Individual level consistency and correlations of fish spatial behaviour assessed from aquatic animal telemetry

The potential for populations to undergo adaptive evolution depends on individual variation in traits under selection and how multiple traits are correlated. While fitness relates to the performance of animals in the wild, most of the research on evolutionary potential of behavioural traits has used captive or mesocosm settings, especially with aquatic organisms. We investigated the individual level consistency (personalities) and correlations (behavioural syndromes) of fitness-related behavioural traits displayed by a harvested marine fish in the natural environment, and the potential of such individual level behaviour to constrain adaptive evolution. For this, we acoustically tracked 303 individuals of Atlantic cod, Gadus morhua, over a period of 3 years from two populations in southern Norway. We then estimated repeatability and correlations between four behavioural traits: diel vertical migration, activity, home range and dispersal. Average autonomy was estimated as a measure of the potential of the observed behavioural syndromes to constrain evolution. We found a moderate to high individual consistency in all behavioural traits (mean repeatability 0.41, range 0.22-0.66) suggesting that they reflect fish personalities. Also, behavioural traits were structured into behavioural syndromes, where one population displayed a more integrated behavioural syndrome (all the four traits correlated) than the other (only home range and dispersal correlated). In both populations, the magnitude of these behavioural syndromes could potentially constrain evolution, as revealed by average autonomy values significantly below 1. Our study provides strong empirical support to the idea that natural and human-induced selection on behavioural traits can indeed drive adaptive evolution, and, further, that such evolutionary responses may be constrained by the correlational structure among multiple traits.