Content uploaded by Joshua P. Scholl
Author content
All content in this area was uploaded by Joshua P. Scholl on Jul 22, 2015
Content may be subject to copyright.
23 Volume 1, Issue 1 Fall 2012
FAURJ
Population Structure and Burrow Placement
of Gopherus polyphemus in a Small, Declining
Southeast Florida Conservation Area
JOSHUA SCHOLL1,2, TOBIN HINDLE3, EVELYN FRAZIER2
1NSF-Undergraduate Research and Mentoring Program; 2Department of Biological
Sciences, Florida Atlantic University, Boca Raton, Florida 33431; 3Department of
Geosciences, Florida Atlantic University, Boca Raton, Florida 33431
Abstract
Gopherus polyphemus has been declining throughout its range since the 1800s
primarily due to urbanization, which often leads to the creation of island habi-
tats. is connes populations and eliminates natural management by wildres
resulting in degraded island habitats. To maximize conservation eorts in rap-
idly developing regions it is critically important to investigate not only the natu-
ral ecology of native species, but also how they are doing in habitats set aside for
them. We studied a gopher tortoise population to determine its sustainability
and burrow placement across dierent soil and vegetation types in a conserva-
tion area on the Florida Atlantic University campus in Boca Raton, Florida. We
conducted complete burrow surveys using belt transects, directly captured tor-
toises, and performed vegetation and soil analyses through aerial photos and
United States Geological Survey data, respec tively. e sustainability of the pop-
ulation was based directly on age structure, gained from carapace length mea-
surements, and indirectly on ratios of active to abandoned burrow categories.
Tortoises burrowed densely in areas of low vegetation and completely avoided
areas with closed canopies, which comprised about 15% of the habitat. Soil types
had a signicant correlation to the spatial distribution of burrows. We found a
high ratio of active to abandoned burrows, which could indicate an active and
healthy population; however, age structure data compiled from captured tor-
toises revealed a lack of sub-adults, suggesting an unsustainable population. We
concluded that tortoise surveys which solely collect data on burrow numbers and
activity level and not tortoise sizes may provide misleading results on the status
of gopher tortoise populations in conned, degraded habitats. More direct pop-
ulation assessment methods such as tortoise captures or burrow measurements
need to be used.
Introduction
Since the 1980s, when conservation biology
gained its contemporary denition, eorts to
protect our planet’s biodiversity through the
establishment of national and local parks and
conservation areas has grown signicantly
(2010a). However, establishing such areas may
not be enough to protect certain species(Ervin
2003, 2010a). For example, the greater
FAURJ
24 Volume 1, Issue 1 Fall 2012
FAURJ
prairie chicken’s declines in the Midwest Unit-
ed States continued in the mid 1990s despite a
signicant increase in the amount of favorable
habitat set aside for its recovery(Westemeier
et al. 1998). Habitat management and species
assessment are equally important to ensure
that our conservation eorts are eective
(Ervin 2003).
South Florida is home to an abundant array
of imperiled species and habitats, such as Go-
pherus polyphemus, the gopher tortoise (Hold-
er et al. 2007). e gopher tortoise is a long
lived, slow maturing species, whose burrows
provide shelter to more than 300 invertebrate
and 60 vertebrate species (Breininger et al.
1991, Butler and Sowell 1996, Schwartz and
Karl 2005). As a result of their disproportion-
ately large impact on other species, they are
considered a keystone species and receive a
great deal of conservation attention (Ashton
and Ashton 2008). Tortoise populations have
been declining since the early 1800s as a result
of human activity (Ashton and Ashton 2008).
Populations west of the Mobile and Tombigbee
Rivers in Alabama, Mississippi, and Louisiana
have been listed as threatened since 1987 by
the United States Fish and Wildlife Service
(USFWS) while those to the east of the rivers
in Florida Georgia, Alabama, South Carolina
were listed as a candidate species for threat-
ened status (Service 2011, USFWS 2011).
Evidence suggests that tortoises continue to
decline throughout their range today despite
increases in land set aside for the recovery of
gopher tortoises (McCoy et al. 2006, Ashton
and Ashton 2008). Monitoring and manage-
ment have been incomplete and it has been
shown that tortoises on some protected lands
in northern and central Florida continue to
decline (McCoy et al. 2006).
Little is known about the status of tortoises
in southeastern Florida and no data are avail-
able on population change over time. Preserves
have been set aside for gopher tortoises but
monitoring eorts have been incomplete as
they have focused solely on locating and count-
ing sub-adult and adult burrows(McCoy et al.
2006, Ashton and Ashton 2008). Observation
of tortoise burrow numbers alone may not be
indicative of the sustainability of a population
(McCoy et al. 2006). More information such
as population age structure obtained either
through indirect measurements of burrow
widths or direct measurements of tortoise
carapace length should also be obtained (Alford
1980, Mushinsky et al. 1997, McCoy et al.
2006). King conducted a study in 2005 on a
tortoise population in a conservation area
managed for tortoises in southeastern Florida.
Using systematic stratied sampling King
surveyed the entire habitat for tortoises and
also captured most of them (King 2005). King
also measured the carapace lengths of captured
tortoises and found that the population con-
sisted almost entirely of adults (King 2005). In
addition active to abandoned burrow ratios can
be used to elucidate some factors about popu-
lations in conjunction with age structure data
(McCoy et al. 2006). Our rst goal in this study
was to investigate hypotheses of population
declines on protected habitats by surveying the
tortoise population that King studied in 2005
and comparing our results with King’s.
One factor speculated to contribute to the
continued decline of tortoises on conserva-
tion lands is insucient habitat management
(Mushinsky et al. 1997, McCoy et al. 2006).
Burrowing preferences have not been investi-
gated in southeastern Florida and thus existing
management strategies may not be maximizing
eorts to provide the tortoise with a suitable
habitat (King 2005). A suitable tortoise habitat
is composed of upland, dry scrub habitat which
is also prized by humans because it is ideal
for real estate development (Mushinsky et al.
2006, Ashton and Ashton 2008). Such habitat
consists of well drained, sandy soils, low can-
opy coverage, and high herbaceous vegetation
which aids tortoise burrowing, thermoregula-
tion and foraging, respectively (Mushinsky et
al. 2003). Low canopy coverage and abundant
herbaceous vegetation are maintained natural-
ly by wildres. Fragmentation of habitats and
proximate human settlement eliminates
25 Volume 1, Issue 1 Fall 2012
FAURJ
natural management regimes causing canopy
closure and associated reduction of herbaceous
vegetation, both unfavorable to tortoises
(McCoy et al. 2006).
Our second goal in this study was to gain
insight into burrowing preferences of gopher
tortoises in degraded habitats such as ours.
We compared those to preferences found in
northern Florida populations to determine if
management strategies at our site would bene-
t from being tailored to our population.
For our rst goal, we hypothesized that
the population of gopher tortoises would still
consist primarily of adults though the popu-
lation probably increased as a result of illegal
relocations to our study site. For our second
goal we hypothesized that tortoises would
exhibit burrowing preferences for well drained,
sandy soils and areas of low vegetation, similar
to those observed in the northern portion of
their range.
Methods
Study site
Our study was conducted at the Florida
Atlantic University (FAU) conservation area
in Boca Raton, Florida (26 ̊ 23’ N, 80 ̊ 7’ W).
Although its acreage is not protected in perpe-
Figure 1. Aerial photograph of the FAU Conservation Area on the Florida Atlantic University
campus in Boca Raton, Florida.e site consists of two habitat fragments separated by Palm Beach
Avenue and a long parking lot. e larger fragment to the west of Palm Beach Avenue consists
of xeric oak scrub mixed with patches of oak hammock and saw palmetto (Serenoa repens)
stands. Invasive species including Brazilian pepper tree (Schinus terebinthifolius), umbrella tree
(Scheera spp.), and Acacia exist throughout the scrub. An old, drift-fence line from past tortoise
relocations crosses the center of the fragment. Other trails exist from human disturbance such as
dirt bikers and hikers. e smaller portion lies to the east of Palm Beach Avenue and is regularly
mowed. It consists of grasses and a few lone slash pine and palm trees. e black bar represents
200 meters. Both fragments are bordered by aregional airport to the north and west, FAU to the
south, and Palm Beach State College to the east.
FAURJ
26 Volume 1, Issue 1 Fall 2012
FAURJ
tuity for tortoises it is currently protected
and managed for them. e areas herbaceous
vegetation changes seasonally with green grass
and other herbaceous vegetation primarily
evident during South Florida’s rain season
from late spring to late fall. During the dry
season most annual grasses and plants wither
and only perennial grasses remain along with
shrubs and trees. e site consists of two
habitat fragments separated by Palm Beach
Avenue and a long parking lot. In our study, the
two sites were considered as one as numerous
tortoise crossings were observed between the
sites. e larger fragment is to the west of Palm
Beach Avenue and consists of xeric oak scrub
mixed with patches of oak hammock and saw
palmetto stands. Invasive species including
Brazilian pepper tree, umbrella tree, and acacia
exist throughout the scrub. An old path bisects
the larger fragment, remnant of the path of a
former drift-fence line created to increase site
delity of past tortoise relocations to the site.
Other trails exist from human disturbance
such as dirt bikers and hikers. e smaller
portion lies to the east of Palm Beach Avenue
and is regularly mowed. It consists of grasses
and a few lone slash pine and palm trees. Both
fragments are bordered by a regional airport to
the north and west, FAU to the south, and Palm
Beach State College to the east.
e area is composed of 36.83ha (368,300
m2) of upland habitat lled with well-drained
soils suitable for the gopher tortoise (King
2005). e area is an island habitat as articial
barriers that include a busy road, an airport,
and the FAU and Palm Beach State College
campuses, prevent tortoises from dispersing
(King 2005).
Burrow survey
We surveyed our site during the nal
month of the tortoises’ 2010 active season to
ensure that burrows located were representa-
tive of tortoise preference and selection. We
conducted a complete burrow survey using
belt transects (McCoy et al. 2006, Ashton and
Ashton 2008). Our GPS unit (Trimble ProXH)
was only accurate to +/- 5m. We placed our belt
transects inside 50 meter by 50 meterplots
to ensure that transects were traced precisely
with low error (+/- 1m). To ensure complete
coverage within each plot while maintaining a
manageable width, we set our transect width
to ve meters and length to 50 meters (McCoy
et al. 2006). We used the dependent double
observer method in which two observers
survey the same belt transect(Nichols et al.
2000). Burrows were determined as belonging
to tortoises based on their evidence of a dome
shaped burrow (Mushinsky et al. 2006). All
burrows found were categorized as either active
or abandoned; the inactive burrow category has
been lumped into the active category because
it has been demonstrated to be very dicult to
dierentiate reliably between the two (Ashton
and Ashton 2008, FWC 2008). Active burrows
were described by very little vegetative growth
in the burrow entrance, footprints or plastral
drag marks, or tortoise presence (Ashton and
Ashton 2008). Abandoned burrows lacked all of
the characteristics of active burrows and were
collapsed and required signicant excavation
eort by a tortoise for re-use (Ashton and
Ashton 2008).Burrows were digitally marked
using the GPS unit and mapped using ArcGIS
9.3 software.
Habitat characterization
Vegetation characteristics were determined
from aerial images (2.54 cm = 91.44 m) taken
in 2009 and veried by ground surveys. We
characterized vegetation into three height
categories: Canopy coverage (>3m), shrub
cover (1.5m-3.0m), and herbaceous ground
cover (<1.5m). Soil maps were derived from the
Natural Resource Conservation Service and soil
types are explained in King’s papers(King 2005,
2010b). ese data were projected, as digital
layers, over burrow points in ArcGIS 9.3 to cre-
ate maps for analysis of the spatial distribution
of burrows across soil and vegetation classes.
Tortoise survey
Gopher tortoises were captured from June
27 Volume 1, Issue 1 Fall 2012
FAURJ
2010 to March 2011. We visited every burrow
at our site once a month and attempted to
capture every tortoise we saw inside or outside
of their burrows. Tortoises that we spotted
inside of their burrows were captured by
coaxing them to their burrow entrance by
eliciting a territorial response. Such a response
is displayed equally by males and females
(Mushinsky et al. 2006, Ashton and Ashton
2008). However, during the breeding season,
typically between March and November in
South Florida, males are more active and thus
more likely to be captured (Ashton and Ashton
2008). However, given our exhaustive eorts
and their duration throughout a calendar year
we are condent this did not signicantly im-
pact our results. We marked tortoises’ scutes,
with a le according to an established coding
system for long term identication (Ashton
and Ashton 2008). We used water based paint
to write numbers on tortoise scutes which
allows in-burrow re-identication and thus
avoids unnecessary recapture and stress to
the animal. We also measured the carapace
(top shell) length of all captured tortoises and
sexed each according to plastral (bottom shell)
concavity. Sub-adults (<23cm) were not sexed
due to the lack of distinguishing features
between males and females (Mushinsky et al.
1994). All research eorts for this particular
project were carried out under FWC permit
number LSSC-10-00208 and an approved
animal care (IACUC) protocol number A10-28
from FAU.
Population estimates
We used the carapace length measurements
to create a population demographic prole. We
classied tortoises into three life stages based
on carapace length: Juveniles (<13 cm), sub
adults (13-22 cm) and adults (>22 cm)(Diemer
1992, Mushinsky et al. 1994). Tortoises in
the adult categories are considered sexually
mature and thus are potential contributors to
population growth (Mushinsky et al. 1994).
Tortoises in the sub-adult and juvenile cate-
gories are usually sexually immature but their
relative abundances allow estimation of the
reproductive success of the population(Mush-
insky et al. 1994). Gopher tortoises mature at
a carapace length of about 22-24 cm which, in
South Florida, correlates to about 9-15 years
of age (Mushinsky et al. 1994).
Analysis
We used chi-squared goodness of t tests
for independent samples to evaluate burrow
placement across the dierent soil and vegeta-
tion types. We used the statistical package R to
calculate chi-squared and p-values for our sam-
ples. Our null hypothesis stated that burrow
placement was random and did not depend on
soil or vegetation types. We did not include
the high vegetation category in our analysis
because it did not contain any burrows. We
also removed any portion covered by high
vegetation from the areas of the respective soil
types in which the high vegetation occurred.
Beyond this, we were not able to account
for interactions between soil and vegetation
types. We analyzed active burrows and their
distribution across the dierent soiland vege-
tation types to elucidate burrowing preferenc-
es in current conditions. Abandoned burrows
were analyzed against soil and vegetation
types to shed light on abandonment rates in
these characteristics.
Results
Tortoise Population
Our gopher tortoise burrow survey revealed
199 active burrows and 63 abandoned bur-
rows. is translates to an active to abandoned
burrow ratio of about 3:1. Using a burrow
correction factor of active burrows divided
by two, as established by the Florida Fish and
Wildlife Conservation Commission (FWC), we
estimated that 100 tortoises inhabit this 36.8
ha site (FWC 2008). We captured 63 tortoises
during our study period from June 2010 to
March 2011(Fig.3). Of those captured, ve
were juveniles, two were sub-adults and
FAURJ
28 Volume 1, Issue 1 Fall 2012
FAURJ
Figure 2. Carapace length distribution of gopher tortoises captured in the FAU conservation area.
We caught 63 tortoises from June 2010 to March 2011. An additional 5 juveniles or hatchlings
and 12 adults were observed but not captured.
56 were adults. We observed ve additional
juveniles and 12 adults but were not able
to capture them for direct measurements.
Tortoise carapace length ranged from 6 cm to
36 cm. We captured several small juvenile tor-
toises with a carapace length of < 7cm ranging
between 1 and 2 years old based on growth
annuli (Mushinsky et al. 2006). One older
juvenile was captured with a carapace length of
7cm and an estimated age of 4 years. We also
captured two sub-adults with a carapace length
of 22cm and about 14-16 years old based on
carapace length (Mushinsky et al. 1994). No
tortoises with carapace lengths between 7 and
22 cm were captured.
Burrowing Preferences
Our site contained four dierent soil types.
Urban Land, Basinger Fine Sand, Pompano
Fine Sand, and Immokalee Fine Sand covered
0.81ha, 3.24ha, 4.45ha, and 28.33ha, respec-
tively (Table 1). In Urban Land we found 18
total burrows, in Basinger we located 31, Pom-
pano contained 48, and Immokalee had 165.
Seventeenactive burrows were located in Urban
Land, 20 in Basinger Fine Sand, 40 in Pompano
Fine Sand, and 122 in Immokalee Fine Sand
(Table 1 & Fig. 2). One abandoned burrow was
located in Urban Land, 11 in Basinger Fine
Sand, eight in Pompano Fine Sand, and 43 in
Immokalee Fine Sand (Table 1 & Fig. 2).
We did not nd a signicant dierence in the
distribution of abandoned burrows across
dierent soil types (x2 = 4.4102, df = 3, p-value
= 0.2204) (Table 2). We found a statistical sig-
nicance in the distribution of active burrows
across the dierent soil types (x2 = 35.7096, df
= 3, p-value = 8.625e-08) (Table 3). Active
29 Volume 1, Issue 1 Fall 2012
FAURJ
burrows were most dense in Urban land at
21.0 burrows/ha. Active burrow densities were
also high in Pompano ne sand at 9.0/ha. In
Immokalee ne sand and Basinger ne sand,
the density of active burrows was 5.3/ha and
4.3/ha, respectively.
Low vegetation covered 21.04ha, medium
vegetation covered 10.12ha, and high vegeta-
tion covered 5.67ha (Table 4). We found 190
total burrows in low vegetation, 72 in medium
vegetation, and zero in high vegetation. High
vegetation occupied 5.67ha or about 15% of
the study site (Table 4). 141 Active burrows
were located in low vegetation and 58 in me-
Table1. Burrow distribution across dierent soil types in the FAU conservation area. Immokalee
soil covered 28.33ha and made up the greatest portion of our study area. Basinger and Pompano
soil types made up small portions of the habitat at 3.24 ha and 4.45ha, respectively. With 0.81ha,
Urban Land had the least representation at our site.
Soil types Active Burrows Abandoned Burrows Area (ha)
Urban 17 1 0.81
Basinger 20 11 3.04
Pompano 40 8 4.45
Immokalee 122 43 22.86
Totals 199 63 31.16
Table 2.Contingency table summarizing the chi-squared test for abandoned burrow distribution
among the four dierent soil types in the FAU conservation area. Expected frequencies were
based on area. Not signicant (x2= 4.4102, df = 3, p-value = 0.2204).
Pompano Basinger Urban Immokalee Row
Fine Sand Fine Sand Land Fine Sand Totals
Observed count 8 11 1 43 63
Expected frequencies 8.9971 6.1463 1.6377 46.2189 63
(0.1428) (0.0976) (0.0260) (0.7336) (1)
Area (ha) 4.45 3.04 0.81 22.86 36.83
Table 3. Contingency table summarizing the chi-squared test for active burrow distribution
among the four dierent soil types in the FAU conservation area. Expected frequencies were
based on area. Signicant (x2 = 35.7096, df = 3, p-value = 8.625e-08).
Pompano Basinger Urban Immokalee Row
Fine Sand Fine Sand Land Fine Sand Totals
Observed count 40 20 17 122 199
Expected frequencies 28.4194 19.4146 5.1730 145.9929 199
(0.1428) (0.0976) (0.0260) (0.7336) (1)
Area (ha) 4.45 3.04 0.81 22.86 36.83
FAURJ
30 Volume 1, Issue 1 Fall 2012
FAURJ
Figure 3. e distribution of gopher tortoise burrows in dierent soil types in the FAU conserva-
tion area. Four soil types were found to occur in the FAU conservation area, namely Urban Land,
Basinger Fine Sand, Pompano Fine Sand, and Immokalee Fine Sand. Active burrows were very
dense in Urban Land at 21.0/ha. Active burrow densities were also high in Pompano Fine Sand at
9.0/ha. In Immokalee and Basinger Fine Sand, the density of active burrows was 5.3/ha and 6.6/
ha, respectively.
Table 4. Burrow distribution across dierent vegetation types in the FAU conservation area. Low
vegetation covered 21.04ha and covered the largest portion of our study site. Medium and high
vegetation covered 10.12ha and 5.67ha, respectively.
Vegetation types Active Burrows Abandoned Burrows Area (ha)
Low 141 49 21.04
Medium 58 14 10.12
High 0 0 5.67
Totals 199 63 36.83
31 Volume 1, Issue 1 Fall 2012
FAURJ
Table 5. Contingency table summarizing the chi-squared test for the distribution of abandoned
burrows among the three dierent vegetation types in the FAU conservation area. Expected fre-
quencies were based on area. Not signicant (x2 = 3.02, df = 1, p-value = 0.08).
Low Medium Row Totals
Observed counts 49 14 63
Expected frequencies 0.6752 0.3248 1
Area (ha) 21.04 10.12 31.16
Figure 4. e spatial distribution of gopher tortoise burrows in dierent vegetation heights in
the FAU conservation area. A total of 199 active and 63 abandoned burrows were found. We found
no signicant dierence between the spatial distribution of burrows in medium and low vegeta-
tion (x2 = 3.0; df = 1; p < 0.05). Active burrow densities were 6.7/ha in low vegetation and 5.7/ha
in medium vegetation. No burrows were located in high vegetation.
FAURJ
32 Volume 1, Issue 1 Fall 2012
FAURJ
dium vegetation (Table 1, Fig. 3). In low vege-
tation we found 49 abandoned burrows and in
medium vegetation we located 14 abandoned
burrows (Table 4& Fig. 3).
We did not nd a signicant dierence in the
distribution of abandoned burrows across
dierent vegetation types (x2 = 3.02, df = 1,
p-value = 0.08) (Table 5). In addition, we found
no statistical signicance in the distribution of
active burrows across the dierent vegetation
types (x2 = 1.01, df = 1, p-value = 0.31) (Table
6). Active burrow densities were 6.7/ha in low
vegetation and 5.7/ha in medium vegetation.
Table 6. Contingency table summarizing the chi-squared test for the distribution of active bur-
rows among the three dierent vegetation types in the FAU conservation area. Expected frequen-
cies were based on area. Not signicant (x2 = 1.01, df = 1, p-value = 0.31).
Low Medium Row Totals
Observed counts 141 58 199
Expected frequencies 0.6752 0.3248 1
Area (ha) 21.04 10.12 31.16
Discussion
Our data coupled with King’s data from
2005, presents evidence that the tortoise
population in the FAU conservation area may
not be sustainable. In addition, we were able to
shed light on some of the tortoises’ burrowing
preferences, albeit more research must be
conducted on this topic in South Florida before
conclusions can be drawn to inform manage-
ment strategies.
e age structure prole created from our
direct captures further corroborates specula-
tion of an unsustainable population because
reproduction and juvenile survival seem to
be very low, even for this small population.
Our age structure graph shows that the FAU
population consists almost entirely of adult
gopher tortoises unlike healthy population-
swhich tend to have continuous representation
throughout the life stages (Diemer 1992,
Mushinsky et al. 1997). Our tortoise captures
show only two sub-adults and a gap of about
ten years between the youngest juvenile and
smallest sub-adult. On account of the ten juve-
niles and lack of sub-adults found, we suggest
that the population is reproducing successfully
but the ospring are not surviving. is demo-
graphic is consistent with King’s who captured
36 adults and one sub-adult and observed
an additional 30 adults (2005). Contrasted
to King’s survey, the number of tortoises
captured or observed in our study increased
by two adults, one sub-adult and ten juveniles.
e increase in sub-adults and adults is likely
due to relocations. From the FAU-employed
consultant, we know that several tortoises
were relocated to our site from other areas of
the FAU Boca Raton campus (M. Brandenburg,
Miller Legg and Assoc., unpubl. data). We also
speculate that tortoises have been moved to
the site illegally by the public. Juveniles may
have increased for several reasons. King con-
ducted her surveys during the late spring and
early summer months while we conducted ours
at the end of summer and the beginning of
fall (2005). Hatchling tortoises tend to hatch
in the fall (Mushinsky et al. 2006, Ashton and
Ashton 2008). In addition, juveniles are very
secretive and do not necessarily burrow (Dou-
glass 1978, Epperson and Heise 2003, Pike
2006, Ashton and Ashton 2008). ey may
have simply gone undetected in King’s survey
and we too may have missed some. Conse-
quently we base our discussion of juveniles on
the relative abundance of sub-adult and adult
tortoises and their burrows which are much
easier to locate.
33 Volume 1, Issue 1 Fall 2012
FAURJ
Juvenile tortoises have very low survival
rates and it is possible that the lack of sub-
adults seen in our study is a result of natural
uctuations (Butler and Sowell 1996, Epperson
and Heise 2003). However, the consistency
in the lack of sub-adults between 2005 and
2011 weaken this hypothesis by suggesting
that almost all juveniles perished during this
period.Reasons for theconsistently low surviv-
al rate of juveniles may include a complex of
factors involving habitat loss, degradation, and
predators. Adult tortoises are robust, versatile
foragers consuming over 200 plant genera and
able to forage across home ranges greater than
3 acres (Mushinsky et al. 2003, Mushinsky et
al. 2006, Ashton and Ashton 2008). Juvenile
tortoises are less versatile foragers and have
home ranges under 0.5 acres as a result of their
smaller size and increased susceptibility to
predators (Mushinsky et al. 2003, Mushinsky
et al. 2006). Since a habitat is fragmented
and begins decreasing in available herbaceous
vegetation, adult tortoises may stand a better
chance of survival than juveniles. In particu-
lar, the lower abundance of food may require
hatchling tortoises to forage in a wider range
and as a result make them more susceptible
to predation. Adult reproductive success may
also be lowered due to increased stress levels
imposed by a declining habitat. Assessment
methods usually focus on adult tortoise counts
and fail to assess their reproductive success or
survival of the juvenile tortoises (McCoy et al.
2006). Due to the longevity of adult tortoises,
population declines can take several decades
to become noticeable (McCoy and Mushinsky
2007).
e high active to abandoned burrow ratio
in this particular habitat is a cause for concern
as well. Active to abandoned tortoise burrow
ratios can be used to indicate the tendency of
gopher tortoises to dig new burrows which
relates directly to the status of a population
(Mushinsky et al. 1997, McCoy et al. 2006).
Our active to abandoned burrow ratio was
about 3:1 and similar to studies on other small
habitats and true island habitats that support
gopher tortoises (Mushinsky et al. 1997,
Eubanks et al. 2003). Studies in unrestricted
tortoise habitats generally report ratios in
which the number of abandoned burrows is
very similar to the number of active burrows
(Mushinsky et al. 1997, Jones and Dorr 2004).
e high ratio found at the FAU site and other
small habitats suggests that tortoises in these
habitats are limited in their ability to disperse
freely to create new burrows and abandon
old ones. is pattern of behavior has been
demonstrated to be a result of declininghabitat
conditions (Mushinsky et al. 1997, McCoy et
al. 2006).
Burrow activity changes over time are
important. In 2005, King reported 181 active
burrows and 38 abandoned burrows (we
have lumped King’s inactive burrows into the
active burrow category as described in the
methods section). e gures translate to an
active to abandoned burrow ratio of about 4:1
and reveal a relative increase in the number
of abandoned burrows. Other studies have
demonstrated that increases in the abandoned
burrow category accompanied by decreases
in the active burrow category suggest overall
habitat declines (Mushinsky et al. 1997,
McCoy et al. 2006). Such habitat declines may
also be evidenced by a population age structure
represented primarily by adults (McCoy et
al. 2006). Both predictors are evident in the
tortoise population at FAU.
e signicant correlation found between
the spatial distribution of gopher tortoise
burrows and soil types was expected. Tortoises
have been shown to choose well-drained soils
supportive of their extensive burrowing habits
(Mushinsky et al. 2006). e high density of
tortoise burrows in urban land suggests that
this soil type is well-suited for tortoise bur-
rowing. Immokalee, Basinger, and Pompano
ne sand appear to be less suitable which may
be due to poorer drainage thus accumulating
more water than urban land and threatening
burrow infrastructure.
FAURJ
34 Volume 1, Issue 1 Fall 2012
FAURJ
Spatial distribution of burrows throughout
dierent vegetation types (Fig. 4) displayed
unexpected trends for the FAU site. e lack of
signicance seen in the spatial distribution of
burrows in low versus medium vegetation sug-
gests overcrowding or novel habitat preferenc-
es among gopher tortoises. Contrary to typical
gopher tortoise ecology, tortoises in certain
parts of the conservation area preferred to dig
their burrows in areas with medium vegetation
(1.5 -3.0m) although open patches with low
herbaceous vegetation were available (Mushin-
sky et al. 2006, Ashton and Ashton 2008). We
suggest that this may be due to distribution of
a combination of dierent soil types and the
presence of more roots in dense vegetation.
Higher root densities may help to sustain
tortoise burrow infrastructure where certain
soil types would not. Roots may also increase
moisture content of the burrow, a potentially
important aspect in such dry environments as
our study site (Ashton and Ashton 2008). e
relatively high density of burrows located in
medium vegetation suggests novel habitat use
among tortoises atthe FAU site and warrants
further investigation. Complete avoidance
of habitat with complete canopy coverage is
frequently reported in gopher tortoise ecology
research and was corroborated by our study
(Diemer 1986, King 2005, Mushinsky et al.
2006, Ashton and Ashton 2008).
Based on active to abandoned burrow
ratios and the demographic prole we suggest
that the FAU tortoise population may not be
sustainable without increased habitat man-
agement. Additional management may include
mechanical clearing of tall woody plants and
removal of raccoons and other unnaturally
large concentrations of predators such as feral
cats.
Additionally, adult gopher tortoises are
able to survive for decades in habitat patches
unsuitable for juveniles: therefore burrow
observations alone may be deceiving (McCoy
and Mushinsky 2007). Without investigating
burrows beyond their mere existence, it is not
possible to determine whether a population
consists of primarily adults or comprises a con-
tinuous distribution from juveniles to adults.
Ultimately, when burrow surveys are coupled
with age structure proles more accurate
results may be obtained. However, capturing
individual tortoises is a time consuming eort
because they primarily remain underground.
An alternative solution may be to measure
burrow widths as these have been shown to
correlate well with carapace length and could
be used to indirectly obtain an age structure
prole (Alford 1980, McCoy et al. 2006). We
are currently investigating these correlations
in southeastern Florida habitatsto determine
how dierent soils throughout the tortoises’
range may or may not skew them.
In conclusion, a combination of active to
abandoned gopher tortoise burrow ratios,
direct tortoise captures, and habitat analyses
should be used to comprehensively assess tor-
toise population status and, when necessary,
develop management techniques. Southeast
Florida contains several fragmented habitats
that support tortoise populations. Unfortu-
nately most of these sites are degraded and
not monitored thoroughly. e remaining
tortoises on these sites, like those on the FAU
preserve, may simply be the last remnants of
a once thriving population now unsustainable
without signicant human intervention. Nev-
ertheless, small populations such as the one we
studied can maintain healthy, sustainable pop-
ulations with proper habitat management and
are valuable to the conservation of the tortoise
and its numerous commensal species (McCoy
and Mushinsky 2007, Mushinsky et al. 1997).
In fact, a site in Jupiter Florida hosts a healthy,
sustainable tortoise population on a much
smaller tract of land than the FAU site (Moore
et al. 2009). e Jupiter site is managed heavi-
ly by a combination of roller chopping, or clear
cutting, and invasive species removal by hand
(per. comm. J. Moore). Additionally, small hab-
itats are isolated from disease outbreaks, and
may serve as potential sources for genetic
35 Volume 1, Issue 1 Fall 2012
FAURJ
diversity and restocking eorts for other larg-
er tracts of land where tortoises have become
extinct (Mushinsky et al. 1997).
e primary driver of worldwide species
declines is habitat loss and fragmentation
(2010a). In addition, human development
often contributes to the elimination of bene-
cial natural disturbances such as re which
among other things prevents the establish-
ment of invasive species (2010a). We must
also monitor more eectively our conservation
areas and the status of the targeted, imperiled
species (Bruner et al. 2001, McCoy et al.
2006). Consequently, setting aside protected
conservation areas to aid species recovery is
only a rst step and must be supplemented by
intense habitat management to mimic natural
disturbances as well as thorough evaluation of
our eorts (Mushinsky et al. 1997, Bruner et
al. 2001, 2010a).
Acknowledgements
is study was supported by the National Sci-
ence Foundation through the Undergraduate
Research & Mentoring Program, the Gopher
Tortoise Council, Florida Atlantic University,
and the Terrestrial Ecology Laboratory at
FAU. We thank Jeanette Wyneken and Tricia
Meredith for their help in editing. We also
thank Leonardo Calle and all the volunteers
and students involved in direct and indepen-
dent studies that helped with data collection
and analysis.
References
2010a. Conservation Biology for All. Oxford
University Press, New York.
2010b. Soil Survey Geographic (SSURGO) \
Database for Palm Beach County, Florida. U.S.
Department of Agriulture, Natural Resources
Conservation Service.
Alford, R. A. 1980. Population-Structure of
Gopherus polyphemus in Northern Florida.
Journal of Herpetology 14:177-182.
Ashton, R. E. and P. S. Ashton. 2008. Natural
History and Management of the Gopher Tor
toise Gopherus polyphemus (Daudin).
Original Edition edition. Krieger Publishing
Company, Malabar, Florida.
Breininger, D. R., P. A. Schmalzer, and C. R.
Hinkle. 1991. Estimating Occupancy of Go
pher Tortoise (Gopherus-polyphemus)
Burrows in Coastal Scrub and Slash Pine Flat
woods. Journal of Herpetology 25:317-321.
Bruner, A. G., R. E. Gullison, R. E. Rice, and G.
A. B. da Fonseca. 2001. Eectiveness of parks
in protecting tropical biodiversity. Science
291:125-128.
Butler, J. A. and S. Sowell. 1996. Survivorship
and predation of hatchling and yearling go
pher tortoises, Gopherus polyphemus. Jour
nal of Herpetology 30:455-458.
Diemer, J. E. 1986. e Ecology and Manage
ment of the Gopher Tortoise in the South
eastern United-States. Herpetologica 42:125-
133.
Diemer, J. E. 1992. Demography of the
Tortoise Gopherus-polyphemus in Northern
Florida. Journal of Herpetology 26:281-289.
Douglass, J. F. 1978. Refugia of Juvenile
Gopher Tortoises, Gopherus-polyphemus
(Reptilia, Testudines, Testudinae). Journal of
Herpetology 12:413-415.
Epperson, D. M. and C. D. Heise. 2003. Nesting
and hatchling ecology of gopher tortoises
(Gopherus polyphemus) in southern Missis
sippi. Journal of Herpetology 37:315-324.
Ervin, J. 2003. Protected area assessments in
perspective. Bioscience 53:819-822.
Eubanks, J. O., W. K. Michener, and C. Guyer.
2003. Patterns of movement and burrow use
in a population of gopher tortoises (Gopherus
polyphemus). Herpetologica 59:311-321.
FWC. 2008. Gopher Tortoise Permitting
Guidelines: Gopherus polyphemus. Florida
Fish and Wildlife Conservation Commission
(FWC), Tallahassee, Florida.
Holder, G., M. Allen, J. Berish, M. Hinkle, A.
FAURJ
36 Volume 1, Issue 1 Fall 2012
FAURJ
Kropp, R. McCann, A. Williams, and M.
Yaun. 2007. Gopher Tortoise Management
Plan: Gopherus polyphemus. Florida Fish and
Wildlife Conservation Commission, Tallahas
see, FL.
Jones, J. C. and B. Dorr. 2004. Habitat associa
tions of gopber tortoise burrows on industrial
timberlands. Wildlife Society Bulletin 32:456-
464.
King, E. H. 2005. Gopher Tortoise Habitat Use
and Spatial Distribution in a Southeastern
Florida Population. Florida Atlantic Universi
ty, Boca Raton.
McCoy, E. D. and H. R. Mushinsky. 2007. Es
timates of minimum patch size depend on the
method of estimation and the condition of
the habitat. Ecology 88:1401-1407.
McCoy, E. D., H. R. Mushinsky. 1994. Eects of
fragmentation on the richness of vertebrates
in the Flroida Scrub Habitat. Ecology 75:
446-457.
McCoy, E. D., H. R. Mushinsky, and J. Lindzey.
2006. Declines of the gopher tortoise on
protected lands. Biological Conservation
128:120-127.
Moore, J. A., M. Strattan, and V. Szabo. 2009.
Evidence for Year-round Reproduction in the
Gopher Tortoise (Gopherus polyphemus) in
Southeastern Florida. Bulletin of the Peabody
Museum of Natural History 50:387-392.
Mushinsky, H. R., E. D. McCoy, J. E. Berish, R.
E. J. Ashton, and D. S. Wilson. 2006. Go
pherus polyphemus - Gopher Tortoise.
Biology and Conservation of Florida Turtles
3:350-375.
Mushinsky, H. R., E. D. McCoy, and D. S.
Wilson. 1997. Patterns of Gopher Tortoise
Demography in Florida. Pages 252-258 in
Conservation, Restoration, and Management
of Tortoises and Turtles. New York Turtle and
Tortoise Society, New York.
Mushinsky, H. R., T. A. Stilson, and E. D.
McCoy. 2003. Diet and dietary preference of
the juvenile gopher tortoise (Gopherus poly
phemus). Herpetologica 59:475-483.
Mushinsky, H. R., D. S. Wilson, and E. D.
McCoy. 1994. Growth and Sexual Dimor
phism of Gopherus-polyphemus in Central
Florida. Herpetologica 50:119-128.
Nichols, J. D., J. E. Hines, J. R. Sauer, F. W.
Fallon, J. E. Fallon, and P. J. Heglund. 2000.
A double-observer approach for estimating
detection probability and abundance from
point counts. Auk 117:393-408.
Pike, D. A. 2006. Movement patterns, habitat
use, and growth of hatchling tortoises, Go
pherus polyphemus. Copeia:68-76.
Schwartz, T. S. and S. A. Karl. 2005. Popula
tion and conservation genetics of the gopher
tortoise (Gopherus polyphemus). Conserva
tion Genetics 6:917-928.
Service, U. S. F. a. W. 2011. Endangered and
reatened Wildlife and Plants; 12-Month
Finding on a Petition To List the Gopher
Tortoise as reatened in the Eastern Portion
of Its Range; Final Rule. Pages 45129-45162
in D. o. t. Interior, editor. Federal Register.
USFWS. 2011. Endangered and reatened
Wildlife and Plants; 12-Month Finding on a
Petition To List the Gopher Tortoise as
reatened in the Eastern Portion of Its
Range; Final Rule. Pages 45129-45162 in D. o.
t. Interior, editor. Federal Register.
Westemeier, R. L., J. D. Brawn, S. A. Simpson,
T. L. Esker, R. W. Jansen, J. W. Walk, E. L.
Kershner, J. L. Bouzat, and K . N. Paige. 1998.
Tracking the long-term decline and recovery
of an isolated population. Science 282.
