Journal of Tropical Ecology (2006) 22:385?395. Copyright ? 2006 Cambridge University Press
doi:10.1017/S0266467406003191 Printed in the United Kingdom
Seasonal shift in the foraging niche of a tropical avian resident:
resource competition at work?
Julie A. Jedlicka?1, Russell Greenberg?, Ivette Perfecto?, Stacy M. Philpott?1 and Thomas V. Dietsch?#
? Department of Ecology and Evolutionary Biology, University of Michigan, 830 N. University Ave., Ann Arbor, MI 48104, USA
? Smithsonian Migratory Bird Center, National Zoological Park, 3001 Connecticut Ave NW, Washington DC 20008, USA
? School of Natural Resources and the Environment, University of Michigan, 430 E. University, Ann Arbor, MI 48109, USA
# Center for Tropical Research, University of California, La Kretz Hall, Suite 300, Box 951496, Los Angeles, CA 90095-1496, USA
(Accepted 23 December 2005)
Abstract: This study examined the foraging behaviour of a resident bird species, the rufous-capped warbler (RCWA,
Basileuterus ru?frons), in a shaded-coffee farm in Chiapas, Mexico. Unlike many resident species that use shaded-coffee
agroecosystems seasonally, RCWAs do not move to other habitats when migrants are present. RCWA foraging was
compared when migrant birds were present (dry season) and absent (wet season). It was hypothesized that RCWAs
would exhibit a seasonal foraging niche shift because of resource competition with migrants. Observations from both
the canopy and coffee understorey show that RCWAs foraged almost equally in both vegetative layers during the wet
season although they were more successful foraging in the canopy. In the dry season, migrants foraged primarily in
the canopy and RCWAs shifted so that 80% of RCWA foraging manoeuvres were in the understorey. At that time
RCWAs foraged less successfully in both vegetative layers. Avian predation in the dry season was found to reduce
densities of arthropods by 47?79% in the canopy, as opposed to 4?5% in the understorey. In the canopy, availability
of large (>5 mm in length) arthropods decreased by 58% from the wet to dry season. Such resource reductions could
have caused the RCWA foraging niche shift yet other alternative or additional hypotheses are discussed. Shifts in
foraging niche may be a widespread mechanism for some small insectivorous residents to avoid seasonal competition
with abundant migrant species.
Resumen: Este estudio examino? el comportamiento de forrajeo de un ave residente, Basileuterus ru?frons (RCWA), en
una ?nca de cafe? con sombra en Chiapas, Me?xico. A diferencia de muchas aves residentes que usan los agroecosistemas
de cafe? con sombra solamante durante una estacio?n, RCWAs no se van a otros ha?bitats cuando los aves migrantes esta?n
presente. El forrajeo de RCWA fue comparado cuando las aves migrantes eran presente (la e?poca seca) y ausente (la
e?poca de lluvia). La hipo?tesis fue que los RCWA exhibir??an un cambio de forrajeo con los cambios de estaciones a causa
de la competencia de recursos con los migrantes. Observaciones en el docel y el sotobosque en un cafetal muestran
que durante la e?poca de lluvia, los RCWAs forrajean igualmente en los dos niveles de vegetacio?n, pero tienen ma?s
e?xito forrajeando en el docel. Durante la e?poca seca, los migrantes forrajean principalmente en el docel y los RCWAs
se mueven al sotobosque donde efectu?an 80% de las maniobras de forrajeo. Durante ese tiempo, los RCWAs tuvieron
menos e?xito forrajeando tanto en el docel como en el sotobosque. Durante la e?poca seca el nu?mero de artro?podos
bajo? entre 47?79% en el docel y entre 4?5% en el sotobosque a causa de la depredacio?n de las aves. En el docel, la
disponibilidad de artro?podos grandes (>5 mm en longitud) bajo? en un 58% de la e?poca de lluvia a la e?poca seca. Tales
reducciones de recursos podr??an causar el cambio de lugar a forrajeo observado en RCWA, pero otras explicaciones e
hipo?tesis son discutidas. Pueden ser que este cambio de forrajeo sea comu?n en algunas aves residentes pequen?as que
comen artro?podos para evitar la competencia con los migrantes durante la e?poca seca.
Key Words: avian insectivory, birds, breeding demand, Chiapas, Mexico, coffee agroecosystem, direct interference,
foraging, indirect competition, migrant?resident interaction, resource competition, rufous-capped warbler, sexual
segregation, shaded coffee
1 Corresponding author. Current address: Environmental Studies
Department, University of California, 1156 High Street, Santa Cruz,
CA 95064, USA. Email: jenvs@ucsc.edu
INTRODUCTION
Each year millions of insectivorous birds arrive in tropical
habitats during what is often the resource-poor time
386 JULIE A. JEDLICKA ET AL.
of year. Not surprisingly, avian ecologists have focused
considerable effort on understanding the mechanisms by
which migratory and resident populations are able to
coexist (Greenberg 1995, Johnson et al. 2005, Rappole &
Warner 1980, Ricklefs 1989). The interaction of the two
avifaunas may be connected to the population dynamics
of organisms at lower trophic levels in tropical areas. A
large annual in?ux of insectivores with no detectable
competitive effects would suggest that predator?prey
populations are not tightly coupled. Studies examining
the foraging behaviour of overwintering migrant birds in
tropical habitats are common (see Greenberg 1986 for
a review), yet surprisingly few have investigated how
foraging behaviour of tropical residents is affected by
migrant presence. Previous comparative studies showed
that many resident birds were specialists and required
more speci?c microhabitat conditions than did migrant
species (for review see Greenberg 1986, Herrera 1978,
Leisler 1990, Remsen & Parker 1983, Stotz et al.
1996). Accordingly, most migrants have been thought
to avoid resource competition while in the tropics by
foraging for resources underused by resident species
and underrepresented in their foraging microhabitats
(Herrera 1978, Lack 1986, for review see Leisler 1990,
1992; O?Connor 1981). However data from many studies
fail to support these hypotheses (Johnson & Sherry
2001, Lack 1986, Post 1978, Strong 2000) and these
hypotheses are now recognized as overly simplistic
(Greenberg pers. comm.).
The four studies that examine how tropical residents
forage before and after migrant arrival found that while
a majority of resident species did not appear to change
their foraging behaviour upon migrant arrival, a few
generalist resident species exhibited seasonal shifts in
foraging niche (Chipley 1976, Rab?l 1987, 1993; Waide
1981). In these cases, certain resident bird species were
found to forage higher in the vegetative strata (in
more favourable or preferred locations) when migrants
were absent than when they were present. Contrary to
conventional wisdom that resident birds out-compete
migrants, these resident resource generalists may shift
to under-exploited niches to avoid resource competition
when migrants are very abundant.
The resource competition hypothesis is commonly offered
as a possible explanation for this behavioural shift
(Chipley 1976, for review see Leisler 1990, 1992;
Rab?l 1987, Waide 1981). When migrants arrive they
consume large numbers of arthropods in the canopy,
and the resource competition hypothesis holds that this
consumption results in such low densities of arthropods
that it is no longer pro?table for certain residents to
forage in the canopy. In turn, residents forage lower
in the vegetative strata. If this indirect competition
with migrants is driving the displacement, the following
predictions should be met: (1) shared food resources
should be reduced where migrants primarily forage and
(2) resident birds should forage less successfully at those
locations once migrants are present. One would expect
that with time after arrival, migrants would also forage
less successfully due to resource depletion, but testing this
prediction is outside the realm of this study.
Shade-grown coffee farms in the Americas usually
support a few resident bird species and a large number
of insectivorous neotropical migrant species that arrive
in high densities around the beginning of the wet season
(Calvo & Blake 1998, Greenberg et al. 1994, 1997a,b;
Johnson 2000, Johnson & Sherry 2001, Perfecto et al.
2003, Petit et al. 1993, Robbins et al. 1992, Vannini
1994, Willis 1966, Wunderle 1999, Wunderle & Latta
1996, Wunderle & Waide 1993). Because shaded-coffee
farms support such large densities of migrants and
because the vegetation is oftentimes heavily pruned,
eliminating much of the vegetative complexity found
in a tropical forest, it is a comparatively easier system
to look for shifts in foraging niche. This study was
conducted on a shaded-coffee farm where the rufous-
capped warbler (RCWA, Basileuterus ru?frons) was the
numerically dominant foliage-gleaning resident species.
Additionally the RCWA was chosen for this study because
it spends a considerable amount of time foraging in the
coffee understorey. Unlike many resident species that
use shaded-coffee agroecosystems seasonally, RCWAs
maintain territories year-round and do not move to
other habitats when migrants are present (Dietsch
2003). They breed during April?July when migrants
are largely absent (Stiles & Skutch 1989). By observing
both foraging behaviour and location of RCWA in the
wet and dry season, this study tests the hypothesis that
rufous-capped warblers exhibit a seasonal foraging niche
shift. Also, by coupling these data with measurements
of arthropod abundance and migrant presence, this
study tests the hypothesis that the seasonal niche
shift is caused by resource competition from migrants.
Alternative hypotheses for this seasonal niche shift, such
as climatic and seasonal effects, breeding demand, sexual
segregation, predator avoidance and direct interference
hypotheses are also examined in the discussion.
METHODS
Overview of study site
This research was conducted in Finca Irlanda (1040 m
asl, 15?10?N, 92?20?W), a shade-grown, organic coffee
farm located on the Paci?c slope of the Sierra Madre del
Sur, 40 km north-east of Tapachula in the Soconusco
region of Chiapas, Mexico. In this region there is a
pronounced wet season from March to October that
receives approximately 4400 mm y?1 rainfall. The dry
Seasonal shift in the foraging niche of a tropical avian resident 387
Figure1.Finca Irlanda photograph illustrating the two distinct vegetative
layers.
season lasts from November to February and receives
close to 60 mm of rain during the entire 4-mo period
(Lin unpubl. data, Richter 2000). The study took place
where the canopy consists of approximately seven tree
species that are regularly pruned to a height of roughly
9 m, creating a consistent canopy cover of about 60%
(Mas & Dietsch 2003, Perfecto et al. 2004) with scattered
emergent trees. Approximately 60% of the canopy trees
belong to the genus Inga (Fabaceae). The farm is classi?ed
as a commercial polyculture (sensu Moguel & Toledo
1999, Perfecto & Vandermeer 2002, Philpott et al. 2004),
which means that coffee is grown under a planted canopy.
The coffee understorey and canopy form two distinct
strata of vegetation (Figure 1).
Foraging observations
In June 2002, 22 adult and hatch year rufous-capped
warblers were banded with split plastic Avinet colour
bands on a plot (approximately 275 ? 250 m) ?agged at
25 m intervals. Foraging observations were made on ban-
ded and unbanded RCWAs from 18 June 2002?17 July
2002, and from 18 December 2002?13 January 2003.
No more than one foraging observation was made
per banded individual per day. Two or more foraging
observations were recorded on unbanded RCWA
individuals on the same day only if the distance between
observations was over 150 m, a distance assumed to
minimize the probability of observing the same individual.
This distance was chosen based on territory maps drawn
from banded RCWAs in the area.
The foraging protocol used was adopted from
Greenberg et al. (1999) and Dietsch (2003) based on
Remsen & Robinson (1990). In order to reduce any
bias towards conspicuous individuals, RCWAs were
located both aurally and visually. RCWAs were followed
until a foraging manoeuvre was observed. Foraging
manoeuvres were de?ned as actual foraging attacks, not
merely searching for food. For each foraging manoeuvre
recorded, the bird?s height from the ground was estimated,
the plant species (where the manoeuvre took place)
identi?ed, and RCWA colour bands were recorded.
Because only male RCWAs sing to defend territories, it was
noted whether the bird was singing. Attack manoeuvre
was noted as either standing, aerial, or hanging. If the
foraging attempt was ?clearly successful? the prey was
identi?ed, usually to insect order. If a bill wipe immediately
followed the foraging manoeuvre it was recorded as
?successful not seen?. If prey capture or bill wipes were
not visible, the manoeuvre was considered unsuccessful.
Arthropod sampling
In the same study site, two different bird exclosure
experiments were carried out to determine the effects of
birds on arthropods in the coffee understorey (Perfecto
et al. 2004) and canopy (Philpott et al. 2004). In the
coffee understorey, ?shing gill nets (35 ? 35-mm mesh)
were used to form bird exclosure cages of approximately
10 ? 5 ? 3 m over at least 10 coffee plants in November
2000. Arthropods were collected from ten plants inside
the exclosure and ten adjacent plants outside the
exclosure (controls). Arthropods were vacuumed from
two branches per plant in November 2001 and May 2002
into ?ne mesh bags with a 10-cm-diameter reversed leaf-
blower (D-vac). Ethyl acetate was added to the bags to
kill arthropods, which were later identi?ed to order (or
family) and measured in length (mm). Leaf biomass
was estimated (as described by Perfecto et al. 2004)
and arthropod data were standardized as number of
individuals g?1 of wet foliage.
Bird exclosures in the canopy were established in Inga
micheliana trees in the dry season (12 December 2001 ?
12 February 2002) and wet season (2 May 2002 ? 15 July
2002; see Philpott et al. 2004). The exclosures were placed
in 18 trees in the dry season and 20 in the wet season.
Two branches that were 3?4 m above ground with 4?8
leaves were selected and randomly assigned to a control or
exclosure treatment. Birds were excluded from foraging
on Inga foliage by placing mono?lament nylon ?shing
nets (35 ? 35-mm mesh) over entire branches and tying
nets to form a bag. Arthropods were collected in February
and July 2002. Branches were covered with 60 ? 90-mm
plastic bags and cut. Cotton balls saturated with ethyl
acetate were placed inside to kill all arthropods. The
arthropods were then collected, identi?ed to order (or
family) and measured in length (mm). Foliage was dried
and weighed and used to standardize all arthropod data
388 JULIE A. JEDLICKA ET AL.
as number of individuals g?1 of dry foliage. To compare
with coffee arthropod data, all arthropod data from Inga
trees were converted to number of individuals g?1 of wet
foliage with the following empirically derived, signi?cant
regression: wet leaf weight (g) = 2.29 ? dry leaf weight
(g) ?18.76 (R2 = 0.861, P < 0.001).
Analysis
Kolmogorov?Smirnov tests were performed on arthropod
abundances. Some data were not distributed normally
and were resistant to transformations so more conserva-
tive, non-parametric tests were used. Because arthropods
in the understorey and canopy were sampled differently,
we did not attempt a direct comparison between
densities in the two vegetative layers. Instead we limit
arthropod comparisons to (1) effect of season within
each vegetative layer and (2) effect of bird exclosure.
Seasonal differences in arthropod abundances (both total
number of arthropods and the number of large arthropods
>5 mm in length) were tested using Mann-Whitney
U-tests. Differences in exclosure and control arthropod
abundances in the canopy and understorey were tested
with paired Wilcoxon tests. RCWA foraging heights (total
observations and male-female comparisons) were tested
using Mann-Whitney U-tests. Chi-square contingency
tests were used to compare all categorical variables:
seasonal foraging layer, proportion of successful foraging
attempts (by layer and season), and type of successful
foraging manoeuvres. Bonferroni calculations were
performed with a Texas Instruments TI-85 calculator and
all other statistical tests were performed using SPSS 11.0
for Macintosh.
RESULTS
Seasonal foraging location
In the wet season, RCWAs were observed foraging equally
as often in the canopy (47%, n = 34) and understorey
(53%, n = 39), but in the dry season they foraged
signi?cantly more in the coffee understorey (81% of
observations;?2 = 12.1, df = 1, P = 0.0005, n = 72). The
average foraging height of RCWAs signi?cantly decreased
from 4.6 m (SE = 0.5 m; median = 2 m) in the wet season
to 2.0 m (SE = 0.2 m; median = 1.5 m) in the dry
season (Mann-Whitney U = 1753, P = 0.0005, Figure 2).
The average foraging height of male birds (sexed either
by territorial singing or presence of a large cloacal
protuberance while banding; mean = 4.0 m, SE = 0.5 m,
n = 52) did not signi?cantly differ from the average of
female birds (sexed by the presence of a brood patch while
banding; mean = 3.7 m, SE = 0.9 m, n = 22, U = 485,
P = 0.30). Seasonal comparisons between male and
Figure 2. Histograms of rufous-capped warbler seasonal foraging
heights. Each count represents separate foraging observations. With
the exception of one observation in the wet season and two in the dry
season, all foraging heights in the 0?1-m range were above the ground
in low branches.
female average foraging heights were also not signi?cant
(dry season, male mean = 2.52 m, SE = 0.39 m, n = 28,
female mean = 2.83 m, SE = 1.56 m, n = 4, U = 51,
P = 0.81; wet season, male mean = 5.6 m, SE = 0.9 m,
n = 24, female mean = 3.8 m, SE = 1.1 m, n = 18,
U = 156, P = 0.13). Using the Bonferroni method to cor-
rect to the 0.05 level, the critical P-value for the above
data on foraging height would be 0.017. For the 0.01
level, P would equal 0.003.
Successful foraging manoeuvres
When the seasons are pooled and success is de?ned as
manoeuvres that were either clearly successful (CS) or
successful not seen (SNS), RCWAs were proportionally
more successful foraging in the canopy (44%, n = 48)
than the understorey (27%, n = 97; ?2 = 4.21, df = 1,
P = 0.04). Clearly successful manoeuvres were observed
more in the canopy during the wet season than any other
combination of season and vegetative layer, representing
38% of all documented foraging attempts (?2 = 4.34,
df = 1, P = 0.04, Figure 3). Additionally there was a
higher proportion of successful manoeuvres (SNS and CS)
in the wet season than in the dry season (Figure 3), but the
differences were not statistically signi?cant to the 0.05
level (?2 = 2.09, df = 1, P = 0.15). Finally, successful
30 - 
25 
20 
2 15 
Wet season 
I    1     I 
1      2      3      4      5      6      7      8      9     10    11    12    13    14    15    16 
30 
25 
20 - 
15 1 
10 
Dry season 
1      2      3      4      5      6      7      8      9     10    11    12    13    14    15    16 
Foraging height (m) 
Seasonal shift in the foraging niche of a tropical avian resident 389
Figure 3. Per cent of successful manoeuvres out of total rufous-capped
warbler foraging attempts in vegetative layers separated by season. The
striped area below represents per cent of clearly successful manoeuvres
and darker area above represents ?successful not seen? manoeuvres
(manoeuvres followed by a bill wipe).
Figure 4. Foraging manoeuvre used to successfully capture prey in
understorey and canopy. If the rufous-capped warbler was foraging
while standing upright, the manoeuvre was classi?ed as standing. If
the bird foraged by either leaping or ?ying a short distance (striking) the
manoeuvre was classi?ed as aerial. If the bird was hanging on vegetation
with the head positioned laterally or downwards, the manoeuver was
classi?ed as hanging.
foraging events were more often performed with standing
manoeuvres in the understorey and aerial manoeuvres
in the canopy (?2 = 20.7, df = 2, P < 0.0001, Figure 4).
Using the Bonferroni method to correct to the 0.05 level,
the critical P-value for the above data on successful
foraging manoeuvres would be 0.012. For the 0.01 level,
P would equal 0.002.
Arthropod abundance
Effects of season in theunderstoreyandcanopy. Total number
of arthropods g?1 of leaf biomass in the coffee understorey
ranged from 0.52 in the wet season to 0.43 in the dry
season (Mann?Whitney U = 4375, P = 0.13, Figure 5).
In the canopy, total number of arthropods varied from
0.30 g?1 in the wet season to 0.34 g?1 in the dry season
(Mann?Whitney U = 162, P = 0.60).
Abundance of arthropods >5 mm (large arthropods)
did not signi?cantly change seasonally in the understorey
(0.023 g?1 in the dry season and 0.018 g?1 in the wet
season, Mann?Whitney U = 4325, P = 0.08, Figure 5),
however in the Inga canopy there were signi?cantly fewer
arthropods in the dry season (0.021 g?1) than the wet
season (0.049 g?1, Mann?Whitney U = 64, P = 0.001).
Effect of bird exclosure in the understorey and canopy. In the
canopy, total arthropod abundance was higher without
bird predation (exclosure treatments) in both the wet
season (Wilcoxon Z =?3.55, P = 0.0004, Figure 5)
and dry season (Wilcoxon Z = ?2.37, P = 0.02). In the
understorey, arthropod abundance was not reduced by
bird predation in the dry season (Wilcoxon Z = ?1.06,
P = 0.29), but in the wet season arthropod abundance
was slightly lower without bird predation (Wilcoxon
Z = ?2.08, P = 0.038).
Large arthropod abundance was signi?cantly higher
on canopy trees without bird predation during the wet
season (Wilcoxon Z = ?3.44, P = 0.0006, Figure 5) and
dry season (Wilcoxon Z = ?3.72, P = 0.0002), but there
was no exclosure-control difference in the understorey.
Using the Bonferroni method to correct to the 0.05 level,
the critical P-value for the above data on arthropod
abundance would be 0.025. For the 0.01 level, P would
equal 0.005.
DISCUSSION
RCWA seasonal niche shift
RCWAs clearly exhibited a seasonal foraging niche shift.
In the wet season, RCWAs were observed foraging equally
as often in the coffee understorey and canopy, but in the
dry season they were observed foraging in the understorey
80% of the time. Similarly, Dietsch (2003) observed a
signi?cant decrease in the dry season foraging height of
RCWAs during the 2 y prior to this study on the same
farm. He also observed that during this time over 60%
of foraging manoeuvres for migrant species were in the
canopy and that most migrants were insectivorous.
The RCWA seasonal niche shift occurred despite the
fact that total arthropod abundance in both vegetative
layers did not change seasonally either on control
|"0?M, 
SO 
40 
a* 
I" 
0 
=34.2= ^3 
Undorilorey    |        Canopy Undmntormy    | Canopy 
Standing Aerial Hanging 
390 JULIE A. JEDLICKA ET AL.
Wlth  Birds
0.9
0.8
N
um
be
r o
f a
rth
ro
po
ds
 p
er
 u
ni
t
fo
lia
ge
 b
io
m
as
s 
(in
div
. g
-
1 ) 0.7
0.6
0.5
0.4
0.3
0.2
0.1
U
Wet season Wet season
Total arthropods Arthropods over 5 mm
Dry season Dry season
C U C U
a
b
C U C
0
No  Birds
Figure 5. Number of arthropods g?1 of leaf biomass found seasonally in each vegetative layer (U = Understorey and C = Canopy). Total number
of arthropods is represented by the left set of bars, and arthropods over 5 mm by the right set. Letters indicate signi?cant seasonal differences in
background arthropod abundance (control plants) and asterisks indicate statistical differences in arthropod abundance within season between
paired control (with birds) and exclosure (no birds) plants due to bird predation (? = 0.05).
branches or in experimental exclosures without foraging
birds. However in the canopy, availability of large
arthropods on control branches signi?cantly decreased
by 58% from wet season to dry season. Previous research
in the same study location analysed arthropod data
collected from stomach contents of all insectivorous birds
sampled. The average length of consumed arthropods
was estimated to be 3.24 mm (Greenberg et al. unpubl.
data). Large arthropods (>5 mm) compose a very small
proportion of the entire arthropod community, so a mean
length of 3.24 mm may suggest that insectivorous birds
are preferentially foraging on larger arthropods.
RCWAs were more successful when foraging in the
canopy, perhaps due to the greater quantities of large
arthropods found there. It is unlikely that this difference
represents observer bias towards recording conspicuous
manoeuvres in the canopy because RCWAs were most
commonly observed foraging in the understorey. RCWAs
primarily used aerial manoeuvres in canopy and standing
manoeuvres in coffee for successful attacks further
supporting the hypothesis that RCWAs were foraging
on larger arthropods in the canopy. Larger prey are
more likely to be targeted by aerial rather than standing
manoeuvres simply because larger prey are more visible
from farther away. Furthermore, birds are likely to incur
a higher foraging cost (by using aerial manoeuvres) to
obtain larger prey. Consequently, RCWAs were probably
foraging on larger, more energetically favourable prey in
the canopy.
Resource competition hypotheses
The resource competition hypothesis assumes that bird
foraging behaviour will change as a result of lower
resource availability. Some studies investigating the effect
of ants on bird foraging have found evidence for this
hypothesis. In separate studies Haemig (1992, 1994)
found that number and duration of visits by insectivores
and foliage-gleaners (but not by seed-eaters) to trees
without ants was longer, presumably because there were
fewer resources on trees with ants. Philpott et al. (2005)
investigating the effects of aggressive ants on bird foraging
in the same study site discussed here found that length
of visits made by insectivorous birds, but not by other
feeding guilds or overall, were reduced on trees with high
ant densities, again suggesting that foraging was affected
by a lower availability of shared resources, not because of
direct interference or aggression which would be expected
to affect all bird guilds equally.
In the context of the study presented here, the resource
competition hypothesis would mean that because
migrants lower arthropod abundance in the canopy,
RCWAs forage elsewhere due to resource depletion. Large
arthropods showed a four-fold decrease in abundance in
the canopy due to avian predation when migrant birds
were present. Moreover, abundance of large arthropods
on control canopy branches signi?cantly decreased by
58% from the wet to dry season. The observed foraging
success rate of RCWAs in the canopy dropped from being
Seasonal shift in the foraging niche of a tropical avian resident 391
clearly successful 38% of the time in the wet season to
only 7% in the dry season when migrants are present. It
is likely that this is due to migrants preferentially foraging
on large arthropods in the canopy and lowering their
abundance and availability. This evidence supports the
resource competition hypothesis.
Seasonal shifts in foraging niche and resource competition
in other studies
RCWAs are not the only generalist avian residents
found to exhibit a seasonal foraging niche shift. What
is particularly striking is that each of the following
empirical studies found that certain residents foraged
at lower heights upon the arrival of migrants. For
example, in the Colombian Andes, Chipley (1976)
documented that three resident species (slate-throated
redstart, Myioborus miniatus; brown-capped vireo, Vireo
leucophrys; tropical parula, Parula pitiayumi) foraged
at lower heights when migrant birds were present in
dry season. Migrant competitors generally preferred the
upper, more favourable vegetative strata where 90%
of all clearly successful manoeuvres were observed.
Chipley concluded that the shift to forage more in
the canopy by slate-throated redstarts and tropical
parulas in April/May was probably related to the
corresponding migrant departure. Likewise, in August
and early September, when there were no migrants
present, both species had ?nished breeding and used the
understorey the least. When migrants arrived in late
September and October both species abruptly shifted back
to using the understorey signi?cantly more, supporting
the resource competition hypothesis. Chipley ?nally
concluded that migrant arrival probably had little to no
effect on the majority of residents except for some small
insectivorous species where competition with migrants
may be important.
Waide (1981) studied a dry tropical forest in Campeche,
Mexico and found that seasonal changes in the foraging
behaviour of permanent resident species occurred. Six
resident bird species foraged at lower heights in the
dry season: tropical gnatcatcher (Polioptila plumbea),
grey-throated chat (Granatellus sallaei), white-bellied
wren (Uropsila leucogastra), tropical kingbird (Tyrannus
melancholichus), boat-billed ?ycatcher (Megarynchus
pitangua), social ?ycatcher (Myiozetetes similis). However,
by calculating the niche overlap between resident and
migrant species, only the grey-throated chat signi?cantly
reduced its niche overlap with migrants by foraging
at lower heights, foraging lower in the canopy crown,
and changing foraging manoeuvres when migrants were
present. Although he did not publish the data on prey
abundance in this study, he noted that arthropods were
scarcer in a drier wet season when compared with the
wetter following year. He concluded that there was
not much evidence for resource competition between
residents and migrants and suggested that perhaps
competition is only noticeable later in the dry season
(March?April) or in years with food shortages.
At Lake Naivasha, Kenya, Rab?l (1987) noted how
the resident species black-breasted apalis (Apalis ?avida)
and grey-backed camaroptera (Camaroptera brevicaudata)
foraged lower in less-preferred heights of the vegetation
and that the black-breasted apalis clearly diverged in
foraging behaviour from the migrant willow warbler
(Phylloscopus trochilus) upon its arrival. Upon the arrival of
willow warblers, Rab?l also reported decreased arthropod
abundance, a diet switch by the black-breasted apalis,
increased aggression among guild members, and a change
in migrant foraging manoeuvres. Consequently Rab?l
considered resource competition a likely explanation for
the shifts in foraging niche.
In a follow-up study, however, Rab?l (1993) recon-
sidered resource competition as the primary explanation,
because changes in brood rearing corresponded with
the arrival of migrants and potentially explained the
foraging shift (see below). His later study did not show
as great a niche shift as his earlier research and arthropod
abundances did not appear limiting. It is important to
acknowledge that relative densities of willow warblers in
two of the three habitats studied were 41% and 52% of
their previous densities in the 1987 study, partly due
to their late arrival. Even with the fewer competitors
when compared with the resident black-breasted apalis
in the woodlands, the migrant willow warbler was found
signi?cantly less in small trees and bushes, higher up in
trees and their foraging height was much higher.
Analysis of alternative or additional hypotheses
Climatic and seasonal effects. As noted by Waide (1981)
and Rab?l (1993), year to year variation in arthropod
abundance may heighten competitive effects. In Chiapas,
extended dry seasons occur periodically in association
with El Nin?o that may exacerbate the effects observed in
this study. Though the current study was not conducted
during an El Nin?o event, Dietsch (2003) observed lower
abundances of residence birds following a particularly
harsh El Nin?o in 1997?8. Drought-induced resource
limitations have been shown to induce rapid evolutionary
response in birds (Grant & Grant 2002, Price et al. 1984).
Consequently, climate-induced inter-annual variation in
arthropod abundance may produce critical bottlenecks
for RCWAs with important evolutionary implications.
In fact, although RCWAs are considered generalist
insectivores, the species may be specialized to forage in
the understorey during the dry season. A more detailed
study of RCWA diet and foraging behaviour may reveal
392 JULIE A. JEDLICKA ET AL.
additional niche characteristics that indicate RCWAs
exploit some resources more ef?ciently than migratory
warblers.
It is possible that avian foraging shifts correspond
to seasonal changes in rain or wind patterns or
phenological changes in the environment. At the end
of the dry season, Inga produces new leaves followed
shortly by nectar-rich ?owers which attract high densities
of arthropods, especially Homoptera (Greenberg et al.
1997b, Johnson 2000, Koptur 1994). It is possible that
larger arthropods respond strongly to these phenological
changes while smaller arthropods may not, resulting
in a different seasonal trend between the two controls.
Increased densities of large arthropods in the canopy
could in?uence avian insectivory. However, any climatic
variation cannot account for the differences in exclosure
and control arthropod densities demonstrating that
avian insectivory strongly affects arthropod abundance.
Additionally, Chipley (1976) noted that shifts in foraging
niches of resident avian species corresponded with
migrant presence during periods when the weather did
not change.
Breeding demand hypothesis. In northern Latin America,
most insectivorous residents breed primarily during the
wet season, in the absence of migrants (Skutch 1950).
Consequently, seasonal shifts in foraging niches may be
partly a result of constraints imposed by reproductive
activities (Chipley 1976, Leisler 1992, Lo?vei 1989, Rab?l
1987, 1993; Waide 1981). Adult birds may prefer
different food sources than those they feed to their young,
so they may forage in different locations depending on
breeding status.
Chipley (1976) hypothesized that the shift to lower
foraging heights in April/May for the slate-throated
redstarts was related to changes in breeding demands
because three slate-throated redstart nests were found
in late May. Chipley also suggested that the shifts in
foraging niche of the brown-capped vireo resulted from
both breeding demands and migrant presence. Rab?l
(1987) speculated as to whether a change in breeding
demands contributed to the foraging niche shift for the
grey-backed camaroptera and Waide (1981) noted that
there was an abrupt decline in the number of resident
species breeding upon migrant arrival.
The breeding demand hypothesis is compatible with
the resource competition hypothesis. Although indirect
competition may limit the foraging niche for RCWAs
during the dry season, the absence of migrants may allow
arthropods in the canopy to grow larger or to persist in
the canopy during the breeding season. These arthropod
resources would then be available if RCWAs need to
expand their foraging niche to search for the necessary
resources for feeding young. A more complete chronology
of foraging behaviour of the resident species through the
breeding season, in particular during the periods when
migrants arrive and depart, is needed to better document
how readily RCWAs forage in the canopy. Additionally
one should know the approximate date when adults stop
feeding young. Although attempts were made in this
study to document when ?edglings were fed by adults,
these data were too sparse to address the breeding demand
hypothesis.
Regardless, in the case of the RCWA, it is unclear
what the expected seasonal foraging pattern would be
if breeding demands were a signi?cant determining factor
in their foraging location. During the breeding season
RCWAs nest on the ground (pers. obs., Stiles & Skutch
1989), and consequently may forage more in the coffee
understorey as a way to remain closer to the nest and
young. However, this may depend on how sub-optimal
the coffee understorey is for arthropod abundance.
Coffee plants contain alkaloid compounds that discourage
herbivory. This is one possible explanation for the minimal
impact of birds observed in the coffee understorey.
Additionally, protection from avian predation within the
exclosure may result in an osmotic effect of arthropods
dispersing from exclosures to the point that control
abundances are equal or even slightly greater than
exclosure abundances. The canopy is very resource-rich
with high densities of large arthropods and one may
predict that RCWAs should forage there to increase their
foraging effectiveness. Obviously future research in this
area is warranted.
Direct interferencehypothesis.This hypothesis assumes that
migrants force residents to forage at lower, unfavourable
levels due to agonistic interactions (Chipley 1976, Rab?l
1993). No aggressive behaviours between migrants and
RCWAs were witnessed. Migrants were not observed
to defend territories; rather, most were foraging in the
canopy in large mixed species ?ocks. RCWAs did not
join these ?ocks but foraged either alone, with one other
RCWA, or in small intraspeci?c ?familial? groups (with
their hatch years) in the coffee understorey during the
dry season. Although it is possible that due to timing
such agonistic interactions were missed, this study does
not support the direct interference competition as a
cause of the niche shift. Additionally, Chipley (1976)
observed that agonistic encounters between residents
and migrants were extremely uncommon (85% of the
agonistic behaviours between individuals during the dry
season were intraspeci?c interactions).
Sexual segregation hypothesis. Males of territorial avian
species can often be found singing on high perches
during the breeding season, claiming their territory.
After singing, these males may stay and forage in
Seasonal shift in the foraging niche of a tropical avian resident 393
the higher vegetative strata (Waide 1981). When the
breeding season is over and singing is softer and less
constant, males may return to forage in lower vegetative
strata. Consequently the territorial behaviour of singing
males may be misinterpreted as a foraging niche shift
of the species. Because Waide (1981) could not sex the
individual birds he was observing, he could not test this
hypothesis. In this study the differences in the wet season
between average foraging height of males and females did
not differ signi?cantly although perhaps a larger sample
size may be required to test this hypothesis.
Predator avoidance hypothesis. In order to avoid aerial
predators, susceptible birds may concentrate their time
foraging lower in the vegetative strata. During the dry
season, the proportion of avian predators to potential
prey is lower than during the wet season because
passerine birds comprise the large majority of all migrants.
Consequently foraging lower in the vegetative strata
during the dry season is unlikely a result of increased
predatory hawk and falcon abundance. However, in
relation to other predators such as snakes it is unclear
where RCWAs would be expected to forage to reduce risk
of predation. Chipley (1976) ruled out this hypothesis
because of the very low predator abundance during
his study. More information would have to be known
about the seasonal ?uctuations of predator populations
to address this hypothesis.
CONCLUSIONS
The results of this study support the claim of Chipley
(1976) that migrants may be a signi?cant competing force
with small, generalist, insectivorous residents. Seasonal
niche shifts, such as lower foraging heights, could be
a relatively widespread mechanism for resident birds to
reduce competition with migrant species, a subject that
warrants further research. However, this interaction is
rarely considered in migrant-resident studies, generally
because these studies only include the dry season.
Further studies should attempt to monitor any behaviour
changes in residential species upon migrant arrival and
consider the predatory effects this shift may have on
the abundance of organisms at lower trophic levels.
If resource competition with migrants is continually
supported, this study among others will support the
hypothesis that predator?prey populations are coupled.
Perhaps the best explanation for seasonal shifts in
foraging niches of residents is the combined effects
of indirect resource competition, the breeding demand
hypothesis, and other alternative hypotheses acting in
unison. If seasonal displacement is widespread in various
tropical communities, these causal agents likely vary in
their contribution on a case-by-case basis and may be
strong seasonal forces in?uencing community structure.
ACKNOWLEDGEMENTS
We thank J. Vandermeer, B. Rathcke and three anony-
mous reviewers for advice and comments on the manu-
script. P. Raimondi provided statistical assistance. We
thank P. Bichier for providing ?eld training in how to con-
duct avian foraging observations. We thank B. E. Chilel
and P. Bichier for help banding birds and W. Peters, owner
of Finca Irlanda, for gracious hospitality. J. A. Garcia-
Ballinas, P. Bichier, G. Lopez, J. Maldonado, B. E. Chilel,
R. Velasquez, J. C. Mendez Lopez, A. Mendez Mendizabal,
L. H. L. Ramirez, A. Gonzalez, G. Dominguez, H. Yard,
J. Vandermeer, G. Ibarra- Nu?n?ez and S. Uno participated
in arthropod sampling and biomass collection. Funding
was provided by the Smithsonian Institution, Horace
Rackham Graduate School of the University of Michigan
and a National Science Foundation grant DEB-9981526
awarded to I. Perfecto, R. Greenberg and G. Ibarra-
Nun?ez.
LITERATURE CITED
CALVO, L. & BLAKE, J. 1998. Bird diversity and abundance on two
different shade coffee plantations in Guatemala. Bird Conservation
International 8:297?308.
CHIPLEY, R. M. 1976. The impact of wintering migrant wood warblers
on resident insectivorous passerines in a subtropical Colombian oak
woods. Living Bird 15:119?141.
DIETSCH, T. V. 2003. Conservation and ecology of birds in coffee
agroecosystems of Chiapas, Mexico. Ph.D. Thesis. University of
Michigan, Ann Arbor.
GRANT, P. R. & GRANT, B. R. 2002. Unpredictable evolution in a
30-year study of Darwin?s Finches. Science 296:707?711.
GREENBERG, R. 1986. Competition in migrant birds in the nonbreeding
season. Pp. 281?307 in Johnston, R. F. (ed.). Current ornithology.
Plenum Press, New York.
GREENBERG, R. 1995. Insectivorous migratory birds in tropical
ecosystems: the breeding currency hypothesis. Journal of Avian
Biology 26:260?264.
GREENBERG, R., SALGADO-ORITZ, J. & MACIAS-CABALLERO, C.
1994. Aggressive competition for critical resources among migratory
birds in the Neotropics. Bird Conservation International 4:114?127.
GREENBERG, R., BICHIER, P., ANGON, A. C. & REITSMA, R. 1997a.
Bird populations in shade and sun coffee plantations in Central
Guatemala. Conservation Biology 11:448?459.
GREENBERG, R., BICHIER, P. & STERLING, J. 1997b. Bird populations
in rustic and planted shade coffee plantations of Eastern Chiapas,
Mexico. Biotropica 29:501?514.
394 JULIE A. JEDLICKA ET AL.
GREENBERG, R., PRAVOSUDOV, V., STERLING, J., KOZLENKO, A. &
KONTORSCHIKOV, V. 1999. Divergence in foraging behavior of
foliage-gleaning birds of Canadian and Russian boreal forests.
Oecologia 120:451?462.
HAEMIG, P. D. 1992. Competition between ants and birds in a Swedish
forest.Oikos 65:479?483.
HAEMIG, P. D. 1994. Effects of ants on the foraging of birds in spruce
trees.Oecologia 97:35?40.
HERRERA, C. M. 1978. Ecological correlates of residence and non-
residence in a Mediterranean passerine bird community. Journal of
Animal Ecology 47:871?890.
JOHNSON, M. D. 2000. Effects of shade-tree species and crop structure
on the winter arthropod and bird communities in a Jamaican shade
coffee plantation. Biotropica 32:133?145.
JOHNSON, M. D. & SHERRY, T. W. 2001. Effects of food availability on
the distribution of migratory warblers among habitats in Jamaica.
Journal of Animal Ecology 70:546?560.
JOHNSON, M. D., SHERRY, T. W., STRONG, A. M. & MEDORI, A. 2005.
Migrants in Neotropical bird communities: an assessment of the
breeding currency hypothesis. Journal of Animal Ecology 74:333?
341.
KOPTUR, S. 1994. Floral and extra?oral nectars of Costa-Rican Inga
trees ? a comparison of their constituents and composition. Biotropica
26:276?284.
LACK, P. C. 1986. Ecological correlates of migrants and residents in a
tropical African savanna. Ardea 74:111?119.
LEISLER, B. 1990. Selection and use of habitat of wintering migrants.
Pp. 156?174 in Gwinner, E. (ed.). Bird migration: the physiology and
ecophysiology. Springer-Verlag, Berlin.
LEISLER, B. 1992. Habitat selection and coexistence of migrants and
Afrotropical residents. Ibis 134:77?82.
LO?VEI, G. L. 1989. Passerine migration between the Palaearctic and
Africa. Pp. 143?174 in Johnston, R. F. (ed.). Current ornithology.
Plenum Press, New York.
MAS, A. H. & DIETSCH, T. V. 2003. An index of management intensity
for coffee agroecosystems to evaluate butter?y species richness.
Ecological Applications 13:1491?1501.
MOGUEL, P. & TOLEDO, V. M. 1999. Biodiversity conservation in
traditional coffee systems of Mexico. Conservation Biology 13:11?
21.
O?CONNOR, R. J. 1981. Comparisons between migrant and non-migrant
birds in Britain. Pp. 167?195 in Aidley, D. J. (ed.). Animal migration.
Cambridge University Press, Cambridge.
PERFECTO, I. & VANDERMEER, J. 2002. Quality of agroecological
matrix in a tropical montane landscape: ants in coffee plantations
in Southern Mexico. Conservation Biology 16:174?182.
PERFECTO, I., MAS, A., DIETSCH, T. & VANDERMEER, J. 2003.
Conservation of biodiversity in coffee agroecosystems: a tri-taxa
comparison in southern Mexico. Biodiversity and Conservation
12:1239?1252.
PERFECTO, I., VANDERMEER, J. H., BAUTISTA, G. L., NUNEZ, G. I.,
GREENBERG, R., BICHIER, P. & LANGRIDGE, S. 2004. Greater
predation in shaded coffee farms: the role of resident neotropical
birds. Ecology 85:2677?2681.
PETIT, D. R., LYNCH, J. F., HUTTO, R. L., BLAKE, J. G. & WAIDE,
R. B. 1993. Management and conservation of migratory landbirds
overwintering in the Neotropics. Pp. 70?90 in Finch, D. M. & Stangel,
P. W. (eds.). Status and management of Neotropical migratory birds.
U.S. Forest Service General Technical Report R.M. 229, Fort Collins,
Colorado.
PHILPOTT, S. M., GREENBERG, R. & BICHIER, P. 2005. The in?uence
of ants on the foraging behavior of birds in an agroforest. Biotropica
37:468?471.
PHILPOTT, S. M., GREENBERG, R., BICHIER, P. & PERFECTO, I.
2004. Impacts of major predators on tropical agroforest arthropods:
comparisons within and across taxa. Oecologia 140:140?149.
POST, W. 1978. Social and foraging behavior of warblers wintering in
Puerto Rico scrub. The Wilson Bulletin 90:197?214.
PRICE, T. D., GRANT, P. R., GIBBS, H. L. & BOAG, P. T. 1984. Recurrent
patterns of natural selection in a population of Darwin?s ?nches.
Nature 309:787?789.
RAB?L, J. 1987. Coexistence and competition between overwintering
Willow Warblers Phylloscopus trochilus and local warblers at Lake
Naivasha, Kenya. Ornis Scandinavica 18:101?121.
RAB?L, J. 1993. Competition between overwintering Willow Wablers
Phylloscopus trochilus and local warblers in the acacia-savanna in
Kenya. Pp. 76?96 in Jadsen, M. (ed.). Proceedings of the Seventh
Nordic Congress of Ornithology. Nordic Congress of Ornithology, Skive,
Denmark.
RAPPOLE, J. & WARNER, D. W. 1980. Ecological aspects of migrant bird
behavior in Veracruz, Mexico. Pp. 353?394 in Keast, A. E. & Morton,
E. S. (eds.). Migrantbirds in theneotropics: ecology, behavior, distribution,
and conservation. Smithsonian Institution Press, Washington, DC.
REMSEN, J. V. & PARKER, T. A. I. 1983. Contributions of river created
habitats to bird species richness in Amazonia. Biotropica 15:223?
231.
REMSEN, J. V. & ROBINSON, S. K. 1990. A classi?cation scheme for
foraging behavior of birds in terrestrial habitats. Studies in Avian
Biology 13:144?160.
RICHTER, M. 2000. The ecological crisis in Chiapas: a case study from
Central America. Mountain Research and Development 20:332?339.
RICKLEFS, R. E. 1989. The megapopulation: a model of demographic
coupling between migrant and resident landbird populations.
Pp. 537?548 in Hagan, J. M. & Johnston, D. W. (eds.). Ecology and
conservation of neotropical migrant landbirds. Smithsonian Institution
Press, Washington, DC.
ROBBINS, C. S., DOWELL, B. A., DAWSON, D. K., COLON, J. S.,
ESTRADA, R., SUTTON, A., SUTTON, R. & WEYER, D. 1992.
Comparison of Neotropical migrant landbird populations wintering
in tropical forest, isolated forest fragments, and agricultural habitats.
Pp. 207?220 in Hagan, J. M. & Johnston, D. W. (eds.). Ecology and
conservation of neotropical migrant landbirds. Smithsonian Institution
Press, Washington, DC.
SKUTCH, A. 1950. The nesting period of Central American
birds in relation to climate and food supply. Ibis 120:480?
498.
STILES, G. F. & SKUTCH, A. F. 1989. A guide to the birds of Costa Rica.
Cornell University Press, Ithaca. 511 pp.
Seasonal shift in the foraging niche of a tropical avian resident 395
STOTZ, D. F., FITZPATRICK, J. W., PARKER, T. E. & MOSKOVITS, D. K.
1996. Neotropical birds: ecologyand conservation. University of Chicago
Press, Chicago. 478 pp.
STRONG, A. M. 2000. Divergent foraging strategies of two neotropical
migrant warblers: implications for winter habitat use. The Auk
117:381?392.
VANNINI, J. P. 1994. Nearctic avian migrants in coffee plantations
and forest fragments of south-western Guatemala. Bird Conservation
International 4:209?232.
WAIDE, R. B. 1981. Interactions between resident and migrant birds in
Campeche, Mexico. Tropical Ecology 22:134?154.
WILLIS, E. O. 1966. The role of migrant birds at swarms of army ants.
Living Bird 5:187?231.
WUNDERLE, J. 1999. Avian distribution in Dominican shade
coffee plantations: area and habitat relationships. Journal of Field
Ornithology 70:58?70.
WUNDERLE, J. & LATTA, S. 1996. Avian abundance in sun and shade
coffee plantations and remnant pine forest in the Cordillera Central.
Ornitologia Neotropical 7:19?34.
WUNDERLE, J. & WAIDE, R. B. 1993. Distribution of overwintering
Nearctic migrants in the Bahamas and Greater Antilles. Condor
95:904?933.