Understanding intrinsic and extrinsic drivers of aggressive behaviour in waterbird assemblages: a meta-analysis

Aggressive behavioural interactions between animals are widespread in nature, with ecological and evolutionary consequences of such interactions reported for both individuals giving and receiving aggression. Yet despite the importance of aggressive interactions in determining social dominance and conferring fitness benefits to successful individuals, we lack a general understanding of the conditions that influence the variation in agonistic behaviours among individuals and species. We conducted a global meta-analysis of published time-activity budgets, using a data set comprising 555 values from 88 studies, to determine variation in the time that waterbirds engaged in aggressive interactions. The mean ± SD percentage of time devoted to aggression was 2.0 ± 3.6%, with a range of 0.0-35.0%. We used our data set to test four predictions regarding avian aggression, based on the findings of earlier site-based studies. We predicted that the time spent on aggression would be influenced by four factors: (1) age class, (2) sex, (3) seasonal timing and (4) functional feeding group. A comparison of linear mixed-effects models using an information theoretic approach indicated that the proportion of time waterbirds spent engaged in aggressive interactions was best explained by the age class and sex of the focal waterbird species. More time was spent on aggression by males than females, and by adults than juveniles. We found no evidence that the time spent on aggressive behaviours varied across latitudes or body mass, with seasonal timing, sexual dimorphism, migration or breeding strategies, or between different functional feeding groups. Our findings highlight the high levels of variation in the time devoted to aggression across species, feeding groups, latitudes and seasons. Furthermore, our study demonstrates the utility of the numerous published time-activity budget studies that are available as a valuable source of data that can be used to answer broad-scale questions regarding animal behaviour.