1 The original publication is available at http://www.springerlink.

com (DOI

2 10.1007/s12562-008-0054-0)

3 Occurrence of heavy metals in the sediments of Uranouchi Inlet, Kochi

4 prefecture, Japan

6 De Jaysankar1*Ω, Fukami Kimio1, Iwasaki Kozo2 and Okamura Kei3

1
7 Graduate School of Kuroshio Science (GRAKUS), Kochi University, B200 Monobe,

8 Nankoku, Kochi 783 8502, Japan

2
9 Faculty of Agriculture, Kochi University, B200 Monobe, Nankoku, Kochi 783 8502,

10 Japan

3
11 Center for advanced marine core research, Kochi University, Nankoku, Kochi 783 8502,

12 Japan

13

14

15 RUNNING TITLE: Heavy metals in the Uranouchi Inlet.

16

17
Ω
18 Current address: Vanevan Institute; 14 Yerevanyan Street, Gegharkunik Martuni, Armenia.

19 Tel & Fax: 00374 10 227089; Email: jaysankarde@yahoo.com.

*
20 Corresponding author
1 ABSTRACT IN JAPANESE:

2 高知県浦ノ内湾における底泥中の重金属の分布とその状況

3 ジェイサンカール・デ，深見公雄（高知大学大学院黒潮圏海洋科学研究科），

4 岩崎貢三（高知大学農学部），岡村 慶（高知大学海洋コア総合研究センター）

5 魚類養殖が盛んな高知県浦ノ内湾における底泥中の重金属の分布を調べたところ，Zn

6 は178±4.8 mg/kg dry wt.,Cu は125±1.2 mg/kg dry wt.，Pb は50.7±0.77 mg/kg

7 dry wt. と，いずれも養殖の行われていない湾口近くの測点と比較して養殖場近くで高

8 い値を示した．しかしながらCo，Cr，Fe では測点間で有意な差は見られなかった．塩

9 酸で容易に遊離してくる“易分解性画分”はZn で約56％，Cu で約40％と高い値を示し

10 たことから，生物への影響が懸念された．以上の結果から，長年にわたる魚類養殖によ

11 り，内湾底泥環境に重金属の蓄積が起こっていることが明らかとなった．

2
 1 ABSTRACT:

2 The present status of contamination by heavy metals, and the impact of cage culture on

3 sediments at the Uranouchi inlet, Kochi Prefecture, Japan were investigated from two

4 stations influenced by intensive aquaculture and a control station, during May-July, 2006.

5 The moisture content of the sediments was over 65%, and the organic matter was always

6 over 100 mg/g dry wt. at the aquaculture stations. In contrast, the highest moisture

7 content and organic matter at the control station was 45.5% and 62.9 mg/g dry wt.

8 respectively. Concentration of Zinc (178±4.8 mg/kg dry wt.) and Cu (125±1.2 mg/kg dry

9 wt.) were highest at the aquaculture stations. Pb was highest (50.7±0.77 mg/kg dry wt.) at

10 the aquaculture station though it was as high as 33.2±0.77 mg/kg dry wt. at the control

11 station. One way ANOVA showed that the differences in concentrations of Zn and Cu in

12 sediments from the aquaculture and control stations were highly significant (p=0.01),

13 whereas Pb showed no such trend. Occurrence of a large fraction of labile Zn (56.1%)

14 and Cu (40.3%) in these sediments warrants attention. Although factors other than metals

15 may explain the distribution observed, the information presented here may be useful in

16 predicting long-term effects of heavy-metal contamination from aquaculture in the

17 marine environment.

18

19 KEYWORDS: Uranouchi Inlet, copper, zinc, lead, aquaculture, labile metals

3
 1 INTRODUCTION

2 Environmental concern pertaining to heavy metals relates to their toxicity, labile nature,

3 bioaccumulation in organisms and ultimately to their effect on the aquatic environment

4 and human beings. Through the processes of precipitation and sedimentation a portion of

5 the heavy metals introduced into the aquatic environment is adsorbed onto the sediment.

6 Thus, concentrations of heavy metals in sediment play an important role in providing

7 information about overlying water quality. The long-term ecological implications of high

8 concentration of heavy metals in fish farm sediment are poorly studied although its

9 impact on sediments is much greater than its effects on water quality. Analyses of

10 physical, geochemical and biological characteristics of the sediment are sensible

11 approaches to evaluate impacts on sediments and the surrounding environment. There are

12 several methods and indicators for estimation of environmental impact on sediments from

13 simple visual analysis to sophisticated methods. The most common indicators for analysis

14 of impact on sediments are the organic materials1, concentration of heavy metals in the

15 impacted sediment (www.scotland.gov.uk/cru/kd01/green/reia-04.asp).

16

17 Farming fish in cages is essentially an open system. In marine farms the inputs are

18 juvenile fish, fish feed, medicines, disinfectants and anti-foulants, and the outputs (losses)

19 are harvested fish, escaped fish, uneaten feed, faeces, excreted metabolic wastes, and

20 residual chemical species e.g. medicines2. The open nature of this culture system allows

21 these outputs to participate in external biological, chemical and ecological systems where

22 they may cause unwanted effects. These effects are often complex, varying by orders of

23 magnitude on temporal and spatial scales. For example, major effects of particulate inputs

4
 1 to sediments on benthic communities are typically restricted to a relatively small area

2 around the farm but may persist for several years3-4. Briefly, particulate organic material

3 (uneaten feed, faeces and bio fouling biomass detached from cage structures) settles on

4 the seabed where it is degraded by microbes utilizing a variety of electron acceptors.

5 Oxygen in sediment pore waters is rapidly depleted and sulphides are generated by

6 sulphate reduction, which is the dominant anaerobic process in coastal sediments5. These

7 effects on sediment biogeochemical processes have a profound effect on the seafloor

8 fauna that becomes dominated by a few small, opportunistic species, often at very high

9 abundances, and confined to the upper few centimeters of the sediment3,6.

10

11 While the fate and consequences of organic wastes to the benthic ecosystem are relatively

12 well understood, less attention has been given to metallic wastes from fish farms and to

13 the changes in sediment geochemistry induced by high organic matter fluxes and

14 sediment anoxia/hypoxia. Elevated levels of zinc and copper have been found in fish

15 farm sediments7-9. These reports suggest that the zinc and copper sediment enrichment

16 around fish farms is a result of feed and faecal inputs, with significant copper also

17 released from anti-foulant products. Fish feeds include metals that are elemental

18 constituents of the meals from which the food is derived, but these are supplemented with

19 metal-containing mineral additives that serve secondary purposes, such as copper

20 sulphate, which acts as a preservative, as well as to meet perceived nutritional

21 requirements. Manufactured feeds can contain zinc, copper, cadmium, iron, manganese,

22 cobalt, nickel, lead, magnesium, selenium and mercury10-11. Salmon feeds are

23 supplemented with mineral pre-mixes, but these may be over-formulated because of a

5
 1 general lack of information of nutritional requirements and it may be that an un-

2 supplemented diet would be sufficient10,12.

4 Analyses of sediment quality adjacent to fish cages are therefore carried out which are

5 indicators of health of the environmental conditions. In addition to high concentrations of

6 antibiotics in aquatic environments near fish farms13, metal contamination is ubiquitous

7 and is still on the increase14-15. Unlike most organic pollutants, metals remain unaltered in

8 environment, and so may represent a long-term selective pressure, depending on the

9 chemical form and bioavailability. Several metals are essential for biological processes

10 but become toxic above a threshold value thus causing changes in community structure as

11 species of lower tolerance are excluded. Sublethal concentrations of Zn delay sexual

12 maturity and reproduction in fish16. Copper, in certain concentrations, is toxic to algae

13 and some other aquatic organisms, in particular embryonic and larval stages of

14 invertebrates6. Health effects on shellfish can take place even when only very small

15 concentrations of lead are present. Body functions of phytoplankton can be disturbed

16 when lead interferes. Since benthic invertebrates are a food source for many organisms,

17 there is also the potential for the transfer of metals through the food chain2. Furthermore,

18 the environmental impact of aquaculture has been studied using structural measures but

19 not in terms of the functioning of the system (i.e. measuring rates of different processes).

20 For example, hardly anything is known on the bioavailability of different metals which

21 plays a major role in determining the extent of heavy metal stress. Risk assessment of

22 these contaminated sediments requires knowledge of contaminant bioavailability to

23 sediment- ingesting aquatic animals. The bioavailability of metal contaminants associated

6
 1 with sediment has therefore been studied very extensively over the past two decades17.

2 Metals are bound with different geochemical fractions in sediment, including the

3 operationally defined easily exchangeable, carbonate, reducible, organic matter, and

4 residue phases. Several experimental studies have generally shown that labile metals

5 (such as those extracted easily by HCl) are more bioavailable to benthic animals17.

7 Uranouchi Inlet, Kochi Prefecture in Japan, where many fish cages of aquaculture are set,

8 is a semi enclosed small inlet that is representative of a typical aquaculture site18. In

9 Uranouchi Inlet the oxygen is depleted over the bottom layer in summer. The bottom

10 layer in this inlet becomes anoxic and rich in organic and inorganic nutrients19. Another

11 very important feature of this inlet is that there are no other activities than marine

12 aquaculture. Thus, the only inputs received by water are: aquaculture outputs, water

13 renovation from rainwater, and then any non-natural alteration of the sediments, must be

14 related to the aquaculture activities. As a consequence, the site becomes an ideal natural

15 laboratory to study the environmental effects of marine aquaculture, and then, to establish

16 direct correlations between aquaculture and accumulation of heavy metals in sediments.

17 The present work aimed at determining the high-resolution spatial distribution of the

18 potentially ecotoxic metals zinc, copper, lead and few others in sediments around cage

19 farms in the Uranouchi inlet to investigate the potential risk therein.

20

21 MATERIALS AND METHODS

22 Sampling site

7
 1 The present study was carried out at the Uranouchi Inlet, Kochi Prefecture in Japan,

2 where many fish cages of aquaculture are set20. It is a semi enclosed inlet with a

3 maximum depth of 16–19 m. The bottom layer in this inlet becomes anoxic and rich in

4 organic and inorganic nutrients especially during the summer months19. Samples were

5 collected from three stations (Fig. 1) from this semi enclosed small inlet. Two of these

6 stations, Menokuso (332522N/1332361E) and Mitsumatsu

7 (332543N/1332408E) were within the aquaculture area whereas the third station,

8 Osaki (332591N/1332491E) was away from aquaculture related activities. The

9 salinity of the seawater above the sediments was 32-34 PSU and its temperature ranged

10 from 24-26°C (measured onboard using YSI 6600- multi-parameter water quality

11 monitor; model no 6600-D, USA). The ORP values were between -380 to -410 mV in the

12 aquaculture stations whereas at Osaki (non-aquaculture station) it was between -300 to -

13 320. Hardly any tidal exchange of water takes place at the first two stations whereas

14 Osaki experiences tidal exchange regularly. There is apparently no input of contaminants

15 from the adjacent shoreline.

16

17 Sample collection

18 Sediment samples were collected during the summer months (May-July) of 2006 by

19 Ekman Birge grab. For each sampling station, triplicate sediment samples were collected

20 and the outer sediment that were in contact with the grab, were removed to ensure that

21 there was no contamination from the sampling devise. The unrepresentative materials

22 (e.g., twigs, leaves, stones, wood chips and shells) were removed and the sediments were

23 mixed well manually by an ethanol-cleaned plastic spatula. After measurement of pH,

8
 1 temperature (by Horiba pH meter, Japan; model D-13) and redox condition (by TOA

2 ORP meter, Japan; model RM-12P), sub samples in two parts were collected in zip

3 locked plastic bag. The samples were kept on ice and were immediately transported to the

4 laboratory for analyses. Finally, an artificial fish feed used in the farms was analyzed to

5 determine their contribution to the accumulation of heavy metals (Zn, Cu and Pb etc.) in

6 sediments, mainly via uneaten food and faeces.

8 Preparation of samples

9 Approximately 50 gm sediment (slurry) from one part was freeze dried immediately and

10 later on stored at -20C for analyses of heavy metals. The other part of the sediment was

11 divided into two portions. One portion of these was used for analyses of sedimentary

12 organic matter, moisture content whereas the other portion was used for microbial and

13 molecular biological studies. One set of sample was stored at -20C for future reference.

14

15 Determination of moisture content and Organic matter (OM)

16 Aliquots of sediment samples (ca. 5 g) in triplicate were weighed using an analytical

17 balance and oven dried for 48 h at 90°C. The oven-dried samples were then weighed

18 again to determine the moisture content of the sediment. These sediment samples were

19 then transferred into a muffle furnace (FO 300, Yamato, Japan) and carbonized at 550°C

20 for 4 h. After the sediments cooled to ambient temperature, samples were reweighed, and

21 the organic-matter content was determined as the ash-free dry weight.

22

23 Analyses of heavy metals

9
 1 Total heavy metals

2 Two replicates of freeze-dried homogenized (powdered) sediments for each sample were

3 acid digested for the determination of metal by the following procedure. Approximately

4 300 mg of sediment was weighed into a TFM vessel (type M, 100ml) and then 9.0 ml of

5 nitric acid (HNO3) plus 3.0 ml of hydrofluoric acid (HF) was added. The vessels were put

6 into the holder and closed tightly. The system was placed in the multiwave microwave

7 (Perkin Elmer, MA, USA) on the rotating tray and a specific program (M=30 bar,

8 power=1000 W, temperature= maximum 220°C, time= 1h) was run. The system was

9 cooled and unscrewed. The digested samples were collected into acid-cleaned

10 polyethylene volumetric flask and for each transfer the digestion vessel was washed with

11 distilled water for a minimum of three times. The total volume of this material was

12 diluted to 50 ml with distilled water. This was further filtered through 5C filter paper

13 (quantitative ashless filter paper; Advantec Corporation, Japan) and stored in dark in

14 polyethylene bottles till further analysis.

15

16 Labile (HCl-extractable) metals

17 Two replicates of freeze-dried homogenized sediments for each sample were acid-treated

18 for the determination of HCl-extractable metal by the following procedure.

19 Approximately 300 mg of sediment was weighed using an analytical balance and were

20 placed in acid-washed 50 ml centrifuge tube. Ten (10) ml of 0.5 M HCl was poured into

21 the tube and capped immediately. The sediments were then agitated at 200 rpm for 1 h at

22 ambient temperature using an orbital shaker following Gillan et al.21. Sediments attached

23 to the inner wall of the tubes were washed down using distilled water and the total

10
 1 volume was made upto 15 ml. The whole material was then filtered with a 5C filter paper

2 and the filtrate was collected in an acid-cleaned polyethylene bottle till further analysis.

3 The 0.5 M HCl extraction removes sorbed metals but does not remove metals from the

4 matrix sediment particles themselves and thus represent the easily exchangeable (labile)

5 fraction of the metals.

7 Analytical measurements

8 The samples were analyzed for Zn by using an air acetylene flame atomic absorption

9 spectrophotometer (AA-6800; Shimadzu, Kyoto, Japan; equipped with ASC-6100 auto-

10 sampler). Analyses of Cu, Pb and other metals were performed using inductively coupled

11 plasma-atomic emission spectrometer (ICP-AES; Perkin Elmer Optima 4300 DV).

12 Samples were diluted with distilled water prior to analyses and reagent blanks were

13 prepared with the same amounts of reagents used in the analytical procedures. Calibration

14 was performed by analyzing a reagent blank sample and standard solutions within 0.5 to

15 2.0 mg/kg. Each analysis was repeated for three times including triplicate samples to

16 obtain the average value and its relative standard deviation was calculated throughout this

17 work.

18

19 Quality control

20 A quality control sample was routinely run through during the period of metal analysis.

21 To avoid possible contamination, all glassware and equipment used were acid washed

22 and the accuracy of the analysis was checked with the standard addition testing procedure.

23 The quality of the method used was checked with a Certified Reference Material (CRM-

11
 1 12) for marine sediment (National Institute of Environmental Studies, Japan). The

2 agreement between the analytical results for the reference material and its certified values

3 for each metal was satisfactory, with a recovery of at least 80% as shown in Table 1.

5 Chemicals

6 Super Special Grade HNO3 and HCl were procured from Wako Pure Chemical Industries,

7 Ltd., Tokyo Japan, whereas HF for atomic absorption spectrometry grade was form

8 Kanto Chemical, Tokyo, Japan. The standard solutions of 1000 mg/l were from Japan

9 Calibration Service System (JCSS). The heavy metals salts were from Nacalai Tesque

10 Incorporation, Japan.

11

12 RESULTS

13 Sediment quality

14 The sediments from the aquaculture stations were always dark black with strong foul

15 smell indicating to low to no oxygen condition. Visual observation showed that the

16 sediments form both of the aquaculture stations were clay dominated whereas the

17 sediments from the non-aquaculture station were dominated by silt and clay. There were

18 hardly any live benthic organisms found in the sediment samples from the aquaculture

19 stations though the control station did not show any great difference.

20

21 Moisture content and OM

22 The moisture content of the sediments from the aquaculture stations was always over

23 65% (Fig. 2). Moisture content in the sediments of Menokuso ranged from 66.1% (in

12
 1 May) to 70.2% (in July). Similar was the case at Mitusmatsu 63.0% (in May) to 68.0%

2 (in July). Interestingly enough at the non-aquaculture station (Osaki), moisture content

3 was around 44.0% showing the lowest value of 37.3% in July. The organic matter (OM)

4 showed significant difference (p=0.01) in between the aquaculture and non-aquaculture

5 stations (Fig. 3). Sediments at Menokuso had an average of 119±3.32 mg/g organic

6 matter whereas Mitsumatsu was recorded for an average of 107±3.71 mg/g OM. The

7 average organic matter in sediments at Osaki was only 55.8±10.1 mg/g. There was a

8 moderate correlation in between occurrence of heavy metals and organic matter

9 indicating that OM plays a role in binding metals, though it might not be the only

10 controlling factor.

11

12 Total Heavy metals

13 The average concentrations of heavy metals in the sediments from the study area during

14 the three months are presented in the Figures 4-9. We emphasize mainly on three metals

15 (Zn, Cu and Pb), which are of main concern at the study sites due to their continuous

16 input from fish feed, antifouling paints and probably fuel burning.

17

18 Zinc

19 Though the highest concentration (178 mg/kg dry wt.) of Zn was recorded from

20 Menokuso during May, the average concentration (165 mg/kg dry wt.) of this metal was

21 higher at Mitsumatsu (Fig. 4). The concentration of Zn at Osaki was quite lower (75.9

22 to118.4 mg/kg dry wt.) when compared to these two aquaculture stations during the

23 whole study period. One way analysis of variance (ANOVA) showed significant

13
1 difference (p=0.01) in concentrations of Zn in sediments from the three stations. The

2 trend in the concentration of this metal showed a decrease from May to July at all stations.

4 Copper

5 Occurrence of Cu at the aquaculture stations showed different trend from Zn. Mitsumatsu

6 showed the highest concentration (125 mg/kg dry wt.) of Cu in July and the Cu

7 concentration at this station was always near or above 100 mg/kg dry wt. (Fig. 5). The

8 concentration of Cu at Menokuso was comparatively lower (highest in May; 74.4 mg/kg

9 dry wt.) during the sampling period, whereas Osaki showed much lower concentration

10 (highest in May; 54.8 mg/kg dry wt.). The total Cu concentration was generally higher at

11 the aquaculture stations during the later months. There was significant difference (p=0.01)

12 in concentrations of Cu in sediments from the three stations.

13

14 Lead

15 Occurrence of Pb showed no specific trend though it showed the maximum concentration

16 during June at all the stations (Fig. 6). Like the other two heavy metals Pb was higher at

17 the aquaculture stations but the difference in concentrations in between the stations was

18 not significant (p=0.01) unlike the other two heavy metals. The highest concentration

19 (50.7 mg/kg dry wt.) of Pb was recorded at Mitsumatsu in June and the lowest at this

20 station was 33.7 mg/kg dry wt. in July. The minimum concentration (25.4 mg/kg dry wt.)

21 of Pb at Menokuso was recorded in May, whereas the highest concentration (41.0 mg/kg

22 dry wt.) occurred in June. Osaki had the highest Pb (33.2 mg/kg dry wt.) in June, whereas

23 the lowest (11.8 mg/kg dry wt.) was recorded in May.

14
 1

2 Other heavy metals

3 Three more heavy metals namely Cobalt (Co), Chromium (Cr), and Iron (Fe) were also

4 estimated from this study area. There was almost no difference in occurrence of Co over

5 the three stations during the study period (Fig. 7). The concentration of Co ranged from

6 23.3 to 31. 6 mg/kg dry wt.. The highest concentration (150 mg/kg dry wt.) of Cr was

7 recorded in June from Mitsumatsu whereas the lowest value (108.3 mg/kg dry wt.)

8 occurred in July at Osaki (Fig. 8). The concentration of Fe was recorded as high as 4.9%

9 in July at Menokuso and was similarly high during the other months mainly at the

10 aquaculture stations (Fig. 9). However there was no specific temporal or spatial trend in

11 the distribution of these heavy metals.

12

13 Labile metals

14 Estimation of bioavailability of heavy metals using the HCl-extractable fraction is

15 important, because the majority of the metals in sediments are included in various

16 mineral phases or chelated to organic matter and is consequently less bioavailable. The

17 labile fraction (HCl-extractable) of these heavy metals was interestingly quite high (Fig.

18 4-6). The labile fraction of Zn at Menokuso and Mitsumatsu kept increasing from May

19 and reached the maximum during July whereas there was no such trend at Osaki. There

20 was no specific trend for occurrence of labile fraction of Cu in all the stations though the

21 maximum was recorded in June from all the stations. The occurrence of labile fraction of

22 Pb at all the stations was also quite high during all the sampling months indicating to fact

23 that this occurrence is due probably to recent deposition. The labile fraction of Co and Cr

15
 1 in all the samples was always less than 10% and 7% respectively. Quite interestingly, the

2 labile fraction of Fe was also quite low which indicates to the fact that the Fe was bound

3 tightly in the sediment.

5 DISCUSSION

6 A major worldwide environmental issue is the introduction of persistent toxic substances,

7 including heavy metals into our environment22-24. Of the metals present in fish farm

8 sediments, elevated concentrations of zinc and copper have been reported in Scotland

9 (www.scotland.gov.uk/news/1999/12/se1670.asp), Canada25 and USA26. The

10 concentration of heavy metals in the Uranouchi inlet is much higher than their average

11 background level in Japanese river sediments, which was found to be 27.7 mg/kg dry wt.

12 for Cu, 17 mg/kg dry wt. for Pb and 121.9 mg/kg dry wt. for Zn27-28. Chou et al.29 studied

13 trace metals in the sediments around a salmon cage in New Brunswick, Canada. They

14 found a drastic increase in both copper and zinc concentration in a heavily sedimented

15 area. In anoxic conditions, zinc and copper concentrations reached to 253±85.7 µg/g and

16 54.5±5.1 µg/g respectively, while in normal conditions beyond 50 m from the cage the

17 concentrations were 2 to 3 times lower. Interaction of the metal and sulphide renders the

18 zinc harmless to benthos6,1. However, risk of high accumulation in semi closed water

19 bodies with low water exchange such as Uranouchi Inlet should be considered in farm

20 management practices. A comparison to the Interim Sediment Quality Guidelines of the

21 Canadian Government30, which states the concentrations of Zn, Cu and Pb are 123 mg/kg

22 dry wt., 35.7 mg/kg dry wt. and 35 mg/kg dry wt. respectively, the concentrations of

23 these heavy metals in the Uranouchi inlet appears to be reasonably high. As per the

16
 1 Netherlands Environmental Quality Guidelines31 the target concentrations of Zn, Cu and

2 Pb are 140 mg/kg dry wt., 36 mg/kg dry wt. and 85 mg/kg dry wt. respectively. Current

3 risk assessment of polluted soils and sediments is often based on normalization on clay

4 and organic carbon content, which is justified because organic carbon and clay provide

5 important sorption surfaces under aerobic conditions. This normalization procedure,

6 however, does not account for precipitation of trace metals with sulphides under reduced

7 conditions32. Thus our study without inclusion of this normalization procedure is

8 instrumental environmental risk assessment.

10 The environmental impact of heavy metals is closely related to their solubility since the

11 readily soluble forms of metals are considered to be the most bioavailable33. Detection of

12 labile fraction of the heavy metals in this study area thus reflects on the bioavailability of

13 the metals and indicates to their potential hazard. The trends of increasing sediment

14 concentration for Zn, Cu, and organic matter relate to the inputs from fecal and metabolic

15 waste from aquaculture feed. Quite interestingly, Zn concentrations at the most degraded

16 sites indicate that the seafloor is nearly saturated with Zn as a result of input from feed

17 and waste. The increasing sediment Cu in degraded conditions as seen at the aquaculture

18 stations is not surprising. Cu is known to bind tightly with organic material and a high

19 concentration factor is expected, especially in the presence of high organic carbon

20 concentrations26. The aquaculture feeds contain much lower concentrations of Fe, Mn,

21 and Al, which are at microgram per gram levels while the background sediments contain

22 percent (%) levels. The interrelationships between organic matter and sediment metals in

23 our study showed moderate correlations, which is not surprising because they come from

17
 1 the fish feed and the feces play a major role in the changes of the sediment conditions

2 under and nearby areas of the cage. Analysis of a sample of fish feed from one of the

3 farms at the Uranouchi inlet showed presence of Zn (87.59±3.15 mg/kg), Cu (14.44±0.3

4 mg/kg) and Pb (4.11±1.28 mg/kg) as integral constituents of the artificial feed. Dean et.

5 al.2 reported the presence of Zn (196.44±12.12 mg/kg), Cu (8.94±0.52 mg/kg) and Pb

6 (0.72±0.07 mg/kg) from a salmon feed that ultimately resulted in much higher

7 concentration in the sediments under and around the cages. Therefore, a causal

8 relationship between increase of metal concentration and practice of fish farming can be

9 confirmed. This can be also supposed from chemical variations such as of total organic

10 matter, total nitrogen and total sulfur. Although the later two parameters were not

11 included in our study and thus making it difficult to discuss on the reason of the

12 correlation, it is clear that the eutrophication, reduction state of the surface sediments, and

13 increase of metal contents were introduced after construction of fish farms. It cannot be

14 denied that seawater circulation in the bay is hampered by fish farms. The stagnation of

15 the seawater current at the area of inner inlet causes finer-grained distribution of the

16 sediments in and around the areas of aquaculture activity.

17

18 The extent of the environmental impact on the sea bottom is a function of local

19 assimilative capacity and amount of organic waste generated from aquaculture

20 activities34-35. The effect of physical–chemical parameters on metal availability is

21 complex and is a function not only of their interactions with the metals but also of these

22 parameters with each other. Occurrence of little portion of Fe as labile in our study

23 further indicates to the fact that the major portion is bound tightly in these nearly anoxic

18
 1 sediments. This in turn explains the occurrence of high levels of labile Zn or Cu as the

2 available sulphide is probably bound to Fe. The levels of labile heavy metals, however,

3 were not directly correlated with total heavy metals present. One possible explanation for

4 this discrepancy is some of the heavy metals may exist in insoluble organic and/or

5 inorganic complexes that could not be readily extractable36. Nevertheless this present

6 study reports for the first time the occurrence of several toxic heavy metals from the

7 Uranouchi inlet and it shows that this area is moderately polluted in terms of

8 concentrations of heavy metals when compared to some other studies (Table 2) and

9 according to the Shaanning pollution index (Table 3), which was developed by

10 Schaanning37 from 10 years of field data on electrode and elemental analyses of

11 sediments in the vicinity of salmon farms, describing the environmental situation of the

12 sediment in relation to normal. Moreover earlier studies have shown the presence of

13 tributyltin38 and antibiotics39 in the Uranouchi inlet.

14

15 The timely and cost-effective remediation of metal and organic co-contaminated sites

16 mandates an understanding of the extent and mechanisms by which toxic metals inhibit

17 organic biodegradation. Past attempts to quantify the impact of metals on biodegradation

18 are difficult to interpret because they have generally been based on total metal rather than

19 labile or bioavailable metal concentrations. Currently, our best approximation is to

20 measure and use labile metal data. Despite the enormous variance among reported

21 inhibitory concentrations of metals, it remains clear that metals have the potential to

22 inhibit organic biodegradation in both aerobic and anaerobic systems. Moderate to poor

23 correlation in between the organic matter content and the concentrations of the heavy

19
 1 metals, in particular, degradation of organic material and reduction in sulphide

2 concentrations, may increase metal bioavailability in the sediments, and might also result

3 in the release and further dispersal of metals away from fish farm sites. Although factors

4 other than metals may explain the distribution observed, the information presented here

5 may be useful in predicting long-term effects of heavy-metal contamination from

6 aquaculture in the marine environment. Slowing or reversing the effects of historical

7 practices in this region that harbors moderate level of heavy metal concentrations will

8 take many years but, with continued improvements, the natural and commercial values of

9 the Uranouchi inlet can be preserved and protected.

10

11 ACKWLEDGEMENTS

12 We thank the staff of the Marine Biological Research Center, Usa, Kochi, Japan for their

13 help during sampling. We also thank the students of the Graduate School of Kuroshio

14 Science (GRAKUS) for their kind cooperation. Critical reviews of the two anonymous

15 reviewers are gratefully acknowledged. De acknowledges JSPS Postdoctoral fellowship

16 for the financial support.

17

20
 1 REFERENCES

2 1. Carroll ML, Cochrane S, Fieler R, Velvin R, White P. Organic enrichment of

3 sediments from salmon farming in Norway: environmental factors, management

4 practices, and monitoring techniques. Aquaculture 2003; 226: 165–180.

5 2. Dean RJ, Shimmield TM, Black KD. Copper, zinc and cadmium in marine cage fish

6 farm sediments: An extensive survey. Environ. Poll. 2007; 145: 84-95.

7 3. Karakassis I, Hatziyanni E, Tsapakis M, Plaiti W. Benthic recovery following

8 cessation of fish farming: a series of successes and catastrophes. Mar. Ecol. Prog. Ser.

9 1999; 184: 205-218.

10 4. Pereira PMF, Black KD, Mclusky DS, Nickell TD. Recovery of sediments after

11 cessation of marine fish farm production. Aquaculture 2004; 235: 315-330.

12 5. Holmer M, Kristensen E. Impact of marine fish cage farming on metabolism and

13 sulphate reduction of underlying sediments. Mar. Ecol. Prog. Ser. 1992; 80: 191-201.

14 6. Brooks KM, Stierns AR, Mahnken CVW. Chemical and biological remediation of the

15 benthos near Atlantic salmon farms. Aquaculture 2003; 219: 355-377.

16 7. Brooks KM, Mahnken CVW. Interactions of Atlantic salmon in the Pacific Northwest

17 environment III. Accumulation of zinc and copper. Fish. Res. 2003; 62: 295–305.

18 8. Schendel EK, Nordstrom S, Lavkulich LM. Floc and sediment properties and their

19 environmental distribution from a marine fish farm. Aquac. Res. 2004; 35: 483-493.

20 9. Smith JN, Yeats PA, Milligan TG. Sediment geochronologies for fish farm

21 contaminants in Lime Kiln Bay, Bay of Fundy. In: Hargrave, B.T. (Ed.),

22 Environmental Effects of Marine Finfish Aquaculture. Springer, Berlin, 2005; pp.

23 221-238.

21
 1 10. Lorentzen M, Maage A. Trace element status of juvenile Atlantic salmon Salmo salar

2 L. fed a fish-meal based diet with or without supplementation of zinc, iron,

3 manganese and copper from first feeding. Aquac. Nutr. 1999; 5: 163-171.

4 11. Naylor SJ, Moccia RD, Durant GM. The chemical composition of settleable solid fish

5 waste (manure) from commercial rainbow trout farms in Ontario, Canada. North Am.

6 J. Aquacult. 1999; 61: 21–26.

7 12. Naylor RL, Goldburg RJ, Primavera JH, Kautsky N, Beveridge MCM, Clay J, Folke

8 C, Lubchenco J, Mooney H, Troell M. Effect of aquaculture on world fish supplies.

9 Nature 2000; 405: 1017-1024.

10 13. Kolpin DW, Furlong ET, Meyer MT, Thurman EM, Zaugg SD, Barber LB, Buxton

11 HT. Pharmaceuticals, hormones, and other organic wastewater contaminants in US

12 streams, 1999–2000: a national reconnaissance. Environ. Sci. Technol. 2002; 36:

13 1202–1211.

14 14. Nriagu JO. A history of global metal pollution. Science. 1996; 272: 223–224.

15 15. Stepanauskas R, Glenn TC, Jagoe CH., Tuckfield RC, Lindell AH, King CJ,

16 McArthur JV. Coselection for microbial resistance to metals and antibiotics in

17 freshwater microcosms. Environ. Microbiol. 2006; 8(9): 1510–1514.

18 16. Hogstrand C, Wood CM. The physiology and toxicology of zinc in fish. In: Taylor

19 EW (Ed) Toxicology of aquatic pollution, physiological, molecular and cellular

20 approaches. Cambridge University Press, Cambridge, 1996; 61–84.

21 17. Fan W, Wang WX, Chen J. Geochemistry of Cd, Cr, and Zn in highly contaminated

22 sediments and its influence on assimilation by marine bivalves. Environ. Sci. technol.

23 2000; 36: 5164-5171.

22
 1 18. Ruangdej U, Fukami K. Stimulation of photosynthesis and consequent oxygen

2 production in anoxic bottom water by supply of low-intensity light through an optical

3 fiber. Fish. Sci. 2004; 70: 421–429.

4 19. Karim MA, Fukami K, Patel AB. Enhancement of inorganic nutrient regeneration in a

5 eutrophic sediment–bottom water complex system by adding effective indigenous

6 bacteria. Fish. Sci. 2003; 69: 1146–1157.

7 20. Fukami K, Ruangdej U, Patel AB, Nishijima T. Improvement of eutrophic coastal

8 bottom environments by using an optical fiber and effective psychrophilic bacteria.

9 Fish. Sci. 2002; 68: 617–620.

10 21. Gillan DC, Danis B, Pernet P, Joly G, Dubois P. Structure of sediment-associated

11 microbial communities along a heavy-metal contamination gradient in the marine

12 environment. Appl. Environ. Microbiol. 2005; 71(2): 679–690.

13 22. Herrman JL. The role of the World Health Organization in the evaluation of

14 pesticides. Regul. Toxicol. Pharmacol. 1993; 17: 282–286.

15 23. Ramaiah N, De J. Unusual rise in mercury resistant bacteria in coastal environs.

16 Microb. Ecol. 2003; 45: 444-454.

17 24. De J, Sarkar A, Ramaiah N. Bioremediation of toxic substances by mercury resistant

18 marine bacteria. Ecotoxicology 2006; 15(4): 385-389.

19 25. Yeats P, Milligan T, Sutherland TF, Robinson S, Smith J, Lawton P, Levings CD.

20 Lithium-Normalized Zinc and Copper Concentrations in Sediments as Measures of

21 Trace Metal Enrichment due to Salmon Aquaculture. In: Hargrave BT (Ed.).

22 Handbook of Environmental Chemistry: The Environmental Effects of Marine

23 Finfish Aquaculture. 2005; 5(M): 207-220.

23
 1 26. Chou CL, Haya K, Paon LA, Moffatt JD. A regression model using sediment

2 chemistry for the evaluation of marine environmental impacts associated with salmon

3 aquaculture cage wastes. Mar. Poll. Bull. 2004; 49: 465–472.

4 27. Shibahara M, Yamazaki R, Nishida K, Suzuki J, Suzuki S, Nishida H, Tada F.

5 Distribution of heavy metals in bottom mud of river. III. Bottom mud of principal

6 rivers in Japan. J. Hygien. Chem. 1975; 21: 173-182 (in Japanese with English

7 summary).

8 28. Baasansuren J, Okazaki M, Choi IS, Toyota K. Tin concentrations in River/Bay

9 sediments of Tokyo in 1984 and 2000. Soil. Sci. Plant. Nutr. 2002; 48(5): 667-672.

10 29. Chou CL, Haya K, Paon LA, BurridgeL, Moffatt JD. Aquaculture-related trace metals

11 in sediments and lobsters and relevance to environmental monitoring program ratings

12 (EMP) for near-field effects. Mar. Poll. Bull. 2002; 44(11):1259-1269.

13 30. CCME, 1997. Recommended Canadian Soil Quality Guidelines. CCME, ISBN 1-

14 895-925-92-4.

15 31. Department of Soil Protection, Netherlands. The Netherlands Soil Contamination

16 Guidelines. Netherlands Intervention Values for Soil Remediation. Reference #

17 DBO/07494013. 1994.

18 32. Griethuysen CV, Meijboom EW, Koelmans AA. Spatial variation of metals and acid

19 volatile sulfide in floodplain lake sediment. Environ. Toxicol. Chem. 2003; 22(3):

20 457-465.

21 33. Hickey MG, Kittrick JA. Chemical partitioning of cadmium, copper, nickel and zinc

22 in soils and sediments containing high levels of heavy metals. J. Environ. Qual. 1984;

23 13: 372-376.

24
 1 34. GESAMP (IMO/FAO/UNESCO-IOC/WMO/WHO/IAEA/UN/UNEP Joint Group of

2 Experts on the Scientific Aspects of Marine Environmental Protection). Towards safe

3 and effective use of chemicals in coastal aquaculture. Reports and Studies, GESAMP.

4 1997; No. 65. Rome, FAO. p.40.

5 35. Black KD. (Ed.). Environmental Impacts of Aquaculture. Sheffield Academic Press,

6 Sheffield, UK. 2001; pp.214.

7 36. Burton GA. Sediment toxicity evaluations: their niche in ecological assessments.

8 Environ. Sci. Tech. 1992; 26: 2068–2074.

9 37. Schaanning MT. Distribution of sediment properties in coastal areas adjacent to fish

10 farms and environmental evaluation of five locations surveyed in October 1993.

11 Norwegian Institute for Water Research (NIVA), Report No. O-93205, O-93062.

12 Oslo, Norway. 1994; pp. 29 + appendices.

13 38. Murai R, Takahashi S, Tanabe S, Takeuchi I. Status of butyltin pollution along the

14 coasts of western Japan in 2001, 11 years after partial restrictions on the usage of

15 tributyltin. Mar. Poll. Bull. 2005; 51: 940–949.

16 39. Maki T, Hasegawa H, Kitami H, Fumoto K, Munekage Y, Ueda K. Bacterial

17 degradation of antibiotic residues in marine fish farm sediments of Uranouchi Bay

18 and phylogenetic analysis of antibiotic-degrading bacteria using 16S rDNA sequences.

19 Fish. Sci. 2006; 72(4): 811-820.

22

25
 1 LEGENDS TO FIGURES

2 Figure 1. Location of sampling stations

3 Figure 2. Moisture content in sediments of the Uranouchi Inlet

4 Figure 3. Organic matter content in sediments of the Uranouchi Inlet

5 Figure 4. Total and labile Zn in sediments of the Uranouchi Inlet

6 Figure 5. Total and labile Cu in sediments of the Uranouchi Inlet

7 Figure 6. Total and labile Pb in sediments of the Uranouchi Inlet

8 Figure 7. Total Co in sediments of the Uranouchi Inlet

9 Figure 8. Total Cr in sediments of the Uranouchi Inlet

10 Figure 9. Total Fe in sediments of the Uranouchi Inlet

11

12

13

14

26
1 Table 1: Recovery of heavy metals from reference material (CRM-12; NIES, Japan)

Heavy metal Conc. (mg/kg dry Recovered conc. Recovery %

wt.) in CRM-12 (mg/kg dry wt.)

Zn 738 670 90.8

Cu 104 100 96.2

Pb 101 95.4 94.5

Co 16.6 20 120.5

Cr 201 177 87.9

Fe* 4.31 3.51 81.0

3 *concentration is in percentage (g/100g) level

27
1 Table 2: Status of the Uranouchi Inlet in comparison with world wide reports

Area Zn Cu Pb Reference

Tokyo Bay (CRM- 738 104 101 http://www.nies.go.jp/labo/crm-

12) e/index.html

Uranouchi >177 >125 >50 This study

British Columbia 200 - - Brooks & Mahnken, 2003

& US-Salmon

cage

New Burnswick, 253 54.5 - Chou et al., 2002

Canada

Baia de 235 163 52 Stringer et al, 2000

Guanabara, Brazil

Hong Kong coastal 197.2 199.5 151.1 Wong et al., 2000

sediment

Gold Coast 30.1- 8.3-194 16.3- Burton et al., 2005

Broadwater, 220 74.8

Australia

28
1 Table 3: Pollution index of sediment (modified from Shaanning et al., 1994)

Degree Index pH Zn(µg g-1) Cu(µg g-1)

Large 3 <6.9 >650 >150

Moderate 2 6.9-7.2 150-650 25-150

Small 1 7.21-7.77 5-150 5-25

No 0 >7.7 <5 <5

29
1

5 Fig. 1

30
 80

May

May
60

June

June
July

July
Moisture (%)

May
40

June

July
20

0
Menokuso Mitsumatsu Osaki
Station
1
2

3 Fig. 2.

31
 15
OM-May
OM-June
OM-July

10
OM (mg/g dry wt.)

0
Menokuso Mitsumatsu Osaki
Station
1
2

3 Fig. 3.

32
 200

Total-May
180
Total-June
Total-July
160 Labile-May
Labile-June
140 Labile-July
Conc. (mg/kg dry wt.)

120

100

80

60

40

20

0
Menokuso Mitsumatsu Osaki
Stations
1
2

3 Fig. 4.

33
 140
Total-May
Total-June
120
Total-July
Labile-May
Labile-June
100
Labile-July
Conc. (mg/kg dry wt.)

80

60

40

20

0
Menokuso Mitsumatsu Osaki
Stations
1

2 Fig. 5.

34
 60
Total-May
Total-June

50 Total-July
Labile-May
Labile-June
Labile-July
40
Conc. (mg/kg dry wt.)

30

20

10

0
Menokuso Mitsumatsu Osaki
Stations
1

2 Fig. 6.

35
 35
Total-May
Total-June
30
Total-July

25
Conc. (mg/kg dry wt.)

20

15

10

0
Menokuso Mitsumatsu Osaki
Stations
1

3 Fig. 7.

36
 160
Total-May
Total-June
140
Total-July

120
Conc. (mg/kg dry wt.)

100

80

60

40

20

0
Menokuso Mitsumatsu Osaki
Stations
1
2

3 Fig. 8.

37
 6

Total-May
Total-June
5 Total-July

4
Conc. (%)

0
Menokuso Mitsumatsu Osaki
Stations
1

2 Fig. 9.

38