The final countdown: Continuous physiological welfare evaluation of farmed fish during common aquaculture practices before and during harvest

Welfare of farmed fish has become of increasing concern for consumers, producers, interest groups and authorities. To improve fish welfare, it is necessary to find indicators that can identify stressful situations early enough so that an intervention can take place before detrimental effects occur. By using heart rate bio-loggers in freely swimming rainbow trout (Oncorhynchus mykiss), combined with plasma cortisol levels and a range of haematological and blood chemistry parameters, we assessed the severity of stress responses induced by a range of common aquaculture practices before and during harvest. Following surgery, transportation and reintroduction with conspecifics in the sea cage, it took ~4 days for heart rate to stabilize and for a clear circadian rhythm in heart rate to emerge (i.e. average circadian fluctuation in heart rate of ~25 to 27 beats min−1). The presence or absence of this circadian rhythm in heart rate could inform researchers in the aquaculture industry whether or not specific farming routines induce unnecessary and prolonged stress. The elevations in heart rate caused by common farming practises such as crowding and transportation (e.g. heart rate increased by ~8 and 9 beats min−1 above what would normally be expected for that time of day, respectively) corresponded well with increases in plasma cortisol levels. Stressful farming practises or events (indicated by elevated plasma cortisol levels) such as air exposure during brailing and aquatic hypoxia triggered a hypoxic bradycardia until fish were released back into oxygenated water whereupon heart rate significantly increased to repay the accumulated oxygen debt. Repeated stress induced by multiple farming practises (i.e. combined stressors of crowding, brailing and transportation) clearly had a cumulative and long-lasting effect as heart rate peaked at ~25 beats min−1 above what would normally be expected for that time of day. Heart rate also remained significantly elevated by ~9 beats min−1 the following morning, suggesting that if rainbow trout need to be subjected to multiple stressors during consequtive days, it is recommended that sufficient time for recovery is provided between stressors. This study demonstrates that heart rate monitoring can be useful to assess stress levels of freely swimming fish in sea cages. Moreover, the use of implantable bio-loggers opens up a broad range of possible applications that will allow researchers to investigate the effects of environmental and/or anthropogenic stressors on the welfare of fish under conditions more realistic to the aquaculture industry.