Identifying shark mating behaviour using three-dimensional acceleration loggers

Abstract

Little is known about the reproductive strategies and mating behaviour of most sharks. Understanding mating behaviour is important as it can determine reproductive success and possibly rates of multiple paternity and fecundity. Additionally, some sharks appear to have specific habitat requirements for mating activities. We tested the utility of a 3-dimensional acceleration logger to identify mating events in free-living nurse sharks Ginglymostoma cirratum in an area where behav- iours can be observed directly, thus allowing corroboration of acceleration data. Loggers were attached to 4 adult females and were recovered after recording periods of 23.2 to 99.8 h (mean ± SD = 50.0 ± 35.1 h). We used acceleration data to classify several behaviours, including swimming, rest- ing, resting in a surge zone, and mating, with examples of each behaviour confirmed via direct obser- vation. Twenty-six mating events were inferred from acceleration data, ranging from 21 s to 20.1 min in duration (median = 2.22 min), with no events taking place during the nighttime hours between 23:00 and 07:00 h. Four events lasted longer than 9 min and took place during periods when logger- equipped sharks were not acoustically detected within the shallow study site. The similarity in mat- ing behaviours between this and some other species raises the possibility that accelerometry could be used to quantify mating in a variety of shark species. This is the first study to classify mating and other behaviours in free-living sharks from acceleration data.