Federal regulation and coordination of bird
banding in the United States and Canada began in 1920 (Pitts 1994).
We retrieved from the Bird Banding Laboratory all available
records that documented the death of Eastern Bluebirds (Sialia
sialis) that had been banded during 1920-1990. We present here a summary of the dates of mortality, causes
of death, and geographic locations of mortality.
Additionally, we consider the effects of sex and age on
mortality.
Methods
In
response to our request for banding and recovery records, Danny
Bystrak of the U.S. Fish and Wildlife Service's Bird Banding
Laboratory provided us (in March 1993) with records of Eastern
Bluebirds that had been banded during 1955-1990 and records of
recoveries for the years 1920-1991. Banding records for the period 1920-1954 had not been
computerized and were not available.
The
North American Bird Banding manual (Canadian Wildlife Service and U.S.
Fish and Wildlife Service 1991) defines the term "recovery"
as a banded bird that is killed or found dead and then reported to the
Bird Banding Laboratory. However,
many biologists, including the writers of the Bird Banding manual, use
the term "recovery" more generally to refer to either a
banded bird that is found dead or a banded bird that is captured alive
outside of the area where it had been banded.
In fact, the Bird Band Recovery Report form from the Bird
Banding Laboratory asks whether the bird was alive or dead at the time
the band number was read. Many of the "recovery" records we obtained from the
Bird Banding Laboratory actually describe the recapture of live birds
that had been previously banded.
We
entered each recovery record into a database with files containing:
Permit Number of the Bander, Age when Banded, Sex, Region Where
Banded, Date Banded, Region of Recovery, Date Recovered, How Obtained,
and Present Condition. By
examining the "How Obtained" and "Present
Condition" codes we determined whether or not a recovery record
referred to a dead bird. We
selected for our analysis only the records of adult birds that had
died. We defined an
"adult" as any bird that had lived at least 90 days.
We calculated the age at which each bird died by subtracting
the date it was banded from the date of its recovery.
Since most bluebirds are banded as nestlings, the ages we
calculated are probably accurate to within 2-4 weeks unless a bird was
found long after its death. We
did not determine how many of the birds recovered in a region had also
been banded in that region.
Results
and Discussion
We
estimate that during 1920-1990 approximately 290,000 Eastern Bluebirds
were banded. We arrived
at this figure by combining the number of Eastern Bluebirds that
Klimkiewicz et a/. (1983) reported had been banded through 1981
(140,099) with the annual totals in the Bird Banding Laboratory files
for 1982-1990 (150,037).
The
Bird Banding Laboratory records indicate that 256,453 Eastern
Bluebirds were banded in 1955-1990, with the smallest number, 1,062,
being banded in 1960 and the largest number, 24,780, being banded in
1990 (Figure 1). A
tabulation of the number of Eastern Bluebirds banded each year was not
a major objective of this paper.
We feel, however, that information about the number of
bluebirds banded is important when attempting to understand variations
in the number of banded bluebirds that have been recovered.
The probability of a banded bluebird being found
dead or recaptured away from the area where it was banded is low.
Klimkiewicz et a/. (1983) noted that of the 140,099 Eastern
Bluebirds that were banded prior to 1982, 1,618 (1.15%) had been
recovered. The recovery
rate for many other species is even lower.
For example, the recovery rate for Blue-gray Gnatcatchers (Polioptila
caerulea) was 0.2%, and the recovery rate for Northern Parulas (Parula
americans) was 0.06% (Kllimkiewicz et a/. 1983).
The higher recovery rate for Eastern Bluebirds may be partially
due to the facts that (1) they do not migrate into Central America or
South America where the probability of someone finding and reporting
the band number of a dead bird is lower than in the United States, and
(2) they sometimes die in nest boxes or other sites where they are
likely to be found by humans.
Dates of
deaths
We
obtained from the Bird Banding Laboratory files a total of 1,931
records of recovered Eastern Bluebirds.
We determined that 725 of these records represented dead birds.
The monthly mortality totals are given in Figure 2, and the
annual mortality totals are presented in Figure 3. Fifty-nine percent
of the deaths occurred during the four months of March-June.
Many deaths also occurred in February.
Mortality is known to be high during this time of year (Pitts
1978; Sauer and Droege 1990), but this is also a time when more people
inspect their nest boxes and are, therefore, more likely to find
bluebirds that died while roosting or on an early nest.
We believe that much of the mortality in February, March, and
early April can be attributed to the stress of late winter and early
spring weather.
We
suspect that mortality is heavier in the late summer and early fall
months than indicated in Figure 2. Many young of-the-year are becoming
independent of their parents and, in some areas, are beginning their
first migration at this time. Even
though these young birds hatched earlier in the year, many of them are
older than 90 days and are treated as adults in this paper.
Much of the mortality at this time of year is not associated
with nest boxes, and therefore, is less likely to be detected and
reported.
The
number of banded bluebirds found dead each year has varied greatly.
The general trend, as shown in Figure 3, has been toward an
increase in the number of recoveries each year.
We think this trend is not due to an increased
rate of mortality. Instead, we think this trend is to be expected as
more bluebirds are now being banded (Figure 1) and more people are
inspecting roost sites and nest boxes, thus increasing the probability
of finding dead bluebirds. The number of recoveries per year has,
however, not increased uniformly. For example, in 1978 high mortality
was recorded, primarily as a result of the harsh winter weather. But
in 1979 and 1980, mortality appeared to be much lower than expected.
This is probably due to the fact that fewer bluebirds were alive
during these two years. Some states had record low populations during
these two years (Pitts 1981). The same sequence of higher than
expected mortality one year followed by lower than expected mortality
the next year has previously occurred (e.g., 1942 and 1958). We
believe that the alternating high and low annual mortality totals
shown in Figure 3 reflect fluctuations in bluebird populations.
Cause of death
Based on the "How Obtained" and
"Present Condition" codes in the band records, we were able
to determine the probable cause of death for 354 individuals. Eight
factors accounted for 293 (83%) of these deaths (Table 1).
These factors were (1) shooting, (2) died on nest, (3) caught by cat, (4) found dead inside a building, (5) hit by
automobile, (6) struck stationary object other than TV tower, (7) died
due to injury, and (8) weather. Small
numbers of bluebirds died from various other causes.
For example, 5 died
due to disease, 3 were captured by raptors, 3 were captured by dogs, 2
drowned, 2 were entangled in fishing gear, and 1 struck a TV tower.
We think the data shown in Table 1 do not
accurately reflect the manner of death for the majority of Eastern
Bluebirds. We believe
that some factors caused more mortality than shown while other factors
caused, proportionately, less mortality than indicated by Table 1. Our
thinking was influenced by the fact that a specific cause of death was
given for fewer than half of the dead bluebirds. Of the 725 banded
Eastern Bluebirds whose death was reported, the cause of death was
indicated for only 354 (49%). The
other 371 deaths (51%) were simply recorded as "found dead." We suspect
that predators, diseases, and parasites are responsible for many more
deaths than Table 1 indicates. Winter
weather was listed as the cause of death for only 14 (4%) of 354
birds. Severe weather in
late winter and early spring is known to decimate bluebird
populations, sometimes over large areas (Sauer and Droge 1990).
So why did the Bird Banding Laboratory have so few records of
winter-killed banded Eastern Bluebirds?
Many
of the birds whose cause of death was listed without explanation as
"found dead" possibly died due to winter weather. This speculation is supported by examination of the month of death (Figure 2). Large numbers of
bluebirds were reported as "found dead" in February and
March. Early spring
storms in April may also be responsible for many of the birds reported
as "found dead on nest." Pitts (1981) estimated that the
number of Eastern Bluebirds nesting in Tennessee declined by at least
70,000 individuals during the severe winters of 1977-1979. We
believe it would be unrealistic to assume that four times that number
died from shooting, which is what an examination of the data in Table
1 might suggest.
Another factor that has possibly biased the
results summarized in Table 1 involves the reporting policies of the
Bird Banding Laboratory. Each dead banded bird that is recovered is assigned a number
code indicating how the bird was obtained.
In many cases the selection of a proper code involves an
interpretation of the circumstances and a subjective decision as to
the cause of the bird's death. Consider
a banded bluebird that is found dead in a nest box in March.
Which of the following codes should be assigned?
Table 1. Cause of death for 354 banded adult Eastern
Bluebirds.
|
CAUSE OF
DEATH |
NUMBER REPORTED |
| Shot |
55 |
| Found dead in nest |
51 |
| Domestic cat |
51 |
| Found dead in building |
50 |
| Collision with automobile |
42 |
| Collision with stationary object |
16 |
| Weather (Oct. through Mar., only) |
14 |
| Injury |
14 |
Miscellaneous (no more than 5 deaths
due to any one cause) |
16 |
Code
00-found dead; Code 02-found dead due to starvation; Code 15-found
dead due to weather; Code 21-found dead in an enclosure; Code 50-found
dead: band with skeleton only; or, Code 97-method of recovery not
covered by other codes. Conceivably,
any one of these six codes could be applied to this recovery.
Since the cause of death is not obvious, most people would
probably select the most general category, Code 00-found dead. The fact that this code is the first code on the list might
cause it to be selected more frequently than some of the other codes,
particularly if doubt exists as to which is the most appropriate code.
Based on Table 1, the major cause of death
in adult Eastern Bluebirds is shooting.
Many of these records are from the years 1931-1940 (Figure 4)
when, even though the shooting of bluebirds may have been illegal, the
regulations were not rigidly enforced in many areas, and most shooters
were probably not reluctant to be truthful. (I am continually amazed
at the number of students in my introductory ornithology classes who
are not aware that the shooting of songbirds is illegal. Ignorance of the law in the 1930's is more understandable
than a similar type of ignorance in the 1990's.) The Bird Banding
Laboratory records show that the greatest number of Eastern Bluebirds
was shot in the 1960 Is. If
this decade is ignored, Figure 4 shows a steady decline in the number
shot each decade. However,
the decade of the 1960's cannot be ignored, so why the increase during
these years? We at first
suspected one person at one site might be responsible (e.g., a kid
with a new air rifle). This
theory was not supported by the data.
When we determined the location where each of these bluebirds
had been killed we found no concentration.
Bluebirds were reported shot from 10 different locations with
no more than 3 records from any one location.
At this point we can only guess as to why so many bluebirds
were shot in the 1960's. Perhaps
the number of bluebirds being shot reflects the political and social
unrest of that time. Possibly
the small sample size biased the results and the peak seen in
1961-1970 is an artifact.
Unknown
numbers of bluebirds are still shot, but the cause of death for those
reported is listed under some non-incriminating code, such as
"found dead," in order to avoid potential legal
consequences. In the past, the general public was less familiar with
bird bands and, out of curiosity, people frequently shot
banded Eastern Bluebirds. Unfortunately,
some bluebirds are still shot because they wear bands.
On the study area of TDP a new resident shot a color banded
bluebird in order to "...get a closer look at the weird color on
its legs." The bird was reported, thankfully, by the shooter to
the Bird Banding Laboratory; the cause of death was listed as
"found dead."
Location of
mortality
The
recovery locations for 725 dead
Eastern Bluebirds are summarized in Table 2. Tracing each of the
individual birds from its banding location to the site of its death
and analyzing the resulting patterns is
beyond the scope of this paper.
The numbers in Table 2 almost certainly do not correspond to
the size of the Eastern Bluebird populations in an area.
Neither do these data describe the actual amount of bluebird
mortality occurring in each area.
We believe, instead, that the results shown in Table 2
more likely reflect the numbers of banders and their activities.
For example, our home state, Tennessee, is the location of the
largest number of recovered dead bluebirds.
This is primarily due to the banders such as Mrs. Amelia Laskey
who over
a, 40 year period handled thousands of Eastern Bluebirds.
Table 2. Recovery locations of banded adult Eastern
Bluebirds found dead.
|
LOCATION |
NO. |
LOCATION |
NO. |
| Alberta |
1 |
Mississippi |
12 |
| Manitoba |
1 |
Connecticut |
13 |
| Dist. of Col. |
1 |
South Carolina |
14 |
| Maine |
1 |
Missouri |
16 |
| Vermont |
3 |
Georgia |
19 |
| Oklahoma |
3 |
Virginia |
19 |
| Quebec |
3 |
North Carolina |
26 |
| New Hampshire |
4 |
Alabama |
28 |
| Kentucky |
4 |
Illinois |
29 |
| Florida |
5 |
Massachusetts |
29 |
|
Indiana |
6 |
Arkansas |
30 |
| Ontario |
6 |
New York |
36 |
| Kansas |
7 |
Ohio |
45 |
| Bermuda |
7 |
Michigan |
46 |
| West Virginia |
7 |
Pennsylvania |
47 |
| Louisiana |
10 |
Maryland |
52 |
| Texas |
11 |
Wisconsin |
56 |
| Minnesota |
12 |
Tennessee |
92 |
| New Jersey |
12 |
........................ |
....... |
| Iowa |
12 |
Total |
725 |
Sex of
recovered Eastern Bluebirds
The
sex was recorded for 295 of the 725
recovered dead bluebirds; 162 were females and 133 were males. A chi-square test shows that these results are not
significantly different from a 1:1 ratio (X2 = 2.85, df = 1, P >
0.10). We cannot think of logical reasons why most causes of mortality
(Table 1) would influence one sex more than the other.
However, two mortality factors, "Found Dead in Nest"
and "Killed by Domestic Cat" might result in the death of a
higher percentage of females because females spend more time in the
nest cavities than do males.
Of
51 bluebirds found dead in
the nest, 8 were identified as females and 8 as males; the sex of the
others was not recorded. While
the sex ratio, of those birds whose sex was determined, is equal to
1:1, we would feel much more comfortable about this conclusion if a
larger sample size were available.
Considering
that each bird was handled at least twice, once when banded and once
when found dead, why was the sex not determined on more of these
birds? Most of the
bluebirds found dead in the nest had been banded as nestlings during a
previous nesting season. Since
most nestlings are banded before their plumage has developed
sufficiently to indicate sex, their banding records do not indicate
their sex. Likewise, if
an adult bird dies on the nest and then decays or is mutilated before
being found, sex determination can be difficult.
Of
51 Eastern Bluebirds killed
by domestic cats, 19 were females and 5 were males; the sex of the
other birds was not indicated. While
this sample size is not much larger than that of bluebirds found dead
in the nest (24 versus 16), the trend seen here shows a
significant difference in the number of males and females killed (X2
= 8.16, df = 1, P <0.005). Figure 5
shows that mortality due to domestic cats is highest during the
nesting season. Considering that the female bluebirds, who construct each
nest, lay eggs, and incubate the eggs, receive significant amounts of
help from the male only when caring for nestlings, we are not
surprised that more females are captured by domestic cats. (Domestic
cats kill an unknown, but probably large, number of nestling
bluebirds. However, since
this paper deals only with adult mortality we have not explored the
effects of cat predation on nestlings.)
Age
of recovered Eastern Bluebirds
We
were able to calculate an approximate age at the time of death for 516
Eastern Bluebirds. The youngest bird died at 91 days (please
recall that we limited our analysis to birds that were at least 90
days old); the oldest individual died at 2836 days (7.76 years), We
are excluding from consideration a record of an Eastern Bluebird that,
according to the Bird Banding Laboratory records, lived at least 7834
days (21.4 years). We suspect, but have not been able to
confirm, that this record contains an error. The mean age at
death was 499 days (1.37 years). As summarized in Figure 6,
slightly more than half of the bluebirds died before they were 1 year
old. Fewer than 20% were over 2 years old at the time of their
death. Although the age composition of local populations of
Eastern Bluebirds may vary, we believe the data in Figure 6
approximate the actual age composition of Eastern Bluebirds in North
America. We assume that mortality is proportional to the number
of individuals in each age class. The argument could be made
that the young birds have less experience than old birds and are more
likely to die. While this is probably true for young birds that
have recently fledged, we considered only individuals who were at
least 90 days old. These birds may not have as much experience
as their parents, but they have the advantages of youth (e.g., good
eyesight, rapid reflexes, and fewer parasites). The average
lifespan of 134 females was 449 days (1.23 years); the average
lifespan of 120 males was 474.75 days (1.30 years). A T-test
indicates that no significant difference exists between the age at
death for females versus males (df = 119, 133; P = 0.51).
Summary
Of
approximately 290,000 Eastern Bluebirds banded during 1920-1990, the
deaths of 725 adults (i.e., individuals at least 90 days old) are
documented by Bird Banding Laboratory records.
-
Approximately 60% of these deaths occurred
during the 4 months of March -June. We suggest that the
monthly mortality figures we calculated may be biased because the
deaths of many Eastern Bluebirds are associated with nest boxes
during March-June, a time when the nest boxes are frequently
inspected by people. Deaths of Eastern Bluebirds are
probably less likely to be associated with nest boxes at other
times of the year and, consequently, are less likely
to be detected by humans.
-
The number of banded Eastern Bluebirds
reported dead each year has varied greatly and has generally
reflected the severity of winter weather and the number of birds
that had been banded the previous year.
-
The most frequently reported causes of death
for Eastern Bluebirds are events (such as being shot, found dead
in a nest box, and capture by a domestic cat) that are more likely
to result in a detectable carcass than are deaths due to
parasites, diseases, and predators. We believe the Bird
Banding Laboratory data underestimate, by a large factor, deaths
due to many natural factors.
-
We suspect that the reported locations of
deaths more likely reflect the activities of banders rather than
actual numbers of Eastern Bluebirds that have died in an area.
-
Most causes of death in Eastern Bluebirds do
not affect one sex more frequently than the other sex.
Mortality due to domestic cats appears to be an exception to this
generality; significantly more females have been reported as taken
by cats.
-
The average age at death for adult Eastern
Bluebirds was 1.4 years with no significant difference between
females and males. Fewer than 20% of the banded adult
Eastern Bluebirds lived more than 2 years.
Acknowledgments
We thank the staff of the Bird Banding
Laboratory for their assistance during the visit of RWS to the
Lab. We especially thank Danny Bystrak for the retrieval of
banding and recovery data. We also wish to acknowledge the help
that T.D. Pitts, Jr. provided with computer file manipulations.
Literature
Cited
Canadian Wildlife Service and U.S. Fish and
Wildlife Service. 1991. North American Bird Banding. Vol.
1. Environment Canada, Can. Wildl. Serv., Ottawa, Canada
and U.S. Fish and Wildl. Serv., Washinc ,iton, D.C.
Klimkiewicz,
M, Kathleen, R,B, Clapp, and A-G, Futcher. 1983. Longevi@
records of North American birds: Remizidae through Parulinae. J. Field
Omithol. 54:287-294.
Pitts, T. David. 1978. Eastern Bluebird
mortality at winter roosts in Tennessee. Bird-Banding 49:77-78.
Pitts,
T. David. 1981. Eastern Bluebird population fluctuations in
Tennessee during 1970-1979. Migrant 52:29-37.
Pitts,
T. David.1994. Eastern Bluebird Banding:1900-1920. Sialia
16:123-126, 129.
Sauer,
John R., and Sam Droege. 1990. Recent population trends of the
Eastern Bluebird. Wilson Bull. 102:239-252.
Department of Biological Sciences
University of Tennessee at Martin
Martin, Tennessee 38238
Reprinted, with permission, from "Sialia/Bluebird" Journal
of the North American Bluebird Society. NABS is a membership
organization for persons interested in bluebirds and other North
American birds which use cavities for nesting. For membership
information, send a message to nabluebird@aol.com
or go to the NABS web site at http://www.nabluebirdsociety.org/
