Aquaculture 187 Ž2000. 185–198 www.elsevier.nlrlocateraqua-online
Pacific oyster žCrassostrea gigas / feeding responses to a fish-farm effluent Sebastien Lefebvre a,) , Laurent Barille´ b, Muriel Clerc a ´ a
´ Centre de Recherche en Ecologie Marine et Aquaculture (CNRS-Ifremer), BP 5, 17137 L’Houmeau, France b Laboratoire de Biologie Marine, ISOMER, Faculte´ des Sciences, 2 rue de la Houssiniere, ` 44322 Nantes Cedex 3, France Accepted 7 December 1999
Abstract Bivalves have often been used in integrated fish-farming to enhance the economical value of by-products andror to improve water quality. However, no physiological studies have dealt with the contribution of the two main sources of organic matter potentially present in a fish-farm effluent: living cells of phytoplankton and detritical matter Žfish-faeces and uneaten feed.. This study evaluated feeding responses of the Pacific oyster Ž Crassostrea gigas . to a land-based fish-farm effluent comprised mainly of fish-faeces Ž Dicentrarchus labrax . and compared them with those obtained with a diatom Ž Skeletonema costatum.. A particular distinction of the main sources of organic matter was made in the experiments, the two diets being evaluated separately and mixed. Feeding responses were evaluated using the biodeposit method with special attention being paid to pre-ingestive processes and absorption efficiency ŽAE.. Experiments were carried out between May and July 1998 in the laboratory at constant water temperature Ž208C.. Results showed that food quality Žnature of particulate organic matter, POM, organic content of the total suspended matter, TSM. had no influence on clearance rates. However, clearance rates were significantly reduced for oysters in advanced stage of gametogenesis. S. costatum was preferentially ingested compared to the fish-faeces when offered in a mixed diet. Absorption efficiency and energy content Ž56% and 15 J mgy1, respectively. of the fish-faeces were lower than those estimated for S. costatum Ž66–70%, 20 J mgy1 .. Nevertheless, these values were high for such a detritic type of food. This study confirms that both concepts of biomechanical filter and economical value improvement of fish-farming waste using oysters are of interest. q 2000 Elsevier Science B.V. All rights reserved. Keywords: Oyster; Fish-faeces; Integrated aquaculture; Biological treatment; Nutrition
)
Corresponding author. Tel.: q33-0-5465-094-40; fax: q33-0-5465-006-00. E-mail address:
[email protected] ŽS. Lefebvre..
0044-8486r00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 4 - 8 4 8 6 Ž 9 9 . 0 0 3 9 0 - 7
186
S. LefebÕre et al.r Aquaculture 187 (2000) 185–198
1. Introduction Intensive fish-farming continuously generates large amounts of both particulate organic waste Žuneaten feed and faeces., and dissolved excretory products owing to an inefficient use of diet ŽHandy and Poxton, 1983; Porter et al., 1987; Dosdat et al., 1996.. In particular, dissolved products can lead to an uncontrolled phytoplankton production generated by eutrophication around cage systems ŽFolke and Kautsky, 1989. or landbased pond farms ŽKrom et al., 1985.. The use of suspension-feeders in intensive fish or shrimp aquaculture, Že.g., mussels or oysters. is generally done to enhance the economical value of aquaculture in land-based systems ŽShpigel et al., 1993; Bodvin et al., 1996; Sandifer and Hopkins, 1996. or in off-shore systems ŽJones and Iwama, 1991; Stirling and Okumus, 1995.. Bivalves can also be used to improve water quality by removing particulate organic matter ŽPOM. ŽShpigel and Blaylock, 1991; Shpigel et al., 1997.. This ‘‘biomechanical filter’’ can reduce the impact of facilities on the surrounding ecosystem and the fish-farm itself. POM produced by intensive fish-farming has two origins: particulate organic waste from uneaten feed and faeces, and transformation of the dissolved excreted products to phytoplankton through regenerated primary production. Although previous fish-farming studies have considered the possibility of using bivalves to treat and improve the economical value of aquaculture wastes, none have made a clear distinction between these two potential food sources for bivalves ŽShpigel et al., 1997.. As a consequence, bivalve feeding responses remain unclear in these situations. This study was designed to determine the potential value of a fish-farm effluent characterized by fish-faeces as a food source for the Pacific oyster, and compare it with a diet of the diatom Skeletonema costatum. In particular, we intended to determine pre-ingestive food selection and absorption of these two types of food. These mechanisms are important in order to determine oyster growth in a treatment system, but are also fundamental processes common to the oyster population from the shellfish ecosystem surrounding the fish farm.
2. Materials and methods 2.1. Measure of feeding processes Experiments were performed from May to July 1998 in a controlled temperature room of around 208C Ž"18C.. During this period, the farm reached its maximum production. This temperature was chosen because it was close to that recorded in the farm ponds Ž19.78C "1.8 S.D... Furthermore, it matched the thermic optimum for the clearance rate of Crassostrea gigas ŽBougrier et al., 1995.. The oysters used in the experiment were adults cultivated in the Marennes–Oleron Bay; their main biometric ´ characteristics are given in Table 1. They were stocked and acclimatized to the temperature for 2 weeks before the beginning of each experiment.
Table 1 Conditions of concentrations Žmg ly1 . of total suspended matter ŽTSM., volatile suspended matter ŽPOM., in the three diets: gross fish farm effluent ŽE., phytoplankton culture ŽP. and mixed ŽM.. Oyster biometrics Žmean and s.d.. and level of gametogenesis stage according to the scale of Marteil Ž1976.. Žn.m.. not measured POM
Total live weight
Flesh live weight
Dry flesh weight
Shell weight
Gam. stage
32.0 45.0 26.0 37.8 15.2
5.8 6.8 5.0 6.2 3.4
59.8 Ž6.8. 75.8 Ž19.5. 67.6 Ž6.7. 59.4 Ž14.7. 62.1 Ž8.9.
11.8 Ž0.9. 13.5 Ž3.8. 12.2 Ž1.2. 11.0 Ž3.8. 10.3 Ž2.2.
1.5 Ž0.3. 1.9 Ž0.7. 1.6 Ž0.3. 1.4 Ž0.7. 1.9 Ž0.8.
36.9 Ž3.9. 45.4 Ž12.6. 39.3 Ž3.1. 33.3 Ž8.2. 35.7 Ž5.1.
3–4 3–4 3–4 4 1
P1 P2
15.9 12.3
11.3 8.6
77.0 Ž18.1. 81.1 Ž17.5.
10.4 Ž1.8. 13.4 Ž3.3.
1.1 Ž0.3. 1.6 Ž0.6.
44.0 Ž8.3. 47.1 Ž11.5.
1 1
Mixed ŽM.
Source
80.0 Ž10.7. 70.6 Ž13.8. 61.0 Ž8.9.
12.9 Ž2.2. 11.7 Ž2.1. 9.4 Ž2.1.
1.8 Ž0.5. 1.6 Ž0.5. 1.2 Ž0.6.
48.3 Ž7.9. 41.4 Ž9.3. n.m.
1 4 3
Effluent ŽE. E1 E2 E3 E4 E5 Phytoplankton ŽP.
M1 M2 M3
Source
E
P
4.9 5.3 13.2
18.3 15.4 12.3
E 0.7 1.1 2.2
P 10.8 9.2 7.4
S. LefebÕre et al.r Aquaculture 187 (2000) 185–198
TSM
187
188
S. LefebÕre et al.r Aquaculture 187 (2000) 185–198
Different diets were distributed through a peristaltic pump in feeding chambers similar to those used by Barille´ et al. Ž1997. and Pastoureaud et al. Ž1996., where five oysters and an empty shell were placed. Diets had been kept in several 1-m3 tanks, with strong aeration to avoid sedimentation, in the same controlled temperature room as the feeding chambers. Before the beginning of each experiment, oysters were acclimatized for 24 h to the diet. Pseudo-faeces ŽPF. and faeces ŽF. were collected separately after a production period of 8 h. The oysters’ feeding responses were estimated using the biodeposit method ŽIglesias et al., 1998.. Thus, clearance rate ŽCR; l hy1 indy1 . was estimated by dividing the sum of the particulate inorganic matter ŽPIM. in the PF and F by the concentration of PIM in water. Filtration rate ŽFR; mg hy1 indy1 . of POM was obtained by multiplying CR by POM concentration in water. Ingestion rate ŽIR; mg hy1 indy1 . of POM was FR of POM minus POM in PF. Finally, absorption efficiency ŽAE; mg hy1 indy1 . was calculated by dividing, the difference between IR of POM and POM in F, by IR of POM. All feeding processes estimated for individuals were standardized to those of an equivalent individual of 1-g dry tissue weight applying an allometric exponent of 0.67 ŽShpigel et al., 1992.. Water flow rate varied from 4 to 6.6 l hy1 per oyster. Flow rate was regulated manually, depending on oyster filtration rates and according to the instructions for the method: particle concentration was maintained at 30% higher at the inflow than at the outflow. Water samples, as well as biodeposits ŽPF and F., were filtered onto Whatman GFrC filters, dried at 608C and ashed at 4508C to determine the amount of organic and mineral matter. Chlorophyll-a and pheopigments were analysed both in seawater and biodeposits using fluorometry after methanol extraction in a Turner fluorometer AU-10. The sum of chlorophyll-a and pheopigments will be referred to as total pigments. Particulate organic carbon ŽPOC. and particulate organic nitrogen ŽPON. contents of the different food sources were determined by combusting filters in a Carlo Erba analyser. Particle-sized distributions were determined with a Coulter Multisizer fitted with a 100-mm aperture and were used to calculate oyster retention efficiencies ŽRE; Barille´ et al., 1993.. RE were measured in individual flow-through chambers for each diet by subtracting for each class of particle size Ževery 0.3 mm. the outflow of a chamber occupied by an oyster with the outflow of a reference with an oyster shell. Results are expressed as a percentage relative to the class that was most retained. Particle weight distribution was estimated using a 50-mm mesh sieve. Oyster level of gametogenesis was estimated by observations of the gonad and was classified according to the scale of Marteil Ž1976.. This scale ranges from 1 to 5. Level 4 represents the most advanced level of gonadal development. Although level 5 represents gonads after spawning, it is not possible to distinguish level 5 from level 1 and thus they are noted similarly. Dry weights were obtained from freeze-dried soft tissues. 2.2. Experimental diets 2.2.1. Effluent sampling Wastewater effluents were sampled directly from a private farm fish-pond Ž Dicentrarchus labrax . located in salt marshes on the French Atlantic coast and brought back to the laboratory in several 0.6-m 3 polyethylene tanks. The volume of an average
S. LefebÕre et al.r Aquaculture 187 (2000) 185–198
189
pond was 8.000 m3, its depth 3 m, with a fish biomass of 100 tons. The water renewal rate was around 400% a day. Daily feeding was between 1% and 2% of fish biomass. Feeding was made manually two or three times a day with floating pellets in order to satiate fish with minimum feed waste. Budget balances of an earlier study ŽHussenot et al., 1995. showed that most of the POM in suspension in pond water came from fish particulate excretion. Fishponds were shaded to limit photosynthesis production Ž- 10 mg ly1 of total pigments mean concentrations.. 2.2.2. Phytoplankton culture The diatom Ž S. costatum) was used as a reference for a phytoplankton source of organic matter in the experiments. This diatom was chosen because of its importance in the phytoplankton population both on the Atlantic coast and in outdoor culture of natural populations using land-based fish-farm effluent as a source of nutrients ŽLefebvre et al., 1996.. This diatom was grown on decanted and filtered seawater, with nitrogen, phosphorus and silicate added in an atomic ratio of 10r1r4 with 250 mmol ly1 ŽmM. of NHq 4 as a source of nitrogen. Culture was conducted outdoors in several 500-l tanks with light aeration. Culture was completed in 2 or 3 days, at an average temperature of 228C. 2.2.3. Mixed diet For the mixed diet treatment, the two food sources described above were stocked separately and mixed just before entering the feeding chamber. In this way, the food quality of the sources and of the mixing and proportion of each source was known exactly in the water column. Proportions of each diet were chosen in the range of that encountered in phytoplankton outdoor continuous culture, using fish farm effluent as a source of nutrients ŽHussenot et al., 1998; Lefebvre and Hussenot, 1998.. As for the biodeposits, non-algal organic particulate matter Že.g., fish-faeces. was estimated by subtracting algal organic matter from total POM. The former was estimated
Fig. 1. Mean Ž"CI 95%. of the ratio particulate organic matter ŽPOM. over total suspended matter ŽTSM. for the three types of diets. Results from ANOVA showed significant differences between the means Ž df 48, P - 0.001.. Means not sharing a common superscript are significantly different Ž P - 0.05, Newman–Keul’s test..
S. LefebÕre et al.r Aquaculture 187 (2000) 185–198
190
Table 2 Mean percentage Ž"SE. of nitrogen ŽN. and carbon ŽC. relative to the organic matter for the fish farm effluent, the fish feed and the phytoplankton. The CrN ratio Žmg mgy1 . is also presented. Results from ANOVA showed significant differences between the means Ž df 32, P - 0.001.. Means of each column not sharing a common superscript are significantly different Ž P - 0.05, Newman–Keul’s test.
Effluent Fish feed Phyto.
N
C
CrN
6.18 Ž"0.30. c 7.78 Ž"0.07. b 9.30 Ž"0.63. a
32.40 Ž"1.59. c 43.87 Ž"0.29. b 63.02 Ž"4.96. a
5.25 Ž"0.07. b 5.65 Ž"0.08. b 6.75 Ž"0.28. a
by multiplying total pigment weight Žmg. by 27.5 Ž r 2 s 0.78, n s 62.. This corresponds to 17 mg of carbon per mg of total pigments and 0.63 mg of carbon per mg of POM. 2.3. Statistical analysis Statistical comparisons of experimental data were performed by one-way analysis of variance ŽANOVA. and Newman–Keul’s test. Student’s t-test was used to test a statistical significant difference between Ži. means for each diet when comparison of PF composition was made with water composition and Žii. slopes of linear regressions. Statistics were performed using the software Statgraphics plus 6.0. 3. Results 3.1. Diet and oyster qualities Total suspended particulate matter concentrations ŽTSM. varied from 12.3 to 45 mg l -1 and were linked to the type of diets ŽTable 1.. TSM concentrations were lower in the
Fig. 2. Particle size distributions of the fish farm effluent Žblack diamond-shaped. and phytoplanktonic diets Žwhite squares..
S. LefebÕre et al.r Aquaculture 187 (2000) 185–198
191
Fig. 3. Mean Ž"CI 95%. relative retention efficiency of oysters fed on effluent diets.
phytoplankton diets for which the diatom S. costatum represented the main source of particulate matter. Consequently, POM concentrations were high compared to TSM, with mean organic fraction of 70% ŽFig. 1.. Unlike the high quality diet, the effluent was characterized by a significantly lower mean organic fraction of 18% ŽFig. 1.. The mixed diet presented an intermediate value between both sources. ŽTable 1 and Fig. 1.. Carbon and nitrogen contributions to POM were estimated for the effluent and the
Fig. 4. Mean oyster clearance rates Ž"CI 95%. in the three types of diet conditions. For each diet, a distinction was made according to the stage of gametogenesis development. Oysters belonging to stages 3 and 4 Žadvanced gametogenesis, shown by an ) . have been pooled and compared with oysters at stage 1. Results from ANOVA showed significant differences between the means Ž df 47, P - 0.001.. Means not sharing a common superscript are significantly different Ž P - 0.05, Newman–Keul’s test..
192
S. LefebÕre et al.r Aquaculture 187 (2000) 185–198
phytoplanktonic diets ŽTable 2.. The percentage of C and N was significantly lower in the effluent, mainly consisting of fish-faeces. The fish feed composition is given in Table 2 for comparison to that of fish-faeces. The quality of the diets also differed by the size distributions of the particles. The effluent diet had a size distribution characterized by a mode of around 5 mm Žequivalent spherical-diameter, ESD., whereas that of phytoplankton had a bimodal aspect with two modes of around 10 and 16 mm ESD ŽFig. 2.. In addition, a fractionation of the effluent water samples using a 50-mm mesh sieve showed that 21% of the POM of the effluent was associated with particles larger than 50 mm.
Fig. 5. Pacific oyster preingestive selection. ŽA. Mean Ž"CI 95%. of the ratio POM over TSM for each diet in the water column Žw. and in the pseudo-faeces Žpf.. ) ) and ) show a significant difference between the means ŽStudent’s t-test; P - 0.01 and P - 0.05, respectively.. ŽB. Mixed diets. Mean Ž"CI 95%. ratio POM from the fish faeces over POM from the phytoplankton in the water column Žw. and in the pseudo-faeces Žpf.. )) shows a significant difference between the means ŽStudent’s t-test, P - 0.01..
S. LefebÕre et al.r Aquaculture 187 (2000) 185–198
193
Oyster biometric characteristics did not show significant differences between the experiments, except in their gonadal development ŽTable 1.. The oysters presented two types of gonadal developments: advanced and post-spawning or beginning. 3.2. Retention efficiency (RE) and clearance rate (CR) The relative RE was estimated from effluent diets and revealed that oysters were able to capture particles above 4 mm ESD ŽFig. 3. with an efficiency close to 100%. For each diet, CR has been calculated by making a distinction according to the oyster’s stage of gametogenesis. Oysters belonging to stages 3 and 4 Žadvanced gametogenesis. have been pooled and compared with oysters at stage 1. Mean CR calculated for oysters showing little or no gonadal development Žstage 1. did not differ between the effluent and the phytoplanktonic diets ŽFig. 4.. For the same diet, oysters recognized as being in an advanced stage of gametogenesis Ž3 or 4. had significantly lower CR ŽFig. 4.. 3.3. Pre-ingestiÕe processes The rejection of the different sources of organic matter through PF production before ingestion was compared when offered alone to the bivalves or in a mixed diet. The ratio POMrTSM was significantly lower in the PF than in the water column for the effluent and mixed diets ŽFig. 5A.. This suggests that the proportion of POM ingested was greater than the proportion retained on the oyster gill. The ratio was not significantly
Fig. 6. Rejection of organic matter identified as fish faeces in the oyster pseudofaeces as a function of the bivalve total filtration rates Žorganicqinorganic. in the effluent diets Žblack circle, r 2 s 0.83, P - 0.001, ns 24. and in the mixed diet conditions Žwhite triangle, r 2 s 0.5, ns15.. A Student’s t-test on slopes showed a significant difference Ž P - 0.001..
194
S. LefebÕre et al.r Aquaculture 187 (2000) 185–198
Fig. 7. Mean Ž"CI 95%. of the absorption efficiency ŽAE. for the three types of diets. Results from ANOVA showed significant differences between the means Ž df 45, P - 0.05.. Means not sharing a common superscript are significantly different Ž P - 0.05, Newman–Keul’s test..
different for the phytoplankton diet ŽFig. 5A.. In the mixed experiments, the analyses of the relative proportions of organic sources constituting the organic matter of the mixed diets were estimated by the following ratio: fish-faeces organic matter over phytoplankton organic matter. The comparison of these ratios showed a significant increase in the PF compared to the seawater ŽFig. 5B., suggesting that fish-faeces were preferentially rejected in the PF. This result is supported by the comparison of the regression lines, calculated for the effluent and mixed diets, between the rejection of organic matter identified as fish-faeces in the PF and the oyster total FR ŽFig. 6.. The fish-faeces appeared significantly more rejected in the PF when the bivalves were offered a mixed diet 3.4. Absorption efficiency (AE) Mean AE calculated for the phytoplankton and mixed diets reached high values of 66% and 70%, respectively, the latter being significantly higher than the mean value of 56% obtained for the effluent diet ŽFig. 7..
4. Discussion The results obtained in this study, where two main diets were offered to Pacific oysters Žparticles from a fish-farm effluent andror a phytoplanktonic microalgae. confirmed the importance of food quality on bivalve feeding responses ŽDame, 1993.. This food quality can be characterized by the nature of POM, the proportion of this POM relative to TSM and particle size.
S. LefebÕre et al.r Aquaculture 187 (2000) 185–198
195
The fish-farm effluent was directly sampled in the farm pond, and its associated POM essentially came from fish particulate excretion. POM produced by the fish rearing was in moderate concentrations Ž3 to 7 mg l -1 . due to the high water exchanges. This POM was diluted with inorganic fine clay particles derived mainly from the erosion of the earthen ponds. This explained the average organic fraction of the effluent of 18%. These inorganic particles were probably associated to the mode around 4 mm ŽESD. of the effluent particle size distribution. However, it should be noted that most of fish-faeces particles were found below 50 mm ESD. The effluent particle size distribution was comparable with the results of Cripps Ž1995., who reported that in a land-based salmon farm, the distribution of particles ranged from 5 to 120 mm ESD, with numerous fine particles. The phytoplanktonic diet of the diatom S. costatum, was characterized by a much higher percentage of organic matter. Furthermore, since the microalgae was used alone, the inorganic matter essentially represents the silicon of the frustules. The phytoplanktonic diet size distribution revealed larger particles than in the effluent, which correspond to diatom chains of 2 to 10 cells with chain length varying from 15 to 100 mm ŽSauriau and Baud, 1994.. A previous study on C. gigas retention efficiency ŽBarille´ et al., 1993. suggests for the phytoplankton diet that all algal cells, which compose the size spectra shown in Fig. 2, are potentially fully retained by the oyster gill. In our study, the estimation of particle retention efficiency from the effluent diets suggests that oysters retain particles above 4–5 mm ESD on their gills with an efficiency rate close to 100%. These results indicate that oysters are capable of filtering most of the faecal particles in effluents from land-based fish-farms. This idea is in accordance with the concept of a biomechanical filter using oysters, as suggested by Shpigel et al. Ž1993. on a land-based fish-farm in Israel. The period chosen for the experiment ŽMay–July. corresponded to the maximum production for fish aquaculture on the French Atlantic coast and thus required more biological treatment. However, this period coincides with the development of gametogenesis in C. gigas on the French Atlantic coast ŽDeslous-Paoli and Heral, ´ 1988.. This led to a reduction in CR for the most mature stages, a result, which has also been reported for C. gigas by Soletchnik et al. Ž1997.. Nevertheless, CR for individuals, which were not in an advanced stage of gametogenesis, were lower than those reported by Bougrier et al. Ž1995. at the same temperature Ž4.2 l h -1 g -1 .. This can be explained by the fact that the method we used to measure FR ŽBiodeposit method; Iglesias et al., 1998., gave values representing an average activity of oyster filtration during several hours Ž8 h in measuring periods.. On the contrary, Bougrier et al. Ž1995. measured instantaneous filtration using flow-through systems during active periods of filtration and excluding non-active periods. Two distinct processes of pre-ingestive food selection have been identified. Firstly, a preferential selection of organic vs. inorganic ones ŽNewell and Jordan, 1983; DeslousPaoli et al., 1992.. The oyster is therefore able to compensate for the dilution of fish-faeces by inorganic clay particles by preferentially rejecting the latter in the PF. This selection was obviously not possible with the phytoplankton diet only consisting of microalgae. Secondly, there is a selection among the two types of organic particles in the mixed diet experiments, indicating that S. costatum was preferentially ingested over fish-faeces. Such a selection has been previously shown between estuarine detritic
196
S. LefebÕre et al.r Aquaculture 187 (2000) 185–198
material and phytoplankton ŽPastoureaud et al., 1996; Barille´ et al., 1997.. The underlying mechanism has been demonstrated by Ward et al. Ž1998. using endoscopic examination of the processing of both types of particles on oyster gills. Although the AE of the effluent diet was lower than that of the phytoplankton diet, they remained high for a food, which can be qualified as detritic. This high absorption may be explained by a bacterial mediation as suggested by Crosby et al. Ž1990. for detritic vegetal products. The presence of an important bacterial activity in such an environment is well known ŽGaret et al., 1997.. Fish-faeces results from the degradation in the digestive tract of feed pellets, which were composed mainly of fish meal and products or by-products of cereal grains. The present trend is to increase the vegetable protein sources in fish feeds ŽWatanabe et al., 1993.. Nevertheless, the introduction of cereal grains is limited to about 20% by the high protein content of this type of fish feed. Digestibility of these two protein sources is often equivalent, although cereal grain digestibility can be low because of its ash content ŽRobaina et al., 1997.. The digestibility of the main organic dietary components, 55% protein and 15% lipid, averaged 90% and 85%, respectively ŽHoar et al., 1979.. Theoretically, the same proportion of protein and lipid should be found in the F as in the feed. This also suggests that fish feed and fish-faeces reach the same gross energy per unit biomass as shown by Henken et al. Ž1986.. The similarity of the CrN ratio between the effluent and the feed pellet can be explained by this comparable digestibility of the protein and the lipid. It appears that the CrN ratio of the effluent is lower than that of an estuarine environment ŽBarille´ et al., 1997. and this is explained by a higher content in nitrogen. This suggests that the amount of energy per mg of POM is close to that given by Henken et al. Ž1986. for fish feed as well as fish-faeces, i.e., around 20 J mg -1 of organic material. Nevertheless, in our study, the proportion of carbon, as well as nitrogen in POM ŽTable 2., was higher in the fish feed than the fish-faeces. By simple calculation, we should attribute a gross energy of around 15 J mg -1 in the fish-faeces in our effluent. This theoretical gross energy is twice as high as that generally reported for an estuarine detritical food ŽBarille´ et al., 1997.. Due to low fish feed wastes in the fish farm where we sampled effluents, it seemed that POM consisted mainly of fish-faeces. However, concerning POM of uneaten fish feed, i.e., fish feed waste, there is apparently no evidence as to why oyster responses may be different from fish-faeces responses since particle sizes are below 100 mm ŽBarille´ et al., 1993.. All the results obtained in this study suggest that detritical waste from intensive fish-farming can contribute to the growth of bivalves, in this case Pacific oysters C. gigas. This is in accordance with the hypothesis suggested by Jones and Iwama Ž1991. and Stirling and Okumus Ž1995. both working on off-shore systems with a salmon–mussel association. However, this study provides further information and precision on feeding processes related to this particular type of detritic food. This knowledge is important to understand and explain bivalve growth in integrated cultures or treatment systems, in land-based, as well as off-shore systems. Actually, it indicates that the distinction between the two organic forms generally found in aquaculture effluents or surrounding environments, e.g., fish-faeces or uneaten feed and phytoplankton, has to be made since the bivalve feeding responses are different for the two types of food. This distinction becomes particularly relevant when the quantification of the energy fluxes in
S. LefebÕre et al.r Aquaculture 187 (2000) 185–198
197
such integrated fish-bivalve cultures are analyzed using a modelling approach ŽLefebvre, unpublished data.. Removal and improvement of the economical value of lost organic products by intensive fish-farming could be advantageous both for economical and environmental aspects. Acknowledgements We thank Dr. Heral for ´ Maurice, Dr. Hussenot Jerome ´ ˆ and Dr. Pouvreau Stephane ´ interesting comments on our study. We would also like to thank Robert Knutsen for his English assistance, Lucette Joassard, Marcel Guillaut and Michel Prineau for technical assistance. Special gratitude is extended to the private farm company and its managers Bernard Houin and Andre´ Zwaga, who allowed Sebastien Lefebvre access to their ´ facilities and gave financial support for the study. This study forms part of the PhD Thesis of S. Lefebvre. References Barille, ´ L., Prou, J., Heral, ´ M., Bougrier, S., 1993. No influence of food quality, but ration-dependent retention efficiencies in the Japanese oyster Crassostrea gigas. J. Exp. Mar. Biol. Ecol. 171, 91–106. Barille, ´ L., Prou, J., Heral, ´ M., Razet, D., 1997. Effects of high natural seston concentrations on the feeding, selection, and absorption of the oyster Crassostrea gigas. J. Exp. Mar. Biol. Ecol. 212, 149–172. Bodvin, T., Indergaard, M., Norgaard, E., Jensen, A., Skaar, A., 1996. Clean technology in aquaculture — a production without waste products?. Hydrobiologia 326–327, 83–86. Bougrier, S., Geairon, P., Deslous-Paoli, J.M., Bacher, C., Jonquieres, G., 1995. Allometric relationships and effects of temperature on clearance and oxygen consumption rates of Crassostrea gigas ŽThunberg.. Aquaculture 134, 143–154. Cripps, S.J., 1995. Serial particle size fractionation and characterisation of an aquacultural effluent. Aquaculture 133, 323–339. Crosby, M.P., Newell, R.I.E., Langdon, C.J., 1990. Bacterial mediation in the utilization of carbon and nitrogen from detrital complexes by Crassostrea Õirginica. Limnol. Oceanogr. 35, 625–639. Dame, R.F., 1993. Bivalve Filter Feeders in Estuarine and Coastal Ecosystem Processes. Springer, Berlin, 579 pp. Deslous-Paoli, J.M., Heral, M., 1988. Biochemical composition and energy value of Crassostrea gigas ´ ŽThunberg. cultured in the bay of Marennes–Oleron. Aquat. Living Resour. 1, 239–249. ´ Deslous-Paoli, J.M., Lannou, A.-M., Geairon, P., Bougrier, S., Raillard, O., Heral, ´ M., 1992. Effects of the feeding behaviour of Crassostrea gigas ŽBivalve Molluscs. on biosedimentation of natural particulate matter. Hydrobiologia 231, 85–91. Dosdat, A., Servais, F., Metailler, R., Huelvan, C., Desbruyeres, ´ ` E., 1996. Comparison of nitrogenous losses in five teleost fish species. Aquaculture 141, 107–127. Folke, C., Kautsky, N., 1989. The role of ecosystems for a sustainable development of aquaculture. Ambio 18, 234–243. Garet, M.J., Reymond, H., Delmas, D., 1997. Stimulation of bacterial exoproteolytic activity by fish farming in coastal marine ponds: effects on dissolved protein cycling. Aquat. Living Resour. 10, 221–229. Handy, R.D., Poxton, M.G., 1983. Nitrogen pollution in mariculture: toxicity and excretion of nitrogenous compounds by marine fish. Rev. Fish Biol. Fish. 3, 205–241. Henken, A.M., Lucas, H., Tijssen, P.A.T., Machiels, M.A.M., 1986. A comparison between methods used to determine the energy content of feed, fish and faeces samples. Aquaculture 58, 195–201. Hoar, W.S., Randall, D.J., Brett, J.R., 1979. Bioenergetics and growth, Fish Physiology 8 Academic Press, New York, 786 pp. Hussenot, J., Brossard, N., Lefebvre, S., Benet, A., 1995. Traitement par lagunage des effluents d’une
198
S. LefebÕre et al.r Aquaculture 187 (2000) 185–198
pisciculture marine intensive : qualite´ des eaux et des sediments. In: Amiard-Triquet, C., Hamon, T. ŽEds.., ´ Actes du colloque pluridisciplinaire, la Qualite´ de l’eau, 26–28 octobre 1995, Universite´ de Nantes. Comite´ du Pilotage de l’Environnement, Nantes. pp. 139–143. Hussenot, J., Lefebvre, S., Brossard, N., 1998. Open-air treatment of wastewater from land-based marine fish farms in extensive and intensive systems: current technology and future perspectives. Aquat. Living Resour. 11, 297–304. Iglesias, J.I.P., Urrutia, M.B., Navarro, E., Ibarrola, I., 1998. Measuring feeding and absorption in suspensionfeeding bivalves: an appraisal of the biodeposition method. J. Exp. Mar. Biol. Ecol. 219, 71–86. Jones, T.O., Iwama, G.K., 1991. Polyculture of the pacific oyster, Crassostrea gigas ŽThunberg., with chinook salmon, Oncorhynchus tshawytscha. Aquaculture 92, 313–322. Krom, M.D., Porter, C., Gordin, H., 1985. Description of the water quality conditions in a semi-intensively cultured marine fish pond in Eilat, Israel: daily and diel variations. Aquaculture 49, 141–157. Lefebvre, S., Hussenot, J., 1998. Biological treatment of intensive fish farm effluents using mass phytoplankton production. In: Hussenot, J., Buchet, V. ŽEds.., Marais maritimes et aquaculture: activite´ durable pour la preservation et l’exploitation des zones humides littorales, Actes de colloque, 6–8 juin 1996, Rochefort, ´ France Vol. 19 pp. 224–233, Ifremer, Plouzane, France. Lefebvre, S., Hussenot, J., Brossard, N., 1996. Water treatment of land-based fish farm effluents by outdoor culture of marine diatoms. J. Appl. Phycol. 8, 193–200. Marteil, L., 1976. Shellfish culture in France: Part 2. Oyster and mussel biology. Rev. Trav. Inst. Peches ´ Marit. 40, 149–346. Newell, R.I.E., Jordan, S.J., 1983. Preferential ingestion of organic material by the American oyster Crassostrea Õirginica. Mar. Ecol. Prog. Ser. 13, 47–53. Pastoureaud, A., Heral, ´ M., Prou, J., Razet, D., Russu, P., 1996. Particle selection in the oyster Crassostrea gigas ŽThunberg. studied by pigment HPLC analysis under natural food conditions. Oceanol. Acta 19, 79–88. Porter, C.B., Krom, M.D., Robbins, M.G., Brickell, L., Davidson, A., 1987. Ammonia excretion and total N budget for gilthead seabream Ž Sparus aurata. and its effect on water quality conditions. Aquaculture 66, 287–297. Robaina, L., Moyano, F.J., Izquierdo, M.S., Socorro, J., Vergara, J.M., Montero, D., 1997. Corn gluten and meat and bone meals as protein sources in diets for gilthead seabream Ž Sparus aurata.: nutritional and histological implications. Aquaculture 157, 343–355. Sandifer, P.A., Hopkins, J.S., 1996. Conceptual design of a sustainable pond-based shrimp culture system. Aquacult. Eng. 15, 41–52. Sauriau, P.G., Baud, J.P., 1994. Artificial filament breakage of the diatom Skeletonema costatum intended for mollusc aquaculture. Aquaculture 123, 69–81. Shpigel, M., Barber, B.J., Mann, R., 1992. Effects of elevated temperature on growth, gametogenesis, physiology, and biochemical composition in diploid and triploid Pacific oysters, Crassostrea gigas Thunberg. J. Exp. Mar. Biol. Ecol. 161, 15–25. Shpigel, M., Blaylock, R.A., 1991. The Pacific oyster, Crassostrea gigas, as a biological filter for a marine fish aquaculture pond. Aquaculture 92, 187–197. Shpigel, M., Gasith, A., Kimmer, E., 1997. A biomechanical filter for treating fish-pond effluents. Aquaculture 152, 103–117. Shpigel, M., Neori, A., Popper, D.M., Gordin, H., 1993. A proposed model for ‘‘environmentally clean’’ land-based culture of fish, bivalves and seaweeds. Aquaculture 117, 115–128. Soletchnik, P., Razet, D., Geairon, P., Faury, N., Goulletquer, P., 1997. Ecophysiologie de la maturation Žrespiration. et sexuelle et de la ponte de l’huıtre metaboliques ˆ creuse Crassostrea gigas: reponses ´ ´ alimentaires Žfiltration, absorption. en fonction des differents stades de maturation. Aquat. Living Resour. ´ 10, 177–185. Stirling, H.P., Okumus, I., 1995. Growth and production of mussels Ž Mytilus edulis L.. suspended at salmon cages and shellfish farms in two Scottish sea lochs. Aquaculture 134, 193–210. Ward, J.E., Levinton, J.S., Shumway, S.E., Cucci, T., 1998. Particle sorting in bivalves: in vivo determination of the pallial organs of selection. Mar. Biol. 131, 283–292. Watanabe, T., Pongmaneerat, J., Satoh, S., Takeuchi, T., 1993. Replacement of fish meal by alternative protein sources in rainbow trout diets. Bull. Jpn. Soc. Sci. Fish. 59, 1573–1579.
