Behavioral Ecology Vol. 11 No. 4: 437-443
© 2000 International Society for Behavioral Ecology
Can nest predation and predator type explain variation in dispersal of adult birds during the breeding season ?
a Department of Biology, University of Dubuque, 2000 University Avenue, Dubuque, IA 52001, USA b Institute of Ecology, University of Georgia, Athens, GA 30602, USA
Address correspondence to L. A. Powell. E-mail : lpowell{at}dbq.edu .
Received 13 January 1999; accepted 2 December 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Many types of predators depredate bird nests and thus potentially influence the spatial distribution of their prey. We used a simulation model of a double-brooded songbird's nesting season to test three predictions about the selective advantage of dispersing different distances after nest predation by predators with varying home range sizes. Our results supported the predictions that (1) dispersing birds had higher success than nondispersing birds after predation of the first nest, (2) dispersing beyond the home range of the nest predator increased the success of the second nest, and (3) birds whose first nests were depredated early in the nesting cycle did better by dispersing farther than birds whose nests were depredated later in the nesting cycle. Our results provide evidence that predation and predator characteristics may cause variation in adult dispersal distances during the breeding season. However, we did not find an advantage for long-distance dispersal when predators with small- or medium-sized home ranges were responsible for the predation event. The critical decisions of dispersal and dispersal distance made by adult birds are complex, but our model demonstrates that predation events can create a selective advantage to disperse.
Key words: birds, dispersal, home ranges, nesting cycle, predation, predators..
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Within-year breeding dispersal is common to many avian species, including those species that attempt to raise multiple broods in a single season and species that renest after a predated first nest. Biologists have documented substantial variation in within-year dispersal distances (Beletsky and Orians, 1991
;
Greig-Smith, 1982 ;
Howlett and Stutchbury, 1997 ;
Powell, 1998 ;
Shields, 1984 ;
Stenzel et al., 1994 ;
Thompson and Hale, 1989), but
much of this variation remains unexplained. Variation in dispersal distances
may often be higher than reported ; radio telemetry studies
(Lang, 1998 ;
Powell, 1998 ;
Vega Rivera et al., 1998) have
confirmed that many longdistance dispersals go undetected with mark-recapture
study designs (Greenwood and Harvey,
1982).
We have 6 years of empirical data on wood thrushes (Hylocichla
mustelina) in Georgia (Powell et al.,
2000 ; Frasch L, unpublished data). Wood thrush pairs leave
territories after successful and unsuccessful nests
(Powell, 1998). Therefore the
pair, or female in cases when pairs do not remain intact, must make two
decisions after the first nesting attempt : (1) to remain at the original nest
site or to disperse and, (2) if the choice is to disperse, how far ?
Jackson et al. (1989) made
three predictions about the dispersal response of birds to nest predation.
First, they predicted that an individual should move to a new location after
nest predation if the move will reduce the chances of subsequent predation.
Second, birds with nests taken by predators with large hunting ranges should
move farther than birds with nests taken by predators with smaller hunting
ranges. Last, dispersal should depend on the period in the nesting cycle when
the nest is lost. If the female is nesting late in the season and time is of
essence, Jackson et al. (1989)
predicted that females should not disperse to save the time of finding a new
territory and possibly a new mate. Conversely, if time is of little concern,
Jackson et al. (1989)
suggested that late-cycle failures may indicate a better-than-average
territory (if the average nest fails earlier). Females should then disperse
more often after early-cycle losses. Although Jackson et al.
(1989) found support for their
predictions based on a literature review and analysis of prairie warbler
(Dendroica discolor) data, they did not evaluate the benefit to
fitness of dispersing by comparing nest success rates for dispersing and
nondispersing prairie warblers.
Dispersal distances may be affected by a variety of factors, including
local resource depletion (Grieg-Smith,
1982), conspecific resource competition
(McCarthy, 1997), competition
for breeding territories (Lemel,
1997), sex ratios and gene flow of local populations
(Marzluff and Balda, 1989),
environmental heterogeneity (McPeek and
Holt, 1992), and predictability of habitat quality
(Switzer, 1993). In addition,
Jackson et al. (1989)
suggested that dispersal and dispersal distance could be affected by the type
predation event that caused the need for dispersal.
Nest predation appears to play a key role in determining whether females of
some avian species decide to disperse
(Greenwood and Harvey, 1982).
Powell (1998) reported
breeding season dispersals of 1-17388 m for male and female wood thrushes and
found that wood thrush females dispersed more often and greater distances
after nests failed. Gowaty and Plissner
(1997) found that within-year
dispersal was more common for female eastern bluebirds (Sialia
sialis) that experienced nest predation than for successful females.
Female barn swallows dispersed farther after failed than after successful
nests (Shields, 1984). None of
the above studies reported the benefits of dispersal in terms of subsequent
success rates of dispersing females as compared to nondispersing females.
However, Howlett and Stutchbury
(1997) suggested that
predation does not play a role in the selection of renesting sites of hooded
warblers (Wilsonia citrina), which dispersed farther after successful
nesting attempts than after failed nesting attempts. Beletsky and Orians
(1991) reported that female
red-winged blackbirds (Agelaius phoeniceus) breeding season dispersal
distances were unrelated to nest success, and the females did not move farther
in response to predators with larger home ranges.
Modeling exercises can be useful during the development of ecological
theory (Conroy, 1993). We
believe there are benefits to using modeling to refine initial hypotheses when
determining the selective advantage of dispersal in the context of a variety
of predation events. First, field biologists must use predator-sensing cameras
(Picman, 1987) to accurately
determine the cause of nest failure. Second, in field experiments predation
events cannot be controlled to ensure adequate, even sample sizes of nests
depredated by different types of predators. Third, field biologists must use
intense field observations of banded birds (which usually produces only a
small proportion of the subsequent nest locations) or radio telemetry to
document the female's dispersal movements. In sum, the best-designed field
study, using predator cameras and radio telemetry, may not result in adequate
samples of nests depredated by certain types of predators to test all of
Jackson et al.'s (1989)
hypotheses. Therefore, in this study we chose to use a simulation model to
evaluate the predictions of Jackson et al.
(1989). Our goal was to test
their predictions about the selective advantage of dispersal after nest
predation. Our modeling exercise allows (1) large sample sizes, (2) the
ability to "determine" fates of nests and "follow"
females between nests, (3) the ability to detect relative increases or
decreases in nest success under a range of dispersal distances, and (4)
sensitivity analyses using various levels of predator densities. Our model
system's level of simplicity versus generality was thus affected by the need
to test specific hypotheses. As field biologists, our intent was to produce
data that could be compared to natural populations.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Model structure
We simulated the nesting cycle of a double-brooded species, with characteristics based on the life history of the wood thrush. The model consisted of a single female bird that nested at a random location within a 100-km2 area. During a 24-day nesting period (Roth et al., 1996
), predators
searched for the nest. We defined nest failure as a single predatory event
during which a predator found the nest. After a successful nesting period or
nesting failure, the female renested at a fixed distance from the original
nest within the same 100-km2 area. The second nest location was
chosen randomly from all possible locations at the required dispersal distance
from the first nest. If the location chosen was outside the 100-km2
area, another location was selected. The model then simulated a second 24-day
nesting period during which predation of the nest could occur. We simulated predation by randomly placing a predetermined density of predators within the 100-km2 nesting area. Each predator hunted within a home range defined by maximum movements from a central location and a "daily kill area" (see below). During every day of the nesting cycle each predator moved from its central location to a point determined by x- and y-axis values randomly selected from a uniform distribution. The predator caused nest failure if the nest was located within a fixed area surrounding the hunt location, the "daily kill area" (Figure 1). After hunting, the predator returned to the central location at the end of the day. Predators' central locations remained fixed throughout the simulation of both nests.
|
We simulated nesting, predation, and dispersal under 12 sets of parameter
values, consisting of combinations of three types of predators and four
dispersal distances. The predator types included : (1) 30 predators with large
home ranges (LHR), 1400 x 1400-m home ranges and 400 x 400-m daily
kill areas (e.g., raptor), (2) 400 predators with medium home ranges (MHR),
400 x 400-m home ranges and 100 x 100-m daily kill areas [e.g.,
mid-sized mammal, crow (Corvus brachyrynchos), blue jay
(Cyanocitta cristata), or snake], and (3) 10,000 predators with small
home ranges (SHR), 30 x 30-m home ranges and 10 x 10-m daily kill
areas [e.g., fire ant colony (Solenopsis invicta), mouse]. Each of
the above simulations were "monoculture" predator models, with
only one type of predator in each simulation. In addition, we developed a
mixed predator model in which all three types of predators, at the same
respective densities as in the monoculture models, simultaneously hunted for
nests. The model selected daily predator movements for each individual from a
uniform distribution. We chose predator densities for the monoculture models
by running practice simulations until we found a density of predators that
would result in 20-50% mortality of the first randomly placed nest, given the
predator's home range size and daily kill area. We used avian breeding season
dispersal distances of (1) no dispersal, (2) 50 m, (3) 500 m, and (4) 5000 m,
based loosely on wood thrush movements
(Powell, 1998).
Model assumptions
For modeling simplicity and testing specific hypotheses, we assumed that
(1) predators' movements are random and not affected by other prey, (2)
predators will cause nest failures if they are in the vicinity of the nests,
(3) predators are not attracted to the nest by cues that may change temporally
(e.g., nestlings' begging calls and parents' feeding flights), and (4)
dispersal distances are not affected by food resource or territory
availability.
The interpretation of our results requires a number of assumptions about
the natural history of female birds and nest predators. Our underlying
assumption is that birds must (1) genetically inherit or (2) efficiently learn
the tendency to disperse within a breeding season and the ability to correctly
choose a dispersal distance when dispersing as a response to predation
(Arcese, 1989 ;
Greenwood et al., 1979). Payne
and Payne (1993) found no
evidence that indigo buntings (Passerina cyanea) had an inherited
tendency to disperse, but migratory direction seems to be innate for juveniles
(Perdeck, 1967). Natal
dispersal, presumably, occurs before much learning can take place. Still, many
biologists have described large, unexplained variation in natal dispersal
distances (Lang, 1998 ;
Plissner and Gowaty, 1996 ;
Vega Rivera et al., 1998),
including independent movements and dispersal distances of siblings
(Lang, 1998 ;
Vega Rivera et al., 1998).
Genetic transfer of dispersal characteristics may not be necessary if
"public knowledge" information can be distributed among
individuals (Templeton and Giraldeau,
1995). Therefore, mechanisms may be available for transmittal of
avian responses to predators via genetic or nonheritable means.
We also assume that female birds are always able to ascertain which
predators are responsible for their nest failures. To satisfy this assumption,
birds may (1) witness the predation event, (2) determine from constant
nest-guarding duties which predators are most likely to depredate the nest, or
(3) determine the identity of an unseen predator from visual or other cues.
Wood thrush pairs divide nest-guarding duties
(Roth et al., 1996), and
female wood thrushes are sometimes killed at the nest site
(Powell, 1998). This evidence
would suggest that disturbances at the nest site are usually observed by one
or both members of the pair. At present, no evidence exists about birds'
ability to evaluate threats from potential predators or to use visual or other
cues to determine the identity of a predator.
Analysis
The model outputs consisted of the fate of each nest and the day of failure
(days 1-24). We conducted 200 simulations of the model for each set of initial
parameters ; parameters (dispersal distance and predator type) were kept
constant throughout the 200 simulations. Nest success rates were the
proportion of sample (usually n = 200) with successful (nonpredated)
nests. We used frequency tables (SAS,
1990) to describe the proportion of first and second nests that
failed and the proportion of females that were successful twice, once, or
never. Because this was not a field experiment, we could not evaluate the true
dispersal response of each female to predation events (dispersal distances
were constant parameters in our model). But we could determine the benefits of
dispersal after nest failure (prediction 1) by evaluating the success of each
dispersal strategy for the monoculture and mixed predator models. We used 95%
confidence intervals for each binomial sample proportion
(Burleson, 1980) to describe
the proportion of second nests that were successful, given that the first nest
failed, and proportion of second nests that were successful, given that the
first nest was successful. We pooled all dispersal categories (5000, 500, and
50 m) in a chi-square test to determine the benefits of dispersing. To
determine the sensitivity of the model to predator density, we compared the
results of the monoculture MHR predator simulations with a second set of
simulations that incorporated a density (n = 800) double the original
predator density (n = 400).
We tested for a selective advantage to disperse beyond the predator's home range (prediction 2) by comparing the success of second nests of birds whose first nest had been depredated by different-sized predators. We used 95% confidence intervals of the proportion of successful nests to make comparisons among the three predator types. To test whether the benefits of dispersal changed during the nesting cycle (prediction 3), we determined the proportion of successful second nests following failures, caused by MHR, monoculture predators, in the egg (days 1-12) or nestling (days 13-23) stage. We used chi-square statistics and 95% confidence intervals to make comparisons. We also used frequency tables to determine the distribution of nest failures during the nesting cycle (discrete categories : days 0-6, 7-12, 13-18, and 19-24), and we tested for constant daily nest success using the chi-square statistic.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Simulations with monoculture LHR and SHR predators resulted in overall nest success rates of 54-61% and 73-81%. Monoculture MHR predator simulations resulted in 56-60% and 27-37% overall nest success rates for normal and double predator densities, respectively. Fifty-six to sixty percent of the nests were successful during the mixed-predator simulations. For each predator type, overall nest success rates were similar among all dispersal distances (X2 test, each p [UNK].05).
After a nest failure, dispersing birds (all distances pooled) did better
than nondispersers in each monoculture predator model (LHR predator :
X2 = 6.68, df = 1, p =.010 ; MHR predator : X2
= 35.70, df = 1, p =.001 ; SHR predator : X2 = 59.68, df =
1, p =.001), supporting Jackson et al.'s
(1989) first prediction
(Figure 2A-C). Dispersers also
did better than nondispersers in the mixed-predator model (X2 =
8.90, df = 1, p =.001 ; Figure
2D). Birds with successful first nests did better by remaining at
the same site for their second nest (LHR predator : X2 = 10.76, df
= 1, p =.001 ; MHR predator : X2 = 7.36, df = 1,
p =.007 ; SHR predator : X2 = 23.18, df = 1, p
=.001 ; mixed predator : X2 = 15.44, df = 1, p =.001),
also supporting Jackson et al.'s
(1989) first prediction
(Figure 2). The benefits of
long- and medium-distance dispersal after a nest failure were negated by
higher predator densities in the MHR predator model. Under high predator
densities, short-distance dispersers and nondispersers were still more
successful after a successful first nest than medium- and long-distance
dispersers (Figure 3).
|
|
Nondispersers tended to have a higher chance of recording no successful nests during a breeding season than short-, medium-, and long-distance dispersers. The trend was less significant for birds in LHR predator models than for those in MHR and SHR predator models. Dispersers in all monoculture predator models tended to have a better chance than nondispersers of recording one successful nest during the breeding season (Figure 4). In the mixed predator model, 49% of nondispersers and short-distance dispersers recorded at least one success, while 55% of medium-distance dispersers and 69% of long-distance dispersers had at least one success.
|
The type of predator causing the first nest's failure had an impact on the
dispersal distance necessary to ensure the success of the second nest,
supporting Jackson et al.'s
(1989) second prediction. In
general, the most successful birds dispersed beyond the home range size of the
predator responsible for the first nest's failure (LHR predator, 2400-m wide
kill area : X2 = 38.81, df = 1, p <.001 ; MHR predator,
700 m : X2 = 94.06, df = 1, p <.001 ; SHR predator, 50
m : X2 = 59.68, df = 1, p <.001). Birds whose nests
were depredated by an LHR predator had to disperse 5000 m, the longest
distance, to achieve high second-nest success rates (52%). Birds whose nests
were depredated by an SHR predator only had to disperse 50 m to achieve
second-nest success rates similar to long distance dispersers (60-83%,
Figure 2). In the MHR predator
model, medium-distance dispersers did better than long-distance dispersers
after a failed nest (long-distance, 55%, versus medium-distance, 88% :
X2 = 22.80, df = 1, p <.001).
Birds in the MHR predator model whose nests were depredated in the egg
stage had to disperse at least 500 m to increase the chances of the second
nest's success. However, birds whose nests were depredated in the nestling
stage only had to disperse 50 m to increase the success of the second nest
(X2 = 7.44, df = 1, p =.006 ;
Figure 5). Because birds in our
model were not constrained by the approaching end of the breeding season, this
supports Jackson et al.'s
(1989) third prediction. It
also suggests that the selective benefit of longer dispersal is not as high
for birds whose nests are not discovered until the end of the nesting cycle.
Daily nest failure rates were not constant in our model, as predators found
more nests early in the cycle than late in the cycle (first and second nests
pooled : X2 = 184.60, df = 3, p <.001 ;
Table 1,
Figure 6). Egg and nestling
stage failure rates of first nests were 19.5% and 5.6% for SHR predators,
29.0% and 20.1% for MHR predators, and 29.6% and 18.5% for LHR predators.
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our model provides evidence that predator characteristics may cause variation in adult avian dispersal distances during the breeding season, as Jackson et al. (1989
)
predicted. Specifically, longer dispersal distances could be adaptive in LHR
monoculture and mixed-predator situations, and shorter dispersal distances
could be adaptive when predators with smaller home ranges cause nest failure.
Our results emphasize that a correct set of dispersal strategies, following
failed or successful first nests, can be highly advantageous. Also, our
simulations provide evidence that breeding season dispersal, of any distance,
results in a higher chance of having at least one successful nest during the
breeding season than a strategy of nondispersal. Wood thrushes disperse after
successful and failed nests (Powell,
1998). Nondispersing birds could experience a boom (2 successful
nests) or bust (no successful nests) breeding season, depending on the
landscape characteristics and abundance of predators at the nesting site
(Tewksbury et al., 1998).
Our results suggest that different types of predators can promote a
selective advantage for variation in adult dispersal distance during the
breeding season. However, in the monoculture models there was no advantage for
dispersing farther than a threshold distance, just beyond the home range size
of the predator (Figure 2). The
added energy and time required to make the longer dispersal would be a high,
prohibitive cost (Greenwood and Harvey,
1982), especially when longer dispersals provide no advantage to
fitness. Our mixed-predator model, perhaps more real than the monoculture
models, showed that long-distance dispersals were advantageous. Powell
(1998) documented extremely
long-distance dispersals (e.g., > 10 km) for wood thrushes, and our model
did not provide comparisons for such long distances. Predators probably do not
operate on such large spatial scales, and these dispersals would seem to be
caused by another mechanism, unless a traumatic predation event (e.g.,
predator causes death of mate in addition to depredating the nest) has the
potential to trigger an unnecessary, extremely long-distance dispersal
response by the female. Savill and Hogeway
(1998) modeled predator-prey
interactions and found that ever-increasing dispersal distances were
advantageous as prey attempted to escape the local predator population, but
this result was mainly a function of the structure of their model, which did
not include predator populations throughout the landscape at time zero.
The structure of our model must be considered when drawing conclusions from
our results. First, we based most of our conclusions on monoculture predator
simulations to avoid confounding the effects of one type of predator with
another. Most predator guilds are, of course, more complex. Predators with
large, medium, and small home ranges are often responsible for the predation
of bird nests (e.g., Leimgruber et al.,
1994 ; Picman,
1987 ; Simons and Farnsworth,
1996), and predator densities may be variable across the
landscape. However, our mixed-predator model also showed an advantage for
dispersing after a failed nest and remaining at the same spot after a
successful nest. Our monoculture model showed that increased predator
densities can lower the advantage of dispersal
(Figure 3), which is applicable
to mixed populations of predators.
Second, we selected a combination of density, movements, and daily kill areas for each predator type to cause 20-50% nest failure in the 100-km2 nesting area. It is not possible, nor was it the purpose of this model, to draw conclusions about the overall effect of each predator type on avian nesting success because the threat of each predator type could have been adjusted simply by changing predator densities or resizing the daily kill area. However, because each predator's characteristics were held constant throughout the simulations of different dispersal distances, it is valid to draw conclusions on the effect of different types of predators on the reproductive success of birds using a variety of dispersal strategies.
Third, we did not incorporate between-female variation in nest vigilance,
or the ability to defend the nest against predators. Nest vigilance also
remained the same after a failed or successful first nest. If females increase
their vigilance after a predation event, the need to disperse might be less.
Natural between-female variation in nest vigilance could also make dispersal
less profitable for highly vigilant females. Therefore, variation in vigilance
may be responsible for some of the unexplained variation in dispersal
strategies reported by Beletsky and Orians (1992), Greig-Smith
(1982), Howlett and Stutchbury
(1997), Powell
(1998), Shields
(1984), Thompson and Hale
(1989), and Stenzel et al.
(1994).
Last, the distribution of nest failures throughout the nesting cycle (Figure 6) was caused by the predator's characteristics that we chose and the structure of the model that allowed the predator to randomly hunt in a home range centered on a fixed location. Predators in our model had no memory of nests found earlier, but the high proportion of second nests that failed under the nondispersing strategy indicates that predators were quite efficient at finding nests located in their territories (Figure 2). SHR predators found fewer nests and found most nests during the first 6 days, which was different from MHR or LHR predators (Table 1). This indicates that SHR predators in our model covered less of the available space and covered more previously searched area later in the nesting cycle than LHR or MHR predators. Despite these differences, all types of predators depredated 80-90% of the nests of nondispersing birds after a failed first nest (Figure 2).
The fixed, central location of our predator created higher rates of predation for nests near the central location than near the boundary of the predator's home range. This characteristic of the model, although biologically realistic (hunting patterns of predators with dens or nests would be centered near those areas), probably was the reason that birds in the MHR model with nests predated later had a higher chance of success after 50-m dispersals than birds with nests predated early (Figure 5).
Lang (1998) reported that
the nestling stage was less successful than the egg stage in wood thrushes,
contrary to the assertion by Jackson et al.
(1989) that the egg stage is
the most vulnerable. Predation in our model was higher in the egg stage
(Figure 6), which is
biologically realistic. Cowbirds (Molothrus ater) must parasitize
nests early in the nesting cycle, and other predators such as blue jays,
eastern chipmunks (Tamias striatus), squirrels (Sciurus
spp.), and American crows that steal eggs from nests must find them during the
egg stage. It is possible that methodical, rather than random, hunting tactics
might produce higher chances of failure later in the nesting cycle. Therefore,
our model should not be used to predict nest success rates ; our intent was to
compare relative successes under a variety of dispersal strategies.
The critical decisions of dispersal and how far to disperse are probably
very complex. Nonetheless, our model demonstrates that, all else being equal,
predation events can create a selective advantage to disperse. In addition,
the predator type and the timing of the predation event during the nesting
cycle created variation in success of dispersers of various distances in our
model. This suggests that birds that respond to predation events correctly by
choosing the best dispersal distance will have an advantage over others when
all else is equal. Given the variability in adult dispersal during the
breeding season and the apparent importance of predation on dispersal choices,
field experiments designed to examine variability in dispersal distances
during the breeding season and the relative importance of predation on
dispersal may be warranted. Further, dispersal of songbirds can have critical
effects on conservation of species (Donovan
et al., 1995) especially when available dispersal habitats vary in
food abundance and predator or nest parasite levels. Our results suggest that
management decisions that could affect populations of predators across a
landscape may also affect songbird movement.
|
ACKNOWLEDGEMENTS |
|---|
We thank P. A. Gowaty, C. W. Beck, and two anonymous referees for helpful comments on earlier versions of this manuscript. The Georgia Cooperative Fish and Wildlife Research Unit, University of Georgia, provided computer facilities.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Arcese P, 1989. Intrasexual competition, mating system and natal dispersal in song sparrows. Anim Behav 38 : 958-979.
Beletsky LD, Orians GH, 1991. Effects of breeding experience and familiarity on site fidelity in female red-winged blackbirds. Ecology 72 : 787-796.
Burleson DR, 1980. Elementary statistics. Cambridge, Massachusetts : Winthrop Publishers.
Conroy MJ, 1993. The use of models in natural resource management : prediction, not prescription. Trans N Am Wildl Nat Res Conf 58 : 509-519.
Donovan TM, Lamberson RH, Kimber A, Thompson FR III, Faaborg J, 1995. Modeling the effects of habitat fragmentation on source and sink demography of neotropical migrant birds. Conserv Biol 9 : 1396-1407.
Gowaty PA, Plissner JH, 1997. Breeding dispersal of eastern bluebirds depends on nesting success but not on removal of old nests : an experimental study. J Field Ornithol 68 : 323-330.
Greenwood PJ, Harvey PH, 1982. The natal and breeding dispersal of birds. Annu Rev Ecol Syst 13 : 1-21.[Web of Science]
Greenwood PJ, Harvey PH, Perrins CM, 1979. The role of dispersal in the great tit (Parus major) : the causes, consequences and heritability of natal dispersal. J Anim Ecol 48 : 123-142.
Greig-Smith PW, 1982. Dispersal between nest-sites by stonechats (Saxicola torquata) in relation to previous breeding success. Ornis Scand 13 : 232-238.
Howlett JS, Stutchbury BJM, 1997. Within-season dispersal, nest-site modification, and predation in renesting hooded warblers. Wilson Bull 109 : 643-649.
Jackson WM, Rohwer S, Nolan V, Jr. 1989. Within-season breeding dispersal in prairie warblers and other passerines. Condor 91 : 233-241.
Lang JD, 1998. Effects of thinning and prescribed burning in pine habitat on nesting success, fledgling dispersal, and habitat use by wood thrushes (MS thesis). Athens : University of Georgia.
Leimgruber P, McShea WJ, Rappole JH, 1994. Predation on artificial nests in large forest blocks. J Wild Manage 58 : 254-260.
Lemel JY, Belichon S, Clobert J, Hochberg ME, 1997. The evolution of dispersal in a two patch system : some consequences of differences between migrants and residents. Evol Ecol 111 : 613-629.
Marzluff JM, Balda RP, 1989. Causes and consequences of female-biased dispersal in a flock-living bird, the Pinyon Jay. Ecology 70 : 316-328.
McCarthy MA, 1997. Competition and dispersal from multiple nests. Ecology 78 : 873-883.
McPeek MA, Holt RD, 1992. The evolution of dispersal in spatially and temporally varying environments. Am Nat 140 : 1010-1027.
Payne RB, Payne LL, 1993. Breeding dispersal in Indigo Buntings : circumstances and consequences for breeding success and population structure. Condor 95 : 1-24.
Perdeck AC, 1967. Orientation of starlings after displacement to Spain. Ardea 55 : 194-202.
Picman J, 1987. An inexpensive camera set-up for the study of egg predation at artificial nests. J Field Ornithol 58 : 372-382.
Plissner JH, Gowaty PA, 1996. Patterns of natal dispersal, turnover and dispersal costs in eastern bluebirds. Anim Behav 51 : 1307-1322.
Powell LA, 1998. Experimental analysis and simulation modeling of forest management impacts on wood thrushes, Hylocichla mustelina (PhD dissertation). Athens : University of Georgia.
Powell LA, Lang JD, Conroy MJ, Krementz DG, 2000. Effects of forest management on density, survival, and population growth of wood thrushes. J Wildl Manage 64 : 11-23.
Roth RR, Johnson MS, Underwood TJ, 1996. Wood thrush (Hylocichla mustelina). In : The Birds of North America, no. 246 (Poole A, Gill F, eds). Washington, DC : The American Ornithologists' Union.
SAS Institute, 1990. SAS/STAT user's guide, 4th ed. Cary, North Carolina : SAS Institute Inc.
Savill NJ, Hogeweg P, 1998. Spatially induced speciation prevents extinction : the evolution of dispersal distance in oscillatory predatorprey models. Proc R Soc Lond B 265 : 25-32.[Medline]
Shields WM, 1984. Factors affecting nest and site fidelity in Adirondack barn swallows (Hirundo rustica). Auk 101 : 780-789.
Simons T, Farnsworth G, 1996. Evaluating Great Smoky Mountains National Park as a population source for wood thrush. 1995 annual report. Raleigh, North Carolina : National Biological Service Cooperative Park Studies Unit.
Stenzel LE, Warriner JC, Warriner JS, Wilson KS, Bidstrup FC, Page GW, 1994. Long-distance breeding dispersal of snowy plovers in western North America. J Anim Ecol 63 : 887-902.
Switzer P, 1993. Site fidelity in predictable and unpredictable habitats. Evol Ecol 7 : 533-555.
Templeton JJ, Giraldeau L-A, 1995. Patch assessment in
foraging flocks of European starlings : evidence for the use of public
information. Behav Ecol 6 :
65-72.
Tewksbury JJ, Hejl SJ, Martin TE, 1998. Breeding productivity dows not decline with increasing fragmentation in a western landscape. Ecology 79 : 2890-2903.[Web of Science]
Thompson PS, Hale WG, 1989. Breeding site fidelity and natal philopatry in Redshank Tringa totanus. Ibis 131 : 214-224.
Vega Rivera JH, Rappole JH, McShea WJ, Haas CA, 1998. Wood thrush postfledging movements and habitat use in northern Virginia. Condor 100 : 69-78.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||