Facebird. Instagrebe. Tikstork. Linkedpenguin. Twitter. Only one of these is real, and I’m pretty sure that even on Twitter the number of actual birds participating is negligible. Birds do have social networks though—the old-fashioned analog kind, made up of flockmates and siblings and reproductive partners and rivals. Some species, like the Black-capped Chickadee, have tight-knit flocks with elaborate social hierarchies; others, like the Hermit Thrush, live most of their lives alone. These differences are surely fundamental to shaping the lives of birds, from their mundane daily experiences to how they tackle life-or-death challenges. If you spot a deadly predator like a Cooper’s Hawk, do you try to disappear quietly into the brush or do you risk your safety by giving an alarm call to warn your companions?
Social behavior is an odd combination of obvious and very hard to study. On the one hand, if I asked you about a bird species you see occasionally, you could probably tell me whether it is highly social or not simply based on whether you tend to see it in groups. On the other hand, if you wanted to know what determines the make-up of those groups—are they related? similar ages? the same sex? more or less dominant than others? are some learning from others?—you would have trouble answering that without substantial effort. (Birds with plumages that differ by age and sex are an exception here: color-coded for ornithologists’ convenience. But few birds wear all their secrets in plain sight.) Too, observations can easily miss things: for example, what if only some individuals form flocks, but those flocks are much easier to see than lone birds?
Color-banding birds (so you can tell them apart) and then watching them is the classic solution to this problem, and has yielded most of our knowledge about avian social behavior. It takes a lot of time and effort, and it’s not perfect (spoiler: none of these methods are)—if some birds are more secretive than others, color-banding them won’t help you.
In the last decade or so, researchers have started using technology to study social behavior. RFID tags are tiny electronic devices, small and light enough to be glued to a bird’s leg band, that can be detected by an RFID reader if the tag and reader are close enough. Put RFID tags on your birds, put readers somewhere the birds tend to go—like a birdfeeder or nestbox—and you’ll get a log of which individual birds were at the reader at any given time. You can see if certain individuals like to feed at the same time, or if some individuals always leave when the other shows up. You can even do sneaky things like program a feeder to open only when specific RFID tags are registered, so you can make certain feeders only function for certain birds, and then see what that does to the birds’ social relationships. This is extremely cool and I would do it in a heartbeat. However, it does require your birds to use birdfeeders and/or nestboxes, which affects the questions you can ask; and it is, sadly, still quite expensive. (Petition for tech companies to take a break from developing The Internet Of Things long enough to get costs down on the Internet Of Birds please…)
Is there any way that we could know where individual birds are at a given time, that doesn’t rely on human observers’ acuity or expensive tech or birdfeeders? Any other way to get a record of birds that are associated with each other?
Bird banding! We mark the birds with uniquely-numbered leg bands, so we knows who each individual is. We check the nets at least every 30 minutes, usually closer to every 10-15 minutes, so any birds caught near each other must have been near each other within the same <30 minutes. Every bird bander knows, anecdotally, that you catch flocks of flocking birds. You also catch birds associated for other reasons.
Bird banding data won’t be affected by an observer’s ability to spot birds, and it doesn’t require birds to use birdfeeders. The data usually include age, sex, and other individual attributes, so you can investigate how those influence social associations. When I looked at bird banding data from Coyote Creek Field Station, I found patterns like: Song Sparrows tend to hang out with individuals who are a similar age; young Bushtits tend to have more associates than older Bushtits, but only in the breeding season; Song, Golden-crowned, and White-crowned Sparrows do form winter mixed flocks, but they still each associate more with individuals of their own species than with others.
Bird banding data has its own downsides, of course. You probably only detect a small percentage of social associations this way; different species are more or less likely to get caught in a mist net; and you’ll get a few false positives, birds caught in the same net at the same time by coincidence rather than due to an association. It’s noisy data, essentially, just like most behavior data.
There’s another crucial attribute of bird banding data: we have a lot of it already, and we have a lot of it from the past. Some bird banding stations have 40+ years of data. This means that 1) we can partially compensate for things like our low detection rate of associations with the sheer amount of data, and 2) we can look into the past, at trends and changes in behavior. So often we wish we could know what had happened in the past, and so rarely is it possible to actually go back and do a study—but banding data from the past may let us do that!
When I looked at 1995-2019 data from Coyote Creek Field Station, I found that the overall number of associates birds had seems to be decreasing over time. I’ll need to do a lot more work to figure out why that could be happening; but it’s a glimpse at the kind of long-term patterns we can detect with these data.
I’ll continue to poke around in our data, because I’m insatiably nosy about birds’ lives and I still have a lot of questions; but my hope is that other researchers will also use this approach, adapt it to their own interests and ask questions that I haven’t thought of about bird species that I don’t band. Those researchers won’t need to have access to a lot of money or time to use this method—they could be students just starting out. There’s a lot of data out there!
Reference: LaBarbera K, Scullen JC. 2021. Using individual capture data to reveal large-scale patterns of social association in birds. Journal of Ornithology doi.org/10.1007/s10336-021-01863-3 PDF: LaBarbera&Scullen2021