Integrating social network analysis and fine-scale positioning to characterize the associations of a benthic shark

Many shark species form groups; yet the drivers of this behaviour are not well understood. Aggregations have been presumed to be a function of resource availability or phenology. However, recent studies have suggested that some sharks display preferences in their associations with conspecifics and that complex social networks may underpin their movement patterns, demographic distribution and fitness. This study used a network approach to investigate patterns of sociality of a benthic predator, the spotted wobbegong shark, Orectolobus maculatus (Family Orectolobidae), in a small marine reserve. Spatial data obtained from fine-scale passive acoustic telemetry were used to infer association preferences of 15 tagged sharks over a 15-month period. Thirty-five associating dyads emerged, with permutation tests indicating that these were nonrandomly formed. Many dyads were temporally stable, with some persisting even after sharks returned to the reserve from their seasonal migration. A complete but sparse network was observed and although association patterns were evident at the dyadic level, indication of stable communities or network structure was limited. The population was not found to be gregarious and exclusive pairs were not observed, with 14 sharks associating with more than one individual. Assortative mixing (by sex, size and familiarity) was not present and range overlap did not strongly correlate with association indices. This suggests that these relationships are not a random result of nonsocial grouping behaviours but can be explained, at least in part, by genuine social affiliation. As such, anthropogenic influences on the population may have more complex impacts than previously thought.