Back

Falcons use wind assistance and remote islands to mitigate risk during ocean-crossings

CHEN, M.; Viana, D. S.; Figuerola, J.; Gangoso, L.; Vansteelant, W. M. G.

2025-08-15 animal behavior and cognition
10.1101/2025.08.12.669758 bioRxiv
Show abstract

Wind conditions play a major role in determining how birds negotiate ecological barriers during long-distance migration. Migrants typically minimize flights over barriers in opposing winds, while crossing the same barriers directly in supportive winds. Nevertheless, even in supportive conditions, barrier-crossings likely involve a number of risk-mitigating behaviors, including the use of stepping stones. We aimed to understand how seasonal ocean-crossings of Eleonoras falcons (Falco eleonorae) compare in terms of wind support and flight effort, and whether falcons use islands to cross safely in variable wind conditions. To do this we combined GPS-tracking data from 19 individuals over a decade (2012-2022) with wind data from an atmospheric reanalysis model. Despite major differences in wind support, falcons achieved similar ground speeds in both seasons. That is because falcons reduce instantaneous flight effort by flying at lower airspeeds in supportive spring winds (31.5 {+/-} 6.0 km h-1) than in adverse autumn winds (47.4 {+/-} 9.3 km h-1). Overall, it took falcons twice as many flight hours to complete the spring crossings that were 40% greater in terms of air distance compared to autumn crossings. Islands were more frequently used during spring crossings (57.9% individuals in spring vs 21.1% in autumn), and the probability of a falcon using an island significantly increased with weaker wind support in spring (but not in autumn). While wind support partially offsets the extra distance flown over the ocean in spring, isolated islands offer emergency stop-over opportunities when Eleonoras falcons experience relatively weak wind support during protracted ocean crossings.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.