Characterizing movement patterns of killer whales along the Norwegian coast

Abstract

Abstract Background There has been ever-growing research fitting movement models to marine mammal satellite tracking data in recent years, yet little focus on methods attempting to go beyond individual restricted foraging and directed transiting behaviors. Prior studies investigating Norwegian killer whale ( Orcinus orca ) behavior typically used either a continuous behavioral index or two to three discrete behavioral modes. Methods Here we classified killer whale behavior in a more nuanced manner along the Norwegian shelf and within fjords by parsing satellite tracks into five predefined behavioral modes: ranging, encamped, nomadic, roundtrip, and semi-roundtrip. For this, we used a behavioral change point analysis (BCPA) to select homogeneous velocity-persistence states and then modeled the net squared displacement (NSD) over time to differentiate between each mode. Results Whales had consistent use of all classified behavioral modes across differing areas with increased time allocation to herring ( Clupea harengus ) overwintering and spawning grounds. We observed a high degree of intra-individual variability in movement types, particularly within herring overwintering and spawning areas. The timing of major behavioral shifts between more resident, round-trip, and partial round-trip type behaviors to more transiting behavior corresponded with the assumed timing of the herring migration. Conclusions Classifying the tracking data into five candidate modes that incorporate absolute space use expanded the dimensions of analysis and revealed the finer-scale movement and behavioral patterns among individuals and across their habitats. The apparent individual variation across movement patterns highlights the gaps in knowledge regarding the drivers behind variation of finer-scale search and dispersal behavioral movements in this species. By crossing both movement metrics and absolute space use, this paper points towards ways to expand the dimensions of analysis that reveal a nested hierarchical structure.