Sustainable aquafeeds: Progress towards reduced reliance upon marine ingredients in diets for farmed Atlantic salmon ( Salmo salar L.)

Abstract

The salmon farming industry has been criticised for being a net consumer of marine resources, in the form of fishmeals (FMs) and fish oils (FOs) used in feeds. Despite the efforts made to replace FM and FO with alternatives, such as vegetable proteins and oils, the balance is still generally negative, with calculated fish in–fish out (FIFO) values often being over 4. This paper reports on a FM and FO replacement study, with maximum 20 and minimum 10% FM inclusion in high-energy, extruded salmon feeds, and in which 50% of the feed oil was of vegetable origin (rapeseed). Further, half of the dietary FO was oil reclaimed from fish processing waste (herring offal silage oil), the other half being pristine FO (blend of herring and anchoveta oils). Growth and feed utilisation were assessed in a 9 month trial, during which fish weight increased from ca. 1.2 kg to ca. 4.6 kg. There were no significant differences between feed treatments with respect to growth, feed utilisation and mortality, and replacement of FM with vegetable proteins did not compromise the bioavailability of feed nutrients. Salmon given the feed with the highest level of fishmeal replacement (FM10) had a net production of fillet protein relative to feed input in the form of protein derived from FM, indicating that FM supply is not a major factor that would impose serious limits on the quantity and efficiency of production. The inclusion of FO as 50% of the feed oils ensured that the salmon fillets contained levels of n-3 highly-unsaturated fatty acids (n-3 HUFAs) that would be considered adequate from a consumer perspective (at least 1.5 g n-3 HUFAs per 100 g fillet) and the ratio of n-6 to n-3 fatty acids (ca. 0.65) was also favourable from a human health point of view. There was net consumption of marine fish resources when assessed as FIFO calculated on the basis of the amounts of fish required to produce all FOs (FIFO 3.03–3.59) and on fish needed to produce pristine FOs included in the feeds (FIFO 1.53–1.83). Calculations based upon nutrient ratios gave positive outcomes, and salmon in all treatments deposited more fillet fat than the amount of pristine FO consumed. It is concluded that supplies of FOs impose greater limitations on the formulation of salmon feeds than do supplies of FMs. The results of the study also indicate that increased use of fish processing by-products has the potential to reduce some of the predicted short-fall in FOs resulting from reductions in the amounts of small, pelagic marine fish species rendered directly for the production of FMs and FOs.