Roller ring recoveries

Posted on Aug 10, 2016

Our recent analysis of Roller ring recoveries has just been published in the Journal of Ornithology! It’s open access – take a look here.

I started thinking about ring recoveries after we’d successfully tracked Rollers to and from their south African non-breeding grounds using solar geolocators (see the paper here). Although this revealed a great deal about the annual movements of Rollers from across Europe, the picture was still incomplete. Because geolocators are archival (i.e. they don’t transmit data), they must be recovered. This means that we only get data from birds which return to the tagging (breeding) site, so we have no idea where and under what circumstances migrants die. We also don’t know much about what first year migrants do; juveniles generally have a pretty low return rate to the natal site, so it’s usually uneconomical to put geolocators on them.

Ring recovery analysis has its limitations too, but it provides a complementary approach to archival geolocators. It seemed silly not see what ring recoveries revealed, though we should probably have attempted this before using geolocators. Nevermind.

Ornithologists have been ringing Rollers across Europe for ~80 years, though in fairly small numbers. I’m not aware of any national migration atlases which include the Roller, but in theory all records of ringed Rollers re-encountered (dead or alive) away from their original ringing site should be held in the EURING databank. Unfortunately, many records seem to be missing – this is a bit worrying*, but I suppose not much of an issue for other species for which there are relatively many recoveries. However, with only 11 EURING records of Rollers ringed during the breeding season and recovered on migration, it was clear that we needed to widen our search. So, I started getting in touch with Roller ringers from across Europe in an attempt to collate their ring recovery data. As with our geolocator study, I think this pan-European, collaborative approach has been pretty effective.


A Latvian Roller nestling, sporting a coded plastic ring – will it be re-sighted on migration?

So, after poaching data from ringers in Latvia, Hungary, Lithuania, Bulgaria, France and Serbia, and combining this with the EURING data, we had 41 records of ringed Rollers recovered during autumn migration, 16 during spring and (as expected) just 1 from the sub-Saharan non-breeding period.

Just as geolocators suffer from a bias towards telling us only about successful routes, ring recoveries can also be biased; a ringed Roller is more or less likely to be a) encountered and b) reported in certain parts of the world and during certain times of the year. So, the recovery map (below) should be taken with a pinch of salt. It reveals a generally similar pattern to that shown with geolocators – a broad-front autumn migration of birds from Eastern Europe across the Mediterranean basin.


Migratory movements of European Rollers revealed by ring recoveries

The spring movements are also revealing, with several recoveries from further east in Arabia. Our geolocator work already demonstrated that Latvian birds use this route; now we know (or at least, now we’ve published that we know) that Hungarian and Serbian birds also fly home via the guns of Arabia. Whether or not all individuals from these populations use this route remains to be seen – it’s possible that the higher encounter rate (due to hunting) results in birds using this route being over-represented in our data-set (though you might expect a correspondingly reduced reporting rate from these hunters).


Rollers shot on spring migration through Arabia – awful. Shooting was the most common known cause of death in our ring recovery data-set


What about juvenile migration? Our sample sizes are pretty small, but our results point to a couple of interesting conclusions:

  1. The direction of autumn movements appears to be more variable in juveniles than adults. This makes sense given that naive juveniles are potentially migrating using a simple ‘clock-and-compass’, and are pretty bad at compensating for (e.g.) wind drift. In contrast, more experience adults might have a ‘map’ sense too, which allows them to keep a more constant bearing. This is all pretty speculative (though not altogether surprising) and definitely deserves more work. I’d love to track some juvenile Rollers with satellite tags**.
  2. The overwinter survival of juveniles appears to be lower than for adults. In the autumn, our recoveries are juvenile-biased (not surprising given that the breeding season has just finished), but come spring the majority of recoveries are of adult birds. Again, this is far from conclusive, but supports more robust studies from other species which suggest high juvenile mortality during the non-breeding season.

I hope I’ve demonstrated that ring recoveries can complement direct tracking, providing additional insights into bird migration. Whilst the Roller probably isn’t an ideal species for this sort of analysis (there are very few recoveries), we still managed to contribute a bit more to our understanding of what happens during the non-breeding season.

*As far as I can tell there are three reasons why a recovery might not be in the EURING databank. (1) it takes up to a couple of years for the data to filter down from the the local schemes to the central database. (2) many historical records – especially from eastern countries – have not yet been digitised, so can only be accessed following a direct request to the national ringing scheme. (3) colour-ring resightings aren’t necessarily reported to the national schemes.

**Although satellite transmitters are now just about light enough to use on Rollers, they’re quite expensive – particularly if you’re attaching them to juveniles which probably suffer high natural mortality rates.

Leave a Reply

Your email address will not be published. Required fields are marked *

Time limit is exhausted. Please reload CAPTCHA.