The behaviour of Atlantic salmon in relation to efficient cage-rearing

Abstract

To maintain salmon farming as a viable industry it is necessary to continue to focus on the well-being of the fish from hatching to harvesting. This requires intimate knowledge of all aspects of salmon biology. Understanding the behaviour of farmed salmon is an essential part as this could suggest ways of preventing stress and diseases rather than merely treating their symptoms. This review attempts to summarize current knowledge of salmon behaviour in sea cages. Major modes of behaviour have been described and possible control mechanisms have been suggested. Behavioural studies of salmon in this particular system, however, are still at an early stage. Further research is needed to verify and deepen the current understanding. In future research the behavioural ontogeny of cultured salmon from hatching to harvesting should receive more attention. For example, the effect of smolt-rearing procedures on subsequent behaviour in sea cages is unclear. An ontogenetic approach would probably pay off in terms of a better holistic understanding of the mechanisms controlling the behaviour of cultured salmon in different stages of the production process. The present level of understanding should provide a basis for developing simple models of behaviour in cages from which meaningful hypotheses can be formulated. Experiments should primarily be carried out in a realistic rearing environment, but the inherent methodological problems of carrying out full-scale experiments make it difficult to test hypotheses that require detailed description of the behaviour of individuals. The transfer of a small number of individuals from larger groups subjected to different treatments to an experimental arena, might overcome these problems. This approach proved fruitful in an unpublished study that tested the competitive ability of individuals reared under different feeding regimes. This paper proposes several ways to improve the management of salmon behaviour and growth in sea cages. Some of these suggestions should be further investigated before they are adopted by the salmon industry in general. Daily management would be improved by methods to observe the fish complementary to visual inspection. Continuous hydroacoustic monitoring of sea cages provides this and would be especially helpful in large cages holding a considerable value of fish. In addition to demand feeding and routine behavioural observations (e.g. early warning of disease), this would provide useful information about the accumulation of dead fish on the cage bottom and the abundance of wild fish under cages (predators, food-waste eaters). Another important application could be to monitor the fish biomass. Fish counters are sometimes used during stocking of sea cages with smolts, but the error in estimated numbers remains unknown. Unregistered theft, escapees and mortality in the period prior to harvesting can easily increase errors in such estimates. A reliable hydroacoustic method of estimating biomass would thus significantly improve management control. Such work is in progress (Furuzawa et al., 1984; Burczynski et al., 1990; Bjordal et al., 1993; Dunn and Dalland, 1993).