Southeastern Naturalist
S.A. Rush, E.C. Soehren, and M. Miller
2014 Vol. 13, No. 1
92
2014 SOUTHEASTERN NATURALIST 13(1):92–100
Stopover Duration and Habitat Use by Tennessee Warblers
(Oreothlypis peregrina) at a High-Elevation Bald
Scott A. Rush1,*, Eric C. Soehren2, and Mary Miller3
Abstract - Use of high-elevation balds by Nearctic-Neotropical passerines remains a poorly
understood component of migration ecology. We used radio-telemetry to assess the habitat
use of 20 Oreothlypis peregrina (Tennessee Warbler) at a high-elevation bald during September
2011. Our goals were to estimate the duration of stopover, examine variation in stopover length
by age class, and determine patterns of habitat use during fall migration. For radio-tracked
birds, the average length of time spent at the bald was 4.6 days (range: 1–12 days). The majority
of tracked locations corresponded with heath habitat that surrounded the bald opening
(79% of locations), relative to the more distal and less complex, hardwood forest. Collectively,
these results provide information on the important, yet ephemeral role that high-elevation
balds play in supporting the autumnal migration of eastern North American passerines.
Introduction
Nearctic-Neotropical songbirds may spend up to a third of a year in migration, and
most spend the majority of that time at stopover sites (Mehlman et al. 2005). Stopover
sites therefore play a critical role in providing migrating passerines some key resources
needed to complete these epic journeys. As a consequence, understanding the impact
of environmental changes such as habitat loss or fragmentation along their migratory
routes is essential for the implementation of effective conservation schemes (Buler et
al. 2007, Hutto 1998, Mehlman et al. 2005). Although migrating passerines use both
coastal and inland routes, the majority of stopover studies have focused on coastal
locations (e.g., Bonter et al. 2007, Moore and Kerlinger 1987). Focus on coastal stopover
sites has left much of the stopover ecology at inland sites unexplored (Vogt et
al. 2012). We have limited understanding of how songbirds use inland stopover sites,
including the duration of time spent at these locations. For instance, relatively little is
known about age-related differences in fat deposition and habitat selection by migrating
passerines (Jones et al. 2002); a case especially true for inland stopover sites.
Studies focused on the ecology of passerines migrating through inland sites
are limited. Of the existing studies, several have identified differences in stopover
length (Ellegren 1991, Jones et al. 2002) and pattern of mass deposition (Ellegren
1991) between young and adult migrant passerines (Wang et al. 1998). Several
factors can contribute to age-related differences in mass gain among migrant
1Department of Wildlife, Fisheries and Aquaculture, 775 Stone Boulevard, Mississippi State
University, Mississippi State, MS 39762. 2Elhew Field Station, Wehle Land Conservation
Center, State Lands Division, Alabama Department of Conservation and Natural Resources,
4819 Pleasant Hill Road, Midway, AL 36053. 3United States Department of Agriculture,
Forest Service, Cherokee National Forest, 2800 Ocoee Street North Cleveland, TN 37312.
*Corresponding author - srush@cfr.msstate.edu.
Manuscript Editor: Wylie Barrow
Southeastern Naturalist
93
S.A. Rush, E.C. Soehren, and M. Miller
2014 Vol. 13, No. 1
passerines. Young birds may be less efficient fliers than adults, causing them to
burn more energy during migration and arrive at stopover sites with more depleted
energy reserves (Wang and Moore 1994). Further, upon arrival at stopover sites,
young birds may be less efficient foragers than adults, choose less nutritious foods,
or forage in sub-optimal habitats (Chernetsov 2006, Morris et al. 1996, Wunderle
1991). Patterns of habitat use therefore appear to be very important factors for determining
the success of migration. Because the benefits of a stopover site can be
determined by factors measureable at both localized and landscape scales, derived
benefits can also vary along the migratory route (T ankersley and Orvis 2003).
For Oreothlypis peregrina (Wilson) (Tennessee Warbler), the primary pathway
for fall migration appears to be inland, along a route that includes the southern Appalachian
Mountains (Rimmer and McFarland 1998). High-elevation grasslands
or “balds”, an early successional habitat type unique to the southern Appalachian
Mountains represent “habitat islands” that are used by passerines regularly during fall
migration (Vogt et al. 2012). Thus, the inter-year recapture of 14 Tennessee Warblers
(unprecedented for any migrant Parulidae) at one high-elevation bald, Whigg Meadow
in eastern Tennessee, suggests that this location may be along a route consistently
used during this species’ autumnal migration (Vogt et al. 2012). Regular use of Whigg
Meadow by Tennessee Warblers during fall migration highlights the importance of
conserving this site. It may also indicate the additional need for early successional
habitats, and maintenance of other high-elevation balds within the southern Appalachians;
especially those habitats used by other species of conservation need including
Vermivora chrysoptera (L.) (Golden-winged Warbler) and Sylvilagus obscurus Chapman,
Cramer, Deppenaar, and Robinson (Appalachian Cottontail).
Our objective was to identify habitat use by Tennessee Warblers at Whigg Meadow
during fall migration. Our specific goals were to determine patterns of habitat use,
estimate the duration of stopover periods, and examine variation in stopover length
between younger and older Tennessee Warblers using this high-elevation bald.
Study Site
This project was carried out at Whigg Meadow, a high-elevation site in eastern
Tennessee (35°189'N, 84°029'W; elevation: 1490 m; Monroe County; Fig. 1).
Whigg Meadow is located in the Tellico Ranger District of the Cherokee National
Forest and is comprised of 2 forest stands, reflecting a high-elevation bald with an
opening approximately 3 ha in size. The dominant habitat type at Whigg Meadow
is a forest-edge ecotone that transitions between open grassland and northern hardwood
forest. A shrub-dominated assemblage surrounds the herbaceous bald with a
dense thicket of Rubus canadensis L. (Smooth Blackberry) and scattered Vaccinium
corymbosum L. (Highbush Blueberry), encroaching upon the open herbaceous area.
A forest dominated by second-growth or stunted Fagus grandifolia Ehrh. (American
Beech) and Betula alleghaniensis Britton (Yellow Birch) can be found north of
the bald opening, while Acer saccharum Marshall (Sugar Maple), Quercus rubra
L. (Northern Red Oak), and Crataegus crus-galli L. (Cockspur Hawthorn) become
more dominant south of the opening (TDEC 1999).
Southeastern Naturalist
S.A. Rush, E.C. Soehren, and M. Miller
2014 Vol. 13, No. 1
94
Methods
Radio telemetry
Fall mist-netting and banding efforts have been performed annually at Whigg
Meadow since 1998 (Vogt et al. 2012), and fieldwork for this project was carried
out during September 2011.
Data collected from migration monitoring at Whigg Meadow indicates that the
duration of Tennessee Warblers’ migration stopovers at this site differs by age class
(D.F. Vogt et al., Cherokee National Forest, Monroe County, TN, unpubl. data).
Figure 1. Location of Whigg Meadow in eastern Tennessee. The lowest panel,
reflecting the highest level of resolution, shows the distribution of telemetered
Tennessee Warblers (black dots) at Whigg Meadow during September 2011. The
darker shaded polygon reflects edge/heath habitat that encircles the grassy bald
opening (lighter shaded polygon).
Southeastern Naturalist
95
S.A. Rush, E.C. Soehren, and M. Miller
2014 Vol. 13, No. 1
However, several factors such as age-related recapture probability can influence
this estimate. Clarification of this potentially confounded result could be achieved
by radio-tracking both hatch-year (HY) and after-hatch-year (AHY) birds and
evaluating differences in stopover duration. Therefore, a goal of the project was to
outfit 10 HY and 10 AHY birds with radio-transmitters. Transmitters used for this
study (Model A2412, ATS, Isanti, Minnesota) weighed 0.28 g (≈3% of adult body
weight) and had a battery life of 23 days, a longevity that exceeded the estimated
maximum stopover duration of Tennessee Warblers (14 days) calculated from banding
data (analyzed using mark-recapture models) collected during 2001–2008 (D.F.
Vogt, unpubl. data).
Beginning 1 September 2011, we captured Tennessee Warblers incidentally using
mist nets (see Vogt et al. 2012 for description of methods). We banded all captured
birds using USGS aluminum bands and aged them following Pyle (1997). We
examined each bird in hand to determine the amount of subcutaneous fat visible in
the furculum of the clavicle (tracheal pit) and assigned a “fat” score ranging from
0–5, where 0 = no visible fat and 5 = fat overflowing the tracheal pit (Krementz and
Pendleton 1990).
Of the Tennessee Warblers captured, 10 HY and 10 AHY birds were selected for
this radio-telemetry study on the basis of “first-come, first-served”. We attached
radio-transmitters to each bird using a small square of chiffon, which was affixed
to each transmitter and attached to the interscapular region of each bird using
cyanoacrylate glue (Johnson et al. 1991). We used a Biotracker telemetry receiver
(ATS, Isanti, MN) with a three-element Yagi antenna to relocate radio-marked individuals.
We determined the estimated detection range for these transmitters in this
study area to be 0.5–1.5 km.
During this study, we located radio-marked Tennessee Warblers three times a
day by homing in on their location (White and Garrott 1990). Successive tracking
sessions for each bird were separated by at least 3 hrs. During tracking events, we
confirmed the location of each bird visually or by tracking to the location where the
strongest radio signal was identified and species-specific vocalizations were heard.
We recorded each location using a handheld GPS device (Garmin GPS 60, Garmin,
Olathe, KS; average measured error ± 3.7 m). During radio-tracking activities, we
considered a bird to have left the stopover site if no signal was received for at least
three consecutive days throughout the bald, with departure time defined as night
following the last confirmed observation.
Habitat measurements
We created a land-cover map as a GIS layer in ArcMap 9.3 (ESRI, Redlands,
CA) from aerial photographic material completed with field verifications. Based on
direct observations, we considered three different habitats as dominating the study
area: grassy meadow habitat (primarily Smooth Blackberry with scattered Highbush
Blueberry), heath or forest-edge ecotone (transition between open grassland
and northern hardwood forest), and hardwood forest. We plotted the locations of
radio-tracked Tennessee Warblers on these maps to determine habitat use within
each bird’s range of movements, as defined by radio-telemetry locations.
Southeastern Naturalist
S.A. Rush, E.C. Soehren, and M. Miller
2014 Vol. 13, No. 1
96
We assessed structural differences in habitat by overlaying a grid of 50-m x 50-m
squares on top of aerial imagery of the study site. We classified habitat intersecting
each corner of the overlain grid by identifying the dominant vegetation class at each
location. Additional assessments collected for the habitat at each sampling point
included canopy cover, canopy height, density of vegetation <1 m and 1–2 m above
ground, and the distribution of tree sizes (diameter at breast height: DBH).
Following methods detailed in James and Shugart (1970), we centered a 5-mradius
circular plot on the location corresponding to each grid intersection. At each
plot center, we measured the DBH (1.4 m above the ground surface) of all trees
>2 m tall with a dbh >5 cm. At each of the four cardinal directions 5 m from the center
of the plot, we measured the average height of the vegetation using a clinometer
(Forestry Suppliers, Inc. Jackson, MS). We measured the vertical distribution
of foliage in each cardinal direction by placing a 2-m graduated pole, marked at
decimeter intervals vertically in the vegetation, and recording the number of times
foliage hit the pole within each increment. We also calculated canopy cover at 5 m
from plot center in each cardinal direction using a concave spherical densitometer
(Forestry Suppliers, Inc. Jackson, MS). We averaged repeated measurements obtained
for each plot prior to further analysis.
Statistical analysis
We performed all statistical analyses using the statistical package R (Version
2.14.2; R Development Core Team 2012). Prior to analysis, we tested all data for
normality and homogeneity of variance using probability plots. We assessed mean
differences in the duration of time spent at Whigg Meadow between Tennessee
Warbler age classes using t-tests, with the degrees of freedom conditioned for
unequal variance (i.e., Welch’s t-tests; Ruxton 2006). We explored relationships between
the fat index of Tennessee Warblers and time spent at Whigg Meadow using
linear models. We used multiple analysis of variance (MANOVA) to test for the
effect of site (heath vs. forest habitat) on habitat measurements, and then applied
analysis of variance to specify significant dif ferences found in the MANOVA.
Results
During 1–29 September 2011, we captured 20 Tennessee Warblers and outfitted
them with radio-transmitters. For these birds, the average length of time spent at
Whigg Meadow was 4.6 days (min, max: 1–12 days), with a maximum length of
stay similar to that identified through banding efforts (D.F. Vogt, unpubl. data). On
average, older Tennessee Warblers tended to stay longer at Whigg Meadow than
younger birds (AHY mean: 5.7 days; HY mean: 3.4 days), but this difference was
not statistically significant (t = 1.34, df = 17.9, P = 0.20). There was no difference in
the time Tennessee Warblers spent at Whigg Meadow relative to the bird’s fat index
for either young (HY: F1,8 = 0.04, P = 0.85) or older individuals (AHY: F1,8 = 1.46,
P = 0.26).
Multiple analysis of variance of habitat metrics revealed significant differences
between heath and forest habitats (Wilk’s l = 0.32, df = 1, 6, P = 0.01; Table 1).
Southeastern Naturalist
97
S.A. Rush, E.C. Soehren, and M. Miller
2014 Vol. 13, No. 1
Habitat differences between heath and forest locations were less canopy cover
(F1,18 = 6.81, P = 0.02) and shorter canopy height (F1,18=27.95, P < 0.01) at heath
locations. Heath habitat was also characterized by more dense vegetation less than 1 m
above ground (F1,18 = 12.86, P < 0.01) and 1–2 m above ground (F1,18 =16.82, P less than
0.01). There was no difference in basal area measured in heath or forest habitats
(mean DBH per ha heath = 567.91 m2, mean DBH per ha forest = 658.81 m2; t =
-0.66, df = 15.52, P = 0.52). There was, however, a greater proportion of smaller
trees (those with DBH ≤ 20 cm) in heath relative to forest habitat (Fig. 2), but a
greater proportion of larger trees (with DBH > 20 cm) in the forest habitat.
The majority of locations identified for the radio-tracked birds (n = 164 locations
for 20 individual Tennessee Warblers) corresponded with the heath habitat (79% of
locations), with considerably fewer locations in the hardwood forest habitat (21%)
(Fig. 1). There was no apparent directionality relative to the time of the day when
the birds were tracked (Fig. 1).
Discussion
Results from the present study support earlier findings (Vogt et al. 2012) that
Tennessee Warblers use the early successional heath/edge habitat at Whigg Meadow
during fall migration. This heath habitat is characterized by low canopy cover,
shorter canopy height, and greater vegetation structure within 2 m of the ground.
The majority of locations identified for radio-tracked Tennessee Warblers occurred
within this heath habitat, whereas considerably fewer locations were within the
surrounding forest. Furthermore, for the 20 Tennessee Warblers that were radiotracked
during 2011, significantly more locations were determined to occur in the
southern portion of Whigg Meadow.
Little information exists on the directionality of diurnal movements and habitat
use by migrating songbirds during stopover events. For instance, Chernetsov (2011)
found no observable pattern in the directionality of movements among several species
at stopover locations on the Baltic coast. Although our present study is focused
on one location and one migratory period, it is of interest whether the behaviors
observed for Tennessee Warblers are relative to resources at this site alone or reflect
more general behaviors similar among other passerine species at other southern
Appalachian balds.
Table 1. Habitat characteristics evaluated at sample locations within heath or forest habitat at Whigg
Meadow during 2011. Measurements are provided as mean (minimum–maximum).
Heath Forest
Canopy cover (%) 83.84 (51.04–95.84) 95.19 (92.20–96.62)
Canopy height (m) 10.7 (2.5–20.0) 23.8 (17.0–29.0)
Density of vegetation less than 1 m above ground
(# of foliage hits/m) 2.90 (1.00–6.45) 1.24 (0.75–1.75)
Density of vegetation 1 m to 2 m above ground
(# of foliage hits/m) 1.41 (0.50–2.75) 0.37 (0.00–0.75)
Basal area (m2/ha) 567.92 (132.1–1159.4) 658.81 (230.0–1032.1)
Southeastern Naturalist
S.A. Rush, E.C. Soehren, and M. Miller
2014 Vol. 13, No. 1
98
In general, high-elevation balds have received little attention, yet these ecosystems
are used regularly by transient Tennessee Warblers and other species
of migratory passerines (Vogt et al. 2012). Anticipated habitat change at Whigg
Meadow, whether managed or not, will determine the suitability of these locations
for future migrations. For instance, high-elevation balds such as Whigg Meadow
are experiencing a shift in the forest boundary between heath and the forest community
that is threatening these isolated communities (White and Sutter 1998).
Without active management, it is anticipated that woody encroachment will reduce
the herbaceous community and eventually the extent of heath at these sites.
Although the experimental design employed in this study does not provide
a clear assessment of whether increased availability of heath habitat would
Figure 2. Tree distribution
by DBH for heath
(top graph) and forest
(bottom graph) habitats
at Whigg Meadow during
2011. Percentages reflect
the proportion of basal
area in each DBH category
multiplied by 100,
while the y-axis, “Frequency”,
denotes the cumulative
distribution of
each size class.
Southeastern Naturalist
99
S.A. Rush, E.C. Soehren, and M. Miller
2014 Vol. 13, No. 1
definitively benefit Tennessee Warblers, results do show that heath habitat is the
dominant habitat type used by this species during autumn migration. The present
study suggests that habitat management undertaken at Whigg Meadow with the aim
of aiding species such as the Tennessee Warbler should focus on expanding and
conserving the availability of heath/edge habitat. Coupling this management goal
with those aimed at supporting Golden-winged Warblers and Appalachian Cottontail
will likely afford protection to an array of species requiring this early successional
habitat type.
This study, and the findings of others (Vogt et al. 2012), highlight the importance
of Whigg Meadow and the southern Appalachian Mountains as migratory stopover
sites used by Tennessee Warblers. Additional research addressing the role of specific
southern Appalachian habitats, such as high-elevation balds and adjacent forest,
in supporting songbird migration is needed. For instance, inference drawn from
the current study relates to one bald only. Further studies focused on habitat use at
multiple balds could provide greater depth to understanding and developing best
management practices. The placement of additional migratory monitoring stations,
coupled with passive monitoring techniques such as acoustic monitoring arrays
(Blumstein et al. 2011) could provide additional data concerning the importance of
these stopover locations and important information key to the conservation of this
unique ecological community.
Acknowledgments
This research was made possible through the tireless efforts of numerous volunteers,
specifically Joe Carter, Nelson Edwards, Laura Lewis, John Trent, David Vogt, and Hayden
Wilson. Funding and logistical support was provided by the USDA Forest Service, Cherokee
National Forest and the Department of Wildlife, Fisheries and Aquaculture at Mississippi
State University. We are indebted to John Trent for graphical assistance. We gratefully acknowledge
the material support ATS provided to this project.
Literature Cited
Blumstein, D.T., D.J. Mennill, P. Clemins, L. Girod, K. Yao, G. Patricelli, J.L. Deppe, A.H.
Krakauer, C. Clark, K.A. Cortopassi, S.F. Hanser, B. McCowan, A.M. Ali, and A.N.G.
Kirschel. 2011. Acoustic monitoring in terrestrial environments using microphone arrays:
Applications, technological considerations, and prospectus. Journal of Applied
Ecology 48:758–767.
Bonter, D.N., T.M. Donovan, and E.W. Brooks. 2007. Daily mass changes in landbirds during
migration stopover on the south shore of Lake Ontario. Auk 124:122−133.
Buler, J.J., F.R. Moore, and S. Woltmann. 2007. A multi-scale examination of stopover
habitat use by birds. Ecology 88:1789−1802.
Chernetsov, N. 2006. Habitat selection by nocturnal passerine migrants en route: Mechanisms
and results. Journal of Ornithology 147:185−191.
Chernetsov, N. 2011. Daytime movements of nocturnal migrants at stopover between two
nearby capture sites. Journal of Ornithology 152:1007–1011.
Ellegren, H. 1991. Stopover ecology of autumn migrating Bluethroats, Luscinia s. svecica,
in relation to age and sex. Ornis Scandinavia 22:340−348.
Southeastern Naturalist
S.A. Rush, E.C. Soehren, and M. Miller
2014 Vol. 13, No. 1
100
James, F.C., and H.H. Shugart. 1970. A quantitative method of habitat description. Audubon
Field Notes 24:727–736.
Johnson, G.D., J.L. Pebworth, and H.O. Krueger. 1991. Retention of transmitters attached
to passerines using a glue-on technique. Journal Field Ornithol ogy 62:486−491.
Jones, J., C.M. Francis, M. Drew, S. Fuller, and M.W.S. Ng. 2002. Age-related differences
in body mass and rates of mass gain of Passerines during autumn migratory stopover.
Condor 104:49−58.
Hutto, R.L. 1998. On the importance of stopover sites to migrating birds. Auk 115:823–825.
Krementz, D.G., and G.W. Pendleton. 1990. Fat scoring: Sources of variability. Condor
92:500–507.
Mehlman, D.W., S.E. Mabey, D.N. Ewert, C. Duncan, B. Abel, D. Cimprich, R.D. Sutter,
and M. Woodrey. 2005. Conserving stopover sites for forest-dwelling migratory landbirds.
Auk 122:1281−1290.
Moore, F., and P. Kerlinger. 1987. Stopover and fat deposition by North American woodwarblers
(Parulinae) following spring migration over the Gulf of Mexico. Oecologia
74:47−54.
Morris, S.R., D.W. Holmes, and M.E. Richmond. 1996. A ten-year study of the stopover
patterns of migratory passerines during fall migration on Appledore Island, Maine.
Condor 98:395–409.
Pyle, P. 1997. Identification Guide to North American Birds. Part 1. Slate Creek Press,
Bolinas, CA.
R Development Core Team. 2012. R: A Language and Environment for Statistical Computing.
R Foundation for Statistical Computing, Vienna, Austria.
Rimmer, C.C., and K.P. McFarland. 1998. Tennessee Warbler (Vermivora peregrina). No.
350, In A. Poole and F. Gill (Eds.). The Birds of North America. The Academy of Natural
Sciences of Philadelphia, Philadelphia, PA.
Ruxton, G.D. 2006. The unequal variance t-test is an underused alternative to Student’s ttest
and the Mann-Whitney U test. Behavioral Ecology 17:688–690.
Tankersley, R., Jr., and K. Orvis. 2003. Modeling the geography of migratory pathways and
stopover habitats for neotropical migratory birds. Conservation Ecology 7:7. Available
online at http://www.consecol.org/vol7/iss1/art7. Accessed 10 October 2011.
Tennessee Department of Environment and Conservation (TDEC). 1999. An ecological inventory
of selected sites in the Cherokee National Forest. Challenge Cost Share number
99-CCS-0804-001. Division of Natural Heritage, Nashville, TN.
Vogt, D.F., M. Hopey, G.R. Mayfield, E.C. Soehren, L.M. Lewis, J.A. Trent, and S.A. Rush.
2012. Stopover site fidelity by Tennessee Warblers at a southern Appalachian highelevation
site. Wilson Journal of Ornithology 124:366−370.
Wang, Y., and F.R. Moore. 1994. Flight morphology, energetic condition, and the stopover
biology of migrating thrushes. Auk 111:683–692.
Wang, Y., 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.
White, G.C., and R.A. Garrott. 1990. Analysis of Wildlife Radio-tracking Data. Academic
Press, London, UK.
White, P.S., and R.D. Sutter. 1998. Southern Appalachian grassy balds: Lessons for management
and regional conservation. Pp. 375–396, In J.D. Peine (Ed.). Ecosystem Management:
Principles and Practices Illustrated by a Regional Biosphere Cooperative. St
Lucie Press, Delray Beach, FL.
Wunderle, J.M. 1991. Age-specific foraging proficiency in birds. Current Ornithology
8:273–324.