Improving production efficiency of farmed Atlantic salmon ( Salmo salar L.) by isoenergetic diets with increased dietary protein-to-lipid ratio
2018; Wiley; Volume: 49; Issue: 4 Linguagem: Inglês
10.1111/are.13598
ISSN1365-2109
AutoresRúni Weihe, Jens‐Erik Dessen, Regin Arge, Magny Sissel S. Thomassen, Bjarne Hatlen, Kjell‐Arne Rørvik,
Tópico(s)Fish Ecology and Management Studies
ResumoAquaculture ResearchVolume 49, Issue 4 p. 1441-1453 ORIGINAL ARTICLEOpen Access Improving production efficiency of farmed Atlantic salmon (Salmo salar L.) by isoenergetic diets with increased dietary protein-to-lipid ratio Rúni Weihe, Corresponding Author Rúni Weihe ruw@havsbrun.fo orcid.org/0000-0002-2760-4352 Havsbrún, Fuglafjørður, Faroe Islands Department of Animal and Aquaculture Sciences, Norwegian University of Life Sciences, Ås, Norway Correspondence Rúni Weihe, Havsbrún P/F, Fuglafjørður, Faroe Islands. Email: ruw@havsbrun.foSearch for more papers by this authorJens-Erik Dessen, Jens-Erik Dessen orcid.org/0000-0002-0667-4183 Department of Animal and Aquaculture Sciences, Norwegian University of Life Sciences, Ås, Norway Nofima, Ås, NorwaySearch for more papers by this authorRegin Arge, Regin Arge orcid.org/0000-0001-5657-8352 Department of Animal and Aquaculture Sciences, Norwegian University of Life Sciences, Ås, Norway Fiskaaling, Hvalvík, Faroe IslandsSearch for more papers by this authorMagny Sissel Thomassen, Magny Sissel Thomassen Department of Animal and Aquaculture Sciences, Norwegian University of Life Sciences, Ås, NorwaySearch for more papers by this authorBjarne Hatlen, Bjarne Hatlen Nofima, Sunndalsøra, NorwaySearch for more papers by this authorKjell-Arne Rørvik, Kjell-Arne Rørvik Department of Animal and Aquaculture Sciences, Norwegian University of Life Sciences, Ås, Norway Nofima, Ås, NorwaySearch for more papers by this author Rúni Weihe, Corresponding Author Rúni Weihe ruw@havsbrun.fo orcid.org/0000-0002-2760-4352 Havsbrún, Fuglafjørður, Faroe Islands Department of Animal and Aquaculture Sciences, Norwegian University of Life Sciences, Ås, Norway Correspondence Rúni Weihe, Havsbrún P/F, Fuglafjørður, Faroe Islands. Email: ruw@havsbrun.foSearch for more papers by this authorJens-Erik Dessen, Jens-Erik Dessen orcid.org/0000-0002-0667-4183 Department of Animal and Aquaculture Sciences, Norwegian University of Life Sciences, Ås, Norway Nofima, Ås, NorwaySearch for more papers by this authorRegin Arge, Regin Arge orcid.org/0000-0001-5657-8352 Department of Animal and Aquaculture Sciences, Norwegian University of Life Sciences, Ås, Norway Fiskaaling, Hvalvík, Faroe IslandsSearch for more papers by this authorMagny Sissel Thomassen, Magny Sissel Thomassen Department of Animal and Aquaculture Sciences, Norwegian University of Life Sciences, Ås, NorwaySearch for more papers by this authorBjarne Hatlen, Bjarne Hatlen Nofima, Sunndalsøra, NorwaySearch for more papers by this authorKjell-Arne Rørvik, Kjell-Arne Rørvik Department of Animal and Aquaculture Sciences, Norwegian University of Life Sciences, Ås, Norway Nofima, Ås, NorwaySearch for more papers by this author First published: 29 January 2018 https://doi.org/10.1111/are.13598Citations: 10AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat Abstract The effects of isoenergetic diets with high (HP) and low (LP) protein-to-lipid ratios on feeding rate (SFR), feed conversion (FCR), growth (TGC) and relative- and absolute nutrient retention were investigated using both whole-body weight (BW) and carcass weight (CW) to assess the production efficiency. Three different feeding trials in seawater were conducted: two large-scale trials with yearling smolt (S1) and under-yearling smolt (S0) and one small-scale with S1 smolt. The initial body weights in the trials were 105, 319 and 978 g, respectively, and the fish were fed and monitored until they reached harvest weights. In all three trials, the dietary HP group attained significantly higher (p < .05) CW at harvest based on fish with equal BW. Also, fish fed the HP diets significantly improved FCR (p < .05) when based on CW. In the small-scale trial, fish fed HP diet, especially during late autumn and spring, significantly (p < .001) improved FCRBW and FCRCW. Improved FCR coincided with significantly higher (p < .05) relative energy retention in the dietary HP group. In all three trials, the HP groups had significantly higher (p < .05) TGC with regard to both BW and CW. Taken together, the present studies indicate that growth performance and feed utilization in modern salmon farming has the potential to be further improved by increasing the dietary protein-to-lipid ratio. In addition, dietary influence is more precisely assessed when using carcass as the weight denominator when analysing feed utilization and growth performance. 1 INTRODUCTION In modern aquaculture production of Atlantic salmon, the dietary protein-to-lipid ratio generally decreases inversely with increasing body weight. Small salmon, such as, parr and smolt, are usually fed a diet with relative high protein content (>40%) and low lipid content (<30%). The commercial practice, especially in Norway, has been to give the salmon high-fat diets (≥35% lipid, ≤35% protein) from a body weight of approximately 1 kg (grower diets), while the protein content is reduced so that protein-derived energy is spared in favour of fat. A historical retrospective from the Norwegian aquaculture industry displays an approximately four times increase in lipid inclusion in the feed for salmon since the start of the industry in the 1970s (Tacon & Metian, 2009; Torrisen et al., 2011). Thus, during the relative short lifespan of the salmon farming industry, the dietary protein-to-lipid ratio in the grower diets has changed from near 5 to 1. With a shift towards higher content of lipid, the feeds have necessarily become denser in energy. High-fat diets have previously been demonstrated to have beneficial effects on key production parameters such as growth rate and feed conversion (FCR) ratio (Hillestad, Johnsen, Austreng, & Åsgård, 1998; Karalazos, Bendiksen, & Bell, 2011; Karalazos, Bendiksen, Dick, & Bell, 2007). But studies have also indicated that high dietary fat intake may result in increased lipid content in both muscle and intestinal tissues of salmonids (Hillestad & Johnsen, 1994; Jobling, 1998, 2001; Jobling, Larsen, Andreassen, & Olsen, 2002; Refstie, Storebakken, Baeverfjord, & Roem, 2001). This may be undesirable since body lipids may act as a negative feedback signal on feed intake and thus impair growth (Johansen, Ekli, & Jobling, 2002; Johansen, Sveier, & Jobling, 2003; Silverstein, Shearer, Dickhoff, & Plisetskaya, 1999). Also, increased fat deposition in the visceral tissues may reduce the overall production yield. Salmonids are poikilothermic, which means that their feed intake and growth are highly influenced by water temperatures (Brett, 1979; Jobling, 1997). Both sea temperatures and day length vary throughout the year, and previous experiments have demonstrated that Atlantic salmon responds greatly to the seasonal changes with regard to energy demand, feed intake, nutrient retention and growth (Alne, Oehme, Thomassen, Terjesen, & Rørvik, 2011; Hemre & Sandnes, 2008; Lysfjord, Jobling, & Solberg, 2004; Måsøval et al., 1994; Mørkøre & Rørvik, 2001; Oehme et al., 2010). In general, these studies seem to depict a high production efficiency during the autumn, which coincides with decreasing day lengths and peak sea temperatures in the salmon producing countries situated in the North Atlantic Ocean such as Norway, the British Isles and the Faroe Islands. In general, it is a goal for all producers of animal proteins to increase utilization of feed resources. Thereto, a high turnover rate of production is crucial in most businesses. This is especially momentous in animal farming when the production area is limited. The Faroese aquaculture industry encounters significant limitations in biomass growth due to the relative limited coastline of the Faroe Islands (1,117 km), and virtually all potential farming areas are presently utilized. Currently, lack of well-established farming technology makes it difficult to farm salmon in exposed areas that surrounds the islands. Thus, the only realistic, short-term possibility for biomass increase for the Faroese aquaculture industry is through higher growth rate of salmon (shorter production cycle from sea transfer to harvest) and increased carcass-to-body weight yield. Since final carcass is the primary tradable commodity, carcass weight and not only body weight, should be considered as the weight denominator when evaluating the dietary effects on FCR and growth performance. Thus, using the carcass weight as a biometric measurement of dietary effects, a more complete picture, both nutritional and economical, may be achieved when assessing overall feed efficiency in salmon production. Previous experiments have displayed high carcass-to-body weight yields (≥90%) (Einen & Roem, 1997; Hillestad & Johnsen, 1994; Hillestad et al., 1998; Wathne, 1995). Although there might be a lack of detailed definition of carcass weight in these studies, these results may indicate that the carcass-to-body weight ratio has been somewhat higher compared with some of the yields (∼ 83%) recently observed in the industry (Waagbø et al., 2013). Therefore, it may be questioned whether the changes seen in the dietary protein-to-lipid composition have been in favour of obtaining high carcass growth throughout the marine production phase of salmon. In this context, diets with low protein-to-lipid ratios may not utilize the full potential of carcass growth in salmon, and thus the industry has not been assessing what protein-to-lipid composition is needed to achieve a more optimal production throughout the whole seawater phase, especially in the grow-out phase from approximately 1 kg until harvest. During this phase of production, the dietary protein-to-lipid ratio is at the lowest, however, most of the weight gain is generated as the fish is harvested between 4 and 6 kg (Nystøyl, 2017). The aim of the present work was, consequently, to examine the effects of different dietary protein-to-lipid ratios on feed utilization and fish growth rate using both whole-body weight and carcass weight in assessing the feed effects on overall production efficiency. In addition, the effect of seasonal influence on biometric performance was examined together with the potential interaction of dietary effects. 2 MATERIALS AND METHODS 2.1 Experimental design Three dietary high protein-to-lipid ratio (HP) and three lower protein-to-lipid ratio (LP) feeding strategies were first tested in two different commercial large-scale farming sites in the Faroe Islands with yearling (S1) and under-yearling smolt (S0) following a small-scale trial which was conducted in Norway using S1 smolt. In all three experiments, the protein and lipid contents in the LP diets were designed to resemble those of a typical commercial diet for the respective sizes of fish, whereas the HP diets had higher protein and lower lipid contents. The total energy from lipid, protein and carbohydrates were targeted to be equal in the HP and LP diets for each pellet size. Compared with large-scale feeding experiments in commercial conditions in general, small-scale trials ensure more accurate measurements of feed intake, biomass and equal slaughter time. Therefore, the present small-scale trial was conducted to test the reproducibility and validity of the dietary influences as well as to complement the observations from the large-scale experiments with a more scientific approach with regard to feed intake, feed utilization and dietary retention of nutrients. 2.2 Experimental diets All feeds were produced by Havsbrún (Fuglafjørður, Faroe Islands). Multiple batches of feed were produced throughout the large-scale experimental period and two feed batches per dietary treatment were produced for the small-scale trial (Table 1). The main dietary raw materials used in the large-scale experiments, ranked from highest to lowest inclusion level, were fishmeal, fish oil, wheat, soy protein concentrate, wheat gluten and sunflower meal. In the small-scale experiment, the ingredients used were, fishmeal, fish oil, rapeseed oil, wheat, krill meal and porcine blood meal. For all three trials, premixes containing pigments, minerals and vitamins were included in the diets to fulfil the minimum nutritional requirements in accordance with the National Research Council (1993, 2011). The estimated feed digestibility was calculated in compliance with Morris et al. (2003) assuming apparent digestibility coefficients for protein and lipid to be 0.86 and 0.94 (Einen & Roem, 1997), respectively, and 0.50 for nitrogen-free extractives (NFE) (Arnesen & Krogdahl, 1993). The feed production process included standard manufacturing routines regarding the control of physical pellet quality as well as the monitoring and control of proximate feed composition. Table 1 states the proximate composition of the experimental diets. These were based on the weighted mean from each feed batch supplied to the fish farming sites. The 3 and 4 mm HP diets in the S1 large-scale were intended to be the same (52% protein and 24% lipid). The relative large compositional deviation of the 3 mm HP feed was caused by manufacturing problems in addition to wrongful handling of feed during transport, which resulted in the dietary HP fish group being supplied with some 3 mm LP feed instead of HP feed. Thus, the dietary HP group was fed a combination of both HP and LP feed for approximately 4 weeks. Table 1. Proximate feed compositions (wet weight) used in all three experiments. Brackets demonstrate the number of feed batches used in the experiment per pellet size per dietary treatment. Values are given as weighted means per diet. HP: dietary high protein-to-lipid ratio strategy. LP: dietary low protein-to-lipid ratio strategy Smolt group Large-scale S1 Large-scale S0 Small-scale S1 Diet HP LP HP LP HP LP Pellet size 3 mm (n = 4) (n = 2) Dry matter, % 93.3 ± 0.1 93.1 ± 0.2 Crude protein, % 49.9 ± 0.7 46.6 ± 0.3 Lipid, % 25.6 ± 1.4 27.2 ± 0.2 Ash, % 9.4 ± 0.5 8.7 ± 0.2 Starch, %a 6.7 ± 0.1 8.6 ± 0.2 DP. %b 42.4 ± 0.6 40.0 ± 0.2 DE, MJ/kgb 20.3 ± 0.4 20.5 ± 0.0 DP:DE, g/MJb 20.9 ± 0.7 19.5 ± 0.1 Protein-to-lipid ratio 1.95 1.71 Pellet size 4 mm (n = 5) (n = 2) Dry matter, % 94.1 ± 0.1 93.4 ± 0.2 Crude protein, % 52.1 ± 1.4 45.8 ± 0.3 Lipid, % 22.1 ± 1.8 28.7 ± 0.6 Ash, % 11.0 ± 0.2 8.6 ± 0.3 Starch, %a 6.9 ± 0.2 8.7 ± 0.3 DP. %b 44.8 ± 1.2 39.4 ± 0.3 DE, MJ/kgb 19.6 ± 0.4 20.9 ± 0.2 DP:DE, g/MJb 22.9 ± 1.0 18.9 ± 0.3 Protein-to-lipid ratio 2.36 1.60 Pellet size 6 mm (n = 7) (n = 2) (n = 2) (n = 7) Dry matter, % 95.6 ± 0.1 94.2 ± 0.1 94.1 ± 0.3 93.9 ± 0.2 Crude protein, % 46.6 ± 0.5 41.9 ± 0.2 44.4 ± 0.3 42.7 ± 0.5 Lipid, % 27.6 ± 0.4 32.4 ± 0.2 30.8 ± 0.7 31.6 ± 0.4 Ash, % 9.5 ± 0.4 8.1 ± 0.2 8.2 ± 0.2 7.8 ± 0.1 Starch, %a 8.6 ± 0.7 8.9 ± 0.0 8.3 ± 0.4 9.0 ± 0.0 DP. %b 40.1 ± 0.5 36.1 ± 0.2 38.2 ± 0.3 36.7 ± 0.5 DE, MJ/kgb 20.8 ± 0.1 21.6 ± 0.1 21.4 ± 0.2 21.5 ± 0.1 DP:DE, g/MJb 19.3 ± 0.1 16.7 ± 0.1 17.9 ± 0.3 17.1 ± 0.2 Protein-to-lipid ratio 1.69 1.29 1.44 1.35 Pellet size 9 mm (n = 71) (n = 10) (n = 20) (n = 10) (n = 2) (n = 2) Dry matter, % 93.7 ± 0.2 94.1 ± 0.1 94.0 ± 0.2 94.2 ± 0.1 94.1 ± 1.0 94.3 ± 0.5 Crude protein, % 42.0 ± 0.2 35.4 ± 0.1 40.2 ± 0.3 34.5 ± 0.2 42.7 ± 0.1 35.4 ± 0.4 Lipid, % 32.6 ± 0.2 35.9 ± 0.1 34.4 ± 0.2 35.8 ± 0.2 32.1 ± 0.7 36.0 ± 0.6 Ash, % 8.1 ± 0.1 6.4 ± 0.1 8.0 ± 0.1 6.7 ± 0.1 7.9 ± 0.2 7.1 ± 0.2 Starch, %a 8.4 ± 0.2 9.6 ± 0.1 9.1 ± 0.1 9.8 ± 0.8 8.5 ± 0.2 11.0 ± 0.4 DP. %b 36.1 ± 0.1 30.4 ± 0.1 34.6 ± 0.3 29.6 ± 0.2 36.7 ± 0.1 30.4 ± 0.3 DE, MJ/kgb 21.6 ± 0.1 22.0 ± 0.0 21.9 ± 0.1 21.8 ± 0.1 21.6 ± 0.3 21.9 ± 0.3 DP:DE, g/MJb 16.7 ± 0.1 13.9 ± 0.1 15.8 ± 0.1 13.6 ± 0.1 17.0 ± 0.2 13.9 ± 0.0 Protein-to-lipid ratio 1.29 0.99 1.17 0.96 1.33 0.98 a Starch content was not analysed in all feed batches. The stated value is the average of the analysed batches. b Digestible protein and digestible energy were calculated, based on the measured proximate feed composition, assuming 23.7, 39.5 and 17.2 MJ per kg of protein, lipids and nitrogen-free extractives (NFE) respectively. The apparent digestibility coefficients used for protein, lipid and NFE in Atlantic salmon were 0.86 (Einen & Roem, 1997), 0.94 (Einen & Roem, 1997) and 0.50 (Arnesen & Krogdahl, 1993). 2.3 Fish and facilities—large-scale trials In the large-scale S1 trial, salmon smolt were supplied by Bakkafrost hatchery station in Glyvradalur and transferred to the Bakkafrost commercial seawater site at Lambavík (62°08′N, 06°41′W), Faroe Islands, during April 2009. Duplicate 128 m circumference cages with a water volume of 18,500 m3 were used for rearing the fish per dietary treatment. Mean number of fish per net pen was 66,627 (SEM = 213). The fish were subjected to 1000 W artificial light (L:D 24:0) from 10 December 2009 until 21 March 2010. We identified an error regarding the body weight measurement of the stocked fish 5 months after the trial initiation which caused unequal starting weights between the dietary treatments, showing that the dietary LP group was 8% bigger (LP = 104 ± 10 g vs. HP = 96 ± 2 g, n = 2). To achieve equal starting weights per dietary treatment, a triplicate cage, also fed HP diet since sea transfer, was included. This was considered necessary to achieve reliable data to examine dietary influence based on comparable fish groups with equal starting weights. Thus, mean body weight at sea transfer for the fish group fed the LP diet was 104 g (SEM = 10, n = 2) vs. 105 g (SEM = 10, n = 3) after adjustment of the HP fed smolt group. Feeding of the fish in the experimental cages started in week 19 (May 2009). There was a great algal bloom during the period July–August 2009 at the S1 trial site causing a severe decrease in feeding rate within both dietary treatments. The average sea water temperature through the S1 experimental period was 8.5°C with a maximum and minimum of 11.1°C and 5.7°C respectively (Figure 1a). Salmon-fed HP feed had an average production period of 452 ± 11 days and 3,752 ± 109 day degrees, whereas the production duration of the dietary LP group was 477 ± 27 days and 3,971 ± 266 day degrees. Figure 1Open in figure viewerPowerPoint (a) Weekly seawater temperature (°C) for the large-scale S0 trial (solid black line) and the large-scale S1 trial (broken black line) displayed on the y-axis. (b) Daily seawater temperature (solid black line) during the S1 small-scale experiment is displayed on the y-axis where the sampling periods that identify the three feeding periods are noted above the figure. Average day length per week (hr) for the large- and small-scale experiments are illustrated with broken grey line displayed on the z-axis S0 smolt from Luna's hatchery station in Fútaklettur had been transferred to Luna′s commercial sea farming site in Sørvágur (62°04′N, 7°20′W), Faroe Islands, in October 2008. In March 2009, when the feeding trial started, the fish had a mean body weight of 319 g (SEM = 5, n = 4) with a mean number of 60,392 fish per cage (SEM = 245). Duplicate cages per dietary treatment of 24 m × 24 m, with a water volume of 6,912 m3, were used in the beginning of the trial. In June 2009, all the fish were transferred by towing the cages approximately 1 km southwest across the fjord (62°04′N, 07°22′W) and restocked in 128 m circumference cages with a water volume of 18,500 m3, maintaining the same experimental groups. The transportation time was approximately 3.5 hr per cage. The S0 experimental fish were subjected to 1,000 W artificial light (L:D 24:0) from 14 December 2009 until 15 March 2010. The average sea water temperature through the S0 experimental period was 8.4°C where the peak temperature was 10.7°C and the lowest temperate was 5.8°C. The average production period for the dietary HP group was 429 ± 6 days and 3,597 ± 42 day degrees whilst the dietary LP group had a production period of 439 ± 11 days and 3,688 ± 97 day degrees respectively. Figure 1a gives an overview of the temperature and day length in both large-scale trials. Four different pellet sizes were used within the dietary treatments in the S1 large-scale experiment, whereas two pellet sizes were used within the dietary treatments in the S0 large-scale trial (Table 2). The pellet sizes were adjusted to fit the fish weight according to the guidelines of the feed manufacturer. Table 2. Overview of the feeding period for each pellet size within both dietary treatments in the large-scale trials. The pellet sizes are fed in relation to the preferred fish weight intervals which is also given Pellet size used (mm) Preferred fish weight (g) First feed delivery Feeding period Large-scale S1 HP 3 ~ 100–150 07.04.2009 9 weeks (week 15–week 24) 4 ~ 150–300 16.06.2009 11 weeks (week 24–week 35) 6 ~ 300–800 28.08.2009 6 weeks (week 35–week 41) 9 ~ 800+ 08.10.2009 44 weeks (week 41–week 33) LP 3 ~ 100–150 27.03.2009 10 weeks (week 13–week 23) 4 ~ 150–300 04.06.2009 11 weeks (week 23–week 34) 6 ~ 300–800 18.08.2009 7 weeks (week 34–week 41) 9 ~ 800+ 19.10.2009 49 weeks (week 41–week 38) Large-scale S0 HP 6 ~ 300–800 18.03.2009 16 weeks (week 12–week 28) 9 ~ 800+ 09.07.2009 35 weeks (week 28–week 21) LP 6 ~ 300–800 04.03.2009 20 weeks (week 10–week 30) 9 ~ 800+ 26.06.2009 39 weeks (week 26–week 23) 2.4 Fish and facilities—small-scale trial The small-scale experiment with S1 post-smolt was conducted at Nofima's research station at Ekkilsøy (currently owned by Marine Harvest Fish Feed AS) on the west coast of mid-Norway (63°03′N, 07°35′E) in 2012. In all, 150 post-smolt salmon weighing 978 g (SEM = 1, n = 6) were randomly distributed in each of six cages measuring 5 m × 5 m × 5 m. Prior to this, the fish had been transferred to sea as yearling (S1) smolt (95 g) in April 2012 from Salmar's hatchery station in Straumsnes, and then been involved in an earlier feeding trial (Dessen, Weihe, Hatlen, Thomassen, & Rørvik, 2017) and fed the same high-protein diets through three different periods from April to September. During the last period from 23 July to 24 September in this pre-trial, the post-smolt grew 658 g, ending up with a final body weight of 926 g and a whole-body composition of 17.6% protein and 16.0% fat. The experimental diets (HP and LP 9 mm, Table 1) used in the small-scale trial were fed to triplicate groups of fish from 27 September 2012 until trial termination on 10 June 2013. The trial was split into three feeding periods representing three different seasons; 1: 27 September–4 December (late autumn), 2: 7 December–8 April (winter) and 3: 11 April–10 June (spring) respectively (Figure 1b). Fish were fed to satiation daily using automatic feeders four times a day from 27 September to 25 October. Subsequently, until trial termination in June, the fish were fed three rations per day. The daily feed rations were approximately 10% in excess of the feed eaten the day before. Waste feed was collected daily as described by Einen, Mørkøre, Rørå, and Thomassen (1999) and analysed for recovery of dry matter as described by Helland, Grisdale-Helland, and Nerland (1996). The average sea water temperature in the three experimental periods was 9.4°C (612 day degrees), 4.1°C (490 day degrees) and 7.1°C (427 day degrees) respectively. Figure 1b illustrates the changes in temperature and day length during the small-scale trial. 2.5 Sampling procedures large-scale Fish from the experimental cages were harvested following standardized routines of the farming respective companies (Bakkafrost and Luna). This included a starvation period of 3–5 days prior to slaughter, and the average harvesting time per cage in the S1 and S0 trials was two and 4 weeks respectively. In the S1 large-scale trial, the fish were transported with well boat to the Bakkafrost harvesting facilities in Klaksvík (62°23′N, 06°59′W) during the period from week 28 (July) to week 41 (November) 2010. The experimental S0 fish were harvested at Luna′s harvesting facility in Sørvágur (62°07′N, 07°32′W) from week 17 (April) to week 25 (June) 2010 after dragging the experimental cages approximately 2 km from the production site to the harvesting facilities at the head of the fjord. At both harvesting facilities, the salmon were killed and bleed using an automated swim-in system (SI-7 Combo, killing and bleeding machine) and subsequently transported to a bleeding tank with a water temperature between 0°C and −1°C to bleed out. At the first day of slaughter of each experimental cage in the S1 trial, 30 fish were sampled and divided into three weight classes á 10 fish of 4.5 kg, 5.5 kg and 6.5 kg average weight respectively. All the sampled fish were handpicked from the bleeding tank at the harvesting facilities. In one experimental unit (cage no. 4) in the large-scale S1 trial fed HP feed, only 10 fish, respectively, of 4.5 kg and 5.5 kg were sampled. In the S0 experiment, 30 fish from all experimental cages were sampled 8 April (week 14), and divided into the mentioned weight classes. All samples in both large-scale experiments were recorded and measured for body weight, length and carcass weight. Carcass weight was defined as the weight after removal of blood, viscera, heart and kidneys. The measured body weights were corrected for 2.7% blood loss in accordance with Einen, Waagan and Thomassen (1998) to calculate live weight at slaughter. During the harvest period, the total number of fish and gutted biomass was recorded and harvest reports were generated for each experimental unit and the body weight of fish and biomass within each cage was calculated. We chose to use the carcass-to-body weight ratio per cage, measured at first day of harvest, to convert the carcass weights in the harvest reports to whole-body weight and biomass within each experimental cage. The harvest reports depict a difference within the smolt groups regarding the number of production days in the experimental units and thus a difference in day degrees were used to achieve about the same body weight within dietary treatments at harvest. Table 3. Differences in specific feeding rate (SFR), feed conversion (FCR) and growth rate (TGC) based on live body weight (BW) and carcass weight (CW) in S1 and S0 Atlantic salmon in the large-scale experiments. Significant differences between dietary treatments (D) and smolt group (SG) and the interaction (D × SG) in the two-way ANOVA are given whilst the values in brackets depict statistical trends, and non-significant differences are highlighted as ns. Dietary statistics within smolt group is visualized by p Smolt group S1 S0 Two-way ANOVA Dietary group HP (n = 3) LP (n = 2) p HP (n = 2) LP (n = 2) p D SG D ×SG R 2 SFR 0.55 ± 0.01 0.56 ± 0.02 ns 0.51 ± 0.02 0.52 ± 0.02 ns ns 0.03 ns .50 FCRBW 1.29 ± 0.03 1.36 ± 0.03 ns 1.21 ± 0.03 1.25 ± 0.02 ns (.06) 0.01 ns .73 FCRCW 1.47 ± 0.04 1.57 ± 0.01 ns 1.40 ± 0.02 1.47 ± 0.03 ns .03 0.04 ns .67 TGCBW 3.18 ± 0.04 2.98 ± 0.07 (.06) 3.16 ± 0.03 3.09 ± 0.09 ns .04 ns ns .46 TGCCW 3.05 ± 0.03 2.84 ± 0.07 (.06) 2.99 ± 0.03 2.91 ± 0.09 ns .02 ns ns .59 2.6 Sampling procedures small-scale At the end of each feeding period (Figure 1b), all fish within each experimental unit were anaesthetized (MS 222 metacaine 0.1 g/L, Alpharma, Animal Health, Hampshire, UK) and bulk-weighed for determination of specific feeding rate (SFR), growth rate (presented as thermal growth coefficient, TGC) and FCR. When sampling fish in the first two periods, 10 fish representing the average body weight in each unit were stunned with a blow to the head and bled out. These fish were then individually weighed, length measured and gutted, and carcass weight registered. In line with the large-scale trials at trial termination, 30 fish from each cage were collected and divided into three weight classes. Because the experimental fish did not grow as big as the fish in the large-scale trials, the three groups of 10 fish were divided in subgroups of 2.4, 3.2 and 4.0 kg. Also, an additional 10 fish (not bled) representing the mean body weight per experimental unit were sampled for whole-body analysis of protein, fat and energy. The fish were starved for 4 days prior the sampling in December, whereas the fish were starved for 3 days prior to the samplings in April and June. At each sampling, all fish with obvious signs of wounds, runts or sexual maturity were removed (weights and number of these fish was recorded). 2.7 Feed chemical analyses In all three experiments, the feeds were analysed for moisture (drying loss at 103°C to stable weight; ISO 6496), ash (combustion at 550°C, ISO 5984), crude protein (N × 6.25, combustion according to the Kjeldahl principle, ISO 5983) and crude fat was analysed using pre-extraction and post-extraction in petroleum ether after HCL hydrolysis (98/64/EC). In the large-scale trials, total- and gelatinized starch was analysed as d-glucose following enzymatic cleavage with gluco-amylase after full gelatinization by cooking with NaOH. In the small-scale trial, the total starch content was analysed as glucose after enzymatic hydrolysis employing the Megazyme K-TSTA 07/11 kit (Megazyme International, Ireland) in accordance with AOAC method 996.11. The ener
Referência(s)