Changes in body condition and fatty acid composition of ......Marine ﬁsh, especially carnivores, have a natural diet rich in highly unsaturated x3 fatty acids. As a consequence,

MARINE

Marine Environmental Research 63 (2007) 1–18

www.elsevier.com/locate/marenvrev

ENVIRONMENTAL

RESEARCH

Changes in body condition and fatty acidcomposition of wild Mediterranean horse mackerel

(Trachurus mediterraneus, Steindachner, 1868)associated to sea cage fish farms

Damian Fernandez-Jover a,*, Jose Angel Lopez Jimenez b,Pablo Sanchez-Jerez a, Just Bayle-Sempere a,

Francisca Gimenez Casalduero a, Francisco Javier Martinez Lopez b,Tim Dempster c

a Marine Biology Unit, Department of Marine Sciences and Applied Biology, University of Alicante,

P.O. Box 99, E - 03080, Alicante, Spainb Department of Physiology, Faculty of Biology, University of Murcia, Campus of Espinardo,

P.O. Box 30100, Murcia, Spainc SINTEF Fisheries and Aquaculture, 7465 Trondheim, Norway

Received 16 December 2005; received in revised form 8 May 2006; accepted 12 May 2006

Abstract

Net-cage fish farms attract a great number of wild fishes, altering their behaviour and possiblytheir physiology. Wild Mediterranean horse mackerel (Trachurus mediterraneus), sampled from pop-ulations aggregated around two Mediterranean fish farms and from two natural control populations,were analyzed for differences in body condition, stomach content and fatty acid composition. Pelletsused to feed caged fish in both farms were also analyzed to identify their relationship with the fattyacid composition of tissue of wild fish. T. mediterraneus aggregated around the farms throughout theyear although large seasonal changes in abundance and biomass occurred. Wild fish aggregated atfarms mainly ate food pellets while control fish fed principally on juvenile fish and cephalopods.Wild fish that fed around the cages had a significantly higher body fat content than the control fish(7.30 ± 1.8% and 2.36 ± 0.7%, respectively). The fatty acid composition also differed between farm-associated and control fish, principally because of the significantly increased levels of linoleic

0141-1136/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.marenvres.2006.05.002

* Corresponding author. Tel.: +34 96 590 3400(x)2916; fax: +34 96 590 9840.E-mail address: [email protected] (D. Fernandez-Jover).

mailto:[email protected]

2 D. Fernandez-Jover et al. / Marine Environmental Research 63 (2007) 1–18

(C18:2x6) and oleic (C18:1x9) acids and decreased docosahexaenoic acid (C22:6x3) in farm-associ-ated fish. The increased condition of wild fish associated with farms could increase the spawning abil-ity of coastal fish populations, if wild fish are protected from fishing while they are present at farms.The fatty acids compositions could also serve as biomarkers to infer the influence of a fish farm onthe local fish community, helping to better describe the environmental impact of fish farming.� 2006 Elsevier Ltd. All rights reserved.

Keywords: Aquaculture; Fatty acid; Body condition; Mediterranean; Trachurus mediterraneus; Wild fish; Bio-marker

1. Introduction

The presence of wild fish beneath sea-cage fish farms has often been noted, and mayaffect the presence, abundance and residence times of fishes in a given area (Carss,1990; Bjordal and Skar, 1992). Dempster et al. (2002) demonstrated that sea-cage fishfarms act as ‘super-FADs’ (FAD: fish attraction device) in the south-western Mediterra-nean Sea, attracting large multi-species schools of pelagic fish. However, the potential eco-logical and biological effects on wild fish populations have rarely been tested despite thepresent high level of public interest in the issue in many countries. As production fromsea-cages in coastal regions continues to expand rapidly (FAO, 2003, Federation of Euro-pean Aquaculture Producers, www.feap.info/feap), information on such effects willbecome increasingly important for managing the interactions between aquaculture andwild fish populations.

Marine fish, especially carnivores, have a natural diet rich in highly unsaturated x3fatty acids. As a consequence, long-chain x3 polyunsaturated fatty acids occur in higherconcentrations in marine fish muscle (Ackman, 1967). The production of fish, such as sea-bream (Sparus aurata), and sea-bass (Dicentrarchus labrax) in coastal fish farms is carriedout using food pellets composed in part with vegetable oils. The corn or soya used in foodproduction gives a high concentration of oleic acid (18:1x9), linoleic acid (18:2x6) and a-linolenic acid (18:3x3). The introduction of this source of food to the marine environmentcould modify the fatty acid (FA) composition of wild fish that feed on the lost pellets aswell as their total fat content due to the high availability of food.

In this study, we tested if Mediterranean horse mackerel (Trachurus mediterraneus), acarnivorous species of commercial interest in the Mediterranean Sea, aggregate aroundfish farms throughout the year using visual counts. Further, we tested whether the dietof T. mediterraneus differed when it was aggregated around fish farms compared to naturalcontrol locations and if this led to a modification of its muscle nutritional condition andFA composition. Finally, we attempted to identify if fatty acid compositions could func-tion as a ‘physiological tag’ that could be useful in evaluating the impact of the lost feedfrom farms on marine trophic nets.

2. Materials and methods

T. mediterraneus aggregated around two fish farms separated by 50 km in the south-eastof Spain were spear-fished (Fig. 1). The farm at Campello was 3.2 km off the coast at an

http://www.feap.info/feap

Fig. 1. Map of the four study locations along the south-east coast of Spain in the Mediterranean Sea.

D. Fernandez-Jover et al. / Marine Environmental Research 63 (2007) 1–18 3

average depth of 28.6 m, with 12 cages of 17 m diameter and 17 m deep producing300 t yr�1 of fish. The farm at Guardamar was 3.7 km from shore at a depth of 22.6 m,with 24 cages of 19 m diameter and 15 m deep producing 1000 t yr�1. Both farms rearedsea-bass (Dicentrarchus labrax) and sea-bream (Sparus aurata). Individuals from naturalcontrol populations were obtained from commercial, bottom-trawling fisheries in twolocations at least 10 km away from the nearest fish farm (Villajoyosa and Alicante). Tenfish from each locality were used to carry out the analysis. Farm-associated and controlfishes were caught on 3 different random days but during the same period (July 2005)to avoid seasonal variation in their chemical composition (e.g. Bandarra et al., 2001).Farm-associated fish ranged from 179 to 410 mm total length (TL) while control fish ran-ged from 227 to 315 mm TL. Commercial food pellets used to feed the crop were also ana-lyzed; two different types of pellets were used in Campello (Types I and II) while only onewas used in Guardamar (Type III).

To estimate the abundance of T. mediterraneus aggregated around the cages, we con-ducted rapid visual counts (RVCs; Kingsford and Battershill, 1998) using SCUBA. Fulldetails of the count methodology are given in Dempster et al. (2002). At each farm, fishwere counted at 3 random times over a period of 2 months for every season from winter2004 until the fish sampling time in summer 2005 (January and February for winter, Apriland May for spring, August and September for summer and October and November forautumn). Six 5-min rapid visual counts were conducted each time within the farm com-plex. Each count covered a volume of approximately 11,250 m3 (15 m wide · 15 mdeep · 50 long). During counts, the average total length (TL) of each group was noted.Count data were entered into the ecoCEN program (Bayle-Sempere et al., 2002), whereconversions to biomass were made using ecoCEN based on a published length–weightrelationship for T. mediterraneus. Counts and sampling by spear-fishing were conductedon separate days.

Fish were weighed after emptying the stomach content and measured. After dissec-tion, the liver was weighed and stomach contents were analyzed to determine the dietary


composition of farm-associated and control fish. Gut contents were extracted and preyitems were identified under the microscope. The percentage of fish containing each kindof prey item was calculated.

All fish were iced after landing. In the laboratory, a part of the anterior-dorsal whitemuscle (approximately 6 g) was removed, frozen at �18 �C and analyzed within one week.Tissue composition was determined after homogenisation as follows: protein (Kjeldahlmethod with a 6.25 nitrogen to protein conversion factor), fat (ethyl-ether extraction usinga SOXTEC System HT6 extractor), moisture (drying to constant weight in an oven at105 ± 1 �C), total ash (incineration to constant weight in a muffle oven at 450 ± 2 �C)and diet NFE (nitrogen-free extract) was calculated by subtracting the other componentsfrom 100.

After individual tissue homogenisation, the fatty acid composition of the total lipidfraction was determined by fat extraction following the method of Folch et al. (1957), witha mixture of chloroform and methanol (1:1 proportion for the first extraction and 2:1 pro-portion for the second one). Fatty acid methyl esters (FAME) samples were analyzedaccording to the method of Stoffel et al. (1959) by gas-liquid chromatography using aSP� 2560 flexible fused silica capillary column (100-m length and 0.25-mm internal diam-eter and 0.20 lm of film thickness; SUPELCO) in a Hewlett-Packard 5890 gas chromato-graph. The oven temperature was programmed for 5 min at an initial temperature of140 �C, was increased at a rate of 4 �C per min to 230 �C, was further increased at a rateof 1 �C per min to 240 �C and then held at that temperature for 6 min. The injector andflame ionization detector were set at 250 �C. Helium was used as a carrier gas at a pressureof 290 kPa, and peaks were identified by comparison of their retention times with appro-priate FAME standards purchased from Sigma Chemical Company (St. Louis, MO,USA). Individual FA concentrations were expressed as percentages of the total content.

Fulton’s Condition Factor Index (CI = [100,000 · (W)]/L3, where W = weight andL = length), and Liver-somatic Index (LSI = 100 · [liver weight/total weight]) were calcu-lated as indicators of fish condition.

Non-parametric multivariate techniques were used to compare FA compositions offish and pellets. All multivariate analyses were performed using the PRIMER statisticalpackage. Triangular similarity matrices were calculated using the Bray–Curtis similaritycoefficient (Clarke and Warwick, 1994). Non-metric multidimensional scaling (nMDS)was used as the ordination method. Variables that had more influence on similaritieswithin groups and dissimilarities among groups of locations were calculated usingthe SIMPER (similarity percentages) procedure (Warwick et al., 1990; Clarke, 1993).A permutation test (PERMANOVA) was used to assess the significance of the overallfatty acid composition among the considered sources of variation (Clarke, 1993;Anderson, 2004).

To test whether fat content and the main FAs varied among the two farms and two con-trol sites, we used an analysis of variance (ANOVA) which incorporated the factors – feed-ing (fixed) with two treatments (pellets and natural diet) and locations (random and nestedin feeding) with two random localities; Campello and Guardamar for pellets and Villajoy-osa and Alicante for natural diet.

Prior to ANOVA, heterogeneity of variance was tested with Cochran’s C-test. As datawere percentages, they were transformed with arcosen (x + 1). ANOVA is robust to het-erogeneity of variances, particularly when experiments are large and balanced (Under-wood, 1997).


3. Results

3.1. Visual counts

Rapid visual counts indicated that T. mediterraneus aggregated around farms through-out the study period at both Campello and Guardamar, except for winter 2005 at Guard-amar. However, the size of aggregations (abundance and biomass) differed greatly amongtimes. At Campello, abundance and biomass peaked during autumn 2004 (289.9 ± 122.3ind./11,250 m3 and 68.8 ± 26.2 kg/11,250 m3) and large aggregations also occurred inspring and summer 2005 (Fig. 2). Lowest abundances at this farm occurred in spring of2004 and winter of 2005 (8–10 ind./11,250 m3and 1–2 kg/11,250 m3). At Guardamar, var-iation in abundances was also high; the population remained at low levels until spring andsummer of 2005 when abundances reached 110–313 ind./11,250 m3 and biomasses werebetween 30 and 37 kg/11,250 m3.

Fig. 2. Number of individuals and biomass of Trachurus mediterraneus aggregated around the 2 fish farms. Barsrepresent the average of 18 censuses made during 3 sampling days (6 censuses per day). Error bars indicate ±SE.


3.2. Stomach contents

Fish caught by local commercial fisheries far from the farms used a variety of foodsources, principally juvenile fish, crustaceans and cephalopods (Fig. 3). Diet compositiondiffered between the two natural control populations; cephalopods were most important atVillajoyosa while juvenile fish dominated the stomach contents of the fish from Alicante.In contrast, food pellets were the main item found in the stomach of T. mediterraneus

aggregated around both the Guardamar and Campello farms. Other prey items, such asjuvenile fish, made up only a small proportion of the diet.

Fig. 3. Stomach contents of Trachurus mediterraneus aggregated around farms (Campello and Guardamar) andnon-aggregated (Alicante and Villajoyosa). Values are percentages of wet weight.

Fig. 4. Individual total fish fat content (percentage of total body composition) and total length (mm) in farmsand control sites.


3.3. Body condition

Farm-associated fish had on average 3.5 times higher body fat content that control fish(7.30 ± 1.8% vs. 2.36 ± 0.7%). Fat content levels were considerably more variable in thefarm-associated fish (Fig. 4), ranging from 1.5% to 13% in contrast to the fat levels of

Table 1Composition of the three kinds of food pellets used to feed farmed fish; values represent percentages as mean ±standard deviation

Campello Type I Campello Type II Guardamar Type III

Fat 22.57 ± 0.61 17.73 ± 0.06 24.04 ± 0.01Protein 42.62 ± 0.15 44.50 ± 0.31 42.49 ± 0.29Moisture 6.81 ± 0.20 10.51 ± 4.05 5.80 ± 0.15Ash 7.36 ± 0.24 7.00 ± 0.06 8.70 ± 0.02NFE 20.65 ± 0.72 20.26 ± 3.62 18.98 ± 0.15C12:0 0.3 0 0C14:0 6.81 3.38 3.03C15:0 0.97 0.49 0.63C15:1x5 0.16 0 0C16:0 18.17 20.48 17.53C16:1x7 5.8 7.07 3.23C17:0 0.52 0.45 0.5C17:1x7 0.77 0.16 0.2C18:0 3.98 6.19 4.8C18:1x9 10.39 14.57 16.18C18:1x7 2.31 3.41 2.16C18:2x6 14.78 8.7 23.35C20:0 0.3 0.26 0.35C18:3x6 1.84 0.28 2.21C20:1x9 0.87 4.57 1.67C18:3x3 1.87 1.55 2.65C18:4x3 1.87 1.43 0.81C20:2x6 0.73 0.64 0.67C22:0 0.26 0.11 0.26C20:3x6 0 0.38 0.45C22:1x9 0 1.18 1.12C20:3x3 0 0.31 0C20:4x6 0.64 0.68 1.14C24:0 0.22 0 0.2C20:5x3 13.47 6.64 5.59C24:1x9 0.33 1.1 0.49C22:4x6 0 0.08 0C22:5x3 1.1 1.2 0.76C22:6x3 11.52 14.68 10.03x3 29.83 25.81 19.84x6 18 10.78 27.82x5 0.16 0 0x7 8.11 10.48 5.39x9 11.6 21.42 19.45Saturated 31.53 31.35 27.3Monounsaturated 20.64 32.06 25.04Polyunsaturated 47.83 36.59 47.66

NFE: Nitrogen free extract.

Table 2Fulton’s Condition Index (CI) and Liver Somatic Index (LSI) for the fish (n = 10) from the four localities

Location Length (cm) Weight (g) CI LSI

Villajoyosa Control 291.7 ± 27.9 222.9 ± 87.7 0.86 ± 0.06 1.32 ± 0.50Alicante Control 250.3 ± 15.2 134.4 ± 20.3 0.85 ± 0.04 1.60 ± 0.61Campello Farm 378.7 ± 20.6 570.9 ± 107.5 1.04 ± 0.11 1.02 ± 0.34Guardamar Farm 302.9 ± 30.9 275.71 ± 63.0 1.02 ± 0.24 1.11 ± 0.47

Values represent means ± standard deviation.

Table 3Fatty acid composition of associated and non-associated Trachurus mediterraneus at farms and control sites(n = 10, % of total fatty acids)

Name Villajoyosa Control Alicante Control Campello Farm Guardamar Farm

C12:0 0.25 ± 0.51 0.16 ± 0.24 0.05 ± 0.08 0.14 ± 0.15C13:0 1.02 ± 3.23 0.00 ± 0 0.00 ± 0 0.00 ± 0C14:0 2.03 ± 0.42 2.53 ± 1.09 3.58 ± 0.55 3.32 ± 0.69C14:1x5 0.00 ± 0 0.00 ± 0 0.00 ± 0 0.04 ± 0.12C15:0 0.46 ± 0.35 0.57 ± 0.36 0.35 ± 0.19 0.49 ± 0.22C15:1x5 0.00 ± 0 0.00 ± 0 0.00 ± 0 0.02 ± 0.06C16:0 21.84 ± 1.94 23.39 ± 3.11 19.73 ± 1.79 18.81 ± 2.74C16:1x7 1.97 ± 0.4 2.66 ± 1.28 4.49 ± 0.62 4.63 ± 1.83C17:0 0.66 ± 0.24 0.55 ± 0.3 0.48 ± 0.21 0.67 ± 0.29C17:1 0.24 ± 0.27 0.27 ± 0.22 0.21 ± 0.16 0.31 ± 0.28C18:0 8.63 ± 2.67 7.85 ± 1.2 7.28 ± 2.88 7.66 ± 3.69C18:1x9 9.30 ± 1.63 11.34 ± 2.45 18.81 ± 2.68 20.60 ± 9.09C18:1x7 1.65 ± 0.2 1.73 ± 0.99 2.69 ± 0.23 2.70 ± 0.69C18:2x6 2.69 ± 1.22 2.77 ± 2.88 14.26 ± 2.11 12.02 ± 5.59C20:0 0.15 ± 0.2 0.20 ± 0.17 0.23 ± 0.18 0.29 ± 0.14C18:3x6 0.59 ± 1.08 0.40 ± 0.62 0.66 ± 1.11 1.81 ± 3.22C20:1x9 0.45 ± 0.33 0.98 ± 1.31 1.90 ± 0.82 1.93 ± 1.16C18:3x3 0.24 ± 0.27 0.51 ± 0.71 1.60 ± 0.24 1.29 ± 0.61C18:4x3 0.07 ± 0.12 0.19 ± 0.25 0.50 ± 0.31 0.57 ± 0.43C20:2x6 0.21 ± 0.27 0.44 ± 0.33 0.56 ± 0.4 0.64 ± 0.26C22:0 0.35 ± 0.31 0.33 ± 0.22 0.18 ± 0.17 0.19 ± 0.21C20:3x6 0.18 ± 0.31 0.13 ± 0.31 0.23 ± 0.39 0.78 ± 1.32C22:1x9 0.02 ± 0.06 0.12 ± 0.28 0.28 ± 0.25 0.20 ± 0.28C20:3x3 0.00 ± 0 0.06 ± 0.17 0.05 ± 0.09 0.04 ± 0.09C20:4x6 1.45 ± 0.25 1.42 ± 0.36 0.68 ± 0.41 0.68 ± 0.21C23:0 0.00 ± 0 0.00 ± 0 0.00 ± 0 0.15 ± 0.48C24:0 0.72 ± 0.3 0.46 ± 0.31 0.14 ± 0.17 0.18 ± 0.24C20:5x3 5.83 ± 1.09 6.18 ± 1.61 6.04 ± 1.67 4.53 ± 1.74C24:1x9 0.98 ± 0.37 0.90 ± 0.37 0.40 ± 0.23 0.51 ± 0.2C22:4x6 0.05 ± 0.1 0.02 ± 0.07 0.05 ± 0.08 0.02 ± 0.05C22:5x3 1.36 ± 0.53 1.29 ± 0.56 1.39 ± 0.24 1.11 ± 0.43C22:6x3 36.61 ± 4.01 32.56 ± 6.55 13.19 ± 2.63 13.64 ± 3.72x3 44.11 ± 5.2 40.79 ± 6.99 22.76 ± 4.31 21.19 ± 4.63x6 5.17 ± 1.82 5.19 ± 2.82 16.43 ± 2.93 15.95 ± 5.27x5 0.00 ± 0 0.00 ± 0 0.00 ± 0 0.06 ± 0.13x7 3.62 ± 0.57 4.38 ± 2.11 7.18 ± 0.82 7.33 ± 2.47x9 10.75 ± 1.51 13.34 ± 3.75 21.40 ± 3.15 23.25 ± 9.01Saturated 36.10 ± 4.77 36.04 ± 3.16 32.02 ± 4.18 31.91 ± 6.81Monounsaturated 14.62 ± 1.88 17.99 ± 5.45 28.79 ± 3.16 30.95 ± 10.86Polyunsaturated 49.28 ± 5.22 45.98 ± 6.44 39.19 ± 4.17 37.14 ± 7.81

Data are expressed as mean ± standard deviation.



control fish which remained stable within a narrow range (mainly 1–4%). The total lipidamount in farm-associated fish was significantly higher (p < 0.01) than that in control fish,which is a common difference between cultivated and wild fish. Total protein content,however, did not differ between farm-associated and control fish. In spite of the biggeraverage size of farm-associated fish, total fat content was not correlated with fish size fromeither farm-associated or control fish.

The protein and fat content of food pellets were similar for Guardamar and ‘CampelloType I’, but ‘Campello Type II’ contained lower levels of fat and higher levels of protein(Table 1). Significant differences existed in the Condition and Liver Somatic indexesbetween the farm-associated and control fish. Farm-associated fish had a higher ConditionIndex (1.02–1.04) than found in control fish (0.85–0.86, p < 0.01, Table 2). In contrast, theLiver Somatic Index was higher in control fish (1.32–1.60) than in farm-associated fish(1.02–1.11, p = 0.1).

3.4. Fatty acid composition

The total FA composition differed significantly between control and farm-associatedfish (Table 3). The two-dimensional MDS plot (Fig. 5) based on relative presence of dif-ferent FAs revealed a clear separation of the two groups (farm-associated and controlhorse mackerel) with a low stress value (0.08). Further, the separation of the groupsamong localities was clearer in farm-associated fish. The FA composition of food pelletspresented a similar distribution to the farm-associated fish. The composition of pellet TypeII, used to feed the caged fish in Campello, appeared particularly close to the fatty acidcomposition of wild T. mediterraneus aggregated there. Such a close match between pelletand wild fish composition, however, did not occur for Type I pellets at Campello or TypeIII pellets at Guardamar.

PERMANOVA analysis showed that the differences in FA composition between treat-ments were significant (p < 0.001). SIMPER analysis indicated that docosahexaenoic acid(C22:6x3) was the main differentiating FA (32.6% contribution to dissimilarity) due to itselevated presence in control T. mediterraneus (Table 4). The two other main differentiating

Fig. 5. MDS. ¤: Guardamar Farm; n: Campello Farm; ,: Alicante Control; n: Villajoyosa Control; > Type IPellets; · Type II Pellets; + Type III pellets. Stress value = 0.08. Food pellets Types I and II were used inCampello; food pellets Type III were used in Guardamar. Circles with percentages (Bray–Curtis similarities)enclose major groups with similarities greater than the given percentage.

Table 4Contributions of the main fatty acids to overall dissimilarities between associated and non-associated T.

mediterraneus and percentage contributions to the cumulative dissimilarity

Fatty acids Non-associated Associated

Averageabundance

Averageabundance

Averagedissimilarity

Dissimilarity/SD Contribution% Cumulative%

C22:6x3 34.58 13.41 10.59 3.34 32.62 32.62C18:2x6 2.73 13.14 5.26 2.4 16.22 48.84C18:1x9 10.32 19.71 4.77 1.45 14.7 63.54C16:0 22.61 19.27 1.92 1.35 5.92 69.46C18:0 8.24 7.47 1.37 1.03 4.21 73.67C16:1x7 2.31 4.56 1.21 1.82 3.74 77.41C20:5x3 6.01 5.29 0.93 1.34 2.88 80.29C20:1x9 0.71 1.92 0.76 1.53 2.33 82.62C18:3x6 0.49 1.23 0.69 0.62 2.13 84.75C14:0 2.28 3.45 0.64 1.42 1.96 86.7C18:3x3 0.38 1.44 0.59 2.29 1.8 88.51C18:1x7 1.69 2.7 0.54 1.48 1.67 90.18


variables were linoleic acid (C18:2x6; 16.2% contribution) and oleic acid (C18:1x9; 14.7%contribution). Both of the latter two fatty acids were present in high levels in the commer-cial pellet food.

Levels of x3 were lower in farm-associated fish at both farms than in control fish, beinglowest at Guardamar (Fig. 6). Docosahexaenoic acid (C22:6x3) was the main x3 FA pres-ent in all fish, and was significantly higher (p < 0.01) in control fish. In contrast, a-linolenicacid (C18:3x3) was significantly lower (p < 0.05) in control fish (Table 5). However, therewere no significant differences in eicosapentaenoic acid (C20:5x3) between farm-associatedand control fish.

Fig. 6. Fish muscle values of the main fatty acids in the four localities and in the three kinds of food pellets.Control sites are Villajoyosa and Alicante, Farms are Campello and Guardamar. Food pellets Types I and II wereused in Campello; Type III in Guardamar. Error bars show ± SE.

Table 5Analysis of variance (ANOVAS) of the most important fatty acids, total x3 FAs, x3/x6 ratio and total fatcontent

Palmitic 16:0 Oleic 18:1x9 TOTAL n3

DF MS P MS P MS P

Af 1 55.6398 0.0642 1632.6687 0.0078 561.3609 0.0194Lo(Af) 2 3.9462 0.2675 12.9106 0.3376 11.2033 0.4064RES 36 2.8845 11.5339 12.1321C test 39 0.3685 0.7651** 0.3695

Linoleic 18:2x6 a-linolenic 18:3x3 18:3x6

MS P MS P MS P

Af 1 1413.7398 0.0104 184.2325 0.0241 41.3901 0.2314Lo(Af) 2 14.9759 0.3338 4.6053 0.3517 14.3390 0.4599RES 36 13.2388 4.2806 18.0653C test 39 0.5511* 0.5266* 0.4720

Arachidonic 20:4x6 EPA 20:5x3 DHA 22:6x3

MS P MS P MS P

Af 1 58.2873 0.0096 10.4146 0.4290 2126.8104 0.0075Lo(Af) 2 0.5671 0.7244 10.7634 0.0804 16.0832 0.1875RES 36 1.7428 3.9789 9.1676C test 39 0.7570** 0.4095 0.4540

Total fat x3/x6

MS P MS P

Af 1 372.4215 0.1005 1180.4583 0.0000Lo(Af) 2 43.9167 0.0313 0.0041 0.9995RES 36 11.4920 8.9073C test 39 0.4946 0.6166**

Af: Artificial feeding; Lo: Locality; Cochran’s C-test significances: *: significant p < 0.05. **: significant p < 0.01.All data were transformed arcsen (x + 1).


The percentage of x6 FA was higher in farm-associated fish. Nevertheless, concentra-tions of arachidonic acid (C20:4x6) were significantly (p < 0.01) higher in control fish, butlinoleic acid (C18:2x6) showed significantly higher values (p < 0.01) in farm-associated fish(Fig. 6). In the same way, x7 FAs were more elevated in farm-associated individuals(Table 3).

Markedly higher monounsaturated and x9 FAs, such as oleic acid (C18:1x9, Table 3),were recorded from tissues of farm-associated fish (p < 0.05). Control fish tissues presentedthe opposite pattern, with more total saturated FAs, although differences were notsignificant.

The three types of food pellets had similar fatty acid compositions, with high concen-trations of palmitic (C16:0), linoleic (C18:2x6, especially high in the food pellets Type III),oleic (C18:1x9), eicosapentaenoic (C20:5x3) and decosahexaenoic (C22:6x3) acids (Table1). Food pellets used to feed the fish at Campello had more x3 FA content than the fishfood at Guardamar, but had a higher content of x6 FAs (27.82%). Levels of monounsat-urated acids in farm-associated fish exceeded those found in pellets, total saturated acidsapproached those found in pellets and total polyunsaturated acids occurred in lower levelsthan in pellets. This was due to the low amount of docosahexaenoic acid (C22:6x3) in


farm-associated fish tissues, since eicosapentaenoic acid (C20:50x3) levels remained con-stantly below those of the pellets.

4. Discussion

Rearing of fish in coastal sea-cages using food pellets containing terrestrial plant prod-ucts impacts wild fish that aggregate around farms by altering their natural diet in a waythat leads to a change in body condition and FA composition. These changes make theirbody composition more similar to the cultivated fish.

Populations of wild T. mediterraneus occurred at farms throughout the year althoughwith large scale differences in abundance between times. As seen in Dempster et al.(2002), large numbers of this species were seen at eight of nine farms investigated alongthe Spanish Mediterranean coastline, appearing as an important part of the fish assem-blage at many farms.

T. mediterraneus was not seen in visual counts in winter 2005 at Guardamar. This sug-gests that the individuals sampled at Guardamar aggregated after the visual counts madein winter (March at the earliest), pointing out that they could only have been resident for amaximum of three to four months prior to our sampling in July. This was also likely tohave occurred in Campello, since populations decreased markedly at the same time. Thisperiod of time is likely to be sufficient to change the fish FA composition (Izquierdo et al.,2005).

Juvenile fish and cephalopods were the main items found in control fish stomachs, butthere was a drastic change in the feeding behaviour of T. mediterraneus around the cages.As expected, since samples were taken during feeding hours, the vast majority of fish hadpellets in their guts. As a consequence of the decrease in other natural food items, it is pos-sible to assert that T. mediterraneus strongly changes its feeding behaviour while it is asso-ciated with fish farms. Although no quantitative data is available for comparison, directfeeding of T. mediterraneus on pellets lost from farms has been observed previously(Dempster et al., 2002).

Farm-associated fish had markedly higher muscle fat content and condition than con-trol fish. Similarly, Skog et al. (2003) found that wild saithe associated with a single fjord-based farm in Norway had higher condition than control fish taken from within the samefjord. In contrast to condition and muscle fat content, Liver Somatic Index was higher incontrol T. mediterraneus than farm-associated fish. Abnormal LSI index values could becaused by ingestion of hormonally active compounds in the lost fish food beneath farms,which may be capable of modifying liver weight due to the activation of metabolisingenzymes (Sloof et al., 1983). Further research into the physiological effects on wild fishcaused by aggregation at farms should target the mechanisms driving the changes we haveobserved.

Tzikas et al. (2005), found seasonal variations in the muscle lipid content of T. mediter-

raneus off the coast of Greece; the mean lipid content during August was 0.8%, a valueconsiderably lower than the mean content found in the present study for control specimens(2.19%) and much lower than the value obtained for associated fish (6.37%). Differencesbetween the results of the two studies could be due to natural variation in the populations,although farm-associated fish showed much more lipid content than control fish from thecoast of Spain and the Greek populations. This fact could lead to substantial physiologicalchanges, since better body condition increases the spawning success of fish (Izquierdo et al.


(2001). From a previous study carried out at nine farms along the SE coast of Spain(including those in this paper) it is known that 85% of farm-associated fish are of adultsize (Dempster et al., 2002). Therefore, we hypothesise that the better condition theseadults gain while associated with farms enables greater production of eggs and sperm,which translates to greater spawning success. T. mediterraneus is known to spawn in sum-mer and autumn in the Mediterranean (Ragonese et al., 2004) precisely at this time whenwe observed their condition to be enhanced by farms. Further research is required to testthe quality of eggs and larvae. This important source of spawners of adult size could func-tion as both spawning product and biomass exporters in the same way as the numerouslarge adult fish protected within Marine Protected Areas (Gell and Roberts, 2003;Garcıa-Charton et al., 2004). Subsequently, our results add weight to the managementconcept of protecting wild fish assemblages around fish farms, when they are possiblymore susceptible to capture due to their aggregating behaviour. Furthermore, the fact thatwild fish help to remove up to 80% of wastes that a farm produces (Vita et al., 2004) sup-ports the idea of protecting the farm-associated fauna following the dual objectives ofenhancing local fishing and reducing the environmental impacts caused by farm wastes(Dempster et al., 2005).

The high variation in fat contents in farm-associated fish could be due to differing res-idence times of individuals around the farms before capture or even due to feeding on dif-ferent kinds of food, natural and pellets, during short migrations. Different individualslikely consumed pellets over different periods of time; less fatty natural food for longerperiods prior to sampling in some fish was probably responsible for the variability weobserved (e.g. Deudero and Morales-Nin, 2001). Fat levels in control fish were relativelystable among individuals, suggesting they had similar dietary profiles rich in low-fat nat-ural items. The total lipid amount in cultivated fish is usually higher than that of wild fish,a feature common in studies that compare reared and wild fish of the same species (Chan-mugam et al., 1986; Nettleton et al., 2000; Saglik et al., 2003). However, little is knownabout the effect on wild fishes that aggregate around fish farms and feed on lost pellets.Fish aggregate around sea-cages mainly due to the availability of food (Tuya et al.,2006). We have demonstrated that if wild fish that occur beneath farms change their nat-ural feeding patterns to consume food pellets, a modification in their physiology willoccur.

Ackman and Takeuchi (1986) reported that the percentage of x3 PUFA in farmed mar-ine fish is usually lower than in their wild relatives, presumably because of the lack of lipidsoriginating from algae and marine phytoplankton (Nichols et al., 1989; Muje et al., 1989).In this study, we have demonstrated that this also happens in wild T. mediterraneus thataggregate and feed around fish farms. Skog et al. (2003) found that wild saithe (Pollachiusvirens) feeding around a salmon farm in a Norwegian fjord had similar fatty acid profilesto the food pellets used at the farm, with increased levels of oleic and a-linolenic acids aswell as a comparatively low ratio of x3/x6 which reflected that in pellets. They alsodetected an increase in palmitic acid in the flesh of fish from reference areas, as we detectedwith T. mediterraneus, pointing out de novo synthesis of these acids (Henderson and Sar-gent, 1985).

The increased levels of x6 acids in farm-associated fish may have been due to the lin-oleic acid (C18:2x6) present in pellets. It seems that unsaturated acids are to some degreeregulated by the fish, leading to a change in the FAs of farm-associated fish with respect topellet composition, especially in linoleic acid (e.g. Linko et al., 1992).


The clear differences in FA composition were mainly due to docosahexaenoic acid(C22:6x3), because of the low values for farm-associated fish, and linoleic acid(C18:2x6), which was found in higher levels than in the control T. mediterraneus becauseof their presence in the commercial food pellets. The several-fold decrease in docosahexa-enoic acid content in farm-associated fish may be due to their changed feeding behaviour,since natural food is rich in x3 FAs (e.g. Shirai et al., 2002).

The MDS analysis (Fig. 5) showed that the FA composition of T. mediterraneus aggre-gated around the Guardamar farm was more variable than that of the fish aggregated atthe Campello farm. This could be related to the increased abundances around the farmprior to the capture of samples in summer 2005, indicating the recent arrival of new indi-viduals at the farm; the differing influence of artificial (pellets) and natural food itemswould explain the different muscle FA composition. The visual count data suggest thatthe observed physiological changes in wild T. mediterraneus condition, body fat contentand FA composition occurred over a maximum time scale of three to four months.

The x3/x6 ratio is generally lower in cultivated than in wild fish (Van Vliet and Katan,1990). Despite the higher relative quantity of x6 in farm-associated fish, the total contentof x3 was also higher because of their elevated fat contents. However, the relationshipbetween x3/x6 was significantly lower. Rueda et al. (2001), nevertheless, found higher lev-els of x3 FAs in cultivated Diplodus puntazzo, an omnivorous species, with a slightly non-significant increment of x3 FAs in their wild counterparts, however, linoleic acid(C18:2x6) was much higher in cultivated fish.

The fatty acids 20:1x9 and 22:1x11 have been found to be possible biomarkers of cod-farms (Gadus morhua) since they are abundant in the feed and are not well digested by fishso they can be detected in the dispersed wastes (Van Biesen and Parris, 2005). However,22:1x11 was not a component in the pellets used to feed sea bass and sea bream in thefarms tested here. We suggest that it may also be possible to detect the influence that farmshave on wild fish in an area by analyzing the FA profile of the local ichthyofauna (e.g.Skog et al., 2003). FAs are generally taken in by tissues, and it is possible to distinguishbetween dietary and non-dietary components. Some fatty acids are deposited in adiposetissue with little modification and in a predictable way (Iverson et al., 2004). The specificFA patterns are passed from prey to predator near the bottom of the food web (Sargentet al., 1988; Fraser et al., 1989; Graeve et al., 1994: Navarro et al., 1995; St. John andLund, 1996; Kirsch et al., 1998), determining the FA composition of higher predator levels(Hooper et al., 1973) and indicating the presence of specific prey in predator diets (Colbyet al., 1993; Pond et al., 1995; Raclot et al., 1998). Tracking of dietary componentsthrough the food web cannot be entirely achieved using other methods such as stomachcontents, which are modified by digestion, or stable isotopes, which are useful in estimat-ing the trophic level of a predator but cannot determine the species composition of the diet(Hobson, 1993; Gilmore et al., 1995; Koch et al., 1995).

FA could, therefore, be used as biomarkers in the study of the structure and dynamicsof fish food webs around fish farms, as alternatives to direct or indirect methods that pro-vide only a snapshot of the most recent meal and may not be representative of the longerterm diet. Their qualitative use has inferred trophic levels and spatial and temporal differ-ences in diets both within and among species (Kakela et al., 1993; Smith et al., 1996: Smithet al., 1997; Iverson et al., 1997a,b). Similar methods could provide quantitative estimatesof predator diets in fish assemblages aggregated around cages. To do this requires under-standing the FA dynamics of each species. This study is the first contribution in that direc-


tion to the knowledge of trophodynamics of fish assemblages around cages in the Medi-terranean Sea.

To develop the concept of using fatty acids as biomarkers, it is necessary to undertakelaboratory and field experiments to measure the accumulation times of FAs in tissue. Fur-ther, the residence times of the aggregated species around cages and the magnitude of theirmigrations is important information to determine the extent to which pellet-originatedFAs are transported over a broader scale than the immediate vicinity of the cages. Sucha method would be a key tool to infer the scale and magnitude of the influence of anet-cage fish farm on the local fauna, not only on fish that directly feed on lost food pelletsbut on the whole food web.

5. Conclusion

T. mediterraneus aggregated around sea-cage fish farms underwent ecological and phys-iological changes. Their diet differed from control fish, which affected both their body con-dition and fatty acid composition. Differences in residence times at farms or shortmigrations from individual fish may explain the large differences in fat content amongthe aggregated fish. The x3/x6 ratio and 22:6x3 level were significantly lower in farm-associated fish, reflecting the food pellet composition, while control fish showed signifi-cantly lower levels of x6 fatty acids. Increased levels of 18:2x6 and 18:1x9 and decreasedlevels of 22:6x3 are promising candidates for biomarkers to study the influence of fishfarms on the local food web.

Acknowledgements

We thank Carlos Valle, Juan Antonio Martinez and Almudena Haro for assistanceduring sampling, and the staff of the Cudomar (El Campello) and Martorres (Guardamar)fish farms for their help. The study forms part of the ACUFISH project (Unidad de Bio-logıa Marina, University of Alicante) and was funded by Spanish Ministry of Sciencegrant MYCT-REN2003-00794.

References

Ackman, R.G., 1967. Characteristic of the fatty acid composition and biochemistry of some fresh-water fish oilsand lipids in comparison with marine oils and lipids. Comparative Biochemistry and Physiology 22, 907–922.

Ackman, R.G., Takeuchi, T., 1986. Comparison of fatty acids and lipids of smelting hatchery-fed and wildAtlantic salmon (Salmo salar). Lipids 21, 117–120.

Anderson, M.J., 2004. PERMANOVA_2factor: a FORTRAN computer program for permutational multivariateanalysis of variance (for any two-factor ANOVA design) using permutation tests. Department of Statistics,University of Auckland, New Zealand.

Bandarra, N.M., Batista, I., Nunes, M.L., Empis, J.M., 2001. Seasonal variation in the chemical composition ofhorse-mackerel (Trachurus trachurus). European Food Research and Technology 212, 535–539.

Bayle-Sempere, J.T., Valle, C., Verdu, A., 2002. EcoCEN: application for managing fish visual counts. Informes yestudios. COPEMED, no. 7.

Bjordal, A., Skar, A.B., 1992. Tagging of saithe (Pollachius virens L.) at a Norwegian fish farm: preliminaryresults on migration. ICES Council Meeting Papers, 1992/G:35.

Carss, D.N., 1990. Concentrations of wild and escaped fishes immediately adjacent to fish farm cages.Aquaculture 90, 29–40.


Chanmugam, P., Boudreau, M., Hwang, D.H., 1986. Differences in the x3 fatty acid contents in pond-reared andwild fish and shellfish. Journal Food Science 51, 1556–1557.

Clarke, K.R., 1993. Non-parametric multivariate analyses of changes in community structure. American Journalof Ecology 18, 117–143.

Clarke, K.R., Warwick, R.M., 1994. Change in marine communities: an approach to statistical analysis andinterpretation. Natural Environmental Research Council, Plymouth.

Colby, R.H., Mattacks, C.A., Pond, C.M., 1993. The gross anatomy, cellular structure and fattyacid composition of adipose tissue in captive polar bears (Ursus maritimus). Zoo Biology 12,267–275.

Dempster, T., Sanchez-Jerez, P., Bayle-Sempere, J.T., Gimenez-Casalduero, F., Valle, C., 2002. Attraction ofwild fish to sea-cage fish farms in the south-western Mediterranean Sea: spatial and short-term temporalvariability. Marine Ecology Progress Series 242, 237–252.

Dempster, T., Fernandez-Jover, D., Sanchez-Jerez, P., Tuya, F., Bayle-Sempere, J.T., Boyra, A., Haroun, R.J.,2005. Vertical variability of wild fish assemblages around sea-cage fish farms: implications for management.Marine Ecology Progress Series 304, 15–29.

Deudero, S., Morales-Nin, B., 2001. Prey selectivity in planktivorous juvenile fishes associated with floatingobjects in the western Mediterranean. Aquaculture Research 32 (6), 481.

FAO (Food and Agriculture Organisations). 2003 Fishstat Plus. Aquaculture production: quantities 1950-2002.FAO, Rome; also available at www.fao.org/fi/statist/FISOFT/FISHPLUSH.asp.

Folch, J., Lees, M., Stanley, G.A., 1957. A simple method for the isolation and purification of total lipids fromanimal tissues. Journal of Biological Chemistry 226, 497–509.

Fraser, A.J., Sargent, J.R., Gamble, J.C., Seaton, D.D., 1989. Formation and transfer of fatty acids in anenclosed marine food chain comprising phytoplankton, zooplankton and herring (Clupea harengus L.) larvae.Marine Chemistry 27, 1–18.

Garcıa-Charton, J.A., Perez-Ruzafa, A., Sanchez-Jerez, P., Bayle-Sempere, J.T., Renones, O., Moreno, D., 2004.Multi-scale spatial heterogeneity, habitat structure, and the effect of marine reserves on WesternMediterranean rocky ref. fish assemblages. Marine Biology 144, 161–182.

Gell, F.R., Roberts, M., 2003. Benefits beyond boundaries: the fishery effects of marine reserves. Trends inEcology and Evolution 18 (9), 448–455.

Gilmore, I., Johnston, M.A., Pillinger, C.T., Pond, C.M., Mattacks, C.A., Prestrud, P., 1995. The carbon isotopiccomposition of individual fatty acids as indicators of dietary history in arctic foxes on Svalbard. PhilosophicalTransactions of the Royal Society of London B 349, 135–142.

Graeve, M., Kattner, G., Hagen, W., 1994. Diet-induced changes in the fatty acid composition of Arcticherbivorous copepods: experimental evidence of trophic markers. Journal of Experimental Marine Biologyand Ecology 182, 97–110.

Henderson, R.J., Sargent, J.R., 1985. Fatty acids metabolism in fish. In: Cowey, C.B., Mackie, A.M., Bell, J.G.(Eds.), Nutrition and Feeding in fish. Academic press, London, UK, p. 253.

Hobson, K.A., 1993. Trophic relationships among high Arctic seabirds: insights from tissue-dependent stable-isotope models. Marine Ecology Progress Series 95, 7–18.

Hooper, S.N., Paradis, M., Ackman, R.G., 1973. Distribution of trans-6-hexadecanoic acid, 7-methyl-7-hexadecanoic acid and common fatty acids in lipids of the ocean sunfish Mola mola. Lipids 8, 509–516.

Iverson, S.J., Arnould, J.P.Y., Boyd, I.L., 1997a. Milk fatty acid signatures indicate both major and minor shiftsin diet of lactating Antarctic fur seals. Canadian Journal of Zoology 75, 188–197.

Iverson, S.J., Frost, K.J., Lowry, L.L., 1997b. Fatty acid signatures reveal fine scale structure of foragingdistribution of harbor seals and their prey in PrinceWilliam Sound, Alaska. Marine Ecology Progress Series151, 255–271.

Iverson, S.J., Field, Ch., Bowen, W.D., Blanchard, W., 2004. Quantitative fatty acid signature analysis: a newmethod of estimating predator diets. Ecological Monographs 74 (2), 211–235.

Izquierdo, M.S., Fernandez-Palacios, H., Tacon, A.G.C., 2001. Effect of broodstock nutrition on reproductiveperformance of fish. Aquaculture 197, 25–42.

Izquierdo, M.S., Montero, D., Robaina, L., Caballero, M.J., Rosenlund, G., Gines, R., 2005. Alterations in filletfatty acid profile and flesh quality in gilthead seabream (Sparus aurata) fed vegetables oils for a long termperiod. Recovery of fatty acid profiles by fish oil feeding. Aquaculture 250, 431–444.

Kakela, R., Hyvarinen, H., Vainiotalo, P., 1993. Fatty acid composition in liver and blubber of the Saimaa ringedseal (Phoca hispida saimensis) compared to that of the ringed seal (Phoca hispida botanica) and grey seal(Halichoerus grypus) from the Baltic. Comparative Biochemistry and Physiology part B 105, 553–565.

http://www.fao.org/fi/statist/FISOFT/FISHPLUSH.asp


Kingsford, M., Battershill, C., 1998. Studying Marine Temperate Environments: A Handbook for Ecologists.Canterbury University Press, Christchurch, New Zealand.

Kirsch, P.E., Iverson, S.J., Bowen, W.D., Kerr, S., Ackman, R.G., 1998. Dietary effects on the fatty acidsignatures of whole Atlantic cod (Gadus morhua). Canadian Journal of Fisheries and Aquatic Sciences 55,1378–1386.

Koch, P.L., Heisinger, J., Moss, C., Carlson, R.W., Fogel, M.L., Behrensmeyer, A.K., 1995. Isotopic tracking ofchange in diet and habitat use in African elephants. Science 267, 1340–1343.

Linko, R.R., Rajasilta, M., Hiltunene, R., 1992. Comparison of lipid and fatty acid composition in vendace(Coregonus albula L.) and available plankton feed. Comparative Biochemistry and Physiology A 103 (1), 205–212.

Muje, P., Agren, J.J., Lindqvist, A.V., Hanninen, O., 1989. Fatty acid composition of Vendace (Coregonus albula

L.) muscle and its plankton feed. Comparative Biochemistry and Physiology B 92 (1), 75–79.Navarro, J.C., McEvoy, L.A., Amat, F., Sargent, J.R., 1995. Effects of diet on fatty acid composition of body

zones in larvae of the sea bass Dicentrarchus labrax: a chemometric study. Marine Biology 124, 177–183.Nettleton, J.A., 2000. Fatty acids in cultivated and wild fish. IIFET (International Institute of Fisheries

Economics and Trade), Microbehavior and Macroresults, Oregon State University (USA) 2000, Oralpresentation.

Nichols, P.D., Holdsworth, D.G., Volkman, J.K., Daintith, M., Allanson, S., 1989. High incorporation ofessential fatty acids by the rotifer Brachionus plicatilis feed on the prymesiophyte alga Pavlova lutheri.Australian Journal of Marine and Freshwater Research 40, 645–655.

Pond, C.M., Mattacks, C.A., Gilmore, I., Johnson, M.A., Pillinger, C.T., 1995. Chemical and carbon isotopiccomposition of fatty acids in adipose tissue as indicators of dietary history in wild foxes (Alopex lagopus) onSvalbard. Journal of Zoology 236, 611–623.

Raclot, T., Groscolas, R., Cherel, Y., 1998. Fatty acid evidence for the importance of myctophid fishes in the dietof king penguins, Aptenodytes patagonicus. Marine Biology 132, 523–533.

Ragonese, S., Fiorentino, F., Garofalo, G., Gristina, M., Levi, D., Gancitano, S., Giusto, G.B., Rizzo, G.,Sinacori, G., 2004. MedSudMed Technical Documents, FAO. Available from http://www.faomedsudmed.org.

Rueda, F.M., Hernandez, M.D., Egea, M.A., Aguado, F., Garcıa, B., Martınez, F.J., 2001. Differences in tissuefatty acid composition between reared and wild sharpsnout sea bream, Diplodus puntazzo (Cetti, 1777). BritishJournal of Nutrition 86, 617–622.

Saglik, S., Alpaslan, M., Gezgin, T., Cetinturk, K., Tekinay, A., Cemal-Guven, K., 2003. Fatty acid compositionof wild and cultivated gilthead sea bream (Sparus aurata) and sea bass (Dicentrarchus labrax). EuropeanJournal of Lipid Science and Technology 105, 104–107.

Sargent, J.R., Parkes, R.J., Mueller-Harvey, I., Henderson, R.J., 1988. Lipid biomarkers in marine ecology. In:Sleigh, M.A. (Ed.), Microbes in the Sea. Ellis Horwood, Chichester, UK, pp. 119–138.

Shirai, N., Terayama, M., Takeda, H., 2002. Effect of season on the fatty acid composition and free amino acidcontent of the sardine Sardinops melanostictus. Comparative Biochemistry and Physiology B 131, 387–393.

Skog, T.E., Hylland, K., Torstensen, B.E., Berntssen, M.H.G., 2003. Salmon farming affects the fatty acidcomposition and taste of wild saithe Pollachius virens L. Aquaculture Research 34, 999–1007.

Sloof, W., Van Kreijl, C.F., Baars, A.J., 1983. Relative liver weights and xenobiotic-metabolizing enzymes of fishfrom polluted surface waters in the Netherlands. Aquatic Toxicology 4, 1–14.

Smith, R.J., Hobson, K.A., Koopman, H.N., Lavigne, D.M., 1996. Distinguishing between populations of fresh-and salt-water harbour seals (Phoca vitulina) using stable isotope ratios and fatty acid profiles. CanadianJournal of Fisheries and Aquatic Sciences 53, 272–279.

Smith, S., Iverson, S.J., Bowen, W.D., 1997. Fatty acid signatures and classification trees: new tools forinvestigating the foraging ecology of seals. Canadian Journal of Fisheries and Aquatic Sciences 54, 1377–1386.

St. John, M.A., Lund, T., 1996. Lipid biomarkers: linking the utilization of frontal plankton biomass to enhancedcondition of juvenile North Sea cod. Marine Ecology Progress Series 131, 75–85.

Stoffel, W., Chu, F., Edward, H., 1959. Analysis of long-chain fatty acids by gas–liquid chromatography.Micromethod for preparation of methyl esters. Analytical Chemistry 31, 307–308.

Tuya, F., Sanchez-Jerez, P., Dempster T., Boyra, A., Haroun, R., 2006. Changes in demersal wild fishaggregations beneath a sea-cage fish farm after the cessation of farming. Journal of Fish Biology, in press.

Tzikas, Z., Amvrosiadis, I., Soultos, N., Georgakis, Sp., 2005 Seasonal variation in the chemical composition andmicrobiological condition of Mediterranean horse mackerel (Trachurus mediterraneus) muscle from the NorthAegean Sea (Greece). Food Control, in press.

http://www.faomedsudmed.org

http://www.faomedsudmed.org


Underwood, A.J., 1997. Experiments in Ecology. Their Logical Design and Interpretation. Cambridge UniversityPress, UK.

Van Biesen, G., Parris, C.C., 2005. Long-chain monounsaturated fatty acids as biomarkers for the dispersal oforganic waste from a fish enclosure. Marine Environmental Research 60, 375–388.

Van Vliet, T., Katan, M.B., 1990. Lower ratio of x3 to x6 fatty acids in cultured than wild fish. American Journalof Clinical Nutrition 51, 1–2.

Vita, R., Marın, A., Madrid, J.A., Jimenez-Brinquis, B., Cesar, A., Marın-Guirao, L., 2004. Effects of wild fisheson waste exportation from a Mediterranean fish farm. Marine Ecology Progress Series 277, 253–261.

Warwick, R.M., Platt, H.M., Clarke, K.R., Agard, J., Gobin, J., 1990. Analysis of macrobenthic and meiobenthiccommunity structure in relationship to pollution and disturbance in Hamilton Harbour, Bermuda. Journal ofExperimental Marine Biology and Ecology 138, 119–142.