Sparus aurata

Aquaculture 464 (2016) 111–116

Contents lists available at ScienceDirect

Aquaculture journal homepage: www.elsevier.com/locate/aquaculture

Controlling feed losses by chewing in gilthead sea bream (Sparus aurata) ongrowing may improve the fish farming environmental sustainability M. Ballester-Moltó a,⁎, P. Sanchez-Jerez b, B. García-García a, J. García-García a, J. Cerezo-Valverde a, F. Aguado-Giménez a a b

Instituto Murciano de Investigación y Desarrollo Agrario y Alimentario (IMIDA), Estación de Acuicultura Marina, Puerto de San Pedro del Pinatar, 30740, Murcia, Spain Departamento de Ciencias del Mar y Biología Aplicada, Universidad de Alicante, PO Box 99, 03080 Alicante, Spain

a r t i c l e

i n f o

Article history: Received 10 December 2015 Received in revised form 8 March 2016 Accepted 13 June 2016 Available online 16 June 2016 Keywords: Aquaculture Chewing Environmental sustainability Feed losses Feeding behaviour Sparus aurata

a b s t r a c t Gilthead sea bream (Sparus aurata) usually chew their feed before swallowing it. Under rearing condition, this feeding behaviour generates significant feed waste in the form of pellet fragments. The study aimed to experimentally quantify feed losses by chewing considering fish weight and the feed pellet size. Gilthead sea bream with a body weight of 28–1019 g were fed with differently sized pellets (2–8 mm) from a commercial aquafeed gamma. Feed wastes were collected and data were used to create a model able to estimate the waste derived from this particular feeding behaviour. The results pointed to a substantial feed loss, which increased proportionally with fish size as the pellets are larger. Simulations for a complete ongrowing cycle carried out following the aquafeed manufacturer's recommendations with regard to pellet size revealed that feed wastes by chewing represent 8% of the feed delivered. Alternative feeding regimes based on the use of smaller pellets would lead to a significant reduction in losses (up to 50%) through chewing. Improving the feeding strategy would help to minimize waste output, increasing gilthead sea bream aquaculture sustainability. Statement of relevance: This study deals with the wastes generated by chewing in the ongrowing of gilthead seabream of different body weight fed with different pellet sizes. This is the only work modelling the losses by chewing phenomenon, also suggesting alternative feeding regimes to improve environmental sustainability of the aquaculture process. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Gilthead sea bream (Sparus aurata), GHSB hereafter, is one of the main marine cultured fish in the Mediterranean Sea, with a yield in 2013 of close to 180,000 t (APROMAR, 2014). Between the beginning of the industrial rearing of GHSB in the late 1970′s (Bernabé, 1991) and now, considerable improvements have been achieved with regards to its husbandry, stock management, feeding and nutrition, leading to its becoming the most consolidated fish specie in Mediterranean aquaculture. In the last decade, research into this fish species has mainly focused on immunology (Tort et al., 2010), metabolomics (Picone et al., 2011), pathology (Colorni and Padrós, 2011) and nutrition (Teles et al., 2011). Nevertheless, despite our ample knowledge of the biology of this species and rearing techniques, there is scope for improvement regarding its husbandry, especially in the feeding process. GHSB, with its specialized chewing apparatus, has a particular way of feeding, which involves a high degree of food manipulation ⁎ Corresponding author. E-mail addresses: [email protected], [email protected] (M. Ballester-Moltó).

http://dx.doi.org/10.1016/j.aquaculture.2016.06.018 0044-8486/© 2016 Elsevier B.V. All rights reserved.

(Andrew et al., 2003). Other sparid species with well-developed molar teeth also exhibit this behaviour in the wild, e.g., white sea bream (Diplodus sargus; Vandewalle et al., 1995), red porgy (Pagrus pagrus; Castriota et al., 2006) or red-banded sea bream (Pagrus auriga; Chakroun-Marzouk and Kartas, 1987). While feeding under aquaculture conditions, GHSB processes the pellets by opening and closing its mouth repeatedly, as if savouring. Usually, GHSB crushes the pellets before swallowing or swallows them as a whole, depending on its size and appetite. Sometime the whole pellets or fragments are discarded, so that they can be re-consumed by the same or other individuals. These actions generate a variable degree of food loss and waste (Andrew et al., 2003, 2004a), which may affect the utilization efficiency of the feed. Industrial GHSB ongrowing is carried out using extruded pellets of different sizes. Fish feed producers recommend the most appropriate pellet size according to fish size. For some fish species it has been determined that optimum pellet size ranges from 25 to 50% of the fish mouth width (Knights, 1985; Linnér and Brännäs, 1994; Smith et al., 1995; Tabachek, 1988; Wankowski and Thorpe, 1979). However, many fish species are able to take in a wide range of prey sizes (Dörner et al., 2007; Dörner and Wagner, 2003; Scharf et al., 2000; Specziár, 2011). In captivity, GHSB become accustomed to a certain pellet size and may

112

M. Ballester-Moltó et al. / Aquaculture 464 (2016) 111–116

refuse to accept significantly smaller pellets even when hungry (Ballester-Moltó, pers. observations), which underlines how previous experience significantly affects feeding behaviour (Bryan, 1973; Clarke and Sutterlin, 1985; Cox and Pankhurst, 2000). Particulate organic wastes from fish farming, normally in the form of uneaten (whole or fragmented) food pellets and faeces, tend to settle in the vicinity of the farms. This is the main cause of any environmental effects on the seabed derived from the activity (Bureau and Hua, 2010; Buryniuk et al., 2006; Holmer et al., 2007; Islam, 2005). While quantification of waste production is an essential task for an accurate environmental impact assessment (Bureau and Hua, 2010; Fernandes et al., 2001; Xu et al., 2007), quantifying the portion of residues corresponding to uneaten food is complicated since it largely depends on the care taken when feeding. This has not been quantified accurately in GHSB, but some producers intuitively estimate it at around 5–10% of the supplied food in open-sea conditions (Piedecausa et al., 2009). Considering GHSB feeding behaviour, fragmented pellets lost by chewing may represent a considerable source of pollution. In this work we aimed to quantify experimentally the food losses resulting from chewing in a wide range of GHSB sizes fed with commercially available feed pellets of different sizes. Through simulations, we also compare the losses by chewing that may occur when following the feeding indications provided by the manufacturer (with regard to the optimum pellet size according to fish size) and the losses which might occur if alternative feeding regimes based on the management of pellet sizes are followed. 2. Material and methods 2.1. Experimental conditions and procedures Experiments were carried out at the Marine Aquaculture Station belonging to the IMIDA, in San Pedro del Pinatar, Murcia (SE Spain). Fish were stocked in 600 l troncoconical tanks with a purge system to collect uneaten feed pellets and pellet fragments after settling. Tanks were connected to a recirculating seawater system with biological and mechanical filtration. Approximately 20% of the water volume was renewed daily. Water temperature was maintained at 23 ± 1 °C with a heat exchanger, and oxygenation was always above 90% saturation. A 12:12 light/dark photoperiod was followed. Six different GHSB size groups (Table 1) and four extruded pellet sizes (2, 4, 6 and 8 mm; Table 2) were used in the experiments. In order to prevent possible effects derived from palatability and/or physical features, and to resemble real rearing conditions, fish were fed with commercially available aquafeed commonly used in GHSB ongrowing in the Mediterranean. The raw materials used in the feed were fishmeal, fish and vegetables oils, cereal legumes and oilseeds products and byproducts, vitamins and minerals. Each GHSB size group was fed with different pellet sizes. Pellet size acceptability was evaluated previously for each fish size: XS, S and M fish did not accept 8 mm pellets, and XXL fish did not accept 2 mm pellets. Only L and XL fish accepted all pellet sizes (see Table 1 for explanation of sizes). Fish were fasted two days before the beginning of the trials. Each of the six trials was three times replicated so fish of each size were distributed in three batches (Table 1 shows the number of fish per batch in each trial). Every batch of each trial was fed for four consecutive days

Table 2 Macronutrient composition of aquafeed used in the experiments, as declared by the manufacturer. Pellet size (mm)

Crude protein (%) Crude lipid (%) Ash (%) Cellulose (%) Total phosphorus (%) Digestible energy (MJ/kg)

2

4

6

8

48.5 18.0 6.4 2.7 1.0 18.5

46.0 19.0 5.9 4.5 1.0 18.0

44.0 20.0 6.2 4.5 0.9 18.0

41.5 20.0 6.1 4.5 0.9 17.4

and the feed lost by chewing (as pellet fragments) from these four days was averaged. The experimental procedure includes that each day tanks were purged just before feeding to start the trials with clean end-cones. Fish were fed until apparent satiety once a day at 9:30 a.m. avoiding overfeeding: feeding was stopped right when loss of appetite was perceived. The feed remains were collected from the settling cone using a 40 μm mesh (the smallest waste particles derived from fish farming range between 31 and 63 μm; Gowen and Bradbury, 1987) 1 h after feeding, reflecting GHSB waste sinking rates (Magill et al., 2006; Piedecausa et al., 2010b). Faeces and whole uneaten pellets were easily separated with forceps from the mass of pellet fragments. Faeces were discarded, and then whole uneaten pellets and fragments were dried in an oven (105 ± 1 °C, 24 h) until constant weight and weighed separately to ascertain the food ingested and the losses by chewing. Previously, both fragments and uneaten pellets were rinsed with distilled water to remove sea salt. The following corrections were made. A blank tank without fish was sampled in the same way every day to ascertain background particulate matter levels entering the experimental tanks. Also, whole pellet stability was assessed by submerging 1 g of each pellets size (previously dried in an oven (105 ± 1 °C, 24 h) until constant weight) in 50 ml sea water with aeration for 1 h. After this, the pellets were removed and dried again in an oven (105 ± 1 °C, 24 h) until constant weight and weighed; the water was filtered (Whatman filter, 45 μm pore size, previously weighed) and, the filters were also dried in an oven (105 ± 1 °C, 24 h) until constant weight and weighed. Then, the disaggregation and leaching of pellets after 1 h were calculated as weight loss from the initial pellet sample. Loss by chewing was calculated as a percentage of the supplied food, and corrected by subtracting background level and full pellets disaggregation and adding leaching. 2.2. Statistical analyses 2.2.1. Loss by chewing model Losses by chewing (LbC) data were fitted by mean of backward stepwise multiple linear regression analysis to an equation as follows: LbC ð%Þ ¼ a þ b Fw þ c Ps þ d Fw Ps ;

ð1Þ

where Fw is the fish weight in g, Ps is the pellet size in mm, a is the intercept in the model, and b, c and d are the respective independent variable coefficients provided by the regression analysis. Backward stepwise

Table 1 GHSB stocking conditions and tested pellet sizes in chewing experiments (Fw: fish weight; Ps: pellet size). GHSB size

Number of fish per batch

Fw (g)

Stocking density (kg m−3)

Ps (mm)

Very small (XS) Small (S) Medium (M) Large (L) Very large (XL) Extra large (XXL)

27 20 17 16 10 4

27.59 ± 0.44 70.82 ± 0.27 151.92 ± 1.11 343.20 ± 27.48 617.31 ± 17.05 1025.03 ± 35.17

1.24 ± 0.02 2.36 ± 0.03 4.30 ± 0.01 9.15 ± 0.10 10.28 ± 0.20 8.35 ± 0.02

2, 4, 6 2, 4, 6 2, 4, 6 2, 4, 6, 8 2, 4, 6, 8 4, 6, 8


procedure removes elements (one per step; P N 0.05) from the equation until the best fit is reached. 2.2.2. Simulations Several simulations of LbC were made using the best fitting equation obtained as mentioned above. The LbC model was run for the ongrowing of GHSB from 20 g initial weight to 500 g final weight (above which sexual maturation leads to bad estimates for growth in GHSB, so that LbC estimates would not be accurate). Simulations were performed considering the most favourable environmental and husbandry conditions for the ongrowing of GHSB in the Mediterranean (Piedecausa et al., 2010a). For growth estimates, the model described by Mayer et al. (2012) was selected because it was validated for GHSB ongrowing under intensive culture conditions. These authors proposed two equations for fish growth, below (Eq. (2a)) and above (Eq. (2b)) 170 g:

Table 3 Feeding regimes for simulations based on the manufacturer recommendations and three alternatives in which pellet sizes differs for different fish weights. Pellet Size (mm)

Manufacturer Alternative 1 Alternative 2 Alternative 3

2

4

6

8

20–70 g 20–120 g 20–120 g 20–200 g

71–220 g 121–300 g 121–500 g 201–500 g

221–500 g 301–500 g –

N500 g – –

the supplied feed would be the RIF plus the LbC estimated from the LbC model for each alternative. 3. Results 3.1. Loss by chewing model

3 W f ðtÞ ¼ W0 1=3 þ TGC1 STðt0 ; tÞ ;

ð2aÞ

3 W f ðtÞ ¼ W0 2=3 þ TGC2 STðt0 ; tÞ ;

ð2bÞ

where Wf and W0 are the final and initial weights in g, respectively; ST(t0,t) is the effective cumulative temperature minus 12 in °C (below 12 °C GHSB growth is zero; Mayer et al., 2008, 2009); t0 and t are the initial time and the time at a given moment, respectively; TGC1 and TGC2 are the thermal growth coefficients (0.00164561 and 0.0160949, respectively). A feeding model was developed from the feeding schedule (Specific Feeding Rates: SFR) provided by the feed manufacturer (GHSB weight range: 1–500 g; temperature range: 14–28 g), and the best fitting equation was: lnSFR ð%Þ ¼ −4594–0; 289 ln ð Fw Þ þ 2034 ln ðTÞ;

113

ð3Þ

where Fw is the fish weight in g, T is the water temperature in °C (Adj R2: 0.988). From this equation we calculate the really ingested food (RIF) by subtracting the LbC estimated from the LbC model for the manufacturer conditions to the supplied feed. Fig. 1 illustrates the simulation conditions for temperature and growth and food intake development. Simulations were carried under different feeding regimes, considering the manufacturer's recommendations and three alternative feeding regimes (Table 3). For the alternative simulations to be comparable among themselves and versus the manufacturer recommendations,

Unlike fish size, LbC values below 1% of the supplied food were observed when fish were fed 2 mm pellets. GHSB below 150 g fed with 4 or 6 mm pellets produced LbC values of 1–4% and 1–6% of the supplied food, respectively. Above 150 g and also fed with 4 or 6 mm pellets, the observed LbC values were between 4-15% or 6–17% of the supplied food, respectively. The observed LbC values represented 21–31% of the supplied food in GHSB larger than 350 g fed 8 mm pellet (Table 4). Backward stepwise regression excluded the term Fw (P N 0.05) from the model and maintained the remaining terms (Table 5). The best fitting equation included Ps (P b 0.01) and its interaction with Fw (Ps × Fw; P b 0.001). The resulting equation was: LbC ð%Þ ¼ −3:9074 þ 1:3869 Ps þ 0:0029 Fw Ps ;

The significance of the interaction term (Ps × Fw) means that LbC increases as the pellet size increases, particularly (i.e. the slope is more pronounced) as the fish weight increases. The larger the pellet, the greater the LbC unlike fish size, but not always the opposite. Fig. 2 illustrates this interaction effect. The model always estimates some LbC except for fish weighing b 195 g fed with pellets smaller than 2.66 mm. When the model provides negative values, it is assumed that LbC is zero. LbC can reach values as high as 30% of the supplied feed in the largest fish fed with the largest pellets.

Table 4 Average observed values (± se: standard error) of Losses by Chewing (LbC) for XS, S, M, L, XL and XXL GHSB fed with different pellet sizes. GHSB size

Pellet size (mm)

LbC ± se (%)

XS

2 4 6 2 4 6 2 4 6 2 4 6 8 2 4 6 8 4 6 8

0,96 ± 0,12 2,52 ± 0,28 4,03 ± 0,29 0,35 ± 0,12 1,03 ± 0,15 1,36 ± 0,19 0,59 ± 1,09 4,21 ± 0,19 5,47 ± 0,38 0,98 ± 1,09 8,13 ± 0,71 7,53 ± 0,30 20,51 ± 0,32 0,29 ± 1,06 13,10 ± 0,32 12,81 ± 0,65 27,55 ± 1,39 15,09 ± 2,45 17,32 ± 2,31 31,48 ± 2,44

S

M

L

XL

Fig. 1. Simulation of the development of GHSB from 20 to 500 g from Eqs. (2a) and (2b) and (3). SGR: Specific Grow Rate; SFR: Specific Feeding Rate and monthly average temperature.

ð4Þ

XXL

114


Table 5 Results of the backward stepwise regression analysis for LbC. **P b 0.01, ***P b 0.001). Steps

Effect

Effect status

d.f.

F

P

Adj. R2

Step 1

Ps Fw Ps × Fw Ps Ps × Fw

In Removed In In In

1 1 1 1 1

3.8056 0.4054 10.8259 10.4030 59.5251

0.0688 0.5332 ** ** ***

0.8938

Step 2

0.8975

3.2. Simulations for alternative regimes Simulation (Fig. 2) carried out using the above mentioned growth (Eqs. (2a) and (2b); Mayer et al., 2012) and feeding (Eq. (3)) models for the manufacturer's recommendations (Table 3) and ongrowing conditions, indicates that 928.95 g have to be supplied for one GHSB specimen to increase its weight from 20 g to 500 g, in 468 days. This means a feeding conversion rate (FCR) of 1.92 for the whole ongrowing process. This simulation overestimates the feed needed since it does not consider any LbC. Running Eq. (4), the estimated average LbC would be 8.45% of the supplied feed (Fig. 3) for the whole ongrowing period, so the corrected FCR improves to 1.81 when LbC is considered. Therefore, the really ingested feed (RIF) without LbC would be 850.44 g. For alternative regimes 1, 2 and 3, the supplied feed was 912.94, 890.08 and 886.32 g, respectively, which involved an average LbC for the whole ongrowing process of 6.85, 4.45 and 4.05%, respectively (Fig. 3). Likewise, FCR for the alternative regimes were 1.89, 1.85 and 1.84, respectively.

4. Discussion GHSB fed with feed pellets generate substantial quantities of feed waste as a result of their chewing behaviour. The amount of waste increases with increasing pellet size, but differs depending on fish body weight: the differences in LbC as pellets increase in size become greater as fish size increases. Simulations under Mediterranean fish farming conditions up to a final body weight of 500 g reveal that LbC may reach 8.45% of the supplied food on average using the pellet sizes recommended by the aquafeed producer. Rearing larger fish would involve even higher levels of feed waste. However, alternative pellet size management may reduce the LbC considerably. In the wild, GHSB is able to modify its diet depending on prey availability. The development of its chewing apparatus allows GHSB to process increasingly harder feed types, from nematodes and polychaetes

Fig. 2. Average estimated values of Losses by Chewing (LbC) for 28, 76, 150, 343, 658 and 1019 g GHSB fed with different pellet sizes.

Fig. 3. Simulated monthly average accumulated LbC for the manufacturer recommendations and the three alternative feeding regimes (Table 3).

when juveniles to shellfish and crustaceans at more advanced ages (Russo et al., 2007; Tancioni et al., 2003). This feeding plasticity has favoured its easy adaptation to rearing in captivity, since it accepts a wide range of feed types. Hence, GHSB is considered as a generalist predator (Pita et al., 2002). GHSB feeding behaviour involves repetitive strong crushing and considerable oral manipulation to reduce the feed size before swallowing (Andrew et al., 2003). So, GHSB is used to grinding its feed as needed. Under intensive conditions, most reared fish are fed with extruded feedstuffs. This type of feed has many advantages including the use of specific formulations, high stability, low density, appropriate floatability, easiness of storage and handling, etc. (Jobling et al., 2001). Several works mention the need to use different pellet sizes to feed successive growth stages (Pillay, 1990). Optimum feed size has an important influence on fish rearing efficiency (Hasan and Macintosh, 1992; Wankowski and Thorpe, 1979). Normally, optimum pellet size has been determined based on mouth width; for example, Scophthalmus maximus larvae (Cunha and Planas, 1999), Cyprinus carpio fries (Hasan and Macintosh, 1992), Scophthalmus maximus juveniles (Irwin et al., 2002), Salmo salar juveniles (Wankowski and Thorpe, 1979), Salvelinus alpinus adults (Linnér and Brännäs, 1994; Tabachek, 1988). Following this approach, optimum food size for GHSB has only been established for larvae (Fernández-Diaz et al., 1994), and we have found no information concerning the criteria that should be followed for determining optimum pellet size for the development of GHSB. It seems that until now the most favourable pellet sizes for GHSB juveniles and adults have been estimated from subjective observations (as for other fish species; Hasting and Dickie, 1972) and by comparison with other species. Whatever the case, the particular chewing behaviour of GHSB seems not to have been taken into account for defining the feeding strategy of this species. When feeding, fish go through different sequential phases, including stimulation, identification-localization and consumption. The last involves food intake and ingestion or rejection (Lamb, 2001). Particularly in the case of GHSB, chewing would be an action included in the consumption phase and would depend on the feed size. Andrew et al. (2003, 2004a) described GHSB feeding behaviour and the consequent loss of particles caused by pellet chewing, which results in an increase in feed waste and feeding conversion rate. These authors also tested softer pellets for feeding GHSB to investigate whether this strategy could help reduce feed losses (Andrew et al., 2004b), finding that GHSB were more efficient at crushing. However, this strategy did not reduce feed waste. These authors also suggested that there is scope for varying pellet texture to improve feeding efficiency and to reduce waste production. Our work empirically demonstrates that the smaller


the pellet size, the lower the resulting waste, and also theoretically shows that an alternative strategy such as the management of pellet size would lead to a significant reduction of LbC. We hypothesize that small pellets could be swallowed directly, or, if chewed, lead to lower or negligible losses even when provided to large GHSB. Nevertheless, the effect of chewing differs among different fish sizes as a consequence the development of the chewing apparatus: the larger the fish, the longer the manipulation and the more effective the mastication (as Andrew et al. (2003) postulated), which greater feed wastes. Thus, we hypothesize that feeding GHSB with the smallest possible pellets would reduce the resulting wastes to within a range of acceptability and handling viability. However, large fish feeding small pellets will require more time and energy for food intake, which could affect ongrowing efficiency. Therefore, alternative pellet size regimes as those proposed in this work should involve the re-formulation of feed in terms of energy and nutrients supply to compensate the extra feed trapping effort. Nevertheless, this conjecture needs to be assessed experimentally. Following the manufacturer's recommendations with regard to pellet size without taking into account GHSB chewing behaviour, a substantial LbC was estimated, representing about 80 g of feed per kg of fish produced. Curiously, fish producers know about the GHSB chewing behaviour but the significance of these losses have never been assessed with regards to potential environmental effects and profitability. However, by considering alternative pellet sizes it is possible to reduce wastes by up to 50%, which means saving up to 42 g of feed per kg of reared fish. This benefit may be even greater if GHSB is reared up to larger sizes. Obviously, these considerations have both environmental and economic consequences. Nevertheless, it should be taken into account that our simulations were performed considering that fish will accept the alternative pellet size regimes, which needs to be validated under intensive rearing conditions. Most waste output predictions for GHSB are based on mass balance (Alongi et al., 2009; Lupatsch and Kissil, 1998) or bioenergetic (Piedecausa et al., 2010a) models to obtain nutrient gross waste production. Usually, the feed supplied (but not necessarily ingested) is estimated intuitively by farmers since it largely depends on the care taken in handling (e.g. 5–15%: Beveridge et al. (1997), Cho and Bureau (1997), Findlay and Watling (1994)), and only in a few cases has it been experimentally measured in the field (Rapp et al., 2007), and never for GHSB. Most environmental impact assessments for GHSB projects include both gross waste production and estimated uneaten feed (Mateus and Neves, 2013), but the LbC has never been included. We suggest the application of alternative pellet size management to mitigate the environmental impact derived from organic wastes, and also incorporating LbC estimates in environmental assessments, considering that these losses may be significant. Acknowledgements This research was funded by the Autonomic Government (Department of Agriculture and Water) of Murcia, Spain (Regional Programme cofunded by FEDER, project grant numbers RM-POI-07-043 and 1420-10). The study was also partially sponsored by the student-grant Sub-programme of Researcher Formation of the Spanish Institute of Agro-Forestry Research (FPI-INIA). References Alongi, D.M., McKinnon, A.D., Brinkman, R., Trott, L.A., Undu, M.C., Muawanah, Rachmansyah, 2009. The fate of organic matter derived from small-scale fish cage aquaculture in coastal waters of Sulawesi and Sumatra, Indonesia. Aquaculture 295, 60–75. http://dx.doi.org/10.1016/j.aquaculture.2009.06.025. Andrew, J.E., Anras, M.L.B., Kadri, S., Holm, J., Huntingford, F.A., 2003. Feeding responses of hatchery-reared Gilthead Sea bream (Sparus aurata L.) to a commercial diet and natural prey items. Mar. Freshw. Behav. Physiol. 36, 77–86. http://dx.doi.org/10.1080/ 1023624031000109864. Andrew, J., Holm, J., Huntingford, F., 2004a. The effect of pellet texture on the feeding behaviour of gilthead sea bream (Sparus aurata L.). Aquaculture 232, 471–479. http:// dx.doi.org/10.1016/S0044-8486(03)00490-3.

115

Andrew, J., Holm, J., Kadri, S., Huntingford, F., 2004b. The effect of competition on the feeding efficiency and feed handling behaviour in gilthead sea bream (Sparus aurata L.) held in tanks. Aquaculture 232, 317–331. http://dx.doi.org/10.1016/S00448486(03)00528-3. APROMAR, 2014. La acuicultura en España 2014. Asociación Empresarial de Productores de Cultivos Marinos de España. http://www.apromar.es/content/informes-anuales. Bernabé, G., 1991. Acuicultura. 1991st ed. Omega, Barcelona. Beveridge, M.C.M., Phillips, M.J., Macintosh, D.J., 1997. Aquaculture and the environment: the supply of and demand for environmental goods and services by Asian aquaculture and the implications for sustainability. Aquac. Res. 28, 797–807. http://dx.doi. org/10.1046/j.1365-2109.1997.00944.x. Bryan, J.E., 1973. Feeding history, parental stock, and food selection in rainbow trout. Behaviour 45, 123–153. Bureau, D.P., Hua, K., 2010. Towards effective nutritional management of waste outputs in aquaculture, with particular reference to salmonid aquaculture operations. Aquac. Res. 41, 777–792. http://dx.doi.org/10.1111/j.1365-2109.2009.02431.x. Buryniuk, M., Petrell, R.J., Baldwin, S., Lo, K.V., 2006. Accumulation and natural disintegration of solid wastes caught on a screen suspended below a fish farm cage. Aquac. Eng. 35, 78–90. http://dx.doi.org/10.1016/j.aquaeng.2005.08.008. Castriota, L., Finoia, M.G., Campagnuolo, S., Pipitone, C., Andaloro, F., 2006. Diet of juvenile Pagrus pagrus (Sparidae), from sandy bottoms of the southern Tyrrhenian Sea. Cybium Int. J. Ichthyol. 30, 291–295. Chakroun-Marzouk, N., Kartas, F., 1987. Denture et regime alimentaire des espèces du genre pagrus (Pisces, Sparidae) des côtes Tunisiennes. Cybium 11, 3–19. Cho, C.Y., Bureau, D.P., 1997. Reduction of waste output from salmonid aquaculture through feeds and feeding. Progress. Fish Cult. 59, 155–160. http://dx.doi.org/10. 1577/1548–8640(1997)059b0155:ROWOFSN2.3.CO;2. Clarke, L.A., Sutterlin, A.M., 1985. Associative learning short-term memory and colour preference during first feeding by juvenile Atlantic salmon. Can. J. Zool. 9–14. Colorni, A., Padrós, F., 2011. Diseases and health management. Sparidae: Biology and Aquaculture of Gilthead Seabream and Other Species. Wiley-Blackwell, pp. 321–357. Cox, E.S., Pankhurst, P.M., 2000. Feeding behaviour of greenback flounder larvae, Rhombosolea tapirina (Günther) with differing exposure histories to live prey. Aquaculture 183, 285–297. http://dx.doi.org/10.1016/S0044-8486(99)00302-6. Cunha, I., Planas, M., 1999. Optimal prey size for early turbot larvae (Scophthalmus maximus L.) based on mouth and ingested prey size. Aquaculture 175, 103–110. http://dx.doi.org/10.1016/S0044-8486(99)00040-X. Dörner, H., Wagner, A., 2003. Size-dependent predator–prey relationships between perch and their fish prey. J. Fish Biol. 62, 1021–1032. http://dx.doi.org/10.1046/j.1095-8649. 2003.00092.x. Dörner, H., Hülsmann, S., Hölker, F., Skov, C., Wagner, A., 2007. Size-dependent predator– prey relationships between pikeperch and their prey fish. Ecol. Freshw. Fish 16, 307–314. http://dx.doi.org/10.1111/j.1600-0633.2006.00223.x. Fernandes, Eleftheriou, Ackefors, Eleftheriou, Ervik, Sanchez-Mata, Scanlon, White, Cochrane, Pearson, Read, 2001. The scientific principles underlying the monitoring of the environmental impacts of aquaculture. J. Appl. Ichthyol. 17, 181–193. http:// dx.doi.org/10.1046/j.1439-0426.2001.00315.x. Fernández-Diaz, C., Pascual, E., Yúfera, M., 1994. Feeding behaviour and prey size selection of gilthead seabream, Sparus aurata, larvae fed on inert and live food. Mar. Biol. 118, 323–328. http://dx.doi.org/10.1007/BF00349800. Findlay, R., Watling, L., 1994. Toward a process level model to predict the effects of salmonnet-pen aquaculture on the benthos. In: Hargrave, B.T (Ed.)Modelling Benthic Impacts of Organic Enrichment from Marine Aquaculture, Canadian Technical Reportof Fisheries and Aquatic Sciences. (Minister of Supply and Services (ISSN 0607-6457), Canada), p. 136. Gowen, R.J., Bradbury, N.B., 1987. The ecological impact of salmonid farming in coastal waters: a review. Oceanogr. Mar. Biol. Annu. Rev. 25, 563–575. Hasan, M.R., Macintosh, D.J., 1992. Optimum food particle size in relation to body size of common carp, Cyprinus carpio L., fry. Aquac. Res. 23, 315–325. http://dx.doi.org/10. 1111/j.1365-2109.1992.tb00774.x. Hasting, W.H., Dickie, L.M., 1972. Feed formulation and evaluation. In: Halver, J.E. (Ed.), Fish Nutrition. Academic Press, New York, pp. 327–374. Holmer, M., Marba, N., Diaz-Almela, E., Duarte, C.M., Tsapakis, M., Danovaro, R., 2007. Sedimentation of organic matter from fish farms in oligotrophic Mediterranean assessed through bulk and stable isotope (δ13C and δ15N) analyses. Aquaculture 262, 268–280. http://dx.doi.org/10.1016/j.aquaculture.2006.09.033. Irwin, S., O'Halloran, J., FitzGerald, R., 2002. Mouth morphology and behavioural responses of cultured turbot towards food pellets of different sizes and moisture content. Aquaculture 205, 77–88. http://dx.doi.org/10.1016/S0044-8486(01)00671-8. Islam, M.S., 2005. Nitrogen and phosphorus budget in coastal and marine cage aquaculture and impacts of effluent loading on ecosystem: review and analysis towards model development. Mar. Pollut. Bull. 50, 48–61. http://dx.doi.org/10.1016/j. marpolbul.2004.08.008. Jobling, M., Gomes, E., Dias, J., 2001. Feed types, manufacture and ingredients. Food Intake in Fish. Blackwell Science Ltd., pp. 25–48. Knights, B., 1985. Feeding behaviour and fish culture. In: Cowey, C.B., Mackie, A.M., Bell, J.G. (Eds.), Nutrition and Feeding in Fish. Academic Press, London, pp. 223–241. Lamb, C.F., 2001. Gustation and feeding behaviour. Food Intake in Fish. Blackwell Science Ltd., pp. 108–130. Linnér, J., Brännäs, E., 1994. Behavioral response to commercial food of different sizes and self-initiated food size selection by Arctic char. Trans. Am. Fish. Soc. 123, 416–422. http://dx.doi.org/10.1577/1548–8659(1994)123b0416:BRTCFON2.3.CO;2. Lupatsch, I., Kissil, G., 1998. Predicting aquaculture waste from gilthead seabream (Sparus aurata) culture using a nutritional approach. Aquat. Living Resour. 11, 265–268. http://dx.doi.org/10.1016/S0990-7440(98)80010-7.

116


Magill, S.H., Thetmeyer, H., Cromey, C.J., 2006. Settling velocity of faecal pellets of gilthead sea bream (Sparus aurata L.) and sea bass (Dicentrarchus labrax L.) and sensitivity analysis using measured data in a deposition model. Aquaculture 251, 295–305. http://dx.doi.org/10.1016/j.aquaculture.2005.06.005. Mateus, M., Neves, R. (Eds.), 2013. Ocean Modelling for Coastal Management - Case Studies with MOHID. IST Press. Mayer, P., Estruch, V., Blasco, J., Jover, M., 2008. Predicting the growth of gilthead sea bream (Sparus aurata L.) farmed in marine cages under real production conditions using temperature- and time-dependent models. Aquac. Res 39, 1046–1052. http:// dx.doi.org/10.1111/j.1365-2109.2008.01963.x. Mayer, P., Estruch, V., Martí, P., Jover, M., 2009. Use of quantile regression and discriminant analysis to describe growth patterns in farmed gilthead sea bream (Sparus aurata). Aquaculture 292, 30–36. http://dx.doi.org/10.1016/j.aquaculture.2009.03. 035. Mayer, P., Estruch, V.D., Jover, M., 2012. A two-stage growth model for gilthead sea bream (Sparus aurata) based on the thermal growth coefficient. Aquaculture 358–359, 6–13. http://dx.doi.org/10.1016/j.aquaculture.2012.06.016. Picone, G., Engelsen, S.B., Savorani, F., Testi, S., Badiani, A., Capozzi, F., 2011. Metabolomics as a powerful tool for molecular quality assessment of the fish Sparus aurata. Nutrients 3, 212–227. http://dx.doi.org/10.3390/nu3020212. Piedecausa, M.A., Aguado-Giménez, F., García-García, B., Ballester, G., Telfer, T., 2009. Settling velocity and total ammonia nitrogen leaching from commercial feed and faecal pellets of gilthead seabream (Sparus aurata L. 1758) and seabass (Dicentrarchus labrax L. 1758). Aquac. Res. 40, 1703–1714. http://dx.doi.org/10.1111/j.1365-2109. 2009.02272.x. Piedecausa, M.A., Aguado-Giménez, F., Cerezo-Valverde, J., Hernández-Llorente, M.D., García-García, B., 2010a. Simulating the temporal pattern of waste production in farmed gilthead seabream (Sparus aurata), European seabass (Dicentrarchus labrax) and Atlantic bluefin tuna (Thunnus thynnus). Ecol. Model. 221, 634–640. http://dx. doi.org/10.1016/j.ecolmodel.2009.11.011. Piedecausa, M.A., Aguado-Giménez, F., García-García, B., Telfer, T.C., 2010b. Total ammonia nitrogen leaching from feed pellets used in salmon aquaculture. J. Appl. Ichthyol. 26, 16–20. http://dx.doi.org/10.1111/j.1439-0426.2009.01352.x. Pillay, T.V.R., 1990. Aquaculture: Principles and practices. Fishing News Books. Pita, C., Gamito, S., Erzini, K., 2002. Feeding habits of the gilthead seabream (Sparus aurata) from the Ria Formosa (southern Portugal) as compared to the black seabream (Spondyliosoma cantharus) and the annular seabream (Diplodus annularis). J. Appl. Ichthyol. 18, 81–86. http://dx.doi.org/10.1046/j.1439-0426.2002.00336.x.

Rapp, P., Ramírez, W.R., Rivera, J.A., Carlo, M., Luciano, R., 2007. Measurement of organic loading under an open-ocean aquaculture cage, using sediment traps on the bottom. J. Appl. Ichthyol. 23, 661–667. http://dx.doi.org/10.1111/j.1439-0426.2007.00900.x. Russo, T., Costa, C., Cataudella, S., 2007. Correspondence between shape and feeding habit changes throughout ontogeny of gilthead sea bream Sparus aurata L., 1758. J. Fish Biol. 71, 629–656. http://dx.doi.org/10.1111/j.1095-8649.2007.01528.x. Scharf, F.S., Juanes, F., Rountree, R.A., 2000. Predator size — prey size relationships of marine fish predators: Interspecific variation and effects of ontogeny and body size on trophic-niche breadth. Mar. Ecol. Prog. Ser. 208, 229–248. Smith, I.P., Metcalfe, N.B., Huntingford, F.A., 1995. The effects of food pellet dimensions on feeding responses by Atlantic salmon (Salmo salar L.) in a marine net pen. Aquaculture 130, 167–175. http://dx.doi.org/10.1016/0044-8486(94)00207-5. Specziár, A., 2011. Size-dependent prey selection in piscivorous pikeperch Sander lucioperca and Volga pikeperch Sander volgensis shaped by bimodal prey size distribution. J. Fish Biol. 79, 1895–1917. http://dx.doi.org/10.1111/j.1095-8649.2011.03127.x. Tabachek, J.-A.L., 1988. The effect of feed particle size on the growth and feed efficiency of Arctic charr [Salvelinus alpinus (L.)]. Aquaculture 71, 319–330. http://dx.doi.org/10. 1016/0044-8486(88)90201-3. Tancioni, L., Mariani, S., Maccaroni, A., Mariani, A., Massa, F., Scardi, M., Cataudella, S., 2003. Locality-specific variation in the feeding of Sparus aurata L.: evidence from two Mediterranean lagoon systems. Estuar. Coast. Shelf Sci. 57, 469–474. http://dx. doi.org/10.1016/S0272-7714(02)00376-1. Teles, A.O., Lupatsch, I., Nengas, I., 2011. Nutrition and feeding of Sparidae, in: Sparidae: Biology and aquaculture of gilthead seabream and other species. Wiley-Blackwell, pp. 199–232. Tort, L., Pavlidis, M.A., Woo, N.Y.S., 2010. Stress and welfare in sparid fish. In: Pavilidis, M.A., Mylonas, C.C. (Eds.), Sparidae: Biology and Aquaculture of Gilthead Seabream and Other Species. Wiley-Blackwell, New Delhi, India, pp. 75–94. Vandewalle, P., Saintin, P., Chardon, M., 1995. Structures and movements of the buccal and pharyngeal jaws in relation to feeding in Diplodus sargus. J. Fish Biol. 46, 623–656. http://dx.doi.org/10.1111/j.1095-8649.1995.tb01101.x. Wankowski, J.W.J., Thorpe, J.E., 1979. The role of food particle size in the growth of juvenile Atlantic salmon (Salmo salar L.). J. Fish Biol. 14, 351–370. http://dx.doi.org/10. 1111/j.1095-8649.1979.tb03530.x. Xu, Z., Lin, X., Lin, Q., Yang, Y., Wang, Y., 2007. Nitrogen, phosphorus, and energy waste outputs of four marine cage-cultured fish fed with trash fish. Aquaculture 263, 130–141. http://dx.doi.org/10.1016/j.aquaculture.2006.10.020.