Figure 1

Plasma cortisol and cortisol release rate after long term hypoxia (Experiment 1). This experiment aimed at mimicking an overall situation often found in sea cages when water flow decreases and DO levels are reduced to levels as low as around 50%. Decreased water exchange rate may also cause increased concentrations of toxic metabolites. Four experimental oxygen regimes in triplicate tanks were created, fixed oxygen levels at 50%, 60%, 70% and 80%, automatically regulated by adjusting inflow (range 41-137 L min^-1) in response to oxygen consumption of the fish. Blood was sampled for plasma cortisol measurements from all treatment groups at days 9, 29 and 48 and from the 80% and 50% DO between days 41-43. Non-invasive measurement of cortisol status of the fish was conducted by measuring the cortisol release rate into the water. Plasma cortisol levels were analysed using a general linear model with sampling occasion and DO treatment levels (with tank nested within treatment) as factors. DO levels had an effect on plasma cortisol levels (A) (p < 0.001) but were not affected by time (p = 0.607) and no interaction could be observed (p = 0.128). Further, SNK post hoc test grouped the 50% and 60% DO groups in one subset and the 70% and 80% DO groups in one subset at day 9, 29 and 48. No differences could be observed in plasma cortisol between 80% and 50% DO levels in fish sampled for intestinal barrier function between days 41-43. No major differences could be observed in the cortisol release rate between treatment groups (B). All data are expressed as means ± SEM and p < 0.05. The overall effect of treatment is indicated by ◊.