Seasonal adjusted diets to Atlantic salmon (Salmo salar): Evaluations of a novel feed based on heat-coagulated fish mince, fed throughout 1 year in sea: Feed utilisation, retention of nutrients and health parameters

The present paper reports the results of a 1 year experiment using Atlantic salmon with a starting weight of 1.7 kg, and final weight of 7.4 to7.8 kg dependent of diet group. The experiment aimed to evaluate a direct production method of feed where landings from fishing vessels are used directly in feed processing. The dietary compositions followed the natural rhythm in lipid level of the catches of herring in the North Sea (Catch group) adjusted to be stable throughout one season only, and compared to a control diet (constant composition throughout the year) and a designed diet (opposite to the natural biorhythm of salmon lipid stores) (diet group Design). Four seasons were defined: spring (March, April, May), summer (June July August), autumn (September, October, November), winter (December, January, February). All diets varied in protein and lipid due to the adjustment according to season. The experiment was run in triplicate sea cages for each diet, using natural light and temperature regimes (21°N). All diet groups showed equal growth throughout each season, except for the Catch group during autumn showing lower SGR-values. SGR values varied significantly also between seasons for all diet groups, ranging from 0.29 during spring to 0.81 during autumn. Feed conversion ratio varied dependent on season and the dietary protein to energy ratio, at all seasons except winter. Quite a large variation between diet groups and seasons were registered in condition factor. Whole body lipid varied between seasons, as a consequence of dietary lipid levels, and increased as fish weight increased. After autumn growth only, whole body protein levels were significantly lower in the Catch group compared to the two other diet groups. Muscle lipid levels did not vary as a consequence of fish size alone, but varied substantially between seasons in both the Catch and Control groups, but not in the Design group. Muscle alpha-tocopherol and fatty acid profile varied due to diet in spring and autumn seasons, but with minor variations due to diet in summer and winter periods. Only minor influences from diet were registered in liver and heart compositions. Haematological parameters and plasma nutrient values ranged within normal values for salmon, and together with close to no mortality throughout the experiment, fish health was concluded to be good in all diet groups at all seasons. However, plasma asparagine aminotransferase (ASAT) values were quite high during summer and autumn, especially in the control group. These results point to the need of optimizing antioxidant composition of diets during the summer and early autumn season. Overall results show the possibility to obtain acceptable production results when feeding Atlantic salmon a diet adjusted to each season, as long as the fish protein to energy (P:E) requirements are met, further that total body lipid stores varied more than muscle lipid concentrations, and that the lipid levels were highly dependent on season, diet composition and fish size.