After sending the message yesterday, I thought "why didn't I look at CBC
results for all 30 years?" So I went back into American Birds and tallied
BTPI counts for 3 decades, as follows. Average per count in 1963-72, 78.7
(range 2-182); 1973-82, 193.3 (range 45-409); 1983-92, 145.0 (20-495).

Much more interesting are the raw data, which I present graphically by
letting a dot represent 10 birds.

63 - .....47
64 - .................170
65 - ............115
66 - ..............143
67 - .......74
68 - ...34
69 - 2
70 - .8
71 - .12
72 - ..................182
73 - ...............................308
74 - .........................................409
75 - ..........103
76 - ..........................259
77 - ..........................256
78 - ......................221
79 - .....45
80 - ......55
81 - .........89
82 - ...................188
83 - ..................................................495
84 - ..........102
85 - ....................204
86 - ..20
87 - .........90
88 - .............................291
89 - .......73
90 - ....37
91 - .......71
92 - .......67

Darn, I never resubscribed to Audubon Field Notes after they changed it.
Can anyone furnish the 1993 count?

It seems as if there are pigeon peaks and pigeon poops, with high numbers
from 1964-66, 72-78, and 82-85, low numbers inbetween, and an isolated peak
in 1988 (I think that was the year I started feeding them). Or were 86 and
87 anomalously low during a peak period (bad weather on the counts?)? Is
this species really cyclical, and if so, why? Are any of the fruiting
trees at which it feeds (madrona one of the most prominent) cyclical, with
boom and bust years, as we know is the case for some conifers and broadleaf
trees? The tree cycles are typically shorter than the pigeon fluctuations
shown here. And the prevalence of bird feeders should cancel out the
cycles of natural foods.

If one extended this data set, one would expect pigeon numbers to go up in
a few years. If they don't, maybe that is a sign of decline. By the way,
note that the peak was higher in 1984 than in 1974, which was in turn
higher than in 1963. More feeder watchers? But the peak numbers in the
early 70s were overall higher than those in the 80s. Of course, these data
are merely indicative and should be supplemented by looking at
breeding-bird survey data and whatever other information we can muster. I
wonder if anyone who has the BBS data wants to try a correlation with CBC
data. You could try correlating both the year before and the year after
the CBC with it.

Dennis Paulson, Director phone: (206) 756-3798
Slater Museum of Natural History fax: (206) 756-3352
University of Puget Sound e-mail: dpaulson at ups.edu
Tacoma, WA 98416