Effects of the timing of the introduction of feeds containing different protein and lipid levels on the performance and quality of Atlantic salmon, Salmo salar, over the entire seawater phase of growth

Abstract

A 430-day study was undertaken to evaluate whether adjusting the timing of introduction of feeds with varied protein and energy content could constitute a long-term strategy for steering the quality of Atlantic salmon throughout the entire seawater cycle. Feed regimes were designed with changes in proportions of protein and oil implemented in response to fish growth. For the Control regime (CON), stepwise changes in protein and oil level reflected practice in Scottish salmon aquaculture. By comparison with the Control regime, the Retard (RET) programme was characterised by diets with a higher level of protein and lower level of oil for a given fish size. Conversely, for the same weight of fish, the Advance (ADV) programme was characterised by diets with a lower level of protein and higher level of oil than was found under the Control regime. A further regime was created for salmon of more than 2.3 kg round weight by the introduction of a lower protein variant of the Advance diet designated ADVLP. The final diet regime (LOW) was characterised by a high level of crude protein (50%) and very low oil level (16%), which was not adjusted according to fish size. Twenty extruded feeds were produced with DP/DE ratios from 25.6 to 16.0 g MJ −1 (sizes 3.0–8.5 mm). The trial took place in 24 net pens (5×5×6 m), each with an adaptive feeding system and each pen receiving 800, 70 g, S 1/2 smolts. After 14 months, the salmon weighed between 2.5 and 3.6 kg for those fed according to the LOW and ADV regimes, respectively (SGR=0.83–0.92% day −1; TGC=2.39–2.77). The cumulative growth rate of salmon fed according to the LOW regime was significantly lower than that of fish fed the higher energy feeds, and the cumulative growth of the salmon fed according to the ADV programme was significantly greater than that of fish fed according to the RET, CON and ADVLP programmes ( P<0.001). At harvest in April 2001, there were small diet-related effects on the basic yield and quality indices amongst fish fed according to the RET, CON, ADV and ADVLP programmes, though the quality of fish fed according to the LOW regime was consistently different to that of the other four programmes. Despite starting the process at smolt transfer, the compounding effects of season- and size-related increases in inter-individual variability made it difficult to steer the yields, composition and quality of Atlantic salmon simply by conservative changes in the gross composition of the feeds. Differences in fish quality achieved using the LOW feed regime were achieved but at the cost of substantial growth and FCR penalties.