by Darryl Jones
“This is a particularly interesting area to watch for cardinals,” says David, pointing to the eastern US-Canada border region. As the animation advanced gradually from 1989 to the present, the spread of yellow dots, slowly though unsteadily, pulsates northward until the data stop at last year’s records; yellow dots are now spread all over Nova Scotia and the Maritimes as well as central Quebec and Ontario. What this demonstrates—explicitly—is that Northern Cardinals are now seen regularly in February much farther north than just a couple decades earlier. “Cardinals are iconic and everyone wants to see them in their gardens, so their movement north in winter—or at least their willingness to stay put during the depths of winter—has been of particular interest to Project FeederWatch participants,” explains David. The same thing is happening with less charismatic species too: Tufted Titmice, Carolina Wrens, Red-bellied Woodpeckers, and others have all shown a similar pattern of being observed regularly farther north than their traditional distributions.
The obvious question is “why?” We need to be careful: the most obvious answers are not necessarily the best explanations. As tempting as it might be to claim that these spatial patterns are all about feeders—or, indeed, to maintain that it has nothing at all to do with feeders—the reality is likely to be a lot more complicated. Certainly, many observers and feeders have noted the northward movements of the cardinal over recent years, but a major range expansion of this species has been under way since the late 1800s. The first Arizona Cardinals were recorded in the 1870s. The birds apparently first reached Toronto during the 1930s and spread west into Minnesota around the same time. These significant movements were occurring long before bird feeding was a common practice, though there is equally no doubt that feeding has played an important role. A recent study of cardinals in Ohio found them to prefer urban areas to other habitats in colder periods and concluded that this was probably due to general characteristics of cities such as dense undergrowth and warmer microclimates.8 The availability of food was found to be particularly important in winter, but, perhaps surprisingly, fruit-bearing trees and shrubs had a much greater influence on the presence of the birds than the number of bird feeders. In other words, feeders are just one of the features that cardinals weigh up when they are assessing a location. Obviously food supply is a fundamental factor, but it is not all about the stuff humans provide. Naturally occurring berries, seeds, buds, and fruit—including those found in gardens—are always going to be crucial; the significance of feeders will probably depend on the severity of the conditions.
In cold conditions places with concentrated human activities are also warmer than the woods and countryside that surround them. Many species—including the cardinals of Ohio—do not have to read the scientific literature to appreciate the heat island effect of towns and cities and move in to take advantage of the milder temperatures in winter. This is also an effect occurring on a global scale. Everywhere, as the climate warms, thousands of species are steadily shifting their distributions pole-ward, as the historical climatic barriers associated with the cold move farther north (or south for the Southern Hemisphere, or up mountains). This is not the place for further discussion of this profound global phenomenon, though it is important to note that huge numbers of birds, mammals, amphibians, reptiles, fish, and insects are on the move. For our purposes, climate-induced changes in bird distributions add yet another possible influence to the mix.
David Bonter draws my attention back to the screen in front of us: a new animation is showing the Project FeederWatcher records of the Carolina Wren, a small, seemingly delicate species, gradually moving above the Canadian border since the turn of the century. “We always knew that this species sometimes ventured north a little in mild winters, but this data shows that some of them are now living permanently in the Montreal area—and part even further north—in February!” That’s a significant shift in their range in little over a decade; something significant is behind movements like these. The Carolina Wren is a largely insectivorous species, a bird far less interested in the seeds of typical feeders, though they are partial to suet and fat balls. For Carolina Wrens, the influence of feeders may be even lower than for other species, but we can’t be certain yet. “To some extent, it’s the same with all the birds moving north. Is it climate, garden resources, natural food supply, warmer city environments? All of these? Can we actually separate them in any meaningful way?” David’s candid questions suggest an unsettling mix of scientific curiosity and practical difficulty; these are really important queries but so hard to investigate.
David Bonter pushes back his chair and continues on this theme of thwarted ambition. “My biggest professional frustration with Project FeederWatch is in its extraordinary success in collecting enormous amounts of data but the lack of time and resources we have to do the necessary research that could answer so many really important questions. I have so many half-written manuscripts but no time to finish them! But it is just as much my fault too: I’m always trying new ways to investigate birds on feeders.” Suddenly, David is describing a recent study, his frustration forgotten in his enthusiasm to describe what he has recently discovered.
As anyone with a feeder knows well, the number of birds making use of the food provided can change dramatically throughout the year. Sometimes there are clouds of them bickering for a perch or space on the platform, while at other times there is hardly a visit all day. Understanding the reasons behind this variation is one of the fundamental dimensions of bird feeding research, but the challenges are formidable. One of the main difficulties is knowing what individual birds are getting up to. Do certain birds visit continuously, others only occasionally? Or is there a constant flow of new birds, stopping briefly before moving on? To be able to address these sorts of questions, we need to be able to recognize individuals. Elsewhere (see Chapter 5) I describe the work of Margaret Brittingham, who transformed our understanding of the behavior of Black-capped Chickadees at feeders in winter. Her insights were only possible because she was able to distinguish between individuals, based on the color-banding (or “ringing”) of hundreds of birds. This is standard practice among ornithologists today, but it only works if it is possible to detect and “read” the tiny, plastic colored stripes on a constantly moving bird. That can be hard work, limited by human sight and the availability of observers. But what is happening when the researchers are not around?
A new approach to this problem is to have the feeding apparatus itself automatically record the visits of individuals birds. Using technology similar to the PIT tags most of our pets now carry, minute, individual-ized electronic tags can be attached to the leg bands of the birds. When these tags are detected by a receiver set up at a feeder, the identity of each bird’s individual radio frequency is logged, providing a precise record of each visit of each tagged bird. This technique, known as radio frequency identification (RFID) is already providing extraordinary new information on the foraging behavior of birds in relation to feeders. “Let me illustrate what I mean,” said David, pulling out a writing pad and pencil. “We tagged a big sample of chickadees in a woods near houses and fixed the RFID receivers to feeders in back yards nearby. Our main interest was in the use made of feeders over the whole year, a timescale that would be very difficult simply using human observers. The technology is so robust and the tags so reliable you can set up the system and it hums away for months, diligently logging every bird.”
On the pad David sketches a graph showing the visits per day for the first year of the RFID study. The rate of visits was typically moderate from January and dipped even lower during late summer. But in August and September, daily visits suddenly shot up dramatically, with David’s drawing showing a huge and abrupt spike, before falling just as rapidly to “normal” levels by October. “This spike coincided perfectly with our pre-diction of the prewinter seed-caching period. These tiny birds are preparing for the cold that is just around the corner by storing seed away. When I saw these data I remember thinking:
‘We are so good!’ This is just the sort of evidence we needed to demonstrate the significance of providing supplementary food for chickadees. But I should have known it is never that simple. In the second year, it was as though all of the feeders (and all those expensive detection gadgets) just didn’t exist; instead of that spike in visits we saw in the previous fall, this year numbers stayed way down. What were they doing instead? Feasting on the massive harvest of ash tree seeds that occurred that year! It was a sobering lesson in assessing our influence on the lives of these birds. Again, thankfully, the birds were sensibly assessing and utilizing all the available foraging resources, and this year, it was all about the natural stuff.” Sometimes, it seems, feeders are crucial; at other times, trivial.
Sometimes, on the other hand, feeders really appear to be essential for the very existence of a species in certain locations. As another example of the changing distribution of a bird species revealed by Project FeederWatch data, David selects the February records of a West Coast favorite, the exquisite Anna’s Hummingbird. Although their stronghold is in California (where 75% of project participants have hummers visiting their gardens), their distribution has been extending ever northward. The animation shows patchy sightings in Oregon and Washington during the 1990s, but this fills out in the 2000s, and by 2014 Anna’s have a solid presence throughout the western coast and well into southern British Columbia. “This is certainly one of the strongest movements from the FeederWatch records,” says David. But the reasons underlying the phenomenon seem a little clearer for these hummers: the widespread planting of winter-blooming ornamentals and the sheer abundance of hummingbird feeders. A little consideration of the map and the continuous yellow dots spreading into lower eastern Canada—in the depths of winter—brings up the incongruous image of exotic hummingbirds flitting around a snow-bound landscape. Friends from Seattle recently sent photos of Anna’s Hummingbirds amid large icicles, sipping eagerly from a specially heated sugar-water feeder. This was such an unlikely, almost unnatural, scene, probably only possible through direct and sustained human intervention. The admittedly marvelous sight of hummingbirds in the bitter cold also generates its own questions: Would these birds even be there without the feeders? Should they be? These birds are famously territorial, so what happens when a feeder freezes or is neglected? Do the feeders actually improve their chances of survival over the winter? And given that hummingbirds are adored everywhere, and that people will never stop wanting to see them—even (especially?) in colder parts—does it really matter?
“There are just so many important questions to ask,” says Emma Greig, as we sit in an observation room overlooking Sapsucker Woods Pond. The pond, which was hosting dozens of waterbirds only a few days earlier, is now fully iced over. A few hardy ducks remain, hunkered down in the drifts. It is bitter out there, and the numerous feeders in sight are busy: chickadees, Tree Sparrows, House Finches, and nuthatches apparently amicably sharing the bounty. “Look at these poor birds. We sometimes feel sorry for them because we are safe and warm inside while they are stuck out there. But they are used to these conditions. They have evolved to cope with all sorts of weather or they wouldn’t be here.” This simple observation raises one of the really big questions that continuously cir-culate around the effect of feeders, and an issue that Emma is constantly asked about: Does feeding lead birds to abandon migration?
“Oh yes, that’s a big one!” Emma sighs and glances out at the feeders. “Certainly, there are examples of migratory birds staying put through the winter—Canada Geese are probably the best-known example—and feeding has been suggested as one of the reasons. But in almost all of these cases, it seems to be mainly some individuals rather than a whole species or population that are remaining behind, or perhaps groups of birds in specific circumstances. I can’t think of any species or even population where migration has just been abandoned entirely.” Emma pauses to reflect for a moment, possibly recalling her time spent in Australia where she completed her doctoral studies on tiny fairy-wrens. Like the vast majority of Australia’s avifauna, these birds are decidedly nonmigratory; most are either permanently sedentary, living their entire lives in the same place, or are nomadic. The epic movements of huge numbers of birds north and south, so familiar and predictable throughout the Northern Hemisphere, are just not a prominent phenomenon in the southern parts of the globe, although the long-distance movements of waders and shorebirds are notable exceptions. My interpretation of Emma’s expression suggests that she is considering a quick tutorial on bird migration for the visiting antipodean: Bird Migration 101.
The extraordinary migratory journeys made by many birds are ultimately controlled by the internal hormonal and physiological cycles of the birds. These inner cycles, which function daily as well as seasonally, are fairly strictly linked to grand external cycles such as the slight but utterly rigid variation in day length. But birds are not robots; no year or season is exactly the same, and they must be able to alter decisions about when to act (start building up reserves for the flight, or gather in groups, for example) if the conditions are not suitable. There are many such potential influences that may trigger these migratory behaviors in birds— temperature, changes in vegetation, food supplies—all of which can vary from day to day and year to year. If the birds were to respond primarily on these fairly volatile signals, they could fly off too early or too late, both with big risks. “After all,” explains Emma, “the reality is that they need to time their travels correctly to the breeding areas or it may be too cold, too early, or too late when they arrive. Similarly, once breeding is over, they need to move at a suitable time to maximize the survival chances of their young but avoiding the coming cold, storms, and general lack of food. To get the timing right, most songbirds seem to have evolved over the eons to respond to the small but crucial changes in day length as they prepare and eventually leave.”
But this otherwise sound, reliable system seems to be coming unstuck. Across the globe, birds of all types, from tiny warblers to large shorebirds, are changing their age-old traditions, once regarded as fixed and immutable, and are arriving increasingly earlier in spring as well as de-laying their autumn departure dates. This general pattern is most clearly being attributed to the warmer weather of the past few decades; yes, climate change is also changing ancient migration patterns. This seems to suggest that the apparently fundamental day-length trigger, which does not alter, can be overridden by the influence of temperature. “Just how far this might go is really unknown at the moment “ says Emma. “But it is certainly going to affect the pattern of migratory birds using feeders. What used to be traditional arrival times for all sorts of species are already very different from a decade ago. It’s a big point of discussion.”
Emma pauses, perhaps regretting the decision to summarize several centuries of international research so succinctly. I, however, am grateful for such a concise explanation. “Given that grand scenario,” I offer, “feeders do seem to be less significant than other influences that migrating birds are assessing.” Emma agrees. “Foraging resources are always going to be of critical importance, but it you are a small and vulnerable species, winter is likely to be just too tough. Better to head south, even if there is plenty of food. It would not be a sensible idea to become overly reliant of any particular food type, including feeders.” She pauses, looks thoughtful and adds quietly, “Of course, there are plenty of apparently small and vulnerable species which don’t move and are even moving further and further north. Oh, it seems so complicated, but that is what you get with functioning ecosystems. To survive birds must be able to adapt to changing circumstances. They have been migrating back and forth for millennia. People and their feeders are really very recent additions to the landscape, so it’s probably too early to know whether all that additional food is going to have large-scale evolutionary effects.”
Our conversation returns to the difficulties in trying to isolate the possible effect of feeders from all the other influences, human and
natural, that play a part in the way that birds interact with feeders. From a scientific research perspective, it is virtually impossible to control all these factors with any meaningful precision, especially in complex places like towns and cities. This is one of the reasons that much of the research on bird feeding has been conducted away from people and where true experimental conditions can be imposed. This simplifying of the variables is what defines the supplementary feeding experiments that will be discussed in the next chapter. Standard experimental design requires a “control” where everything is the same as the “treatment” but without whatever that treatment might be. The big problem for those of us trying to discern the effect of feeders is that control sites are difficult to find. As David Bonter exclaims, in places like the Northeast and along the west of the United States, “feeders are everywhere!” Elsewhere, this is certainly not the case. Assuming that the distribution of Project FeederWatch contribu-tors reflects that of bird feeders generally, the map of participants across North America (easily found on the PFW website)9 shows that the density of feeders tends to mirror the broad density of people across this vast landscape. This variation in the number of feeders per area actually provides an opportunity for investigating the feeder effect on a grand scale.
