Content uploaded by Gary Ritchison
Author content
All content in this area was uploaded by Gary Ritchison on Dec 24, 2013
Content may be subject to copyright.
The Hunting Behavior of Eastern Screech-owls (Otus asio)
Carlo M. Abbruzzese1 and Gary Ritchison2
Abstract.—We studied the nocturnal hunting behavior of eight radio-
tagged Eastern Screech-owls (Otus asio; five females and three males)
during the period from November 1994 through March 1995.
Screech-owls selected low perches when hunting (x = 1.66 m),
presumably to obtain a clear view of the ground and an unobstructed
flight path to prey. Low perches may also improve the ability of
screech-owls to hear and locate prey. Screech-owls used perches at
different heights when hunting different types of prey and also
tended to perch higher when moonlight was available, perhaps
because increased light levels permit owls to rely more on vision.
Only 8 of 35 attacks were successful, and this low success rate
suggests that owls were more often attempting to capture small
mammals rather than invertebrates. Male and female screech-owls
exhibited similar hunting behavior, with no differences observed in
the types of prey hunted or in giving up times. Weather conditions
and season (early winter vs. late winter) had little effect on the
hunting behavior of screech-owls.
While the hunting behavior of diurnal preda-
tors has been studied by several investigators
(e.g., Fitzpatrick 1981, Greig-Smith 1983, Rice
1983, Sonerud 1992), few studies of the
hunting behavior of nocturnal predators have
been conducted (Bye et al. 1992). The hunting
behavior of owls, in particular, is little known,
with most information anecdotal or speculative
(Bent 1938, Voous 1989). Consequently, little
is known about how factors such as weather,
moon phase (i.e., light levels), snow cover, sex,
and temporal distribution of prey might influ-
ence the hunting habits of owls.
Eastern Screech-owls (Otus asio) are small,
nocturnal predators found throughout eastern
North America (Johnsgard 1988). Researchers
have examined several aspects of their behavior
and ecology, including food habits (Craighead
and Craighead 1956, Ritchison and Cavanagh
1992), home range sizes and habitat use
(Belthoff et al. 1993, Sparks et al. 1994), nest
site selection (Belthoff and Ritchison 1990),
and the postfledging behavior of adults and
young (Belthoff 1987). Very little is known,
however, about their hunting behavior.
__________________________________
Gehlbach (1994) gathered information concern-
ing the hunting behavior of Eastern Screech-
owls in central Texas, however, observations
were made only in suburban yards and in the
immediate vicinity of nests. Thus, little is
known about how screech-owls hunt in more
natural habitats and at locations some distance
from nest sites.
Eastern Screech-owls apparently hunt in a sit-
and-wait fashion, using short flights to capture
prey (Marshall 1967, Gehlbach 1994). It has
also been suggested that Eastern Screech-owls
may rely primarily on sight when searching for
prey because they have symmetrical ear
openings (Marshall 1967, Norberg 1987). Bye
et al. (1992) suggested that Boreal Owls
(Aegolius funereus) used low perches while
hunting because they rely primarily on their
sense of hearing. Thus, at least in Boreal
Owls, sensory capabilities may play an impor-
tant role in determining how prey are located
and captured. This may also be the case for
Eastern Screech-owls.
The objective of this study was to describe the
hunting behavior of Eastern Screech-owls
during the non-breeding period. We specifically
sought to examine: (1) possible differences
between the hunting behavior of males and
females, (2) seasonal and temporal variation in
1 Biologist with the City of Austin, TX.
2 Professor of Biology, Eastern Kentucky
University, Richmond, KY.
21
2nd Owl Symposium
screech-owl hunting behavior, and (3) the
possible effects of weather and moon phase on
hunting behavior.
STUDY AREA
The hunting behavior of Screech-owls was
studied from November 1994 through March
1995 at the Central Kentucky Wildlife Manage-
ment Area, located 17 km southeast of
Richmond, Madison County, Kentucky. The
study area was composed of a patchwork of
small deciduous woodlots, old fields, agricul-
tural fields, and mowed fields connected by
woodrows. Dominant woodland canopy tree
species included shagbark hickory (Carya
ovata), bitternut hickory (C. cordiformis), post
oak (Quercus stellata), chinquapin oak (Q.
prinoides), boxelder (Acer negundo), and red
oak (Q. borealis). The mid-story layer was
composed primarily of red maple (Acer rubra),
flowering dogwood (Cornus florida), pawpaw
(Asiminia triloba), spice bush (Lindera benzoin),
silky dogwood (C. amomum), and hackberry
(Celtis occidentalis). Common edge and old
field tree species included black locust (Robinia
pseudo-acacia), white ash (Fraxinus americana),
smooth sumac (Rhus glabra), and sweetgum
(Liquidambar styraciflua); while American
sycamore (Plantanus occidentalis) and black
willow (Salix nigra) were frequently found in wet
and riparian areas. The eastern red cedar
(Juniperous virginiana) was widely distributed
throughout the study area. Two vines, summer
grape (Vitus aestivalis) and heart-leaf
ampelopsis (Ampelopsis cordata), were common
in woodlots.
Terrestrial crayfish (Cambarus spp.) are
common throughout much of the study area,
particularly in low, poorly drained areas and
along streams (Ritchison and Cavanagh 1992,
pers. observ.). These crayfish typically produce
small mounds of soil around the entrances to
their burrows, and these mounds clearly
indicate the presence of active crayfish. Other
prey species that are potentially available to
screech-owls in our study area include various
small mammals, birds, and invertebrates
(Ritchison and Cavanagh 1992).
METHODS AND MATERIALS
Beginning in November 1994, Eastern Screech-
owls were captured by checking nest boxes
distributed throughout the study area. Nest
boxes were checked during daylight hours
when screech-owls are typically docile and
more easily handled. After capture, radio-
transmitters (Wildlife Materials, Inc.,
Carbondale, IL) were attached backpack style
(Smith and Gilbert 1981). Transmitters
weighed approximately 6 g (3 to 4 percent of
total body mass). The owls appeared to adjust
quickly to the transmitters, and did not appear
to behave abnormally (pers. observ.). Owls
were allowed at least 1 week to become
accustomed to the transmitter before
observations began. To facilitate observations,
a 5 to 6 cm piece of reflective tape was attached
to the transmitter antennas.
Observations typically began shortly after
sunset and continued for 3 to 5 hours. During
owl observations, the general location of a
radio-tagged owl was determined using a
receiver (TR-2; Telonics, Inc., Mesa, AZ) and a
hand-held, two-element antenna. Then, a
Petzel headlamp or portable wheat lamp was
used to scan the area from which the signal
originated. Normally the light would strike the
reflective tape on the transmitter’s antenna and
permit us to precisely locate the owl. While
observing an owl, a light with a red cellophane
filter was used to minimize disturbance as owls
are known to have limited vision at the red end
of the light spectrum (greater than 600 nm;
Martin et al. 1975).
Eastern Screech-owls are relatively tame and
typically do not fly until an observer comes
within 3 or 4 m (pers. observ.). Our
observations of screech-owls were usually
made at a distance of 10 to 12 m using 10 x 25
binoculars. At this distance, our presence
appeared to have little effect on the owls. The
owls sometimes allowed us to watch them for
more than an hour before moving, and owls
often flew in our direction or directly over us
when changing perches.
To examine possible differences in hunting
behavior over time, November and December
were categorized as early winter, and January,
February, and March as late winter. For each
hunting observation, we recorded the time the
owl landed on a perch and the time until the
owl either initiated an attack or left for another
perch (i.e., giving up time). If an owl was not
observed landing on a perch, we sometimes
estimated their time of arrival at the perch by
monitoring the transmitter’s signal. Trans-
mitters had activity switches so pulse rates
typically changed when an owl stopped moving.
22
After the owl moved to another location,
perches and attack sites (the point on the
ground where the owl hit or attempted to hit
the prey) were marked with flagging and plotted
on an aerial photograph of the study area. We
returned to these marked perch sites during
daylight hours to measure perch heights,
identify and measure the tree, shrub, or vine in
which the perch was located, measure attack
distances (the distance from the perch to the
prey item attacked), and measure distances
between consecutive perches.
We also noted whether or not attacks were
successful and, if so, attempted to identify the
prey species. We noted and flagged the search
area (where the focal owl appeared to be
looking for prey). For perches where no attack
was made or where the attack was unsuccess-
ful, we noted the type of prey (e.g., crayfish,
small mammal, or bird) that the owl was
probably hunting. We made this judgement
based either on where the owl appeared to be
searching or, less frequently, on prey species
that we actually observed the owl to be
watching. For example, an owl searching the
ground in a low, poorly drained area with
numerous crayfish burrows was assumed to be
hunting crayfish, while an owl searching the
ground in a drier area with no crayfish burrows
was assumed to be hunting small mammals.
Owls peering intently into eastern red cedars or
dense shrubs were thought to be hunting
birds.
During each observation period, we recorded
the air temperature and, for subsequent
analysis, categorized temperature as either
above freezing or below freezing. We also noted
the presence or absence of precipitation and,
on that basis, categorized conditions as:
snowing, raining, or no precipitation. We also
noted the presence or absence of snow cover
and whether or not the moon was visible.
Analysis
Because the number of observations on each
owl varied, and to avoid bias from pooling such
data (Leger and Didrichsons 1994), overall
mean values were calculated using the means
for each bird. We used different subsets of the
data for other analyses. Owls were included in
a particular analysis only if represented in all
categories. For example, an owl would be
included in an analysis of possible differences
in perch height with season only if observed
hunting during both seasons (early winter and
late winter).
Multiple comparisons were made using non-
parametric analysis of variance (analysis of
variance on ranked data which is equivalent to
the Kruskal-Wallis test; SAS Institute 1989).
Paired comparisons (e.g., males versus females)
were made using Wilcoxon rank sum tests
(which are equivalent to Mann-Whitney U tests;
SAS Institute 1989). All analyses were per-
formed using the Statistical Analysis System
(SAS Institute 1989). All values are presented
as mean + standard deviation.
RESULTS
Capture and Observation of Owls
Eight adult Eastern Screech-owls (five females
and three males) were tracked during the
period between November 10, 1994 and March
11, 1995. These owls were observed for a total
of 168.5 hours during 91 evenings. Hereafter,
these owls will be referred to by the last three
digits of their U.S. Fish and Wildlife Service
aluminum bands (table 1). Based on obser-
vations during previous breeding seasons and
during the breeding season that followed our
study, it was determined that two of these owls
(male 099 and female 285) represented a mated
pair. The mated status of the other owls was
unknown. One owl (female 285) died during
the study, and the cause of death could not be
determined.
Table 1.—Tracking period and number of nights
tracked for each Eastern Screech-owl.
Owl Sex Tracking Number of nights
period tracked
931 Female 11/10/94 - 2/24/95 19
016 Female 1/4/95 - 3/5/95 8
285 Female 11/14/94 - 11/24/94 4
959 Female 1/8/95 - 2/22/95 9
041 Female 11/18/94 - 2/25/95 13
099 Male 11/10/94 - 3/11/95 11
307 Male 11/11/94 - 3/8/95 14
215 Male 11/12/94 - 3/2/95 13
Hunting Behavior - Overall
Eastern Screech-owls (N = 8) used a total of
338 hunting perches, and these perches were
an average of 1.66 + 0.22 m high. The mean
23
2nd Owl Symposium
height of the vegetation (e.g., tree, shrub, or
vine) in which these perches were located was
5.04 + 1.81 m, and the mean d.b.h. was 11.07
+ 6.13 cm. Most hunting perches (63.2
percent) were on an open branch (25 cm or
more away from the trunk), rather than near
the trunk (less than 25 cm from the trunk)
(23.0 percent) or on the stub or top of a plant
(13.8 percent).
Owls (N = 6) initiated 35 attacks from hunting
perches, with 27 being unsuccessful (no prey
captured) and eight successful. The successful
attacks resulted in the capture of one bird, one
crayfish, one small mammal, and five moths.
The mean attack distance was 3.41 + 0.79 m.
Owls (N = 8) remained on perches for an
average of 393.7 + 204.2 sec before either
attacking or giving up. Screech-owls (N = 8)
leaving perches without making an attack (i.e.,
giving up time) remained on perches for an
average of 278.9 + 99.3 sec, while owls (N = 5)
that attacked prey were on perches for an
average of 226.1 + 108.9 sec prior to initiating
the attack. Mean giving up time for these latter
five owls was 323.9 + 90.4 sec. When owls (N =
6) changed perches (either after an attack or
after giving up), the mean distance between
consecutively used perches was 9.28 + 4.12 m.
Screech-owls appeared to hunt six different
types of prey: birds, insects, crayfish, small
mammals, leeches, and fish. Owls appeared to
hunt primarily crayfish (49.7 percent of all
observations), small mammals (29.7 percent),
and birds (5.4 percent) (fig. 1). Four owls (041,
215, 931, and 959) were observed hunting all
three of these primary prey items, and perch
height varied significantly with type of prey
being hunted (F2,9 = 17.69, P = 0.0008). Mean
perch height when hunting birds was 2.86 +
1.11 m. By contrast, mean perch height was
1.90 + 0.45 m when owls were hunting small
mammals and 1.27 + 0.10 m when hunting
crayfish. We found no significant differences in
the perch position (open branch, near trunk, or
at the top of the plant) used by owls hunting
the three primary prey items (χ2 = 6.7, df = 4, P
= 0.155). Giving up times for these four owls
varied significantly with type of prey hunted
(F2,6 = 6.35, P = 0.033): a mean of 308.7 +
539.7 sec for crayfish, 486.8 + 522.7 sec for
small mammals, and 721.9 + 1175.9 sec for
birds.
Hunting Behavior - Males versus Females
Hunting perches used by male and female
screech-owls did not differ significantly in
height (z = 0, P = 0.99), with a mean perch
height of 1.65 + 0.18 m for females (N = 5) and
1.66 + 0.35 m for males (N = 3). Similarly,
there were no differences in either the mean
height (z = 0.89, P = 0.37; x = 5.69 + 1.98 m
for females and 3.97 + 0.93 m for males) or
mean d.b.h. (z = 1.49, P = 0.14; x = 13.56 +
6.60 cm for females and 6.92 + 1.70 cm for
males) of the vegetation in which these perches
were located. Males and females did, however,
exhibit a significant difference in choice of
perch positions (χ2 = 7.6, df = 2, P = 0.023),
with females more likely to perch near the
trunk and males more likely to perch on the
stub or top of a plant (fig. 2).
We found no difference between male and
female screech-owls (z = 0.6, P = 0.55) in giving
up time: a mean of 238.8 + 122.7 sec for males
(N = 3) and 303.0 + 88.4 sec for females (N = 5).
Figure 1.—Apparent prey items hunted by East-
ern Screech-owls.
0
10
20
30
40
50
60
70
Near Trunk Open Branch Stub
Percent Use
Females
Males
Figure 2.—Choice of perch positions by male
and female Eastern Screech-owls.
24
We also found no difference (z = 0.29, P = 0.77)
between males and females in mean attack
time, with a mean of 273.3 + 122.4 sec for
females (N = 3) and 155.4 + 28.4 sec for males
(N = 2).
Male and female screech-owls also did not
differ (z = 0, P = 0.99) in mean attack distance.
The mean attack distance for females (N = 4)
was 3.64 + 0.91 m, while for males (N = 2) the
mean distance was 2.96 + 0.01 m. Similarly,
the mean distance between consecutive
perches did not differ (z = 0.23, P = 0.82)
between the sexes, with females (N = 4) moving
a mean distance of 8.3 + 4.8 m and males (N =
2) a mean distance of 11.3 + 2.0 m.
Both males and females appeared to hunt
primarily for crayfish, followed by small
mammals and birds. There was no difference
(χ2 = 1.8, df = 2, P = 0.4) between the sexes in
the frequency with which they appeared to be
hunting for the three primary prey items.
Hunting Behavior - Effect of Outcome
The height of hunting perches from which
attacks were initiated (x = 1.54 + 0.36 m; N = 6
owls) did not differ significantly (z = 0.8, P =
0.42) from that of perches from which owls did
not initiate attacks (x = 1.59 + 0.16 m; N = 6
owls). The mean height of perches from which
successful attacks were initiated was 1.58 +
0.59 m (N = 4 owls), while that for perches from
which unsuccessful attacks were initiated was
1.40 + 0.71 m (N = 4 owls). This difference was
not significant (z = 0.43, P = 0.67). The mean
attack distance was 3.10 + 1.34 m (N = 4 owls)
for successful attacks and 4.03 + 2.31 m (N = 4
owls) for unsuccessful attacks, and this
difference was not significant (z = 0.14, P =
0.89).
Screech-owls (N = 6) that eventually initiated
an attack remained on perches for an average
of 614.6 + 956.5 sec, while those that did not
initiate an attack remained on perches an
average of 299.9 + 100.0 sec. This difference in
perch time was not significant (z = 0.56, P =
0.58). One owl (Female 285) had a longer
perch time prior to attack than the other owls
because she once spent 84 minutes on a perch
before making an attack. It is likely that this
female, while roosting, happened to locate and
then attack a prey item. If female 285 is
removed from the analysis, the mean time until
attack decreases to 226.1 + 108.9 sec (N = 5
owls) and the mean time until giving up
increases slightly to 323.9 + 90.4 s (N = 5
owls). Nonetheless, this difference in perch
time is still not significant (z = 1.2, P = 0.21).
Hunting Behavior - Probability of Prey
Detection and Pattern of Giving Up
For a screech-owl on a perch searching for a
prey item, the probability of detecting prey may
increase, decrease, or remain the same as time
passes. As described by Bye et al. (1992:270):
“The cumulative distribution of detection times
may be used to determine which of these pos-
sibilities is the case. This distribution decays
exponentially if the probability of detecting prey
remains constant. If the probability of prey
detection increases or decreases, this will
appear as a concave or convex deviation,
respectively, from the exponential model.” For
all screech-owls that initiated attacks, we
plotted the proportion of owls still remaining on
their perches versus perch time (total time
spent on the perch). These proportions were
log-transformed to facilitate the assessment of
fit to an exponential model (Bye et al. 1992).
The decay in the distribution of attack times
(detection times) was, in fact, close to expon-
ential (fig. 3). The straight line representing the
0.01
0.1
1
012345678910
Time on Perch (min)
Proportion remaining on perch
Figure 3.—Proportion of Eastern Screech-owls
remaining on perch in relation to time on
perch for perches from which an attack was
made (N = 35). The straight line pattern
indicates that how soon an owl is likely to
attack is not predicted by perch time.
25
2nd Owl Symposium
best-fitting exponential model for attack times
up to 10 min was log y = -0.09x - 0.056. In
other words, prey were attacked by screech-
owls at a constant rate.
The decay in the distribution of giving-up times
was also close to exponential (fig. 4). The
straight line representing the best-fitting
exponential model for giving-up times up to 10
min was log y = -0.09x - 0.036. Thus, screech-
owls also gave-up at a constant rate.
Hunting Behavior - Effect of Season
Examination of the hunting behavior of owls (N
= 5) observed during both early and late winter
revealed no significant seasonal variation in
perch height, height or d.b.h. of the vegetation
in which perches were located, perch time
(before either giving up or attacking), attack
distance, or distance between consecutive
perches (table 2).
Table 2.—Seasonal variation in the hunting
behavior of Eastern Screech-owls. (Numbers
represent mean + standard deviation.)
Early winter Late winter
(Nov.-Dec.) (Jan.-March)
Perch height (m) 1.65 + 0.16 2.14 + 0.90
Perch tree/shrub
height (m) 3.93 + 1.21 5.88 + 3.62
Perch d.b.h. (cm) 7.84 + 2.68 11.16 + 10.34
Perch time (sec) 319.3 + 155.0 368.8 + 305.5
Attack dist. (m) 3.42 + 0.72 3.47 + 2.07
Interperch
distance (m) 8.98 + 4.73 9.95 + 2.08
We also examined possible seasonal variation
in the types of prey that the four most fre-
quently observed owls appeared to be hunting.
Although three of these owls exhibited signifi-
cant seasonal variation (χ2 tests, P < 0.035),
and variation for the fourth owl approached
significance (χ2 = 5.45, df = 2, P = 0.065), no
clear trends were apparent. For example, two
owls (307 and 041) appeared to be hunting for
small mammals more than expected during
early winter, while, in contrast, the other two
owls (931 and 215) hunted for small mammals
more than expected during late winter. Also
illustrating the absence of any trends, one owl
(931) hunted for crayfish more than expected in
early winter, while two owls (307 and 041) did
the same in late winter.
0.01
0.1
1
012345678910
Time on Perch (min)
Proportion remaining on perch
Figure 4.—Proportion of Eastern Screech-owls
remaining on perch in relation to time on
perch for perches that were given up (N =
303). The straight line pattern indicates that
how soon an owl is likely to give up is not
predicted by perch time.
Hunting Behavior - Effect of Moonlight and
Weather
Screech-owls chose significantly higher perches
when moonlight was available (i.e., 1/4 moon,
1/2 moon, 3/4 moon, or full moon) than when
moonlight was not available (z = 2.09, P =
0.036), with a mean perch height of 1.88 + 0.35
m (N = 5 owls) when moonlight was available
and 1.46 + 0.08 m (N = 5 owls) when moonlight
was not available.
Three owls (016, 215, and 931) were observed
hunting during all weather conditions
(snowing, raining, and no precipitation) and
these varying conditions had no apparent effect
on the type of prey that owls hunted (χ2 = 1.68,
df = 4, P = 0.79). Similarly, temperature (above
versus below freezing) had no effect on the type
of prey being hunted by Screech-owls (χ2 =
4.57, df = 2, P = 0.102). Screech-owls
appeared to hunt for crayfish at similar rates
both when air temperatures were above and
below freezing. Temperature also had no
significant effect on giving up times (z = 1.46, P
=0.14), with owls giving up after 366.2 + 40.3
sec (N = 5 owls) when temperatures were above
freezing and after 264.2 + 148.4 sec (N = 5
owls) when below freezing.
Hunting Behavior - Effect of Time of Night
Overall, screech-owls exhibited significant
temporal variation in type of prey hunted (χ2 =
38.1, df = 6, P < 0.0001), with owls more likely
to hunt for crayfish early in the evening (1800 -
2300 h) and small mammals later in the
26
evening (2300 - 0300 h). We had sufficient
numbers of observations on six owls (016, 041,
215, 307, 931, and 959) to permit examination
of possible temporal variation in type of prey
hunted, and three of these owls (041, 215, and
959) exhibited the same significant (χ2 tests, P
< 0.035) tendency to hunt for crayfish earlier
and small mammals later. In addition, this
tendency approached significance (χ2 = 9.6, df
= 6, P = 0.14) for another owl (931). The
remaining two owls (016 and 307) appeared to
hunt for crayfish and small mammals at
similar levels throughout the night.
We found little evidence that giving up times
varied with time of night. For five owls with
sufficient number of observations for analysis
(041, 215, 307, 931, and 959), only one
exhibited significant temporal variation in
giving up time. This female (041) had
significantly shorter giving up times early in the
evening (x = 172.1 + 144.8 sec; N = 26) than
later in the evening (x = 595.5 + 666.9 sec; N =
13) (z = 2.55, P = 0.011). Three other owls
(307, 931, and 959) also had shorter giving up
times early in the evening, but differences in
giving up times between early and late evening
were not significant (Wilcoxon tests, P > 0.09).
DISCUSSION
Hunting Perches
The mean perch height for hunting Eastern
Screech-owls in this study was 1.66 m. By
contrast, screech-owls roosting on open limb
perches in the same study area were found at a
mean height of 10.2 m (Belthoff and Ritchison
1990). Gehlbach (1995) also reported that
screech-owls used higher perches when
roosting (x = 4 m) than when hunting (x = 2.6
m). Boreal Owls also use higher perches for
roosting than for hunting (Bye et al. 1992).
Owls may select higher perches for roosting
because such sites may provide more cover
than lower sites (Bye et al. 1992). In fact,
Belthoff and Ritchison (1990) noted that
screech-owls typically selected roost sites that
provided concealment. The lower perch sites
selected by hunting owls probably provide an
unobstructed view of, and unobstructed access
to, the ground (Bye et al. 1992), an important
consideration for predators that largely depend
on ground-based prey.
Screech-owls in this study used lower hunting
perches than reported for other species of owls.
For example, the mean height of hunting
perches was found to be 3.3 m for Boreal Owls
(Bye et al. 1992), 5.5 m for Great Gray Owls
(Strix nebulosa; Bull and Henjum 1990), and
8.5 m for Northern Hawk Owls (Surnia ulula;
Sonerud 1992). Norberg (1970), however,
reported that the mean height of hunting
perches used by Boreal Owls was 1.7 m,
similar to that for screech-owls in this study.
Several factors may influence the height of
perches selected by hunting owls. An owl’s
sensory capabilities may be one such factor.
For example, owls that depend heavily on
acoustic cues to localize ground-dwelling prey
may need to be a short distance from potential
prey (Andersson 1981, Rice 1982, 1983).
Thus, the use of low hunting perches by
screech-owls in this study suggests that they
may depend on hearing to locate prey.
Anatomical evidence, however, suggests that
screech-owls should be more dependent on
vision. That is, screech-owls have symmetrical
ears and Norberg (1987) suggested that owls
with such ears may rely more on vision while
those with asymmetrical ears may rely more on
hearing. This may be true because asym-
metrical ear openings would permit an owl to
simultaneously locate prey on both the hori-
zontal and vertical planes, but symmetrical ear
openings do not preclude the use of hearing to
locate prey. Owls with symmetrical ears could
simply determine horizontal and vertical
directions one after the other with an inter-
vening tilting of the head (Norberg 1987). This
technique would require more time but might
still be effective when hunting relatively slow
moving prey like terrestrial crayfish. It is
likely, therefore, that screech-owls hunting
from low perches may, in part, be seeking
auditory cues concerning the location of prey.
Other investigators have also reported observ-
ations suggesting that Eastern Screech-owls
hear well and may, at times, depend on hearing
to locate prey. For example, Bent (1938) and
Gehlbach (1994) reported that screech-owls
were able to locate prey hidden in leaf litter.
The type of prey being hunted may also
influence the height of hunting perches. We
found that screech-owls that appeared to be
hunting crayfish perched lower than when
apparently hunting small mammals or birds.
Screech-owls that appeared to be hunting birds
used relatively high perches, perhaps because
avian prey were located in dense vegetation
(e.g., cedars) rather than on the ground.
Although both crayfish and small mammals are
27
2nd Owl Symposium
found on the ground, screech-owls perched
significantly lower when apparently hunting
crayfish. One reason for such behavior may be
that slower moving crayfish are more difficult
to detect. Similarly, Pinkowski (1977) found
that Eastern Bluebirds (Sialia sialis) perched
lower in the spring (March 15 - April 15),
because “smaller, fewer, or less active” insects
were more difficult to detect at greater heights.
Also, crayfish and, perhaps, invertebrates (or
ectotherms) in general may be less likely to
detect a nearby predator (particularly during
cooler weather) than would more active (and
alert) small mammals.
Light levels may also influence the height of
hunting perches. As already noted, Great Gray
Owls and Northern Hawk Owls use relatively
high hunting perches compared to screech-
owls. These two species both hunt during the
day when visibility is relatively high. In
contrast, screech-owls are primarily nocturnal
(and all our observations were made after
sunset). Although screech-owls and other
nocturnal owls do have eyes well-adapted for
seeing under low-light conditions (Norberg
1987), they may have to perch lower to detect
and accurately locate prey. Supporting this
view, the hunting perches of screech-owls in
this study were significantly higher when
moonlight was available.
The height of hunting perches may also be
influenced by perch availability. For example,
screech-owls in suburban Waco, Texas, used
higher hunting perches (x = 2.6 m; Gehlbach
1994) than did screech-owls in this study. One
possible factor for such differences may be that
the vegetation in suburban areas has been
substantially altered and typically has little
understory (Beissinger and Osborne 1982,
Gehlbach 1994). Thus, fewer low branches
may be available as perch sites for hunting
screech-owls.
The hunting perches of screech-owls were
primarily on open branches of small trees or
shrubs and at least 25 cm from the main
trunk. Such sites, in contrast to typical roost
sites (Belthoff and Ritchison 1990), provided
little concealment and, as a result, may
increase an owl’s vulnerability to predation
(e.g., by Great Horned Owls, Bubo virginianus).
As already noted, however, hunting owls must
have an unobstructed view that may not be
available on perches located closer to, or
against, the main trunk. Perhaps in an
attempt to reduce the chances of being spotted
by a larger predator (as well as to reduce the
chances of being spotted by potential prey),
screech-owls typically remain motionless (and
silent) when on hunting perches (pers. observ.).
Attack Distance and Success
The mean attack distance for screech-owls in
this study was 3.41 m. Similarly, mean attack
distances for Boreal Owls were reported to be
4.4 m (Norberg 1970) and 5.6 m (Bye et al.
1992). Bye et al. (1992) suggested that such
relatively short attack distances indicate a
restricted search area and, further, also
indicate that prey are being located using
acoustic cues.
Only 8 of 35 attacks (22.8 percent) by screech-
owls were successful. Similarly, Ural Owls
(Strix uralensis) hunting rodents were
successful 27 percent of the time (Nishimura
and Abe 1988) and Great Gray Owls also
hunting rodents were successful 33 percent of
the time (Bull and Henjum 1990). Eastern
Screech-owls in Texas successfully captured 56
percent of vertebrate prey attacked and 83
percent of invertebrate prey attacked (Gehlbach
1994). Such results indicate that attack
success rates vary with prey type, with
invertebrate prey more likely to be successfully
captured than vertebrate prey. The limited
success of screech-owls in this study may
indicate that they were attacking primarily
small mammals. Whereas success rates may
be higher for invertebrate prey, screech-owls
may hunt small mammals because smaller
invertebrates may represent less energy and, at
times (e.g., during the winter months), may not
be available.
Search Time
For the five screech-owls we observed both
when giving up and when attacking, the mean
giving up time was 323.9 sec while the mean
time until attack (or detection time; Carlson
1985) was 226.1 sec. Similarly, giving up times
were longer than detection times for
insectivorous, or primarily insectivorous,
Spotted Flycatchers (Muscicapa striata; Davies
1977), Eastern Bluebirds (Pinkowski 1977),
and American Kestrels (Falco sparverius;
Rudolph 1982). In contrast, detection times
were found to be longer in duration than giving
up times for two species of owls that prey
primarily on small mammals, Boreal Owls (Bye
28
et al. 1992) and Northern Hawk Owls (Sonerud
1989). Several investigators have noted that
small mammals are more difficult for predators
to catch than are insects (e.g., Sonerud 1980
cited in Bye et al. 1992, Temeles 1985). As a
result, Sonerud (1989) suggested that an owl
must wait longer before initiating an attack. If
so, observed detection times would appear
longer than actual detection times because
owls are waiting for an undetermined period of
time before launching an attack on already
detected prey. Thus, one possible explanation
for the short detection times (relative to giving
up times) observed in this study is that
screech-owls were hunting primarily insects
and other invertebrates (i.e., crayfish).
Screech-owls hunting primarily small
mammals might have longer mean detection
times. Unfortunately, we were unable to
identify a sufficient number of prey to deter-
mine if attack times (detection times) for
screech-owls varied with type of prey.
A predator attempting to optimize its hunting
effort might be expected to abandon a perch as
the probability of detecting prey begins to
decline (Fitzpatrick 1981). However, screech-
owls in this study gave up and attacked prey at
a constant rate, and similar results have been
reported for Boreal Owls (Bye et al. 1992).
Fitzpatrick (1981) explained such behavior by
suggesting that birds can assess the complexity
of the search area around each perch indepen-
dently after landing and estimate how long it
will take to search it thoroughly. If a prey item
appears during this time, the bird attacks and,
if not, the bird gives up. If an owl’s home range
includes a random selection of perches with
respect to the quality of search areas, the perch
survivorship curve will decline exponentially
(Bye et al. 1992).
Males versus Females
We found few differences in the hunting
behavior of male and female screech-owls.
Males and females did differ significantly in
perch location, with males more likely to perch
on the very top of plants and females more
likely to perch near the trunk. This difference
may be due, at least in part, to differences in
body size. Female screech-owls typically weigh
more than males (Gehlbach 1994, Henny and
VanCamp 1979, pers. observ.) and, as a result,
may have to perch on the slightly larger and
stronger branches located closer to the trunk of
small trees and shrubs.
The similar hunting techniques of male and
female screech-owls in our study may be due to
similarities in the types of prey being hunted.
We found that males and females appeared to
hunt primarily for crayfish and small mam-
mals. Similarly, Hofstetter (1995) reported that
male and female screech-owls on the same
study area captured the same types of prey
during the breeding season.
Bye et al. (1992) found that female Boreal Owls
had longer giving up times than males, and
suggested that larger females should wait
longer than smaller males because the cost of
flight is greater for larger females. We found no
difference between male and female screech-
owls in giving up time. This apparent
difference in the behavior of Boreal Owls and
Eastern Screech-owls may be due to differ -
ences in the degree of sexual dimorphism.
Female Boreal Owls are 4 percent larger than
males by wing length and 64 percent larger by
body mass (Korpimaki 1986, Lundberg 1986).
In contrast, female screech-owls are typically
only 16-17 percent larger in body mass than
males (Henny and VanCamp 1979, Gehlbach
1994). As with screech-owls in this study, Bye
et al. (1992) found that male and female Boreal
Owls did not differ significantly in the mean
height of hunting perches.
Moon Phase, Weather, and Season
Screech-owls in this study perched significantly
higher when moonlight was available, perhaps
because more light may permit hunting owls to
see greater distances and, therefore, perch
higher. In addition, however, more light may
permit screech-owls to rely more on vision than
on hearing. Because visual hunters need not
be as close to prey as acoustic hunters,
screech-owls relying on vision to locate prey
would be able to perch higher.
Screech-owls in this study exhibited no
changes in hunting behavior with season or
with changing weather conditions. Our study
was conducted over a 4-month period, and
conditions during that limited time may not
have varied sufficiently to influence prey
availability or hunting behavior. Studies
conducted over longer periods have revealed
that the food habits of Eastern Screech-owls do
vary seasonally, with more invertebrates taken
during the breeding period (March-August)
than during the non-breeding period
(September-February; Ritchison and Cavanagh
29
2nd Owl Symposium
1992). Such changes in prey use would, as
observed in this study, cause corresponding
changes in hunting behavior.
An important factor in the seasonal variation in
prey used by screech-owls is that invertebrates
are less likely to be available during colder
weather (Ritchison and Cavanagh 1992).
Screech-owls in our study hunted primarily for
crayfish, an invertebrate whose activity (and
therefore availability to screech-owls) may be
influenced by temperature. If so, fewer crayfish
should have been available at lower temper-
atures and, as a result, screech-owls should
have altered their hunting behavior (perhaps,
for example, by hunting for endotherms like
small mammals and birds). No such change in
behavior was observed. A likely explanation for
this is that the crayfish hunted by screech-owls
in our study apparently remained active and
available as prey even when temperatures were
below freezing.
Time of Night
Four of six screech-owls observed hunting both
early and late in the evening exhibited a
tendency to hunt for crayfish early in the
evening (1800 - 2300 h) and small mammals
later in the evening (2300 - 0300 h). Because
crayfish are ectotherms, declining tempera-
tures as the evening progresses might reduce
crayfish activity levels and, therefore, availabil-
ity. As noted previously, the crayfish on our
study area apparently remain active even when
temperatures are below freezing. Another pos-
sible explanation is that the availability of cray-
fish and small mammals varies with nightly
variation in their normal activity patterns, with
crayfish more active early in the evening and
small mammals later in the evening. We have
no information, however, concerning the
normal activity patterns of these prey species.
ACKNOWLEDGMENTS
We thank Jeff Hawkins, Michael Moeykens,
Charla Mutchler, Dawn Wilkins, Eric Williams,
Randy Mowrer, and Ann Abbruzzese for field
assistance. Rick Gerhardt and Chris Hill
provided helpful comments on an earlier draft
of this manuscript.
LITERATURE CITED
Andersson M. 1981. On optimal predator
search. Theoretical Population Biology. 19:
58-86.
Belthoff, J.R. 1987. Post-fledging behavior of
the Eastern Screech-owl (Otus asio).
Richmond, KY: Eastern Kentucky
University. M.S. thesis.
Belthoff, J.R.; Ritchison, G. 1990. Roosting
behavior of postfledging Eastern Screech-
owls. Auk. 107: 567-579.
Belthoff, J.R.; Sparks, E.J.; Ritchison, G. 1993.
Home ranges of adult and juvenile Eastern
Screech-owls: size, seasonal variation and
extent of overlap. Journal of Raptor
Research. 27: 8-15.
Bent, A.C. 1938. Life histories of North
American birds of prey, Part 2. Bull. 170.
Washington, DC: U.S. National Museum.
Beissinger, S.R.; Osborne, D.R. 1982. Effects
of urbanization on avian community
organization. Condor. 84: 74-83.
Bull, E.L.; Henjum, M.G. 1990. Ecology of the
Great Gray Owl. Gen. Tech. Rep. GTR-
PNW-265. Portland, OR: U.S. Department
of Agriculture, Forest Service, Pacific
Northwest Research Station.
Bye, F.N.; Jacobsen, B.V.; Sonerud, G.A. 1992.
Auditory prey location in a pause-travel
predator: search height, search time, and
attack range of Tengmalm’s Owls (Aegolius
funereus). Behavioral Ecology. 3: 266-276.
Carlson, A. 1985. Prey detection in the Red-
backed Shrike (Lanius collurio): an experi-
mental study. Animal Behavior. 33: 1243-
1249.
Craighead, J.J.; Craighead, F.C., Jr. 1956.
Hawks, owls, and wildlife. Harrisburg, PA:
Stackpole Co.
Davies, N.B. 1977. Prey selection and the
search strategy of the Spotted Flycatcher
(Muscicapa striata): a field study on optimal
foraging. Animal Behavior. 25: 1016-
1033.
30
Fitzpatrick, J.W. 1981. Search strategies of
tyrant flycatchers. Animal Behavior. 29:
810-821.
Gehlbach, F.R. 1994. The Eastern Screech-owl:
life history, ecology, and behavior in
suburbia and the countryside. College
Station, TX: Texas A&M University Press.
Gehlbach, F.R. 1995. Eastern Screech-owl
(Otus asio). In: Poole, A.; Gill, F., eds. The
birds of North America. No.165. Philadel-
phia, PA: The Academy of Natural Sciences,
and Washington, DC: The American
Ornithologists’ Union.
Greig-Smith, P.W. 1983. Use of perches as
vantage points during foraging by male and
female Stonechats, Saxicola torquata.
Behaviour. 86: 215-236.
Henny, C.J.; Van Camp, L.F. 1979. Annual
weight cycle in wild screech owls. Auk. 96:
795-796.
Hofstetter, S.H. 1995. The provisioning
behavior of adult Eastern Screech-owls:
prey use, sex differences, and the effect of
nestling behavior. Richmond, KY: Eastern
Kentucky University. M.S. thesis.
Johnsgard, P.A. 1988. North American owls:
biology and natural history. Washington,
DC: Smithsonian Institution Press.
Korpimaki, E. 1986. Reversed size dimorphism
in birds of prey, especially in Tengmalm’s
Owl Aegolius funereus: a test of the
“starvation hypothesis.” Ornis
Scandinavica. 17: 326-332.
Leger, D.W.; Didrichsons, I.A. 1994. An
assessment of data pooling and some
alternatives. Animal Behavior. 48: 823-832.
Lundberg, A. 1986. Adaptive advantages of
reversed sexual dimorphism in European
owls. Ornis Scandinavica. 17: 133-140.
Marshall, J.T., Jr. 1967. Parallel variation in
North and Middle American screech owls.
Western Foundation of Vertibrate Zoology
Monograph. 1: 1-72.
Martin, G.R.; Gordon, I.E.; Cadle, D.R. 1975.
Electroretinographically determined
spectral sensitivity in the Tawny Owl
(Strix aluco). Journal of Comparative
Physiological Psychology. 89: 72-78.
Nishimura, K.; Abe, M.T. 1988. Prey suscep-
tibilities, prey utilization and variable
attack efficiencies of Ural Owls. Oecologia.
77: 414-33.
Norberg, R.A. 1970. Hunting technique of
Tengmalm’s Owl Aegolius funereus. Ornis
Scandinavica. 1: 51-64.
Norberg, R.A. 1987. Evolution, structure, and
ecology of northern forest owls. In: Nero,
R.W.; Clark, R.J.; Knapton, R.J.; Hamre,
R.H., eds. Biology and conservation of
Northern Forest Owls. Gen. Tech. Rep. RM-
142. Fort Collins, CO: U.S. Department of
Agriculture, Forest Service, Rocky
Mountain Forest and Range Experiment
Station.
Pinkowski, B.C. 1977. Foraging behavior of the
Eastern Bluebird. Wilson Bulletin. 89: 404-
414.
Rice, W.R. 1982. Acoustical location of prey by
the Marsh Hawk: adaptation to concealed
prey. Auk. 99: 403-413.
Rice, W.R. 1983. Sensory modality: an example
of its effect on optimal foraging behavior.
Ecology. 64: 403-406.
Ritchison, G.; Cavanagh, P.M. 1992. Prey use
by Eastern Screech-owls: seasonal variation
in central Kentucky and a review of
previous studies. Journal of Raptor
Research. 26: 66-73.
Rudolph, S.G. 1982. Foraging strategies of
American Kestrels during breeding.
Ecology. 63: 1268-1273.
SAS Institute. 1989. SAS user’s guide:
statistics. 1989 ed. Cary, NC: SAS Institute.
Smith, D.G.; Gilbert, R. 1981. Backpack
radio transmitter attachment success in
Screech Owls (Otus asio). North American
Bird Bander. 6: 142-143.
31
2nd Owl Symposium
Sonerud, G.A. 1980. Jaktstrategier hos
flyvende smapattedyr predatorer i barskog.
Cand. Oslo, Norway: University of Oslo.
Cand. real. thesis.
Sonerud, G.A. 1989. Search strategies of
predators and avoidance strategies of prey,
illustrated by birds and mammals in the
boreal ecosystem of Fennoscandia. Oslo,
Norway: University of Oslo. Ph.D.
dissertation.
Sonerud, G.A. 1992. Search tactics of a pause-
travel predator: adaptive adjustments of
perching times and move distances by
Hawk Owls, Surnia ulula. Behavioral
Ecology and Sociobiology. 30: 207-217.
32
Sparks, E.J.; Belthoff, J.R.; Ritchison, G. 1994.
Habitat use by Eastern Screech-owls in
central Kentucky. Journal of Field
Ornithology. 65: 83-95.
Temeles, E.J. 1985. Sexual size dimorphism of
bird-eating hawks: the effects of prey
vulnerability. American Naturalist. 125:
485-499.
Voous, K.H. 1989. Owls of the northern
hemisphere. Cambridge, MA: M.I.T. Press.
