The American Robin is celebrated in the popular imagination as a harbinger
of Spring, but the species is proving quite numerous lately on Audubon’s
annual Christmas Bird Counts (CBCs). In Washington, D.C., for example,
recent years have seen some striking totals: 1,259 robins last winter,
for example. The most casual observers have remarked on the change. We
have even gotten calls to our hotline from concerned citizens wondering
what is going on (not to worry folks, this isn’t Alfred Hitchcock’s The
Birds come to life). The question is, has there been a real increase or
have we for some reason only recently been paying more attention? Or do
their numbers simply fluctuate widely, up one year and down the next, as
they move about? These are exactly the kinds of questions that a review
of CBC data can help answer.

These counts first began in 1900 in a small number (25) of U.S. locations,
but they have since expanded greatly in number (over 1,800 with over
50,000 participants annually) and are now international in scope, so there
is a lot of information to consider. Although I have heard the counts
described as “unscientific,” (most recently by the Washington Post) I am
puzzled by this characterization. What exactly does it mean? Does it
mean less so than a laboratory experiment, or a public opinion poll, or a
study of American eating habits carried out on a group of twenty people?
Such a comment seems to ignore the real potential of the CBCs to
contribute to our knowledge of bird population dynamics.

True, there are many variables in such a real-world enterprise, but one
would be hard-pressed to gather information on a comparable scale in the
confines of a laboratory, or by means of a necessarily smaller and more
controlled study (even if one could get the funding to carry it out). No
study of bird behavior reveals absolute answers--they are too complicated
and varied for that--but in the case of bird population trends, for
example, the CBCs can signal a problem before it is noticed by the
scientific community at large, and indicate the need for action before
it’s too late. And that’s only one of the broader applications that come
to mind. Furthermore, now that all this data--over a century’s worth for
many count circles--is online in a single database that can be queried, I
think it is much more amenable to scientific inquiry than is popularly
understood. If you doubt this, visit the Audubon site, which lists
literally hundreds of scientific articles based on CBC data in the
bibliography section.

The fact that robins were not always so numerous at this time of year, at
least in our latitude, came to my attention recently while reading the
pioneer naturalist William Bartram. In his discussion of bird migration
in his Travels (Philadelphia, 1791), he notes that relatively few land
bird species were found in his native Pennsylvania in the colder months.
After presenting his short list, he continues, “add to these the blue
bird, mock bird, and sometimes the robin red breast, in extraordinary warm
winters.” As I read this, I remembered hearing about the calls to the
hotline, and recalled media reports of global warming, and wondered if
there was a way to use CBC data to confirm what Bartram asserted (he’ll
forgive me if I focus on D.C. as an example, rather than Pennsylvania).

Data for Washington’s Christmas Count goes back to 1943. Not until 1950,
when 10 were found, is the American Robin first recorded. Twice during
the 1950’s, in ’55 and again in ’57, the number jumps to over 200 birds,
but in some years falls back to zero; the average is around 20 or so. In
the ’60s the average drops to just over ten birds but, as the ’70s begin,
there is a slow, steady increase, and by the 1980’s we are into triple
digit numbers. In 1990’s, however, things really get going: the yearly
average for the “go-go” decade is a whopping 476! Based on my first pass
at the data, then, it did seem that the bird’s winter numbers were up
significantly.

If we want to be more scientific we should adjust for level of effort,
since the exact number of observers and time spent in the field varies
from year to year. Using Audubon’s database, this is simple to do by
comparing the figures--fractions, actually--for birds per party hour. For
the first 20 years after robins make their appearance the number seen per
hour hovers around .25 (i.e., it took about 4 participant hours to get 1
bird). In the next twenty year period this figure more than doubles, to
.60, indicating it took less than half that time to get a robin. So a
real increase occurred during that period, and it is not merely an
artifact of a change in level of effort, since any variation in the
popularity of CBCs has been accounted for. As for variations in observer
skill, I doubt anyone would claim that the average CBC participant is
significantly better or worse at recognizing an American Robin today than
in any former count period.

Now let’s take a look at the 20 year period 1983 to 2002 to bring
ourselves fully up to date (data from the Dec. 2003 count is not yet
compiled). Here are the figures for the American Robin in D.C. during
that period from the Audubon website. [Note the count number is typically
1 higher than the year (in December). The site also provides the number
of participants (usually over 100), but this column has been omitted for
simplicity’s sake.]

Count Number
Number
Number/Party Hr.

84 (Dec. 1983)1140.37
85590.21
863441.54
87440.17

882931.2
891810.76
901750.96
911380.88
922170.99

933391.77
943151.39
955062.51
965663.17
977093.65

986684.02
996933.93
1006132.74
1011340.7
1029345.25

103 (Dec. 2002)12597.55
AVERAGE4152.2

So how can we explain this substantial jump in the wintering robin
population? Keeping in mind that the same basic factors that increase
human populations also operate in the animal world, i.e., increased food
supply, improvements in the economy (in the case of birds, the ecology) I
want to suggest a link between our success and theirs. In short, they
have increased their numbers as our own fortunes have boomed (remember,
I’m speaking long-term here). Now before you dash off a buy order to your
broker, let me explain.

As the 20th century progressed, the economy grew and so did the U.S.
population; mid-century saw a post-war boom (remember the decade the
robins were first reported?). With this expansion came an increase in the
need for housing and, as a result, wooded areas were increasingly cleared
for suburban developments. Where tall trees had been, lawns and
ornamental plantings grew up. Now, what bird was poised to take advantage
of this American approach to prosperity? None other than the American
Robin, a species that has a distinct preference for cleared areas, and
which could now freely hunt these new lawns for worms and search the
shrubbery for berries. In short, human changes to the landscape
coincidentally resulted in an increased food supply for these birds, even
though other species more at home in woodlands probably suffered as a
consequence. If that is true, it suggests a lack of balance in our
approach to growth; a mid-course correction is therefore needed if we wish
to maintain real biodiversity.

As the quote from Bartram above suggests, there is another key factor:
those “extraordinary warm winters” are not so extraordinary any more
(again, I’m speaking long-term). Whatever you may think of theories of
human-induced global warming (another by-product of prosperity?) it is
true that robins must seek out unfrozen ground in the winter months if
they are going to probe the dirt for their typical prey. As the current
data indicates, they rarely need to fly farther South than Washington in
order to find it. Of course, such a complex phenomenon probably has other
contributing causes, as well, and more detailed analysis would be needed
to reach a final conclusion, but we did reach a valid starting point, in
any case.

To carry out your own analysis of the American Robin, or any other
species, visit Audubon’s CBC website:
http://www.audubon.org/bird/cbc/index.html [1]
and click on “Historical
Results.” Users can query the database in several ways, and many types of
maps, tables and graphs can be created directly from the site.