Providing evidence to improve practice

Action: Use an alternative oil source: plant-based Sustainable Aquaculture

Key messages

  • Three replicated studies (two controlled) in Norway found growth rates were similar in salmon that were fed diets containing fish oil and vegetable oil. One replicated study in Norway found growth rates were higher in fish fed diets containing vegetable oil.  One replicated, controlled study found salmon growth rates were both lower and higher in vegetable oil diets compared to fish oil diets, dependant on family genetics.
  • Two replicated studies (one controlled) in Norway found similar average final body weights between groups of salmon fed both fish oil and vegetable oil diets.
  • Three studies (two replicated, one controlled) in Norway and Scotland found that the fatty acid profile of salmon flesh reflected oil source within diets.
  • A study in Norway found that oil source in diets did not affect salmon broodstock fecundity levels, egg weights, fertility rates, as well as the weights and development of resultant fry.
  • One replicated Norwegian study found that salmon fed vegetable oil diets had high liver lipid and low plasma lipoprotein compared to the fish oil diet. One replicated, controlled study found high levels of n-3 highly unsaturated fatty acids compared to diets containing rapeseed oil.
  • A replicated, controlled study in Scotland found salmon fed vegetable oil-based diets had lower concentrations of dioxin and polychlorinated biphenyls within flesh, compared with diets containing fish oil.
  • A replicated study in Norway found that fresh, frozen and smoked salmon flesh from fish fed vegetable oil- and fish oil-based diets had similar levels of gaping, texture and liquid holding capacity. Pigment concentration was lower in vegetable oil diets.


Supporting evidence from individual studies

1 

In 2002, a randomised, replicated study in Norway (Ng et al., 2004) found there were similar growth rates and feeding efficiency in salmon, Salmo salar, fed one of two experimental diets containing different blends of fish oil, palm oil and rapeseed oil. Both experimental diets contained 50 % fish oil with either 10 % or 25 % palm oil. Rapeseed oil made up the remainder of the oil in the diet. Over 12 weeks, 880 salmon were fed one of the experimental diets. Growth rates and feeding efficiency were measured.

 

2 

Between 2000 and 2002, a replicated, controlled study  in Scotland (Bell et al., 2005) found salmon, Salmo salar, fed diets containing low fish oil, and low-high vegetable oil concentrations had lower levels of dioxin and polychlorinated biphenyls within flesh. This was in comparison to feeds containing high fish oil concentration. The highest dioxin and PCB flesh concentrations in salmon fed the high fish oil diet was 0.53 ng TEQ per kg and 1.48 ng TEQ per kg, respectively. The lowest dioxin and PCB flesh concentrations in salmon fed the high vegetable oil diet was 0.10 ng TEQ per kg and 0.58 ng TEQ per kg, respectively. Over 115 weeks, salmon were fed one of four diets comprised of low fish oil (17%), high fish oil (35%), low vegetable oil (17% linseed and rapeseed oil) and high vegetable oil (35% linseed and rapeseed oil).

3 

Between 2002 and 2003, a study in Norway (Rennie et al., 2005) found that salmon, Salmo salar, broodstock fed diets containing pure fish oil and a mix of fish and rapeseed oil had similar levels of adult fecundity, egg weights plus development and body weight in resultant fry.  The fatty acid profiles of both the eggs and fry reflected the oil source and were different. Rates of fertilisation, eyeing and hatching plus fry survival to first feeding were similar between diets. Approximately 315 salmon were tagged and distributed across three sea cages. One cage was fed a broodstock diet containing 100 % fish oil. The other two cages were fed a diet containing 50% fish oil and 50% rapeseed oil. Fish were weighed intermittently. Fecundity and egg weights were measured in mature salmon transferred to fresh water. Eggs were fertilised using a pool of milt from three males of the same treatment. Pre-fertilised eggs and fry underwent fatty acid chemical analysis. Rates of fertilisation were recorded, alongside eyeing, hatching and survival rates in the fry.

4 

A randomised, replicated study in Norway (Rora et al., 2005) found similar growth rates in salmon, Salmo salar, fed diets containing either 29% fish oil or 29% soybean oil. Muscle fatty acid profile reflected dietary oil source with malondialdehyde being four times higher in the fish oil diet. Pigment concentration was lower in salmon fed the soybean oil diet. Gaping, texture and liquid holding capacity of fresh, frozen and smoked muscle was similar between diets. A consumer panel detected no differences between dietary treatments and end products. Three groups of 400 salmon were fed one of two experimental diets for 120 days. Diets contained either 29% fish oil or 29% soybean oil and were identical in composition otherwise. Chemical analyses were conducted on fresh muscle. Colour, texture and LHC analyses were performed on fresh, frozen and smoked muscle.

5 

Between 2002 and 2004, a replicated, controlled study comprising two trials in Scotland and Norway (Torstensen et al., 2005) found salmon, Salmo salar, fed fish oil and vegetable oil diets had different fatty acid compositions within flesh that reflected oil source. However, final body weights were similar (2.5kg on average). Growth rates were similar between diets. It was calculated a 200 g portion of salmon fed 75% vegetable oil would meet 80% of the recommended weekly human intake for very long chain n-3 polyunsaturated fatty acids. Groups of salmon juveniles were fed diets containing 100% fish oil, 75% vegetable oil or 100% vegetable oil, followed by a finishing diet period when all groups were fed 100% fish oil. The vegetable oil was composed of blended rapeseed, palm and linseed oil. The trials were conducted over 22 months in Norway and 25 months in Scotland.

6 

Between 2002 and 2004, a replicated study in Norway (Jordal et al., 2007) found that liver lipid and plasma lipoprotein levels were affected when salmon, Salmo salar, were fed vegetable oil-based diets as a complete replacement for fish oil. Salmon weights were similar and increased from 890g to 2.3kg and 2.7kg in the fish oil and vegetable oil groups, respectively. Lipid liver stores were higher in fish fed the vegetable oil diet after 14 and 22 months of feeding. In contrast, plasma lipid levels were lower. Over 22 months, 2,000 salmon juveniles were fed diets containing either 100% fish oil or 100 % vegetable oil (comprising 55 % rapeseed oil, 30 % palm oil and 15 % linseed oil). Fish meal was used as a protein source. Fish were weighed at 0, 6, 9, 14 and 16 months. Liver and plasma were analysed for lipid and lipoprotein content.

7 

In 2005, a randomised study in Scotland (Schlechtriem et al., 2007) found that salmon, Salmo salar, fed diets containing vegetable oil had lower proportions of n-3 highly unsaturated fatty acids than diets containing fish oil. Flesh from salmon fed the fish oil diets contained 26.0% of n-3 highly unsaturated fatty acids compared to 17.7% in vegetable oil fed fish. HUFA is essential for protecting humans from cardiovascular disease. Over 12 weeks, groups of 30 salmon were fed diets containing either 100% fish oil or vegetable oil (comprised of rapeseed, linseed and palm oils at a 2:2:1 ratio). Flesh was analysed for n-3 highly unsaturated fatty acid content.

8 

A replicated, controlled study in Norway (Pratoomyot et al., 2008) found salmon, Salmo salar, fed fish oil, decontaminated fish oil and different vegetable oil diets had similar growth rates and final average body weights. In terms of n-3 highly unsaturated fatty acids, essential to humans for protecting against cardiovascular disease, soybean-based vegetable oil produced higher levels in salmon flesh compared to rapeseed oil. One of five experimental diets were fed to groups of 120 salmon for 10 weeks: 100% fish oil, 100% decontaminated fish oil and three different of vegetable/fish oil blends (using soybean or rapeseed oil). Growth rates and n-3 highly unsaturated fatty acid content were measured.

9 

In 2010, a replicated study in Norway (Bell et al., 2010) found that salmon, Salmo salar, with different breed characteristics (fat and lean flesh) had different growth rates and flesh lipid content from each other and the control group, dependant on whether they were fed fish or vegetable oil diets. Growth rates ranged from 0.89 in the fat group fed the 100% vegetable oil diet to 1.01 in the control group fed the same diet. Flesh lipid levels were in the order fat>control>lean in diets containing 100% fish oil and fat=lean>control in diets containing 100% vegetable oil. It was calculated that a 140g portion of fish oil-fed salmon would provide 63–76% of the recommended weekly human intake of n-3 PUFA, while the vegetable oil-fed salmon would provide 46–61% of this value. Over 55 weeks, three groups of trait-bred salmon smolts were grown on diets with reduced fish meal that contained either 100% fish oil or vegetable oil (comprised of rapeseed, palm and Camelina oils in a ratio of 5:3:2). After a 15 week finishing diet, flesh lipid content was analysed. Growth rates were also measured over the course of the study.

10 

A replicated study in Norway (Karalazos et al., 2011) found that salmon, Salmo salar, fed diets containing rapeseed oil had higher growth rates and average final body weights compared to those fed diets containing fish oil. Growth rates were 0.94 and 0.87, respectively.  Average final weights were 3.64kg and 3.34kg, respectively. Protein content did not affect growth parameters. Six experimental diets were fed to salmon over a 10 week period. Three contained high, medium and low protein content and 100% fish oil. Three contained high, medium and low protein content and 100% rapeseed oil. The protein/lipid contents were high: 350g per kg/350g per kg, medium: 330g per kg/360g per kg and low: 290g per kg/380g per kg. Growth rates and final body weights were measured.

Referenced papers

Please cite as:

Jones, A.C., Mead, A., Austen, M.C.V.  & Kaiser, M.J. (2013) Aquaculture: Evidence for the effects of interventions to enhance the sustainability of aquaculture using Atlantic salmon (Salmo salar) as a case study. Bangor University