The wheatear's remarkable migration: Alaska to East Africa

A great piece of research came my way today, detailing the migration of the individual wheatears from their Alaskan breeding areas to winter territories in East Africa. We've long known this must happen, as pretty much all the world's Northern Wheatears Oenanthe oenanthe spend the winter in Africa, but now technology has allowed us to follow individual birds on their 14,600km long migration from Alaska to East Africa and back. It's a remarkable story, not least that a 20g songbird can repeatedly do this sort of movement, but that we now have devices that can be attached to such small birds and record their journey. Wheatears are also a favourite of mine, and their migration has been the subject of one of my student's research so I almost feel qualified to make a few comments!

Migration routes and wintering grounds of three northern wheatears breeding in Alaskan (AK) and one in the eastern Canadian Arctic (CN; grey dot, breeding area, blue, autumn migration, orange, spring migration, dashed lines indicate uncertainty in migration routes close to equinoxes). Fifty per cent kernel densities of winter fixes (beginning of December 2009-end of February; purple, bird AK-1; green, bird AK-2; orange, bird AK-3; blue, bird CN-1) are given depending on the sun elevation selected (with 228 for most southern and with 24.58 for most northern densities). Pie charts indicate the proportion of individuals (AK: n ¹/₄ 9, CN: n ¹/₄ 4) originating from one of the three pre-defined wintering regions (red, western; orange, central; yellow, eastern) [8] based on stable-hydrogen isotope (dD) values in winter grown feathers and the dD values within each wintering region (mean+s.d. shown); Credit: F. Bairlein et al. 'Global migration of wheatears' (doi:10.1098/rsbl.2011.1223) in Biology Letters

Female wheatear by Adam Seward. Who couldn't love these birds?!

 It's a simple enough study - they caught 30 wheatears in Alaska (and another 16 in NE coastal Canada) and fitted them with harnesses that had geolocators attached. These devices aren't GPS devices, nor are they satellite loggers, rather they record simply the time and whether or not it is light or dark. From that information it's easy to calculate day length, and if you know the day length and the time of sunrise for any given date you can work out more or less where you are. It's not the meter level precision of a GPS device, but when you're tracking a bird from one continent, across a second and into a third it's precise enough! The problem is, you've then got to catch the birds when they come back to breed the next year. And from our studies I can assure that survival between is only about 60% for an adult wheatear, lower for a juvenile. Plus the bird has to come back to the same place. And what's more, once you've trapped a bird once it becomes much harder to trap it a second time! Unsurprisingly, not so many came back (though 5 of 30 seems rather low), and one had lost it's geolocator, whilst the another outwitted the scientists. [Of course, in the good old days a shot-gun would have been required...] So they only got three loggers back from the Alaskan birds, and one from the Canadian birds. But even these results are amazing!

Wheatear pair by Adam Seward. It's cold up north!

They backed up these results with an additional study of the chemicals in the bird's feathers. Many wheatears moult some of their feathers whilst they're spending the winter in Africa. As they grow new feathers they will be making use of chemicals picked up in their food. Now, just like when we discussed carbon dating we mentioned chemical isotopes, the same principal is used here. In this case it's not Carbon we're interested in, but the hydrogen in water: the same hydrogen element can be found in slightly different forms, and the frequency of the two forms varies spatially. People have mapped this spatial variation, and once ingested by an animal and fixed into biological materials (in this case feathers) the ratio of the two is fixed and by comparison with global maps it's possible to work out where the feather was grown. Neat! So the authors also did this - the results of that are the little coloured wheels in the breeding ground with the source of the material the same colour mapped across the African wintering ground: it's clear that c.50% of Alaskan birds wintered in East Africa, whilst a similar proportion of Canadian birds wintered in West Africa, exactly as we'd always expected.  All in all a very nice study.

This isn't the longest migration of any bird (Arctic Terns must hold that record), but it probably is for any passerine. And there are a few more interesting details we can fit in too, such as the flight speed. That bird crossing the Atlantic to Britain averaged 850km per day, for four days (perhaps stopping in Greenland), which is pretty remarkable. (Actually, I happen to know that the flight speed of a wheatear is 47kmh, so that's about 18hrs flying time per day, if there's no tail-wind. And I also know they like to make this hop when the wind is in their favour, so if it stopped in Greenland it will have bee a short stop unless there was a strong tailwind). The movement across Asia is also nice to study, showing typical travel times of migrations, being 91 days on the way south (160km / day), and a faster 55 days (250km / day) when the rush is on to get to the breeding grounds in the spring. (Curiously, the Canadian bird was faster in autumn). Knowing that the lean weight of a wheatear is about 20g (not too different from a house sparrow), and what the fuel efficiency rate is, if a wheatear were to do the 14,600km trip in a single flight it would need about 20 times it's own weight in fat. It's unusual to see a wheatear with this much fat(!), with 50-60% of lean body mass the typical departure fuel load for long flights in this species, and much less when there's good feeding to be found en route. In fact, our birds on autumn migration carried loads of fat that would carry them on average about 2000km before needing to stop and refuel, so if we assume this is much more reasonable, 2000km requires about 10g of fat (50% body mass) the total trip might well have been carried out in 7 or 8 stages, interrupted by feeding stops in nice areas. We've found that this race of wheatear can, if the conditions area really good, deposit an average of about 5% of their body mass (1g) of fat per day (exceptionally the Canadian race can get up to 20-25%). There's at least 300hrs of flying time required to do the distance (~13 days). So each stop over to reach 10g and the next flight might be about 10 days, giving about 73 days of travel. Obviously to make  the trip in 55 days they've got to be faster, making sure they pick the best places to refuel and average stop overs of only 5 or six days. Pretty had work though, with no time to rest! And imagine repeatedly gaining and losing 50% of your body weight - I think it would kill us!

How we know the extra stuff! Colour-ringed wheatear weighing itself
and snacking on mealworks. Big brother is watching... Thanks Adam!

Anyway, what surprises me most about these results is the fact that some wheatears breeding in Alaska seem to have wintered in west Africa (having completed an even loinger migration!) and some Canadian birds here in east Africa. That's particularly interesting to me as it means that some of 'our' colour-ringed birds from Europe and Greenland might have be wintering on the plains near me! I'm going to have keep my eyes open...

PS, thanks to the very nearly Dr Adam for providing the photos of his wheatears! I'm looking forward to cute squirrels next...

Main reference:

Bairlein, F., Norris, D., Nagel, R., Bulte, M., Voigt, C., Fox, J., Hussell, D., & Schmaljohann, H. (2012). Cross-hemisphere migration of a 25 g songbird Biology Letters DOI: 10.1098/rsbl.2011.1223