Three-dimensional assessment of hunting strategies in a deep diving predator, southern elephant seal Mirounga leonina

• Published in: Marine Ecology Progress Series
• Main Authors: Jouma'A, Joffrey, Le Bras, Yves, Picard, Baptiste, Guinet, Christophe
• Other Authors: Centre d'Études Biologiques de Chizé - UMR 7372 (CEBC), Université de La Rochelle (ULR)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE), École pratique des hautes études (EPHE), Université Paris sciences et lettres (PSL)
• Format: Article in Journal/Newspaper
• Language: English
• Published: HAL CCSD 2017
• Subjects: Data mining; Unsupervised classification; Area restricted search; Foraging; Accelerometer; Acousonde™; Machine learning; Spherical first passage time; [SDE]Environmental Sciences; Kerguelen; Kerguelen Islands; Elephant Seal; Mirounga leonina; Southern Elephant Seal
• Online Access: https://hal.archives-ouvertes.fr/hal-01573651; https://doi.org/10.3354/meps12191

International audience Few studies have examined the foraging ecology of top marine predators by taking into account the vertical and horizontal dimensions simultaneously. In this study, using 3-dimensional reconstruction of underwater animal paths, we investigated the relationship between area restricted search (ARS) and foraging behaviors in the southern elephant seal Mirounga leonina. We equipped 6 post-breeding females on the Kerguelen Islands with an Acousonde™ recording simultaneously sounds, pressure, acceleration and magnetic field-strength in 3-dimensions (n = 879 dives). Using the spherical first passage time analysis, we identified ARS in 46.4% of the dives. They represented 36.9% of the bottom time, but due to a higher prey encounter rate, 67.6% of catch attempts took place in ARS. These areas, occurring within a sphere of 48.2 ± 25.7 m radius, were characterized by an increase in energy expenditure and a strong variability in the roll and heading compared to non-ARS regions. While within an ARS, we identified 3 hunting modes from unsupervised classification based on acceleration properties previous to captures. We found that 32.8% of attempts matched a passive-gliding approach from above the target. The other 2 modes corresponded to an actively swimming approach associated with increases of variance in heading and/or rolling movements for the first one (56.6% of captures), and to a much larger swimming effort for the second one (10.6% of captures). These data revealed a strong link between seals’ fine-scale movements and feeding activity, suggesting that seals fed mainly on patchily distributed resources, with different hunting strategies that may be indicative of several prey species targeted.