The social and spatial organisation of avian societies is often complex and dynamic with individuals socialising with others in a local population. Although social interactions can readily be described in colonial breeders through the location of nests, social interactions regularly take place in other contexts that are often not considered. Social behaviour in the colonially breeding zebra finch, Taeniopygia guttata, has been the focus of much work in the laboratory, but very little is known about their social organisation in free-living populations, especially outside the breeding context. Here we characterise semi-permanent gathering locations, or ‘social hotspots’ in the zebra finch in the wild. We determined the use of such social hotspots and the resulting group dynamics by quantifying movements to and from these locations through direct observation and by quantifying the vocal activity at these locations using acoustic recorders. We show that, throughout the day, zebra finches regularly visit these hotspots, and the hotspots are occupied for a substantial proportion of the day. Individuals typically arrived and left in pairs, or small groups, indicating that these social hotspots do not function just for flock formation. Instead, the high levels of vocal activity at these hotspots indicate that they may potentially function as local hubs for socialisation and information exchange, whilst also perhaps providing safety-in-numbers benefits to individuals during periods of resting. These findings characterise an important component of the natural social life of one of the most widely studied birds in captivity. The characterisation of these social hotspots highlights the use of landmarks by birds to facilitate social contacts, cohesion, and behaviour, in a social bird. Similar hangouts and social hotspots may be a feature of social behaviour in other multi-level aggregative species in which the fission and fusion of groups is an important component of daily life.

This dataset collection describes data from three different methods that we combined in order to describe the presence and usage of social hotspots in wild zebra finches. These hotspots are temporally stable locations, either in bushes or (shrub-like) trees, where zebra finches come and go, and spend substantial time. Hangouts, periods of time when the hotspot is occupied, regularly take place at these meeting locations. The data was collected from October–December 2019 at UNSW Fowlers Gap Arid Zone Research Station in New South Wales, Australia, home to a well-studied population of wild zebra finches. The three methods are:

a) Dropping counts
We counted zebra finch droppings under bushes/trees in 10x10 cm squares in order to identify and/or confirm which vegetation was substantially used by zebra finches. For each site we identified also a 'control tree' which was nearby and similar, but where we observed a lower amount of droppings. These 'social tree' – 'control tree' pairs were then used for the rest of the study. The droppings.csv dataset contains repeated (over several days) dropping counts for these pairs (n = 10 tree pairs).

b) Focal observations
We conducted two full-day observations (from before dawn till dusk) of social hotspot and control tree pairs (n = 7 tree pairs), where we quantified the group size of each arriving and departing group of zebra finches, as well as kept track of the number of zebra finches on the social hotspot and the control tree. This dataset describes the group dynamics at these sites over time. The hangout_observations.csv dataset contains all the arrival and departure events that we observed during these observation days.

c) Audio recordings
We put time-programmed recorders in the social hotspots and control trees (n = 10 tree pairs) to quantify how much time zebra finches were spending at each respective tree, and more precisely, how often vocal activity took place. To do this, we measured for one full day of recording (about 12h) every bout of zebra finch vocal activity (where a silence of 5 minutes demarcates the end of a vocal activity bout). Several of these recorded days were the same as the days during which we did the focal observations, so that we could also compare and validate the results of both methodologies. The hangout_acoustics.csv dataset contains all the vocal activity bouts.

For further details on how the data was collected, we refer to the (open-access) related article in Journal of Avian Biology.

We included a README file that contains descriptions of all variables in all files that are part of this dataset collection here on Dryad, and also of the software elements (R and python code) that are uploaded to Zenodo. The readme also describes which files are the raw datasets, which are the processed ones, and which are the files to process raw into processed files.