RESUMEN
In this study we present the first attempt at modelling the feeding behaviour of whale sharks using a machine learning analytical method. A total of eight sharks were monitored with tri-axial accelerometers and their foraging behaviours were visually observed. Our results highlight that the random forest model is a valid and robust approach to predict the feeding behaviour of the whale shark. In conclusion this novel approach exposes the practicality of this method to serve as a conservation tool and the capability it offers in monitoring potential disturbances of the species.
Asunto(s)
Conservación de los Recursos Naturales/métodos , Conducta Alimentaria/fisiología , Aprendizaje Automático , Tiburones/fisiología , AnimalesRESUMEN
The ability to produce estimates of the metabolic rate of free-ranging animals is fundamental to the study of their ecology. However, measuring the energy expenditure of animals in the field has proved difficult, especially for aquatic taxa. Accelerometry presents a means of translating metabolic rates measured in the laboratory to individuals studied in the field, pending appropriate laboratory calibrations. Such calibrations have only been performed on a few fish species to date, and only one where the effects of temperature were accounted for. Here, we present calibrations between activity, measured as overall dynamic body acceleration (ODBA), and metabolic rate, measured through respirometry, for nurse sharks (Ginglymostoma cirratum), lemon sharks (Negaprion brevirostris) and blacktip sharks (Carcharhinus limbatus). Calibrations were made at a range of volitional swimming speeds and experimental temperatures. Linear mixed models were used to determine a predictive equation for metabolic rate based on measured ODBA values, with the optimal model using ODBA in combination with activity state and temperature to predict metabolic rate in lemon and nurse sharks, and ODBA and temperature to predict metabolic rate in blacktip sharks. This study lays the groundwork for calculating the metabolic rate of these species in the wild using acceleration data.