BREEDING BIRD RESPONSE TO GREENTREE
RESERVOIR MANAGEMENT
STEVEN P. CHRISTMAN, U.S. Fish and Wildlife Service, Denver Wildlife Research Center, 412 N.E. 16th Avenue, Gainesv
FL 32601
Abstract: Breeding season song counts in 1980 and 1981 revealed significant differences in relative abundance
for 12 of 28 nongame bird species between a greentree reservoir and an adjacent control area in
eastern Arkansas. Bird species that forage primarily in the understory were absent or occurred at lower
frequencies in the greentree reservoir. Species that hawk or forage primarily in the canopy were either not
affected or showed slight increases in frequency on the greentree reservoir transects. There were fewer
species of breeding birds and fewer total birds encountered on the greentree reservoir than on the control
area. Greentree reservoir management caused a reduction in understory vegetation that reduced foraging
and nesting opportunities for bird species that use the ground and lower levels of the forest.
J. WILDL. MANAGE. 48(4): 1164-1172
A greentree reservoir (GTR) is a temporary,
seasonal impoundment in a forested
river-bottom created to attract and
provide habitat for migrating waterfowl.
GTR's are usually flooded in the autumn
to a depth of a meter or so, and dewatered
in the spring before the new growing season.
Because impoundment occurs only
during the dormant season, tree mortality
is low (Broadfoot 1958) and the growth of
some native hardwood species may be enhanced
during times of drought (Broadfoot
and Williston 1973).
The first GTR's were constructed in
eastern Arkansas in the 1930's by private
duck hunting clubs (Rudolph and Hunter
1964). The level land, the impervious clay
soils, the abundance of oaks with small
acorns, and the thousands of ducks already
attracted to the ricefields of the
Grand Prairie and the bottomland hardwoods
of the Mississippi Delta provided
an ideal situation for this new management
tool.
Southern bottomland hardwood forests
typically flood naturally in late winter
when the rivers and bayous overflow their
banks due to reduced evaporation and
transpiration brought about by lower temperatures
and dormant vegetation (Wharton
and Brinson 1978, Wharton 1980). But
in years when flooding is late or never occurs,
GTR's provide dependable habitat
for waterfowl and improve duck hunting.
Greentree reservoirs have proven to be
an excellent management tool for wintering
and migrating waterfowl (Rudolph
and Hunter 1964, Hunter 1978), but little
is known about the effects of GTR's on
nontarget species. Fredrickson (1979,
1980) has provided some data on the effects
of various water management regimes
on forest wildlife and plants in
southeastern Missouri, but few observations
concerning GTR's. Krull (1969) and
Hubert and Krull (1973) contributed information
on aquatic macroinvertebrates
in temporary impoundments in New York
and Illinois. Filer (1975) described the effects
of seasonal impoundment on tree root
mycorrhizae and soil microflora in Mississippi
GTR's. Thompson et al. (1968)
provided some qualitative information
concerning the effects of seasonal impoundments
in New York on nongame
wildlife and plants.
1164 J. Wildl. Manage. 48(4):1984
This content downloaded from 147.251.87.220 on Tue, 03 Dec 2019 14:16:58 UTC
All use subject to https://about.jstor.org/terms
GREENTREE RESERVOIRS AND SONG BIRDS * Christman 1165
The present study was designed to determine
whether GTR management has
any effect on breeding songbird densities.
STUDY AREA
The study was conducted in eastern Arkansas
on the southern one-third of the
White River National Wildlife Refuge
(NWR), Arkansas and Desha counties. Located
within the floodplain of the lower
White River, just above its confluence with
the Mississippi River, the 45,902-ha refuge
consists of bottomland hardwood forests
interspersed with more than 160 lakes
(mostly old oxbows) and many kilometers
of sloughs and bayous. The White River
NWR contains the largest tract of bottomland
hardwood forest in public ownership
in the entire Mississippi Delta and is surrounded
by intensive agriculture where
soybeans, rice, winter wheat, and cotton
are produced.
All drainage is into the White River,
which flows roughly down the center of
the refuge. The area is extremely flat with
elevations ranging from 41 to 48 m. The
soils are poorly drained clays and silts:
Dundee silt loam is underlain by Sharkey,
Portland, or Dundee clays. Summers are
hot and winters are cool with an average
of about 80 days/year above 32 C and
about 50 days/year below 0 C. Mean annual
precipitation is 122 cm, with more
than half falling in the summer.
The habitat is characterized by annual,
prolonged flooding resulting from the
overflow of the White and the backingup
of the Mississippi Rivers. The inundation
may occur at any time of the year,
but typically begins in late winter or early
spring and persists for 1-4 months. Flood
levels in excess of 46 m above MSL may
inundate over 80% of the forest for several
weeks every year.
Forest types in the bottoms are described
by Eyre (1980) as overcup oakwater
hickory (SAF type 96), oak-American
elm-green ash (not recognized by
SAF), sugar hackberry-American elmgreen
ash (SAF type 93), sweetgum-Nuttall
oak-willow oak (SAF type 92), and
several other types of less frequent occurrence
(U.S. Fish and Wildl. unpubl. for.
manage. plan, White River NWR., 1980).
Sugar hackberry (Celtis laevigata) is the
most abundant tree species on the White
River NWR (Christman, unpubl. data).
Other important trees include overcup and
Nuttall oak (Quercus lyrata and Q. nuttallii),
water hickory (Carya aquatica),
sweetgum (Liquidambar styraciflua),
honeylocust (Gleditsia triacanthos), common
persimmon (Diospyros virginiana),
baldcypress (Taxodium distichum), and
several species of elm (Ulmus), ash (Fraxinus),
and maple (Acer). Common understory
species include possumhaw holly
(Ilex decidua), swamp privet (Ligustrum
acuminata), and buttonbush (Cephalanthus
occidentalis). The ground cover is
dominated by poison-ivy (Toxicodendron
radicans).
The White River NWR is managed for
wildlife, especially migratory waterfowl
(U.S. Fish and Wildl. Serv. 1980). Although
most of the forested bottomlands
within the refuge usually flood naturally
every year, about 4,000 ha are managed
as GTR's and inundated each October to
provide additional habitat (usually before
natural flooding) for waterfowl, principally
mallards (Anas platyrhynchos) and
wood ducks (Aix sponsa). The largest GTR
is formed by impounding Honey Locust
Bayou near its confluence with the White
River. Flooded every October since 1967,
the Honey Locust GTR inundates about
1,200 ha until April or until natural flood
waters recede (usually by late spring or
early summer), whichever comes later.
J. Wildl. Manage. 48(4):1984
This content downloaded from 147.251.87.220 on Tue, 03 Dec 2019 14:16:58 UTC
All use subject to https://about.jstor.org/terms
1166 GREENTREE RESERVOIRS AND SONG BIRDS * Christman
Table 1. Means number of song registrations for breeding bird species per 20-25-minute traversals of 5-ha strip transects in
the Honey Locust Greentree Reservoir (GTR) and nearby control areas (CON) for 1980 and 1981 surveys.
CON GTR
1980 1981 1980 1981
(N = 42) Mean (N = 48) (N = 32) Mean (N = 48)
Mourning dove
(Zenaida macroura) 0.42 0.55
Yellow-billed cuckoo
(Coccyzus americanus) 1.84 1.67
Eastern wood-pewee
(Contopus virens) 0.38 0.35 0.75 0.19
Acadian flycatcher
(Empidonax virescens) 1.87 2.10
Great crested flycatcher
(Myiarchus crinitus) 0.66 0.89
Carolina chickadee
(Parus carolinensis) 1.38 1.37
Tufted titmouse
(P. bicolor) 2.18*b 1.59*
White-breasted nuthatch
(Sitta carolinensis) 0.09 0
Carolina wren
- (Thryothorus ludovicianus) 2.70* 1.16*
Wood thrush
(Hylocichla mustelina) 0.62* 0.04 0* 0.17
Gray catbird
(Dumetella carolinensis) 0.10 0
White-eyed vireo
(Vireo griseus) 0.48 1.25* 1.28 0.42*
Yellow-throated vireo
(V. flavifrons) 0.26 0.33
Red-eyed vireo
(V. olivaceus) 0.27 0.50
Northern parula
(Parula americana) 0.43* 0.86*
Cerulean warbler
(Dendroica cerulea) 0.05 0.04
American redstart
(Setophaga ruticilla) 0 0.01
Prothonotary warbler
(Protonotaria citrea) 4.45* 2.54 1.78* 2.90
Swainson's warbler
(Limnothlypis swainsonii) 0 0.29* 0.03 0*
Kentucky warbler
(Oporornis formosus) 0.43* 0.08*
Hooded warbler
(Wilsonia citrina) 0.07 0
Yellow-breasted chat
(Icteria virens) 3.10* 0.79* 0.31* 0.04*
Summer tanager
(Piranga rubra) 0.84 0.80
Northern cardinal
(Cardinalis cardinalis) 1.65* 1.06*
Indigo bunting
(Passerina cyanea) 1.98 1.76
Rufous-sided towhee
(Pipilo erythrophthalmus) 0.17* 0*
Orchard oriole
(Icterus spurius) 3.69* 0.77 0.91* 0.67
J. Wildl. Mana
This content downloaded from 147.251.87.220 on Tue, 03 Dec 2019 14:16:58 UTC
All use subject to https://about.jstor.org/terms
GREENTREE RESERVOIRS AND SONG BIRDS * Christman 1167
Table 1. Continued.
CON GTR
1980 1981 1980 1981
(N = 42) Mean (N = 48) (N = 32) Mean (N = 48)
Northern oriole
(I. galbula) 0.06 0
Mean total registrations 32.95* 20.56 21.66* 17.35
Mean species richness 12.5* 10.80*
a The unweighted means of both years are provided when there
year are provided.
b * denotes pairs of treatment means that are different at an expe
This results in a bottomland hardwood
forest that is flooded for nearly twice as
long each year as the adjacent, nonmanipulated
forests.
METHODS
The Honey Locust GTR was not surveyed
before initial impoundment, and
there are no data concerning pretreatment
bird populations. In lieu of baseline
information, I have drawn inferences concerning
the response of songbird populations
to the management of Honey Locust
GTR by comparing relative abundances
with an adjacent bottomland forest in the
same drainage system.
I established 10 strip transects in the
Honey Locust GTR (4 in 1980 and 6 in
1981) and 13 control strips in a nearby
forested area (7 in 1980 and 6 in 1981)
not subject to GTR management. The
control areas were selected to represent
habitats similar to Honey Locust GTR except
for the water management regimes
imposed by man. Both areas are described
by Eyre (1980) as overcup oak-water
hickory (U.S. Fish and Wildl. Serv. 1980)
and have similar elevations. Each strip
transect was 500 m long by 100 m wide,
representing a sampling area of 5 ha. Both
the GTR and control areas were dry at
the time of the study.
Each transect was traversed on six or
eight rain-free mornings between sunrise
and 1000 local time in May and June 1980
and 1981 and territory-defending singing
birds were counted. I made no attempt to
count woodpeckers (Picidae), brownheaded
cowbirds (Molothrus ater), or blue
jays (Cyanocitta cristata), and one species
of singing bird (blue-gray gnatcatcher,
Polioptila caerulea) was excluded because
they were so abundant that counts were
unreliable. I recorded bird observations on
a portable tape recorder as I walked along
a transect at about 1 km/hour, resulting
in a sample of 20-25 minutes on each 5-ha
strip transect. For the purposes of comparing
two areas (GTR and control) simple
counts of singing birds (i.e., relative
measures of abundance) are adequate and
often preferred over more difficult, timeconsuming,
and sometimes questionable
estimates of absolute density (Temple 1981
and others in Ralph and Scott 1981). Relative
abundance is proportional to actual
density for a given species.
A two-way ANOVA was used to test the
hypotheses that the mean number of song
registrations for each of 28 species, the
mean number of species encountered, and
the mean total number of song registrations
for the two treatments (i.e., GTR and
control) over the 2 years were equal. Because
the data consisted of counts, often
including zeros, a square root transformation
(\/song registrations + 0.5) was
performed before statistical analysis (Steel
J. Wildl. Manage. 48(4):1984
This content downloaded from 147.251.87.220 on Tue, 03 Dec 2019 14:16:58 UTC
All use subject to https://about.jstor.org/terms
1168 GREENTREE RESERVOIRS AND SONG BIRDS * Christman
and Torrie 1960). To avoid an inflated
Type I error rate while maintaining an
experimentwise significance level of P <
0.05, I used the Bonferroni critical value
method for multiple comparisons (Harris
1975). Differences in relative abundance
were deemed significant when P < 0.0017
(0.05 divided by 30).
All trees over 15 cm DBH and within
10 m of the transect center line (=1-ha
sample) were identified, measured, and
counted. I used the plotless point-quarter
method (Ashby 1972) at each of 13 evenly
spaced points along each 1980 transect to
estimate the density of saplings in each of
three diameter size classes: <2.5 cm, 2.6-
7.6 cm, 7.7-15 cm. At each of these points
I also visually estimated the percentage of
cover by green vegetation within a circle
of radius = 2 m in two strata: ground to 1
m and 1-2 m above the ground. The untransformed
means were compared with
the t-test procedure.
RESULTS
The Honey Locust GTR did not differ
from the control area in mean dbh (39
and 49 cm, respectively) or density of trees
(360 trees/ha for both). However, the percentages
of cover by green vegetation from
the ground to 1 m and from 1 to 2 m
above the ground were significantly different
between the two study areas. The
GTR (N = 52) had an average of 18%
green cover in the lower stratum and 20%
in the higher, compared with 49% and
40%, respectively, for the same two layers
in the control area (N = 78). These means
are different (0-1 m: P < 0.007; 1-2 m:
P < 0.001) and show that the Honey Locust
GTR has less green vegetation in the
understory and ground layers than the adjacent,
nonmanipulated forest. The Honey
Locust GTR had fewer vines and
shrubs, fewer low-level tree branches, and
the ground was often bare of green vege-
tation.
The two-way ANOVA showed that 7 of
the 28 breeding bird species considered
had significant treatment (i.e., GTR or
control) by year interactions. For these
species, and for the total number of registrations
(all species combined), it was
necessary to examine the effects of GTR
management separately for each year of
the study. The remaining 21 species and
the total number of species encountered
did not have significant treatment by year
interactions, therefore I was able to combine
the results for the 1980 and 1981 surveys
(Table 1).
The relative abundances of tufted titmice,
Carolina wrens, Kentucky warblers,
yellow-breasted chats, northern cardinals,
and rufous-sided towhees were lower (P _
0.0017) in the GTR than in the control
area for both years of the study (see Table
1 for scientific names). Also, the total
number of species encountered (=species
richness) was significantly lower in the
GTR than in the control area for both
years.
The total number of registrations from
all species combined was lower in the GTR
for 1980, as were the relative abundances
of wood thrushes, prothonotary warblers,
and orchard orioles. White-eyed vireos and
Swainson's warblers had lower relative
densities in the GTR in 1981. No species
showed reversal in relative abundance
pattern between the 2 years of the study,
and only the southern parula was more
abundant in the GTR, and it was so for
both years.
Five of the species considered were apparently
not affected by GTR management.
These included the mourning dove,
yellow-billed cuckoo, Carolina chickadee,
summer tanager, and indigo bunting. The
number of observations were not suffiJ.
Wildl. Manage. 48(4):1984
This content downloaded from 147.251.87.220 on Tue, 03 Dec 2019 14:16:58 UTC
All use subject to https://about.jstor.org/terms
GREENTREE RESERVOIRS AND SONG BIRDS * Christman 1169
cient to identify any tendencies for the
white-breasted nuthatch, gray catbird,
cerulean warbler, American redstart,
hooded warbler, and northern oriole. The
remaining five species, the eastern woodpewee,
Acadian flycatcher, great crested
flycatcher, yellow-throated vireo, and redeyed
vireo had slightly higher relative
abundances in the GTR than in the control
area, but the differences were not sig-
nificant.
DISCUSSION
Most observers (but see Anderson and
Shugart [1974] for an exception) consider
the Carolina wren and yellow-breasted
chat to be species of the forest understory,
dependent on shrubs and understory
vegetation for nesting and foraging (Bent
1948, 1953; James 1971; Whitmore 1977;
Dickson and Noble 1978). These two
species occurred in lower relative densities
in the Honey Locust GTR than in the
adjacent control areas for both years of
the study. The wood thrush and the rufous-sided
towhee are both ground foragers
(Bertin 1977, Dickson and Noble
1978, Terres 1980), and, although neither
species was especially abundant on the
White River NWR, both were virtually
excluded by GTR management.
The prothonotary, Swainson's, and
Kentucky warblers are also species of the
forest understory (Bent 1953, Meanley
1971, Terres 1980), and each of these was
reduced by GTR management. Dickson
and Noble (1978) considered the whiteeyed
vireo to be a midstory species, but
my observations suggest and Terres (1980)
stated that this species forages mainly in
low-level shrubs and is seldom seen far
above the ground. The different results I
obtained in the 1980 and 1981 surveys
may be explained by small-scale habitat
heterogeneity. On three of the 1980 GTR
transects, small (<100-m2) brushy areas,
consisting mainly of fallen trees covered
with vines (Campsis radicans, Ampelopsis
arborea, Berchemia scandens, etc.)
harbored white-eyed vireos that were recorded
on nearly every traversal. The 1981
GTR transects happened (by chance) to
include fewer of these gaps and thus fewer
white-eyed vireo territories. I believe
white-eyed vireos require a shrubby (or
vine) understory for nesting and foraging
but can utilize a small patch, probably less
than 100 m2. The relative abundance of
white-eyed vireos was lower in the GTR
than in the control area for 1981, but not
in 1980.
The orchard oriole is one of the most
abundant breeding birds on the White
River NWR and seems to be flexible in its
habitat selection (Dennis 1948, Bent 1958).
Orchard orioles are known to nest in colonies
(Dennis 1948) and a concentration
of nesting orchard orioles was encountered
on the control area in 1980. Orchard
orioles were observed in relatively equal
numbers on both areas in 1981, however.
The northern cardinal, considered a
ground and understory forager by Dow
(1969) and a forager of the low canopy by
Samson (1979), had lower relative abundances
in the Honey Locust GTR for both
years of the study. The tufted titmouse
also showed a reduced relative abundance
in the GTR for both years. Hardin and
Evans (1977) claimed that titmice usually
occurred in "thick growth" but Anderson
and Shugart (1974:834) stated that titmice
selected areas with "an open understory
and a well developed sub-canopy." The
Honey Locust GTR has an open understory,
but tufted titmice were less common
there than in the more shrubby control
areas. This suggests that tufted titmice
at White River are more dependent on
understory vegetation than commonly beJ.
Wildl. Manage. 48(4):1984
This content downloaded from 147.251.87.220 on Tue, 03 Dec 2019 14:16:58 UTC
All use subject to https://about.jstor.org/terms
1170 GREENTREE RESERVOIRS AND SONG BIRDS * Christman
lieved for other areas. The gray catbird
and the hooded warbler are both ground
and shrub-level foragers (Bent 1953,
Terres 1980) and at White River NWR
usually are encountered in somewhat better-drained
sites than my study areas. Both
species occurred sporadically in the control
area but not at all in Honey Locust
GTR.
The breeding season relative abundances
of all three species of flycatchers
tended to be higher in the GTR than in
the control areas. According to Hespenheide
(1971), eastern wood-pewees and
Acadian flycatchers select breeding habitats
that are structurally different from
each other and the two species rarely coexist
with the larger great crested flycatcher.
In the White River bottoms, all
three species frequently occurred on the
same transects and their respective responses
to GTR management were similar.
In a general way, the reduction of
understory vegetation, including shrubs
and saplings, probably improves the foraging
habitat for the sallying flycatchers.
The yellow-throated vireo, red-eyed
vireo, and northern parula can be characterized
as species of the larger trees
(Samson 1979), if not the canopy proper
(see James 1971). Higher breeding season
densities in the GTR for these species were
not predicted on the basis of known effects
of GTR management (i.e., a reduction
in understory vegetation). The reason
for the observed differences (significant
only for the parula) in relative densities
between the control and GTR study areas
for these three species is not apparent.
Stauffer and Best (1980) provided a table
of bird species (14 of which are in
common with this study) and the predicted
effects on breeding season densities of
various habitat alterations. Among the
habitat alterations included was, "shrubs/
saplings thinned." Based on the literature
and their own research, they indicated that
thinning the shrubs and saplings would
lower breeding season densities of rufoussided
towhees, wood thrushes, northern
cardinals, and gray catbirds. The Honey
Locust GTR is a bottomland hardwood
forest in which the shrubs and saplings
have been thinned by prolonged winter
and spring flooding. The results of my
study confirm that these species are reduced
by GTR management. They also
predicted that breeding season densities of
the indigo bunting would be depressed by
thinning the understory vegetation, but
my data contradict this prediction and
suggest that bunting densities are not affected
at all by GTR management.
Data presented here tend to support
Stauffer and Best's (1980) prediction that
breeding season densities of great crested
flycatchers will increase with understory
thinning, but their prediction that tufted
titmouse densities should also increase is
strongly contradicted. They further predicted
that breeding season densities of
mourning doves, northern orioles, redeyed
vireos, and eastern wood-pewees
would not be affected by understory thinning.
Based on the present study, this may
be true for mourning doves and northern
orioles, but is probably not true for redeyed
vireos and eastern wood-pewees,
which seemingly have higher breeding
season densities in the GTR. For the yellow-billed
cuckoo, white-breasted nuthatch,
and American redstart, they were
unable to predict an effect from shrub and
sapling thinning. My data suggest that
yellow-billed cuckoos are not affected by
GTR management; the data are insufficient
to allow comment on the other two
species.
Samson (1979) listed several species of
North American forest birds and assigned
J. Wildl. Manage. 48(4):1984
This content downloaded from 147.251.87.220 on Tue, 03 Dec 2019 14:16:58 UTC
All use subject to https://about.jstor.org/terms
GREENTREE RESERVOIRS AND SONG BIRDS * Christman 1171
each to a "feeding guild," based on: (1)
primary food, (2) primary foraging stratum,
and (3) primary foraging behavior.
His list included all but two (orchard oriole
and summer tanager) of the species
included in this study. Among the 17
species for which differences in breeding
season relative abundance were suggested
(Table 1) or detected between the GTR
and the control, all of the "salliers" and
all of the species that forage in the "high
canopy" had slightly higher relative densities
in the GTR. The northern parula,
which, according to Samson (1979), forages
in the "middle canopy," had a higher
density in the GTR. On the other hand,
all of the "ground foragers" and all of the
"low canopy" species (except the sallying
Acadian flycatcher) had lower relative
densities in the GTR as compared to the
control area. (I assume that Samson erred
in assigning the rufous-sided towhee to the
"middle canopy" guild; Willson [1974]
provided a similar list of bird guilds, but
included the towhee as a "ground forager.")
Thus, the sallying and high-canopy
gleaning guilds probably are favored by
GTR management, but the ground and
low-canopy guilds certainly are adversely
impacted. The species for which I was unable
to detect a GTR effect included one
sallier, one high-canopy gleaner, four
middle-canopy gleaners, two low-canopy
gleaners, one ground forager, and one bark
forager.
Wildlife managers in the Mississippi
Valley create GTR's for migratory waterfowl
by impounding water during the fall
and winter in forested river-bottom habitats.
Although the larger trees seemingly
are not affected, the understory vegetation
is noticeably reduced. As a result,
densities of breeding forest birds that use
the lower levels of vegetation for nesting
and foraging also are reduced. Because a
higher proportion of birds in the forest
community are in the ground and lowlevel
guilds (Samson 1979), the number of
species and the total density of nongame
birds nesting in greentree reservoirs will
be lower than in comparable, nonmanipulated
habitats.
LITERATURE CITED
ANDERSON, S. H., AND H. H. SHUGART. 1974. Habitat
selection of breeding birds in an east Tennessee
deciduous forest. Ecology 55:828-837.
ASHBY, W. C. 1972. Distance measurements in
vegetation study. Ecology 53:980-981.
BENT, A. C. 1948. Life histories of North American
nuthatches, wrens, thrashers, and their allies. U.S.
Natl. Mus. Bull. 195. 459pp.
1953. Life histories of North American
warblers. U.S. Natl. Mus. Bull. 203. 734pp.
1958. Life histories 6f North American
blackbirds, orioles, tanagers and allies. U.S. Natl.
Mus. Bull. 211. 547pp.
BERTIN, R. I. 1977. Breeding habitats of the wood
thrush and veery. Condor 79:303-311.
BROADFOOT, W. M. 1958. Reaction of hardwood
timber to shallow-water impoundments. Mississippi
State Univ. Agric. Exp. Stn. Inf. Sheet 595.
2pp.
-- , AND H. L. WILLISTON. 1973. Flooding effects
on southern forests. J. For. 71:584-587.
DENNIS, J. V. 1948. Observations on the orchard
oriole in lower Mississippi Delta. Bird-Banding
19:12-21.
DICKSON, J. G., AND R. E. NOBLE. 1978. Vertical
distribution of birds in a Louisiana bottomland
hardwood forest. Wilson Bull. 90:19-30.
Dow, D. D. 1969. Home range and habitat of the
cardinal in peripheral and central populations.
Can. J. Zool. 47:103-114.
EYRE, F. H. 1980. Forest cover types of the United
States and Canada. Soc. Am. For., Washington,
D.C. 148pp.
FILER, T. H., JR. 1975. Mycorrhizae and soil microflora
in a green-tree reservoir. For. Sci. 21:
36-39.
FREDRICKSON, L. H. 1979. Floral and faunal
changes in lowland hardwood forests in Missouri
resulting from channelization, drainage, and impoundment.
U.S. Fish Wildl. Serv. FWS/OBS-
78/91. 130pp.
1980. Impact of water management on the
resources of lowland hardwood forest. Annu. For.
Symp., Louisiana State Univ., Baton Rouge 27:
51-64.
HARDIN, K. I., AND K. E. EVANS. 1977. Cavity nesting
bird habitat in the oak-hickory forest-a reJ.
Wildl. Manage. 48(4):1984
This content downloaded from 147.251.87.220 on Tue, 03 Dec 2019 14:16:58 UTC
All use subject to https://about.jstor.org/terms
1172 GREENTREE RESERVOIRS AND SONG BIRDS * Christman
view. U.S. For. Serv. Gen. Tech. Rep. NC-30.
23pp.
HARRIS, R. T. 1975. A primer of multivariate statistics.
Academic Press, New York, N.Y. 332pp.
HESPENHEIDE, H. A. 1971. Flycatcher habitat selection
in the eastern deciduous forest. Auk 88:
61-74.
HUBERT, W. A., AND J. N. KRULL. 1973. Seasonal
fluctuations of aquatic macroinvertebrates in
Oakwood Bottoms Greentree Reservoir. Am.
Midl. Nat. 90:177-185.
HUNTER, C. G. 1978. Managing green tree reservoirs
for waterfowl. Int. Waterfowl Symp., New
Orleans, La. 3:217-223.
JAMES, F. C. 1971. Ordinations of habitat relationships
among breeding birds. Wilson Bull. 83:215-
236.
KRULL, J. N. 1969. Seasonal occurrence of macroinvertebrates
in a green-tree reservoir. N.Y.
Fish and Game J. 16:119-124.
MEANLEY, B. 1971. Natural history of the Swainson's
warbler. North Am. Fauna 69. 90pp.
RALPH, C. J., AND J. M. SCOTT, editors. 1981. Estimating
numbers of terrestrial birds. Stud. Avian
Biol. 6. 630pp.
RUDOLPH, R. R., AND C. G. HUNTER. 1964. Green
trees and greenheads. Pages 611-618 in J. P.
Lindusky, ed. Waterfowl tomorrow. U.S. Dep.
Inter., Washington, D.C.
SAMSON, F. B. 1979. Lowland hardwood bird communities.
Pages 49-66 in R. M. DeGraaf, tech.
coord. Management of north central and northeastern
forest for nongame birds. U.S. For. Serv.
Gen. Tech. Rep. NC-51.
STAUFFER, D. F., AND L. B. BEST. 1980. Habitat
selection by birds of riparian communities: evaluating
effects of habitat alterations. J. Wildl.
Manage. 44:1-15.
STEEL, R. G. D., AND J. H. TORRIE. 1960. Principles
and procedures of statistics. McGraw-Hill
Book Co., New York, N.Y. 481pp.
TEMPLE, S. A. 1981. Summarizing remarks: estimating
relative abundance. Part II. Page 112 in
C. J. Ralph and J. M. Scott, eds. Estimating numbers
of terrestrial birds. Stud. Avian Biol. 6.
TERRES, J. K. 1980. The Audubon Society encyclopedia
of North American birds. Alfred A. Knopf
Co., New York, N.Y. 1109pp.
THOMPSON, D. Q., P. B. REED, JR., G. E. CUMMINGS,
AND E. KIVISALU. 1968. Muck hardwoods as
green-timber impoundments for waterfowl.
Trans. North Am. Wildl. and Nat. Resour. Conf.
33:142-159.
WHARTON, C. H. 1980. Values and functions of
bottomland hardwoods. Trans. North Am. Wildl.
and Nat. Resour. Conf. 45:341-353.
), AND M. M. BRINSON. 1978. Characteristics
of southeastern river systems. Pages 32-40 in R.
R. Johnson and J. F. McCormick, tech. coords.
Strategies for protection and management of
floodplain wetlands and other riparian ecosystems.
U.S. For. Serv. Gen. Tech. Rep. WO-12.
WHITMORE, R. C. 1977. Habitat partitioning in a
community of passerine birds. Wilson Bull. 89:
253-265.
WILLSON, M. F. 1974. Avian community organization
and habitat structure. Ecology 55:1017-
1029.
Received 28 December 1982.
Accepted 8 December 1983.
J. Wildl. Manage. 48(4):1984
This content downloaded from 147.251.87.220 on Tue, 03 Dec 2019 14:16:58 UTC
All use subject to https://about.jstor.org/terms