Interview with a rhino : establishing communication structure and call functionality in a semi-social species (Ceratotherium simum simum)
Acoustic communication is a complex model that is most essential in social animals. Its structure can be applied to describe social bonds and group dynamics and is therefore substantial for animal monitoring and management. However, for species with rather loose social bonds and low reactivity the establishment of a distinct communication structure is challenging. White rhinoceroses (Ceratotherium simum), as semi-social megaherbivores, are such a species, which in addition to their declining populations draws major research interest to them. Within this framework, the present study aimed to examine the communication structure and the functionality of calls in the white rhinoceros (WR). For this purpose, behaviour and vocalisations of 30 Southern WR (21♀♀, 9♂♂) were simultaneously video- and audio recorded in 7 EAZA-zoos. The analysis focused on 3 most common call types: snorts, hisses and pants. The behavioural context and the nearest neighbours during vocalisations were determined. By considering neighbours at a proximity of <1 body length to the sender as potential receivers, directed call rates were calculated and effects of dyad and interaction type were tested (LME model). Moreover, communication networks were constructed for each call type in order to analyse directionality and arrangement of vocal interactions. The results demonstrate an approach to establish a distinct communication structure in WR based on the proximity of surrounding conspecifics. The communication networks demonstrate that call rates are not randomly distributed but are rather depending on the potential receiver, indicating an adjustment to present group members. Furthermore, vocal interactions proved to be unidirectional. Moreover, it could be shown that while snort call rates were not affected by the dyad or interaction type, hiss call rates could be best predicted by dyad type and the rate of agonistic interactions, revealing that females were significantly more often hissing to males while the hiss call rate correlated with the rate of aggressive behaviour. For the pant call rates dyad type and the rate of affiliative interactions were the strongest predictors, emphasising higher pant call rates from males to females during socio-positive interactions. In conclusion, each call type proved to have distinct functionality as well as social relevance, which represents a substantial ground for further studies on communication and socio-sexual behaviour in WR.