 |
|
|||||
Variation in Black-capped
Chickadee Group Size
FeederWatch data shed light on the winter biology of this Every two weeks Project FeederWatch participants count and record the numbers and kinds of birds they see at their backyard feeders. What do we do with these data? Past analyses have emphasized unusual events, such as last winter’s "Red Invasion" of redpolls and Red-breasted Nuthatches (Birdscope, Autumn 1994). But FeederWatch data also provide insights into the winter biology of widespread bird species. By "widespread," we mean species that a large proportion of FeederWatchers see in their yards. "Widespread" doesn’t necessarily mean "numerous." For some widespread species, the number of birds that any one person sees can be relatively small. Why? The birds are resident on a winter territory. Species such as chickadees, tit-mice, nuthatches, and woodpeckers defend territories during the winter as well as the breeding season. Although breeding territories are usually occupied by a single pair, in winter a group of birds may defend a territory. We used FeederWatch data to investigate variation in winter group size in the Black-capped Chickadee, a common feeder bird. Other population biologists, studying individually marked birds, found that winter flock sizes vary greatly, from an average of 8.05 chickadees per flock in western Massachusetts to 3.3 chickadees per flock in southern Illinois. In her book, The Black-capped Chickadee, ornithologist Susan Smith identifies some factors that seem to influence flock sizes for birds in the family Paridae (which includes the Black-capped Chickadee), both in North America and in Europe. Larger flocks tend to occur in northern latitudes, or where supplemental food is offered. Flocks tend to be smaller when a variety of parid species occur in the same area. The results are suggestive, but they’re based on just a few studies, from a handful of localities. The data collected by Project FeederWatch offer a golden opportunity to examine whether these trends exist on a continentwide scale. We describe here the results of some preliminary analyses of variation in chickadee flock sizes. We looked at differences in group sizes over the course of one winter, from year to year, and from region to region. Seasonal Variation in Group Size After they become independent from their parents, young Black-capped Chickadees disperse and look for places to settle. They settle in by September, usually joining a former breeding pair on their territory. Thus, a flock’s size is determined early in autumn. The flock remains fairly stable throughout the winter, although sometimes it shrinks as birds die.
To describe winter changes in flock sizes, we examined data from FeederWatchers who recorded Black-capped Chickadees during each count period of the 1993-94 season. In the analyses that follow, we refer to the maximum number of Black-capped Chickadees counted by a FeederWatcher as the "group size," and we assume that it reflects the number of birds living around the feeder (the flock size). Thus we consider Feeder- Watch counts to be an index of flock size: where flocks are larger, feeder counts will be higher. By using data from Feeder- Watchers who always saw chickadees, we could compare average flock sizes from one count period to the next, because all factors except the date remained the same. Figure 1 shows the mean group sizes for U.S. states in which more than 50 FeederWatchers provided data. Two results are apparent. First, groups remained very stable over the entire FeederWatch season, although the data suggest that a small decrease occurred toward the end of winter. (These results are a little surprising; 1993-94 was a very cold winter, and we might have expected higher mortality.) Second, pronounced and consistent differences exist between states. Regional Variation We analyzed the variation between regions and between years using only data collected during count period #5 (mid-January)—a total of 13,906 counts over six winters. The basic unit of analysis was a state or province, but we grouped some states and provinces into regions to obtain adequate sample sizes. In the end we analyzed data from 24 different regions. The average size of Black-capped Chickadee groups counted at feeders varied greatly from region to region. In the New Hampshire-and-Vermont region, the average number over the course of six winters was 7.3 birds, compared to 3.0 in Ohio, 2.9 in Illinois-and-Indiana, and 2.7 in Utah-and-Colorado. We wanted to know whether these differences were meaningful, so we carried out statistical tests called analyses of variance. We found that in each of the six years, the probability that the differences between regions were simply due to chance was less than 1 in 10,000. In other words, the differences between regions are real. Figure 2 shows that in general, chickadee group size decreases from north to south. But the pattern is not really simple. The region with large groups (averaging more than five birds per feeder over six years) does not include Saskatchewan and Manitoba, but does include the northern Rocky Mountain states. Large flocks occurred farther south along the East Coast than in the West. Medium-sized groups (averaging 3.5 to 4.9 birds) occurred in Washington, Minnesota, Wisconsin, Michigan, New York, Pennsylvania, Connecticut, and Rhode Island. Medium-sized groups also occurred in five more- southerly states where Black-capped Chickadees (the Appalachian subspecies Parus atricapillus practicus) live only in the mountains, and Carolina Chickadees live in the surrounding lowlands. Small groups (2.7 to 3.2 birds) occurred along the entire southern part of the chickadees’ range, with the exception of the Appalachian states just mentioned.
Variation Between Years Chickadee group size also showed strong variations from year to year. These variations did not occur in parallel in the East and the West, however. Figure 3 shows the mean number of birds in six eastern and three western regions during the last three winters. In all six eastern regions, flocks were largest in 1994, reflecting an important irruption of Black-capped Chickadees that occurred in that winter. In one western region (Washington), bird numbers were also highest in 1994; however, in two other western regions (Western Canada and Oregon), bird numbers were lower in 1994 than they had been in 1992 or 1993. Conclusions If our assumption that Feeder-Watch counts reflect chickadee flock sizes is correct, we can draw four conclusions from our analyses of 13,906 observations over six winters. First, chickadee flock size varies by region. Second, although flock size generally decreases from North to South, other factors probably are also important. Winter temperatures, the number of other parids, or overall breeding density might play a role. Third, flock sizes vary from year to year, probably a reflection of overall population density and irruptions. Finally, eastern and western chickadee populations fluctuate independently of each other. |
