RESPONSE OF FOREST SONGBIRD COMMUNITIES TO A GRADIENT OF
OVERSTORY RETENTION IN NORTHEASTERN ALABAMA
by
ADRIAN A. LESAK
A THESIS
Submitted in partial fulfillment of the requirements
for the degree of Master of Science
in the Department of Natural Resources and Environmental Science
in the School of Graduate Studies
Alabama A & M University
Normal, AL 35762
MAY 2004
CHAPTER 1
INTRODUCTION
The conservation of forest songbirds has elevated to an issue of international
importance (Martin and Finch 1995, Rappole 1995). Neotropical migrants in particular
have become the subject of intense ecological research on their breeding, stopover, and
wintering grounds (e.g. Askins et al. 1990, Hagan and Johnston 1992, Finch and Stangel
1993, Yong et al. 1998). For many species, these areas encompass several countries and
land-use regimes, and require private, governmental, and academic cooperation across
borders to approach their management and preservation comprehensively.
The study of the consequences of habitat alteration is invaluable to the field of
avian ecology, and a greater understanding is essential for the conservation of birds. The
value of research in this area is evident in its potential function as a tool for land
managers charged with applying appropriate forest management plans or wildlife
conservation measures in a variety of contexts. Habitat loss to alternative land uses and
fragmentation caused by natural resources management are major factors influencing the
populations of many species of songbirds (e.g. Robbins et al. 1989, Askins 1993,
Robinson et al. 1995). Migrants may encounter these situations in their breeding habitats,
on their wintering grounds, as well as at crucial migration stopovers. In many cases,
1
changes in one or more of these settings may have a limiting influence on the populations
of a species (Sherry and Holmes 1995).
Research has solidified the notion that some species of songbirds in forested
landscapes have shown substantial population declines in recent decades (Ambuel and
Temple 1982, Robbins et al. 1989, Askins et al. 1990). Anthropogenic habitat alterations
in the form of agriculture, residential and commercial development, and timber
harvesting, have been widely suspected as causal agents in these declines. As a result of
changes in land use and demands for resources from our forested ecosystems, songbird
communities within these systems have changed in complex ways. As interest and
awareness of these trends have grown, they have fueled changes in forest management
practices and continued research in avian ecology and conservation.
A thorough understanding of the breeding habitat of a bird species or community
can be very useful. The pressures of obtaining a mate and raising young, compounded by
resource competition, greatly increase the physiological stress on birds during the
breeding season (Robinson 1992). Maintaining suitable habitat for birds in their breeding
range is important not only for the sustenance of the breeding adults, but these habitats
also hold the potential to produce a positive input of individuals that can sustain or
augment populations.
Sound management and silvicultural practices on forested
breeding grounds that maintain habitat integrity while providing timber can be crucial to
the preservation of viable populations of non-game landbirds.
The maintenance of bird populations is important to forest ecosystems they
inhabit. As the majority of forest songbirds are insectivorous, they provide an invaluable
service to these systems and their resources by consuming a significant number of
2
arthropods and their larvae, some of which may be detrimental to forest health (Williams
1936, Sherry and Holmes 1995, Conner et al. 1999). Other species may increase the
germination rates of fleshy-fruited plants through the physical removal of the skin and
pulp surrounding their seeds (Greenberg et al. 2001) and by gastrointestinal scarification
during digestion (Traveset et al. 2001). Many forest plants rely on songbirds to perform
necessary seed dispersal functions and in the case of hummingbirds, to act as pollinators.
Woodpeckers, nuthatches, and other bark-foraging and excavating birds can actively
contribute to the decomposition of standing dead trees and down woody debris. Birds
may also serve as indicators of ecosystem complexity as some species’ numbers or
reproductive effort change when the amount, continuity, composition, and health of their
habitats are altered (Maurer 1993, Crozier and Gawlik 2003). This, in turn, may initiate
measures necessary to protection and restoration of these key habitats. Finally, songbirds
undoubtedly have recreational and aesthetic value. Their song, appearance, and their
presence as part of a familiar woodlot or treasured, functioning forest ecosystem draws
the attention and revenue of many outdoor enthusiasts and conservation groups.
Since 1966, the Breeding Bird Survey of the USGS Wildlife Research Center has
maintained records of breeding bird populations throughout North America. The trends
for the songbird species encountered in the hardwood forests of northern Alabama are
mixed.
A number of species have declined while others have remained stable or
increased. The population trends for some species have similarities with the estimates of
their trends over their entire range, while other species appear to be opposing these trends
3
on their breeding grounds in Alabama (Sauer et al. 2001). This illustrates the complexity
of the populations and communities of this group of birds and of their conservation as a
whole.
The forests of north Alabama are largely owned by non-industrial private forest
(NIPF) landowners. Approximately 85% of the land area in forest cover belongs to this
group, while commercial and public interests combined oversee 15% (Hartsell and
Vissage 2001). Some of the largest and least fragmented tracts of mature upland forest in
the region are in a mixed matrix of state, industrial, and NIPF ownership and are very
important for sustaining bird populations. Large areas of forested land under private
ownership, as is the case in Alabama, can be instrumental to the conservation of forest
songbirds. As owners of smaller NIPF tracts are marginalized in a more competitive,
industrial timber industry geared toward large volume operations and markets, these
people look toward other values of their property and how to enhance them. One of these
values is wildlife habitat. According to local managers, this has become an increasingly
important factor driving management decisions in the region (Greg Janzen, personal
comm., Stevenson Land Company). It is possible that while managing for timber and
game, many songbirds could also benefit from sound management.
Alabama has the second largest commercial forest in the nation (Alabama
Forestry Commission 2003). As a result, the motives and operations of forestry in the
region have a significant impact on the habitat of songbirds. Foresters have conservation
and management goals for the timber resources under their stewardship; but water,
wildlife, and soil issues have increased in importance through a shift toward a more
holistic approach of ecosystem management.
4
The industry-based Sustainable Forestry Initiative of the American Forest and
Paper Association provides a measure of incentive to practice sound forest management
in the commercial sector.
This framework of objectives and evaluation requires
participating companies to “manage the quality and distribution of wildlife habitats” and
“improve the science and understanding of wildlife management, …ecosystem functions,
and the conservation of biological diversity” (Sustainable Forestry Board and American
Forest and Paper Association 2002). To meet the stipulation of “improving the science
and understanding of wildlife management,” members are required to support, conduct,
or assist with wildlife research that explores new ways to provide for both economic (e.g.
timber production) and forest ecosystem values (e.g. wildlife habitat).
Forestry practices often result in a change in density of the dominant canopy trees
in the managed stand. The overstory density of a stand (i.e. the basal area of the
dominant and co-dominant trees in the upper levels of a forest) is inextricably linked to
canopy closure. Canopy closure, in turn, dictates the amount of light penetrating to the
mid-story, shrub, and ground cover layers of the forest thus affecting the entire structure
and productivity of its vegetation. This affects foraging and nesting opportunities for
birds as well as their exposure to predators, brood parasites, and adverse weather
conditions. As the overstory is arguably the most ecologically important component of a
forest affecting songbird communities (Crawford et al. 1981) and the most commonly
altered in forest management, knowledge of the suitable overstory conditions for a
songbird species, or suite of species, could help foresters with the problem of applying
stand-level treatments to maintain songbird habitat.
5
Previous Studies
A number of earlier studies have explored the habitat associations and numerical
response of bird populations and communities to different stages of forest development
following disturbance (e.g. Odum 1950, Johnston and Odum 1956, Shugart and James
1973, Crawford et al. 1981, Thompson and Capen 1988, Keller et al. 2003). These
studies are useful in describing temporal differences in the songbird communities at
different sites in response to extensive disturbance (usually clearcutting or agriculture)
but do not describe the changes in different intensities of forest management over time.
The majority of avian research describing the direct effects of habitat alteration related to
timber harvest has compared clearcut sites to untreated stands (e.g. Conner and Adkisson
1975, Thompson and Fritzell 1990, Thompson et al. 1992, Welsh and Healy 1993, and
see Harlow et al. 1997). Conner et al. (1979) explored the effects of clearcutting on
winter bird populations, while Keller and Anderson (1992) and Germaine et al. (1997)
examined how breeding bird communities in small patch or strip clearcuts differed from
those in uncut forests. These studies focused on the songbird community response to one
level of intensity of forest management at several post-treatment intervals, in breeding
and non-breeding seasons, and at a range of spatial extents and configurations. However,
they do not demonstrate the songbird community changes associated with intermediate
levels of harvest intensity.
Few studies have concentrated on the various intensities of harvest common in
today’s silvicultural systems practiced on public and private lands (Thompson et al.
1995).
The shelterwood and seed tree methods are two examples of even-aged,
regeneration prescriptions employing partial removal of the overstory in the initial cut.
6
Single tree and group selection are uneven-aged forest regeneration techniques with
intermediate levels of initial harvest intensity that create a more diverse forest age
structure (Smith 1986). Dickson et al. (1995) noticed that bird diversity in shelterwood
and seed-tree regenerated stands can be higher as a result of the immigration of earlysuccessional species into the stands while some late-successional species remain and
utilize the residual overstory. In West Virginia, Nichols and Wood (1995) found that
two-age stands, with residual overstory and a regenerating cohort underneath, harbored
greater numbers of birds and species than clearcuts or untreated stands.
However,
another study showed that songbird communities had greater indices of diversity, species
richness, and abundance in clearcuts and high- and low-leave two-aged forests than in unharvested stands (Baker and Lacki 1997).
Gaps and inconsistencies exist in the current body of knowledge concerning the
influence of various silvicultural treatments on the songbird communities inhabiting
managed forests.
This research attempts to further investigate the alternative
regeneration strategies with intermediate harvest components in the understudied oakhickory forests of the southern Cumberland Plateau.
Objectives
The objective of this study was to explore the relationship between songbird
community structure and a gradient of stand manipulation.
Bird populations were
sampled in control (100% retention) and clearcut (0% retention) treatments along with
treatments of 25, 50, and 75% residual basal area that approximate potential leave-tree
levels associated with the even-aged, shelterwood regeneration method. I compared the
7
effects that these varying levels of overstory retention had on overall bird species
abundance, richness, diversity, and breeding success, as well as the changes in the guildlevel community infrastructure. The similarity of the community composition between
treatments was also examined.
It was hypothesized that bird community composition and structure would change
predictably with graded levels of forest canopy retention, as bird species and guilds
differentially select breeding territories on the treatments that best suit their habitat
requirements.
The intermediate disturbance hypothesis (Connell 1978) states that
community diversity will be highest in those areas with intermediate frequency or
intensity of disturbance. In accordance with this, it would be expected that the songbird
communities occupying the treatments with intermediate levels of retained overstory
would be the most diverse. The increased structural and floristic diversity of these stands
would sufficiently fulfill the habitat requirements of a more abundant and species rich
assemblage of songbirds.
To test this hypothesis, I quantified the populations of
migratory and permanently-resident breeding songbirds by measuring each species’
territorial density and detection rate (use) during the breeding season to determine the
abundance, composition, similarity, richness, and diversity of the communities. It was
expected that sometimes gradual, and at times abrupt, shifts in community composition
and abundance would occur, from species-to-species or guild-to-guild, among the five
levels of overstory retention. It was probable that patterns of species turnover would be
discovered among the different treatments for certain species and groups of species that
share similar ecological preferences and strategies.
8
A goal of this research is to equip land managers with the information necessary
to apply the appropriate measures when conservation of forest songbirds and timber
management coincide. Although the experimental shelterwood treatments used in this
study may not precisely mimic other silvicultural practices such as seed tree, group
selection, or single tree selection, they could be used as guidelines of harvest intensity
when applying these methods where songbird habitat management is considered.
Another goal of this study is to provide a concrete understanding of the community
structure of these treatments that may serve as a foundation for future research, or longterm studies of this area as forest succession and further harvest occurs on these sites.
Ultimately, I would like to find compromises in forest stand manipulation that will
incorporate the preservation of bird species diversity into the goals of natural resource
management plans.
9
CHAPTER 2
DESIGN AND METHODOLOGY
Study Area
The study area was situated in the Mid-Cumberland Plateau of the southern
Appalachian Mountains in northern Jackson County, Alabama (Figure 1). This region’s
physiography is characterized by narrow, flat plateaus dissected with numerous deep
valleys. Two sites were used, one located at Miller Mountain, Alabama [Figure 2. (34˚
58′ 30″ N, 86˚ 12′ 30″ W)] and one at Jack Gap, Alabama [Figure 3(34˚ 56′ 30″ N, 86˚
04′ 00″ W)]. The elevation range of the study sites was 260 to 520 m. Both sites were at
middle to upper slope positions on the sides of the Plateau. Jack Gap had a northern
slope aspect and Miller Mountain had generally southern to southwestern exposure.
Upland hardwood is the dominant forested land cover type in the northern half of
Jackson County, with many large continuous tracts throughout. The forests of the sites
and much of the surrounding area were composed of mature (80-100 years old) oakhickory (Quercus spp. and Carya spp.) with yellow-poplar (Liriodendron tulipifera),
sugar maple (Acer saccharum), red maple (Acer rubrum), and American beech (Fagus
grandifolia) as associates (Hartsell and Vissage 2001).
10
Location of Study Sites
^
^
JACKSON
MADISON
MORGAN
­
DE KALB
MARSHALL
ALABAMA
CULLMAN
CHEROKEE
ETOWAH
BLOUNT
CALHOUN
ST CLAIR
CLEBURNE
JEFFERSON
TALLADEGA
Figure 1. Northeastern Alabama. Study sites (stars) shown in northern Jackson County.
11
Miller Mountain - Block 1
25%
50%
clearcut
75% control
­
Figure 2. Topographic map of one complete block replicate showing treatment
assignments at Miller Mountain, Jackson County, AL.
12
Jack Gap - Blocks 2 & 3
clearcut
control
25%
75% 50%
50%
25%
control
clearcut
75%
­
Figure 3. Topographic map of two complete block replicates showing treatment
assignments at Jack Gap, Jackson County, AL.
13
Design
This experimental study consisted of three randomized complete block replicates
of five overstory retention treatment units: 0 (clearcut), 25, 50, 75, and 100 percent
(control). Jack Gap had two blocks (Figure 2) and Miller Mountain had one (Figure 3).
Each unit was roughly square in shape and the units were arranged adjacently within each
block. The treatment units were approximately four hectares (10 acres) in size making
each replicate block 20 hectares, and thus, a total study area of 60 hectares.
The three treatments of intermediate harvest intensity were established as
shelterwood stands to examine the effects of different overstory conditions on the
regeneration of desired tree species composition (Schweitzer 2004). These treatments
could then be compared along with the clearcuts and controls before and after the release
and subsequent removal cut.
The clearcut, 25, and 50 percent treatments were accomplished by conventional
chainsaw felling and skidding (December 2001-March 2002). The marking of leave trees
in these stands favored oak (Quercus spp.), ash (Fraxinus spp.), and other species of high
vigor with dominant crowns and boles of merchantable diameter. Common persimmon
(Diospyros virginiana) was left as a species valuable to wildlife. In the 75% retention
units, an herbicide (Arsenal®, active ingredient: imazapyr) was applied mainly to the
small-diameter midstory trees to reduce competition and increase light intensity for oak
regeneration without creating large overstory gaps (November 2001). The majority of
trees culled from the midstory were red maple (Acer rubrun), blackgum (Nyssa
sylvatica), sugar maple (Acer saccharum), hickory (Carya spp.), and black locust
(Robinia pseudoacacia).
14
Each treatment unit had three bird survey transects spaced evenly across its width
and parallel with the slope. Along each transect, marked reference points were placed at
25-m intervals to facilitate bird territory mapping. To adequately sample the entire
treatment unit during spot-mapping, the distance between transects was ≤50 m, as was
the distance from the outer transects to the parallel unit boundaries.
Overstory Measurements
Pre-treatment basal area was derived in each treatment unit from five
systematically arranged 0.08-ha circular plots. Measurements within each plot from all
overstory trees ≥14.2 cm diameter at breast height (1.4 m) were used to calculate initial
basal area. Following treatment, diameter measurements of the remaining trees were
used to determine residual basal area and the percentage of overstory retention.
The same five plots within each treatment unit were used to measure the
percentage of canopy cover following overstory removal. Five readings from a handheld
spherical densitometer were taken at breast height; one at plot center and one taken 3.6 m
from plot center in each of the cardinal directions. The average of these readings was
used to quantify the canopy cover for each plot.
Songbird Community Assessment
Spot-Mapping
Territory spot-mapping was used to determine each songbird species’ territorial
density and detection rate based on the methods first described by Williams (1936) and
outlined by the International Bird Census Committee (1970) and Ralph et al. (1993).
From these data, the composition, species and guild breeding density and use, species
15
richness, community diversity, evenness, and similarity can be calculated and compared
for the bird communities within each treatment.
Most forest songbirds vigorously defend nesting and foraging territories during
the breeding season and display territorial behavior that is often regular and observable.
Recording this behavior and its position on a map, as well as its relative position to the
territorial behavior of simultaneously-occurring conspecifics is the basis of the spotmapping procedure. The use of the spot-mapping technique to estimate the number of
breeding bird territories was appropriate for this study for the following reasons:
1.
The manageable treatment size (4 ha) used in this study made the intensive effort
involved in accurate spot-mapping feasible.
2.
The treatment units had equal areas. This helps to eliminate area-dependent
variation in the bird community.
3.
All of the treatment units were roughly square in shape and therefore had
comparatively short total edge lengths, thus reducing edge effects on the bird
communities. This fact, coupled with the equal area of the treatments gave the
units similar boundary lengths and relative edge areas, diminishing this
confounding influence among and within treatments. Furthermore, the straight
unit boundaries of the square units simplified the interpretation of territories
straddling the edges.
4.
Intensive spot-mapping surveys can yield data on breeding success.
16
5.
One person conducted all aspects of the spot-mapping duties.
This reduced
inter-observer bias and differences in map interpretation, while at the same time
increasing the familiarity and recent knowledge of bird activity on the treatments.
Each of the 15 treatment units received 10 spot-mapping visits between late April
and July 2002 and again in 2003. Ralph et al. (1993) suggested this amount of sampling
effort was sufficient to obtain the data necessary for the interpretation of mapped
territories in forested habitats. One block of five units was visited each morning. One
rotation of visits through all of the blocks was completed before moving on to the next
visit. The order of visits to the 3 blocks was randomized for each of the 10 rotations
leaving 5 days between visits to the same block. Morning mapping surveys began
between 05:00 and 05:30 depending on light conditions that varied with sunrise time and
cloud cover. All surveys were completed by approximately 10:30. Within each block, the
treatment unit visits were ordered systematically ensuring that every unit received equal
sampling at all times of the morning survey period. Each treatment unit of the block
received one hour of surveying per visit. This gave every unit one 10-hour spot-mapping
sample over the course of the breeding season in which to develop the certainty of
territory occupation. In addition, each of the one-hour visits was used to estimate the
average detection rate for all species within each treatment over the 10 visits. This data
supplemented the territory mapping results and provided greater insight into the range of
treatment use by all species present.
While traveling along the transects of a unit during a spot-mapping visit, the
species and position of adult singing males were recorded on a topographic map (scale
17
1:1400).
In addition to these contacts, nest locations and behaviors (intraspecific
aggression, pair feeding, copulations, nest-building, distraction displays, food-carrying,
fecal sac disposal, etc.) by either sex that suggested an active territory were also noted. At
the end of a survey day, data recorded on the visit maps were transferred to separate
species maps for each unit. The collection of registrations for the 10 visits of a unit on
each species map was used to delineate territorial activity centers. Territory density per 4
ha was assigned for each species based on the number of interpreted territorial clusters (to
the nearest half-territory) in each treatment unit.
Spot Map Interpretation
Using 10 valid visits for territory interpretation, at least three registrations of a
particular species within an acceptable territory radius were required to represent a
territorial cluster. Two of these records must be at least 10 days apart and at least two
must also signify high territorial significance such as singing (International Bird Census
Committee 1970).
The interpretations of territorial clusters within one unit were
evaluated simultaneously with those in the units adjacent to it. Territories on the edge of
these treatment units were subject to the same requirements as those completely enclosed
within the unit and were divided into half-territories if roughly half of the registrations
occurred in each of the adjoining units. Within a treatment unit, interpretation of adjacent
territory clusters of the same species was dependent on the satisfaction of one of three
requirements: 1) at least one simultaneous registration supported by additional
observations; 2) at least two pairs of non-simultaneous registrations; or 3) at least one
pair of non-simultaneous registrations and knowledge of territory size for species with
18
low detection probabilities or few valid visits (late breeders). A few exceptions to these
minimum requirements were accepted solely on the basis of strong evidence of breeding
(e.g. nest, nest building, food carrying).
19
Guilds
The breeding songbird communities in this study were categorized into guilds in
four ways (Table 1). In the first of these categorizations, species were classified into
three migrant guilds based on summarized distribution maps from the National Audubon
Society Christmas Bird Count (Sauer et al. 1996) and records from Imhof (1976). The
Neotropical migrant guild included those species that migrate to Central and South
America in the fall. Temperate migrants were defined as species with at least part of
their winter range in the southern United States and northern Caribbean islands. The last
migrant guild included the non-migratory permanent resident species.
Based on Ehrlich et al. (1986), the songbird communities were grouped into
guilds according to their nesting location.
There were five major nesting guilds
represented: ground, low-shrub, midstory-subcanopy, canopy, and cavity nesters. In
addition, the Brown-headed Cowbird (Molothrus ater) was the sole member of the brood
parasite guild.
Four foraging guilds were present in this study. Species were classified into them
according to their primary foraging behavior and substrate (Ehrlich et al. 1986). Foliage
gleaners take mostly insect prey from the leaves, flowers, buds, and small stems and
twigs of vegetation. Ground foraging birds glean food from the soil, leaf litter, and lowgrowing vegetation. Bark gleaners drill, flake, peck, and pry prey from the bark and
wood of the trunks and limbs of trees. Hawking foragers mainly sally from perches or
drop onto their prey but may actively pursue insects in the air.
The final three guilds categorized species based on their primary habitat
association with reference to Blake and Karr (1987) and Freemark and Collins (1992)
20
with some modifications derived from this study. The open-edge guild was made up of
songbird species that prefer forest openings, fields, early-successional scrub, and forest
edge habitats. Interior-edge species occurred in forested habitats but are less sensitive to
forest area or disturbance and occupy edge habitats as well. Forest interior species nest in
mature closed canopy forests and often have minimum forest area requirements.
Table 1. Guild memberships of all forest songbird species encountered on the study sites
classified by: fall migratory destination (N, Neotropical migrant; T, temperate
migrant; R, resident), nest location (G, ground; L/S, low-shrub; M/S, midstorysubcanopy; CN, canopy; CV, cavity), foraging strategy (F, foliage glean; G,
ground glean; B, bark glean; H, hawking), and habitat association (O/E, openedge; I/E, interior-edge; I, interior).
Migrant
Guild
Nest
Location
Guild
Foraging
Guild
Habitat
Guild
Acadian Flycatcher, Empidonax virescens
N
M/S
H
I
American Goldfinch, Carduelis tristis
R
L/S
F
O/E
American Redstart, Setophaga ruticilla
T
M/S
H
I
American Robin, Turdus migratorius
R
M/S
G
O/E
Black-and-white Warbler, Mniotilta varia
T
G
B
I
Blue-gray Gnatcatcher, Polioptila caerulea
T
M/S
F
I/E
Brown-headed Cowbird, Molothrus ater
R
-
G
O/E
Blue Grosbeak, Passerina caerulea
N
L/S
G
O/E
Blue Jay, Cyanocitta cristata
R
CN
G
I/E
Blue-winged Warbler, Vermivora pinus
N
G
F
O/E
Carolina Chickadee, Poecile carolinensis
R
CV
F
I/E
Carolina Wren, Thryothorus ludovicianus
R
L/S
G
O/E
Species
21
Table 1. (continued).
Migrant
Guild
Nest
Location
Guild
Foraging
Guild
Habitat
Guild
Cerulean Warbler, Dendroica cerulea
N
CN
F
I
Chipping Sparrow, Spizella passerina
R
L/S
G
O/E
Common Yellowthroat, Geothlypis trichas
T
L/S
F
I/E
Downy Woodpecker, Picoides pubescens
R
CV
B
I/E
Eastern Bluebird, Sialia sialis
R
CV
H
O/E
Eastern Phoebe, Sayornis phoebe
R
-
H
I/E
Eastern Towhee, Pipilo erythrophthalmus
R
G
G
I/E
Eastern Wood-Pewee, Contopus virens
N
CN
H
I/E
Field Sparrow, Spizella pusilla
R
G
G
O/E
Great Crested Flycatcher, Myiarchus crinitus
N
CV
H
I/E
Gray Catbird, Dumetella carolinensis
T
L/S
G
I/E
Hairy Woodpecker, Picoides villosus
R
CV
B
I
Hooded Warbler, Wilsonia citrina
N
L/S
F
I
Indigo Bunting, Passerina cyanea
N
L/S
F
O/E
Kentucky Warbler, Oporornis formosus
N
G
G
I/E
Mourning Dove, Zenaida macroura
R
L/S
G
O/E
Northern Cardinal, Cardinalis cardinalis
R
L/S
G
I/E
Ovenbird, Seiurus aurocapilla
N
G
G
I
Pileated Woodpecker, Dryocopus pileatus
R
CV
B
I
Prairie Warbler, Dendroica discolor
T
L/S
F
O/E
Red-bellied Woodpecker, Melanerpes carolinus
R
CV
B
I/E
Species
22
Table 1. (continued).
Migrant
Guild
Nest
Location
Guild
Foraging
Guild
Habitat
Guild
Red-eyed Vireo, Vireo olivaceus
N
M/S
F
I/E
Red-headed Woodpecker, Melanerpes erythrocephalus
R
CV
B
I/E
Ruby-throated Hummingbird, Archilochus colubris
N
M/S
-
O/E
Scarlet Tanager, Piranga olivacea
N
CN
F
I
Summer Tanager, Piranga rubra
N
M/S
F
I/E
Tufted Titmouse, Baeolophus bicolor
R
CV
F
I/E
White-breasted Nuthatch, Sitta carolinensis
R
CV
B
I
White-eyed Vireo, Vireo griseus
T
L/S
F
O/E
Worm-eating Warbler, Helmitheros vermivorus
N
G
F
I
Wood Thrush, Hylocichla mustelina
N
M/S
G
I/E
Yellow-breasted Chat, Icteria virens
N
L/S
F
O/E
Yellow-billed Cuckoo, Coccyzus americanus
N
M/S
F
I/E
Yellow-throated Vireo, Vireo flavifrons
N
CN
F
I/E
Yellow-throated Warbler, Dendroica dominica
N
CN
B
I
Species
Breeding Success by Territory Rank
Measures of community indices such as density, richness, diversity, and similarity
are useful in describing the characteristics of a suite of birds attempting to breed in a
specific location. However, without a measure of breeding success, conclusions drawn
about the habitat suitability of different treatments carry less interpretive power (Van
23
Horne 1983). The spot-mapping method yields many behavioral observations that have
two aims: one is to improve the certainty of territory occupation; the second is a byproduct of the first, in that these behaviors may also serve as assessments of the breeding
success of the birds on that territory. In an attempt to capture this information on
breeding success, nest searching and monitoring was carried out opportunistically during
mapping visits in the 2003 census period. In addition, three 2-hour visits during the
breeding season were dedicated to finding new nests and checking the status of
established nests on all treatments. From this nest data and the breeding behavior and
fledgling observations recorded during spot-mapping, breeding territory ranks were
assigned to every interpreted territory of all species to approximate overall breeding
success in each treatment. The eight breeding ranks used were based on a hierarchy of
phases of the passerine reproductive cycle in a fashion similar to Vickery et al. (1992)
and those used by some breeding bird atlas projects.
Rank 1 constituted a territorial male present or territories with nests known to
have been parasitized by the Brown-headed Cowbird. Rank 2 included territories where
a male and female were present on the territory. This rank was restricted to the territories
of sexually dimorphic species or those in which the sexes are similar but only males are
known to sing. Territories with occupants that showed behavior commonly associated
with breeding (e.g. predator anxiety, courtship behavior, copulation, probable nest site
visits) received rank 3. Territories where the occupant pair was observed carrying nest
material or nest building (not wrens or woodpeckers) were classified as rank 4. Rank 5
territories included those where food-carrying to young or fecal sac disposal were
observed. Territories with incubating females on the nest were considered rank 6. Rank
24
7 territories included nests with nestlings of the occupant species and rank 8 was assigned
to those with dependent fledglings on the territory.
There are assumptions that may influence the comparison of overall breeding
success among treatments using this ranking scheme. First, the probability of success is
assumed to be equal among different species. This is not likely to be true among the
many species encountered on our study sites. Another assumption is that the probability
of recording a particular rank should be equal across treatments and species. This is
unlikely due to the relative ease of locating nests and observing the behavior of an open
nesting species such as the Indigo Bunting (Passerina cyanea) versus the difficulty of
detecting Scarlet Tanagers (Piranga olivacea) quietly delivering food to a nest high in the
forest canopy. A third assumption is that rates of brood parasitism are equal across
species and treatments and do not contribute to an inflated number of “rank 1” territories
under some overstory conditions. Despite these potential biases, careful observation and
interpretation of behaviors in the field may provide the power to detect coarse differences
among the different overstory treatments.
Statistical Analysis
Provided the necessary assumptions of normality and homogeneity of variance
were reasonably met (Shapiro-Wilk p>0.05 and Levene p>0.05, respectively), parametric
data from 2002 and 2003 were analyzed simultaneously using repeated-measures
ANOVA with year as the within-subjects factor. Between-subject factors were block and
treatment effects.
The year-treatment interaction was used to determine if the bird
communities reacted differently to the treatments between years one and two. Because of
25
low sample size and a desire to interpret possible treatment effects, this interaction was
deemed significant if p<0.10. If no significant year-treatment interactions were detected,
data from both years were examined together for differences (p < 0.05). When songbird
response to the treatments differed significantly between years, each year was analyzed
separately using one-way ANOVA (p<0.05) with block effect taken into account.
Measurements analyzed from both years included the mean values of territory density,
detection rate, species richness, Brillouin and Shannon diversity, and the pooled mean
territory density and detection rates of all guild categories. Pre-treatment (2001) and
post-treatment (2002) basal area and canopy cover were analyzed using one-way
ANOVA. Percentages of canopy cover retained were arcsine transformed to improve the
normality of this data. Means for all measurements were separated using Tukey’s HSD
test with a 5% significance level. Trend analysis showed the functional relationship
between the gradient of overstory retention and the bird community indices and guild
data. The Kruskal-Wallis test was used to analyze the non-parametric territory rank data.
Analyses were performed using SPSS version 10.0 (SPSS Inc. 1999). Significant results
are reported at the level of p<0.05.
Because the spot-mapping method is assumed to be a non-random sample of the
entire community of a study site, the Brillouin diversity index (Equation 1) using a
logarithm with base 10 was used to describe the songbird communities on each of our
treatments based on territory density (from Zar 1996).
H = ( log n! - ∑log fi! ) / n
(1)
26
In this equation, n is the total number of territories in the treatment unit and fi is the
number of territories held by species i.
Equation (2) was used as the denominator for the evenness component (Equation
3) of the Brillouin diversity measure.
Hmax = ( log n! - (k-d)log c! - dlog (c+1)! ) / n
(2)
Where k equals the number of species observed and c equals the integer of n/k and d is
the remainder of n/k. From this evenness is calculated.
J = H/Hmax
(3)
The Shannon diversity index was calculated for breeding songbird detection rate
on our treatments using a logarithm with base 10 (Equation 4). Shannon diversity is a
descriptive index, which in this case, takes into account the bird abundance distributed
across the species richness of a particular treatment type. This index was used because
the detection rate measures the abundance of species that not only use the treatments for
nesting but also for other uses such as foraging. Therefore the detection rate is a sample
of the entire suite of species found on the site throughout the census periods.
k
H′ = ( n log n - ∑ fi log fi ) / n
(4)
i=1
27
Where n is the total detection rate for the treatment, k is the number of species detected
and fi is the detection rate for species i (from Zar 1996).
Using equation (5), evenness can be calculated to assess the influence that the
equitability of abundance across species has on the diversity measure.
H′max = log k
(5)
Evenness is then calculated with equation (6).
J′ = H′/H′max
(6)
Percent similarity (Equation 7) was calculated for all pairs of treatments using
mean territory density and detection rate to find the common abundance of shared species
(from Krebs 1998).
P = ∑ minimum of ( p1i, p2i )
i
(7)
For this equation, p1i is the abundance of shared species i in the first treatment of the pair
and p2i is the abundance of the same species for the other treatment.
28
CHAPTER 3
RESULTS AND DISCUSSION
Overstory Conditions
The average basal area of overstory trees in the five treatments was not different
prior to removal. However, post-treatment basal area was different among treatments
(F4,8 = 42.54, p<0.001). The control and 75% treatments had significantly higher residual
basal area than the other three treatments (Table 2). The basal area in the clearcut units
was significantly lower than that in the 50% plots while the 25% treatment was
intermediate between the two. Residual basal area showed a significant linear trend
(p<0.0005), increasing from the clearcut to control treatments. The percentage of basal
area retention showed that the applied treatments closely approximated the desired
retention levels, with the largest departure from the target in the 50% treatments.
29
Table 2.
Average basal area (n=3) for pre- and post-treatment years (2001 and 2002,
respectively) and percentage of basal area and canopy cover retained by
treatment.
Overstory
treatment
Pre-treatment
(m²/ha)
Basal Area
Post-treatment
(m²/ha)
0% (clearcut)
24.68aa
1.15a
25%
22.21a
50%
% retained
Canopy cover
% retained
4.7
31ab
6.14ab
27.6
76b
23.92a
9.05b
37.8
75b
75% (herbicide)
26.39a
18.47c
70.0
98c
100% (control)
23.90a
23.83c
99.7
100c
a
Means in the same column followed by different letters are significantly different, (p<0.05)
Means of percent canopy cover retained were arcsine transformed before analysis
b
Three significantly different levels of canopy cover (F4,8 = 19.52, p<0.001) were
evident in our stands following treatment (Table 2). The control and 75% sites had the
highest remaining cover, the 50 and 25 % were intermediate, and the clearcuts had the
lowest percentage of cover.
Trend analysis showed percent residual canopy cover
increasing linearly from clearcut to control (p<0.0005).
The pattern displayed by residual basal area in our overstory retention treatments
was closely reflected in that of canopy cover. These trends illustrate the gradient of
overstory retention across the treatments. This gradient had three levels that progressed
from low retention in the clearcuts for both measurements, to intermediate in the 25 and
50% treatments, and high in the 75% and controls. The desired results in the 75%
treatments were achieved by reducing an average of 30% of the midstory while
effectively maintaining nearly complete canopy cover. These three levels of overstory
30
create four distinct habitat types: open-scrub (clearcuts), open forest (25 and 50%), closed
canopy forest with a standing dead midstory (75%), and closed canopy forest with an
intact midstory.
Songbird Community
Abundance
Territory density. Over the two years of the study, 35 species held territories on the study
sites.
In 2002 there were 31 species on 271 territories and the average density of
territories for all species across all treatments was 18.1. In 2003, 33 species held 359
territories with an average of 23.9 territories per treatment. Average densities for all
territorial species are presented in Table 3. These data were used to derive community
indices for the territorial species in each treatment.
31
Table 3. Mean territory density of breeding bird species by overstory treatment in 2002
and 2003.
Species
Mean Territory Density (territories/4 ha; n=3)
2002
2003
Treatment (% overstory
Treatment (% overstory
retention)
retention)
0
25
50
75
100
0
25
50
75
100
Acadian Flycatcher
Empidonax virescens
-
-
-
0.3
0.3
-
-
-
0.3
0.3
Black-and-white Warbler
Mniotilta varia
-
0.3
-
0.3
0.3
0.3
-
-
-
0.3
Blue-gray Gnatcatcher
Polioptila caerulea
-
1.2
1.7
1.0
0.7
1.7
2.3
3.3
1.0
1.3
Blue Grosbeak
Passerina caerulea
-
0.2
-
-
-
0.3
-
-
-
-
Blue Jay
Cyanocitta cristata
-
-
-
0.2
0.3
-
-
-
0.7
0.2
Carolina Chickadee
Poecile carolinensis
-
0.7
0.7
1.0
0.5
0.7
0.3
0.7
0.3
0.5
1.0
1.3
2.2
0.7
0.8
1.7
1.0
2.0
0.2
-
Common Yellowthroat
Geothlypis trichas
-
-
-
-
-
0.3
0.3
-
-
-
Downy Woodpecker
Picoides pubescens
-
1.2
0.8
1.5
0.3
-
0.3
0.7
0.7
0.3
Eastern Bluebird
Sialia sialis
-
-
-
-
-
-
0.3
-
-
-
Eastern Phoebe
Sayornis phoebe
-
0.3
-
-
-
-
-
-
-
-
Eastern Towhee
Pipilo erythrophthalmus
0.5
1.0
1.5
0.5
-
1.7
1.7
2.0
0.7
-
Eastern Wood-Pewee
Contopus virens
0.5
2.0
1.7
0.7
0.3
0.7
2.0
1.3
1.7
0.7
Carolina Wren
Thryothorus ludovicianus
32
Table 3. (continued).
Species
Mean Territory Density (territories/4 ha; n=3)
2002
2003
Treatment (% overstory
Treatment (% overstory
retention)
retention)
0
25
50
75
100
0
25
50
75
100
Field Sparrow
Spizella pusilla
-
-
-
-
-
0.3
-
-
-
-
Great Crested Flycatcher
Myiarchus crinitus
-
0.5
0.7
-
0.3
-
0.3
0.3
0.3
0.3
Hairy Woodpecker
Picoides villosus
-
0.5
0.3
-
0.2
-
0.7
0.3
0.3
-
Hooded Warbler
Wilsonia citrina
-
-
-
-
0.5
-
-
-
-
-
Indigo Bunting
Passerina cyanea
2.0
2.8
3.0
1.0
-
5.7
6.3
4.7
0.7
0.7
Kentucky Warbler
Oporornis formosus
-
0.3
0.5
-
-
0.7
1.3
1.2
0.3
-
Northern Cardinal
Cardinalis cardinalis
0.3
0.3
1.7
0.5
0.3
1.3
1.0
1.3
-
0.7
Ovenbird
Seiurus aurocapilla
-
-
-
0.3
1.0
-
-
-
-
0.3
Prairie Warbler
Dendroica discolor
-
-
-
-
-
0.3
0.3
-
-
-
Red-bellied Woodpecker
Melanerpes carolinus
-
0.5
0.3
0.5
0.3
0.3
-
-
0.3
0.3
Red-eyed Vireo
Vireo olivaceus
-
1.8
2.5
3.3
3.3
0.2
1.7
2.3
3.3
3.0
-
0.3
-
-
0.3
0.7
0.7
0.7
-
-
-
0.5
1.7
1.7
1.5
-
0.7
1.5
2.3
1.3
Ruby-throated
Hummingbird
Archilochus colubris
Scarlet Tanager
Piranga olivacea
33
Table 3. (continued).
Species
Mean Territory Density (territories/4 ha; n=3)
2002
2003
Treatment (% overstory
Treatment (% overstory
retention)
retention)
0
25
50
75
100
0
25
50
75
100
Summer Tanager
Piranga rubra
0.8
1.7
1.5
1.2
0.5
0.8
1.3
2.0
0.7
0.2
Tufted Titmouse
Baeolophus bicolor
0.2
1.2
1.8
1.7
2.0
0.7
1.7
1.0
1.7
1.7
White-breasted Nuthatch
Sitta carolinensis
-
0.5
0.7
1.0
0.8
-
0.7
1.0
1.0
1.0
White-eyed Vireo
Vireo griseus
-
-
0.2
-
-
0.3
0.3
-
-
-
Worm-eating Warbler
Helmitheros vermivorus
-
0.3
0.7
1.3
1.5
-
0.7
0.3
0.7
1.7
Wood Thrush
Hylocichla mustelina
-
-
-
-
0.7
-
-
-
-
1.2
Yellow-breasted Chat
Icteria virens
0.7
0.2
-
-
-
3.7
4.0
2.0
-
-
Yellow-billed Cuckoo
Coccyzus americanus
-
0.3
0.2
0.7
0.5
-
0.7
0.3
0.2
0.3
Yellow-throated Vireo
Vireo flavifrons
-
0.7
1.0
1.3
0.3
-
1.3
1.0
1.3
0.3
34
Pooling the interpreted territories for all species yielded a total territory density
for each treatment.
Repeated Measures ANOVA showed a significant interaction
between year and treatment effect (F4,8=8.972,
p=0.005), suggesting the pattern of
territory density related to the treatments shifted between years. When both years were
analyzed separately, territory density was greatest in the treatments with intermediate
harvest intensity [25, 50, and 75%, (F4,8=8.472, p=0.006)] in 2002 (Figure 4). Of the
three treatments with highest territorial abundance, the 50% units had the highest density
and the clearcuts had dramatically fewer territories than all but the control plots. Trend
analysis showed that total territory density had a quadratic relationship (p<0.001) with
respect to overstory retention. Average density was at least three times higher in the
control and partial retention treatments than in the clearcut plots.
35
MEAN TERRITORY DENSITY (terr./4ha)
50
40
c
bc
b
30
ab
ab b
b a
20
10
a
a
YEAR
2002
2003
0
0%
25%
50%
75%
100%
TREATMENT
Figure 4. Average breeding songbird territory density (n=3) of five overstory retention
treatments in 2002 and 2003. Different letters above boxes of the same color
(year) indicate significantly different means (p < 0.05). Box-plots show
interquartile range (box), median (line), and the highest and lowest values
(whiskers), excluding outliers.
The seven most abundant species in 2002 held 54 percent of the territories
encountered (Table 4). The most common breeding species was the Red-eyed Vireo
(Vireo olivaceus) which was the most abundant bird on the 100% (control) and 75%
(herbicide) treatments. The Indigo Bunting (Passerina cyanea), an edge species, was the
second most common songbird and the most abundant species on the 50%, 25%, and
clearcut treatments. Four of these seven species were interior-edge habitat associates
common in the closed canopy and open forested sites. The most common forest interior
species was the Scarlet Tanager (Piranga olivacea).
36
Table 4. Total number of territories, mean density per 4 ha(a), and relative importance (b)
of seven species of songbirds accounting for 54 percent of mapped territories in
five overstory retention treatments in northeastern Alabama (2002).
Mean Territory Density (n=3) and Importance
Treatment
Total
Territories
0%
25%
50%
75%
100%
33
-
1.8 a (3) b
2.5 (2)
3.3 (1)
3.3 (1)
Indigo Bunting
Passerina cyanea
26.5
2.0 (1)
2.8 (1)
3.0 (1)
1.0 (5)
-
Tufted Titmouse
Baeolophus bicolor
20.5
0.2 (7)
1.2 (6)
1.8 (4)
1.7 (2)
2.0 (2)
Carolina Wren
Thryothorus ludovicianus
18
1.0 (2)
1.3 (5)
2.2 (3)
0.7 (6)
0.8 (5)
Summer Tanager
Piranga rubra
17
0.8 (3)
1.7 (4)
1.5 (6)
1.2 (4)
0.5 (7)
Scarlet Tanager
Piranga olivacea
16
-
0.5 (9)
1.7 (5)
1.7 (2)
1.5 (3)
15.5
0.5 (5)
2.0 (2)
1.7 (5)
0.7 (6)
0.3 (8)
Species
Red-eyed Vireo
Vireo olivaceus
Eastern Wood-Pewee
Contopus virens
In 2003, songbird territory abundance was higher in the open forest treatments (25
and 50%) than the closed canopy 75 and 100% sites (F4,8=15.728, p=0.001). The
clearcut treatments increased in abundance between the first and second year, likely due
to vegetative growth on the sites after the first growing season.
Average songbird
territory density was higher in the clearcuts than the closed canopy sites, but this trend
was not significant. The clearcuts had less abundant songbird communities than the 25%
treatments but were not significantly different from the 50%. The most common bird
species in closed canopy sites was again the Red-eyed Vireo; however the Indigo Bunting
37
population nearly doubled in abundance and was the most common species overall.
Indigo Buntings still favored the open-scrub and open-forest sites, but occasionally used
tree-fall gaps in the herbicide (75%) treatments (Table 5). The Yellow-breasted Chat
(Icteria virens) and Eastern Towhee (Pipilo erythrophthalmus) became common due to
increases in abundance on open sites. The Yellow-breasted Chat in particular, responded
appreciably in the second year with 10 territories in 2003 compared to 2.5 in 2002. Other
notable changes on the open canopy and clearcut sites between years include increases in
Kentucky
Warblers
(Oporornis
formosus)
and
Ruby-throated
Hummingbirds
(Archilochus colubris), and the arrival of Prairie Warblers (Dendroica discolor) in the
second post-treatment year. Ovenbirds (Seiurus aurocapilla), a forest interior species,
held four territories on the closed canopy treatments in 2002 but were reduced to a single
territory on one control site in 2003 (and none were detected in 2004, unpublished data).
Although the low sample size of this study did not provide the statistical power necessary
to elucidate significant differences for this species, the trend is worth noting. Several
previous studies have found that timber harvest negatively affects Ovenbird populations
(e.g. Thompson et al. 1992, Welsh and Healy 1993, Germaine et al. 1997, Gram et al.
2003). King et al. (1998) found higher nest predation rates for Ovenbirds nearer to
clearcuts and Manolis et al. (2002) reported that Ovenbird nesting success increased with
distance from clearcut edges in forest-dominated landscapes. The decrease in Ovenbird
territories observed here may have resulted from harvest in the adjacent treatment units;
further supporting claims of their edge sensitivity in extensively forested study areas.
38
Table 5. Total number of territories, mean density per 4 ha(a), and relative importance (b)
of eight species of songbirds accounting for 53 percent of mapped territories in
five overstory retention treatments in northeastern Alabama (2003).
Mean Territory Density (n=3) and Importance
Treatment
Total
Territories
0%
25%
50%
75%
100%
54
5.7 a (1) b
6.3 (1)
4.7 (1)
0.7 (6)
0.7 (6)
31.5
0.2 (8)
1.7 (5)
2.3 (3)
3.3 (1)
3.0 (1)
Yellow-breasted Chat
Icteria virens
29
3.7 (2)
4.0 (2)
2.0 (4)
-
-
Blue-gray Gnatcatcher
Polioptila caerulea
29
1.7 (3)
2.3 (3)
3.3 (2)
1.0 (5)
1.3 (3)
Tufted Titmouse
Baeolophus bicolor
20
0.7 (6)
1.7 (5)
1.0 (8)
1.7 (3)
1.7 (2)
Eastern Wood-Pewee
Contopus virens
19
0.7 (6)
2.0 (4)
1.3 (6)
1.7 (3)
0.7 (6)
Eastern Towhee
Pipilo erythrophthalmus
18
1.7 (3)
1.7 (5)
2.0 (4)
0.7 (6)
-
17.5
-
0.7 (8)
1.5 (5)
1.3 (2)
1.3 (3)
Species
Indigo Bunting
Passerina cyanea
Red-eyed Vireo
Vireo olivaceus
Scarlet Tanager
Piranga olivacea
Detection rate.
Forty-seven breeding bird species were detected in 2002 and 2003
combined. 2472 detections yielded 39 species in 2002 and 45 species were accumulated
from 3317 detections in 2003 (Table 6). These data were used to describe the patterns of
relative use of the five overstory treatments by breeding birds on the sites.
39
Table 6.
Mean detection rate of breeding bird species by overstory treatment in 2002
and 2003.
Species
Mean Detection Rate (detections/visit; n=3)
2002
2003
Treatment (% overstory
Treatment (% overstory
retention)
retention)
0
25
50
75
100
0
25
50
75
100
Acadian Flycatcher
Empidonax virescens
-
0.03
0.03
0.17
0.10
-
-
-
0.13
0.17
American Goldfinch
Carduelis tristis
-
0.20
0.53
-
-
2.53
0.70
0.63
0.07
-
American Redstart
Setophaga ruticilla
-
-
-
-
-
-
-
0.03
-
American Robin
Turdus migratorius
-
-
-
-
-
-
0.03
-
-
-
Black-and-white Warbler
Mniotilta varia
0.20
0.23
0.33
0.40
0.43
0.23
0.13
0.10
0.03
0.10
Blue-gray Gnatcatcher
Polioptila caerulea
0.23
1.27
2.23
1.23
0.67
1.67
2.67
2.90
1.07
1.77
Brown-headed Cowbird
Molothrus ater
0.80
1.13
0.87
0.27
0.20
1.03
1.03
0.80
0.40
0.17
Blue Grosbeak
Passerina caerulea
-
0.17
-
-
-
0.17
0.13
-
-
-
Blue Jay
Cyanocitta cristata
0.10
-
-
0.23
0.33
0.03
0.20
0.03
0.70
0.13
Blue-winged Warbler
Vermivora pinus
-
-
-
-
-
0.03
0.03
0.07
-
-
Carolina Chickadee
Poecile carolinensis
0.23
0.77
0.53
0.63
0.43
0.53
0.43
0.80
0.60
0.33
Carolina Wren
Thryothorus ludovicianus
0.70
1.23
1.27
0.53
0.40
0.90
0.90
1.27
0.10
0.03
-
0.07
-
-
0.03
-
0.03
-
-
-
Cerulean Warbler
Dendroica cerulea
40
-
Table 6. (continued).
Species
Mean Detection Rate (detections/visit; n=3)
2002
2003
Treatment (% overstory
Treatment (% overstory
retention)
retention)
0
25
50
75
100
0
25
50
75
100
Chipping Sparrow
Spizella passerina
-
-
-
-
-
0.03
0.07
-
-
-
Common Yellowthroat
Geothlypis trichas
-
-
-
-
-
0.20
0.10
-
-
-
Downy Woodpecker
Picoides pubescens
0.13
0.90
0.53
0.67
0.33
0.07
0.27
0.47
0.57
0.17
Eastern Bluebird
Sialia sialis
0.47
0.17
0.10
0.07
-
0.17
0.33
0.03
-
-
Eastern Phoebe
Sayornis phoebe
-
0.13
0.07
-
-
-
-
-
-
-
Eastern Towhee
Pipilo erythrophthalmus
0.43
0.30
0.53
0.20
-
1.37
0.90
1.27
0.53
0.03
Eastern Wood-Pewee
Contopus virens
0.57
1.37
0.97
0.77
0.10
0.63
1.30
0.97
1.47
0.37
-
-
-
-
-
0.20
-
-
-
-
0.07
0.40
0.53
0.13
0.17
0.10
0.33
0.33
0.33
0.17
Gray Catbird
Dumetella carolinensis
-
-
-
0.03
-
-
-
-
-
-
Hairy Woodpecker
Picoides villosus
-
0.20
0.17
0.03
0.13
0.03
0.33
0.13
0.23
0.03
Hooded Warbler
Wilsonia citrina
-
-
0.07
0.10
0.50
-
0.07
0.03
0.03
0.07
Indigo Bunting
Passerina cyanea
2.07
3.03
3.40
1.10
0.23
5.27
5.90
4.97
0.77
0.47
Field Sparrow
Spizella pusilla
Great Crested Flycatcher
Myiarchus crinitus
41
Table 6. (continued).
Species
Mean Detection Rate (detections/visit; n=3)
2002
2003
Treatment (% overstory
Treatment (% overstory
retention)
retention)
0
25
50
75
100
0
25
50
75
100
Kentucky Warbler
Oporornis formosus
-
0.40
0.17
-
0.03
0.43
0.97
0.77
0.17
0.07
Mourning Dove
Zenaida macroura
-
0.13
-
-
-
0.20
0.27
0.13
0.03
-
0.27
0.63
1.30
0.70
0.20
0.97
0.50
0.63
0.10
0.53
Ovenbird
Seiurus aurocapilla
-
0.03
-
0.23
1.03
-
-
-
0.03
0.17
Pileated Woodpecker
Dryocopus pileatus
-
-
0.07
0.07
0.10
-
-
0.03
0.10
0.13
Prairie Warbler
Dendroica discolor
-
-
-
-
-
0.43
0.23
0.03
-
-
Red-bellied Woodpecker
Melanerpes carolinus
0.17
0.37
0.27
0.40
0.37
0.17
0.10
0.13
0.30
0.43
Red-eyed Vireo
Vireo olivaceus
0.13
1.33
2.33
3.10
3.33
0.27
1.63
1.80
2.77
2.37
-
-
-
-
-
-
0.03
-
-
-
-
0.20
0.07
0.07
0.07
0.17
0.30
0.27
-
0.03
Scarlet Tanager
Piranga olivacea
0.20
0.67
1.17
2.10
1.53
0.13
0.67
0.87
1.80
1.17
Summer Tanager
Piranga rubra
1.00
1.77
1.40
1.03
0.47
0.80
1.20
1.50
0.93
0.40
Tufted Titmouse
Baeolophus bicolor
0.67
0.90
1.87
1.83
1.83
0.50
1.27
0.93
2.33
1.77
Northern Cardinal
Cardinalis cardinalis
Red-headed Woodpecker
Melanerpes erythrocephalus
Ruby-throated
Hummingbird
Archilochus colubris
42
Table 6. (continued).
Mean Detection Rate (detections/visit; n=3)
2002
2003
Treatment (% overstory
Treatment (% overstory
retention)
retention)
0
25
50
75
100
0
25
50
75
100
Species
White-breasted Nuthatch
Sitta carolinensis
0.13
0.47
0.67
0.73
0.50
0.10
0.40
0.87
1.00
0.90
-
-
0.10
-
-
0.23
0.30
0.07
0.03
-
0.07
0.20
0.40
1.03
1.30
0.10
0.67
0.77
0.83
1.30
Wood Thrush
Hylocichla mustelina
-
-
0.07
0.13
0.83
-
-
-
0.07
0.67
Yellow-breasted Chat
Icteria virens
0.27
0.07
0.07
-
-
3.13
2.93
1.70
0.03
-
Yellow-billed Cuckoo
Coccyzus americanus
-
0.23
0.13
0.43
0.43
-
0.27
0.27
0.17
0.17
Yellow-throated Vireo
Vireo flavifrons
0.03
0.50
0.87
0.83
0.37
0.13
0.73
0.73
0.73
0.30
-
0.03
-
-
-
0.03
-
-
-
-
White-eyed Vireo
Vireo griseus
Worm-eating Warbler
Helmitheros vermivorus
Yellow-throated Warbler
Dendroica dominica
Both years’ detection rate data were analyzed separately due to significant
treatment-year interactions (F4,8=11.615, p=0.002). In 2002, univariate ANOVA showed
results similar to the territory density measures, in that, the 25, 50 and 75% treatments
had the highest detection rates and clearcuts were again the lowest (F4,8=8.137, p=0.006).
A similar response to that of territory density in 2003 was also seen in the species use
data (Figure 5). Detections were highest in the 25% treatment and were again lowest in
43
the closed canopy treatments (F4,8=11.221, p=0.002). The clearcuts showed a positive
response in the second year after treatment, surpassing the closed canopy units in
abundance based on detection rate. Trend analysis for both years showed a quadratic
relationship between detection rate and treatment intensity (p<0.005), with lower values
at the ends of the treatment spectrum and higher values in 25 and 50 % overstory
retentions.
DETECTION RATE (det./hour)
40
c b
bc
30
bc
b ab ab
b
20
a
a
10
YEAR
2002
2003
0
0%
25%
50%
75%
100%
TREATMENT
Figure 5. Average breeding songbird detection rate (n=3) of five overstory retention
treatments in 2002 and 2003. Different letters above boxes of the same color
(year) indicate significantly different means (p < 0.05). Box-plots show
interquartile range (box), median (line), and the highest and lowest values
(whiskers), excluding outliers.
44
Species Richness
By Territory Density. Species richness among the territorial bird community differed on
the treatments between years (F4,8=4.898, p=0.027).
In 2002, species richness of
territorial breeding birds (Figure 6) showed a pattern similar to the abundance indices.
Control, 75, 50, and 25% retention treatments had similar numbers of species while the
clearcuts had lower mean richness (F4,8=8.890, p=0.005). There was no interaction
between block and treatment effects.
Trend analysis also revealed a quadratic
relationship (p<0.002) between species richness and overstory treatment. Maximum
species richness was observed in the 25 percent treatments. The control, 75, 50, and 25%
treatments had at least twice as many species, on average, as clearcuts.
In 2003 the trends of species richness changed (Figure 6). There was a significant
treatment effect (F4,8=4.811, p=0.028), however, there were no significant mean
separations. Clearcuts gained several territorial species and were no longer different
relative to the other treatments.
45
TERRITORIAL SPECIES
30
b
20
10
b
b
b
a
YEAR
2002
2003
0
0%
25%
50%
75%
100%
TREATMENT
Figure 6. Mean territorial species richness (n=3) of five overstory retention treatments
in 2002 and 2003. Different letters above boxes of the same color (year)
indicate significantly different means (p < 0.05). Box-plots show interquartile
range (box), median (line), and the highest and lowest values (whiskers),
excluding outliers.
By Detection Rate.
Treatment effects on the total number of species recorded as
detections interacted significantly with year effect (F4,8=5.360, p=0.021).
Species
richness based on detection rate was quadratically related to overstory retention (p=
0.009) in 2002.
The clearcuts had the fewest species using them; whereas species
detections were highest in the 25 and 50% treatments and intermediate in the 75 and
100% treatments. Maximum richness was attained in the 50% units.
Through 2003, the species richness in the control plots was lower relative to
the other treatments (F4,8=4.490, p=0.034) and the clearcuts increased to levels similar to
46
the 25 and 50% treatments (Figure 7). In this case, species richness showed a decreasing
linear trend with increasing overstory retention (p=0.013). Again, songbirds responded
strongly to the conditions of the clearcuts in the second post-treatment year.
40
b
ab b
DETECTED SPECIES
30
20
b
ab
ab ab
ab a
a
10
YEAR
2002
2003
0
0%
25%
50%
75%
100%
TREATMENT
Figure 7. Average species richness (n=3) based on detections of five overstory retention
treatments in 2002 and 2003. Different letters above boxes of the same color
(year) indicate significantly different means (p < 0.05). Box-plots show
interquartile range (box), median (line), and the highest and lowest values
(whiskers), excluding outliers.
Diversity Indices and Evenness
Brillouin diversity and evenness. Treatment effect on diversity of the territorial bird
community varied between years (F4,8=6.398, p=0.013). Therefore, Brillouin diversity
based on territory density was analyzed separately for each. Clearcuts had the lowest
47
Brillouin diversity in 2002 but none of the other treatments differed (F4,8=9.830, p=
0.004). The clearcuts again had the lowest diversity in 2003 but did not differ from the
control plots (F4,8=9.684, p=0.004). Highest diversity in the second year was in the 50%
treatments (Figure 8).
1.2
BRILLOUIN DIVERSITY
1.0
b bc
b c
b
a
.8
b
ab
a
a
.6
.4
YEAR
.2
2002
2003
0.0
0%
25%
50%
75%
100%
TREATMENT
Figure 8. Mean Brillouin diversity (n=3) for five levels of overstory retention based on
territory density in 2002 and 2003. Different letters above boxes of the same
color (year) indicate significantly different means (p < 0.05). Box-plots show
interquartile range (box), median (line), and the highest and lowest values
(whiskers), excluding outliers.
In addition to the diversity index, eveness was calculated as the proportion of
maximum possible Brillouin diversity for each treatment unit. Eveness did not differ
between years or treatments and all values were above 0.77 suggesting that the observed
48
differences in Brillouin diversity were likely due to differences in species richness rather
than eveness.
Shannon diversity and evenness. For Shannon diversity, treatment effect was significant
(F4,8=4.607, p=0.032) without year interaction. Mean separations showed diversity based
on detection rate was lowest in the clearcut treatments, highest in the 25% and
intermediate in the 50%, 75% and 100% treatments (Figure 9). As with the Brilloun
diversity index no differences in eveness among treatments were detected.
1.4
SHANNON DIVERSITY
1.2
a a
b b abab ab ab
ab
ab
1.0
.8
YEAR
.6
2002
2003
.4
0%
25%
50%
75%
100%
TREATMENT
Figure 9. Mean Shannon diversity (n=3) for five levels of overstory retention based on
detection rate in 2002 and 2003. Different letters above boxes of the same
color (year) indicate significantly different means (p < 0.05). Box-plots show
interquartile range (box), median (line), and the highest and lowest values
(whiskers), excluding outliers.
49
Similarity and Treatment Selection
Similarity. Percent similarity for 2002 (Table 7) and 2003 (Table 8) decreased with
increasing difference in overstory conditions suggesting a graded turnover in species
composition from one end of the treatment spectrum to the other. In 2002, there was very
little similarity between communities of territorial species in the clearcuts and controls,
similar to the near complete turnover in species composition found by Dickson et al.
(1993) before and after clearcutting mature forest. As Crawford et al. (1981) described,
this dissimilarity between the extremes of overstory retention are likely due to habitat
structure and its suitability to the individual species within the community of the study
area.
In 2003, as species richness and abundance increased in the clearcuts, their
compositional overlap with the other treatments also increased; including a doubling in
the similarity of species using both clearcuts and the two closed canopy treatments. For
both years, high community overlap was seen in those treatments most alike in structure
such as between the 25 and 50% treatments and similarly between the 75% and controls.
50
Table 7.
Percent similarity between all treatments based on territory density (above the
diagonal) and detection rate (below the diagonal) for 2002.
Treatment (% overstory retention)
Table 8.
Percent Similarity 2002
Treatment (% overstory retention)
0
25
50
75
100
0
X
47
47
25
14
25
70
X
81
65
52
50
64
80
X
68
55
75
50
64
72
X
74
100
36
46
60
76
X
Percent similarity between all treatments based on territory density (above the
diagonal) and detection rate (below the diagonal) for 2003.
Treatment (% overstory retention)
Percent Similarity 2003
Treatment (% overstory retention)
0
25
50
75
100
0
X
71
63
29
29
25
75
X
78
52
43
50
69
85
X
57
56
75
34
53
62
X
73
100
32
47
54
75
X
51
Treatment selection.
Six species defended territories in just one of the retention
treatments in 2002 (Table 9). A few species selected only one or two of the treatment
types, while other generalist species held territories over a range of overstory conditions.
Four species were recorded holding territories only on the 25% treatments, which may
help explain the high species richness and diversity on those sites. Interestingly, 12
species that maintained territories in all but one treatment type excluded either the control
or clearcut plot in their range of habitat selection. Five songbird species held territories
in all treatment types: Carolina Wren (Thryothorus ludovicianus), Eastern Wood-Pewee
(Contopus virens), Tufted Titmouse (Baeolophus bicolor), Northern Cardinal (Cardinalis
cardinalis), and Summer Tanager (Piranga rubra). No single breeding bird species was
unique to the clearcut treatments in 2002.
52
Table 9. Breeding habitat association of 11 bird species holding territories within a
limited range of treatments (2002).
Habitat Type
Open canopy forest (25% and
Closed canopy forest
50% retention) and clearcut (0%
(75% and 100% retention)
retention)
Species
Blue Jay
Cyanocitta cristata
X
Acadian Flycatcher
Empidonax virescens
X
Wood Thrush
Hylocichla mustelina
Xa
Ovenbird
Seiurus aurocapillus
X
Hooded Warbler
Wilsonia citrina
Xa
Yellow-breasted Chat
Icteria virens
X
Kentucky Warbler
Oporornis formosus
X
Blue Grosbeak
Passerina caerulea
Xb
Eastern Phoebe
Sayornis phoebe
Xb
White-eyed Vireo
Vireo griseus
Xb
Mourning Dove
Zenaida macroura
Xb
a
Occurred only on control treatments
Occurred only on 25 percent treatments
b
In 2003, five species were unique to individual treatments (Table 10).
The
Ovenbird, a forest interior specialist, and the Wood Thrush (Hylocichla mustelina), an
interior-edge species, were only found in control plots. Two species typical of fields and
early-successional habitats, the Blue Grosbeak (Passerina caerulea) and Field Sparrow
(Spizella pusilla), each held one territory on the same clearcut in block 2 at Jack Gap.
53
Table 10. Breeding habitat association of 13 bird species holding territories within a
limited range of treatments (2003).
Species
Habitat Type
Open canopy forest (25% and
Closed canopy forest
50% retention) and clearcut (0%
(75% and 100% retention)
retention)
Blue Jay
Cyanocitta cristata
X
Acadian Flycatcher
Empidonax virescens
X
Wood Thrush
Hylocichla mustelina
Xa
Ovenbird
Seiurus aurocapillus
Xa
Ruby-throated Hummingbird
Archilochus colubris
X
Prairie Warbler
Dendroica discolor
X
Common Yellowthroat
Geothlypis trichas
X
Yellow-breasted Chat
Icteria virens
X
Blue Grosbeak
Passerina caerulea
Xb
Eastern Bluebird
Sialia sialis
Xc
Field Sparrow
Spizella pusilla
Xb
White-eyed Vireo
Vireo griseus
X
Mourning Dove
Zenaida macroura
a
Occurred only on control treatments
b
Occurred only on clearcut treatments
c
Occurred only on 25 percent treatments
X
54
Guild Analyses
Migrant guilds. In 2002, the lowest territory density of Neotropical migrant songbirds
was in the clearcuts (F4,8=6.385, p=0.013). All other treatments had higher densities of
Neotropical migrants but did not differ significantly (Table 11). In 2003, Neotropical
migrants were most abundant in the 25% and lowest in the 75 and 100% treatments
(F4,8=14.500, p=0.001). Detection rates were similar to the results for territory density in
both years.
Temperate migrant territory density differed among treatments (F4,8=12.118,
p=0.002) only in 2003 where their highest territory densities were reached in the open
forested sites (25 and 50%). Their lowest territory density was in the 75% herbicide
treatment. The detection rates were significant in both years. In 2002, the highest
detection rates occurred in the 50% and the lowest were in the clearcuts (F4,8=4.934,
p=0.027). For 2003, the detection rates of temperate migrants increased in the clearcuts,
were highest in the 25%, and lowest in the 75% treatments (F4,8=5.221, p=0.023).
In both years, permanent residents had their highest territory density (F4,8=10.684,
p=0.003) in the 50% treatment units and lowest density in the clearcuts.
In summary, species in all three migrant categories showed a strong preference
for open forest sites in both years.
55
Table 11. Mean abundance (n=3) of migrant guilds in five overstory treatments for 2002
and 2003. Means followed by different letters in the same row are
significantly different (p<0.05).
Treatment
Migrant
Guild
Abundance
Measure
Year
0
25
50
75
100
Neotropical
Neotropical
Neotropical
Neotropical
Terr. Den.
Terr. Den.
Det. Rate
Det. Rate
2002
2003
2002
2003
4.00 a
12.67 ab
4.40 a
11.40 abc
11.67 b
21.00 c
10.50 b
17.17 c
13.33 b
17.67 bc
11.67 b
15.07 bc
11.83 b
11.83 a
10.02 b
10.27 ab
11.17 b
10.33 a
10.53 b
7.87 a
Temperate
Temperate
Temperate
Temperate
Terr. Den.
Terr. Den.
Det. Rate
Det. Rate
2002
2003
2002
2003
0.00
3.00 bc
0.43 a
2.77 ab
1.50
3.33 c
1.50 ab
3.43 b
1.83
3.33 c
2.67 b
3.10 ab
1.33
1.00 a
1.67 ab
1.13 a
1.00
1.67 ab
1.10 ab
1.87 ab
Resident
Resident
Terr. Den.
Det. Rate
2002,2003
2002,2003
4.33 a
6.47
7.92 bc
7.65
9.67 c
8.47
6.67 abc
7.28
5.17 ab
4.75
Nest location guilds.
Ground nesting species showed no response to the overstory
treatments (Table 12), however the members of these guilds generally differed between
open and closed canopy sites.
The Ovenbird, Worm-eating Warbler (Helmitheros
vermivorus), and Black-and white Warbler (Mniotilta varia) nested in greater numbers in
the closed canopy sites while the Field Sparrow, Eastern Towhee, and Kentucky Warbler
appeared to prefer open sites.
Low-shrub nesters were most abundant in the open canopy and clearcut
treatments and least abundant in the closed canopy sites (p<0.007, for both abundance
measures in 2002 and 2003). This is to be expected with the vigorous growth of young
trees and shrubs in the understory of the harvested stands.
56
The lack of a midstory on the clearcuts was most likely the cause of the low
representation of the midstory-subcanopy guild in these treatments (F4,8=21.707,
p<0.0005 for 2002 territory density; F4,8=4.840, p=0.028 for 2003 territory density; and
F4,8=10.804, p=0.003 for 2002 and 2003 detection rate).
Highest territory density
occurred on the 50% treatments in 2003 and few differences were detected between the
other treatments for 2002 territory density and the two-year average detection rates.
The abundance of canopy-nesting species was negatively affected by clearcutting.
Greatest territorial abundance of canopy nesters occurred in the 25, 50, and 75%
treatments (F4,8=9.683, p=0.004) and detection rate peaked in the 75% units (F4,8=5.013,
p=0.026). These results show that the residual overstory of either of these shelterwood
stands support a relatively abundant canopy-nesting guild.
Cavity nesters held fewest territories on the clearcuts in both years (F4,8=10.525,
p=0.003) and the other treatments did not significantly differ with respect to density. The
detection rate of this guild was also expectedly lowest in the clearcuts, but in this case,
the 75% retention treatment had the most abundant cavity-nesting assemblage
(F4,8=4.698, p=0.030), suggesting these species may have been attracted to the deadened
midstory of these stands in search of potential nest sites or foraging opportunities.
The Brown-headed Cowbird did not show a pattern of abundance across
treatments and measured detection rates were relatively low. Although preferred feeding
areas of cowbirds such as pastureland, roadsides, agricultural fields, and rural
developments are within several kilometers of the study areas, the sites themselves are
surrounded by extensive and relatively unfragmented mature forest which may have
limited their local abundance.
57
Table 12. Mean abundance (n=3) of nest location guilds in five overstory treatments for
2002 and 2003. Means followed by different letters in the same row are
significantly different (p<0.05).
Treatment
Nesting
Guild
Abundance
Measure
Year
0
25
50
75
100
Ground
Ground
Ground
Terr. Den.
Det. Rate
Det. Rate
2002,2003
2002
2003
1.75
0.70
2.37
2.83
1.17
2.70
3.08
1.43
2.97
2.08
1.97
1.43
2.58
2.80
1.67
Low-shrub
Low-shrub
Low-shrub
Low-shrub
Terr. Den.
Terr. Den.
Det. Rate
Det. Rate
2002
2003
2002
2003
4.00 ab
13.67 b
3.30 ab
14.07 c
4.83 ab
13.33 b
5.47 bc
12.10 bc
7.00 b
10.00 b
6.73 c
9.47 b
2.17 a
0.83 a
2.47 a
1.33 a
1.67 a
1.33 a
1.33 a
1.10 a
Terr. Den.
2002
0.83 a
5.33 b
5.83 b
6.50 b
6.33 b
Terr. Den.
2003
3.33 a
6.67 ab
8.67 b
5.50 ab
6.33 ab
Det. Rate
2002,2003
2.13 a
5.47 b
6.25 b
5.65 b
5.73 b
Canopy
Canopy
Terr. Den.
Det. Rate
2002,2003
2002,2003
0.58 a
1.00 a
3.58 b
2.78 ab
4.08 b
2.80 ab
4.92 b
4.32 b
2.5 ab
2.17 ab
Cavity
Cavity
Terr. Den.
Det. Rate
2002,2003
2002,2003
0.92 a
1.70 a
4.67 b
3.83 ab
4.67 b
4.23 ab
5.17 b
5.02 b
4.33 b
3.90 ab
Brood
parasite
Det. Rate
2002,2003
0.92
1.08
0.83
0.33
0.18
Midstorysubanopy
Midstorysubanopy
Midstorysubanopy
Foraging strategy.
The foliage-gleaning guild was least abundant in the clearcut
treatments in 2002 and similar in the other treatments [p<0.003 for both territory density
and detection rate, (Table 13)]. In 2003, this guild had its highest abundance in the 25
and 50% treatments and lowest in treatments with closed-canopy conditions (p<0.002 for
both abundance estimates). It is likely that vigorous understory growth and residual
canopy cover in the 25 and 50% treatments provided suitable foraging habitat for a wide
58
range of foliage-gleaning species. In addition, the intermediate-intensity harvests offered
nesting opportunities for high numbers of shrub-nesting species (Table 12). Many of
these low-nesting species are foliage gleaners (Table 1) that contributed to the high
abundances in this foraging guild.
Ground foraging species had highest territorial
abundance in clearcut and 50% treatments in 2003 (F4,8=12.529, p=0.002) and lowest
densities in the 75% treatment. Bark-gleaners were least common in the clearcut likely
due to a lack of substantially-sized foraging substrate (p<0.009 for both years and
abundance measures). Flycatchers, the primary members of the hawking guild, did not
respond significantly to the treatments in this study.
Table 13. Mean abundance (n=3) of foraging guilds in five overstory treatments for
2002 and 2003. Means followed by different letters in the same row are
significantly different (p < 0.05).
Treatment
Foraging
Guild
Abundance
Measure
Year
0
25
50
75
100
Foliage glean
Foliage glean
Foliage glean
Foliage glean
Terr. Den.
Terr. Den.
Det. Rate
Det. Rate
2002
2003
2002
2003
3.67 a
14.33 ab
4.90 a
15.97 bc
11.33 b
22.00 c
11.00 b
19.83 c
14.83 b
19.17 bc
15.10 b
18.07 c
14.17 b
12.17 a
13.43 b
12.17 ab
11.33 b
11.00 a
11.13 b
10.10 a
Ground glean
Ground glean
Ground glean
Ground glean
Terr. Den.
Terr. Den.
Det. Rate
Det. Rate
2002
2003
2002
2003
1.83
6.00 c
2.30
5.33 b
3.17
5.00 bc
4.03
5.0 b
5.83
6.50 c
4.20
4.90 b
2.17
1.83 a
2.47
2.13 a
3.17
2.33 ab
3.03
1.80 a
Bark glean
Bark glean
Terr. Den.
Det. Rate
2002,2003
2002,2003
0.33 a
0.63 a
2.33 b
1.73 b
2.08 b
1.88 b
2.83 b
2.25 b
2.00 ab
1.82 b
Hawking
Hawking
Hawking
Hawking
Terr. Den.
Terr. Den.
Det. Rate
Det. Rate
2002
2003
2002
2003
0.50
0.67
1.10
0.90
2.83
2.67
2.10
1.97
2.33
1.67
1.70
1.37
1.00
2.33
1.13
1.93
1.00
1.33
0.37
0.70
59
Habitat distribution. Table 14 summarizes the response of the habitat guilds to the
gradient of overstory retention. Open-edge species were most common in the open forest
habitat of the 25 and 50% treatments and the scrub habitat of the clearcuts (p<0.004 for
both abundance estimates in 2002 and 2003). Interior-edge species favored the three
treatments of intermediate overstory retention (p<0.005 for both abundance measures)
including the closed canopy 75% treatment. Interior forest specialists showed highest
territory density in the control stands (F4,8=4.331, p=0.038). These findings fit well with
the a priori categorization of the bird community into guilds based on habitat preference,
suggesting the classifications were largely accurate.
Table 14. Mean abundance (n=3) of habitat guilds in five overstory treatments for 2002
and 2003. Means followed by different letters in the same row are
significantly different (p < 0.05).
Treatment
Habitat
Distribution
Abundance
Measure
Year
0
25
50
75
100
Open/edge
Open/edge
Open/edge
Open/edge
Terr. Den.
Terr. Den.
Det. Rate
Det. Rate
2002
2003
2002
2003
3.67 ab
13.00 b
4.30 b
14.50 b
4.83 b
13.00 b
6.33 c
13.17 b
5.33 b
9.33 b
6.40 c
9.97 b
1.67 a
0.83 a
2.03 a
1.43 a
1.17 a
0.67 a
0.90 a
0.70 a
Interior/edge
Interior/edge
Terr. Den.
Det. Rate
2002,2003
2002,2003
5.67 a
5.95 a
15.00 b
12.08 b
17.00 b
13.67 b
13.58 b
12.65 b
10.75 ab
9.78 ab
Interior
Interior
Terr. Den.
Det. Rate
2002,2003
2002,2003
0.17 a
0.62
2.42 ab
2.12
3.25 ab
2.87
4.83 ab
4.53
5.58 b
4.85
60
Breeding Success by Territory Rank
The breeding success ranks for each of the 359 territories in all treatments for
2003 were analyzed using the Kruskal-Wallis test; however no significant differences in
breeding success were found using this approach (X2=3.093, df=4, asymp.sign.=0.542).
Although actual rates of breeding success were not measured by this technique, from this
analysis it appears the treatments had similar capacities to support breeding bird
populations. Whether these treatments or the study area are local population sources or
sinks is not known. More intensive nest searching would likely improve the efficacy of
detecting treatment differences using this breeding success estimate.
61
CHAPTER 4
CONCLUSIONS
Open forest sites with intermediate overstory retention levels had the highest
overall community indices of richness, abundance, and diversity. Many of the guilds also
showed high abundance on these sites. It is evident that these treatments provide suitable
habitat for a complex songbird community. These results certainly support the claims of
the intermediate disturbance hypothesis with respect to high diversity at intermediate
severity of disturbance (Connell 1978). Dickson et al. (1995) reported that bird diversity
in shelterwood and seed-tree regenerated stands can be higher as a result of the
immigration of early-successional species into the stands while some late-successional
species remain and utilize the residual overstory. Nichols and Wood (1995) found that
two-aged stands, with residual overstory and a regenerating cohort underneath, harbored
greater numbers of birds and species than clearcuts or untreated stands. Brawn and Balda
(1988) found that shelterwood logging provided a wider array of habitat types for forest
songbirds than mature, unharvested conifer forests. The results of this research appear to
be congruent with these studies as our intermediate retention units had the highest
community indices and have shown high species overlap with the other treatments.
This study offered an interesting look into the bird community characteristics
during the early stages of clearcut regeneration. Similar to Conner and Adkisson (1975),
62
this study showed that bird abundance, species richness, and diversity were lowest in
clearcuts during the first post-treatment year, likely due to limited vegetative re-growth
before the beginning of the breeding season.
It has been widely reported that
regenerating clearcuts can support abundant and diverse bird communities. Baker and
Lacki (1997) showed that songbird communities in clearcuts and high- and low-leave
two-aged hardwood forests had greater diversity, species richness, and abundance than in
unharvested stands. This differs from our study in that clearcut stands had the lowest
density, richness, and diversity and control stands showed values similar to the
intermediate retention treatments. However, the clearcut treatments in 2003 showed a
strong recovery from the very low levels of diversity and abundance in the first posttreatment year and the closed canopy sites were often lower relative to the other
treatments. Keller et al. (2003) reported that values of avian density and species richness
for 6 year old clearcuts in the northeast were over two times that of mature stands. Our
results were quite different, but it is important to stress that we studied the bird
communities of clearcuts in the first two breeding seasons directly after dormant season
harvest. It is anticipated that the vegetative structure and composition in the clearcut
plots will change considerably in coming years, likely leading to increasing community
indices.
The trends of similarity showed that overlap between similar treatments,
especially between the two open forest treatments (25 and 50%) and among the two types
of closed canopy sites, was high. Similarity decreased regularly as the difference in
overstory retention increased. This illustrated a gradual turnover in species composition
63
across the gradient of overstory retention and suggests songbird community composition
is a function of the basal area and canopy cover conditions following the treatments
applied in this study.
Management Implications
From a management perspective, the patterns of treatment selection by some
species would suggest the mature closed-canopy habitat retained in the control and 75%
treatments may be necessary to maintain a few unique species (Tables 9 and 10). The
songbird community structure is relatively similar between the 75% and control plots;
however the Wood Thrush and Hooded Warbler (Wilsonia citrina) only nested on the
control sites and the Ovenbird was comparatively less abundant in the 75% treatments. It
is important to remember that most of the midstory of the 75% treatment was killed using
an herbicide injection. If the preservation of a species dependent on a fully-developed
and living midstory is a management priority, this should be taken into consideration. On
the other hand, the standing dead trees produced in the 75% treatment provided increased
foraging opportunities for bark-gleaning and excavating birds (unpublished data) that
may benefit from this type of shelterwood.
The open forest habitat of the 25 and 50% retention treatments harbored a number
of unique species and had the highest abundance, richness, and diversity. The treatments
that produced open forest conditions (5-10m2ha-1 basal area and ~75% residual canopy
cover of sawtimber diameter trees with vigorous crowns) could be prescribed to increase
songbird abundance and diversity by supporting a combination of closed-canopy, open
forest, and scrub species.
64
The clearcut plots had very low community indices and no unique species during
the first breeding season following harvest. However, just one year later, species richness
and abundance increased dramatically and the clearcuts harbored strong populations of
early-successional species, some of which are declining range-wide (Askins 1993) and
are of important conservation concern. Forest management practices that retain some
overstory, minimize understory disturbance, and allow vegetation to regenerate beneath
the residual canopy may be beneficial to forest songbird communities. The combination
of these three guidelines could be prescribed to promote desirable timber species and at
the same time, maintain high community indices for forest songbirds throughout a
managed forest rotation by avoiding periods of low abundance and richness as in the
early stages of a clearcut.
65
REFERENCES
Alabama Forestry Commission 2003. Forest Statistics: Alabama Forest Facts.
http://www.forestry.state.al.us/publication/forest_facts.htm.
Ambuel, B. and S.A. Temple. 1982. Songbird populations in southern Wisconsin
forests:1954 and 1979. Journal of Field Ornithology 53:149-58.
Askins, R.A. 1993. Population trends in grassland, shrubland, and forest birds in eastern
North America. Current Ornithology 11:1-34.
Askins, R.A., J.F. Lynch, and R. Greenberg. 1990. Population declines in migratory
birds in eastern North America. Current Ornithology 7:1-57.
Baker, M.D. and M.J. Lacki. 1997. Short-term changes in bird communities in response
to silvicultural prescriptions. Forest Ecology and Management 96:27-36.
Blake, J.G. and J.R. Karr. 1987. Breeding birds of isolated woodlots: area and habitat
relationships. Ecology 68:1724-1734.
Brawn, J.D., and R.P. Balda. 1988. The influence of silvicultural activity on ponderosa
pine forest bird communities in the southwestern United States. Bird
Conservation 3:3-21.
Connell, J.H. 1978. Diversity in tropical rain forests and coral reefs. Science 199:13021310.
Conner, R.N. and C.S. Adkisson. 1975. Effects of clearcutting on the diversity of
breeding birds. Journal of Forestry 73:781-785.
Conner, R.N., J.W. Via, and I.D. Prather. 1979. Effects of pine-oak clearcutting on
winter and breeding birds in southwestern Virginia. Wilson Bulletin 91:301316.
Conner, E.F., J.M. Yoder, and J.A. May. 1999. Density-related predation by the
Carolina chickadee, Poecile carolinensis, on the leaf-mining moth, Cameraria
hamadryadella at three spatial scales. Oikos 87:102-112.
66
Crawford, H.S., R.G. Hooper, and R. W. Titterington. 1981. Songbird population
response to silvicultural practices in central Appalachian hardwoods. Journal of
Wildlife Management 45:680-692.
Crozier, G.E., D.E. Gawlik. 2003. Wading bird nesting effort as an index of wetland
ecosystem integrity. Waterbirds 26:303-324.
Dickson, J.G., F.R. Thompson, III, R.N. Conner, and K.E. Franzreb. 1993. Effects of
silviculture on Neotropical migratory birds in central and southeastern oak pine
forests. Pp. 374-385 in Status and Management of Neotropical Migratory Birds
(D.M. Finch and P.W. Stangel, eds.). Rocky Mountain Forest and Range
Experiment Station, USDA Forest Service, General Technical Report RM-229,
Fort Collins, CO.
Dickson, J.G., F.R. Thompson, III, R.N. Conner, and K.E. Franzreb. 1995. Silviculture
in central and southeastern oak-pine forests. Pp. 245-266 in Ecology and
Management of Neotropical Migratory Birds: a synthesis and review of critical
issues (T.E. Martin and D.M. Finch, eds.). Oxford University Press, New York,
NY.
Ehrlich, P.R., D.S. Dobkin, and D. Wheye. 1988. The birder’s handbook: a field guide
to the natural history of North American birds. Simon and Schuster, New York,
NY. 785 pp.
Finch, D.M. and P.W. Stangel. 1993. Status and Management of Neotropical Migratory
Birds. Rocky Mountain Forest and Range Experiment Station. USDA Forest
Service, General Technical Report RM-229, Fort Collins, CO.
Freemark, K. and B. Collins. 1992. Landscape ecology of birds breeding in temperate
forest fragments. Pp. 443-454 in Ecology and Conservation of Neotropical
Migrant Landbirds (J.M. Hagen III and D.W. Johnston, eds.). Smithsonian
Institution Press, Washington, D.C.
Germaine, S.S., S.H. Vessey, and D.E. Capen. 1997. Effects of small forest openings
on the breeding bird community in a Vermont hardwood forest. The Condor
99:708-718.
Gram, W.K., P.A. Porneluzi, R.L. Clawson, J. Faaborg, and S.C. Richter. 2003. Effects
of experimental forest management on density and nesting success of bird species
in Missouri Ozark forests. Conservation Biology 17:1324-1337.
Greenberg, C.H., Smith, L.M., and Levey, D.J. 2001. Fruit fate, seed germination and
growth of an invasive vine: an experimental test of ‘sit and wait’ strategy.
Biological Invasions. 3: 363-372.
67
Hagan, J.M. and D.W. Johnston, eds. 1992. Ecology and conservation of Neotropical
migrant landbirds. Smithsonian Institution Press, Washington, D.C. 609 pp.
Harlow, R.F., R.L. Downing, and D.H. VanLear. 1997. Responses of wildlife to
clearcutting and associated treatments in the eastern United States. Department of
Forest Resources Technical Paper No. 19. Clemson University, Clemson, SC.
Hartsell, A.J. and J.S. Vissage. 2001. Forest statistics for north Alabama, 2000.
Southern Research Station, USDA Forest Service, Resource Bulletin SRS-64,
Asheville, NC.
Imhof, T.A. 1976. Alabama birds. The University of Alabama Press, University, AL.
445 pp.
International Bird Census Committee. 1970. An international standard for a mapping
method in bird census work recommended by the International Bird Census
Committee. Audubon Field Notes 24:722-726.
Johnston, D.W. and E.P. Odum. 1956. Breeding bird populations in relation to plant
succession on the piedmont of Georgia. Ecology 37:50-62.
Keller, J.K., M.E. Richmond, and C.R. Smith. 2003. An explanation of patterns of
breeding bird species richness and density following clearcutting in northeastern
USA forests. Forest Ecology and Management 174: 541-564.
Keller, M.E. and S.H. Anderson. 1992. Avian use of habitat configurations created by
forest cutting in southeastern Wyoming. The Condor 94:55-65.
King, D.I., R.M. DeGraaf, and C.R. Griffin. 1998. Edge-related nest predation in
clearcut and groupcut stands. Conservation Biology 12:1412-1415.
Krebs, C.J., 1998. Ecological methodology. Addison Wesley Longman, Menlo Park,
CA. 620 pp.
Manolis, J.C., D.E. Andersen, and F.J. Cuthbert. 2002. Edge effect on nesting success of
ground nesting birds near regenerating clearcuts in a forest-dominated landscape.
The Auk 119:955-970.
Martin, T.E. and D.M. Finch, eds. 1995. Ecology and management of Neotropical
migratory birds: a synthesis and review of critical issues. Oxford University
Press, New York, NY. 489 pp.
68
Maurer, B.A. 1993. Biological diversity, ecological integrity, and Neotropical migrants:
new perspectives for wildlife management. Pp. 24-31 in Status and Management
of Neotropical Migratory Birds (D.M. Finch and P.W. Stangel, eds.). Rocky
Mountain Forest and Range Experiment Station, USDA Forest Service, General
Technical Report RM-229, Fort Collins, CO.
Nichols, J.V. and P.B. Wood. 1995. Effects of two-age management and clearcutting on
songbird density and reproductive success. Pp. 225 in Proceedings of the Tenth
Central Hardwood Forest Conference (K.W. Gottschalk and S.L.C. Fosbroke,
eds.). Northeastern Forest Experiment Station, USDA Forest Service, General
Technical Report NE-197, Radnor, PA.
Odum E.P. 1950. Bird populations of the Highlands (North Carolina) Plateau in relation
to plant succession and avian invasion. Ecology 31:587-605.
Ralph, C.J., G.R. Geupel, P. Pyle, T.E. Martin, and D.F. DeSante. 1993. Handbook of
field methods for monitoring landbirds. Pacific Southwest Research Station,
USDA Forest Service, General Technical Report PSW-GTR-144-www, Albany,
CA.
Rappole, J.H. 1995. The Ecology of Migrant Birds. Smithsonian Institution Press,
Washington D.C. 269 pp.
Robbins, C.S., J. Sauer, R. Greenberg, and S. Droege. 1989. Population declines in
North American birds that migrate to the Neotropics. Proceedings of the National
Academy of Science 86:7658-7662.
Robinson, S.K. 1992. The breeding season: introduction. Pp. 405-407 in Ecology and
Conservation of Neotropical Migrant Landbirds (J.M. Hagen III and D.W.
Johnston, eds.). Smithsonian Institution Press, Washington, D.C.
Robinson, S.K., F.R. Thompson, III, T.M. Donovan, D.R. Whithead, and J. Faaborg.
1995. Regional forest fragmentation and the nesting success of migratory birds.
Science 267:1987-1989.
Sauer, J.R., J.E. Hines, and J. Fallon. 2001. The North American Breeding Bird
Survey, Results and Analysis 1966 - 2000. Version 2001.2, http://www.mbrpwrc.usgs.gov/bbs/bbs.html. USGS Patuxent Wildlife Research Center, Laurel,
MD.
Sauer, J.R., S. Schwartz, and B. Hoover. 1996. The Christmas Bird Count Home Page.
Version 95.1, http://www.mbr-pwrc.usgs.gov/bbs/cbc.html. USGS Patuxent
Wildlife Research Center, Laurel, MD.
69
Schweitzer, C.J. 2004. First-year response of an upland hardwood forest to five levels of
overstory tree retention. Pp. 287-291 in Proceedings of the Twelfth Biennial
Southern Silvicultural Research Conference (K. Connor, ed.). Southern Research
Station, USDA Forest Service, General Technical Report SRS-71, Asheville, NC.
Sherry, T.W. and R.T. Holmes. 1995. Summer versus winter limitation of populations:
what are the issues and what is the evidence? Pp. 85-120 in Ecology and
Management of Neotropical Migratory Birds. A synthesis and review of critical
issues (T.E. Martin and D. Finch, eds.). Oxford University Press, New York,
NY.
Shugart, H.H., Jr. and D. James. 1973. Ecological succession of breeding bird
populations in northwestern Arkansas. Auk 90:62-77.
Smith, D.M. 1986. The practice of silviculture. John Wiley & Sons, Inc., New York,
NY. 527 pp.
SPSS Inc. 1999. SPSS Base 10.0 applications guide. SPSS Inc., Chicago, IL. 426 pp.
Sustainable Forestry Board and American Forest and Paper Association. 2002. 20022004 SFI Standard and Verification Procedures. AF&PA. Washington, D.C.
Thompson, F.R., III and D.E. Capen. 1988. Avian assemblages in seral stages of a
Vermont forest. Journal of Wildlife Management 52:771-777.
Thompson, F.R., III; Fritzell, E.K. 1990. Bird densities and diversity in clearcut and
mature oak-hickory forest. North Central Forest Experiment Station, USDA
Forest Service, Research Paper NC-293, St. Paul, MN.
Thompson, F.R., III, W.D. Dijak, T.G. Kulowiec, and D.A. Hamilton. 1992. Breeding
bird populations in Missouri Ozark forests with and without clearcutting. Journal
of Wildlife Management 56:23-30.
Thompson, F.R., III, J.R. Probst, and M.G. Raphael. 1995. Impacts of silviculture:
overview and management recommendations. Pp. 201-219 in Ecology and
Management of Neotropical Migratory Birds. A synthesis and review of critical
issues (T.E. Martin and D.M. Finch, eds.). Oxford University Press, New York,
NY.
Traveset, A., N. Riera, and R.E. Mas. 2001. Passage through bird guts causes
interspecific differences in seed germination characteristics. Functional Ecology.
15: 669-675.
Van Horne, B. 1983. Density as a misleading indicator of habitat quality. Journal of
Wildlife Management 47:893-901.
70
Vickery, P.D., M.L. Hunter, Jr., and J.V. Wells. 1992. Use of a new reproductive
index to evaluate relationship between habitat quality and breeding success. The
Auk 109:697-705.
Welsh, C.J.E. and W.M. Healy. 1993. Effect of even-aged timber management on bird
species diversity and composition in northern hardwoods of New Hampshire.
Wildlife Society Bulletin 21: 143-154.
Williams, A.B. 1936. The composition and dynamics of a beech-maple climax
community. Ecological Monographs 6:317-408.
Yong, W., D.M. Finch, F.R. Moore, and J.F. Kelly. 1998. Stopover ecology and habitat
use of migratory Wilson's Warblers. Auk 115:829-842.
Zar, J.H. 1996. Biostatistical analysis. Prentice-Hall, Upper Saddle River, NJ. 662 pp.
71
VITA
Adrian Lesak, son of Alan Lesak and Laura (Nowak) Lesak, was born July 23,
1976, in Milwaukee, WI. He graduated with a Bachelor of Science degree from the
University of Wisconsin, Madison, WI in December 1998. He enrolled in the Graduate
School at Alabama A&M University, Normal, AL, in August 2002 with the intent of
earning the degree of Master of Science in the Department of Plant and Soil Science.
72