See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/232677803

Ground Versus Canopy Methods for the Study of Birds in Tropical Forest Canopies:
Implications for Ecology and Conservation

Article in The Condor · May 2009

DOI: 10.1525/cond.2009.090032

CITATIONS READS
26 681

1 author:

David L Anderson
The Peregrine Fund
61 PUBLICATIONS 571 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Breeding Ecology of Arctic Raptors Under Global Climate Change View project

Harpy Eagle Distribution Ecology View project

All content following this page was uploaded by David L Anderson on 12 December 2016.

The user has requested enhancement of the downloaded file.

The Condor 111(2):226–237
¡ The Cooper Ornithological Society 2009

GROUND VERSUS CANOPY METHODS FOR THE STUDY OF BIRDS IN TROPICAL

FOREST CANOPIES: IMPLICATIONS FOR ECOLOGY AND CONSERVATION
DAVID L. A NDERSON1
Department of Biological Sciences & Museum of Natural Science, Louisiana State University, Baton Rouge, LA 70803

Abstract. Birds of the forest canopy are important components of tropical forest ecosystems, but difficulty of
access or viewing into the canopy complicates their study. If ground methods are biased against canopy birds, as
has been suggested, this bias could affect our understanding of forest ecology as well as biological monitoring and
conservation practices. This study is the first to quantitatively compare results from ground- and canopy-based
methods of censusing canopy birds. I used three methods to assess differences in ground-based and canopy-based
methods for detecting forest birds in a 100-ha plot of lowland forest in northern Honduras: (1) point counts from
the ground, (2) 22 repeat censuses from two canopy trees, and (3) single censuses from 22 canopy trees. I counted
birds for a full annual cycle from April 2006 to April 2007 and recorded 157 species in over 4000 individual de-
tections. Ground methods significantly underestimated species and familial richness as well as abundance of indi-
viduals in the canopy stratum. On the basis of these results, I predict that the use of ground methods alone misses
25 to 50% of the species richness for some migrant and resident families and underestimates the density of some
species by as much as 25%. These findings highlight the risk of relying on ground-based methods for bird studies
in structurally complex tropical forests; reliance on ground-based methods may negatively affect long-term bio-
logical monitoring and the setting of conservation priorities for tropical forests.

Key words: biological monitoring, birds, community composition, forest canopy, Honduras, rain forest.

Muestreos desde el Suelo y en el Dosel para el Estudio de Aves en el Dosel de Bosques Tropicales:
Implicaciones para el Monitoreo Biológico y la Conservación
Resumen. Las aves de dosel son componentes importantes de los ecosistemas de bosque en el Neotrópico,
pero su estudio ha sido complicado por la dificultad de acceso al dosel o por la dificultad de observarlas desde
el suelo. Si los muestreos desde el suelo son sesgados en contra de la detección de aves del dosel, como se ha
sospechado, este sesgo podría afectar nuestra habilidad de entender la ecología de bosques como también el moni-
toreo biológico y las prácticas de conservación. El presente estudio es el primero que compara de manera cuantita-
tiva los resultados de métodos de censo desde el suelo y desde el dosel para aves de dosel. En una parcela de 100-ha
en el norte de Honduras usé tres métodos de censado para evaluar las diferencias entre los censos desde el suelo y
desde el dosel: (1) conteos en puntos en el suelo, (2) 22 conteos en puntos repetidos desde dos árboles emergentes,
y (3) 22 conteos en puntos no repetidos desde 22 árboles diferentes. Realicé los muestreos de aves durante un ciclo
anual completo desde abril del 2006 hasta abril del 2007, periodo durante el cual detecté 157 especies en más de
4000 observaciones. Los métodos de conteo desde el suelo subestimaron la riqueza de especies y familias como
también las abundancias en el estrato del dosel. Con base en estos resultados, predigo que entre el 25 y el 50% de
la riqueza de especies de algunas familias migratorias y residentes no hubiese sido registrada si el estudio hubiese
estado basado sólo en muestreos desde el suelo, y que la densidad poblacional de ciertas especies pudiese haber
sido subestimada hasta en un 25%. Estos resultados demuestran el riesgo de basarse sólo en métodos de censo re-
alizados desde el suelo en los estudios sobre aves en bosques tropicales estructuralmente complejos. Además, los
métodos de muestreo de aves de dosel basados sólo en conteos desde el suelo no son adecuados para el monitoreo a
largo plazo de las tendencias poblacionales, y pueden afectar la determinación de las prioridades de acciones para
la conservación de bosques tropicales.

INTRODUCTION
Ecosystem Assessment 2005). This understanding, in turn,
As the rapid pace of tropical deforestation continues, im- is directly tied to the quality and variety of methods used to
proving our understanding of the processes that create and observe and study both processes and diversity. In tall and
maintain forest biodiversity is increasingly essential for the structurally complex tropical forests, a complete understand-
continued preservation of the forests that remain (Millennium ing of forest ecosystems must address diversity and ecological

Manuscript received 19 February 2009; accepted 3 April 2009.

1
E-mail: [email protected]

The Condor, Vol. 111, Number 2, pages 226–237. ISSN 0010-5422, electronic ISSN 1938-5422. ‘ 2009 by The Cooper Ornithological Society. All rights reserved. Please direct
all requests for permission to photocopy or reproduce article content through the University of California Press’s Rights and Permissions website, http://www.ucpressjournals.com/
reprintInfo.asp. DOI: 10.1525/cond.2009.090032

226
 METHODS FOR CANOPY BIRDS IN TROPICAL FORESTS 227

interactions at all levels of the forest (Lowman and Rinker major implications for the understanding of forest bird com-
2004). Development of field methods for the study of for- munities, ecosystem processes, and the conservation of both.
est canopies, however, has been hindered by the difficulty of The major goal of this study was to address the basic
gaining access or seeing into the forest canopy. question of the relative validity of ground-based methods for
Birds are a conspicuous and important component of the study of canopy birds. Therefore, I compared the results of
tropical forest ecosystems. Canopy bird communities include ground point counts and two types of canopy point counts in
important functional groups, such as seed dispersers, polli- a 100-ha plot in northern Honduras sampled over a complete
nators, and top predators (Howe 1977, Nadkarni and Matel- annual cycle. To my knowledge, this study is the first to make
son 1989, Blake and Loiselle 2000, Holbrook and Smith 2000, such a comparison and the first to use canopy-based methods
Anderson 2001). Long-distance and elevational migrants also to sample canopy birds in a 100-ha plot. A quantitative defini-
occur in the canopy, and their conservation requires an un- tion of what constitutes a “canopy bird species” has remained
derstanding of their ecology, distribution, and abundance elusive but should prove useful in discussions of canopy bird
(Loiselle 1987, Powell and Bjork 1995, Chaves-Campos et ecology and conservation. To compare canopy and noncanopy
al. 2003). Of further conservation concern is the suggestion birds, I combined the data sets to define quantitatively the core
that canopy bird species may be disproportionately sensitive canopy species of the study area. I then compared patterns of
to forest fragmentation (Castelletta et al. 2000, Robinson et species richness, as well as family and species composition, as
al. 2000, Sodhi et al. 2004), and several canopy species (e.g., represented by the respective data sets derived from the three
large raptors, macaws, and some frugivores) are threatened or methods. Finally, I compared detection rates of canopy birds
endangered in Middle America (Terborgh and Winter 1980, as a whole and of groups of quieter or less conspicuous species
Kattan 1992, Levey and Stiles 1994). It follows that the con- that I suspected would be underrepresented in ground-based
servation of tropical forests will depend in part on an accurate data sets.
appreciation of canopy birds and their interactions within the
forest ecosystem. METHODS
One of the methods used most frequently to study the
STUDY AREA
abundance, distribution, and ecology of forest birds is the
point count (Ralph et al. 1995). Point-count data are used to I delineated a 100-ha study site (15° 43.40` N, 86° 44.08` W) in
make inferences about the presence and abundance of birds, the Río Cangrejal valley on the humid north flank of Pico Bo-
but an important consideration of this method is the proba- nito National Park, Honduras. The park encompasses 107 090
bility of the birds’ detection (Farnsworth et al. 2002), which ha and elevations ranging from 50 to 2480 m (FUPNAPIB
can vary widely with species, habitat, and time of day or year, 2004). A majority of the park is primary forest with no recent
among other factors (Blake 1992, Ralph et al. 1995, Pacifici history of human disturbance. Annual precipitation and mean
et al. 2008). The ability of an observer on the ground to de- temperature for the site are 2900 mm and 26° C, respectively.
tect birds in a tropical forest canopy varies dramatically be- The wet and dry seasons are distinct: the driest months are
cause of (1) the range of conspicuousness of different species April and May with an average monthly rainfall of 89 mm; the
depending on size, coloration, vocalizations, and movements wettest months are November and December with an average
and (2) the dense foliage and distance that separate the ob- monthly rainfall of 510 mm (FUPNAPIB 2004).
server from the canopy (Pacifici et al. 2008). In short, some Slopes in the study area are nearly flat to moderately
canopy species should be harder to detect from the ground, in steep, and elevations range from 100 to 350 m. The forest av-
particular those that have soft vocalizations, call infrequently, erages 35–40 m high, and canopy emergents are rare. Pri-
or remain perched for long periods of time. The difficulty of mary and mature secondary moist forests are both present,
detecting such species has led to the conclusion that canopy with primary forests constituting about 60% of the study area.
species are likely underrepresented in otherwise comprehen- Common overstory tree species include Symphonia globulif-
sive studies of neotropical forest-bird assemblages that rely on era, Vochysia guatemalensis, Virola koschnyi, Tapipira gui-
ground-based sampling (Robinson et al. 2000, Blake 2007). anensis, Astronium graveolens, Bursera simaruba, Pouteria
A few pioneering studies have advanced methods for spp., Ficus spp., Calophyllum brasiliense, Dialium guianen-
studying birds in neotropical forest canopies (Greenberg 1981, sis, and Schizolobium parahybum. Numerous wind-snapped
Loiselle 1988, Naka 2004). Despite such advances, no attempt trees, gaps, and canopy vine tangles suggest a high incidence
has been made to quantify the differences between ground- of weather-related disturbance.
based and canopy-based data on canopy birds. Because of the
continued reliance on ground-based methods, such a compari- SAMPLING METHODS
son is crucial to assessing their value for the study of canopy I tested three methods, one ground-based and two canopy-
birds and to determine what biases or limitations may ex- based, for their effectiveness in detecting canopy birds. The
ist. Any substantial weaknesses that are revealed could have first method (henceforth ground) used point transects located
228 DAVID L. ANDERSON

along pre-existing trails. I established 30 count points along canopy), and (4) canopy (the sum of all tree crowns exposed
two trails that bisected the study area. Stations were separated to the sky above; Bongers 2001, Fig. 1C). Birds flying over the
by 100 m. This spacing was chosen because many neotropi- forest were noted as such and excluded from analyses. Finally,
cal species are hard to hear at distances 30 m (Terborgh et al. I noted for every observation whether the initial detection cue
1990, Robinson et al. 2000, Blake 2007). Censuses started 30 was by sight or sound.
min after sunrise and typically lasted 3 hours, during which I sampled birds over a complete annual cycle from April
I normally covered approximately 1.3 km and 13 points. I se- 2006 to April 2007. I subdivided the year into four seasons—
lected the starting time to standardize sampling times with early and late dry, and early and late wet—to compare seasonal
canopy censuses (see below). I rotated starting points to en- variation in species abundance. This technique has been used
sure, as much as possible, that all points were covered early in previous studies of neotropical forest birds (Greenberg 1981,
in the morning when vocal activity was greatest. I conducted Loiselle 1987, 1988, Blake 1992) to account for changes in for-
counts on days with no rain and little or no wind and termi- est phenology, principally the development and abundance
nated counts when rain or wind interfered with the detection of certain resources used by birds or their prey (e.g., flowers,
of birds. I counted birds for 10 min at each point. Any individ- nectar, fruit, insects, and leaves) that may affect the seasonal
ual detected from more than one point was noted as such, but abundance and distributions of birds. Furthermore, seasonal
only the first detection was used in analyses. The maximum changes in leaf density caused by leaf fall and regeneration, as
number of individuals per species, summed for all points of a well as natural levels of background noise, particularly cica-
given census walk, was the datum used in analyses. das, can affect the detectability of birds (Pacifici et al. 2008).
For the second method (repeat-tree method), I conducted
repeat censuses from the crowns of two trees, a method similar STATISTICAL ANALYSES
to that of previous canopy-bird studies (Greenberg 1981, Loi- Because of differences in the spatial distribution, size, and
selle 1988, Naka 2004). The first tree was a 45-m tall Vochysia number of plots used in each method, I do not attempt to esti-
guatemalensis in mature secondary forest at 115 m. The sec- mate densities of species detected by each method. Rather, I
ond tree was a 60-m tall Virola koschnyi in primary forest at present numbers of individual detections (by sight or sound)
220 m. The trees were separated by 1 km. I conducted 22 cen- per method. This conservative approach focuses on the ability
suses from these two trees. For the third method (single-tree of each method to detect species and individuals in the canopy
method), I conducted single censuses from the crowns of 22 rather than to describe the canopy-bird community per se and
separate trees interspersed throughout the entire study area. is in line with the scope of the current study.
I used a crossbow and single-rope technique to climb canopy I used the method of Neu et al. (1974) to quantify birds’
trees (Sillett and Van Pelt 2000). I selected census trees on preference for the canopy stratum. I established 95% confi-
the criteria that they were safe to climb, had an open crown dence limits, based on Bonferroni’s adjustment of the signif-
structure that allowed views out of the census tree, and were a icance level, around the observed frequency of detection in
minimum of 50 m from other census trees. Census trees closer the canopy stratum for species with q4 detections. A signifi-
than 100 m to each other were censused in different seasons cant preference for the canopy was indicated by expected val-
(see below). ues below the 95% confidence limits for the observed values
All canopy censuses began 30 min after sunrise and (Haney and Solow 1992, Cardoso da Silva et al. 1996). I refer
lasted 3 hours. Following the protocol of Loiselle (1988) and to species that met this criterion as core canopy species. I ex-
Naka (2004), I further subdivided the 3-hr censuses into cluded from analyses nocturnal species and birds flying over
12 consecutive 15-min intervals. The use of short intervals the forest. Taxonomy follows AOU (1998) and supplements.
facilitates tracking individual birds, which can be distin- I used rarefaction analyses to compare rates of species ac-
guished by differences in plumage and location in the forest, cumulation among the three methods. Rarefaction curves are
and avoids double counting (i.e., it is easier to follow individ- produced by repeatedly and randomly resampling the pool of
ual birds and their direction of travel over 15 min than over 3 observations and plotting the average number of species rep-
hr). The maximum number of individuals per species within resented by n individuals; they are therefore a statistical repre-
a 15-min period was used for analysis of census results, un- sentation of species-accumulation curves (Gotelli and Colwell
less additional individuals were identified on the basis of sex 2001, Magurran 2004). I used Chao 1 and Chao 2 nonpara-
or plumage. Canopy census plots had a radius of 150 m and an metric estimators (Magurran 2004) to estimate species rich-
area of 7.1 ha. Additionally, both repeat trees and eight single ness from each method. These analyses were conducted with
trees were paired with count points on the ground. EstimateS version 7.5 (Colwell 2005).
I recorded all birds seen or heard and categorized them I followed the methods of Pitman et al. (2001) and Blake
into one of four forest strata: (1) ground (soil, leaf litter, and (2007) to compare the number of detections per family. This
fallen logs), (2) understory (the space from the ground to method tests the null hypothesis that the three methods are
2 m), (3) midstory (the space between the understory and equivalent in terms of species or family composition. If two
 METHODS FOR CANOPY BIRDS IN TROPICAL FORESTS 229

FIGURE 1. Sample-based rarefaction curves (left column) and corresponding estimates of species richness (right column) for three cen-
sus methods and the combined data set, Río Cangrejal study area, Pico Bonito National Park, Honduras, April 2006–April 2007. Each pair
of cells depicts all detections of all species (a, b), canopy detections of all species (c, d), and all detections of core canopy species (e, f). Each
duo in b, d, and f corresponds to richness estimates from Chao 1 and Chao 2 estimators, respectively. Shapes and vertical bars represent
means and 95% confidence intervals, respectively.

methods are equivalent in the number of detections, then the comparisons to test for significant differences between pairs
slope of the regression line should be equal to 1 (Blake 2007). of methods.
I used two subtly different approaches to compare my ability I used a two-step process to test for an effect of method
to detect canopy birds with each census method: (1) the hourly on the hourly detection rates of three groups of inconspicuous
detection rate of all species and individuals in the canopy canopy species, namely, (1) inconspicuous residents with soft
stratum and (2) the hourly detection rate of core canopy spe- or infrequent vocalizations, (2) inconspicuous migrants, and
cies and individuals in all forest strata. The first approach ad- (3) canopy hummingbirds. For each group I considered only
dresses the question, “are the methods equal in their ability to those species that qualified as core canopy species and com-
detect birds in the canopy?” The second approach addresses pared the detection rate on the basis of all individuals within
a different question, “are the methods equal in their ability to each group that were detected in any stratum. To test for dif-
detect those species that spend a substantial portion of their ferences, I first used a Kruskal–Wallis test to test for an over-
time in the forest canopy?” I used mixed-model ANOVA to all effect of method on detection rates for each group. Upon
control for the effects of season and forest type and to test for finding a significant effect of method, I then used a Wilcoxon
an effect of method on the hourly detection rates of species and signed-rank test to make pairwise comparisons between
individuals. Data were first square-root transformed to meet methods. I used this same approach to test for differences be-
assumptions of normality. I used post hoc Tukey’s pairwise tween detection rates of highly vocal and conspicuous canopy
230 DAVID L. ANDERSON

TABLE 1. Number of species and detections (by sight or sound) by three census methods in 100 ha near the
Río Cangrejal, Pico Bonito National Park, Honduras.

Number of species/number of detections

All species Core canopy species

Canopy Canopy
Method Censuses Census hours All strata stratum All strata stratum

Ground 27 66.7 (2.4 o 0.56) 110/1824 36/300 47/986 31/329

Repeat-tree 22 66 (3.0 o 0) 121/1598 91/936 64/1149 64/900
Single-tree 22 66 (3.0 o 0) 123/1191 96/675 59/863 59/614
Total 71 198.7 157/4613 112/1911 65/2998 65/1843

species. Results of all statistical tests were assumed signifi- and 189, not significantly greater than the 156 actually ob-
cant at P  0.05. served (Fig. 1b). Richness estimates for core canopy spe-
cies by the canopy methods and the combined data set did
RESULTS not differ, but all three of these estimates were significantly
higher than those derived from ground censuses (Fig. 1f).
NUMBERS OF DETECTIONS AND SPECIES Repeat-tree and single-tree censuses detected nearly the full
I conducted a total of 71 censuses from April 2006 to April complement of core canopy species, whereas the ground
2007, resulting in 4613 individual detections of 157 species, censuses detected only 47 (72%) of all core canopy species.
112 of which I observed in the canopy (Table 1). I recorded On the basis of the level of sampling I achieved, richness
an additional 27 species outside standardized surveys but ex- estimates for all canopy species did not differ regardless of
cluded these from analyses. Sixty-five species (60% of all census method (Fig. 1d).
species detected in the canopy) qualified as core canopy spe- Distributions of abundances of the core canopy species ac-
cies. Species-accumulation curves for all species suggest that cording to canopy and ground censuses differed significantly
most species on the plot were detected by the combination of when comparisons were limited to canopy detections only
methods but that sampling by any single method was less (Kolmogorov–Smirnov two-sample tests; Fig. 2b, Table 2).
complete (Fig. 1a). Curves for detections in the canopy stra- Most of the curve for ground censuses lies below the curves
tum do not reach asymptotes, suggesting that some species for canopy censuses, indicating (1) larger differences in abun-
observed in lower strata would eventually be encountered in dance by species and (2) a greater predominance of common
the canopy stratum (Fig. 1c). Sampling of core canopy spe- species and an omission of rare ones. These patterns disap-
cies was more complete, with all curves quickly reaching as- pear when detections of core canopy species in all strata are
ymptotes (Fig. 1e). The actual number of species estimated considered (Fig. 2a, Table 2), in which case there were no sig-
for the study area by the combined data set lies between 162 nificant differences between methods.

FIGURE 2. Rank-abundance curves based on numbers of detections (by sight or sound) of core canopy species in all strata (a) and the
canopy stratum only (b) on the 100-ha Río Cangrejal study plot, Pico Bonito National Park, Honduras, April 2006–April 2007.
 METHODS FOR CANOPY BIRDS IN TROPICAL FORESTS 231

TABLE 2. Kolmogorov–Smirnov two-sample significance the same for the repeat-tree and single-tree methods, as indi-
tests for differences between census methods of rank-abun- cated by the slope of the regression being equal to 1.0 (Fig 3c).
dance distributions of core canopy species.
In contrast, the pattern of species richness per family by either
canopy method differed significantly from that by the ground
Comparison KS statistic P
method, as indicated by slopes deviating substantially from
Detections in the canopy stratum 1.0 (Fig. 3a, b). With only two exceptions, species richness
Ground–repeat-tree 0.16 0.014 per family was greater for both canopy methods than for the
Ground–single-tree 0.18 0.006 ground method. Patterns of detections per family were less
Repeat-tree–single-tree 0.08 0.489
precise, with no relationship between methods approaching
Detections in all strata
Ground–repeat-tree 0.05 0.906 a slope of 1.0 (Fig. 3d, e, f), although once again the canopy
Ground–single-tree 0.07 0.638 methods were most similar. Substantially more individuals
Repeat-tree–single-tree 0.05 0.904 were detected in the canopy during canopy censuses than dur-
ing point counts from the ground.

FAMILY COMPOSITION SPECIES COMPOSITION

I observed 27 families in the canopy stratum (Table 3). I found important differences among dominant species as de-
Ground censuses detected eight fewer families in the canopy tected by the three census methods. Eleven of the top 20 spe-
than repeat-tree censuses and nine fewer families than single- cies were detected by all three methods, but in no case were
tree censuses. The pattern of species richness per family was species ranked the same in numbers of detections by the three

TABLE 3. Numbers of species and individual detections (n) by sight or sound in the canopy stratum for
each of three census methods at Pico Bonito National Park, Honduras, April 2006 to April 2007.

Ground Repeat-tree Single-tree Combined

Family Spp. n Spp. n Spp. n Spp. n

Cracidae (guans) 1 2 0 0 1 1 1 3
Accipitridae (hawks) 2 2 1 5 2 8 2 15
Columbidae (pigeons) 1 9 3 5 1 4 3 18
Psittacidae (parrots) 2 50 2 46 3 74 3 170
Cuculidae (cuckoos) 0 0 2 16 2 8 2 24
Trochilidae (hummingbirds) 0 0 9 36 6 34 9 70
Trogonidae (trogons) 3 4 2 9 4 14 4 27
Momotidae (motmots) 0 0 1 1 1 2 1 3
Galbulidae (jacamars) 0 0 0 0 1 4 1 4
Bucconidae (puffbirds) 0 0 1 2 0 0 1 2
Ramphastidae (toucans) 2 43 4 90 3 50 4 183
Picidae (woodpeckers) 1 1 5 26 3 15 5 42
Funariidae (ovenbirds) a 0 0 3 7 5 7 5 14
Thamnophilidae (antbirds) 0 0 2 2 3 8 3 10
Tyrannidae (tyrant flycatchers) b 5 18 16 108 20 94 21 220
Cotingidae (cotingas) 0 0 1 11 1 4 1 15
Pipridae (manakins) 0 0 1 1 2 3 2 4
Vireonidae (vireos) 3 89 7 141 7 86 7 316
Corvidae (jays) 1 5 1 25 1 12 1 42
Troglodytidae (wrens) 0 0 1 4 1 7 1 11
Polioptilidae (gnatcatchers) 1 5 2 39 2 33 2 77
Turdidae (thrushes) 1 4 1 3 3 7 3 14
Parulidae (wood-warblers) 5 8 11 64 8 46 12 118
Thraupidae (tanagers) c 3 4 7 99 7 60 8 163
Cardinalidae (grosbeaks) 1 12 2 50 2 6 2 68
Icteridae (blackbirds) 2 41 3 128 4 63 4 232
Fringillidae (euphonias) 2 3 3 18 3 25 4 46
Total families 17 25 26 27
a
Includes Dendrocolaptidae.
b
Includes Tityra, Pachyramphus, and Schiffornis.
c
Includes Piranga.
232 DAVID L. ANDERSON

FIGURE 3. Number of species (a–c) and detections (by sight or sound; d–f) per family estimated by three census methods in the Río Can-
grejal study area, Pico Bonito National Park, Honduras, April 2006–April 2007. Data represent detections from the canopy stratum only.
Straight lines indicate a 1:1 relationship between values for two methods. Actual slopes are given.

methods (Table 4). Species that were more evenly distributed fewer species were detected in the canopy during ground
across the study area (e.g., Hylophilus decurtatus, Cyanoco- censuses and that the evenness of dominant species by this
rax morio, Polioptila plumbea) or that were highly conspicu- method was also less. This is evident in the inclusion of Penel-
ous (e.g., H. decurtatus, Ramphastos sulfuratus, Psarocolius ope purpurascens, not characteristically a canopy species, in
wagleri) ranked similarly by all methods. Species that were the list of dominant canopy species for ground censuses.
either less common or less conspicuous (e.g., Thalurania co-
lombica, Piranga rubra, Chlorophanes spiza) tended to rank DETECTION RATES
very differently by different methods. The top 20 species ac- I found important differences between detections rates of indi-
counted for a greater percentage of canopy observations made vidual species as well as between groups of species. Secretive
from the ground (96%) than they did in canopy observations migrant and resident species (Table 5) were detected at signifi-
made from the canopy (76% and 70% for the repeat-tree and cantly greater hourly rates by canopy methods than from the
single-tree methods, respectively). This pattern signals that ground, but between canopy methods detection rates did not
 METHODS FOR CANOPY BIRDS IN TROPICAL FORESTS 233

TABLE 4. Percentage of detections and rank for the 20 most TABLE 5. Groups of core canopy species used in comparisons
frequently detected species in the forest canopy, by method, Río of hourly detection rates by the three census methods.
Cangrejal, Pico Bonito National Park, Honduras.
Groups/species
Method
Secretive or inconspicuous species
Ground Repeat-tree Single-tree Migrants
Myiarchus crinitus
Species % Rank % Rank % Rank Vireo flavifrons
Vireo philadelphicus
Penelope purpurascens 0.7 12 0.0 0.1 Vireo olivaceus
Patagioenas nigrirostris 3.0 7 0.2 0.6 Dendroica petechia
Aratinga nana 3.7 6 1.3 3.0 6 Dendroica pensylvanica
Pyrilia haematotis 13.0 3 3.6 8 7.7 2 Dendroica magnolia
Piaya cayana 0.0 1.5 15 1.0 Dendroica virens
Florisuga mellivora 0.0 0.3 1.5 13 Dendroica castanea
Thalurania colombica 0.0 1.6 14 2.1 9 Mniotilta varia
Trogon violaceus 0.7 12 0.5 0.6 Setophaga ruticilla
Pteroglossus torquatus 3.7 6 4.7 5 4.1 4 Piranga olivacea
Ramphastos sulfuratus 10.7 4 4.3 6 3.1 5 Icterus galbula
Melanerpes pucherani 0.0 2.0 11 1.0 Residents
Ornithion semiflavum 1.0 11 1.9 12 1.9 10 Notharchus macrorhynchos
Zimmerius vilissimus 0.7 12 1.3 1.3 14 Tolmomyias sulphurescens
Megarynchus pitangua 0.0 0.1 1.6 12 Cotinga amabilis
Tityra semifasciata 3.7 6 2.9 9 2.1 9 Chlorophanes spiza
Vireo olivaceus 0.3 6.0 3 2.4 8 Hummingbirds
Hylophilus decurtatus 26.7 1 6.1 2 7.7 2 Florisuga mellivora
Vireolanius pulchellus 2.7 8 1.7 13 0.3 Thalurania colombica
Cyanocorax morio 1.7 9 2.7 10 1.8 11 Amazilia tzacatl
Polioptila plumbea 1.7 9 3.8 7 4.3 3 Heliothryx barroti
Catharus ustulatus 1.3 10 0.3 0.7 Tilmatura dupontii
Dendroica pensylvanica 1.3 10 1.9 12 2.1 9 Highly detectable, conspicuous species
Piranga rubra 0.3 1.5 15 3.0 6 Piaya cayana
Chlorophanes spiza 0.0 2.0 11 1.2 Ramphastos sulphuratus
Cyanerpes cyaneus 0.7 12 3.8 7 2.7 7 Attila spadiceus
Cyanerpes lucidus 0.0 1.6 14 0.0 Cyanocorax morio
Caryothraustes 4.0 5 5.2 4 0.7 Thryothorus maculipectus
poliogaster
Psarocolius wagleri 13.3 2 12.0 1 7.9 1
Euphonia hirundinacea 0.3 0.2 1.5 13
Euphonia gouldi 0.7 12 1.2 2.1 9 and single-tree methods (Fig. 5). For core canopy individuals
Percentage of total 95.7 76.4 70.1 the relationship was more complex, although again the detec-
canopy detections tion rate by ground censuses was greatest (Fig. 5).

DISCUSSION
differ (Fig. 4). I found no statistical difference among any of
the methods in the hourly detection rate of canopy humming- OVERVIEW
birds (Table 5), perhaps because of the overall low detection Because a major goal of this study was to test methods for
rate of this group. Ground censuses, however, detected only the study of canopy birds, I chose to focus specifically on
four of seven core canopy hummingbirds, and none in the can- methods that are either widely available to forest ecologists or
opy stratum. I found that the choice of method significantly which have been used in past studies of canopy birds. Ground-
affected the hourly detection rates of both species (F2,86  based point counts remain the primary tool for observing for-
31.1, P < 0.0001) and individuals (F2,86  21.9, P < 0.0001) in est birds in all levels of the forest because they are easy to
the canopy stratum (Fig. 5). Similarly, the choice of method conduct and because they are readily adaptable to different
significantly affected the hourly detection rates of core canopy environments or particular questions of interest (Ralph et al.
species (F2,86  20.4, P < 0.0001) and individual core canopy 1995). Repeated censuses from one or a small number of can-
birds (F2,86  9.5, P  0.0002) when detections in all strata were opy viewpoints, as used in pioneering studies of canopy birds
considered (Fig. 5). For core canopy species, detection rates (Greenberg 1981, Loiselle 1988, Naka 2004), will continue to
were significantly greater for ground censuses than by either be important when canopy access is constrained by the avail-
canopy method and and did not differ between the repeat-tree ability of canopy towers or cranes. An important consideration
234 DAVID L. ANDERSON

FIGURE 4. Hourly detection rates for numbers of individuals of FIGURE 5. Hourly detection rates of all species and individu-
secretive migrant, secretive resident, and conspicuous resident spe- als detected in the canopy stratum and in all forest strata by three
cies observed (by sight or sound) by three census methods, Río census methods in the Río Cangrejal study area, Pico Bonito Na-
Cangrejal study area, Pico Bonito National Park, Honduras, April tional Park, Honduras, April 2006–April 2007. Means o 1 SE are
2006–April 2007. Means o 1 SE are shown. Different letters indicate shown. Different letters indicate significant differences between
significant differences between groups. groups.

with canopy-based methods is the high temporal and spatial (Pacifici et al. 2008). This balance comes with a tradeoff. In-
variability of food resources in the canopy (Leigh et al. 1996), creasing vegetation density decreases sound transmittance,
which could influence the number of species and individual and increasing distance between the observer and bird less-
birds available to the observer. For this reason I incorporated ens probability of detection (Bibby et al. 2000, Ellinger and
single censuses from trees scattered throughout the study Hödl 2003). In tropical forests, the vocalizations of many
area for comparison with my repeat-tree method. Although a species are hard to detect at distances 30 m, and other spe-
more direct comparison of bird detectability from the canopy cies call infrequently (Robinson et al. 2000, Blake 2007).
and ground could have been obtained by paired canopy and These factors increase the dependence on visual detections.
ground observations of equal length, the intended scope of the Ellinger and Hödl (2003) found that sound waves are scat-
study was a quantitative comparison of methods. The study tered by the uneven canopy surface and that species with
design and statistical methods therefore reflect the larger goals high-frequency vocalizations compensate by singing above
of the study. the canopy. In the Río Cangrejal study area the forest aver-
aged 35 to 40 m tall with some trees reaching heights of over
DETECTION PROBABILITIES 50 m. The combination of forest height and foliage density
The spatial distribution, population density, and behavior of a with the habit of small species with high-frequency voices to
species combined with the choice of method can affect how the sing at the top of the forest undoubtedly decreased the prob-
proportional abundance of that species at a site is characterized. ability of birds in the upper foliage layers being detected.
In this study Hylophilus decurtatus was disproportionately pre- Species underrepresented by ground censuses frequented
dominant in ground censuses and was detected more than twice upper levels of the forest and were disproportionately harder
as often from the ground as from the canopy. This finding is to detect through auditory cues. This result is independent of
consistent with results from ground-based censuses in central observer skill, which cannot reduce the amount of obstruc-
Panama, where Robinson et al. (2000) found H. decurtatus to tion or the distance between a bird at the top of the forest
be one of the eight most abundant species. Despite being con- and an observer on the ground. Furthermore, the terrain of
sidered one of the more abundant species at La Selva, Costa the Río Cangrejal study area is uneven and sloping, facilitat-
Rica (Levey and Stiles 1994), it was not ranked among the most ing viewing into the canopy from the ground. In portions of
abundant species when canopy-based data alone were used Amazonia and wherever the terrain is uniformly flat, a
(Loiselle 1988). In contrast, quieter species were systematically ground-based observer is at an even greater disadvantage.
overlooked. During ground censuses I failed to detect 67% of My results also highlight that the probability of detecting
all canopy species, including 28% of core canopy species, a re- birds in the canopy is a function of both census method and
sult that biased the pattern of familial richness as well. bird behavior. Canopy and ground point counts differ in that
Two factors, vegetation density and distance between ob- a canopy point count is conducted at a single point over a long
server and bird, increase dependence on auditory detections period, whereas a ground transect consists of many distinct
 METHODS FOR CANOPY BIRDS IN TROPICAL FORESTS 235

point counts of short duration. Remaining at a single station was severely misrepresented by census data from the ground
in the canopy for several hours resulted in rates of detection of only. Among the core canopy birds alone three of seven spe-
birds in the canopy higher than by ground censuses. Species cies of nectarivores, two of four obligate frugivores, and seven
richness by canopy methods was also substantially higher. of 33 other potential seed dispersers went undetected during
Ground methods detected conspicuous canopy species more ground censuses. For example, Cotinga amabilis, a medium-
often, partly because during any transect the observer walks sized frugivore of the forest canopy, was observed frequently
through multiple territories of vocal species. during canopy censuses. My observations suggest that, by re-
gurgitating seeds onto tree branches, it plays an important role
ESTIMATES OF POPULATION DENSITY in seed dispersal of certain Loranthaceae (principally Psitta-
Robinson et al. (2000) used multiple ground-based methods to canthus rhyncanthus), hemiparasitic mistletoes that grow on
estimate population densities for 165 of 252 species in central canopy trees. Fruits of these plants are in turn fed on by at
Panama. Their density estimates for three groups of species least 19 species of migrant and resident birds, and their flow-
were likely affected by the use of ground-based census data. ers are favored by hummingbirds (unpubl. data). Despite the
The first group comprised 41 diurnal interior-forest species brilliant plumage of the male cotinga, my ground censuses did
for which they attempted no estimates because of the birds’ not to detect it, and all interactions between birds and the Lo-
high mobility or small sample sizes resulting from difficulty ranthaceae were viewed exclusively from canopy viewpoints.
of observation from the ground. Twenty of these species are Furthermore, many canopy species detected from the ground
characteristic of the canopy, including raptors (Leucopternis through auditory cues were seldom, if ever, observed visually.
albicollis, Falco rufigularis) and visually conspicuous but Understanding the foraging ecology and behavior of canopy
otherwise secretive residents (e.g., Cotinga nattererii). In con- species is best accomplished from canopy viewpoints.
trast, some of these species are readily observed from canopy
viewpoints. In Honduras L. albicollis may be the single bird IMPLICATIONS FOR CONSERVATION
most easily observed from the canopy (this study, Anderson Results of this study demonstrate that estimates of population
2001); some other raptors are easily observed as well. Canopy- density, species distributions, and local species richness can
dwelling hummingbirds are notoriously difficult to detect all be biased by exclusively ground-based methods. I offer two
from the ground, yet from my tree vantage points I could of- scenarios in which data from ground-based censuses alone
ten identify and track individual hummingbirds at distances could affect conservation practices:
up to 80 m. The second group consisted of some quiet canopy 1. Estimates of species richness of birds are often used
species, (e.g., Heliothryx barroti, Dendroica pensylvanica, to determine the conservation importance of particular sites.
Chlorophanes spiza), and Robinson et al. (2000) estimated In sites with identical species richness, differences in forest
population densities for these. Given that these species are stature and structure may affect detectability of birds in upper
either largely silent or best heard at short distances and are strata and, therefore, bias estimates of richness and the priori-
difficult to view from the ground, my results suggest that Rob- tization of conservation potential.
inson et al. (2000) may have underestimated densities of some 2. Ground-based censuses are often used to describe ef-
of them by up to 50%. It is likely that the ground methods fects of disturbances, such as selective logging and storm
of Robinson et al. missed a third group of species altogether, damage, on population densities of birds. Disturbances may
particularly certain nearctic migrants, further canopy hum- affect an observer’s ability to detect canopy birds in two ways.
mingbirds, and quiet residents. In my study ground censuses First, a disturbance that reduces the amount of foliage in a for-
revealed 30% fewer species than either canopy method, and est can enhance the observer’s ability to see and hear birds at
the mean estimate of species richness for the study area was greater heights. Second, a disturbance may affect movement
approximately 10% lower when the canopy methods were patterns of birds and render them more detectable by the ob-
excluded. server. In either case, estimates of population density may be
erroneously biased upward. Although forest disturbance may
FUNCTIONAL ECOLOGY favor some species of birds, changes in detectability may lead
Understanding the roles of birds in an ecosystem is a central to this conclusion when it is actually false.
component of tropical forest ecology. For example, much im- Canopy-based censuses may be critical for biological
portance has been placed on the role of birds in seed disper- monitoring in several respects. Thirty-five species listed as
sal and in the natural regeneration of forests after disturbance endangered, threatened, or vulnerable by the International
(Cardoso da Silva et al. 1996, Howe 1996, Silva et al. 2002, Union for the Conservation of Nature (IUCN) inhabit cano-
Cordiero and Howe 2003). Evaluating the potential of the local pies of lowland neotropical forests (Stotz et al. 1996, BirdLife
avifauna to disperse seeds of varying sizes and characteristics International 2000). Of these, six are raptors whose popula-
requires an accurate depiction of the seed-dispersing guild. tion densities are often naturally low and which may be highly
My study found that the functional composition of the avifauna visible from canopy viewpoints. Another 13 are frugivores or
236 DAVID L. ANDERSON

omnivores that may be important seed dispersers. One is a BLAKE, J. G. 2007. Neotropical forest bird communities: a compar-
hummingbird and likely difficult to detect from the ground. ison of species richness and composition at local and regional
scales. Condor 109:237–255.
If my results have general applicability, then ground-based
BLAKE, J. G., AND B. E. LOISELLE. 2000. Diversity of birds along an
surveys will underestimate the densities of these and other elevational gradient in the Cordillera Central, Costa Rica. Auk
inconspicuous species. Additionally, canopy-based censuses 117:663–686.
may be necessary for monitoring long-term changes in com- BONGERS, F. 2001. Methods to assess tropical rain forest canopy
munity composition, including species loss following forest structure: an overview. Plant Ecology 153:263–277.
CARDOSO DA SILVA, J. M., C. UHL, AND G. MURRAY. 1996. Plant
fragmentation, as on Barro Colorado Island, Panama, where
succession, land management, and the ecology of frugivorous
the difficulty of detecting canopy species may affect estimates birds in abandoned Amazonian pastures. Conservation Biology
of species extirpation and recolonization (Robinson 1999). 10:491–503.
CASTELLETTA, M., N. S. SODHI, AND R. SUBARAJ. 2000. Heavy extinc-
tions of forest avifauna in Singapore: lessons for biodiversity
RECOMMENDATIONS conservation in southeast Asia. Conservation Biology 14:1870–
Canopy-based methods offer obvious advantages for studies 1880.
of canopy birds, but the question remains as to what circum- CHAVES-CAMPOS, J., J. E. A RÉVALO, AND M. A RAYA. 2003. Altitu-
dinal movements and conservation of Bare-necked Umbrella-
stances justify the added effort and expense that field work
bird Cephalopterus glabricollis of the Tilarán Mountains, Costa
in the canopy requires. Cohn-Haft et al. (1997) demonstrated Rica. Bird Conservation International 13:45–58.
how canopy surveys from a single canopy tower improve the COHN-H AFT, M., A. WHITTAKER, AND P. C. STOUFFER. 1997. A new
understanding of an otherwise well-known avifauna. Conse- look at the “species poor” central Amazon; the avifauna north of
quently, long-term data sets at sites of high biological interest, Manaus, Brazil. Ornithological Monographs 48:205–235.
COLWELL, R. K. [ONLINE]. 2005. Estimates: Statistical estimation of
such as tropical field stations, should include canopy methods
species richness and shared species from samples. Version 7.5.
for a better understanding of trends in canopy bird commu- <http://purl.oclc.org/estimates (1 December 2008).
nities. Rapid ecological assessments may also justify canopy CORDIERO, N. J., AND J. R. HOWE. 2003. Forest fragmentation sev-
methods, or, if canopy methods are not used, need to state ex- ers mutualism between seed dispersers and an endemic African
plicitly that ground-based assessments likely miss or underes- tree. Proceedings of the National Academy of Sciences
100:14052–14056.
timate the density of core canopy species.
ELLINGER, N., AND W. HÖDL. 2003. Habitat acoustics of a neotropical
lowland rainforest. Bioacoustics 13:297–321.
ACKNOWLEDGMENTS FARNSWORTH, G. L., K. P. BURNHAM, AND J. D. NICHOLS. 2002. A
removal model for estimating detection probabilities from point-
The Corporación Hondureña de Desarollo Forestal (COHDEFOR) in count surveys. Auk 119:414–425.
Honduras kindly granted research permits. Luis Soto, Mike Beard, FUPNAPIB. 2004. Plan de manejo, Parque Nacional Pico Bonito.
and Juan Gomez provided valuable assistance in the field. Funding Fundación Pico Bonito, La Ceiba, Honduras.
and material support were provided by the USAID MIRA project, GOTELLI, N. J., AND R. K. COLWELL. 2001. Quantifying biodiversity:
Cleveland Zoological Society Small Grants Program, Explorers procedures and pitfalls in the measurement and comparison of
Club, Eagle Optics, Louisiana State University Museum of Natu- species richness. Ecology Letters 4:379–391.
ral Science Big Day Fund, LSU BioGrads, Global Forest Science, GREENBERG, R. 1981. The abundance and seasonality of forest can-
and USFWS Neotropical Migratory Bird Conservation Act funds opy birds on Barro Colorado Island, Panama. Biotropica 13:241–
administered by the Gulf Coast Bird Observatory. Logistical sup- 251.
port was provided by the MIRA project of USAID and by Fundación HANEY, J. C., AND A. R. SOLOW. 1992. Testing for resource use and
Pico Bonito. The manuscript benefited greatly from comments from selection by marine birds. Journal of Field Ornithology 63:43–52.
J. V. Remsen, Jr., Philip C. Stouffer, S. Wong, and two anonymous HOLBROOK, K. M., AND T. B. SMITH. 2000. Seed dispersal and move-
reviewers. ment patterns in two species of Ceratogymna hornbills in a West
African tropical lowland forest. Oecologia 125:249–257.
LITERATURE CITED HOWE, H. F. 1977. Bird activity and seed dispersal of a tropical wet
forest tree. Ecology 58:539–550.
A MERICAN ORNITHOLOGISTS’ UNION. 1998. Check-list of North HOWE, H. F. 1996. Fruit production and animal activity in two
American birds, 7th ed. American Ornithologists’ Union, Wash- tropical trees, p. 189–199. In E. G. Leigh, Jr., A. S. Rand and D.
ington, DC. W. Windsor [EDS.], The ecology of a tropical forest: seasonal
A NDERSON, D. L. 2001. Landscape heterogeneity and diurnal raptor rhythms and long-term changes. Smithsonian Institution Press,
diversity in Honduras: the role of indigenous shifting cultivation. Washington, D.C.
Biotropica 33:511–519. K ATTAN, G. H. 1992. Rarity and vulnerability: the birds of the Cor-
BIBBY, C. J., N. D. BURGESS, D. A. HILL, AND S. H. MUSTOE. 2000. dillera Central of Colombia. Conservation Biology 6:64–70.
Point counts and point transects, p. 91–113. In Bird census tech- LEIGH, E. G., JR., A. S. R AND, AND D. W. WINDSOR [EDS.]. 1996. The
niques. Academic Press, London. ecology of a tropical forest: seasonal rhythms and long-term
BIRDLIFE INTERNATIONAL. 2000. Threatened birds of the world. Lynx changes. Smithsonian Institution Press, Washington, DC.
Edicions, Barcelona, Spain. LEVEY, D. J., AND F. G. STILES. 1994. Birds: Ecology, behavior, and
BLAKE, J. G. 1992. Temporal variation in point counts of birds in a taxonomic affinities, p. 217–393. In L. A. McDade, K. S. Bawa,
lowland wet forest in Costa Rica. Condor 94:265–275. H. A. Hespenheide and G. S. Hartshorn [EDS.], La Selva: Ecology
 METHODS FOR CANOPY BIRDS IN TROPICAL FORESTS 237

and natural history of a neotropical rain forest. University of Chi- R ALPH, C. J., J. R. SAUER, R. JOHN, AND S. DROEGE [EDS.]. 1995.
cago Press, Chicago. Monitoring bird populations by point counts. U.S. Forest Ser-
LOISELLE, B. A. 1987. Migrant abundance in a Costa Rican lowland vice, U.S. Department of Agriculture, Albany, CA.
forest canopy. Journal of Tropical Ecology 3:163–168. ROBINSON, W. D. 1999. Long-term changes in the avifauna of Barro
LOISELLE, B. A. 1988. Bird abundance and seasonality in a Costa Colorado Island, Panama, a tropical forest isolate. Conservation
Rican lowland forest canopy. Condor 90:761–772. Biology 13:85–97.
LOWMAN, M. D., AND H. B. R INKER. 2004. Forest canopies. 2nd ed. ROBINSON, W. D., J. D. BRAWN, AND S. K. ROBINSON. 2000. Forest
Elsevier, Amsterdam. bird community structure in central Panama: influence of spatial
M AGURRAN, A. E. 2004. Measuring biological diversity. Blackwell, scale and biogeography. Ecological Monographs 70:209–235.
Manton, MA. SILLETT, S. C., AND R. VAN PELT. 2000. A redwood tree whose crown
MILLENNIUM ECOSYSTEM ASSESSMENT. 2005. Ecosystems and human is a forest canopy. Northwest Science 74:34–43.
well-being: biodiversity synthesis. World Resources Institute, SILVA, W. R., P. DE M ARCO, JR., E. H ASUIL, AND V. S. M. GOMES.
Washington, DC. 2002. Patterns of fruit–frugivore interactions in two Atlantic for-
NADKARNI, N. M., AND T. J. M ATELSON. 1989. Bird use of epiphyte est bird communities of south-eastern Brazil: implications for
resources in neotropical trees. Condor 91:891–907. conservation, p. 423–435. In D. J. Levey, W. R. Silva and M. Gal-
NAKA, L. N. 2004. Structure and organization of canopy bird assem- etti [EDS.], Seed dispersal and frugivory: ecology, evolution and
blages in central Amazonia. Auk 121:88–102. conservation. CABI Publishing, New York.
NEU, C. W., C. R. BYERS, AND J. M. PEEK. 1974. A technique for anal- SODHI, N. S., L. H. LIOW, AND F. A. BAZZAZ. 2004. Avian extinctions
ysis of utilization-availability data. Journal of Wildlife Manage- from tropical and subtropical forests. Annual Review of Ecologi-
ment 38:541–545. cal and Evolutionary Systematics 35:323–345.
PACIFICI, K., T. R. SIMONS, AND K. H. POLLOCK. 2008. Effects of veg- STOTZ, D. F., J. W. FITZPATRICK, T. A. PARKER III, AND D. K. MOSKO-
etation and background noise on the detection process in audi- VITS. 1996. Neotropical birds: ecology and conservation. Univer-
tory avian point-count surveys. Auk 125:600–607. sity of Chicago Press, Chicago.
PITMAN, N. C. A., J. W. TERBORGH, M. R. SILMAN, P. NÚÑEZ V., D. TERBORGH, J., S. K. ROBINSON, T. A. PARKER III, C. A. MUNN, AND
A. NIELL, C. E. CERÓN, W. A. PALACIOS, AND M. AULESTIA. 2001. N. PIERPONT. 1990. Structure and organization of an Amazonian
Dominance and distribution of tree species in upper Amazonian forest bird community. Ecological Monographs 60:213–238.
terra firme forests. Ecology 82:2101–2117. TERBORGH, J., AND B. WINTER. 1980. Some causes of extinction,
POWELL, G. V. N., AND R. BJORK. 1995. Implications of intratropical p. 119–133. In M. E. Soulé and B. A. Wilcox [EDS.], Conservation
migration on reserve design—a case study using Pharomachrus biology: an evolutionary-ecological perspective. Sinauer Associ-
mocinno. Conservation Biology 9:354–362. ates, Sunderland, MA.

A PPENDIX 1. Common and scientific names of birds referenced in the text and tables.

Scientific name Common name Scientific name Common name

Penelope purpurascens Crested Guan Cotinga nattererii Blue Cotinga

Leucopternis albicollis White Hawk Vireo flavifrons Yellow-throated Vireo
Falco rufigularis Bat Falcon Vireo philadelphicus Philadelphia Vireo
Patagioenas nigrirostris Short-billed Pigeon Vireo olivaceus Red-eyed Vireo
Aratinga nana Olive-throated Parakeet Hylophilus decurtatus Lesser Greenlet
Pyrilia haematotis Brown-hooded Parrot Vireolanius pulchellus Green Shrike-Vireo
Piaya cayana Squirrel Cuckoo Cyanocorax morio Brown Jay
Florisuga mellivora White-necked Jacobin Thryothorus maculipectus Spot-breasted Wren
Thalurania colombica Violet-crowned Woodnymph Polioptila plumbea Tropical Gnatcatcher
Amazilia tzacatl Rufous-tailed Hummingbird Catharus ustulatus Swainson’s Thrush
Heliothryx barroti Purple-crowned Fairy Dendroica petechia Yellow Warbler
Tilmatura dupontii Sparkling-tailed Hummingbird Dendroica pensylvanica Chestnut-sided Warbler
Trogon violaceus Violaceous Trogon Dendroica magnolia Magnolia Warbler
Notharchus macrorhynchos White-necked Puffbird Dendroica virens Black-throated Warbler
Pteroglossus torquatus Collared Aracari Dendroica castanea Bay-breasted Warbler
Ramphastos sulfuratus Keel-billed Toucan Mniotilta varia Black-and-white Warbler
Melanerpes pucherani Black-cheeked Woodpecker Setophaga ruticilla American Redstart
Ornithion semiflavum Yellow-bellied Tyrannulet Piranga rubra Summer Tanager
Zimmerius vilissimus Paltry Tyrannulet Piranga olivacea Scarlet Tanager
Tolmomyias sulphurescens Yellow-olive Flycatcher Piranga leucoptera White-winged Tanager
Attila spadiceus Bright-rumped Attila Chlorophanes spiza Green Honeycreeper
Myiarchus crinitus Great Crested Flycatcher Cyanerpes lucidus Shining Honeycreeper
Megarynchus pitangua Boat-billed Flycatcher Cyanerpes cyaneus Red-legged Honeycreeper
Schiffornis turdina Thrushlike Schiffornis Caryothraustes poliogaster Black-faced Grosbeak
Pachyramphus aglaiae Rose-throated Becard Icterus galbula Baltimore Oriole
Tityra semifasciata Masked Tityra Psarocolius wagleri Chestnut-headed Oropendola
Tityra inquisitor Black-crowned Tityra Euphonia hirundinacea Yellow-throated Euphonia
Cotinga amabilis Lovely Cotinga Euphonia gouldi Olive-backed Euphonia

View publication stats