Sociality of a threatened ‘solitary’ marsupial before and after population reinforcement: associations and links to breeding success

Conservation translocations are becoming an increasingly necessary tool to reverse population declines of threatened species. However, translocations often disrupt the social environment for species, which could have consequences on the fitness of individuals released. Greater bilbies (Macrotis lagotis) are considered solitary and threatened, with conservation translocations a key part of the national recovery plan for the species. Here, we use social network analysis to examine whether the little-known associations of bilbies, before and after a population reinforcement of bilbies to a fenced sanctuary, are important for post-release breeding and could inform future translocations. Although we did not find any relationship between early breeding success and individual sociality, there were several other indications that burrows were a focal area for bilby associations, particularly for breeding. At burrows, bilbies preferred to associate with the opposite sex, males avoided other males, and kin avoidance when breeding opportunities were more limited suggests that concurrent burrow sharing was likely related to breeding. We found no difference in the relative breeding success between release groups, and, although there were some sex-specific differences, overall bilbies associated both within and between release groups non-preferentially, suggesting that bilbies released during the reinforcement integrated successfully. Overall, this work demonstrates the importance of behavioural study for the conservation of socially covert species and underlines the need to incorporate such information into conservation management.

Methods
Data includes the bilby movement data, results of the pairwise genetic relatedness for all (N = 15) greater bilbies included in this study, the weighted degree of each bilby (i.e. node) within each social network with the corresponding number offspring produced, and the total number of dyadic associations between every possible bilby dyad with and without considering similar habitat preferences. We have also provided the R code for generating the sociality fitness modelling and analysis as described in the associated paper. Necessary shapefiles for assigning locational fixes to habitat type are also provided.