Experience-based optimal foraging on planktonic prey in Baikal seals.

BACKGROUND: Understanding how predatory animals efficiently locate prey with limited knowledge of its location is challenging. Optimal foraging theory suggests that animals improve their food intake through experience-based adjustments of search patterns. For example, animals feeding on clustered prey may repeatedly search successful areas and move farther away when unsuccessful (the 'win-stay, lose-shift' strategy). A related concept, area-restricted search, predicts that animals initially search broadly and then switch to a more localized, tortuous search upon finding clustered prey. However, few studies have empirically supported these predictions for large aquatic predators due to difficulties in recording their foraging success on known prey species. METHODS: We used biologging techniques to record the fine-scale foraging behaviour of Baikal seals in Lake Baikal, which hunt tiny, clustered, planktonic amphipods at high rates. We reconstructed their three-dimensional movement paths during dives and estimated the timing of prey capture events based on video-validated body acceleration data. RESULTS: Seals moved shorter horizontal distances and exhibited greater directional changes after more successful dives, supporting the 'win-stay, lose-shift' strategy. Consistent with area-restricted search, successful foraging dives led to decreased speed and increased tortuosity in the horizontal plane. CONCLUSIONS: These findings suggest that experience-based behavioural adjustments at a dive-to-dive scale are crucial for Baikal seals-and possibly other large aquatic predators-to maintain high foraging rates. Furthermore, they illustrate how an exceptionally high predator-prey body mass ratio (> 500,000) for a single-prey-feeding (non-filter-feeding) predator is maintained in the unique Lake Baikal ecosystem.