Northeastern Naturalist Vol. 22, No. 2
J. Curtis and P. Vila
2015
387
2015 NORTHEASTERN NATURALIST 22(2):387–402
The Ecology of the Wood Turtle (Glyptemys insculpta) in the
Eastern Panhandle of West Virginia
Jessica Curtis1,2 and Peter Vila1,*
Abstract - Glyptemys insculpta (Wood Turtle) is listed as imperiled in West Virginia and
designated as endangered under the IUCN Red List. We studied a population of Wood
Turtles in the Eastern Panhandle of West Virginia from May 2010 to July 2011. We captured
25 turtles at 2 sites: 12 males, 9 females, 3 juveniles, and 1 hatchling. We fitted 10 adults
with radio transmitters, 9 of which also received temperature-data loggers. Our 249 total observations
include the 10 initial captures of the equipped turtles with 218 subsequent radio
locations, and 15 unequipped turtles that were captured, and then recaptured 6 times. Morphometrics
of Wood Turtles at the study location were similar to those reported for another
West Virginia population. The home range for both males and females was approximately 6
ha; there were no significant intersexual differences. Turtles entered hibernacula in late October
when temperatures were approximately 10 °C and remained inactive until mid-March
when temperatures were approximately 5 °C. Our study provides a baseline assessment of
the Wood Turtle in an unstudied upper-elevation Eastern Panhandle location.
Introduction
Glyptemys insculpta (LeConte) (Wood Turtle) is found in disjunct, remnant
populations extending from eastern Minnesota and northern Iowa to southern Québec
and south to northern Virginia and West Virginia (Ernst et al. 1994, Harding and
Bloomer 1979, Tessier et al. 2005). Wood Turtles are well studied across their range
(Bowen and Gillingham 2004, Ernst et al. 1994). Like many temperate reptiles
(Ashton and Feldman 2003), Wood Turtles exhibit latitudinal gradients in various
life-history and morphological traits (Brooks et al. 1992, Walde et al. 2003).
The Wood Turtle appears to be declining in some portions of its range (Daigle
and Jutras 2005, Garber and Burger 1995, Saumure et al. 2007). As with many
species of reptiles, Wood Turtles are threatened primarily by the destruction and
degradation of preferred habitat (Garber and Burger 1995, Gibbon et al. 2000).
Moreover, life-history characteristics make turtles particularly sensitive to disturbances;
in the case of Wood Turtles, individuals are long-lived (up to 58 years),
recruitment is low, and breeding begins at a relatively old age (up to 20 years)
(Ernst et al. 1994, Harding and Bloomer 1979).
Due to declining populations, the Wood Turtle is classified as endangered on
the IUCN Red List (Van Dijk and Harding 2011). In West Virginia, the Wood
Turtle occurs only in the Eastern Panhandle and is considered a high priority
species for conservation efforts where it is ranked as an S3 or vulnerable species
1Shepherd University, Department of Environmental Studies, Shepherdstown, WV 25443.
2Current address - Marshall University, Biology Department, One John Marshall Drive,
Huntington, WV 25755. *Corresponding author - pvila@shepherd.edu.
Manuscript Editor: Todd Rimkus
Northeastern Naturalist
388
J. Curtis and P. Vila
2015 Vol. 22, No. 2
(Sargent 2014). Establishing baseline Wood Turtle population data is especially
important in the Eastern Panhandle of West Virginia due to a human-population
increase of 27% between 2000 and 2010 and recent rapid economic development
in the region (US Census Bureau 2012). Wood Turtles are not well studied
in West Virginia; 2 earlier studies (Breisch 2006, Niederberger and Seidel 1999)
looked at Wood Turtles in low-gradient and low-elevation riverine habitat in the
Eastern Panhandle. This study describes the Wood Turtle at the southern part of
its range in West Virginia at a higher-elevation headwater location. The objective
of this study was to describe the natural history of the Wood Turtle at this upperelevation
Eastern Panhandle location, including aspects of: (1) demography,
(2) morphometrics, (3) home-range size, (4) distance traveled from water, (5)
habitat use, and (6) thermal regime.
Methods
Study site
We studied Wood Turtles at 2 locations along the same stream in the Eastern
Panhandle of West Virginia. To protect these animals, we do not provide exact site
locations. Sites were at ~305 m elevation and separated by 8.8 km and a reservoir.
One surveyed area was ~2.4 km upstream of the reservoir, and the other was ~6.4
km downstream of the reservoir.
Capture and radiotelemetry
We initially located Wood Turtles by following transects parallel to the stream
up to 200 m from the water and capturing them opportunistically by hand. We found
basking traps and drift fences to be ineffective capture methods. We recorded Wood
Turtle capture coordinates with a Garmin 76CSx GPS unit (Garmin, Olathe, KS).
We visited sites approximately 5 days per week in summer and 3 days per week in
fall. From October 2010 to early March 2011, shortly before the turtles emerged
from their hibernacula, we visited sites approximately once a month.
We recorded mass, minimum straight-line carapace length, carapace width,
minimum straight-line plastron length, plastron width, height at bridge, and number
of annuli for all Wood Turtles. We determined gender by secondary sexual characteristics
(Harding and Bloomer 1979). We considered Wood Turtles to be juveniles
if the carapace was less than 160 mm long or their mass was less than 600 g (Lovich et al. 1990);
young Wood Turtles with no annuli were designated as hatchlings. We also noted
deformities, injuries, and missing limbs. Employing a modified Cagle (1939) numbering
system to distinguish individuals, we used a file to notch 1 or 2 marginal
scutes of all Wood Turtles captured.
We affixed radio transmitters (Model # R1930, Advanced Telemetry Systems,
Isanti, MN) with PC-11 epoxy (Part # 010112, Protective Coating Company, Allentown,
PA) to 10 adult Wood Turtles (5 males and 5 females) on the posterior portion
of the carapace at the edge of the marginal and the costal scutes. The transmitters
weighed 24 g each and together with the epoxy, added approximately 35 g to the
mass of each turtle.
Northeastern Naturalist Vol. 22, No. 2
J. Curtis and P. Vila
2015
389
We relocated Wood Turtles with transmitters approximately twice a week during
the summer of 2010, every 2 weeks during the fall of 2010, and approximately
every 2 weeks in the spring and summer of 2011. We removed radio transmitters
and data loggers in July 2011.
Temperature loggers
We attached temperature loggers (Onset HOBO Data Logger, Model #UA-002-
8K, Bourne, MA) to 9 of the 10 Wood Turtles with radio transmitters. We glued an
aluminum L bracket—with the front edge rounded to the same shape as the HOBO
temperature logger—to the edge of the marginal and the costal scutes on the opposite
side of the transmitter using PC-11 epoxy. The eyelet of the temperature logger
fit through a hole in the bracket and we secured the loggers to the brackets with a
cotter pin that allowed each logger to swivel in its metal cradle. This arrangement
allowed data to be downloaded in the field without removing the loggers from the
Wood Turtles. Each data logger, with bracket and glue weighed approximately 40 g.
Three temperature loggers, including the glue and bracket, fell off during the study.
The combined mass of the transmitter, temperature logger, bracket, and epoxy on
all but one Wood Turtle was 5–8% of the turtle’s mass (Cochran 1980). Though
the combined total mass of the logger and transmitter apparatus represented 9.4%
of the mass of the smallest Wood Turtle (800 g), we did not observe any mobility
issues with this animal.
Home range and maximum distances
We determined Wood Turtle home-range size using the kernel density and the
minimum convex polygon methods in Geospatial Modeling Environment software
(http://www.spatialecology.com/gme/) (Beyer 2012) from Wood Turtle coordinates.
The kernel density method used the smoothed cross-validation method (SCV) bandwidth
estimation (R Core Team 2013). The area of the 95% volume contour was
calculated in ArcMap (ESRI, Redlands CA, Version 10.0) using X-tools Pro (Data
East LLC, Russian Federation, Version 9.2). We used ArcMap to measure maximum
travel distances from GPS capture points to the stream or other aquatic habitats utilizing
the National Hydrography Dataset (USGS 2010).
Habitat characterization
We documented the surrounding habitat (1-m diameter) at each Wood Turtle
capture location. We characterized terrestrial habitats by basic plant-community
type, and aquatic habitats by the form and intermittency of habitat (Table 1; Arvisais
et al. 2002, Forsythe et al. 2004). We recorded ambient air temperature at each
Wood Turtle capture location.
Statistical analysis
We used Sigmaplot (SigmaPlot, Systat Software, San Jose, CA) for statistical
analyses. A Shapiro-Wilk normality test and an equal variance test were applied to
data prior to statistical analyses when comparing means. If data were normal and
passed the variance test, we conducted an ANOVA; if the normality or the equal
Northeastern Naturalist
390
J. Curtis and P. Vila
2015 Vol. 22, No. 2
variance test failed, we utilized a Kruskal-Wallis ANOVA on ranks analysis. We employed
the Holm-Sidak method for all pairwise multiple comparison procedures to
control for Bonferroni inequality and tested the ratio of male to female Wood Turtles
using Yates’ χ² test. Statistical significance was set at P < 0.05 for all analyses.
Results
We captured 25 Wood Turtles: 12 males, 9 females, 3 juveniles, and 1 hatchling.
The 249 total observations included the 10 initial captures of the equipped Wood
Turtles with 218 subsequent radio locations, and 15 unequipped Wood Turtles that
were captured and then recaptured 6 times. We observed Wood Turtles in 2 areas:
downstream of the reservoir (downstream site) and upstream of the reservoir (upstream
site). We captured 22 different Wood Turtles at the upstream site: 11 males, 7
females, 3 juveniles, and 1 hatchling; whereas, 3 different turtles were caught at the
downstream site—2 females and 1 male. We observed no between-site movement
by any Wood Turtle.
The adult male:female sex ratio of the Wood Turtles captured at the study locations
was 1.3:1 and did not differ significantly from 1:1 (Yates’ χ², P = 0.321).
Juveniles and hatchlings comprised 16% of all Wood Turtles captured; excluding
the hatchlings, which have a high first-year mortality rate, juveniles comprised
12.5% of all Wood Turtles captured.
Influence of methods on Wood Turtles
We observed no mortality in any of the notched Wood Turtles or those with
transmitters. The presence of transmitters or loggers did not seem to hinder movement,
and all Wood Turtles maintained their initial mass or gained mass throughout
Table 1. Description of vegetation and habitat types. Many terrestrial sites included multiple canopy
levels.
Habitat type Description
Aquatic
Stream Any flowing body of water.
Wetland Any non-flowing, permanent body of water.
Oxbow/vernal pool Any non-flowing, seasonal body of water.
Terrestrial
Forest Dominant canopy over the capture location composed of trees of
various species or ages.
Shrubs Dominant canopy over the capture location is woody vegetation of
any age or species. Could contain a partial upper level tree canopy.
Shrubs mixed with herbaceous Dominant canopy over the capture location is woody vegetation
mixed with herbaceous vegetation, including grasses. Could
contain a partial upper level tree canopy.
Herbaceous (including grasses) Dominant canopy over the capture location is herbaceous
vegetation, including grasses. Could contain a partial tree canopy,
but does not contain shrubs.
Open Bare soil or leaf litter with no vegetation at any level.
Road Open soil or gravel road.
Northeastern Naturalist Vol. 22, No. 2
J. Curtis and P. Vila
2015
391
the study period. Moreover, transmitters and data loggers did not appear to prevent
courtship because we observed Wood Turtles with radio transmitters and data loggers
mating.
Morphometrics
Mean carapace length was significantly greater for males (190.6 mm) than for
females (178 mm) (F = 8.78; df = 1, 19; P = 0.008). No other characteristics were
significantly different between males and females (Table 2). In our study area, the
smallest minimum plastron length (MPL) associated with secondary sexual traits was
151 mm; the smallest minimum carapace length was 176 mm. We could not establish
minimum size at maturity for females in the study because we observed no females
laying eggs nor did we palpate for eggs. The smallest female and male observed courting
had a minimum carapace length of 184 mm and 186 mm, respectively.
Injuries, deformities, and parasites
All adult Wood Turtles (>600 g) had minor carapace damage with chips in the
marginal scutes; 1 turtle exhibited a slice that cut through its first marginal scute.
Three adult Wood Turtles (12.5%) were missing 1 limb. We did not observe a predation
attempt and could not determine the cause of limb loss.
Two Wood Turtles had eye injuries: one with an eye that was infected, pustulent,
and swollen throughout the study period; this turtle remained active and feeding in
summer and successfully emerged from hibernation in the spring of 2011. The eye
injury of the second Wood Turtle was old and well healed; the injury modified the
round pupil into a keyhole shape and the pupil did not have the ability to contract
or dilate.
Three Wood Turtles (12.5%) had abnormal carapacial scute arrangements: 2
Wood Turtles had extra marginal scutes, and one had an extra central scute.
Although we observed mosquitoes on and surrounding Wood Turtles, we found
no external parasites such as Hirudinea sp. (leeches) on any of the animals.
Table 2. Morphometric data for all Wood Turtles at the study site. Mean ± standard deviation and
range are reported.
Adult female Adult male Juvenile Hatchling
(n = 9) (n = 12) (n = 3) (n = 1)
Weight (g) 890 ± 155.0, 997.1 ± 95.3, 341.7 ± 231.0, 9
680–1100 830–1150 155–600
Carapace length (mm) 178.0 ± 10.3, 190.6 ± 9.1, 126.7 ± 30.7, 39
166–196 176–205 98–159
Carapace width (mm) 130.2 ± 7.0, 135.8 ± 8.1, 96.3 ± 18.2, -
121–141 124–150 80–116
Plastron length (mm) 159.8 ± 8.3, 163.3 ± 6.4, 115.3 ± 26.4, 34
147–168 151–172 92–144
Plastron width (mm) 108.8 ± 10.7, 107.2 ± 7.4, 77.3 ± 14.3, -
95–132 94–120 65–93
Height at bridge (mm) 61.9 ± 6.4, 64.5 ± 7.9, 45.0 ± 6.1, -
51–71 58–86 41–52
Northeastern Naturalist
392
J. Curtis and P. Vila
2015 Vol. 22, No. 2
Home-range size
The kernel density home-range estimate for females and males was 5.5 ha and
7.1 ha, respectively (Table 3), and was not significantly different between the sexes
(F = 0.794, df = 1, 8; P = 0.399). Home-range estimates with the minimum convex
polygon (MCP) method were lower—2.7 ha for females and 2.6 ha for males—and
were not significantly different between the sexes (F = 0.022, df = 1, 8; P = 0.887).
Home ranges were elongated, with significant overlap between Wood Turtles.
Maximum distance from water
Females travelled farther from any water body than males, with a mean maximum
distance of 97 m for females versus a mean maximum distance of 46 m for
males (F = 8.99; df = 1, 8; P = 0.017). However, females did not travel farther from
the permanent aquatic pools in the mainstem stream than males—173 m versus 103
m, respectively (F = 3.57; df = 1, 8; P = 0.096; Table 4).
Non-riparian habitat use
At the upstream site, a clear-cut grassy field bordered a tributary of the mainstem.
Three radio-tagged Wood Turtles utilized this field on multiple occasions;
stays lasted as long as a week before the animals returned to the mainstem of the
stream. The Wood Turtles exploited this area from May to September, but spent
more time in the vicinity during the Rubus fruticosus L. (Shrubby Blackberry) fruiting
season. We observed all 3 Wood Turtles eating blackberries or with presumed
blackberry stains on their mouths while the plants were fruiting.
Habitat use and activity
In winter, Wood Turtles used only stream habitats (Fig. 1). In the spring and fall,
stream use was similar (46 and 54%, respectively), while percent time in stream
habitat decreased to 28% in summer. Increased use of stream habitat in the fall may
have been due to preparation to enter winter hibernacula. In the warmer summer
period, Wood Turtles spent less time in the stream habitat as they foraged in the surrounding
forest, and we encountered them in additional habitats as they increased
Table 3. Home ranges calculated with the kernel density (KD) and minimum convex polygon (MCP)
method for the 10 radio-telemetered Wood Turtles. Mean ± standard deviation is reported for each sex
class. *Male turtle 20 was located 1 km away, and the home range was calculated without this distant
coordinate and with the distant coordinate (in parenthesis).
Number of Home range (ha) Number of Home range (ha)
Females sightings KD MCP Males sightings KD MCP
3 38 6.9 4.7 2 31 5.9 3.3
6 17 5.7 2.2 4 30 4.8 2.7
8 19 8.2 3.3 5 24 6.5 2.7
10 22 4.1 2.2 7 22 5.2 2.2
60 15 2.5 0.9 20* 8 (10) 13.3 (57.8) 1.9 (13.7)
5.5 ± 2.2 2.7 ±1.4 7.1 ± 3.1 2.6 ± 0.5
(16.0 ± 23.4) (4.9 ± 4.9)
Northeastern Naturalist Vol. 22, No. 2
J. Curtis and P. Vila
2015
393
their activity (Fig. 1). Wood Turtle activity varied with season (Table 5) and is summarized
below.
Winter. The winter inactive stage included all time spent in the hibernacula.
During this time, Wood Turtles were entirely aquatic and all of the Wood Turtles
we captured were in a stream habitat (Table 5). Wood Turtles did not utilize the
permanent wetlands habitat as a hibernaculum. We located all radio-tagged Wood
Turtles in the mainstem and not in any of the tributaries.
Table 4. Maximum distance from water is the straight-line distance from the capture coordinates to
the nearest water source, determined with ArcMap.
Turtle # Maximum distance from water (m) Maximum distance from stream (m)
Females
3 146.9 272.8
6 64.7 143.6
8 111.5 130.3
10 60.0 201.4
60 101.8 115.6
Mean ± SD 97.0 ± 35.3 172.7 ± 64.7
Males
2 49.8 87.1
4 35.9 50.9
5 50.8 166.7
7 63.2 66.0
20* 27.6 146.0
Mean ± SD 45.5 ± 13.9 103.3 ± 50.6
*Male turtle 20 was located 1 km away and this distance was not used in the calculation of the maximum
distance from water or stream.
Figure 1. Percent of Wood Turtle captures by habitat type. The category “other” includes
wetland, oxbow/vernal pool, open, and road.
Northeastern Naturalist
394
J. Curtis and P. Vila
2015 Vol. 22, No. 2
Spring. After emergence, Wood Turtles stayed close to the aquatic habitat at a
maximum distance from water of 52 m. We observed that Wood Turtles utilized
aquatic and terrestrial habitats approximately equally: 46% and 54%, respectively.
Summer. In the active summer stage, Wood Turtles foraged at greater distances
from the stream (147 m), and we observed them foraging at a greater number of
terrestrial habitats than in other seasons: 6 habitats in summer, 5 in spring, and 4 in
fall (Fig. 1). Terrestrial habitat use during this period was 70% compared with 30%
aquatic habitat use.
Fall. The fall stream-restricted active stage was similar to the spring stage, with
activity limited to a narrow band around the stream as Wood Turtles prepared to
enter hibernacula. In the fall, Wood Turtles were closer to the stream than in the
spring (40 m versus 52 m, respectively), and we observed them spending slightly
more time in the aquatic habitat than the terrestrial habitat (54% versus 44%, respectively).
We observed 3 instances of courtship at this time.
Hibernacula
Of the 10 adult Wood Turtles outfitted with radio transmitters, we radio-tracked
9 of them to their hibernacula. We observed no Wood Turtles overwintering in an
exposed location and detected no Wood Turtle movement during the inactive period.
Eight hibernacula were located upstream of the lake; 6 were clustered together
and 2 hibernacula were ~0.5 km further upstream. The other hibernaculum was at
the downstream site. Hibernacula were located in ~1-m-deep pools that consisted
of undercut banks with or without snags and other debris. Ondatra zibethicus L.
(Muskrat) and Castor canadensis Kuhl (Beaver) had modified the stream banks
in some hibernacula locations. One hibernaculum contained at least 3 adult Wood
Turtles (1 radio-tagged, 2 untagged). After careful searching, we found that the
other 8 hibernacula contained only 1 Wood Turtle.
Some hibernacula, located in burrows with separate entrances, were very
close to one another (approximately 0.5–3 m). We could not ascertain whether
the burrow entrances connected to a larger interior chamber and therefore
treated these hibernacula as distinct. All identified hibernacula were found in
the mainstem stream. Sites used as hibernacula were also utilized during the
summer period. These sites are deeper areas in the mainstem that did not dry up
during the summer period; availability of these deeper sites was limited within
the study area.
Table 5. Distance travelled from water and the percent of time in aquatic habitat during the activity
stages. Maximum distance = maximum distance of any turtle from water within activity period.
Stage Months Maximum distance (m) Percent aquatic
Winter Late October–mid-March 0 100%
Spring Mid-March–mid-May 52 46%
Summer Mid-May–mid-September 147 30%
Fall Mid-September–late October 40 54%
Northeastern Naturalist Vol. 22, No. 2
J. Curtis and P. Vila
2015
395
Thermal regime
We observed 4 distinct temperature periods (Fig. 2): (1) temperature declined
from ~25 °C in July 2010 to approximately 17 °C in late September 2010, (2)
temperature then declined to almost 0 °C in early December 2010, (3) temperature
stayed at ~0 °C until mid-February 2011, and (4) temperature increased to ~25 °C
in July 2011. Five-degree to 25 °C variations were superimposed on this general
trend. Turtles were in hibernacula from late October 2010, when the temperature
was approximately 10 °C, until mid-March 2011, when the temperature reached 5
°C. The highest temperature recorded on basking Wood Turtles was 42 °C.
We have 63 simultaneous readings of logger temperature, ambient air temperature,
and habitat at time of capture; 32 readings from 4 males and 31 readings from
4 females. Aquatic Wood Turtle logger mean temperatures (°C) for spring, summer,
fall, and winter were 20.9 (n = 2), 23.1 (n = 4), 20.0 (n = 11), and 6.7 (n = 2), respectively;
logger temperatures were within 5 degrees of ambient air temperature.
Terrestrial Wood Turtle logger mean temperatures (°C) for spring, summer, and fall
were 37.0 (n = 2), 28.8 (n = 17), and 25.1 (n = 25), respectively; logger temperatures
were within 8 degrees of ambient air temperature.
Discussion
Morphometrics
Adult Wood Turtles tend to have greater mean carapace lengths with increasing
latitude (Brooks et al. 1992, Walde et al. 2003). The morphometrics of the Wood
Turtles in this study area were similar to populations studied at similar latitudes
Figure 2. Temperature data for a representative Wood Turtle at the upstream site for which
we collected a complete seasonal data set from 15 Jul 2010 to 31 Jul 2011. The different
monthly spacing is due to different temperature-recording intervals; temperature was recorded
every 2 min during the summer and every 15 min during the fall and winter. The
straight line is a line drawn to indicate temperature periods.
Northeastern Naturalist
396
J. Curtis and P. Vila
2015 Vol. 22, No. 2
(Breisch 2006, Lovich et al. 1990). We did not find larger morphometric characteristics
due to expected altitudinal decreases in temperature.
As seen in other Wood Turtle populations, male carapace length at our study
sites was significantly larger than female carapace length: New Jersey (Breisch
2006); Michigan (Farrell and Graham 1991, Harding and Bloomer 1979); New
Hampshire (Harding and Bloomer 1979); Ontario (Tuttle and Carroll 1997, Brooks
et al. 1992); and Quebec (Walde et al. 2003).
Development of secondary sexual traits in male Wood Turtles coincides with
maturation (Harding and Bloomer 1979). The smallest minimum plastron length
(MPL) and smallest minimum carapace length (MCL) at maturation are expected
to increase with latitude (Brooks et al. 1992, Farrell and Graham 1991, Walde et
al. 2003). However, our smallest male displaying secondary sexual characteristics
—MPL = 151 mm and MCL = 176 mm—was similar to the smallest MCL reported
by Ross et al. (Ross et al. 1991) in Wisconsin but larger than Breich (2006) in a
regional West Virginia location.
Injuries, deformities, and parasites
Missing appendages are common in Wood Turtle populations and are likely due
to aborted predation attempts (Saumure and Bider 1998, Saumure et al. 2007). Our
observation of 12.5% amputation is high compared to many other sites: 6% in West
Virginia (Breisch 2006), 8.6% in New Jersey (Farrell and Graham 1991), 6.8%
in Pennsylvania (Ernst 2001), 4.5% in New York (Hunsinger 2002), 9% in New
Hampshire (Tuttle 1996), 9.7% in Michigan (Harding and Bloomer 1979), 12.5%
in a later Michigan study (Harding 1985), and 9.6% in Quebec (Walde et al. 2003).
Studies reporting higher amputation were sites impacted by agricultural or recreational
activities (15% and 32% at 2 different sites in Quebec [Saumure and Bider
1998] and 35.5% in Vermont [Parren 2013]).
Direct observations and incidental evidence implicate Procyon lotor (L.) (Raccoon)
in the mutilation of adult Wood Turtles (Harding 1985). We never directly
observed Raccoons at our study site, but we found their footprints and partially
eaten prey items. Martes pennanti (Erxleben) (Fisher; Parren 2013) and Lontra
canadensis (Schreber) (North American River Otter; Carroll and Ultsch 2006) have
also been implicated in Wood Turtle mutilation and death, but we did not observe
activity of either species in the study area.
Our site has minimal impact by agricultural machinery because the nearby
clearcut site is mowed only once a year. In agricultural areas, mortality caused
by machinery is a significant problem (Saumure and Bider 1998, Saumure et al.
2007). Increased predation by mesopredators in edge habitats, like Mephitis mephitis
(Schreber) Striped Skunk and Raccoon (Ross et al. 1991, Saumure and Bider
1998), were likely minimized in this study area given that the clear-cut areas are
surrounded by forest and not anthropogenic edge habitats where their diets would
have been supplemented by humans. However, predator density may be increased
by the clearcut area; this possibility coupled with the higher capture probability of
injured Wood Turtles may explain the high percentage of amputees observed.
Northeastern Naturalist Vol. 22, No. 2
J. Curtis and P. Vila
2015
397
The 12.5% frequency of abnormal scute arrangements observed in this study
was within the range found in other studies: 14% in West Virginia (Breisch 2006),
4.5% in Pennsylvania (Ernst 2001), and 2% in Michigan (Harding and Bloomer
1979). Scute-arrangement abnormalities do not appear to reduce fitness in affected
turtles (Harding and Bloomer 1979).
Leeches are frequently reported on Wood Turtles. Breisch (2006) found that
16% of Wood Turtles studied had leeches at least once, and Niederberger and Siedel
(1999) reported that Wood Turtles in water were “often (>50%)” parasitized by
leeches, although they did not provide the overall incidence of parasitism for all
captures or the number of individuals parasitized. Infestation rates were much higher
in Pennsylvania at 38.6% (Ernst 2001) and almost 90% in New Jersey (Farrell
and Graham 1991). These rates differ from our study, during which we observed no
leeches on Wood Turtles; it may be that headwater streams at this location have low
leech density. Leeches on Wood Turtles are generally observed when Wood Turtles
are in aquatic habitats from October to May (Breisch 2006, Hulse and Routman
1982, Koffler et al. 1978, Tuttle and Carroll 1997). However, the majority of our
captures were between May and October when Wood Turtles were predominantly
in terrestrial habitats and encounters with leeches would be low. Leech parasitism
could be high for the October to May period, but direct observation of Wood Turtles
is more difficult in deeper waters or in hibernacula.
Home range and habitat use
Mean home-range sizes in this study were small and within the ranges reported
for Wood Turtles of both genders; home-range size follows latitudinal gradients,
with smaller ranges in the South and larger ranges in the North (Bowen and Gillingham
2004, Remsberg et al. 2006). It is likely that habitat quality at our study site
is high and not limiting (Arvisais et al. 2002).
Maximum distance from water was similar to other West Virginia populations at
~200 m (Breisch 2006, Niederberger and Seidel 1999), but smaller than distances
for Wood Turtles in Vermont (425 m; Parren 2013) or for Wood Turtles in Pennsylvania
(600 m; Kaufmann 1995). The wider dispersal from permanent water bodies
seen in females did not appear to be associated with a search for suitable nesting
sites as reported by several authors for various turtle species (Lewis and Faulhaber
1999, Morreale et al. 1984, Quinn and Tate 1991). Rather, movement was
frequently associated with rainfall and may have been related to foraging activities
that capitalized on increased numbers of invertebrate prey available after rainfall
(Kaufmann 1989). As indicated by the small range-size, the small maximum distance
from water traveled by our Wood Turtles indicates high-quality forage near
water (Arvisais et al. 2002).
Upland-site fidelity has been documented in both genders (Parren 2013, Remsberg
et al. 2006). It is possible that extended movement away from water by
females could be due to idiosyncratic differences between individuals and not an
effect of gender (Compton et al. 2002). Increasing sample size and measurements
over several seasons would allow for a more precise discernment of movement
Northeastern Naturalist
398
J. Curtis and P. Vila
2015 Vol. 22, No. 2
patterns between sexes and allow hypotheses concerning Wood Turtle movement
to be tested.
Wood Turtle habitat use in this study was similar to habitat use described for
other eastern Wood Turtle populations (Harding and Bloomer 1979). It included
permanent streams with a mosaic of forest and open-canopy habitats adjacent to
the stream (Ernst et al. 1994, Jones and Sievert 2009). Wood Turtles in our studies
shared the same active period and terrestrial period as animals at other sites
in the species’ southern range (Breisch 2006, Kaufmann 1992, Niederberger and
Seidel 1999).
In our study, Wood Turtles spent approximately equal time in stream and terrestrial
habitats in the spring and fall most likely due to the ability to reliably
thermoregulate in the stream corridor during these periods of wide temperature
fluctuations (Compton et al. 2002, Dubois et al. 2009). Use of stream habitat in
the fall may also be due to preparation to enter winter hibernacula, and this fall
congregation is when we observed courting. The clear-cut area at our site, which
was utilized by 3 different Wood Turtles during the summer, may also provide
foraging and thermogegulatory habitat for Wood Turtles (Compton et al. 2002,
Dubois et al. 2009) and providing similar habitats could be a useful management
tool for this species.
We found low numbers of Wood Turtles (1–3) in hibernacula in this study
compared to those at other sites that contained 5–70 turtles in a single occluded hibernaculum
(Bloomer 1978, Harding and Bloomer 1979). While Wood Turtles have
been observed overwintering exposed on stream beds in Massachusetts (Farrell and
Graham 1991) and in Ontario (Greaves and Litzgus 2008), individuals at our site
used some type of cover.
Wood Turtles in our study area may have been limited by the number of permanent
aquatic habitats available for hibernacula. Wood Turtles appeared to be
restricted to stretches of the mainstem of the creek that do not dry up in the summer,
including flooded areas behind beaver dams. Beaver dams were located above and
below the lake, and we observed Wood Turtles utilizing these areas. Damming activities
by Beavers provide aquatic habitat throughout the year; thus, Beavers may
be functioning as a keystone species in the study area.
Thermal regime
The 1–8 °C difference found between temperatures on the Wood Turtles’ data loggers
and the ambient air temperatures is similar to differences found by for Wood
Turtles in Lancaster County, PA (Ernst 1986). Although Wood Turtles thermoregulate,
regulation is imprecise even at their northern range limit where they would be
expected to be thermally constrained (Dubois et al. 2009).
Population viability and future directions
Low recapture rates precluded accurate estimates of the Wood Turtle population
in the study area; however, given the low number of Wood Turtles captured,
the population is most likely small—likely smaller than populations in other parts
of the state (Niederberger and Seidel 1999). Juvenile Wood Turtles are typically
Northeastern Naturalist Vol. 22, No. 2
J. Curtis and P. Vila
2015
399
not captured (Daigle 1997, Ross et al. 1991). Our observation of 3 juveniles and 1
hatchling along with the observation of no observed adult mortality suggested persistence
of Wood Turtles at the study site. Further research is required to determine
if recruitment into the adult cohort is occurring because our data showed that the
hatchling/juvenile to adult ratio is low.
While low numbers may indicate declining or relict populations with a concomitant
loss of genetic diversity, it may be that habitat availability is limiting in
the study area and cannot sustain a large Wood Turtle population. High levels of
polymorphism and heterozygosity may still be maintained in small populations
(less than 50; Tessier et al. 2005), so this population although likely to be small, could
maintain genetic diversity. The reservoir is probably a barrier to movement because
we observed no movement across the lake. Additionally, Wood Turtles did
not utilize terrestrial habitats adjacent to the lake, suggesting these 2 groups may
be relict populations separated by reservoir formation. This study documented a
baseline count of the Wood Turtles; further study is warranted in order to determine
reproduction, recruitment, and population viability.
While determination of population size, structure, and viability of Wood Turtles
in protected private and public areas is needed, there is also a need to investigate the
status and viability of Wood Turtle populations across the wider landscape. Habitat
destruction and degradation (Ernst et al. 1994, Garber and Burger 1995, Gibbon
et al. 2000) pose the greatest threats to Wood Turtles. Populations that occupy
less-protected habitats located in areas impacted by development and agricultural
activity are at the greatest risk. Important questions for the maintenance of a viable,
regional Wood Turtle metapopulation remain, including whether there is genetic
exchange between nearby populations and what, if any, connectivity of the Wood
Turtles described in this study have to the regional metapopulation.
Acknowledgments
We gratefully acknowledge the assistance of Larry Hines, whose knowledge of the area
and wildlife and enthusiasm and logistical support were instrumental for the success of this
study. We also express our appreciation to Alan Temple at the US Fish and Wildlife Service
for use of the radiotelemetry equipment that made this study possible. This study was
funded by a grant from the West Virginia Department of Natural Resources. Both authors
contributed equally to this paper.
Literature Cited
Arvisais, M., J.-C. Bourgeois, E. Lévesque, C. Daigle, D. Masse, and J. Jutras. 2002. Home
range and movements of a Wood Turtle (Clemmys insculpta) population at the northern
limit of its range. Canadian Journal of Zoology 80:391–398.
Ashton, K.G., and C.R. Feldman. 2003. Bergmann’s rule in nonavian reptiles: Turtles follow
it, lizards and snakes reverse it. Evolution 57:1151–1163.
Beyer, H.L. 2012. Geospatial Modelling Environment. Available online at http://www.
spatialecology.com/gme. Accessed 1 July 2013.
Bloomer, T.J. 1978. Hibernacula congregating in the Clemmys genus. Journal of North Ohio
Association of Herpetology 4:37–42.
Northeastern Naturalist
400
J. Curtis and P. Vila
2015 Vol. 22, No. 2
Bowen, K.D., and J.C. Gillingham. 2004. R9 Species Conservation Assessment for Wood
Turtle – Glyptemys insculpta (LeConte, 1830). Conservation Assessment. Eastern Region
of the US Forest Service, Milwaukee, WI.
Breisch, A.N. 2006. The natural history and thermal ecology of a population of Spotted
Turtles (Clemmys Guttata) and Wood Turtles (Glyptemys Insculpta) in West Virginia.
M.Sc. Thesis. Marshall University, Huntington, WV.
Brooks, R.J., C.M. Shilton, G.P. Brown, and N.W.S. Quinn. 1992. Body size, age distribution,
and reproduction in a northern population of Wood Turtles (Clemmys insculpta).
Canadian Journal of Zoology 70:462–469.
Cagle, F.R. 1939. A system of marking turtles for future identification. Copeia 3:1 70–173.
Carroll, D.M., and G.R. Ultsch. 2006. Glyptemys insculpta (Wood Turtle): Predation. Herpetological
Review 37:215–216.
Cochran, W.W. 1980. Wildlife telemetry. Pp. 507–520, In S.D. Schemnitz (Ed.). Wildlife
Techniques Manual. The Wildlife Society, Inc., Washington, DC. 1136 pp.
Compton, B.W., J.M. Rhymer, and M. McCollough. 2002. Habitat selection by Wood
Turtles (Clemmys Insculpta): An application of paired logistic regression. Ecology
83:833–843.
Daigle, C. 1997. Size and characteristics of a Wood Turtle, Clemmys insculpta, population
in southern Québec. Canadian Field Naturalist 111:440–444.
Daigle, C., and J. Jutras. 2005. Quantitative evidence of decline in a southern Québec Wood
Turtle (Glyptemys insculpta) population. Journal of Herpetology 39:130–132.
Dubois, Y., G. Blouin-Demers, B. Shipley, and D. Thomas. 2009. Thermoregulation and
habitat selection in Wood Turtles Glyptemys insculpta: Chasing the sun slowly. Journal
of Animal Ecology 78:1023–1032.
Ernst, C.H. 1986. Environmental temperatures and activities in the Wood Turtle, Clemmys
insculpta. Journal of Herpetology 20:222–229.
Ernst, C.H. 2001. Some ecological parameters of the Wood Turtle, Clemmys insculpta, in
southeastern Pennsylvania. Chelonian Conservation and Biology 4:94–99.
Ernst, C.H., R.W. Barbour, and J.E. Lovich. 1994. Turtles of the United States and Canada.
Smithsonian Institution Press, Washington, DC. 682 pp.
Farrell, R.F., and T.E. Graham. 1991. Ecological notes on the turtle Clemmys insculpta in
Northwestern New Jersey. Journal of Herpetology 25:1–9.
Forsythe, P., B. Flitz, and S.J. Mullin. 2004. Radio telemetry and post-emergent habitat
selection of neonate Box Turtles (Emydidae: Terrapene carolina) in central Illinois.
Herpetological Review 35:333–335.
Garber, S.D., and J. Burger. 1995. A 20-yr study documenting the relationship between
turtle decline and human recreation. Ecological Applications 5:1151–1162.
Gibbon, J.W., D.E. Scott, T.J. Ryan, K.A. Buhlmann, T.D. Tuberville, B.S. Metts, J.L.
Greene, T. Mills, Y. Leiden, S. Poppy, and C.T. Winne. 2000. The global decline of
reptiles, déjà vu amphibians. BioScience 50:653–666.
Greaves, W.F., and J.D. Litzgus. 2008. Chemical, thermal, and physical properties of sites
selected for overwintering by northern Wood Turtles (Glyptemys insculpta). Canadian
Journal of Zoology 86:659–667.
Harding, J.H. 1985. Clemmys insculpta, Wood Turtle predation–mutilation. Herpetological
Review 161:30.
Harding, J.H., and T.J. Bloomer. 1979. The Wood Turtle, Clemmys insculpta: A natural history.
Bulletin of the New York Herpetological Society 15:9–26.
Hulse, A.C., and E.J. Routman. 1982. Leech (Placobdella parasitica) infestations on the
Wood Turtle, Clemmys insculpta. Herpetological Review 13:116–117.
Northeastern Naturalist Vol. 22, No. 2
J. Curtis and P. Vila
2015
401
Hunsinger, T.W. 2002. Demography and life history of a WoodTurtle (Clemmys insculpta)
population in the Hudson River watershed. Section VIII:1–25, In W.C. Neider and J.R.
Waldman (Eds.). Final Reports of the Tibor T. Polgar Fellowship Program, 2001. Hudson
River Foundation, New York, NY.
Jones, M.T., and P.R. Sievert. 2009. Effects of stochastic flood disturbance on adult Wood
Turtles, Glyptemys insculpta, in Massachusetts. Canadian Field-Naturalist 123:313–322.
Kaufmann, J.H. 1989. The Wood Turtle stomp. Natural History 98:8–13.
Kaufmann, J.H. 1992. Habitat use by Wood Turtles in central Pennsylvania. Journal of
Herpetology 26:315–321.
Kaufmann, J.H. 1995. Home ranges and movements of Wood Turtles, Clemmys insculpta,
in Central Pennsylvania. Copeia 1995:22–27.
Koffler, B.R., R.A. Seigel, and M.T. Mendonça. 1978. The seasonal occurrence of leeches
on the Wood Turtle, Clemmys insculpta (Reptilia, Testudines, Emydidae). Journal of
Herpetology 12:571–572.
Lewis, T.L., and C.A. Faulhaber. 1999. Home ranges of Spotted Turtles (Clemmys guttata)
in southwestern Ohio. Chelonian Conservation and Biology 3:430–434.
Lovich, J.E., C.H. Ernst, and J.F. McBreen. 1990. Growth, maturity, and sexual dimorphism
in the Wood Turtle Clemmys insculpta. Canadian Journal of Zoology 68:672–677.
Morreale, S.J., J.W. Gibbons, and J.D. Congdon. 1984. Significance of activity and movement
in the Yellow-bellied Slider Turtle (Pseudemys scripta). Canadian Journal of Zoology
62:1038–1042.
Niederberger, A.J., and M.E. Seidel. 1999. Ecology and status of a Wood Turtle (Clemmys
insculpta) population in West Virginia. Chelonian Conservation and Biology 3:414–418.
Parren, S.G. 2013. A twenty-five-year study of the Wood Turtle (Glyptemys insculpta) in
Vermont: Movements, behavior, injuries, and death. Herpetological Conservation and
Biology 8:176–190.
Quinn, N.W.S., and D.P. Tate. 1991. Seasonal movements and habitat of Wood Turtles
(Clemmys insculpta) in Algonquin Park, Canada. Journal of Herpetology 25:217–220.
R Core Team. 2013. R: A language and environment for statistical computing. R Foundation
for Statistical Computing, Vienna, Austria. Available online at http://www.R-project.
org/. Accessed 7 July 2013.
Remsberg, A.J., T.L. Lewis, P.W. Huber, and K.A. Asmus. 2006. Home ranges of Wood
Turtles (Glyptemys insculpta) in Northern Michigan. Chelonian Conservation and Biology
5:42–47.
Ross, D.A., K.N. Brewster, R.K. Anderson, N. Ratner, and C.M. Brewster. 1991. Aspects of
the ecology of Wood turtles, Clemmys insculpta, in Wisconsin. Canadian Field Naturalist
105:363–367.
Sargent, B.D. 2014. Rare, threatened, and endangered animal species. West Virginia Department
of Natural Resources, Charleston, WV.
Saumure, R.A., and J.R. Bider. 1998. Impact of agricultural development on a population of
Wood Turtles (Clemmys insculpta) in Southern Québec, Canada. Chelonian Conservation
and Biology 3:37–45.
Saumure, R.A., T.B. Herman, and R.D. Titman. 2007. Effects of haying and agricultural
practices on a declining species: The North American Wood Turtle, Glyptemys insculpta.
Biological Conservation 135:581–591.
Tessier, N., S.R. Paquette, and F.-J. Lapointe. 2005. Conservation genetics of the Wood Turtle
(Glyptemys insculpta) in Quebec, Canada. Canadian Journal of Zoology 83:765–772.
Tuttle, S.E. 1996. Ecology and natural history of the Wood Turtle (Clemmys insculpta) in
Southern New Hampshire. M.Sc. Thesis. Antioch University, Keene, NH.
Northeastern Naturalist
402
J. Curtis and P. Vila
2015 Vol. 22, No. 2
Tuttle, S.E., and D.M. Carroll. 1997. Ecology and natural history of the Wood Turtle
(Clemmys insculpta) in southern New Hampshire. Chelonian Conservation and Biology
2:447–449.
US Census Bureau. 2012. 2010 Census of population and housing, population and housing
unit counts, CPH-2-50, West Virginia. US Census Bureau, Washington, DC. Available
online at http://www.census.gov/prod/cen2010/cph-2-50.pdf. Accessed 3 July 2013.
US Geological Survey (USGS). 2010. National hydrography dataset (NHD - 24k). Available
online at http://wvgis.wvu.edu/data/dataset.php?ID=235. Accessed 15 July 2013.
Van Dijk, P.P., and J.H. Harding. 2011. Glyptemys insculpta (Wood Turtle). IUCN 2013.
IUCN Red List of Threatened Species. Available online at http://www.iucnredlist.org/
details/summary/4965/0. Accessed 24 January 2014.
Walde, A.D., J.R. Bider, C. Daigle, D. Masse, J.-C. Bourgeois, J. Jutras, and R.D. Titman.
2003. Ecological aspects of a Wood Turtle, Glyptemys insculpta, population at the northern
limit of its range in Québec. The Canadian Field-Naturalist 117:377–388.