Providing evidence to improve practice

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

Key messages

 

Supporting evidence from individual studies

1 

A replicated, controlled study in Norway comprised of two salmon, Salmo salar, feed trials (Opstvedt et al., 2003). In trial one the digestibility of crude protein from fish meal, full-fat soybean meal and maize gluten meal were similar at 89.3%, 76.9% and 87.0%, respectively. In trial two, increasing levels of fish meal substitution by plant-based protein led to a decrease in growth rates, protein digestibility and salmon condition. Additionally, the feed conversion ratio increased. Therefore salmon that were fed diets containing the highest levels of plant protein required 50g more feed to gain 1kg of weight. Trial one was conducted over 14 days. Groups of salmon were fed one of four different diets: a reference diet containing 89% fish meal and three experimental diets where 45 % of the reference diet was substituted by fish meal, full-fat soybean meal and maize gluten meal. Digestibility of protein was determined. Trial two was conducted over 104 days. Three diets were formulated that fish meal with a combination of full-fat soybean meal and maize gluten meal. Fish meal derived crude protein content in the three diets was 89%, 65% and 40% respectively. Eighty salmon were present in each treatment tank and were fed by automatic feeders under controlled environmental conditions. Growth rates and digestibility were measured.

 

2 

A randomised, controlled study in Norway (Refstie et al., 2003) found that salmon, Salmo salar, fed diets containing a mix of fish-meal, wheat and low SGA potato protein concentrate grew at similar rates to salmon that were fed diets of fish meal and wheat. At the end of the experiment average final body weights ranged from 249-256g across all groups.  In addition, food intake and feed conversion ratio were not adversely affected. Over 84 days, groups of salmon were fed one of four experimental diets: 0% SGA potato protein concentrate (fish meal control diet) and increments of 7%, 14% and 21% SGA potato protein concentrate. Fish were fed using automated feeders under controlled environmental conditions. Growth rate was measured.

 

3 

A randomised, replicated study in Norway (Mundheim et al., 2004) found lower final average body weights when salmon, Salmo salar, were fed diets containing reduced fish meal content, replaced with a vegetable protein blend. Salmon fed 34.7% fish meal were 13.3% lighter than those fed 85.1% fish meal. Growth rates, feeding efficiency and digestibility of protein also decreased with increasing percentages of vegetable protein. Feed intake was higher when food contained a lower proportion of fish meal. Salmon were fed eight diets (four high and four lower quality fish meal) where the percentage of fish meal to vegetable protein was either 85.1%, 68.6%, 51.9% or 34.7%. The vegetable protein blend was composed of full-fat soybean meal and maize gluten meal at a ratio of 1:2. Fish were fed for 11 weeks using automated feeders under controlled environmental conditions. Final body weight, growth rates, feeding efficiency and protein digestibility were measured.

 

4 

In 2002, a randomised, replicated, controlled study in Norway (Espe et al., 2006) identified slower growth rates in salmon, Salmo salar, fed three plant-based protein diets compared to a control diet containing fish meal. This was attributed to a reduced intake of the plant-based feed. Feed conversion ratios did not differ. Flesh fat content was higher in salmon fed the fish meal-based diets. Over three months, groups of 40 salmon were fed one of four diets: a reference diet containing 49% fish meal and three diets containing 5% solubilised fish protein and different levels of plant protein coming from plants. Growth rates were 0.97, 0.87, 0.86 and 0.87, respectively. Salmon were fed using automated feeders. Growth rates, feed conversion ratios and flesh fat content were measured.

5 

A randomised, replicated, controlled study in Canada (Gill et al., 2006) found similar growth and survival rates and feed efficiency between groups of fingerling salmon, Salmo salar, fed diets containing fish meal compared with diets containing sunflower meal. Growth rates were, on average, 1.39-1.45% per day. Average survival rates were 96.0-99.3% and average feed efficiency was 1.19-1.26. Groups of 50 fingerling salmon were fed one of four experimental diets containing 8.25%, 16.5%, 24.75% or 33% of sunflower meal. A control diet was given to a fifth group of fish containing 68.2% low-temperature dried anchovy (fish) meal. Squid meal (70g per kg) and FinnstimTM (10g per kg) were incorporated into each diet to reduce problems associated with palatability. Salmon were fed for 84 days and were weighed on day 0, 42 and 84.

 

6 

Between 2002 and 2003, a randomised, replicated study in Scotland (Young et al., 2006) found that diets containing maize- and pea-based protein plus carbohydrates (to replace fish oil) achieved similar growth rates in salmon smolts, Salmo salar. Over 377 days, fish across all groups grew from an average 52g to between 1.9kg and 2.5kg. Salmon smolts fed low-energy feeds (containing less fish oil and more carbohydrate) had a lower body condition compared with those fed high-energy feeds (containing more fish oil and less carbohydrates). Nine experimental diets were produced containing different levels of vegetable protein, carbohydrate and fish oil. Groups of salmon smolts were fed one of the experimental diets for 377 days during their marine grow-out phase. Fish were fed using automated feeders. Growth rates, weights and body condition were measured.

 

7 

A replicated, controlled study in Norway (Øverland et al., 2009) found similar levels of weight gain and feed intake in salmon, Salmo salar, fed diets containing fish meal, soybean meal or pea protein concentrate mixed with crude protein. Weight gain ranged from 228g to 274g in salmon fed the control and vegetable protein diets. Levels of protein, fat, starch and essential amino acid digestibility were similar between the control and pea protein diets. However, the soybean meal diet had a reduced digestibility in comparison and induced morphological change in the distal intestine. Over twelve weeks, 600 salmon were fed the experimental diets using automated feeders.  One group was fed a control diet based on high quality fish meal. The other diets contained either 200g per kg soybean meal or 200g per kg pea protein concentrate mixed with 350 or 500g per kg crude protein. Final body weights and digestibility of feed components were measured.

 

8 

Between 2007 and 2008, a randomised, replicated and controlled study in Scotland (Pratoomyot et al., 2010) found salmon, Salmo salar, fed three plant-based protein diets had lower growth rates and final body weights compared to those fed a high fish meal content diet. Growth rates were reduced by between 5-23% in the plant-based protein diets when compared to the high fish meal diet.  After 19 weeks, the overall average weight gain within the plant-protein diets were between 1.67-2.08kg compared to 2.53kg in the high fish meal diet. Differences were attributed to decreased food intake. Over 19 weeks 1,800 salmon were fed a high fish meal diet (55%) or one of three experimental plant protein diets, containing between 50% and 60% plant protein. Growth rates and final body weights were measured.

 

9 

A replicated, controlled study in Norway (Penn et al., 2011) found lower growth rates and weight gain in salmon (Salmo salar) fed diets containing plant-based protein sources compared to a fish meal diet (control). Salmon fed the maize and pea protein diets showed lower specific growth rates of 0.53 and 0.56, compared to the control diet (0.63). Feed conversion ratio was highest in fish receiving the maize or pea protein diets and final average weight gain was highest in the control group (1.05kg) compared to fish fed with soybean protein (0.96kg), maize gluten (0.85kg), pea protein (0.94kg) and a plant-based combination diet (1.01kg). Fat digestibility was highest in the fish meal diet, however protein digestibility was the same across all diets. High inclusion of pea protein inflamed the distal intestine in a similar manner to soy enteritis. Over eight weeks, five diets were fed to salmon. The control diet was based on fish meal (250g per kg). Three low fish meal diets (100g per kg) contained either 350g per kg pea protein concentrate, 300g per kg soybean protein concentrate or 300g per kg maize gluten. A further low fish meal combination diet comprised of 130g per kg pea protein concentrate, 105g per kg soybean protein concentrate and 105 g per kg maize gluten. Growth rates, final body weight, digestibility and morphological parameters were measured.

 

10 

A randomised, replicated, controlled study in Norway (Sørensen et al., 2011) found salmon, Salmo salar, fed soybean meal diets (positive control) had a higher food conversion ratio and lower weight gain than those fed pure fish meal (negative control) or fish meal plus supplements (experimental diets). Average weight gain in the group fed soybean meal was 142kg compared to the negative control diet (237kg) or supplemented diets (180-214kg). The supplemented diets did not interfere with fat or protein digestibility and no morphological changes were recorded in the distal intestine. Over 68 days, groups of salmon were fed one of six diets. These comprised a negative control fish meal diet, a positive control soybean meal diet and one of four diets based on the negative control diet and added supplements (raffinose, stachyose, a combination of raffinose and stachyose and the same combination further supplemented with soya-saponins). Feed conversion ratios, final body weight, fat and protein digestibility and morphological changes were measured.

 

11 

A controlled study comprising of two trials in the USA (Burr et al., 2012) found varied results in final body weight when feeding different stage fingerling salmon, Salmo salar, plant-based protein diets compared to a fish meal control. Trial one found early stage fingerlings fed a plant protein diet had lower average final body weights (19.23-22.97g) compared to the control (27.29g). Trial two found that late stage fingerlings had similar final average body weights across all diets. Survival rates ranged from 92.4-96.3% and were similar across all treatment groups. In trial one, 50 early stage fingerling salmon were fed six experimental diets prepared using plant protein blends (soybean, maize gluten, wheat gluten).  Plant protein was added to five diets in increasing increments (50%, 66% and 84%) to substitute fish meal. The control diet contained fish meal only. Feeding took place over 18 weeks, using automated feeders. In trial two, 112 late stage fingerling salmon were fed diets containing 100% plant protein blend for 12 weeks. All fish were bulk weighed and counted every four weeks during the study

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