No effect of ocean acidification on individual-level variation in behaviour and susceptibility to predation in a Great Barrier Reef damselfish

Dominique G. Roche, Josefin Sundin , Ben Speers-Roesch, Shinichi Nakagawa, Timothy D Clark, Sandra Ann Binning

1) Ocean acidification, caused by rising carbon dioxide (CO2) in the atmosphere, has been reported to negatively impact a wide variety of behaviours in fishes, including activity, exploration, and predator avoidance. 2) These effects have been documented at the population level, but many animal species naturally show large and repeatable individual-level differences in behaviour. How environmental stressors, such as ocean acidification, affect behavioural variation at the individual level remains largely unknown but is critically important to understand adaptation given natural selection operates on variation at the individual rather than population level. 3) Using a statistical approach allowing variation in means and variation in variance to be modeled within a single framework, we quantified individual-level differences across five behaviours in the coral reef damselfish Pomacentrus amboinensis (emergence time, activity level, time spent sheltering, thigmotaxis, novel object inspection). We measured behaviour in a novel environment assay, twice before (CO2 ~450 µatm) and twice following acclimation to predicted end-of-century ocean acidification conditions (~1,100 µatm). 4) Following behavioural assays, we tested individual survival in a live predation experiment. We used predatory rock cod, Cephalopholis microprion, acclimated to the same CO2 treatments as Ambon damsel and examined predictors of survival probability. 5) All behaviours in damselfish were moderately and significantly repeatable, with no marked differences in repeatability estimates between the ambient CO2 and elevated CO2 treatment groups. Exposure to end-of-century ocean acidification conditions had no effect on any of the five behaviours measured, both in terms of group means and residual (within-individual) variance. 6) The probability of survival in the predation trials was similar for damselfish in the elevated and ambient CO2 treatment groups. Smaller damselfish as well as those that spent a greater amount of time inspecting a novel object (i.e., bolder individuals) had a lower probability of survival regardless of their CO2 treatment. 7) Our results challenge assumptions about the impacts of ocean acidification on coral reef fish behaviour and susceptibility to predation, both at the population and individual level. They also provide support for a trade-off between boldness and predation risk in fish.