

3.2.4 Yellow-eye mullet (Aldrichetta forsteri) ....................................................................................... 19

3.3 SAMPLE COLLECTION AND PROCESSING.............................................................................................. 20

3.4 AGE DETERMINATION ......................................................................................................................... 22

3.4.1 Preparing of otolith sections ......................................................................................................... 22

3.4.2. Increment counting........................................................................................................................ 22

3.4.3 Validation and precision of age estimates..................................................................................... 24

3.5 HEAVY METAL ANALYSIS ................................................................................................................... 25

3.5.1 Analysis of general heavy metal suite .......................................................................................... 25

3.5.2 Analysis of mercury....................................................................................................................... 26

3.5.3 Quality control of metal analyses.................................................................................................. 27

3.6 DATA ANALYSIS ................................................................................................................................. 28

4. RESULTS................................................................................................................................................... 30

4.1 OVERALL HEAVY METAL LEVELS........................................................................................................ 30

4.2 INTER-SPECIFIC COMPARISONS ........................................................................................................... 33

4.3 INTRA-SPECIFIC COMPARISONS ........................................................................................................... 36

4.3.1 Gender........................................................................................................................................... 36

4.3.2 Fish age and length....................................................................................................................... 37

4.4 INTER-METAL RELATIONSHIPS ............................................................................................................ 39

4.5 FLATHEAD REGIONAL COMPARISON ................................................................................................... 40

4.6 TROPHIC STATUS OF SELECTED SPECIES .............................................................................................. 46

5. DISCUSSION............................................................................................................................................. 47

5.1 FACTORS INFLUENCING METAL LEVELS BETWEEN SPECIES ................................................................. 47

5.2 REGIONAL VARIABILITY IN FLATHEAD................................................................................................ 55

5.3 IMPLICATIONS FOR PUBLIC HEALTH .................................................................................................... 59

5.4 CONCLUSIONS..................................................................................................................................... 62

5

5.5 FUTURE RESEARCH AND MANAGEMENT IMPLICATIONS....................................................................... 62

6. REFERENCES .......................................................................................................................................... 65

APPENDIX 1....................................................................................................................................................... 72

HEAVY METAL ANALYSIS PROTOCOLS .............................................................................................................. 72

AST QUALITY CONTROL SAMPLES (AS OUTLINED BY AST METHODS) .............................................................. 72

APPENDIX 2....................................................................................................................................................... 74

PLOTS OF BETWEEN METAL CORRELATIONS ...................................................................................................... 74

APPENDIX 3....................................................................................................................................................... 77

MERCURY ACCUMULATION IN MARINE FISH: A LITERATURE REVIEW.................................... 77

6

1. Abstract

This study measured levels of mercury, arsenic, cobalt, chromium, copper, iron, manganese,

nickel, lead, selenium and zinc in the muscle tissue of four key recreational fish species;

yellow-eye mullet (Aldrichetta forsteri), black bream (Acanthopagrus butcheri), sand flathead

(Platycephalus bassensis) and sea-run trout (Salmo trutta) from the Derwent Estuary. The

effects of diet, age, length, gender and region on metal levels were examined for each species

and levels were compared to Australian food standards to examine the risk to human health of

consumption of these species. Mean mercury levels in the muscle tissue of black bream (1.57

mg/kg), sea-run trout (0.68 mg/kg) and sand flathead (0.53 mg/kg) exceeded the maximum

permitted level of 0.5 mg/kg for mercury in seafood as prescribed by Food Standards

Australia and New Zealand (FSANZ). The significantly (P<0.05) higher levels in black bream

were considered to be of a particular human health concern. Mean levels for all other metals

were below the maximum permitted and generally expected levels (FSANZ) for all species

and therefore pose little threat to human health. Significant (P<0.05) inter-species and intra-

species differences were apparent for mercury, arsenic, copper, iron, manganese, zinc and

lead. Diet and age were likely to have the largest influence on differences in metal levels

between species. Gender was found to significantly (P<0.05) influence levels of arsenic, iron

and copper within species, whilst age and length were found to significantly (P<0.05)

influence levels of mercury, zinc, and arsenic. Significant (P<0.05) regional differences were

apparent for mercury levels in the muscle tissue of sand flathead. In contrast to sediment

levels, the highest mercury concentrations were in sand flathead from Ralphs Bay which is

some distance from the most contaminated region of the estuary. Age was found to be the best

predictor of mercury in sand flathead from the Derwent Estuary.

7

2. Introduction

Heavy metals occur naturally in the environment. Some are essential for normal function in

humans and animals (Pourang, 1995) such as, copper, iron, manganese and zinc; whereas

other metals such as mercury, cadmium and lead are not required even in small amounts by

any organism (Laws, 2000). Almost all metals, including the essential ones, are toxic to

animals and humans if levels exceed certain thresholds (Laws, 2000; Carvalho et al., 2005).

The toxicity of metals varies substantially and is largely due to their ability to interfere with

enzyme-mediated processes and disruption of cellular structure (Laws, 2000). Health effects

in humans contaminated by elevated metal levels include neurological disorders, bone

deterioration, cancer and immune system disorders (Jarup, 2003). From a human health

perspective, the primary contaminants of concern are mercury, arsenic, cadmium and lead

(Burger, 2007). The major route of exposure of these metals to humans is either through

direct contact or indirectly through ingestion of food, particularly seafood (Jarup, 2003).

The potential for heavy metal contamination to negatively affect human health has resulted in

many studies into heavy metal levels in fish and shellfish species, particularly in regions

heavily impacted by anthropogenic inputs (Ratkowsky et al., 1975; Walker, 1982; Fabris et

al., 1992). Levels of contaminants in fish are of interest not only because of the potential

effects on the fish themselves, but also because of the effects on organisms that consume

them, such as higher order predators and humans (Hylland et al., 2006). Guidelines on the

maximum permitted levels of metals in seafood have been introduced in many parts of the

world for the safe consumption of fish species (Adams and McMichael, 1999). Studies and

monitoring programs examining heavy metal levels in fish are becoming more and more

8

important, especially in less developed parts of the world where fish provide the major source

of protein (Toth and Brown, 1997; Burger et al., 1999a; Burger et al., 1999b). Even in more

affluent areas, fish are being consumed increasingly as an essential health food (Gislason et

al., 2000; Carvalho et al., 2005). Fish is promoted as a healthy and nutritious component of a

balanced diet, and an important source of proteins and lipids, including long chain

polyunsaturated fatty acids, and also of liposoluble vitamins (Egeland and Middaugh, 1997;

Han, 1998; Carvalho et al., 2005). Studies indicate that people who include fish in their diet

have a lower risk of coronary heart disease, hypertension, and cancer (Egeland and Middaugh,

1997). Thus, globally, fish assume great importance and consumption is increasing (Carvalho

et al., 2005). However, in contrast fish can be a source of contamination. In some

circumstances they can contain amounts of heavy metals which are highly toxic (Egeland and

Middaugh, 1997; Carvalho et al., 2005).

The concentration of heavy metals in fish is influenced by several factors; in particular

biological differences (eg. species, size, age, gender, sexual maturity, diet) and environmental

differences (eg. water chemistry, salinity, temperature, and levels of contamination) (Carvalho

et al., 2005). Numerous studies have shown that heavy metal accumulation in fish is strongly

influenced by environmental concentrations, that is, the levels in the water and sediments

(Blevins and Pancorbo, 1986; Calta and Canpolat, 2006). However, metal accumulation has

been found to vary markedly between species in the same area, as a result of differences in

feeding habits (Calta and Canpolat, 2006) and position in the food chain (Asuquo et al.,

2004). Differences within species have been described in response to location (Asuquo et al.,

2004), fish age/size (Calta and Canpolat, 2006), gender (Pourang, 1995) and life history stage.

9

Metals may be introduced into aquatic systems in several ways; including natural weathering

of rocks and volcanic eruptions, and human activities such as mining, ore refining and other

metal-based industries (Laws, 2000). Human activity is increasingly contaminating aquatic

environments with heavy metals (Park and Curtis, 1997). Since some of the most heavily

industrialised areas of the world are sited on the banks of estuaries, these waters are

particularly at risk from heavy metal contaminates (Birch, 2000). In the past, metallic wastes

have been discharged into rivers and estuaries based on the assumption that they would be

carried to the open sea and dispersed (Bloom and Ayling, 1977; Birch 2000). The reality is

somewhat different, with studies showing that estuaries can efficiently trap heavy metals in

sediments (Bryan, 1980; Birch, 2000). The deposition of high concentrations of heavy metals

in the sediments of many estuaries (and other confined water bodies) provides a sink for

continued contamination even in the absence of further input (Bryan, 1980; Williamson and

Morrisey, 2000). Whilst estuaries may be subjected to significant water pollution, their

sheltered waters also support unique communities of aquatic plants and animals (Edgar et al.,

1999) and provide important nursery grounds and habitats for many fish species (Correll,

1978; Edgar et al., 1999; Jones et al., 2003). This is of particular concern because fish and

shellfish living in contaminated waters readily accumulate metals and pose a risk to human

health if consumed (Han et al., 1998).

The Derwent Estuary in south eastern Tasmania is surrounded by the city of Hobart and the

greater metropolitan area. The estuary is particularly important from a recreational and

environmental perspective (Green and Coughanowr, 2003). Many birds, mammals, fish and

invertebrates depend on the estuarine habitats of the Derwent for shelter, food and

reproduction (Green and Coughanowr, 2003). The Derwent Estuary is also used widely for

recreational activities; particularly swimming, water skiing, windsurfing, scuba diving,

snorkelling and fishing (Green and Coughanowr, 2003). In addition, the Derwent Estuary

10

supports a significant recreational fishery; with the upper estuary being extensively fished for

sea-run trout and black bream, whilst the lower reaches are fished for a range of species

including sand flathead, Australian salmon, yellow-eye mullet, whiting and flounder (Green

and Coughanowr, 2003; Lyle, 2005). It has been estimated that approximately 4,000

recreational anglers fish in the Derwent in any year (Green and Coughanowr, 2003). In

addition the lower reaches of the estuary are also open to commercial fishing, particularly for

whiting and flathead (Lyle, 2005).

Despite its natural values and many recreational uses the Derwent Estuary water, sediment

and biota are all severely contaminated with heavy metals (Bloom, 1975; Green and

Coughanowr, 2003). Metals enter the Derwent via a number of pathways including air

emissions, treated effluent, stormwater run off, ground water seepage and spills (Green and

Coughanowr, 2003). However, the main source is historical industrial effluent from the zinc

smelter in the mid estuary and paper mill in the upper estuary (Eustace, 1974; Bloom, 1975;

Green and Coughanowr, 2003). The main heavy metal pollutants from industry are mercury,

cadmium, lead and zinc (Green and Coughanowr, 2003). Metal levels in the Derwent

sediments are among the highest in the world and levels of cadmium, lead and zinc in oysters

and mussels are also extremely high in comparison to food standards (Bloom, 1975; Cooper

et al., 1982; Green and Coughanowr, 2003). Studies by Eustace (1974), Ratkowsky et al.

(1975) and Dix et al. (1975) were the first to examine metal levels in Derwent Estuary fish

and found that levels of most metals were well below maximum permitted levels for seafood,

with the exception of mercury which exceeded maximum permitted levels in several fish

species including flathead and various sharks. A 25-year monitoring program on mercury

levels in sand flathead (Platycephalus bassensis) in the Derwent Estuary as part of the Zinifex

(Nystar) Seafood Monitoring Program showed mercury levels in some regions to be markedly

11

higher than the safe maximum permitted level for consumption (Green and Coughanowr,

2003; FSANZ, 2007). However, there are no recent data on metal levels in other important

recreational fish species in the Derwent Estuary, or on the levels of any metals besides

mercury in flathead. To characterise the potential risk of heavy metals in fish to consumers, it

is essential to obtain contemporary data on the levels of a range of metals in a variety of fish

species, in particular those that are recreationally important. In addition, there is also a need to

understand how environmental conditions and biological factors might influence metal levels

in fish from the Derwent Estuary to identify other species which consumption may be a threat

to human health and to aid in the understanding of how metal levels are accumulated by fish.

Furthermore, no study has comprehensively compared metal levels with age in fish in the

Derwent Estuary. This is despite the fact that there is much evidence for the increased

accumulation of mercury and to a lesser extent arsenic with increasing age and/or size

(Mackay et al., 1975; Ashraf and Jaffar, 1988; Hornung et al., 1993). Understanding the

relationship of age/size with metal levels for a given species is particularly important from a

human health perspective as it may influence the metal loading, and hence, potential health

effects of consumption.

This study will firstly compare metal levels (mercury, arsenic, cobalt, chromium, copper, iron,

manganese, nickel, lead, selenium and zinc) in four key recreational species of fish from the

Derwent Estuary to Food Standards Australia and New Zealand (FSANZ) guidelines on fish

consumption. Secondly, it will examine inter-specific differences in metal levels in relation to

diet/trophic status and age/size as well as within species differences with respect to age, size,

gender and for sand flathead, region of capture.

12

The primary objectives of this study are to:

(1) Compare metal concentrations in four key recreational fish species (muscle tissue)

from the Derwent Estuary to Australian food standards.

(2) Examine relationships between heavy metal accumulation with and age/size,

diet/trophic position, gender and region (sand flathead).

(3) Provide a comprehensive baseline data set for future research and long-term

studies.

13

3. Methods

3.1 Study site

The present study was undertaken in the Derwent Estuary in southern Tasmania, Australia

(Figure 1). The city of Hobart surrounds the Derwent Estuary and approximately 40% of

Tasmania’s population live around the estuary (Green and Coughanowr, 2003). The estuary

effectively spans from the township of New Norfolk to Storm Bay, a distance of

approximately 52 km (Figure 1) (Green and Coughanowr, 2003). For the purposes of this

study the estuary was divided into four sampling regions consistent with those used by

Eustace (1974), Ratkowsky et al. (1975) and the Zinifex (Nystar) seafood sampling program

over the past 30 years (Green and Coughanowr, 2003). The regions were: 1) North of Tasman

Bridge (Tasman Bridge to New Norfolk), 2) Western shore, 3) Eastern shore, and 4) Ralphs

Bay (Figure 1) with two or more sampling sites within each region (Table 1). The regions

were distinguished on the basis of particular differences in their hydrography, sediment heavy

metal loading and previous biological information (Eustace, 1974; Ratkowsky et al; 1975;

Jones et al., 2003).

Region 1 (Tasman Bridge to New Norfolk) is a graduating zone of sea water to fresh water

and is the most heavily impacted from industrial inputs. Regions 2 (Western shore) and 3

(Eastern shore) are primarily marine and can be divided into east (Region 3) and west (Region

2) due to the fact that river flow is more distinct on the east (Jones et al., 2003). Ralphs Bay

can be considered a further subdivision of the marine zone, since it has distinct physical

differences to the other regions (Jones et al., 2003) and previous studies have shown it to

14

contain oysters with relatively high levels of zinc and cadmium (Thrower and Eustace 1973)

and flathead with relatively high mercury levels (Ratkowsky et al., 1975).

Figure 1. Map of the Derwent Estuary showing the location of the study regions, sample sites and major

industries; Norske Skog and Zinifex (Nystar) (Adapted from figure by Green and Coughanowr, 2003).

15

Sand flathead were also sampled from Mickey’s Bay off the south of Bruny Island (Figure 2).

Previous data have indicated that it is relatively unaffected by pollution (sand flathead

mercury levels are well below guidelines) and therefore can be considered a control region

(Green and Coughanowr, 2003).

Figure 2. Map of the D’Entrecasteaux Channel showing the location of the control region, Mickeys Bay off the

south of Bruny Island (Adapted by figure from www.wildthingadventures.com.au).

16

Table 1. Region and sites where sand flathead were sampled from during this study.

Region No. Region Name Sites

1 New Norfolk to Tasman Bridge Cornelian Bay- CB

Newtown Bay- NB

2 Western Shore Kingston Beach- KB

Sandy Bay Beach- SBB

3 Eastern Shore Seacroft Bay- SB

South Arm- SA

Opossum Bay- OB

Marralyne- M

Punches Reef- PR

Bellerive Beach- BB

4 Ralphs Bay Ralphs Bay Spit- RBS

Maria Point- MP

Old Lease- OL

5 (control) Mickeys Bay Mickeys Bay

3.2 Study species

Species were selected based on the following criteria:

1) Recreationally caught in the Derwent Estuary (Morton et al., 2005)

2) Recognised as good eating fish and commonly consumed (DPIW, 2007; IFS,

2007)

3) Relatively easily caught (DPIW, 2007; IFS, 2007)

4) Displaying markedly different foraging behaviours and life history traits

(Morton et al., 2005)

17

3.2.1 Sand flathead (Platycephalus bassensis)

Sand flathead (Figure 3) are extremely abundant throughout Tasmania and the Derwent

Estuary, where they can be caught anywhere south of Bridgewater (Morton et al., 2005). They

are a bottom-feeding species which feed predominantly on shrimps, crabs and small fish

(Jordan, 2001). They are also non-migratory and are believed to spend the majority of their

lives within a relatively limited region (Dix et al., 1975; Morton et al., 2005). Sand flathead

spawn around Tasmania from October through to March in coastal bays and inner continental

shelf waters (Jordan, 2001).

Figure 3. Sand flathead (Platycephalus bassensis) (copied from DPI, 2005)

They are caught in large numbers from spring to autumn by recreational fishers and are also a

relatively important commercial species in Tasmania (Lyle, 2005; Morton et al., 2005). In

2000/01 an estimated 2.1 million flathead were caught by recreational fishers, of those

approximately 65% were retained for personal consumption (Lyle, 2005). Sand flathead can

live up to 17 years of age and may grow to more than 50 cm long and 3kg in weight (Morton

et al., 2005).

18

3.2.2 Black bream (Acanthopagrus butcheri)

Black bream (Figure 4) are commonly found in estuaries and lower reaches of rivers in

southern Australia, including the Derwent Estuary (Edgar, 1997). They are euryhaline and

may sometimes be found in fresh water (Morton et al., 2005). Black bream are primarily a

bottom feeder, consuming prey such as sandworms, mussels, crabs and pilchards (Morton et

al., 2005; R. Sakabe, pers. comm., 2007). Evidence from tagging studies indicates that bream

largely remain within a river or estuary throughout their life with little movement between

systems (Potter and Hyndes, 1999; R. Sakabe, unpubl. data., 2007).

Figure 4. Black bream (Acanthopagrus butcheri) (DPI, 2005)

In Tasmania, black bream spawn from spring through to mid summer (R. Sakabe, pers.

comm., 2007). They are commonly targeted by anglers, particularly in the upper reaches of

the Derwent Estuary (Morton et al., 2005). In 2000/01 the estimated recreational catch was

estimated at 76,500 individuals with approximately 46,000 retained (Lyle, 2005). Black

bream are a long-lived species, with individuals in excess of 20 years being recorded

(Morison et al., 1998; R.Sakabe, pers.comm., 2007). Growth is slow and they mature at about

4 years reaching a maximum size of around 60 cm in length and 4 kg in weight (Morton et al.,

2005).

19

3.2.3 Sea-run trout (Brown trout- Salmo trutta)

The brown trout (Figure 5) is an introduced species keenly sought after by anglers in rivers

and lakes, but they also occur in estuarine and marine waters, then known as sea-run trout

(Edgar, 1997). They are commonly caught in the upper reaches of the Derwent Estuary, from

the Tasman Bridge and above (DPIW, 2007). Sea-run trout smoltify in spring and migrate to

salt water coastal areas to feed during summer (Cucherousset et al., 2005). They are

opportunistic consumers, predominantly feeding on whitebait (IFS, 2007).

Figure 5. Sea-run trout (Brown trout- Salmo trutta)

(http://pond.dnr.cornell.edu/nyfish/Salmonidae/brown_trout.html)

Sea-run trout are commonly caught by line from August to the end of summer (IFS, 2007).

They are known to reach a maximum size of 90 cm (14 kg) and live to around 9 years of age

(Graynoth, 1996; IFS, 2007).

3.2.4 Yellow-eye mullet (Aldrichetta forsteri)

Yellow-eye mullet (Figure 6) are the most common mullet species in Australia (Edgar, 1997).

They are extremely abundant in southern Australian estuaries, and are a popular fish for

20

human consumption, with around 1000 tonnes per year commercially caught in Victoria

(Edgar, 1997).

Figure 6. Yellow-eye mullet (Aldrichetta forsteri) (DPIW, 2007)

Yellow-eye mullet are a migratory species which are known to change diet with age; juveniles

feed on planktonic animals, medium sized fish feed on benthic crustaceans and molluscs and

the larger fish feed almost exclusively on algae (Edgar, 1997). They are highly abundant in

the inshore estuarine areas of the Derwent Estuary, from the Bridgewater Bridge and down

stream (DPIW, 2007). They can be caught all year round in gill nets and by line fishing. They

live to around 13 years of age and may attain a maximum size of 40 cm (DPIW, 2007).

3.3 Sample collection and processing

Sand flathead, black bream, sea-run trout and yellow-eye mullet were sampled both from the

shoreline and from research vessels in the Derwent Estuary between the 28th

of July and the

28th

of November, 2007 by TAFI staff and recreational fishers using both line fishing and gill

netting (Table 2). On each sampling trip the date, location, species and fishing method were

recorded. All fish were euthanized by immersion in a clove oil seawater solution (3 mL of

clove oil to 30 L of seawater) in accordance with University Ethics Approval.

21

Table 2. Collection dates, method of collection, region and number of fish sampled by species in the Derwent

Estuary.

Fish were labelled and held on ice after which they were transported to the Marine Research

Laboratories (MRL) in Taroona, and stored in a -18oC freezer until required for processing. In

addition several fish were collected by recreational fishers, and either whole fish or frames

(with muscle tissue attached and gut contents intact) bagged individually and frozen. Fish

sampling targeted a range of sizes for each species (including below legal size fish). Flathead

were sampled from all regions, bream and mullet were sampled from region 1 and sea-run

trout were sampled from regions 1 and 3 (Table 2).

Each fish was measured to the nearest millimetre (total and fork length) and weighed to the

nearest gram. Fish were dissected and stomach contents removed, weighed and then identified

into the lowest easily recognisable taxonomic groups. The fish were sexed and gonads

removed and weighed. A sample of muscle tissue (approximately 50 g) was removed from

each fish from the area posterior of and adjacent to the pectoral fin. The muscle tissue was

placed in a labelled plastic zip-lock bag and stored at -18oC until further processing. Sagittal

otoliths (here after ‘otoliths’) were extracted using fine tipped forceps, cleaned, dried and

stored in plastic sample containers.

Species Sampling Period Method Region n Total

Sand flathead 6/10/07 to 23/11/07 line 1 30

2 30

3 30

4 30

5 30 150

Black bream 12/9/07 to 29/10/07 line 1 28 28

Sea-run trout 16/8/07 to 11/10/07 line 1 20

3 5 25

Mullet 28/7/07 to 28/11/07 line and gill net 1 27 27

22

3.4 Age determination

3.4.1 Preparing of otolith sections

Methods used for fish aging were based on those reported by Ewing et al. (2007). One otolith

from each fish (either the right or left otolith) was mounted in a block of polyester casting

resin and sectioned with a diamond gem saw transversely through the primordium to a

thickness of 500 to 600 µm. Between 3 and 6 sections were cut from each otolith to ensure the

primordium was represented in at least one of the sections. Sections were cleaned and

mounted in polyester resin on glass microscope slides under cover slips.

3.4.2. Increment counting

The transverse otolith sections showed alternating bands of wide translucent and narrower

opaque zones extending from an inner zone when examined by a stereo-microscope (20× to

25× magnification) under transmitted light (Figure 7). The zones adjacent to the inner zone

were relatively broad and decreased in width out to the growing edge. Ages were estimated by

counting the opaque zones on the section closest to the primordium along a transect between

the primordium and the outer edge of the section on either the dorsal or ventral side of the

sulcus (Figure 7). Otoliths were rejected if the opaque zones were optically unreadable or if

the primordium could not be identified. Leica IM50 image analysis software was used to aid

in the counting and marking of increments and to make measurements on the image. To

eliminate bias, otoliths were examined with no knowledge of fish size, sex or date of capture.

23

Figure 7. A representative transverse otolith section from an Acanthopagrus butcheri sample (transmitted light).

Opaque zones are marked and numbered and the age calculated from this count was 25 years.

Counts of opaque zones commenced from, and included, the first opaque zone (zone

immediately after the first translucent zone) (Figure 7). To aid in the identification of the first

opaque zone, measurements were made from the primordium to the inner side of the first

opaque zone (the transverse radius). The first opaque zone in all species was dimensionally

stable. The mean transverse radius for the various species were; flathead 1300 µm (n = 150,

SE = 17), bream 1355 µm (n = 27, SE = 31), trout 1400 µm (n = 25, SE = 43) and mullet

1270 µm (n = 28, SE = 30). Counts of opaque zones were converted to age estimates (Table

3) by adding one year to the total count of opaque zones to take into account the date of

capture relative to the estimated closing date of the last opaque zone.

Primordium

Sulcus

24

3.4.3 Validation and precision of age estimates

Age determination by counting the increments of sectioned otoliths has been previously

validated for sand flathead (Jordan et al., 1998), black bream (Morison et al., 1998; R.

Sakabe, unpubl. data., 2007), sea-run trout (brown trout) (Graynoth, 1996; IFS, 2003) and

yellow-eye mullet (Curtis and Shima, 2005). This study therefore followed the methods of

increment interpretation used by these authors for the respective fish species.

Otoliths were read by a primary reader (author, Reader A) who examined all otoliths twice

(n=230). A second reader, experienced in examination of transverse sagittal otolith sections

(Reader B), read a random sub-sample of 80 otoliths (20 from each species). The age

estimates between and within readers were examined from tables of difference of readings

against age, and quantified with the index of average percent error (APE) as a recommended

measure of precision (Beamish and Fournier, 1981). APE scores for within reader differences

were all below 1% (flathead 0.57%, mullet 0.15%, trout 0.38%, and bream 0.00%) with a

maximum difference of 1 zone, indicating a very high mean level of agreement. The APE

scores for between reader differences were: flathead 5.53%, mullet 1.84%, trout 13.75% and

bream 1.18%. The maximum difference between counts was 3. These results indicate that age

estimates of all species except trout were accurate and, with little variation in the reading

process. Estimates of trout age differed between readers more so than the other species,

reflecting the sometimes unclear otolith sections. However, variability was still within

acceptable levels.

25

3.5 Heavy metal analysis

All tissue samples were analysed for the following metals: mercury, arsenic, cadmium, cobalt,

chromium, copper, iron, manganese, nickel, lead, selenium and zinc. All the metals, with the

exception of mercury, were analysed from a single tissue sample using the same method.

Mercury analysis required a different method and therefore was undertaken on a separate

tissue sample. All measurements of metal levels were recorded as milligrams per kilogram

(mg/kg) wet weight (wet wt.)

3.5.1 Analysis of general heavy metal suite

Tissue samples were defrosted by placing them in a fridge over-night. Approximately 12 g of

muscle tissue was removed from the anterior end of the fillet using a scalpel on a ceramic

chopping board. Skin and bones were removed from each tissue sample. Individual samples

were placed on weighed, acid washed watch glasses before being weighed on an electronic

balance (Mettler Pj3600 Balance, Switzerland) (all weights in grams to 2 decimal places).

Samples were then dried for a minimum of 18 hours in a 105oC oven (FSE Scientific OG24

SE3, Australia).

Samples were reweighed in order to calculate the dried matter fraction (DMF). The dried

tissue was scraped into an acid washed ceramic mortar and pestle and ground to a fine

powder. The ground sample was then placed in a labelled plastic zip-lock bag and stored

frozen prior to acid digestion. To minimise any chance of contamination, cutting utensils and

boards were thoroughly rinsed with deionised water and dried between samples. The mortar

and pestle were washed using the following process between samples: 1) rinsed with tap

26

water, 2) rinsed with 1% HNO3, 3) rinsed with deionised water, 4) rinsed with acetone, and 5)

dried with paper towel.

Approximately 1.00 g ±0.5 of dried sample was accurately weighed into a 50 mL digestion

tube to which 10 mL of concentrated (65%) nitric acid (HNO3) was added. The tube was

covered with a watch glass and left to stand for 12 h under a fume hood before digestion on

an Aim 500 Digestion Block using program number 2 (appendix 1). Up to 50 samples could

be digested at any one time. Each digestion run included a reagent blank, a laboratory

reference material (LRM), a blank matrix spike, two sample duplicates and two sample matrix

spikes (appendix 1). Once the program had ceased the samples were allowed to cool before 10

mL of hydrogen peroxide (H2O2) was added. The samples were then digested for a second

time using program number 3 (appendix 1). On completion of this program, samples were

again allowed to cool before filling the tubes up to the 50 mL mark with deionised water. The

sample solutions were then mixed thoroughly and transferred to a 50 mL Stardset tube.

Particulate matter was allowed to settle and then the sample was filtered through a 0.45 µm

filter to remove any suspensions. Heavy metals were analysed using an inductively coupled

plasma atomic emission spectrophotometer (Varian 730ES, Australia). All heavy metal

analyses were undertaken by Analytical Services Tasmania, a NATA accredited analytical

service provider. The method reporting limits were 0.1 mg/kg for all metals except selenium

which had a method reporting limit of 0.5 mg/kg.

3.5.2 Analysis of mercury

Samples were defrosted as previously described and approximately 1.00 g ± 0.5 of tissue was

weighed into a 50 mL digestion tube along with 5 mL of mercury digestion acid (HNO3 67%

27

v/v plus H2SO4 33% v/v). All tubes were capped with a small watch glass before being placed

on a digestion block set on program number 4 (appendix 1). As with the total metals, up to 50

samples could be digested at a time, with a reagent blank, a blank matrix spike, two sample

duplicates and two sample matrix spikes included in each digestion. Once the program was

completed, sample tubes were allowed to cool before adding 15 mL of KMnO4 solution and 5

mL of K2S2O8 solution and mixing with a vortex mixer. The sample solutions were left to

stand for 12 h prior to checking for colour change to purple. If the solutions were clear or

brown an additional 5 mL of KMnO4 solution was added and the samples were left to stand

for another 12 h. Sample solutions were then transferred to a 50 mL Stardset tube. Samples

were decolourised by the addition of 10 mL of hydroxylamine-HCl solution. They were then

made up to 50 mL with deionised water and mixed before being diluted 1:5 with 10%

digestion acid. Mercury samples were analysed by Cold Vapour Atomic Fluorescence

Spectrometry (CV- AFS) on an atomic fluorescence analyser (Melenium Merlin, United

Kingdom). All mercury analyses were undertaken by Analytical Services Tasmania and the

method reporting limit was 0.02 mg/kg.

3.5.3 Quality control of metal analyses

The analysis of the heavy metal suite met the quality control standards for all batches, whilst

analysis of mercury met the quality control standards with the exception of two minor

breaches. In one batch a blank matrix spike exceeded the theoretical value by 26% (25% is

allowed) (Appendix 1). This was considered a very marginal breach and no repeat analysis

was deemed necessary. In the same batch, the preparation blank exceeded the minimum

reportable level of (0.02 mg/kg) (Appendix 1). In this instance it was still determined that

blank subtraction was not appropriate as the levels were too low to have any measurable

28

effect on the results. The results of the quality control suggest that overall the metal levels

detected in all the samples were reliable (see Appendix 1 for full detail of quality controls).

3.6 Data analysis

Statistical computations were carried out using SPSS and PRIMER software. Analysis of

variance (ANOVA) was used to test for inter-species differences, intra-species differences

between sexes, and regional differences in metal levels in flathead. In all cases the type III

sums of squares was used to test the null hypothesis. This was appropriate for the study

because the experimental design was unbalanced. Assumptions of homogeneity and normality

of data were assessed through examination of residual plots and data were appropriately

transformed where assumptions of homogeneity were not met (transformations are identified

where applied). For all ANOVAs, significance value was set at P<0.05. Where effects were

significant, TUKEY pair-wise post hoc tests were used to further examine these differences.

Multivariate analysis of metal levels in flathead by region was performed using PRIMER.

Principle Components Analysis (PCA) was applied to the metal data for individuals. The data

were transformed to account for the large differences in the absolute values for mercury.

Regional differences were examined by superimposing the regions on the resultant

distribution plot. The relative contributions of the various metals to the overall differences

between fish groupings were examined in vector plots.

Preliminary evaluation of the absolute levels for each metal identified that only mercury, iron,

zinc, arsenic and lead were at sufficiently elevated levels to be of concern. Consequently,

correlation analysis was undertaken to further examine the relationships between these metals.

29

Linear regression analysis was carried out on the data with SPSS to examine how metal levels

within species responded to the variables of age and length. Regression analysis revealed a

strong positive, linear relationship of mercury concentration with age and length for flathead

and trout. As a result, regional differences in mercury levels in flathead were further

examined using ANCOVA (SPSS) in which fish age was the covariate. The interaction of

region and age was found to be insignificant; hence the covariate was the same for all

treatments. The analysis was then run again without the interaction term in the model and the

covariate was found to be significant (i.e. the covariate explained a significant amount of the

difference between regions). Population marginal means, also known as least-squares means

(LS means), for the mercury concentrations are presented in the results. These means are the

expected concentration in the muscle of the fish that could be expected for a balanced

statistical design with the covariate (age) at its mean value.

30

4. Results

4.1 Overall heavy metal levels

Overall heavy metal levels by species are summarised in Table 3 and compared with health

standard levels. There were significant differences in the levels of metals within and between

species, with several instances where levels were above recommended food standards. Of the

metals significant from a human health perspective (i.e. mercury, lead, arsenic, nickel,

chromium), mean mercury levels were relatively high in bream, trout and flathead, which all

had mean levels in excess of the maximum level (ML) of 0.5 mg/kg prescribed by Food

Standards Australia and New Zealand (FSANZ) (Table 3). An individual bream had a

mercury content of 2.3 mg/kg, almost five times the ML (Table 3), whilst the highest recorded

levels for individual trout and flathead were more than double the ML (Table 3). In contrast,

mullet had a mean mercury level of 0.23 mg/kg (max 0.25) which was well below the ML

(Table 3).

Arsenic concentrations were highly variable between species with the highest mean level in

sea run trout (5.13 mg/kg); one individual trout having a muscle tissue concentration of 11

mg/kg (Table 3). Arsenic concentrations in the other species were generally low with mullet

having the lowest levels (Table 3). However a flathead had the highest individual arsenic

level reported (16 mg/kg) (Table 3). Given the fact that inorganic arsenic represents

approximately 20% of total arsenic in fish, this level would be in excess of the ML.

31

Table 3. Summary of: mean age, fork length and weight, sex ratios and mean ± standard error and range (parenthesis) muscle tissue metal concentrations for sand flathead,

black bream, sea-run trout and yellow-eye mullet from the Derwent Estuary. Also shown are FSANZ maximum levels (ML), FSANZ generally expected levels (GEL; median

and 95th

percentile) and limit of reporting (LOR). Note: Muscle tissue levels of Cd, Co and Ni were all below the detectable level of the analysis and so are not included in

table.

Biometrics Heavy metal levels (mg/kg wet wt.)

Species n

age

(years)

FL

(mm)

weight

(g)

sex ratio

(M:F) Hg As Zn Pb Fe Cu Mn Se Cr

sand

flathead 150 6 284 161 61: 75 0.53 ± 0.05 3.91 ± 0.43 5.89 ± 0.26 0.06 ± 0.04 2.57 ± 0.29 0.16 ± 0.01 0.05 ± 0.01 0.02 ± 0.02 0.01 ± 0.01

(2-13) (0.1-1.4) (0.5-16.0) (3.7-12.0) (0.0-1.7) (1.1-13.0) (0.0-0.3) (0.0-0.3) (0.0-0.7) (0.0-0.5)

black

bream 28 19 342 994 18: 10 1.57 ± 0.08 1.80 ± 0.23 6.06 ± 0.91 0.02 ± 0.01 5.10 ± 0.95 0.21 ± 0.01 0.46 ± 0.09 0.25 ± 0.06 <0.1

(13-28) (0.57-2.30) (0.5-4.8) (2.7-22.0) (0.0-0.1) (2.5-29.0) (0.1-0.3) (0.2-2.7) (0.0-0.7)

sea-run

trout 25 4 426 709 2: 9 0.68 ± 0.08 5.13 ± 0.55 6.24 ± 0.58 0.01 ± 0.01 4.82 ± 0.25 0.39 ± 0.02 0.24 ± 0.03 <0.5 0.03 ± 0.01

(2-7) (0.08-1.70) (2.0-11.0) (3.4-17.0) (0.0-0.1) (2.9-7.1) (0.3-0.7) (0.0-0.8) (0.0-0.2)

yellow-

eye

mullet 27 7 295 335 9: 16 0.23 ± 0.05 1.06 ± 0.13 9.75 ± 0.72 0.35 ± 0.10 6.33 ± 0.29 0.32 ± 0.01 0.23 ± 0.06 <0.5 0.03 ± 0.01

(3-13) (0.05-0.25) (0.4-2.8) (6.2-16.0) (0.0-1.8) (4.8-10.0) (0.2-0.4) (0.0-0.9) (0.0-0.1)

ML 0.5 2.0 - 0.5 - - - - 1.0

GEL 5, 15 0.5, 2 0.5, 2

LOR 0.02 0.1 0.1 0.1 0.1 0.1 0.1 0.5 0.1

32

Zinc levels ranged from highest mean value of 9.75 mg/kg (mullet) to a lowest mean value of

5.89 mg/kg (flathead) (Table 3). The highest individual concentration of zinc was in bream

(22 mg/kg) (Table 3), with individuals of trout, mullet and bream all exceeding the generally

expected level (GEL) for zinc of 5 mg/kg (median) to 15 mg/kg (95th

percentile) (Table 3).

Lead levels were generally very low, with mean levels for all species falling below the ML of

0.5 mg/kg; however, once again there were individual fish that exceeded the requirements

with one mullet and one flathead being particularly high (1.8 mg/kg, 1.7 mg/kg respectively)

(Table 3).

Levels of the metals less significant from a human health perspective showed only slight

variation between species. Mean iron levels were the highest in mullet (6.33 mg/kg) and the

lowest in flathead (2.57 mg/kg) (Table 3). Mean copper levels were generally similar between

species with all mean levels below the GEL prescribed by FSANZ (Table 3). The highest

levels were in trout (0.39 mg/kg) (Table 3). Manganese concentrations were generally low;

bream had the highest levels (mean 0.46 mg/kg) with only one individual recording a muscle

tissue concentration of any significance (2.7 mg/kg) (Table 3). Only in flathead and bream

were levels of selenium above detection limits and in all cases were below or within the GEL

(Table 3). Similarly, concentrations of chromium were well below the ML (Table 3) and

cadmium, cobalt and nickel were below the limit of reporting (LOR) in all fish examined.

This study sampled across a range of sizes of each species in order to examine metal level

against age and size. However, since only legal size fish would generally be taken and

consumed by fishers, we adjusted the mean levels accordingly to fully evaluate the human

health risk. Table 4 shows that of the fish above legal size collected in this study, all of the

bream (100%), 60% of the trout and 46% of the flathead had mercury levels which exceeded

33

the ML (Table 4). In sharp contrast no individual mullet breached permissible mercury levels

(Table 4).

Table 4. Number and percentage of fish (total, above and below legal size) from each species which exceeded

the maximum mercury level of 0.5 mg/kg.

Flathead Bream Trout Mullet

Minimum legal size (Total length) (mm) 300 250 220 250

No. of legal fish 56 28 25 26

No. of legal fish exceeding mercury limit 26 28 15 0

% of legal sized fish over mercury limit 46 100 60 0

No. of sub-legal fish 94 0 0 1

No. of sub-legal fish exceeding mercury limit 24 0 0 0

% of sub-legal fish sized fish over mercury limit 26 0 0 0

Total no. of fish exceeding mercury limit 50 28 15 0

% of total fish exceeding mercury limit 33 100 60 4

4.2 Inter-specific comparisons

Levels of mercury, arsenic, zinc, lead, iron, copper and manganese were compared between

species, as these metals were all detected at elevated levels in the various fish species and

therefore could reflect biological and physiological differences. Region 1 was the only region

from which samples of all species were obtained, and hence the comparisons of metal levels

were made between fish collected from this region only.

Bream had a significantly higher mean mercury concentration than the other species, almost

three times higher than flathead and mullet (Figure 8a and Table 5), whilst trout and flathead

had significantly higher mercury levels than mullet, again almost three times higher (Figure

8a). Flathead, trout and bream on average, all exceed the recommended maximum permitted

mercury level (Figure 8a).

34

Figure 8. Mean muscle tissue metal concentrations ± SE (Flathead n=30, Bream n=28, Trout n=25, Mullet

n=27) for mercury, arsenic, zinc, lead, iron, copper and manganese between species. Different letters indicate

statistically significant (P<0.05) differences between means from post hoc pooling analysis. Graph (a) also has a

broken line indicating the maximum permitted mercury level 0.5 mg/kg (FSANZ, 2007).

a) Hg

0.0

0.5

1.0

1.5

2.0

Trout Mullet Flathead Bream

Hg

(m

g/k

g)

b) As

0.0

1.0

2.0

3.0

4.0

5.0

6.0


As (

mg

/kg

)

c) Zn

0.0

2.0

4.0

6.0

8.0

10.0


Zn

(m

g/k

g)

d) Pb

0.0

0.1

0.2

0.3

0.4

0.5


Pb

(m

g/k

g)

e) Fe

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0


Fe (

mg

/kg

)

f) Cu

0.0

0.1

0.2

0.3

0.4

0.5


Cu

(m

g/k

g)

g) Mn

0.0

0.1

0.2

0.3

0.4

0.5

0.6


Mn

(m

g/k

g)

B

A B

C C

A

C

B

A

B

A A

A

B

A A

BC

C

A

B C C

A

B

AB B

A

B

35

Table 5. Species comparisons of concentrations of mercury, arsenic, zinc, lead and selenium in muscle tissue

from fish collected from Region 1 by one-way ANOVA. Shown are degrees freedom (df), mean squares (MS), F

values (F) and significance (P).

* Indicates data were (Ln) transformed

Trout and flathead had a significantly higher mean concentration of arsenic than the other

species, twice the levels found in bream and more than three times the mean levels in mullet

(Figure 8b and Table 5). Bream had a significantly higher mean arsenic levels than mullet,

almost twice as high (Figure 8b).

Zinc and lead concentrations were significantly higher in mullet than other species (Figure 8c,

d and Table 5). Zinc levels in bream, trout and flathead were similar, with standard errors

within all species being small (Figure 8c). Lead levels were noticeably higher in flathead in

comparison to bream and trout (Figure 8d). Iron levels were significantly higher in mullet

than the other species, whilst trout and bream had higher levels than flathead (Figure 8e and

Table 5). Copper levels were significantly higher in trout and mullet than the other species,

whilst bream had higher levels than flathead (Figure 8f and Table 5). Manganese levels in

mullet and bream were significantly higher than the other species whilst trout had higher

levels than flathead (Figure 8g and Table 5).

Metal Source df MS F P

Hg* Species 3 25.63 92.79 0.000

As* Species 3 14.91 55.71 0.000

Zn Species 3 68.69 6.80 0.000

Pb Species 3 1.90 3.87 0.000

Fe* Species 3 5.59 38.62 0.000

Cu Species 3 0.48 71.99 0.000

Mn* Species 3 0.65 4.60 0.005

36

4.3 Intra-specific comparisons

4.3.1 Gender

There were significant gender related differences in levels of arsenic, copper and iron in

bream and flathead (Figure 9 and Table 6). Arsenic was significantly higher in females in

both species (Figure 9a, b and Table 6), whilst immature flathead had similar arsenic levels to

female flathead (Figure 9b).

Figure 9. Mean muscle tissue metal concentrations ± SE (Bream: male n=18, female n=10; Flathead: male n=61,

female n=75, immature n=14) for arsenic, copper and iron between male, female and immature fish for bream

and flathead. Different letters indicate statistically significant (P<0.05) differences between means from post hoc

pooling analysis.

d) Flathead

0.0

1.0

2.0

3.0

4.0

Male Female Immature

Fe (

mg

/kg

)

c) Flathead

0.0

0.1

0.2

0.3

0.4


Cu

(m

g/k

g)

b) Flathead

0.0

1.0

2.0

3.0

4.0

5.0


As (

mg

/kg

) A

B B

B

A A

B

A A

a) Bream

0.0

1.0

2.0

3.0

4.0

Male Female

As (

mg

/kg

)

A

B

37

Levels of copper (Figure 9c) and iron (Figure 9d) in flathead were both significantly higher in

males than in females or immature fish (Table 6). However, there were no significant

differences with respect to gender between mercury and lead levels for any of the species.

Table 6. Gender comparisons of concentrations of arsenic, iron and copper in muscle tissue from bream and

flathead by one-way ANOVA. Shown are degrees freedom (df), mean squares (MS), f values (F) and

significance (P).

Metal Species Source df MS F P

As* Bream Sex 1 5.06 21.52 0.000

As Flathead Sex 2 57.74 12.47 0.000

Fe Flathead Sex 2 10.60 5.70 0.004

Cu Flathead Sex 2 0.96 3.58 0.030


4.3.2 Fish age and length

Age and length were found to influence metal concentration (Figure 10 and Table 7).

Significant, positive, linear relationships were observed for both age and length with mercury

concentration for flathead and trout (Figure 10a, b, c, d and Table 7). Length explained more

within species variation in mercury levels than age for both flathead and trout. Length

accounted for more than 10% of variation (highly significant) in flathead as opposed to 3%

for age, whilst for trout length accounted for more than 53% (highly significant) of the

variation as opposed to 48% (also highly significant) for age (Table 7). The large proportion

of the variation in mercury levels in trout explained by both age and length indicates a high

degree of autocorrelation between age and length in trout. Similarly, arsenic levels in bream

and trout were found to significantly (P<0.05) increase with length (Figure 10e, f and Table

7). The relationship was stronger in bream than trout (Figure 10e, f and Table 7).

38

Figure 10. Relationships for species with significant regressions of mercury, arsenic and zinc concentrations

with age and length.

a) Trouty = 0.1962x - 0.0066

R2 = 0.4746

0.0

0.5

1.0

1.5

2.0

0 2 4 6 8

Age (years)

Hg

(m

g/k

g)

b) Trout

y = 0.0037x - 0.8832

R2 = 0.5376

0.0

0.5

1.0

1.5

2.0

200 300 400 500 600

Fork length (mm)

Hg

(m

g/k

g)

c) Flathead

y = 0.0173x + 0.3789

R2 = 0.0264

0.0

0.5

1.0

1.5

0 5 10 15

Age (years)

Hg

(m

g/k

g)

d) Flathead

y = 0.0019x - 0.0646

R2 = 0.1055

0.0

0.5

1.0

1.5

150 200 250 300 350 400 450

Fork length (mm)

Hg

(m

g/k

g)

e) Trout

y = 0.0145x - 1.0615

R2 = 0.1942

0.0

2.0

4.0

6.0

8.0

10.0

12.0

200 300 400 500 600

Fork length (mm)

As (

mg

/kg

)

f) Breamy = 0.0392x - 11.599

R2 = 0.3502

0.0

1.0

2.0

3.0

4.0

5.0

6.0

300 320 340 360 380 400

Fork length (mm)

As (

mg

/kg

)

g) Flathead

y = -0.1812x + 7.0734

R2 = 0.0401

0.0

5.0

10.0

15.0

20.0

0 5 10 15

Age (years)

Zn

(m

g/k

g)

39

In contrast, zinc levels in flathead significantly decreased with age (Figure 10g and Table 7).

Although this relationship was significant, it only explained less than 5% of the variability in

zinc concentrations (Figure 10g and Table 7). Metal levels in mullet were not significantly

correlated with either age or length.

Table 7. Significant (P<0.05) linear regressions for metal concentration and age or length for all metals and all

species by regression analysis. Age and length as independent variables and metal concentration the dependent

variable.

Bream Trout Flathead

Regression R2 P R

2 P R

2 P

Hg*Age 0.475 0.000 0.030 0.036

Hg*Length 0.538 0.000 0.103 0.000

Zn*Age - 0.040 0.014

As*Length 0.350 0.001 0.194 0.027

(-) refers to negative correlation

4.4 Inter-metal relationships

Significant, positive correlations were found between zinc and iron in trout, bream and

flathead; iron and copper in trout, mullet and flathead; zinc and lead in mullet; lead and

manganese in mullet; zinc and copper in flathead and iron and mercury in flathead (Table 8

and Appendix 2). There were also negative correlations between iron and arsenic in flathead

and mercury and copper in bream (Table 8). The strongest correlations were between zinc and

iron for trout, bream and flathead as well as iron and copper for trout, mullet and flathead

(Table 8 and Appendix 2).

40

Table 8. Significant (P<0.05) inter-metal correlations within species.

Trout Bream Mullet Flathead

Correlation R2 P R

2 P R

2 P R

2 P

Zn*Fe 0.264 0.009 0.163 0.033 0.250 0.000

Zn*Pb 0.377 0.001

Zn* Cu 0.062 0.002

Zn*Mn 0.164 0.036

Cu*Fe 0.211 0.021 0.261 0.006 0.095 0.000

Cu*Hg - 0.160 0.035

As*Fe - 0.054 0.005

Pb*Mn 0.478 0.000

Fe*Hg 0.053 0.006

(-) refers to negative correlation

4.5 Flathead regional comparison

Mercury, arsenic, zinc, iron and copper were detected in muscle tissue of all flathead from all

of the five regions sampled. Spatial comparison of these key heavy metal levels between the

five regions indicated significant regional differences (Figure 11).

Mercury levels in flathead from the control region (mean 0.22 mg/kg) were significantly

(P<0.05) lower than in flathead from all other regions (Region 5), whereas levels in flathead

from Region 2 were significantly lower (0.33 mg/kg) than in flathead from Regions 1, 3 and

4, which all had similar mean mercury levels (0.62, 0.57 and 0.64 mg/kg respectively) (Figure

11a, Table 9 and 10). Levels in Regions 1, 3 and 4 all exceeded the FSANZ maximum

permitted level (Figure 11a).

41

Table 9. Summary of: mean age, fork length and weight, sex ratios and mean ± standard error and range (parenthesis) muscle tissue metal concentrations for sand flathead

from the five sample regions of the Derwent Estuary. BDL refers to below detectable level of analysis. Note: Muscle tissue levels of Cd, Co and Ni were all below the

detectable level of the analysis and so are not included in the table.

Biological Heavy metal levels (mg/kg wet wt.)

Region n

age

(years)

FL

(mm)

weight

(g)

sex ratio

(M:F) Hg As Zn Pb Fe Cu Mn Se Cr

1 30 6.3 293 195 8:15 0.62 3.74 5.22 0.07 2.23 0.12 0.01 <0.5 0.01

(3-12) (0.25-1.1) (1.0-7.6) (3.7-7.5) (0.0-1.0) (1.1-4.1) (0.0-0.3) (0.0-0.1) (0.0-0.2)

2 30 5.3 272 135 5: 8 0.33 4.54 5.61 0.01 1.74 0.12 0.00 <0.5 <0.1

(3-11) (0.03-0.66) (1.1-13.0) (4.2-7.3) (0.0-0.3) (1.1-4.3) (0.0-0.3) (0.0-0.1)

3 30 4.8 274 140 5: 9 0.57 4.42 6.21 0.11 2.96 0.21 0.12 <0.5 0.03

(2-7) (0.18-1.3) (0.5-16.0) (4.1-12.0) (0.0-1.7) (1.9-5.1) (0.1-0.3) (0.0-0.3) (0.0-0.5)

4 30 4.4 288 159 16: 13 0.64 2.44 7.60 0.08 3.31 0.33 0.12 0.06 0.01

(2-8) (0.27-1.4) (0.7-4.5) (4.9-17.0) (0.0-0.6) (1.7-8.9) (0.1-1.0) (0.0-0.8) (0.0-0.7) (0.0-0.2)

5 30 7.1 294 174 17: 13 0.22 2.39 5.69 0.11 3.01 0.35 0.31 0.06 <0.1

(3-13) (0.06-0.43) (0.5-9.4) (2.2-13.0) (0.0-0.3) (0.8-9.1) (0.2-0.7) (0.0-3.1) (0.0-0.6)

TOTAL 150 5.6 284 161 61: 75 0.53 3.91 5.89 0.06 2.57 0.16 0.05 0.02 0.01

(2-13) (0.1-1.4) (0.5-16.0) (3.7-12.0) (0.0-1.7) (1.1-13.0) (0.0-0.3) (0.0-0.3) (0.0-0.7) (0.0-0.5)

Figure 11. Mean muscle tissue metal concentrations ± SE (n = 30 flathead per region) for mercury, arsenic, zinc,

iron and copper between Regions 1 to 5 for flathead. Different letters indicate statistically significant (P<0.05)

differences between means from post hoc pooling analysis. Graph (a) also has a line indicating the FSANZ

maximum mercury level (0.5 mg/kg).

Mean arsenic levels were significantly lower in flathead from Regions 4 and 5 (2.44 and 2.39

mg/kg respectively) than Region 1 (3.74 mg/kg), while Regions 2 and 3 were significantly

higher (4.54 and 4.42 mg/kg respectively) (Figure 11b, Table 9 and 10). There was less

b) As

0

1

2

3

4

5

6

Region 1 Region 2 Region 3 Region 4 Region 5

As (

mg

/kg

)

a) Hg

0.0

0.2

0.4

0.6

0.8

1.0


Hg

(m

g/k

g)

e) Cu

0.0

0.1

0.2

0.3

0.4

0.5


Cu

(m

g/k

g)

d) Fe

0

1

2

3

4


Fe (

mg

/kg

)

c) Zn

0

2

4

6

8

10


Zn

(m

g/k

g)

C

B

C C

A

AB B B

A A

A A AB

B

A AB A

C C

BC

A A

B

C C

43

variability in mean zinc levels across the regions, however, Regions 3 and 4, were still

significantly (P<0.05) higher than the other regions, with Region 4 being higher (7.60 mg/kg)

than Region 3 (6.21 mg/kg) (Figure 11c, Table 9 and 10). Mean iron levels in Regions 3 and 4

were significantly (P<0.05) higher than Regions 1, 2 and 5 (Figure 11d and Table 10), and

mean copper levels were significantly higher in Regions 4 and 5 than the other regions

(Figure 11e and Table 10). Manganese, selenium and chromium were detected in only a few

individual flathead resulting in low mean levels (Table 9).

Table 10. Regional comparisons of concentrations of mercury, arsenic, zinc and lead in muscle tissue from

flathead by one-way ANOVA. Shown are mean squares, degrees freedom, f values and significance (p<0.05).

Metal Source df MS F P

Hg* Region 4 7.30 34.77 0.000

As Region 4 32.36 7.04 0.000

Zn Region 4 0.56 7.06 0.000

Fe Region 4 1.88 12.19 0.000

Cu Region 4 5.47 42.90 0.000


Principle components analysis shows the complexity of the relationships between the

individual flathead sampled. It was not possible to differentiate regions with respect to the full

metal suite (Figure 12). However, PCA did reveal that the greatest source of the variation

between individual fish was associated with differences in mercury levels, and to a lesser

extent levels of arsenic and zinc (Figure 12) which corresponds with the individual

comparisons (ANOVA) of metals across regions (Figure 11 and Table 10).

44

Figure 12. Ordination of sample sets using principal components analysis (PCA) on the similarity matrix

produced from the raw data. Vector plots show proportional influence of main metals on sample separation.

Mercury data transformed (×10).

Mercury was the only metal for which levels in flathead regularly exceeded maximum

permitted levels (Figure 11). This therefore is the most significant from a human health point

of view. Hence, in this section the interaction of mercury levels with age and length is

explored further. However, age/size ranges varied markedly between regions, for instance

flathead samples from Regions 1 and 5 were represented by a much larger range of age/sizes

than other regions, thus potentially confounding regional comparisons (Table 9). Size and age

structure of samples were not consistent between regions and as mercury is a function of these

parameters they need to be standardised in making valid regional comparisons. Regional

comparisons show that age was a better indicator of mercury concentration (Table 11). Thus,

in order to make valid comparisons between regions, age needed to be taken into account.

This was achieved by ANCOVA with age as the covariate. Results of the ANCOVA showed

that the slopes were not significantly different and that the age accounted for significant

variation between the regions (Table 12).

As Hg

Zn

Fe

45

Table 11. Linear regression results for mercury concentration in response to age and length for flathead from

different regions. Age and length as independent variables and mercury concentration the dependent variable.


Regression R2 P R

2 P R

2 P R

2 P R

2 P

Hg*Age 0.442 0.000* 0.588 0.000* 0.222 0.012* 0.159 0.029* 0.721 0.000*

Hg*Length 0.519 0.000* 0.282 0.003* 0.160 0.054 0.036 0.317 0.333 0.001*

* Indicates significance at the P<0.05 level

Table 12. Comparison of mercury levels in muscle tissue of flathead caught in different regions by one-way

ANCOVA, with length as the covariate and region as a fixed factor. Shown are, degrees of freedom (df), mean

square (MS), f value (F) and significance (P).

Source df MS F P

Region 4 1.81 13.92 0.000

Age (cov) 1 6.81 52.48 0.000

Region × Age

(cov) 4 0.256 1.97 0.102

ANCOVA without interaction term

Source df MS F P

Region 4 9.44 70.80 0.000

Age 1 10.93 81.95 0.000

Data were (Ln) transformed

Furthermore, mercury concentrations in flathead muscle were significantly different between

regions when age was taken into account (Table 12). A summary of the recorded means and

the ANCOVA adjusted means can be seen in Table 13. Examination of the adjusted means

revealed that when age was taken into account, the regional differences were more

pronounced and importantly the adjusted mean mercury levels for Regions 3 and 4 exceeded

the mean for region 1 (Table 13).

46

Table 13. Summary data showing mean mercury concentrations in flathead from different regions. Shown is

sample size (n), mean fish age, mean mercury level, range and adjusted mean (taking into account the covariate

of length). LS mean is the least square mean, calculated using age as a covariate.

Mercury in mg/kg wet wt

Region n Age (years Mean Range LS Mean

1 30 6.3 0.62 0.25-1.10 0.54

2 30 5.3 0.33 0.03-0.66 0.29

3 30 4.8 0.56 0.18-1.30 0.57

4 30 4.4 0.66 0.38-1.40 0.70

5 30 7.1 0.22 0.06-0.43 0.16

4.6 Trophic status of selected species

Gut analysis to identify main food groups for each of the fish species studied revealed marked

differences in their trophic status. Sand flathead were carnivores preying mainly on crabs and

fish (Table 14).

Table 14. Main food groups found in the gut of individuals for each species as well as trophic level.

Species Main food groups found in gut Trophic level

Sand flathead Crabs, fish Carnivore

Sea-run trout Fish, shrimps, insects Carnivore

Black bream Mussels (Mytilus edulis), crabs, weed Omnivore

Yellow-eye mullet Algae, sediment Herbivore/detritivore

Sea-run trout were also carnivores preying mainly on fish, shrimps and insects (Table 14).

Derwent Estuary black bream were classed as an omnivore with a diet consisting of mussels

(Mytilus edulis), crabs and macro algae (Table 14). Yellow-eye mullet were distinctly

different from the other species, with gut analysis showing that they mainly consume green

and red algae as well as sediment (Table 14) and therefore for the purposes of this study were

classed as a herbivore/detritivore species (Table 14).

47

5. Discussion

The results clearly showed that metal levels varied both between and within species, and in

relation to age, length and gender, and that regional differences in metal levels in flathead

muscle tissue and levels of mercury in three key recreational species exceeded Australian

seafood guidelines.

5.1 Factors influencing metal levels between species

Fish are exposed to heavy metal contamination both directly via the water column through

respiration (gills) and indirectly through diet (Burger et al., 2002; Bu-Olayan and Thomas,

2005). Of the two pathways, direct exposure from the water is thought to account for only a

minor proportion of the metal uptake by fish for most metals (Burger et al., 2002). Metal

levels in Derwent Estuary water are relatively low in comparison to sediments (Green and

Coughanowr, 2003), and therefore it could be concluded that water column uptake would

represent a relatively small component of overall uptake. Since the species comparison was

between fish sampled from only from Region 1 (Tasman Bridge to New Norfolk), it might be

expected (assuming they are non-migratory and stay within this region for the majority of

their lives) that they would be exposed to similar levels of water borne contaminants and that

any species differences would not be as a result of different metal exposures through the

water column. In this instance, species variability in metal levels is likely to be a result of

differences in dietary exposure (dietary preferences, trophic level), time of exposure (age) and

fish physiology (elimination) (Watras and Bloom, 1992).

48

From a human health perspective differences in mercury levels between species are the most

important comparison. Mercury is the one metal for which there is substantial evidence for

bioaccumulation and biomagnification in fish (Mason et al., 1995; Hill et al., 1996). It is

highly persistent and readily absorbed by most organisms and its accumulation in fish is

largely a result of exposure through diet (Mason et al., 1995). Trophic level has been found to

strongly influence mercury levels in fish, with species of high trophic level generally having

higher mercury concentrations than species of lower trophic status (Ratkowsky et al., 1975;

Mason et al., 1995).

Although bream were not the highest trophic species in this study, they had a substantially

higher mean mercury level than the other three species, suggesting that trophic level alone is

not the primary influence on different mercury levels between species. It is possible however,

that specific prey items within the bream diet may be particularly susceptible to mercury

accumulation and hence may have particularly influenced levels in this species. This study

suggested that the diet of bream primarily consisted of crabs, bivalves (blue mussels; Mytilus

edulis planulatus) and macrophytes, which is consistent with previous findings (Sarre et al.,

2000; R. Sakabe, pers. comm.). Consumption of the blue mussel was unique to the diet of

bream, and may be an important source of the resultant high mercury levels in this species.

Blue mussels are abundant throughout the Derwent and have been found to contain very high

levels of cadmium, lead and mercury (Bloom and Ayling, 1977; Green and Coughanowr,

2003). Bloom and Ayling (1977) reported the concentration of mercury in the blue mussel

from the Derwent Estuary as 0.02-1.3 mg/kg (mean 0.35 mg/kg) which is relatively high for a

small invertebrate (Bloom and Ayling, 1977). Consequently, the diet of blue mussels could be

a major contributor to the higher than expected mercury contamination levels in bream.

49

Given that both trout and flathead are regarded as carnivorous, opportunistic feeders it might

be expected that the metal levels in these species would be similar; indeed levels of mercury,

arsenic and zinc were not significantly different between the two species. However, levels of

iron, copper and manganese were all significantly higher in trout and these differences may

again be due to a preference towards particular dietary items. Flathead and trout both

consumed fish as part of their diets, but fish appeared to be a bigger component of the trout

diet than the flathead diet. This is supported by research which suggests that sea-run trout

follow whitebait runs and feed largely on other fish (DPIW, 2007). Assuming that metals are

accumulated through the food chain, then the higher proportion of fish in the trout diet could

account for the higher levels of copper, iron and manganese.

The status of mullet as the lowest trophic level of the four species was supported by the

significantly lower levels of mercury and arsenic recorded; for both metals there is evidence

of increased levels in higher trophic species. In contrast, trophic level cannot explain the

higher levels of lead and zinc found in this mullet, but again it may be due to specific dietary

differences relative to the other three species. Previous studies suggest that the diet of yellow-

eye mullet consists primarily of sediment, algae, detritus and small benthic zooplankton

(Eustace, 1974; Edgar, 1997). In the present study the stomachs of mullet were filled with

green algae and some contained small amounts of sediment, reflecting the fact that when

feeding, mullet suck up the surface layer of mud or graze on submerged rock and plant

surfaces (Eustace, 1974). Metals including lead and zinc are strongly tied in with sediments

(Eustace, 1974; Campbell, 1994) and as a result the high levels of these metals in mullet may

be a result of direct ingestion of these metals along with sediment. The strong, positive

correlation between lead and zinc levels in mullet found in this study would tend to support

this hypothesis. It has been proposed that significant correlations of metal concentrations in

50

tissues may reflect similar pathways of accumulation (Brooks and Rumsey, 1974) and as a

result may be associated with related sources of exposure, excretion or sequestration (Kirby,

2001). In addition, mullet may be exposed to higher levels of particular metals as a

consequence of the organisms they consume (plant and animal) being unable or having a

limited ability to regulate metals. In his review on heavy metal accumulation in marine

animals, Bryan (1980), suggested that in the accumulation of metals through the food chain, it

is of some importance whether or not a predator regulates metals and whether its diet consists

of organisms which do or do not regulate (Bryan, 1980). For example, flounder (Platichthys

flesus) from the Severn Estuary (Great Britain) which had a diet of Macoma balthica, a

marine bivalve which is unable to regulate metal levels, contained higher levels of zinc than

those having a diet of crustaceans and small fish which can regulate zinc (Hardisty et al.,

1974). The algae and small zooplankton consumed by mullet would only have a limited

capacity to regulate metal levels (Bryan, 1980) and as a result mullet may have a greater

exposure to metals through its diet than the other species.

The present study suggests that specific differences in diet are likely to be a major factor in

determining differences in metal levels between species. In particular it highlights the strong

influence trophic level may have on the levels of mercury and arsenic in a given species. It

also illustrates the importance of identifying primary prey groups of a particular species as

some prey items may account for a larger proportion of metal exposure than others.

Whilst fish diet on its own may explain a substantial amount of the variation in metal levels

between species, the length of time which a particular fish is exposed to heavy metals will

also greatly influence the final concentration of any given metal. Relationships between metal

concentration (in particular mercury, arsenic and zinc) and age and/or size have been reported

51

previously in many species of fish including; lake trout (Bache et al., 1971), black marlin

(Mackay et al., 1975), deep water sharks (Hornung et al., 1993), largemouth bass (Park &

Curtis, 1997) and pacific cod (Burger et al., 2007). Generally, muscle metal levels have been

shown to increase with age and size (Mackay et al., 1975; Hornung et al., 1993). However,

the one notable exception to this trend, is the negative relationship which has been observed

for zinc (Kirby et al., 2001; Farkas et al., 2003). In this study there were age/size related

differences in zinc, arsenic and mercury levels within flathead, bream and trout.

As a consequence of the persistent nature of mercury, excretion of this metal in fish is slow

(Miettinen, 1973; Sorensen, 1991). Mercury has a strong ability to accumulate over time,

reaching very high levels in long lived fish (Morel et al., 1998; Wiener et al., 2003). In the

present study, mercury levels in flathead and trout increased with age and length. This

relationship is consistent with findings of the majority of studies which have looked at the

influence of age on mercury levels (Hournung et al., 1993; Heuter et al., 1995; Szefer et al.,

2003). The relationship between length and mercury concentration was stronger for trout than

flathead, and this may simply be a reflection of the large size range of trout collected. The

relationships between age, length and mercury concentration in bream may have been clearer

had a larger size range of fish been collected for this species; all bream collected in this study

were in the age range of 13 to 28 years of age (Table 3). Although mercury levels were not

found to significantly increase with age in bream, the higher mercury levels found in this

species may be due to the relatively old age of the species examined. Longevity has been

attributed to the high mercury levels in many species including deep water shark (Hornung et

al., 1993), black marlin in Australia (Mackay et al., 1975), and perch in the Baltic Sea (Szefer

et al., 2003). Bream collected in this study were on average almost three times the age of

other species. However, the fish sampled may not have adequately represented the full age

52

range and future studies should seek to include a wider size range of fish including some

small, young bream. This would allow for a more thorough examination of age/length

relationships with mercury. The lack of any relationship between mercury concentration and

age or length for mullet is probably due to the low overall levels found in this species. Park

and Curtis (1997) suggested that size relationships may not hold for fish species with low

contaminant levels (Park and Curtis, 1997).

Although zinc levels were significantly higher in mullet than the other species this study

found no age/size relationship of zinc levels in mullet. However, zinc levels were found to

significantly decrease with increasing age in flathead. This trend has been previously noted by

several authors including Farkas et al. (2003) who found that zinc levels decreased

significantly with age in bream (Abramis brama L.) and Kirby et al. (2001) whose study

reported that zinc levels decreased with size (weight) in sea mullet (Mugil cephalus).

Negative trends may be due to older/larger fish being more effective in regulating zinc levels

(Farkas et al., 2003).

There are mixed reports in the literature regarding the relationship between age/size and

arsenic. Ashraf and Jaffar (1988) found arsenic levels in tuna increased with size (weight),

whilst, Liao et al. (2003) and Burger et al. (2007) found that arsenic levels in tilapia

(Oreochromis mossambicus) and pacific cod (Gadus macrocephalus) respectively, decreased

with fish size (weight and length). In the present study, arsenic levels increased significantly

with increasing length in bream and trout. If arsenic is being accumulated through diet and

hence increasing through the food chain, then levels would be expected to increase with size

or age. Ongoing conflicting findings on the accumulation of arsenic in fish with age, size and

trophic level, highlights this as a particular area for which more research is needed.

53

Gender specific differences in heavy metal concentration have been described for several

species of fish (Parks and Curtis, 1997; Alquezar et al., 2006). Gender may alter the metal

concentration in a particular species of fish through a combination of factors including dietary

preferences (Parks and Curtis, 1997; Peakall and Burger, 2003) and physiological differences

related to the reproductive cycle (Olsson et al., 1996). The results of this study showed

significant gender related differences in levels of arsenic, iron, copper and zinc in different

species.

Levels of arsenic were almost twofold higher in female fish for both flathead and bream. This

finding is in direct contradiction with several studies. Glover (1979) and Burger et al. (2004)

who studied school and gummy sharks from south eastern Australia, and Florida char

(Lepisosteus platyrhincus), respectively, found that arsenic levels did not differ significantly

with gender. It was suggested that this could simply be a reflection of the generally low levels

of arsenic in the fish (Burger et al., 2004). Arsenic levels in the present study appeared quite

high and therefore the differences may have been more apparent as a result. The opposite

relationship was present for levels of iron and copper, which were significantly higher for

male flathead than female or immature flathead. Higher iron levels in male fish, has been

noted by Alquezar et al. (2006) for toadfish (Tetractenos glaber), who suggested that this

might be due to a combination of factors including; dietary preference, reproductive

metabolism and foraging behaviour (Alquezar et al., 2006). Gender differences can be

expected to occur when one gender is less efficient at regulating essential metal levels than

another, this may be because more energy is being used to meet the demands for sexual

formation or development (Chernoff and Dooley, 1979; Kirby et al., 2001). At the time of

sampling, each of the study species was in reproductive condition and many females were

54

fully mature with ripe ovaries. The additional requirement for iron and copper by female

flathead in egg production, may account for the resultant lower muscle levels than in male

fish (Kirby et al., 2001). There were no gender related differences for mercury, the most toxic

of the metals, and hence it might be concluded that the gender differences were of little

importance from a human health perspective.

Life history, environmental and physiological requirements of certain species of fish mean

that fish may migrate or move over a range of spatial scales (Kestemont et al., 1999;

Cucherousset et al., 2005). As a result, more mobile species of fish may only spend a small

proportion of their time within any given region and therefore would not be exposed to the

same metal contamination, either through food or water, of less mobile species (Blevins and

Pancorbo, 1986). Consequently, comparisons of species from a particular region require

knowledge of the species mobility in order to accurately examine metal differences

(Francesconi et al., 1997). It is considered that sand flathead are a non-migratory species

which remain within relatively localised areas for their entire life cycle (Dix et al., 1975;

Francesconi et al., 1997; Jordan, 2001) and the significant regional differences in metal levels

in flathead from this study provides more evidence that this is indeed the case. If flathead in

the Derwent Estuary were highly mobile, spatial variability in mercury levels would have

been very small. Bream are thought to be restricted to a particular estuary though move

throughout the system (Potter and Hyndes, 1999) and a similar movement pattern is thought

to be occurring with bream in the Derwent Estuary (R. Sakabe, pers. comm.). The lack of

movement of this species out of the Derwent Estuary would suggest that high mercury levels

in the fish sampled are ultimately due to the contamination in the estuary. Examining mercury

levels in bream from other “uncontaminated” estuaries from around Tasmania to provide

background levels would determine if this is indeed the case. However, as a reference, bream

55

from the relatively uncontaminated Gippsland Lakes in Victoria had a mean mercury level of

0.22 mg/kg (Fabris et al., 1999). The comparative mean levels in bream in this study were

1.57 mg/kg further highlighting the extreme nature of mercury levels in bream from this

study. While there is some information on the movements of bream and flathead in the

Derwent Estuary, little is known about the movements of the other two species studied.

Mullet are primarily an estuarine species (Edgar, 1997) but little is known about their

movements within an estuary and the migratory patterns and movements up and down stream

of sea-run trout are poorly understood. Metal levels in these species may be influenced by

mobility, however, due a lack of knowledge of their movement it cannot be known how much

of a factor fish mobility is.

5.2 Regional variability in flathead

A multitude of studies have looked at metal levels in fish from contaminated sites around the

world and the general conclusion is that environment has a large influence on levels (Weiner

et al., 2003; Calta and Canpolat, 2006). The aquatic environment is highly susceptible to

contamination by heavy metals and other pollutants due to land run-off, industry effluent and

other forms of human activity and natural weathering (Laws, 2000; Calta and Canpolat,

2006). Aquatic sediments can act as both a sink and a source of heavy metal contaminants,

and long term input of heavy metals into an aquatic environment can lead to sediment

concentrations which greatly exceed the levels in the water column (Burton et al., 2003). The

Derwent Estuary has been subjected to long term heavy metal contamination from past

industrial practices which released highly contaminated effluent directly into the estuary for

many years (Green and Coughanowr, 2003). Consequently sediment metal levels in the

Derwent Estuary are among the highest in the world (Bloom, 1975). Sediment levels of zinc,

56

cadmium, lead, arsenic, copper and mercury all exceed national sediment standards by

considerable amounts in several regions, particularly in the area immediately adjacent to the

Zinifex Hobart Zinc Smelter (Region 1) (Table 15). In summary there is substantial variability

in metal loads in the sediments of the Derwent and it is likely that this spatial variability

would be reflected in the metal levels of fauna that are associated with the area.

Table 15. Mean sediment heavy metal levels in regions of the Derwent Estuary in mg/kg dry wt. Also shown are

Interim Sediment Quality Guidelines (ISQG). ISQG low, equivalent to effects range low- causing adverse effects

10% of the time. ISQG high, equivalent to effects range median- causing adverse effects 50% of the time. Data

courtesy of the Derwent Estuary Program, 2003. n refers to number of samples taken from each region.

Region n Hg As Cd Cu Fe Mn Pb Zn

1 14 27 316 115 479 44594 1521 2406 15257

2 8 5 14 4 51 18685 171 367 1003

3 7 2 10 2 19 10293 84 128 361

4 13 2 7 2 17 7140 33 124 325

ISQG low 0.15 20 1.5 65 50 200

ISQG high 1 70 10 270 220 410

If metal levels were a direct reflection of sediment levels then we would expect to see very

high metal levels from Region 1 (the most contaminated region) and a substantial decline in

levels as one progresses towards the mouth of the estuary. Supporting this is the significantly

lower mercury levels in flathead from the control region (Mickey’s Bay) than the other

regions. The mean mercury levels in flathead from the control region of 0.22 mg/kg compared

similarly with the levels found in flathead collected from the relatively pristine east coast of

Tasmania (0.24 mg/kg) (Thomson, 1985) suggesting that this region was indeed a good

reflection of background metal levels and reflected low sediment levels. However, within the

Derwent Estuary levels of mercury, zinc, iron and copper in flathead did not follow the

expected pattern apparent in the sediments. Ralphs Bay (Region 4) which is some distance

downstream from Region 1 had the highest levels of mercury, zinc and iron, whilst Region 1

which is the most contaminated region had moderately high levels. Furthermore, mercury

57

levels in fish from Region 2 (western shore) were significantly lower than Regions 1, 3 and 4.

Ratkowsky et al. (1975) also found significant differences in mercury levels between regions

and similar to this study, percentages of fish having mercury concentrations in excess of the

permissible value were higher in Ralph’s Bay (Region 4) and the surrounding eastern shore of

the estuary (Region 3) than in the remainder of the estuary. This trend has since been

observed consistently in seafood monitoring of mercury levels in flathead by Zinifex (Nystar)

in recent years (Green and Coughanowr, 2003). Furthermore, metal levels in bivalve molluscs

in the estuary, particularly oysters (Crassostrea gigas) and mussels (Mytilus edulis

planulatus) show similar regional trends to metal levels in flathead (Green and Coughanowr,

2003). Data on lead levels in mussels from the Derwent Estuary show that levels in Region 1

and Region 4 (Ralphs Bay) are much higher than levels in mussels from the other regions

(Green and Coughanowr, 2003). A similar trend is apparent for metal levels in oysters from

the Derwent Estuary (Green and Coughanowr, 2003) with oysters from Ralphs Bay typically

having the highest lead and mercury levels of the regions (Green and Coughanowr, 2003).

The findings of the present study on metal levels in Derwent Estuary fish and other studies on

metal levels in Derwent Estuary shellfish all highlight Ralphs Bay as a particular anomaly.

All of these studies have shown that the levels of mercury contamination in the Derwent are

not linked to sediment levels. Clearly other factors such as regional differences in diet, age

and metal bioavailability are of importance.

It is now well understood that the underlying factor behind mercury accumulation in fish is

the methylation of inorganic mercury to organic mercury (methylmercury) making it

bioavailable for uptake by fish (Weiner et al., 2003). Methylation is influenced by several

factors including temperature, pH, organic matter and sediment type (Jernelov and Ansell,

1973; Foster et al., 2000). While there is no simple relationship, it appears that enhanced rates

58

of methylation are linked in particular with low pH, low salinity, high temperature and the

presence of decomposable organic matter in reducing environments (Ullrich et al., 2001). It is

possible that some of these factors may be more favourable in Ralphs Bay than the other

regions of the Derwent Estuary and this may be favouring methylmercury production and

hence uptake and accumulation by flathead. Supporting this is that Ralphs Bay has some

distinct physical differences to the other regions in the Derwent Estuary. Firstly, much of the

bay is very shallow (less than 2m in depth) and the sediment at the mouth of the bay consists

largely of fine silt whilst within the bay sediments primarily consist of coarser sand (Jones et

al., 2003). The shallow waters of Ralphs Bay may allow for warmer sediments than the other

regions which in turn may be increasing methylation and hence bioavailability of

methylmercury (Foster et al., 2000; Filazi et al., 2003). Indeed, Jones et al. (2003) reported

that the water in the shallow areas of Ralphs Bay recorded higher salinities and temperatures

than the rest of the rest of the estuary. Furthermore, sediment temperatures were on average 4-

7oC higher in the shallows of Ralphs Bay compared with the rest of the estuary (Jones et al.,

2003). Sediments in Ralphs Bay were also found to be strongly reduced compared with the

cleaner sands of the estuary mouth (Jones et al., 2003). The more reduced sediments of

Ralphs Bay, which are also rich in organic matter, all provide an environment where

methylation rates can increase (Choi and Bartha, 1994) and could be contributing to greater

metal availability and thus higher rates of uptake by fish and shellfish.

The unique physical attributes of Ralphs Bay may mean that the waters in this region are

home to a different group of prey items to the other regions. Different prey items may have

higher metal levels and so be contributing to increased levels in fish. Preliminary findings on

the trophic transfer of metals to sand flathead from prey items by Hunt (unpublished data,

2008) suggest that this could indeed be contributing to higher mercury levels. His findings

59

also suggest that flathead in this Ralphs Bay may be selectively feeding on a certain prey

group which are more inclined to accumulate high metal levels than other species. This could

also be a contributing factor to the comparatively high mercury levels in flathead in Ralphs

Bay.

5.3 Implications for public health

From a public health perspective, mercury, cadmium and lead are considered the most

hazardous metals, however, several others including copper, zinc, silver, arsenic and

chromium may be of equal or greater hazard to the health of humans (Bryan, 1980; Jarup,

2003). Of these, mercury, lead, arsenic, copper and zinc were all detected in muscle tissue of

fish species from the Derwent Estuary. However, only mercury regularly exceeded maximum

permitted levels for human consumption, although levels of arsenic and lead occasionally

exceeded the guidelines. Consequently the major health concern for consumption of Derwent

Estuary fish would be mercury contamination.

Methylmercury contamination was responsible for large human casualties in Minimata, Japan

in the 1950’s where over 100 people died from mercury poisoning after ingesting

contaminated fish and shellfish (Kurland et al., 1960). From incidents such as these, health

experts have been able to reliably predict maximum allowable levels for safe human

consumption. For the majority of countries including Australia the maximum levels for all

fish with the exception of long lived and exceptionally large fish has been set at 0.5 mg/kg

(FSANZ, 2007). Mercury levels in flathead (Regions 1, 3 and 4), trout and bream exceeded

recommended mercury levels of 0.5 mg/kg. As a result of the long term monitoring of

mercury levels in flathead by the Zinifex (Nystar) monitoring program, a health advisory has

60

been issued regarding the consumption of flathead from the Derwent Estuary (DEP, 2007).

The advisory recommends that flathead from the Derwent Estuary should be consumed no

more than three times a week and pregnant women and children should limit their

consumption of the species to one meal per week. Levels in flathead from the sampling

program are consistent with those found in this study (Green and Coughanowr, 2003),

however, the findings of this study as well as data from the monitoring program suggest that

flathead from the western shore (including Kingston Beach and Sandy Bay) and from the

southern most end of South Arm (Seacroft Bay) pose a much smaller risk to human health

than the other areas. In contrast, flathead taken from Ralphs Bay probably pose the highest

risk to consumers. The present study has also highlighted the importance of standardising

mercury levels in flathead with age (or length) in order to provide accurate regional and

temporal comparisons. The fact that mercury levels identified in bream were threefold higher

than levels in flathead (for which a health advisory has been issued for) suggests that bream

from the Derwent Estuary are of particular concern to public health. Adding to the risk is the

fact that bream are highly sought after sport fish and are regularly consumed by anglers and

recreational fishers (Lyle, 2005). The present study has also identified that consumption of

trout may be a risk to consumers with mercury levels similar to those in flathead from the

most contaminated regions of the Derwent. However, it must also be noted that yellow-eye

mullet from the Derwent Estuary had relatively low mercury levels and are unlikely to pose a

health risk to humans.

The majority of arsenic (80% or more) in fish is the relatively non-toxic, organic arsenic,

arsenobetaine (Edmonds et al., 1977; Larsen et al., 1993), however, some individuals of

flathead and trout had arsenic levels which may have exceeded the maximum permitted level

of 2 mg/kg of inorganic arsenic. If the ratio of organic to inorganic arsenic is taken into

61

account, the average levels fell well below the ML. Consequently the health threats associated

with arsenic toxicity would be minimal in relation to the species examined in this study.

However, in future studies it would be desirable to undertake analysis of inorganic arsenic in

tissue samples to see whether the inorganic to organic arsenic ratio is comparable to findings

from previous studies. It may also be worthwhile to consider whether it is appropriate to base

the Australian guidelines on inorganic arsenic considering that analysis of total arsenic is both

expensive and time consuming and that the ratio of inorganic arsenic to organic arsenic is

well known in most fish.

The health effects of lead contamination in humans are potentially severe, and may include

behavioural disturbances, learning and concentration difficulties, with severe cases of

psychosis and reduced consciousness in worst case scenarios (Jarup, 2003). Consequently

elevated levels of lead recorded in mullet in this study may warrant some concern. Despite a

mean lead level in the species of 0.35 mg/kg which fell below maximum recommended lead

levels of 0.5 mg/kg, several individual fish had levels which exceeded this value, suggesting

that there could potentially be some negative health affects to humans eating yellow-eye

mullet. Whilst the risk could be considered minimal, it should be noted that the skin and

bones of this fish may have much higher levels of lead. Studies have reported higher lead

concentrations in fish scales and bones than in the other parts of the fish, as lead preferentially

accumulates in bone and calcium structures (Rashed, 2001). Consequently, people who eat

the skin of the fish as well as the muscle may be exposed to higher levels of lead, and

potentially be at greater risk to their health.

62

5.4 Conclusions

In summary, this study has revealed that metal levels varied between the different species

with diet/trophic level, age and mobility the key factors. Gender differences were apparent but

only for arsenic, iron and zinc and consequently gender was of minor importance with respect

to human health risk. Regional differences in several metal levels were apparent for flathead

but these differences did not reflect regional differences in sediment levels. Metal levels in

flathead from Ralphs Bay were particularly high, with mercury and zinc highest in fish from

this region despite the sediment levels of these metals being comparatively low. It was

hypothesised that diet and mechanisms influencing metal bioavailability were likely to be the

main factors in the regional differences, particularly for mercury. Mercury in three of the

species studied (black bream, sea-run trout and sand flathead) consistently exceeded

Australian guidelines for levels in seafood and may pose a health risk. However, reported

levels of arsenic and lead are likely of little concern to human health.

5.5 Future research and management implications

This study has identified some particular areas that require more research in order to improve

our overall understanding of metal accumulation and to better aid the management of

Derwent Estuary fish, human health and estuary health. Perhaps the most significant finding

of the study was the high mercury levels in bream. The levels found in this species were three

times higher than levels in flathead from the same region where there are already advisories

regarding their consumption. As a result of these findings a public health advisory from the

Tasmanian Department of Health was issued recommending that bream not be consumed at

all and that consumption of trout and flathead from the Derwent Estuary should be limited. As

a consequence of this study, the Department of Health and other Derwent Estuary

63

stakeholders are keen to identify any additional species that may represent a risk to human

health. This study has indicated that susceptible species would be those which are long lived,

high trophic order species and which live in highly contaminated regions. This study has

identified a need to better understand the influences of metal levels in the species studied in

this project. Although the present study has provided evidence for the influence of various

factors such as diet and mobility on metal levels in a particular species, further research is

needed to better understand the mechanisms behind accumulation. Further studies should look

to compare mercury levels found in Derwent Estuary trout and bream to levels in these

species from other areas in the state not impacted by metal contamination to determine

whether or not the high mercury levels are site specific or species specific. Further work

should also look to better categorise the diets of bream and trout and also measure metal

levels in key prey items of each species to examine trophic transfer. Finally, this study has

highlighted the need to better understand the movements and dispersal patterns of fish species

both within the Derwent Estuary and in other waters if applicable. Researchers at the Marine

Research Laboratories are currently undertaking an acoustic tagging study for bream and

flathead which should shed some light on the movements of these species.

The higher mercury levels observed in flatheads from Ralphs Bay compared to other regions

of the Derwent Estuary is of particular interest. This study has put forward several possible

explanations for this anomaly, and these should be explored further. Research in this area

would be best focused on identifying environmental and biological parameters which are

unique to the region and which may be favouring the bioavailability of metals. Parameters to

be measured should include: water depth, water and sediment temperature, organic matter

levels in the water column and sediments, and identification of metal input sources. This will

enable a better understanding of the pathways by which accumulation is occurring in fish

64

species from the Derwent Estuary as well as the underlying mechanisms which influence

uptake rates of metals. In addition there is also a need to determine metal levels in other

species such as flounder which are commonly taken from the area and may be a particular

“risk” species to human health due to their benthic habitat and the fact that they are regularly

taken and consumed from Ralphs Bay.

In conclusion this study has clearly identified some significant management issues for the

Derwent Estuary. Health management research priorities may involve identifying “at risk”

groups of people and determining how best to make the health risks known to these people.

Whilst, environmental management research may involve looking at the potential affects of

any proposed activity on the bioavailability of metals and the uptake by fish.

65

6. References

Adams, D.H., McMichael, R.H., 1999. Mercury levels in four species of sharks from the Atlantic coast

of Florida. Fisheries Bulletin 97, 372-379.

Alquezar, R., Markich, S.J., Booth, D.J., 2006. Metal accumulation in the smooth toadfish,

Tetractenos glaber, in estuaries around Sydney, Australia. Environmental Pollution 142, 123-

131.

Ashraf, M., Jaffar, M., 1988. Weight dependence of arsenic concentration in the Arabian Sea tuna fish.

Bulletin of Environmental Contamination and Toxicology 40, 219-225.

Asuquo, F.E., Ewa-Oboho, I., Asuquo, E.F., Udo, P.J., 2004. Fish species used as biomarker for heavy

metal and hydrocarbon contamination for Cross River, Nigeria. The Environmentalist 24, 29-

37.

Bache, C.A., Gutemann, W.H., Lisk, D.J., 1971. Residues of total mercury and methylmercuric salts

in lake trout as a function of age. Science 172, 951-952.

Beamish, R.J., Fournier, D.A., 1981. A method for comparing the precision of a set of age

determinations. Canadian Journal of Fisheries and Aquatic Sciences 38, 982-983.

Birch, G.F., 2000. Marine pollution in Australia, with special emphasis on central New South Wales

estuaries and adjacent continental margin. International Journal of Environment and Pollution

13, 573-607.

Blevins, R.D., Pancorbo, O.C., 1986. Metal concentrations in muscle of fish from aquatic systems in

east Tennessee, USA. Water, Air, & Soil Pollution 29, 361-371.

Bloom, H., 1975. Heavy metals in the Derwent Estuary. Chemistry Department, University of

Tasmania, pp. 1-121.

Bloom, H., Ayling, G.M., 1977. Heavy metals in the Derwent Estuary. Environmental Geology 2, 3-

22.

Brooks, R.R., Rumsey, D., 1974. Heavy metals in some New Zealand commercial sea fishes. NZ J.

Mar. Freshw. Res. 8, 155-166.

Bryan, G.W., 1980. Recent trends in research on heavy-metal contamination in the sea. Helgoland

Marine Research 33, 6-25.

Bu-Olayan, A.H., Thomas, B.V., 2005. Toxicity and bioaccumulation of heavy metals in mullet fish

Liza klunzingeri (Mugilidae: Perciformes). Chemistry in Ecology 21, 191-197.

Burger, J., 2007. Heavy Metals in Pacific Cod (Gadus macrocephalus) from the Aleutians: Location,

Age, Size, and Risk. Journal of Toxicology and Environmental Health, Part A 70, 1897-1911.

66

Burger, J., Gaines, K.F., Boring, C.S., Stephens, W.L., Snodgrass, J., Dixon, C., McMahon, M.,

Shukla, S., Shukla, T., Gochfeld, M., 2002. Metal levels in fish from the Savannah River:

Potential hazards to fish and other receptors. Environmental Research 89, 85-97.

Burger, J., Gochfeld, M., 2007. Knowledge about fish consumption advisories: A risk communication

failure within a university population. Science of the Total Environment, The.

Burger, J., Orlando, E.F., Gochfeld, M., Binczik, G.A., Guillette, L.J., 2004. Metal levels in tissues of

Florida gar (Lepisosteus Platyrhincus) from Lake Okeechobee. Environmental Monitoring and

Assessment 90, 187-201.

Burger, J., Pflugh, K.K., Lurig, L., Von Hagen, L.A., Von Hagen, S., 1999b. Fishing in urban New

Jersey: Ethnicity affects information sources, perception, and compliance. Risk Analysis 19,

217-229.

Burger, J., Stephens, W.L., Boring, C.S., Kuklinski, M., Gibbons, J.W., Gochfeld, M., 1999a. Factors

in exposure assessment: ethnic and socioeconomic differences in fishing and consumption of

fish caught along the Savannah River. Risk Analysis 19, 427-438.

Burton, G.A., Denton, D.L., Ho, K., Ireland, D.S., 2003. Sediment toxicity testing: issues and

methods. In: Hoffman, D.J., Rattner, B.A., Burton, G.A., Cairns, J. (Eds), Handbook of

Ecotoxicology (2nd

edition). Lewis Publishers, New York, pp. 111-150.

Calta, M., Canpolat, O., 2006. The comparison of three cyprinid species in terms of heavy metals

accumulation in some tissues. Water Environment Research 78, 548-551.

Campbell, K.R., 1994. Concentrations of heavy metals associated with urban runoff in fish living in

stormwater treatment ponds. Archives of Environmental Contamination and Toxicology 27,

352-356.

Carvalho, M.L., Santiago, S., Nunes, M.L., 2005. Assessment of the essential element and heavy metal

content of edible fish muscle. Analytical and Bioanalytical Chemistry 382, 426-432.

Chernoff, B., Dooley, J., 1979. Heavy metals in relation to the biology of the mummichog, Fundulus

heteroclitus. Journal of Fisheries Biology 14, 309-328.

Choi, S.C., Bartha, R., 1994. Environmental factors affecting mercury methylation in estuarine

sediments. Bulletin of Environmental Contamination and Toxicology 53, 805-812.

Cooper, R.J., Langlois, D., Olley, J., 1982. Heavy metals in Tasmanian shellfish. Journal of Applied

Toxicology 2, 99-109.

Correll, D.L., 1978. Estuarine productivity. BioScience 28, 646-650.

Cucherousset, J., Ombredane, D., Charles, K., Marchand, F., Bagliniere, J.L., 2005. A continuum of

life history tactics in a brown trout (Salmo trutta) population. Canadian Journal of Fisheries

and Aquatic Sciences 62, 1600-1610.

Curtis, T.D., Shima, J.S., 2005. Geographic and sex-specific variation in growth of yellow-eyed

mullet, Aldrichetta forsteri, from estuaries around New Zealand. New Zealand Journal of

Marine and Freshwater Research 39, 1277-1285.

67

DEP (Derwent Estuary Program), Should I eat shellfish and flathead from the Derwent?, Health

advisory for recreational fishers, July 2007.

Dix, T.G., Martin, A., Ayling, G.M.., Wilson, K.C., Ratkowsky, D.A., 1975. Sand flathead

(Platycephalus bassensis) an indicator species for mercury pollution in Tasmanian waters.

Marine Pollution Bulletin 6, 142-143.

DPI (New South Wales Department of Primary Industries), Fishing and Aquaculture, Common

Recreational Fish Species. Dec 2005.

<http://www.dpi.nsw.gov.au/fisheries/recreational/saltwater-fishing/sw-species/tiger-

flathead2 >. [accessed 2007 Dec 10]

DPIW (Tasmania Department of Primary Industries and Water), Sea Fishing and Aquaculture Fish

Fact Sheets- Species Information. October 2007.

< http://www.dpiw.tas.gov.au/inter.nsf/WebPages/LVAE-57H8NL?open>. [accessed 2007 Dec

10]

DPIW (Tasmanian Department of Primary Industries and Water), Sea Fishing and Aquaculture Hot

Fishing Spots- Southern Tasmania. October 2007.

< http://www.dpiw.tas.gov.au/inter.nsf/WebPages/ALIR-4YB6HE?open> [accessed 2007 Nov

10]

DPIW (Tasmanian Department of Primary Industries and Water), Sea Fishing and Aquaculture Size

Limits. January 2008.

< http://www.dpiw.tas.gov.au/inter.nsf/WebPages/ALIR-4YAVA8?open> [accessed 2008 Feb

8]

Edgar, G.J., 1997. Australian Marine Life; the plants and animals of temperate waters. Reed Books,

Victoria, 402-500.

Edgar, G.J., Barrett, N.S., Graddon, D.J., 1999. A classification of Tasmanian estuaries and

assessment of their conservation significance using ecological and physical attributes,

population and land use. Technical Report. Marine Research Laboratories, TAFI,

Tasmania.

Edmonds, J.S., Francesconi, K.A., Cannon, J.R., Raston, C.L., Skelton, B.W., White, A.H., 1977.

Isolation, crystal structure and synthesis of arsenobetaine, the arsenical constituent of the

western rock lobster Panulirus longipes cygnus George. Tetrahedron Letters 18, 1543-1546.

Egeland, G.M., Middaugh, J.P., 1997. Balancing fish consumption benefits with mercury exposure.

Science 278, 1904.

Eustace, I.J., 1974. Zinc, cadmium, copper and manganese in species of finfish and shellfish caught in

the Derwent Estuary, Tasmania. Australian Journal of Marine and Freshwater Research 25,

209-220.

Ewing, G.P., Lyle, J.M., Murphy, R.J., Kalish, J.M., Ziegler, P.E., 2007. Validation of age and growth

in a long-lived temperate reef fish using otolith structure, oxytetracycline and bomb

radiocarbon methods. Marine and Freshwater Research 58, 1-12.

Fabris, G.J., Monahan, C., Nicholson, G., Walker, T.I., 1992. Total mercury concentrations in sand

flathead, Platycephalus bassensis Cuvier & Valenciennes, from Port Phillip Bay, Victoria.

Australian Journal of Marine and Freshwater Research 43, 1393-1402.

68

Fabris, G.J., Theodoropoulos, T., Sheehan, A., Abbott, B., 1999. Mercury levels and organochlorines

in black bream, Acanthopagrus butcheri, from the Gippsland Lakes, Victoria, Australia:

Evidence for temporal increases in mercury levels. Marine Pollution Bulletin 38, 970-976.

Farkas, A., Salánki, J., Specziár, A., 2003. Age-and size-specific patterns of heavy metals in the

organs of freshwater fish Abramis brama L. populating a low-contaminated site. Water

Research 37, 959-964.

Filazi, A., Baskaya, R., Kum, C., Hismiogullari, S.E., 2003. Metal concentrations in tissues of the

Black Sea fish Mugil auratus from Sinop-Icliman, Turkey. Human & Experimental

Toxicology 22, 85.

Foster, E.P., Drake, D.L., DiDomenico, G., 2000. Seasonal changes and tissue distribution of mercury

in Largemouth bass (Micropterus salmoides) from Dorena Reservoir, Oregon. Archives of

Environmental Contamination and Toxicology 38, 78-82.

Francesconi, K.A., Lenanton, R.C.J., Caputi, N., Jones, S., 1997. Long-term study of mercury

concentrations in fish following cessation of a mercury-containing discharge. Marine

Environmental Research 43, 27-40.

FSANZ (Food Standards Australia and New Zealand). Regulations for mercury in fish. March 2004.

<http://www.foodstandards.gov.au/newsroom/factsheets/factsheets2004/mercuryinfishfurther2

394.cfm>. [accessed 2008 Feb 2]

Gislason, H., Sinclair, M., Sainsbury, K., O’Boyle, R., 2000. Symposium overview: incorporating

ecosystem objectives within fisheries management. Journal of Marine Science 57, 468-475.

Glover, J.W., 1979. Concentrations of arsenic, selenium and ten heavy metals in School shark,

Galeorhinus australis (Macleay), and Gummy shark, Mustelus antarcticus Günther, from

south-eastern Australian Waters. Australian Journal of Marine and Freshwater Research 30,

505-510.

Graynoth, E., 1996. Determination of the age of Brown and Rainbow trout in a range of New Zealand

lakes. Marine & Freshwater Research 47, 749-756.

Green, G., Coughanowr, C., 2003. State of the Derwent Estuary 2003: a review of pollution sources,

loads and environmental quality data from 1997 - 2003. Derwent Estuary Program, DPIWE,

Tasmania.

Han, B.C., 1998. Estimation of target hazard quotients and potential health risks for metals by

consumption of seafood in Taiwan. Archives of Environmental Contamination and

Toxicology 35, 711-720.

Hardisty, M.W., Huggins, R.J., Kartar, S., Sainsbury, M., 1974. Ecological implications of heavy

metal in fish from the Severn Estuary. Marine Pollution Bulletin 5, 12-15.

Hill, W.R., Stewart, A.J., Napolitano, G.E., 1996. Mercury speciation and bioaccumulation in lotic

primary producers and primary consumers. Canadian Journal of Fisheries and Aquatic

Sciences 53, 812-819.

Hornung, H., Krom, M.D., Cohen, Y., Bernhard, M., 1993. Trace metal content in deep-water sharks

from the eastern Mediterranean Sea. Marine Biology 115, 331-338.

69

Hueter, R.E., Fong, W.G., Henderson, G., French, M.F., Manire, C.A., 1995. Methylmercury

concentration in shark muscle by species, size and distribution of sharks in Florida coastal

waters. Water, Air, & Soil Pollution 80, 893-899.

Hylland, K., Beyer, J., Berntssen, M., Klungsøyr, J., Lang, T., Balk, L., 2006. May organic pollutants

affect fish populations in the North Sea? Journal of Toxicology and Environmental Health,

Part A 69, 125-138.

IFS (Inland Fisheries Service), Fact Sheet for Brown Trout. August 2006. <http://www.ifs.tas.gov.au/ifs/IFSDatabaseManager/SpeciesDatabase/brown-trout> [accessed

2007 Oct 23]

IFS (Inland Fisheries Service), Fisheries Performance Assessment Program Report- Tombs Lake

2003.

Järup, L., 2003. Hazards of heavy metal contamination. British Medical Bulletin 68, 167-182.

Jernelov, A., Asell, B., 1973. The feasibility of restoring mercury-contaminated waters. In: Krenkel,

P.A. (Ed), Heavy Metals in the Aquatic Environment. Pergamon Press, New York, pp. 299-

309.

Jones, B.G., Chenhall, B.E., Debretsion, F., Hutton, A.C., 2003. Geochemical comparisons between

estuaries with non-industrialised and industrialised catchments: the Huon and Derwent River

estuaries, Tasmania. Australian Journal of Earth Sciences 50, 653-667.

Jordan, A.R., 2001. Reproductive biology, early life-history and settlement distribution of sand

flathead (Platycephalus bassensis) in Tasmania. Marine & Freshwater Research 52, 589-601.

Jordan, A.R., Mills, D.M., Ewing, G.P., Lyle, J.M., 1998. Assessment of inshore habitats around

Tasmania for life-history stages of commercial finfish species.

Kestemont, P., 1999. Spawning migrations, sexual maturity and sex steroid levels in female roach

Rutilus rutilus from the River Meuse. Aquatic Sciences-Research Across Boundaries 61, 111-

121.

Kirby, J., 2001. Changes in selenium, copper, cadmium, and zinc concentrations in Mullet (Mugil

cephalus) from the southern basin of Lake Macquarie, Australia, in response to alteration of

coal-fired power station fly ash handling procedures. Archives of Environmental

Contamination and Toxicology 41, 171-181.

Kurland, L.T., Faro, S.N., Seidler, H., 1960. Minamata disease. World Neurology, 1, 370-390.

Larsen, E.H., Pritzl, G., Hansen, S.H., 1993. Arsenic speciation in seafood samples with emphasis on

minor constituents: An investigation using high-performance liquid chromatography with

detection by inductively coupled plasma mass spectrometry. Journal of Analytical Atomic

Spectrometry 8, 1075-1084.

Laws, E.A., 2000. Aquatic Pollution: An Introductory Text (3rd

Edition). John Wiley & Sons, Inc.

New York, pp. 369-425.

Liao, C.M., Chen, B.C., Singh, S., Lin, M.C., Liu, C.W., Han, B.C., 2003. Acute toxicity and

bioaccumulation of arsenic in tilapia (Oreochromis mossambicus) from a blackfoot disease

area in Taiwan. Environmental Toxicology 18, 252-259.

70

Lyle, J.M., 2005. 2000/01 Survey of Recreational Fishing in Tasmania. Tasmanian Aquaculture and

Fisheries Institute.

MacKay, N.J., Kazacos, M.N., Williams, R.J., Leedow, M.I., 1975. Selenium and heavy metals in

black marlin. Marine Pollution Bulletin 6, 57-61.

Mason, R.P., Reinfelder, J.R., Morel, F.M.M., 1995. Bioaccumulation of mercury and methylmercury.

Water, Air, & Soil Pollution 80, 915-921.

Miettinen, J.K., 1973. The accumulation and excretion of heavy metals in organisms. In: Krenkel, P.A.

(Ed), Heavy Metals in the Aquatic Environment. Pergamon Press, New York, pp. 163-165.

Morel, F.M.M., Kraepiel, A.M.L., Amyot, M., 1998. The chemical cycle and bioaccumulation of

mercury. Annual Review of Ecology and Systematics 29, 543-566.

Morison, A.K., Coutin, P.C., Robertson, S.G., 1998. Age determination of black bream,

Acanthopagrus butcheri (Sparidae), from the Gippsland Lakes of south-eastern Australia

indicates slow growth and episodic recruitment. Marine & Freshwater Research 49, 491-498.

Morton, A., Lyle, J.M., Welsford, D.C., 2005. Biology and status of key recreational finfish in

Tasmania. TAFI report No. 25.

Olsson, P.E., Larsen, A., Haux, A., 1996. Influence of seasonal changes in water temperature on

cadmium inducibility of hepatic and renal metallothionein in Rainbow Trout. Marine

Environmental Research 42, 41-44.

Park, J.G., Curtis, L.R., 1997. Mercury distribution in sediments and bioaccumulation by fish in two

Oregon reservoirs: Point-source and nonpoint-source impacted systems. Archives of

Environmental Contamination and Toxicology 33, 423-429.

Peakall, D., Burger, J., 2003. Methodologies for assessing exposure to metals: speciation,

bioavailability of metals and ecological host factors. Ecotoxicology and Environmental Safety

56, 110-121.

Pond.dnr.cornell.Edu: Sep 2007.

<http://pond.dnr.cornell.edu/nyfish/Salmonidae/brown_trout.html&h=382&w=1000&sz=262

&hl=en&start=1&um=1&tbnid=B6JmNrY3tp1gPM:&tbnh=57&tbnw=149&prev=/images%

3Fq%3Dbrown%2Btrout%26um%3D1%26hl%3Den%26sa%3DG>. [accessed 2007 Dec 10]

Potter, I.C., Hyndes, G.A., 1999. Characteristics of the ichthyofaunas of southwestern Australian

estuaries, including comparisons with holarctic estuaries and estuaries elsewhere in temperate

Australia: a review. Australian Journal of Ecology 24, 395-421.

Pourang, N., 1995. Heavy metal bioaccumulation in different tissues of two fish species with regards

to their feeding habits and trophic levels. Environmental Monitoring and Assessment 35, 207-

219.

Rashed, M.N., 2001. Cadmium and lead levels in fish (Tilapia Nilotica) tissues as biological indicator

for lake water pollution. Environmental Monitoring and Assessment 68, 75-89.

Ratkowsky, D.A., Dix, T.G., Wilson, K.C., 1975. Mercury in fish in the Derwent Estuary, Tasmania,

and its relation to the position of the fish in the food chain. Australian Journal of Marine and

Freshwater Research 26, 223-231.

71

Sarre, G.A., Platell, M.E., Potter, I.C., 2000. Do the dietary compositions of Acanthopagrus butcheri

in four estuaries and a coastal lake vary with body size and season and within and amongst

these water bodies? Journal of Fish Biology 56, 103-122.

Sorensen, E.M.B., 1991. Metal Poisoning in Fish, CRC Press, Boston, pp. 1-353.

Szefer, P., Domagala-Wieloszewska, M., Warzocha, J., Garbacik-Wesolowska, A., Ciesielski, T.,

2003. Distribution and relationships of mercury, lead, cadmium, copper and zinc in perch

(Perca fluviatilis) from the Pomeranian Bay and Szczecin Lagoon, southern Baltic. Food

Chemistry 81, 73-83.

Thomson, J.D., 1985. Mercury concentrations of the axial muscle tissue of some marine fishes of the

continental shelf adjacent to Tasmania. Australian Journal of Marine and Freshwater Research

36, 509-517.

Thrower, S., Eustace, I., 1973. Heavy metals in Tasmanian oysters in 1972. Australian Fisheries, Oct

1973, 7-10.

Toth, J.F., Jr., and R.B., Brown., 1997. Racial and gender meanings of why people participate in

recreational fishing. Leisure Science 19, 129 –136.

Ullrich, S.M., Tanton, T.W., Abdrashitova, S.A., 2001. Mercury in the aquatic environment: A review

of factors affecting methylation. Critical Reviews in Environmental Science and Technology

31, 241-293.

Walker, T. I.., 1982. Effects of length and locality on the mercury content of blacklip abalone,

Notohaliotis ruber (Leach), blue mussel, Mytilus edulis planulatus (Lamarck), sand flathead,

Platycephalus bassensis Cuvier and Valenciennes, and yank flathead, Platycephalus

caeruleopuntatus (McCulloch), from Port Phillip Bay, Victoria. Australian Journal of Marine

and Freshwater Research 33, 553-560.

Watras, C.J., Bloom, N.S., 1992. Mercury and methylmercury in individual zooplankton: Implications

for bioaccumulation. Limnology and Oceanography 37, 1313-1318.

Wiener, J.G., Krabbenhoft., D.P., Heinz, G.H., Scheuhammer, A.M., 2003. Ecotoxicology of mercury.

In: Hoffman, D.J., Rattner, B.A., Burton, G.A., Cairns, J. (eds), Handbook of Ecotoxicology

(2nd

Edition). Lewis Publishers, New York, pp. 409-443.

Wild Thing Adventures, Map of Bruny Island. 2008.

< http://www.wildthingadventures.com.au/ >. [accessed 2008 Jan 15]

Williamson, R.B., Morrisey, D.J., 2000. Stormwater contamination of urban estuaries. 1. Predicting

the build-up of heavy metals in sediments. Estuaries 23, 56-66.

72

Appendix 1

Heavy metal analysis protocols

Program No. 2

1. Step to 30oC

2. Hold for 600 mins

3. Ramp to 100oC at 1

oC/min


5. Cool to room temperature

Program No. 3

1. Step to 30oC


3. Ramp to 100oC at 1

oC/min



Program No. 4

1. Step to 30oC

2. Hold 600 mins

3. Ramp 97oC at 1

oC/min

4. Hold 180 mins


AST quality control samples (as outlined by AST methods)

Heavy metal suite- quality control samples run:

One standard reference material (SRM) with each batch of samples

One reagent blank per batch

One blank matrix spike per batch by addition of 1000 µL of Multi-element

standard to blank reagents

One sample duplicate per batch (min)

One sample matrix spike per batch (min) by addition of 1000 µL of multi-

element spiking standard to a sample

Acceptance/rejection of results:

Sample preparation blanks should be less than the MRL (minimum readable

level)

Blank matrix spikes must be within 25% of the theoretical value

Matrix spikes must be within 25% of the theoretical value

73

Duplicates must be within 20% of each other

SRM must be within three standard deviations from the mean with no more

than two consecutive results between two and three standard deviations from

the mean value

Mercury- quality control samples run:

One preparation blank per analysis batch

One blank recovery per analysis batch

One duplicate sample for every 20 samples

One sample recovery for every 20 samples (sample recoveries spiked with 500

µL of spiking standard

A calibration verification standard is to be run immediately after the

calibration at intervals of at least 20 samples

Acceptance/rejection of results:

The preparation blank should be less than or equal to the MRL

Duplicates should agree to within 20% of the mean

Samples and blank recoveries should be within 25% of the theoretical value of

1.00 µg/L

Calibration verification standards must be within three standard deviations

from the mean with no more than two consecutive results between two and

three standard deviations from the mean value

74

Appendix 2

Plots of between metal correlations

d) Bream

R2 = 0.1637

0

5

10

15

20

25

30

35

0 5 10 15 20 25

Zn (mg/kg)

Fe

(m

g/k

g)

b) Trout

R2 = 0.2112

0

2

4

6

8

0.0 0.2 0.4 0.6 0.8

Cu (mg/kg)

Fe

(m

g/k

g)

c) Bream

R2 = 0.1598

0.0

0.5

1.0

1.5

2.0

2.5

0.0 0.1 0.2 0.3 0.4

Cu (mg/kg)

Hg

(m

g/k

g)

a) Trout

R2 = 0.2641

0

5

10

15

20

0 2 4 6 8

Fe (mg/kg)

Zn

(m

g/k

g)

e) Mullet

R2 = 0.2612

0

2

4

68

10

12

14

0.0 0.2 0.4 0.6 0.8

Cu (mg/kg)

Fe

(m

g/k

g)

f) Mullet

R2 = 0.4779

0.0

0.5

1.0

1.5

2.0

0.0 0.2 0.4 0.6 0.8 1.0

Mn (mg/kg)

Pb

(m

g/k

g)

g) Mullet

R2 = 0.1635

0.0 0.2 0.4 0.6 0.8 1.0

Mn (mg/kg)

Zn

(m

g/k

g)

h) Mullet

R2 = 0.3768

0

5

10

15

20

0.0 0.5 1.0 1.5 2.0

Pb (mg/kg)

Zn

(m

g/k

g)

75

Figure 15. Significant correlations of inter-metal relationships of mercury, arsenic, copper, iron, manganese and

lead within species.

i) Flathead

R2 = 0.0535

0

2

4

6

8

10

0 5 10 15 20

Fe (mg/kg)

As

(m

g/k

g)

j) Flathead

R2 = 0.0954

0

2

4

6

8

10

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Cu (mg/kg)

Fe

(m

g/k

g)

k) Flathead

R2 = 0.0615

0

5

10

15

20

0.0 0.2 0.4 0.6 0.8 1.0 1.2

Cu (mg/kg)

Zn

(m

g/k

g)

l) Flathead

R2 = 0.2498

0

2

4

6

8

10

0 5 10 15 20

Zn (mg/kg)

Fe

(m

g/k

g)

m) Flathead

R2 = 0.0528

0

2

4

6

8

10

0.0 0.5 1.0 1.5

Hg (mg/kg)

Fe

(m

g/k

g)

77

Appendix 3

Mercury accumulation in marine fish: A literature

review

Jeremy Verdouw

Literature Review submitted in partial fulfilment of the requirements

for Honours in Aquaculture

National Centre for Marine Conservation and Resource Sustainability

University of Tasmania

2008

Word count: 4200

78

1. Introduction

Heavy metals include the transitional metals (eg. cobalt, copper, iron and manganese) and the

metalloids (eg. arsenic, cadmium, lead, mercury, selenium and tin) (Kennish, 2001). The

transitional metals are essential in low concentrations for normal function in organisms but

may be toxic at high concentrations (Kennish, 2001; Carvalho et al., 2005), whereas the

metalloids are not generally required for normal function in organisms and are toxic at low

concentrations (Kennish, 2001; Jarup, 2003). Another group of metal compounds called

organometals (eg. alkylated lead, tributyl tin and methylmercury) are particularly toxic and

present a danger to marine organisms as well as humans consuming seafood (Kennish, 2001).

Of all the heavy metals, mercury and its compounds are considered the most toxic and hence

have been the most studied (Morel et al., 1998; Carvalho et al., 2005).

Mercury is a volatile metal which occurs in a liquid state at room temperature (Berlin, 1979).

It is found in minute concentrations in all natural systems; however, it is a non essential heavy

metal which is toxic to humans and animals in trace amounts (Holden, 1973; Berlin, 1979).

Mercury is easily evaporated into the atmosphere and is also highly persistent when in the

natural environment. These properties mean that it can be atmospherically transported over

wide areas and thus it is an important environmental contaminant (Foster et al., 2000;

Carvalho et al., 2005). This, along with the fact that mercury and its compounds are readily

absorbed and retained, means human exposure and possible toxicity is likely.

The known health effects of mercury on humans are many and varied (Berlin, 1979; Table 1).

All forms of mercury are toxic and have been associated with a range of serious human health

79

problems and in the most severe cases, even human mortalities (Kurland et al., 1960; Table

1). The most severe human health hazards are associated with the organic mercury compound

methylmercury (MeHg), which is readily absorbed and has a long retention time in the body

(Berlin, 1979; Table 1).

Table 1. The human health effects of mercury poisoning.

MERCURY

FORM

BIOLOGICAL

ACTION HEALTH IMPACTS REFERENCE

methylmercury

efficiently absorbed

through intestinal tract

neurological damage including

hearing and visual impairment,

damage to fetuses

Berlin, 1979; Egeland

and Middaugh, 1997;

Foster et al., 2000

mercury

vapour

lipid soluble, penetrates

membranes of body and is

easily absorbed

neurological damage, kidney toxicity

and immune system deficiency

Berlin, 1979; Bernier et

al., 1995; Goldman and

Shannon, 2001

mercury salts

formed from oxidation of

elemental mercury- little

absorbed

gastrointestinal disturbance and renal

(kidney) failure

Berlin, 1979; Goldman

and Shannon, 2001

The effects of MeHg on humans may be neurotoxic, teratologic (embryotoxic) and genetic

(Clarkson et al., 1973; Jarup, 2003). Neurotoxic effects arise from damage to the central

nervous system especially the visual cortex and the cerebellum (Clarkson et al., 1973).

Symptoms include: loss of vision, loss of hearing, difficulties with speech and loss of

sensation in the hands and feet (Berlin 1979). Children, fetuses and embryos are extremely

sensitive to the neurotoxic effects of MeHg which can lead to deformities in the developing

nervous systems in early life stages (Foster et al., 2000; Jarup, 2003).

Mercury exposure in humans may occur through water, air and food items (Jarup, 2003).

Although the primary route of exposure is through the consumption of contaminated food, in

particular fish (Foster et al., 2000; Jarup, 2003). Fish are increasingly being perceived as a

healthy and nutritious food source and as a result human consumption is increasing (Egeland

and Middaugh, 1997; Han et al., 1998; Carvalho et al., 2005). Fish contain essential fatty

80

acids and omega-3 which have been shown to benefit health in several ways including

reduced risk of cardiovascular disease (Egeland and Middaugh, 1997; Han et al., 1998;

Carvalho et al., 2005). However, at the same time it is well known that fish can be

contaminated with toxins, including mercury (Sweet and Zelikoff, 2001; Carvalho et al.,

2005). Elevated mercury concentrations in fish (ie. muscle tissue concentrations of 6 to

20ug/g or greater) may be toxic for the fish themselves and hazardous for human consumption

(Greenfield et al., 2001; Weiner et al., 2003). The potential for devastating effects associated

with the transfer of mercury from fish to humans first came to light in the Minamata Bay

disaster in Japan in 1953 where 46 people died and 100 were left seriously ill as a result of

mercury poisoning from contaminated seafood (Kurland et al., 1960). This tragic incident

sparked numerous studies into mercury levels in fish and the concomitant health effects on

humans (Wang, 2002). The fact that mercury exposure primarily occurs through consumption

of contaminated fish (Figure 1), places certain populations at greater risk.

Figure 1. Human exposure to methylmercury (adapted from Hartung, 1972).

Water: Methylation by bacteria in sediments

Uptake of methylmercury by aquatic organisms and food chain

FISH

Pathways: Air, water

Human exposure through consumption

Source of mercury: Volcanoes, rocks and soil; industrial processes, mining, fossil fuels, incineration

81

People who live in coastal areas and who consume large amounts of fish are particularly at

risk as they are more likely to accumulate high levels of mercury (Han et al., 1998).

Monitoring of mercury levels in fish is increasing throughout the world and fish consumption

advisories have been issued for areas where mercury concentrations have been found to be

excessively high (Foster et al., 2000). Efforts to understand the mechanisms and factors which

affect accumulation in fish should be a priority, in order to predict and prevent human health

risks associated with the consumption of fish.

1.1. Scope of literature review

The potential of mercury contamination to cause detrimental health affects in humans, the fact

that mercury concentrations are increasing in the environment and the ability of mercury to

bioaccumulate through food chains are the main reasons why mercury has been so widely

studied (Asuquo et al., 2004). Current research efforts are increasingly focusing on the

mechanisms behind mercury accumulation in fish and the influences on these processes

(Asuquo et al., 2004). There is a great need to understand the factors which influence mercury

accumulation in fish in order to predict the environmental conditions, species of fish and life

history characteristics of fish likely to cause human health problems. Studies into mercury

accumulation in marine fish is of great importance as they comprise the majority of fish

consumed by humans (Tidwell and Allan, 2001) and because some marine species have been

found to contain very high levels even in so called pristine locations (Hornung et al., 1993).

Understanding of mercury bioaccumulation in marine fish is therefore of great importance to

protection of human health. However, the most recent and most thorough review on mercury

bioaccumulation in fish was by Downs et al. (1998) which examined mercury in precipitation

and how it influences accumulation, with much of the reviewed research from freshwater

studies. Freshwater systems provide particularly useful study sites for research into

82

contamination because; 1) they tend to be highly impacted (Campbell, 1994), and 2) they are

often relatively closed systems from which trends can be easily observed. As this review is

primarily concerned with bioaccumulation in marine fish, it will overview the studies on

freshwater fish and discuss the findings and observations in terms of their relevance to marine

fish. The aims of this review are to look at the main factors which govern the accumulation of

mercury in marine fish including: levels in the environment; bioavailability, as well as uptake

and excretion and biological factors influencing mercury levels, and to discuss any recent

findings in this area and consider their importance for future research.

2. Mercury in the environment

The main factor which influences mercury accumulation in fish is the level in the

environment (Weiner et al., 2003). In order to be contaminated with mercury, fish must firstly

come into contact with it. The higher the concentration of mercury in the environment the

more an organism will be in contact with the toxin and hence the greater the chance of

accumulation (Weiner et al., 2003).

2.1. Sources of mercury in the environment

Mercury enters the environment naturally through geological weathering (Table 2; 1.1-1.3)

but human activities such as burning of fossil fuels, waste incineration, industry emissions

and mining have all increased its input (Table 2; 2.1-2.8). Mining is the main means by which

mercury enters the environment; both through ore wastes and via atmospheric deposition of

mercury vapour during roasting processes (Table 2; 2.2, 2.3). The ability of mercury to be

transported atmospherically over large distances (Rolfhus and Fitzgerald, 1995; Morel et al.,

83

1998; Boening, 2000) and its capacity to persist in the environment, mean that even

undisturbed waterways may have elevated levels of mercury in fish and wildlife and there is

potential for bioaccumulation in aquatic food webs (Morel et al., 1998; Clarkson and Strain,

2003).

Table 2. References citing natural and anthropogenic inputs of mercury into the aquatic environment.

SOURCE OF INPUT REFERENCE

1. natural input

1.1 geological weathering Thomson, 1985; Park and Curtis, 1997; Calta and Canpolat, 2006

1.2 leaching from soils Thomson, 1985; Sweet and Zelikoff, 2001; Calta and Canpolat, 2006

1.3 volcanic activity Boening, 2000; Sweet and Zelikoff, 2001

2. anthropogenic input

2.1 smelting processes Bloom and Ayling, 1977; Birch, 2000; Calta and Canpolat, 2006

2.2 mining Park and Curtis, 1997

2.3 atmospheric fallout Park and Curtis, 1997; Greenfield et al., 2001

2.4 paper mill effluent Bloom and Ayling, 1977

2.5 pesticides Forstner and Wittmann, 1979

2.6 urban stormwater Thomson, 1985; Campbell, 1994; Fabris et al., 2006

2.7 burning of fossil fuels Egeland and Middaugh, 1997; Boening, 2000; Sanzo et al., 2001

2.8 industrial wastes Holden, 1973; Egeland and Middaugh, 1997; Sanzo et al., 2001

2.2. Forms of mercury in the environment

Several forms of mercury occur in the environment; with varying degrees of toxicity

according to solubility, reactivity and biological effects (Berlin, 1979; Goldman and Shannon,

2001). Mercury occurs as elemental mercury (mercury vapour and mercury liquid), inorganic

compounds (mercury salts) and organic compounds (alkylmercury, alkoxyalkylmercury and

phenylmercury compounds) (Berlin, 1979; Goldman and Shannon, 2001; Sanzo et al., 2001).

In its inorganic forms (metal or metallic salts) mercury is moderately toxic, but it is highly

toxic in organically bound forms such as MeHg (Holden, 1973). Elemental mercury is

transformed into the organic mercury compound MeHg by microbial activity when it enters

84

the marine environment (Berlin, 1979). As well as being one of the most toxic mercury

compounds, MeHg is very efficiently taken up by biota (Jernelov and Asell, 1973; Boening,

2000) and is readily bioaccumulated through aquatic food chains. As a result MeHg may be

found in high levels and generally accounts for 80 to 100% of the total mercury content in

high trophic order fish ie. those which are often among the top predators in aquatic food

chains (Cappon and Smith, 1981; Dallinger et al., 1987; Wiener et al., 2003). The increase in

input of inorganic mercury into the environment means that an increase in MeHg is likely

(Holden, 1973).

3. Bioavailability of mercury

In order to be accumulated by fish and other biota, mercury must be present in an accessible

form. Most forms of mercury in the aquatic environment are readily taken up by marine

animals and are therefore bioavailable. However, the form which is most bioavailable is

MeHg (Wiener et al., 2003). This is primarily because MeHg has a strong affinity for lipids

and proteins and has a rather unique ability among mercury compounds to easily cross cell

membranes (Berlin, 1997). Therefore when it is produced in the sediments of the aquatic

environment, it is readily taken up by the surrounding biota (Forstner and Wittmann, 1979).

The level of MeHg in the environment will greatly influence the uptake of mercury by fish.

3.1 Mercury methylation

In the marine environment, MeHg is produced from the transformation of inorganic mercury

through a process known as methylation (Greenfield et al., 2001), which involves the

reduction of elemental mercury by microbial activity (Foster et al., 2000). This process

85

primarily occurs in the sediments, but has also been thought to occur below the mixed layer in

the oceans by bacteria which are present on marine snow (Downs et al., 1998). The

methylation of inorganic mercury to MeHg strongly influences the accumulation in fish

because it greatly increases the bioavailability and toxicity of mercury and increases the

potential for human exposure to MeHg. The processes which govern MeHg production will

greatly affect its uptake in fish (Wiener et al., 2003). MeHg production is strongly influenced

by factors which favour mercury methylating bacteria such as warmer sediment temperature

and low pH (Jernelov and Asell, 1973; Foster et al., 2000). The processes that govern

methylation rate should therefore be taken into account in studies and monitoring programs to

better understand them.

3.1.1. Affect of Temperature

Water temperature has a significant influence on many environmental and biological

processes including metabolism in fish and the production of MeHg and hence accumulation

of mercury in fish (Filazi et al., 2003; Foster et al., 2000). Warmer temperatures increase the

rate of microbial processes which leads to an increase in the production of the highly toxic

methylmercury which is readily taken up by biota (Foster et al., 2000). Therefore warmer

temperatures may lead to an increase in the accumulation of mercury by increasing the

bioavailability of the metal. Temperature changes are largely seasonal and result in increased

concentrations of MeHg in summer which may be reflected in fish tissue concentrations of

mercury (Foster et al., 2000). Consequently, seasonal/temperature differences in uptake need

to be included in any monitoring program.

86

3.1.2. Affect of pH

Water pH has been found to strongly influence mercury concentrations in fish in freshwater

lakes, with low pH water favouring MeHg production and hence bioavailabilty of mercury

(Suns and Hitchin, 1990). However this is unlikely to be a factor in marine fish as marine

systems maintain relatively constant pH levels (Greenfield et al., 2001). It may however, be a

factor in coastal marine areas and estuaries which are more subject to terrestrial inputs which

can alter water chemistry.

4. Uptake of mercury by fish

Fish readily uptake MeHg from the water column, sediments and dietary items however, diet

is the primary source (Wiener et al., 2003) with absorption rates estimated to be

approximately 90% (Goldman and Shannon, 2001). Consequently this review will be

restricted only to discussion of uptake via diet. Mercury concentrations have been strongly

linked with the feeding behaviour of fish (Ratkowsky et al., 1975; Pourang, 1995), with

specific aspects of feeding behaviour and diet affecting uptake (ie. feeding preference (benthic

versus pelagic) and trophic level).

4.1. Feeding preference

Both the habitat in which a species of fish resides and its source of food will have a large

influence on the levels of mercury it will be exposed to and hence how much it may

accumulate. For example, fish which dwell and feed primarily in the sediments are likely to

uptake more mercury, as aquatic sediments serve as both a sink and a source of organic and

87

inorganic pollutants (Burton et al., 2003), most notably heavy metals, which enter the system

(Campbell, 1994). The majority of benthic invertebrates (which are common prey items of

benthic fish species) obtain nutrition through the digestion of sediment material (Campbell,

1994). Prey items are often relatively unaffected by high levels of mercury and consequently

can obtain very high concentrations (Rainbow, 1990). As a result fish that feed on benthic

invertebrates (benthic feeders) are likely to accumulate high levels of mercury (Blevins and

Pancorbo, 1986). In contrast to this, the pelagic food chain has little or no interaction with the

sediments and mercury uptake begins with the uptake of trace amounts present in the water by

phytoplankton (Burton and Statham, 1990; Wiener et al., 2003). Mercury is still accumulated

in the pelagic food chain but at much lower levels; high levels only occurring in large, long

lived species. Benthic feeders therefore have a much greater potential to accumulate high

levels of mercury than pelagic feeders.

4.2. Trophic level

Uptake and hence accumulation of mercury in fish has also been found to be affected by

trophic level (Watras and Bloom, 1992; Pourang, 1995; Hill et al., 1996). Aquatic systems are

particularly susceptible to pollution due mainly to the structure of their food chains (Forstner

and Wittmann, 1979). The small biomass in aquatic environments is produced through a

greater variety of trophic levels than land systems, allowing for greater accumulation of toxic

substances (Forstner and Wittmann, 1979). All fish contain small natural levels of mercury

and other essential heavy metals in their bodies which have no measurable affect on the fish

(Hornung et al., 1993). Mercury is therefore concentrated in prey items and up though the

food chain (Watras and Bloom, 1992), and the result is accumulation in top order predator

species (Figure 2).

88

Figure 2. Food chain model for mercury accumulation (adapted from Goldman and Shannon, 2001).

This is known as biomagnification where contaminant levels increase through successive

links in the food chain (Blevins and Pancorbo, 1986; Dallinger et al., 1987; Downs et al.,

1998). High levels can be obtained in species of shark even in pristine environments

(Hornung et al., 1993). This is simply a reflection of their high trophic status and size (Adams

and McMichael, 1999). Human consumption of high trophic species such as sharks can

therefore be considerably risky from a health point of view (Hornung et al., 1993; Adams and

McMichael, 1999). Top predator fish species should only be consumed if comprehensive

monitoring and testing of edible tissue for mercury levels has taken place, otherwise they

should be avoided or their intake limited to reduce the risk of mercury poisoning.

Sediment

Water

Algae

Zooplankton

Fish

Predatory Fish

Birds

Mammals

Man

89

5. Excretion of mercury by fish

The potential for heavy metal accumulation is largely governed by the ability of the organism

to either excrete or, store (partition) pollutants (Bryan, 1979). Fish are generally able to

regulate essential metals to maintain optimum levels and prevent toxicity (Giesy and Wiener,

1977), but, like most other animals they are unable to regulate levels of non essential metals

including mercury (Giesy and Wiener, 1977) and particularly MeHg, which is very persistent.

Of all the various forms of mercury, MeHg has the slowest rate of elimination (Miettinen,

1973), with a half life in the muscle of some fish species estimated at 2-3 years (Sorensen,

1991). The highest levels will often be found in the edible muscle tissue as opposed to other

organs because fish are able to move MeHg from organs, such as the liver and kidney, to the

muscle where it tends to accumulate (Wiener et al., 2003). Therefore mercury detoxification

in fish is by sequestration rather than elimination (Arnac and Lassus, 1985). This may be a

protective mechanism by fish to prevent the toxic effects on the nervous system (Wiener et

al., 2003). However, this makes accumulation of MeHg in fish an even greater human health

concern as fish muscle tissue is generally the part of the fish which is consumed.

6. Biological factors influencing mercury levels in fish

Once mercury has been taken up by fish there are two main biological factors which then

govern the final concentration. These are growth rate and age/size of the fish (Hornung et al.,

1993; Harris and Bodaly, 1998).

90

6.1. Growth rate

Fish growth rate can greatly influence mercury levels in fish (Harris and Bodaly, 1998). Most

of the studies which have looked at the effects of growth rate on mercury accumulation in fish

have been on freshwater species, however, the trends/findings are as applicable to marine

species. Higher growth rates in fish can result in reduced mercury concentrations due to

growth dilution (Arnac and Lassus, 1985; Park and Curtis, 1997; Simoneau et al., 2005).

Whereby faster growth means that less time is required to reach a given size, consequently

less time is spent metabolising and eating food for metabolic needs, and therefore both

cumulative mercury exposure and concentration in faster growing fish is lower (Harris and

Bodaly, 1998). The converse is true for slower growth rates; in temperate systems, which

experience cooler water temperatures, fish metabolism is slowed and fish eat less resulting in

a reduction in, or cessation of, growth (Kehrig et al., 1998). Furthermore, studies have shown

that some species of fish will actually lose body mass during winter when using up existing

energy reserves (Kehrig et al., 1998). When body mass is lost any mercury present in the flesh

will become more concentrated (Kehrig), this is known as starvation concentration (Cizdziel

et al., 2002). As undernourished fish catabolise muscle tissue for energy the overall muscle

mass is reduced faster than the bound MeHg therefore effectively increasing the concentration

of mercury in the remaining tissue resulting in internal bioconcentration (Cizdziel et al.,

2002). Muscle tissue may receive MeHg from organs during decontamination periods

(Cizdziel et al., 2003). Therefore in temperate waters where there are distinct seasons, it is

highly likely that fish growth may be reduced and MeHg levels increased. Starvation

concentration can also be observed in dwarf fish. Dwarf fish are individuals which have

slower growth rates than fish of the same species and are much smaller in size than fish of a

similar age (Doyon et al., 1998). Dwarf fish tend to allocate more energy towards

maintenance and less to flesh production and therefore growth rates of dwarf fish are much

91

slower (Doyon et al., 1998). As a result dwarf fish produce proportionally less flesh than

normal fish which means higher mercury concentrations can be found in body tissues (Doyon

et al., 1998). This once again identifies the need for monitoring and testing of fish to be

carried out across different seasons especially in temperate regions of the world as seasonal

temperature changes are likely to have a strong influence on accumulation in marine fish.

6.2. Age / size

Also influencing the accumulation of mercury in a particular fish is age and or size.

Contamination of mercury tends to increase with age and size (Hornung et al., 1993). This

trend is unique to mercury and is simply due to the fact that it is able to bioaccumulate

through the food chain (Hornung et al., 1993) which in turn can be attributed to MeHg’s

affinity for organic matter and high half life (Szefer et al., 2003). The larger a fish is, the more

prey it will consume and hence the higher the concentration of mercury in the body. In

addition, long lived species of fish have more time to accumulate heavy metals throughout

their lifetime (Hornung et al., 1993). Because mercury concentration increases with age,

marine organisms with a long life span (e.g. tuna, sharks, marine mammals) have higher

concentrations then short lived organisms (Hornung et al., 1993). Extremely high levels have

been measured in black marlin (Makaira indica) in Australia (Mackay et al., 1975). These

fish were caught from un-impacted waters yet still attained some of the highest levels of

mercury ever measured in teleost fish with concentrations in muscle tissue ranging from

0.5ppm to 16.5ppm (mean 7.3ppm) (Mackay et al., 1975). These values greatly exceed the

recommended Australian level for safe consumption of 1ppm for large fish (FSANZ). The

high mercury content in muscle tissue in the black marlin from this particular study was

explained by the longevity of the black marlin which enables weight-age related

92

bioaccumulation (ie. accumulation over time) (Mackay et al., 1975). However, it is also likely

to be a function of their high trophic status as well (black marlin being a top predator species).

Table 3. Increase in mercury concentration in fish with age/size.

FISH / LOCATION

EFFECT OF AGE / SIZE ON

ACCUMULATION REFERENCE

Black marlin Extremely high Hg levels explained by longevity Mackay et al., 1975

Atlantic herring, Canada Positive correlations with age, weight, and length Braune, 1987

Deep water sharks High levels of Hg attributed to longevity of species Hornung et al., 1993

Sharks from Florida Positive correlation of Hg levels and shark size Hueter et al., 1995

Fish from Brazilian

estuaries Hg increased with length and weight Kehrig et al., 1998

Perch in Baltic Sea Hg in muscle increased with age Szefer et al., 2003

In another more recent study on various species of shark in Florida, USA, 33.1% of the

samples exceeded the U.S. food guidelines for mercury concentration of (1ppm) (Hueter et

al., 1995). Consequently human consumption of long lived or large fish especially tuna,

marlin and sharks (Hornung et al., 1993) carries with it a significant human health risk in

some cases (Hornung et al., 1993; Hueter et al., 1995). Many fisheries advisories now take

into account fish size and age with limits set on the maximum size of fish which can be safely

consumed as well as some species which should simply not be eaten (Hueter et al., 1995).

Understanding the relationship between mercury concentration and body size/age within a

population is also extremely important for comparison of mercury concentration in fish to

assess variations in contamination levels (Arnac and Lassus, 1985; Francesconi et al., 1997).

93

7. Summary and future research

Mercury and mercury compounds are extremely toxic in low concentrations, with human

exposure responsible for a range of detrimental health effects. The primary route of human

exposure is via consumption of MeHg in contaminated fish. MeHg is produced in aquatic

environments through methylation and is readily taken up by biota. This along with the fact

that it is very persistent means that MeHg will accumulate up aquatic food chains. There are

several environmental factors which influence the accumulation of MeHg in marine fish; 1)

the amount of elemental mercury in the environment, 2) the microbial activity in the

sediments which is responsible for the production of MeHg and 3) environmental temperature

which affects MeHg production, and therefore may increase mercury bioavailability and

hence accumulation by fish. The uptake of MeHg in any particular fish is primarily a factor of

diet. Fish in high trophic orders and benthic feeders are more likely to accumulate high levels

of mercury. Once taken up, mercury levels in fish are governed mainly by size/age of the fish

and growth rate. Increased growth rates may dilute mercury concentration through a process

known as growth dilution and reduced growth rate may increase mercury concentration

through a process known as starvation concentration. Long lived or large fish are likely to

obtain high levels of mercury due to the fact that they consume large amounts of slightly

mercury contaminated prey throughout their lifetimes in a process referred to as

bioaccumulation. The ever increasing world population growth and the increased perception

of fish as a health food means that more fish will be consumed in the future. This along with

the continued input of mercury into the environment through human activity means that

humans will increasingly be exposed to high levels of mercury in the future. This review

highlights several areas which require further attention. Specifically a need to: 1) continue and

increase monitoring of mercury levels in marine fish species, particularly in coastal regions

94

and in less developed countries 2) expand existing monitoring to include seasons and age of

fish and 3) identify at risk marine areas and fish species based on human input, trophic level

and feeding preference and monitor, assess and impose restrictions on fishing from these

areas to protect human health.

References

Adams, D.H., McMichael, R.H., 1999. Mercury levels in four species of sharks from the Atlantic coast

of Florida. Fisheries Bulletin, 97, 372-379.

Arnac, M., Lassus, C., 1985. Heavy metal accumulation (Cd, Cu, Pb and Zn) by smelt (Osmerus

mordax) from the north shore of the St Lawrence Estuary. Water Research, 19, 6, 725-734.

Asuquo, F.E., Ewa-Oboho, I., Asuquo, E.F., Udo, P.J., 2004. Fish species used as biomarker for heavy

metal and hydrocarbon contamination for Cross River, Nigeria. The Environmentalist, 24,

29-37.

Berlin, M., 1979. Mercury. In: Friberg, L., Nordberg, G.F., Vouk, V.B. (Eds), Handbook on the

Toxicology of Metals, Elsevier/North-Holland Biomedical Press, pp. 503-530.

Bernier, J., Brousseau, P., Krzystyniak, K., Tryphonas, H., Fournier, M., 1995. Immunotxicology of

heavy metals in relation to great lakes. Environmental Health Perspectives, 103, 23-32.

Birch, G.F., 2000. Marine pollution in Australia, with special emphasis on central New South Wales

Estuaries and adjacent continental margin. International Journal of Environmental Pollution,

13, 573-607.

Blevins, R.D., Pancorbo, O.C., 1986. Metal concentrations in muscle of fish from aquatic systems in

East Tennessee, USA. Water, Air, and Soil Pollution, 29, 361-371.

Bloom, H., Ayling, G.M., 1977. Heavy metals in the Derwent Estuary. Environmental Geology, 2, 3-

22.

Boening, D.W., 2000. Ecological effects, transport, and fate of mercury: a general review.

Chemosphere, 40, 1335-1351.

Braune, B.M., 1987. Mercury accumulation in relation to size and age of Atlantic herring (Clupea

harengus) from the Southwestern Bay of Fundy, Canada. Archives of Environmental

Contamination and Toxicology, 16, 311-320.

Bryan, G.W., 1979. Bioaccumulation of marine pollutants [and discussion]. Philosophical

Transactions of the Royal Society of London. Series B. Biological Sciences, 286, 483-505.

Burton, G.A., Denton, D.L., Ho, K., Ireland, D.S., 2003. Sediment toxicity testing: issues and

methods. In: Hoffman, D.J., Rattner, B.A., Burton, G.A., Cairns, J. (Eds), Handbook of

Ecotoxicology (2nd

edition). Lewis Publishers, New York, pp. 111-150.

Burton, J.D., Statham, P.J., 1990. Trace metals in seawater. In: Furness, R.W., Rainbow, P.S. (Eds),

Heavy Metals in the Marine Environment, CRS Press, Florida, pp. 5-7.

Calta, M., Canpolat, O., 2006. The comparison of three cyprinid species in terms of heavy metals

accumulation in some tissues. Water Environment Research, 78, 5, 548-551.

95

Campbell, K.R., 1994. Concentrations of heavy metals associated with urban runoff in fish living in

stormwater treatment ponds. Archives of Environmental Contamination and Toxicology, 27,

352-356.

Cappon, C.J., Smith, J., 1981. Mercury and selenium content and chemical form in fish muscle.

Archives of Environmental Contamination and Toxicology, 10, 305-319

Carvalho, M.L., Santiago, S., Nunes, M.L., 2005. Assessment of the essential element and heavy metal

content of edible fish muscle. Analytical Bioanalytical Chemistry, 382, 426-432.

Cizdziel, J.J., Hinners, T.A., Pollard, J.E., Heithmar, E.M., Cross, C.L., 2002. Mercury concentrations

in fish from Lake Mead, USA, related to fish size, condition, trophic level, location, and

consumption risk. Archives of Environmental Contamination and Toxicology, 43, 309-317.

Cizdziel, J., Hinners, T., Cross, C., Pollard, J., 2003. Distribution of mercury in the tissues of five

species of freshwater fish from Lake Mead, USA. Journal of Environmental Monitoring, 5,

802-807.

Clarkson, T.W., Smith, J.C., Marsh, D.O., Turner, M.D., 1973. A review of dose-response

relationships resulting from human exposure to methylmercury compounds. In: Krenkel,

P.A. (Ed), Heavy Metals in the Aquatic Environment. Pergamon Press, New York, pp. 1-9.

Clarkson, T.W., Strain, J.J., 2003. Nutritional factors may modify the toxic action of methyl mercury

in fish eating populations. The Journal of Nutrition, 1539-1543.

Dallinger, R., Prosi, F., Segner, H., Back, H., 1987. Contaminated food and uptake of heavy metals by

fish: a review and a proposal for further research. Oecologia, 73, 91-98.

Downs, S.G., Macleod, C.L., Lester, J.N., 1998. Mercury in precipitation and its relation to

bioaccumulation in fish: a literature review. Water, Air, and Soil Pollution, 108, 149-187.

Doyon, J.F., Schetagne, R., Verdon, R., 1998. Different mercury bioaccumulation rates between

sympatric populations of dwarf and normal lake whitefish (Coregonus clupeaformis) in the

La Grande complex watershed, James Bay, Quebec. Biogeochemistry, 40 203-216.

Egeland, G.M., Middaugh, J.P., 1997. Balancing fish consumption benefits with mercury exposure.

Science, 278, 1904-1905.

Fabris, G., Turoczy, N.J., Stagnitti, F., 2006. Trace metal concentrations in edible tissue of snapper,

flathead, lobster, and abalone from the coastal waters of Victoria, Australia. Ecotoxicology

and Environmental Safety, 63, 2, 286-292.

Filazi, A., Baskaya, R., Kum, C., Hismiogullari, S.E., 2003. Metal concentrations in tissues of the

Black Sea fish Mugil auratus from Sinop-Icliman, Turkey. Human & Experimental

Toxicology, 22, 85-87.

Forstner, U., Wittmann, G.T.W. (Eds), 1979. Metal Pollution in the Aquatic Environment. Springer-

Verlag, New York, 485 p.

Foster, E.P., Drake, D.L., DiDomenico, G., 2000. Seasonal changes and tissue distribution of mercury

in largemouth bass (Micropterus salmoides) from Dorena Reservior, Oregon. Archives of

Environmental Contamination and Toxicology, 38, 78-82.

Francesconi, K.A., Lenanton, R.C.S., Caputi, N., Jones, S., 1997. Long-term study of mercury

concentrations in fish following cessation of a mercury-containing discharge. Marine

Environment Research, 43, 27-40.

FSANZ (Food Standards Australia and New Zealand). Regulations for mercury in fish. March 2004.

<http://www.foodstandards.gov.au/newsroom/factsheets/factsheets2004/mercuryinfishfurther2

394.cfm>. [accessed 2007 Nov 2]

Giesy, J.P., Wiener, J.G., 1977. Frequency distributions of trace metal concentrations in five

freshwater fishes. Transactions of the American Fisheries Society, 106, 393-403.

96

Goldman, L.R., Shannon, M.W., 2001. Technical report: mercury in the environment: implications for

pediatricians. American Academy of Pediatrics, 108, 197-205.

Greenfield, B.K., Hrabik, T.R., Harvey, C.J., Carpenter, S.R., 2001. Predicting mercury levels in

yellow perch: use of water chemistry, trophic ecology, and spatial traits. Canadian Journal of

Fisheries and Aquatic Sciences, 58, 1419-1429.

Han, B.C., Jeng, W.L., Chen, R.Y., Fang, G.T., Hung, T.C., Tseng, R.J., 1998. Estimation of target

hazard quotients and potential health risks for metals by consumption of seafood in Taiwan.

Archives of Environmental Contamination and Toxicology, 35, 711-720.

Harris, R.C., Bodaly, R.A., 1998. Temperature, growth and dietary effects on fish mercury dynamics

in two Ontario lakes. Biogeochemistry, 40, 175-187.

Hartung, R., Dinman, B.D. (eds), 1972. Environmental Mercury Contamination. Ann Arbor Science

Publishers, Inc., Michigan, pp. 1-349.

Hill, W.R., Stewart, A.J., Napolitano, E., 1996. Mercury speciation and bioaccumulation in lotic

primary producers and primary consumers. Canadian Journal of Fisheries and Aquatic

Sciences, 53, 812-819.

Holden, A.V., 1973. Mercury in fish and shellfish a review. Journal of Food Technology, 8, 1-25.

Hornung, H., Krom, M.D., Cohen, Y., Bernhard, M., 1993. Trace metal content in deep-water sharks

from the eastern Mediterranean Sea. Marine Biology, 115, 331-338.

Hueter, R.E., Fong, W.G., Henderson, G., French, M.F., Manire, C.A., 1995. Methylmercury

concentration in shark muscle by species, size and distribution of sharks in Florida coastal

waters. Water, Air and Soil Pollution, 80, 893-899.

Jarup, L., 2003. Hazards of heavy metal contamination. British Medical Bulletin, 68, 167-182.

Jernelov, A., Asell, B., 1973. The feasibility of restoring mercury-contaminated waters. In: Krenkel,

P.A. (Ed), Heavy Metals in the Aquatic Environment. Pergamon Press, New York, pp. 299-

309.

Kehrig, H.A., Malm, O., Moreira, I., 1998. Mercury in a widely consumed fish Micropogonias

furnieri (Demarest, 1823) from four main Brazilian estuaries. The Science of the Total

Environment, 213, 263-271.

Kennish, M.J. (Ed), 2001. Practical Handbook of Marine Science (3rd

Edition), CRC Press, New York,

pp. 630-631.

Kurland, L.T., Faro, S.N., Seidler, H., 1960. Minamata disease. World Neurology, 1, 370-390.

Lizama, M. Delos A.P., Ambrosio, A.M., 2002. Condition factor in nine species of fish of the

characidae family in the upper Parana River floodplain, Brazil. Brazilian Journal of Biology,

62, 113-124.

MacKay, N.J., Kazacos, M.N., Williams, R.J., Leedow, M.I., 1975. Selenium and heavy metals in

black marlin. Marine Pollution Bulletin, 6, 57-61.

Miettinen, J.K., 1973. The accumulation and excretion of heavy metals in organisms. In: Krenkel, P.A.

(Ed), Heavy Metals in the Aquatic Environment. Pergamon Press, New York, pp. 163-165.

Morel, F.M.M., Kraepiel, A.M.L., Amyot, M., 1998. The chemical cycle and bioaccumulation of

mercury. Annual Review of Ecological Systems, 29, 543-566.

Park, J.G., Curtis, L.R., 1997. Mercury distribution in sediments and bioaccumulation by fish in two

Oregon reservoirs: point-source and nonpoint-source impacted systems. Archives of

Environmental Contamination and Toxicology, 33, 423-429.

Pourang, N., 1995. Heavy metal accumulation in different tissues of two fish species with regards to

their feeding habits and trophic levels. Environmental Monitoring and Assessment, 35, 207-

219.

97

Rainbow, P.S., 1990. Heavy metal levels in marine invertebrates. In: Furness, R.W., Rainbow, P.S.

(Eds), Heavy Metals in the Marine Environment, CRS Press, Florida, pp. 67-71.

Ratkowsky, D.A., Dix, T.G., Wilson, K.C., 1975. Mercury in fish in the Derwent Estuary, Tasmania,

and its relation to the position in the food chain. Australian Journal of Marine and

Freshwater Research, 26, 223-231.

Rolfhus, K.R., Fitzgerald, W.F., 1995. Linkages between atmospheric mercury deposition and the

methylmercury content of marine fish. Water, Air and Soil Pollution, 80, 291-297.

Sanzo, J.M., Dorronsoro, M., Amiano, P., Amurrio, A., Aguinagalde, F.X., Azpiri, M.A., 2001.

Estimation and validation of mercury intake associated with fish consumption in an EPIC

cohort of Spain. Public Health Nutrition, 4, 981-988.

Simoneau, M., Lucotte, M., Garceau, S., Laliberte. D., 2005. Fish growth rates modulate mercury

concentrations in walleye (Sander vitreus) from eastern Canadian lakes. Environmental

Research, 98, 73-82.

Sorensen, E.M.B., 1991. Metal Poisoning in Fish, CRC Press, Boston, pp. 1-353.

Suns, K., Hitchin, G., 1990. Interrelationships between mercury levels in yearling yellow perch, fish

condition and water quality. Water, Air, and Soil Pollution, 650, 255-265.

Sweet, L.I., Zelikoff, J.T., 2001. Toxicology and immunotoxicology of mercury: a comparative review

in fish and humans. Journal of Toxicology and Environmental Health, Part B. 4, 161-205.

Szefer, P., Domagala-Wieloszewska, M., Warzocha, J., Garbacik-Wesolowska, A., Ciesielski, T.,

2003. Distribution and relationships of mercury, lead, cadmium, copper and zinc in perch

(Perca fluviatilis) from the Pomeranian Bay and Szczecin Lagoon, southern Baltic. Food

Chemistry, 81, 1, 73-83.

Thomson, J.D., 1985. Mercury concentrations of the axial muscle tissues of some marine fishes of the

continental shelf adjacent to Tasmania. Australian Journal of Marine and Freshwater

Research, 36, 509-517.

Tidwell, J.H., Allan, G.L., 2001. Fish as food: aquaculture’s contribution- ecological and economic

impacts and contributions of fish farming and capture fisheries. EMBO reports, 21, 958-963.

Wang, W., 2002. Interactions of trace metals and different marine food chains. Marine Ecology

Progress Series, 243, 295-309.

Watras, C.J., Bloom, N.S., 1992. Mercury and methylmercury in individual zooplankton: implications

for bioaccumulation. Limnology and Oceanography, 37, 1313-1318.

Wiener, J.G., Krabbenhoft., D.P., Heinz, G.H., Scheuhammer, A.M., 2003. Ecotoxicology of Mercury.

In: Hoffman, D.J., Rattner, B.A., Burton, G.A., Cairns, J. (eds), Handbook of Ecotoxicology

(2nd

Edition). Lewis Publishers, New York, pp. 409-443.

heavy metal contamination in muscle tissue of four key...

Documents