adaptive drift and individual repeatability in the trans-african migrations of eleonora’s falcons
Pale morph Eleonora’s Falcon carrying an UvA-BiTS GPS logger. - Photo by Laura Gangoso.
Eleonora’s Falcons (Falco eleonorae) engage in one of the most remarkable migrations of all Afro-Palearctic raptors. They are highly aerial hunters that breed colonially on islands and coastal cliffs throughout the Mediterranean Basin and in the East Atlantic. While preying on flying insects during most of the year, they raise their offspring almost entirely on a diet of migratory passerines. After their exceptionally late breeding season all Eleonora’s Falcons escape he European winter by migrating to northern Madagascar. During spring migration falcons from all across the breeding range appear to converge through a stop-over hot-spot in the Greater Horn of Africa before diverging into distinct trans-Saharan flyways. The Canarian Eleonora’s Falcons have been studied by Dr. Laura Gangoso and her team for over 15 years, and I had the opportunity to study the migration ecology of this study population with the support of a Juan de la Cierva Formacion Fellowship.
I conducted this postdoctoral project at the Figuerola group at the Dept. of Wetland Ecology at Estación Biológica de Doñana (EBD-CSIC, Sevilla, Spain) and focused on how the combined influence of seasonal wind regimes and ecological barriers shape the falcons’ seasonal flyways and travel schedules. Leveraging multi-year tracking data of individual falcons, I further assessed whether individual falcons adhere to consistent migration routines, or whether routes or timings are highly flexible depending on annual conditions. So far, this project resulted in two scientific publications in Movement Ecology (Vansteelant et al 2021) and Journal of Avian Biology (Vansteelant et al 2023). A third article by M.Sc. student Meixu Chen is currently in review, and focuses on how falcons use remote islands and seasonal winds to cross the Indian Ocean along distinct corridors in autumn and spring. In the near future, I hope to find some time to finalise unpublished analyses on the limits of endurance flight in this remarkable
species.
HOW DO JUVENILE HONEY BUZZARDS LEARN SAFE MIGRATION ROUTES BETWEEN EUROPE AND AFRICA?
Juvenile Honey Buzzards (Pernis apivorus) Ivar and Sven at their natal nest in southwestern Finland. This was an unlikely brood, as Ivar died shortly after leaving the nest in Finland, and Sven perished in Algeria after a two-day flight across the Mediterranean during which he stopped to spend the night on a fishing vessel. - picture by Patrik Byholm
Over the past few decades tracking technology has allowed us to map migration strategies for a wide range of species. However, in order to maximise the amount of data collected with expensive tracking devices most studies have focused on adult migrants. As such, we know very little about the way in which migrant birds learn their often highly complex and flexible migration routines, and how this leads to the emergence of individual and population-specific migration patterns. This is also true for the species on which I conducted most of my migration research so far: the European Honey Buzzard (Pernis apivorus). However, since 2011 Patrik Byholm at the Bioeconomy Research Team of Novia University of Applied Sciences has equipped more than 30 juvenile and 10 adult European Honey Buzzards (Pernis apivorus) born or breeding in southwestern Finland with satellite transmitters and GPS-GSM loggers. The Finnish Honey Buzzard project thus provided an unique opportunity to investigate how innate and external factors shape individual migration routines in a long-lived thermal-soaring migrant.
Routes taken by juvenile Honey Buzzards on their first autumn migration from Finland to sub-Saharan Africa, coloured according to the longitudinal wind speeds they encountered along the way. For full details see Vansteelant et al. 2017 in Proceedings of the Royal Society B: Biological Sciences.
Honey Buzzards make for a particularly interesting system to study the ontogeny of individual migration strategies, because juveniles typically cannot learn from adult conspecifics on their first migration. While adults engage in highly synchronised migrations along narrow overland migration corridors, juveniles depart on their first autumn migration approx. two weeks later and find their own way over a broad front, often engaging in long and risky flights over the Mediterranean Sea. How and at what age they eventually learn to use the traditional overland flyways has long remained a mystery.
As a first step towards unravelling the migratory development in this species I analysed how wind and geography shape the first autumn migration and distribution of juvenile Honey Buzzards across sub-Saharan Africa. This revealed that the wind conditions that these birds encounter on their first migration determine the longitude at which they settle their non-breeding residency sites south of the Sahara, more so than individual (innate) differences in orientation. The work was published in Proceedings of the Royal Society B (Vansteelant et al. 2017). I was further involved in two studies of collaborators at the Max Planck Institute of Animal Behaviour that used a step-selection approach to study how thermal soaring opportunities over the Mediterranean Sea may facilitate the broad-front migration of juvenile Honey Buzzards (Nourani et al. 2021), and whether Honey Buzzards selected routes differently in response to wind and thermal soaring conditions as they age (Bronnvik et al 2022).
Because some of the juveniles could be tracked for up to 5-8 years after fledging, we are also able to reveal remarkable variation in the age at which Honey Buzzard first return from tropical Africa to their European breeding range, and a remarkably long natal dispersal process whereby young buzzards take at least 5-6 years to establish an efficient spring migration routine that allows for them to recruit as breeders into their natal population (Mirski et al. 2025). I am currently wrapping up a final paper on this phenomenal study system, showing how young buzzards learn multiple flyways during their prolonged early-life learning process, and that the learning of traditional seasonal flyways is conditional on them synchronising migrations with adult conspecifics.
comparing migration strategies of marsh harriers from the Low Countries and southern sweden
The adult male Marsh Harrier ‘Walter’ was equipped with an UvA-BiTS GPS logger at his breeding site in Flandres (Belgium). - Photo by Anny Anselin (INBO)
While the number of tracking studies on migrant raptors is rapidly increasing new opportunities are emerging to make comparative studies of migratory behaviour across multiple populations of certain species. I conducted this projects as a free-lance ecologist in collaboration with researchers of the Knowledge Center Grauwe Kiekendief (Netherlands), the Research Institute for Nature and Forest (Flandres, Belgium) and Lund University (Sweden) to make such a comparative analysis of migration corridors and schedules across breeding populations of Marsh Harrier (Circus aeruginosus) in the Low Countries (Flandres and The Netherlands) and Southern Sweden.
Our analyses revealed a marked separation of autumn migration corridors of LC and Swedish harriers all throughout the West African-Eurasian flyway. In spring, harriers departed through overlapping migration corridors, but according to very different schedules depending on their breeding population. More specifically, harriers breeding in the Low Countries departed much earlier from West Africa, and made long stop-overs in NW Africa, whereas Swedish birds depart later and forego spring stop-overs, thereby catching up with the LC harriers in NW Africa, and finally arriving at their respective breeding areas around a similar time. These are remarkable differences considering the proximity of their breeding areas, and them converging at particular hot-spots in West Africa (e.g. Inner Niger Delta, Senegal River, coastal wetlands).
In a more general sense, our study highlights how difficult it is to generalise known migration strategies from one breeding population to another (Standberg et al 2008, Klaassen et al 2010). Our article was published open access in Journal of Ornithology (Vansteelant et al. 2020) along with accompanying data and code.
West African non-breeding sites of adult Marsh Harriers that were GPS-tracked from breeding areas in Belgium (BE), the Netherlands (NL) and Sweden (SW). Birds from all three areas have a largely overlapping non-breeding range. However, our analyses (in prep) reveal surprising differences in migration corridors and schedules between Marsh Harriers breeding in the Low Countries (BE, NL) vs. those breeding in Sweden (SW).