The assessment of the feeding behaviours of free-ranging animals is essential in order to estimate their time budgets, and thus understand how these animals maximize their fitness. However, investigating this foraging behaviour in their natural environment remains a significant challenge, as it is often impossible to obtain long-term behavioural data through visual observations alone.

However, it is possible to use bio loggers that combine in a single device an accelerometer, a gyroscope and a temperature-depth recorder (TDR) to obtain high frequency data about animal behaviour that can be analysed when recovered.

These high frequency data can used to segment and classify into clusters the different activities that animals can do. By mounting a camera in only a few individuals, we can then then understand to what biological behaviours correspond to each of these clusters, and therefore extrapolate our knowledge even when no visual observations are available.

In this study, we demonstrate the validity of the technique for green turtles, for which the equipment with cameras and GPS allowing the safe recovery of all the equipment does not cause a great disturbance in the animal. We plan to apply next this approach to penguins, to gain insights of their foraging behaviour while minimising the observation disturbance.

• Jeantet L, Planas-Bielsa V, Benhamou S, Geiger S, Martin J, Siegwalt F, Lelong P, Gresser J, Etienne D, Hiélard G, Arque A, Regis S, Lecerf N, Frouin C, Benhalilou A, Murgale C, Maillet T, Andreani L, Campistron G, Delvaux H, Guyon C, Richard S, Lefebvre F, Aubert N, Habold C, Le Maho Y, Chevallier D.  (2020). Behavioural inference from signal processing using animal-borne multi-sensor loggers: a novel solution to extend the knowledge of sea turtle ecology. Royal Society Open Science  N°7(5): 200139.

For more information, please contact:

• Dr Victor Planas-Bielsa