CHAPTER 1
LITERATURE REVIEW OF HERPETOFAUNAL RESPONSE TO FOREST
MANAGEMENT PRACTICES ALONG WITH STUDY SITE DESCRIPTION AND
RESEARCH HYPOTHESES
Literature Review
Herpetofaunal responses to forest management practices are dependent upon
several factors. First, disturbance type may be the most important factor in studies
examining herpetofaunal response to forest disturbance, because some forest
management practices are more obtrusive than others. Second, herpetofaunal response is
dependent on the taxa studied. Third, physiographic factors, such as study location, play
vital roles in determining herpetofaunal responses.
With nearly 39 million hectares (19%) of all timberland in the United States
managed by the National Forest Service (Smith et al., 2001), there is an increasing need
to evaluate effects of disturbance on forest ecosystems. Amphibians and reptiles play
important roles in these ecosystems as both predators and prey, and evaluation of forest
disturbances on herpetofauna has been increasingly recognized as important by natural
resource managers (deMaynadier and Hunter, 1994; Russell et al., 2004).
Forest management practices comprise a broad range of disturbance scales and
intensities and it is important to factor in the type of disturbance when evaluating
herpetofaunal response (Barrett and Guyer 2008). A majority of herpetofaunal forest
management research has evaluated amphibian response to clearcutting. Three seminal
studies initially indicated negative responses of amphibians to clearcut harvesting (Enge
and Marion 1986; Ash 1988; Petranka et al. 1993). Since these studies, additional
research has also indicated that population parameters of adult amphibians (Clawson et
al. 1997; Perison et al. 1997; Grialou et al. 2000; Knapp et al. 2003; Karraker and Welsh
2006; Patrick et al. 2006; Perkins et al. 2006, Homyack and Haas 2009) and juvenile
amphibians (Knapp et al. 2003; Patrick et al. 2006) respond negatively to these
treatments. Furthermore, clearcuts cause drastic changes in environmental conditions
(i.e., increased temperature and decreased soil moisture), which may increase the risk of
dessication (Rothermel and Luhring 2005), and appear to negatively impact amphibian
populations up to 25 years after disturbance (Karraker et al. 2006) and may take up to 60
years for certain amphibian species to recover to pre-harvest counts (Homyack and Haas
2009). Conversely, Ash (1997) estimated that salamander counts may recover to preharvest densities 20-24 years after harvesting and re-population of harvested stands likely
occurs from adjacent unharvested forest stands (Ash et al. 2003). Other studies have
found minimal impacts of clearcuts on adult amphibian population parameters (Diller and
Wallace 1994; Chazal and Niewiarowski 1998; Renken et al. 2004; Goldstein et al. 2005)
or species-specific responses (Graeter et al. 2008) and recommend that clearcut
harvesting and amphibian biodiversity can be sustained by concentrating harvests into
relatively small patches throughout the landscape (Knapp et al. 2003; Renken et al.
2004).
Intermediate forestry operations are more difficult to classify because most
publications either lump together regeneration harvests (e.g., shelterwood and seed-tree
harvests), thinning operations (e.g., thinning from above and from below), and forms of
uneven aged management strategies (e.g., gap creation) as intermediate forestry
operations or do not adequately describe the management strategy. Regeneration harvests
such as shelterwood cuts (Harpole and Haas 1999; Knapp et al. 2003) appear to have
mixed effect on amphibians, whereas intermediate operations such as thinning (Naughton
et al. 2000; Grialou et al. 2000) either negatively affect certain salamander species or
show no negative impacts to salamander populations (Brooks 1999). Conversely, unevenaged management, such as group selection and canopy gap creation through single tree
selection appear to minimally affect amphibian populations in some studies (Messere and
Ducey 1998; Greenberg 2000; McKenny et al. 2006), but have also shown speciesspecific negative impacts to some amphibian populations (Cromer et al. 2002;
MacCracken 2005). The creation of canopy gaps has been shown to positively influence
the presence of Green Treefrogs (Hyla cinerea; Cromer et al. 2002; Horn et al. 2005),
whereas selective harvesting (i.e., high-grading) has species-specific impacts on streamdwelling frog species (Lemckert 1999). Unfortunately, classification of intermediate
operations in the published literature is not well described and more emphasis needs to be
placed on describing the management method.
Amphibian responses to prescribed burning is largely unknown because most
analyses have been limited to small plots with unreplicated treatments, retrospective
study designs, and biases towards certain ecoregions (i.e., southeastern coastal plain;
Russell et al. 1999; Russell et al. 2004). Prescribed fire appears to have negative short-
term effects on amphibian species that inhabit areas that are not fire prone ecosystems
(Kirkland et al. 1996; Cole et al. 1997; Mcleod and Gates 1998) or areas where the
natural fire regime has been altered (Schurbon and Fauth 2003; Means et al. 2004).
Prescribed burning has also been found to have negligible impacts on amphibians (Ford
et al. 1999; Moseley et al. 2003; Greenberg and Waldrop 2008) and short-term positive
effects for certain species (Mushinsky 1985; Wilgers and Horne 2006; Hossack et al.
2009). Prescribed fire impacts are likely influenced by animal life stage (adult or larvae),
environmental conditions, season of burn, and whether a certain species has evolved
under a fire regime (Pilliod et al. 2003). This last point is very important because
herpetofauna that are historically adapted to fire-dependent ecosystems (i.e., Coastal
Plain) are likely to benefit from fire-induced habitat changes (Russell et al. 1999). Fire
may also alter environmental regimes that indirectly influence amphibian populations;
post-fire landslides have been shown to alter foraging dynamics in adult California newts
(Taricha tarosa) (Kerby and Kats 1998). Clearly more research is necessary to better
understand amphibian response to prescribed burning.
Far fewer studies have evaluated reptile response to forest management practices
in the eastern and southeastern United States when compared to amphibians. A majority
of studies have mostly found that lizard species respond positively to tree harvesting
operations (Adams et al. 1996; McLeod and Gates 1998; Renken et al. 2004; Goldstein et
al. 2005; Greenberg and Waldrop 2008; Kilpatrick et al. 2010), or demonstrate speciesspecific responses (Greenberg et al. 1994). Additionally, the manner in which forest gaps
are created, whether through natural means (treefalls) or anthropogenic means
(timbering) causes drastic changes in tropical lizard communities (Vitt et al. 1998).
Clearcutting has also been shown to negatively affect some reptile species (Enge and
Marion 1986), such as litter-dwelling snakes (Todd and Andrews 2007). Reptile response
to prescribed burning appears to be either positive (Moseley et al. 2003; Wilgers and
Horne 2006; Kilpatrick et al. 2010) or negligible (Mcleod and Gates 1998). However,
areas with high reptile species richness are predicted to have a wide array of responses to
wildfire disturbance (Driscoll and Henderson 2009). Fires with higher intensities and
frequent burning regimes appear to benefit certain lizard species including the six-lined
racerunner (Aspidoscelis sexlineata; Mushinsky 1985; Ruthven et al. 2008) and the fence
lizard (Sceloporus sp.; Greenberg 1994; Greenberg and Waldrop 2008). In regions such
as the southeast, prescribed burning is essential to maintain habitat for reptiles native to
longleaf pine ecosystems (Russell 1999; Means et al. 2004), such as the Gopher Tortoise
(Gopherus polyphemus; Yager et al. 2007); however our knowledge of herpetofaunal
responses to fire disturbances throughout other areas of the United States remains an
enigma (Bury 2004). No published studies exist regarding herpetofaunal response to
prescribed burning along physiographic provinces such as the Appalachian Mountains
and the Southern Cumberland Plateau. Increasing research efforts along these geographic
provinces is essential because herpetofaunal response to these disturbances has not been
adequately evaluated (Russell et al. 2004) and these regions have very diverse
herpetofaunal communities.
Forest fragmentation and forest edge creation are greatly increased by forest
management practices and may differentially affect amphibians and reptiles. Fragmented
habitats decrease gene exchange and increase susceptibility of amphibian and reptile
populations to disturbance events, which may lead to the loss of biological diversity
(Cunningham and Moritz, 1998; Dodd 2003). Forest edges have been found to negatively
impact most amphibians (Gibbs 1998a; DeGraaf and Yamasaki 2002; Semlitsch et al.
2007), but not all edge types illicit the same response (Gibbs 1988b). For example, forest
edges associated with roads decreased amphibian movements (described as permeability)
more drastically than forest edges in close proximity with fields and other open habitats
(Gibbs 1998b). Forest edge effects are not limited to the adjacent forest, but have been
found to extend approximately 70 m into the forest stand (deMaynadier and Hunter,
1998). Edge effects in tropical regions appear to be season specific, with edge use
depending on the shift between wet and dry seasons (Schlaepfer and Gavin 2001).
Forest types along with disturbance history are important factors in studies
examining vertebrate responses to forest management. For example, hardwood stands
support greater numbers of amphibian species than natural pine stands (Degraaf and
Rudis 1990) and managed pine stands (Bennett et al. 1980; Hanlin et al. 2000; Ryan et al.
2002). Stand history (i.e. time since last disturbance) was correlated with species richness
and counts; terrestrial salamander counts were greater in mature second-growth forests
than in recently clearcut stands (Buhlmann et al. 1988; Harper and Guynn 1999; Herbeck
and Larsen 1999; Duguay and Wood 2002; Karraker and Welsh 2006). An association
between greater counts of three late-seral stage amphibian species and increasing forest
age was found, illustrating the importance of old-growth forest stands (Welsh 1990;
Stoddard and Hayes 2005; Ashton et al. 2006). In addition, these late-seral stage
amphibians were found to be sensitive to degrees of stream sedimentation, supporting
their use as bio-indicators (Welsh and Ollivier 1998). Conversely, Loehle et al. (2005)
suggested that recently disturbed forest stands had higher herpetofaunal species richness
than older forest stands.
Herpetofaunal studies coupled with forest disturbance provide useful information;
although our ability to draw predictable conclusions has been limited by factors such as
inadequate stand replication, lack of pre-treatment data, and short sampling periods (<2
years). Conventional studies relate organismal abundance and population viability to
forest composition, but do not examine mechanisms responsible for causing observed
population fluctuations (Figure 1.1 in text; Marzluff et al., 2000). In addition, studies of
herpetofaunal response to ecological disturbance in the southeastern Appalachians focus
on salamander populations, whereas reptile species are largely ignored (Greenberg 2000).
Figure 1.1 Proposed mechanisms linking overall forest condition to herpetofauanal
viability. Bottom curved line represents the route of conventional studies failing to
identify causal mechanisms. Adapted from Marzluff et al. 2000.
Overall
Forest
Condition
Canopy coverage
Litter composition
Vegetation structure
Coarse woody debris
Forest Stratification
Air and soil temperature
Relative humidity
Light intensity
Population
Viability
Study Site Description
My study took place in the William B. Bankhead National Forest (BNF), located
in Lawrence, Winston, and Franklin Counties, of northwestern Alabama. Bankhead
National Forest is a 72,800 ha multi-use forest located in the southern Cumberland
Plateau (Gaines and Creed 2003). Mixed forests of the southern Cumberland Plateau tend
to be dominated by oak-hickory forest types (McWilliams 1991) except in areas where
pines were actively planted for commercial purposes. Loblolly pine (Pinus taeda L.) was
used to re-establish forests in abandoned agricultural and heavily timbered areas (Gaines
and Creed 2003). Reforestation efforts along with natural growth have resulted in 31,600
ha of Loblolly Pine throughout BNF (Gaines and Creed, 2003). For the past decade,
Southern Pine Beetle (Dendroctonus frontalis Zimmermann) infestations have affected
Loblolly Pine stands, producing large numbers of standing dead trees and increased fuel
loads, elevating the risk of damaging wildfires (Gaines and Creed 2003). Because
harvesting and fires have been prevented in forests throughout the study area for decades,
BNF has initiated a Forest Restoration Plan to reduce wildfire risk and promote natural
forest growth through tree thinning and prescribed fire disturbance. The BNF has not
traditionally utilized prescribed fire as a management tool, but has opted to include
prescribed burning in the forest restoration plan due to administrative recommendations.
Forest restoration plans in BNF mirror regulations set forth in the Healthy Forest
Restoration Act, which authorizes advanced vegetation management projects when
specified conditions (existence of insect or disease epidemic) pose a significant threat to
ecosystem health (Healthy Forest Restoration Act 2003). Collaboration with BNF has
provided a unique cost-effective opportunity to implement my research.
The selected forest stands were generally located on upland sites that were
composed of loblolly pine 25-50 years of age that also possessed a considerable
hardwood component (Gaines and Creed 2003). At the time of this study, these stands
had not been recently harvested and each stand had varying levels of damage from the
southern pine beetle. Other past disturbances included the clearing of hardwood stands
throughout the region for loblolly pine plantations during the early 1970’s (Gaines and
Creed 2003).
Experimental Design
Experimental design for this project consisted of a before-after and control-impact
(BACI) randomized complete block design. Forest manipulation treatments consisted of
three thinning levels (no thin, 17 m2ha-1 residual basal area [BA], and 11 m2ha-1 residual
BA) along with two burn treatments (no burn and burn; Table 1.1 in text). Each of the six
treatments in this experiment was replicated three times across the landscape (Figure 1.2).
Each experimental stand was approximately 9 ha in size and was blocked accordingly by
time of treatment (year) and location. Pre-treatment data were recorded within each stand
for one field season, whereas post-treatment data were collected over a period of three
field seasons.
Table 1.1 Experimental design: two-factor, complete block design. Treatments include
two burn treatments and three thinning levels. Each treatment was replicated three times.
(Residual basal area)
Thinning treatment
Burning treatment
Burn
No Burn
Control
3
3
17 m2 ha-1
3
3
11 m2 ha-1
3
3
Forest Treatments
All information regarding forest treatment description was taken from
(Schweitzer and Tadesse, 2004). Forest treatments followed the prescriptions set forth in
the BNF’s forest restoration plan (see Gaines and Creed 2003). All thinning procedures
were completed from below and were performed at two levels of BA retention (Figure
1.2). Tree thinning was implemented during fall and summer months and was continued
until the desired residual basal area had been achieved. Hardwood tree species, such as
Quercus spp. and Carya spp. were preferentially retained during the thinning as much as
possible. Thinning procedures in thin*burn plots was completed in the summer or fall
before prescribed burns were implemented.
Prescribed burns consisted of low-intensity fires, which were completed during
the dormant season (December-February) while air temperatures were low and relative
humidity was high. Backing fires were initiated to ensure that prescribed burns affected
understory and litter layers. Prescribed burns were ignited by drip torches and were
managed by the BNF U.S. Forest Service.
Hypotheses
Based on literature reviews and previous research experience, I developed the
following research hypotheses:
Hypothesis 1. Microhabitat and microclimate parameters will be affected be affected by
forest management practices.
(1-A) Presence of cover objects (logs and slash) will be differentially affected by
treatment type and will be greatest in thinned stands and lowest in burn stands due to
combustion during treatment. (1-B) Litter depth will be greatest in control and thinned
stands and lowest in burned stands. (1-C) Vegetation cover will increase with elevated
thinning levels and will be lowest in control and burned stands. (1-D) Microclimate
parameters such as air temperature and soil temperature will increase with greater canopy
removal. (1-E) Thinned stands (i.e., 11 m2/ha residual basal area [BA]) will show the
most drastic fluctuations in the above parameters due to reduction of leaf litter layer and
forest understory. (1-F) Microhabitat complexity and heterogeneity will be greatest in
thinned plots and lowest in burned and control stands.
Hypothesis 2: Herpetofaunal species richness, biomass, and relative abundance will vary
throughout treatment types and will be correlated with habitat heterogeneity.
(2-A) As habitat complexity and heterogeneity will be greatest in thinned plots,
herpetofaunal species richness will also be greatest in these stands. (2-B) Amphibian
species richness, biomass, and counts will be greatest in control plots and lowest in
prescribed burn stands. (2-C) Relative abundance and biomass of reptiles will be greatest
in thin-only stands. (2-D) Additional interactions will be found in stands receiving
burning and thinning treatments; herpetofaunal response in these stands may be unique in
comparison to stands receiving a single treatment.
Hypothesis 3: Microhabitat, microclimate, and herpetofaunal population variables
(species richness, biomass, and relative abundance) will vary temporally throughout
treatment stands.
(3-A) Due to combustion during burning, litter depth will change most drastically
in prescribed burn stands. (3-B) Habitat complexity in thinned stands will continue to
increase after treatments and will remain greatest in intermediate retention stands. (3-C)
As tree canopy increases, air temperature and soil temperature will decrease significantly
in thinned stands (3-D) Amphibian population variables will increase in burned plots as
understory cover and litter depth increase. (3-E) In thinned stands, reptile biodiversity
measures (e.g., species richness and diversity) will continue to increase after treatments.
Hypothesis 4. Home range and habitat use patterns of northern copperheads will depend
on gender and will be affected by forest management practices.
(4-A) Snake home ranges will depend largely on gender and reproductive
condition, and therefore will be larger for male individuals compared to gravid females.
(4-B) At the microhabitat scale, male snakes will prefer shadier sites with greater cover,
whereas female snakes will prefer more open sites with less cover. (4-C) Male and
female snakes will select different macrohabitat attributes and selected sites will be
different than the available habitat. (4-D) Both male and female snakes will select
macrohabitats that benefit physiological processes (e.g., gestation or foraging sites) and
the selection of these sites may be related to microhabitat availability throughout the
landscape.
Hypothesis 5. Lizard response and habitat use patterns will be dependent on
environmental and habitat characteristics in treated forest stands.
(5-A) Lizards will exhibit species-specific responses to prescribed burning and
thinning. (5-B) Each lizard species will be associated with different microhabitat and
microclimate characteristics. (5-C) Lizards will exhibit unique responses in thin and burn
interaction plots compared to other single treatment plots.
Study Significance
Evaluating changes in forest structure is necessary for predicting vertebrate
responses to forest disturbances. In eastern and southeastern forests amphibians and
reptiles reach high densities and in the case of some salamander species, may be the most
abundant vertebrate group (Burton and Likens 1975a; Petranka and Murray 2001).
Herpetofauna play important roles in forest ecosystems; salamanders are important as top
predators in forest detrital food webs (Burton and Likens 1975b; Wyman 1998), whereas
snakes are essential for population control in small mammal communities (Fitch et al.
1949). Evaluation of herpetofauanal population trends is difficult because many species
go through periods of decline and increase (Wake 1991), and longterm experimental
studies with pre-treatment data are essential to fully evaluate herpetofaunal response to
forest ecosystem alteration (Pechmann et al. 1991; Russell et al. 2004).
My study provides information on herpetofaunal response to forest disturbance
and has established species records for the Alabama Herpetological Atlas. In addition to
my research, other colleagues from Alabama A&M University have evaluated effects of
forest disturbance on avian, mammalian, arthropod, vegetation composition and structure,
and soil characteristics. This multidisciplinary approach will permit BNF to weigh
tradeoffs and benefits of each management strategy. The BNF and its partners are
enthusiastic about the comprehensive results to which all partners have contributed.
The proposed study will examine the effects of burning and thinning disturbances
on herpetofaunal community structure and function. Although previous studies have
examined herpetofaunal response to forest management, very few studies have examined
herpetofaunal response to prescribed burning and thinning simultaneously (but see
Greenberg and Waldrop 2008). In order to determine the impacts of forest management
on herpetofauna, it is essential to have a robust experimental design (Gardner et al. 2007).
Experimental design for this study will make it possible to determine factors responsible
for causing population changes in amphibian and reptile populations in disturbed forest
stands. No studies of herpetofaunal response to forest management in the Southern
Cumberland Plateau exist in the published literature. Specifically, this study will explore
the mechanisms contributing to herpetofaunal population fluctuations in forest
ecosystems. Identification of these mechanisms is essential because herpetofaunal
declines are more common than previously noted, with at least 43.2% of all amphibian
species experiencing population declines (Stuart et al. 2004). Some estimates have
indicated that 89% of all declining amphibians are due to large-scale habitat loss (Young
et al. 2004). Although most of the current focus has been placed on amphibians, Gibbons
et al. (2000) suggest that reptiles are declining at even faster rates than most amphibian
populations. The authors claim that habitat loss and degradation, introduced invasive
species, environmental pollution, disease, unsustainable use, and global climate change
are negatively affecting reptile populations. Declines of this magnitude are alarming
because herpetofauna function as indicator species, signaling the presence of an existing
environmental problem (Welsh and Olivier 1998). By recognizing factors that affect
herpetofauna, natural resource managers can develop plans that address these factors in
combination with forest health goals.
Bibliography
Adams, J. P., M. J. Lacki, and M. D. Baker. 1996. Response of herpetofauna to
silvicultural prescriptions in the Daniel Boone National Forest. Proc. Annu. Conf.
SEAFWA 50:312–320.
Ash, A. N. 1988. Disappearance of salamanders from clearcut plots. The Journal of the
Elisha Mitchell Scientific Society 104:116–122.
Ash, A. N. 1997. Disappearance and return of Plethodontid salamanders to clearcut plots
in the southern Blue Ridge Mountains. Conservation Biology 11:983–989.
Ash, A. N., R. C. Bruce, J. Castanet, and H. Francillon-Vieilott. 2003. Population
parameters of Plethodon metcalfi on a 10-year-old clearcut and in nearby forest in the
southern Blue Ridge Mountains. Journal of Herpetology 37:445–452.
Ashton, D. T., S. B. Marks, and H. H. Welsh, Jr. 2006. Evidence of continued effects
from timber harvesting on lotic amphibians in redwood forests of northwestern
California. Forest Ecology and Management 221:183–193.
Barrett, K., and C. Guyer. 2008. Differential responses of amphibians and reptiles in
riparian and stream habitats to land use disturbances in western Georgia, USA.
Biological Conservation 141:2290–2300.
Bennett, S. H., J. W. Gibbons, and J. Glanville. 1980. Terrestrial activity, abundance and
diversity of amphibians in differently managed forest types. The American Midland
Naturalist 103:412–416.
Brooks, R. T. 1999. Residual effects of thinning and high white-tailed deer densities on
northern red-back salamanders in southern New England oak forests. Journal of
Wildlife Management 63:1172–1180.
Buhlmann, K. A., C. A. Pague, J. C. Mitchell, and R. B. Glasgow. 1988. Forestry
operations and terrestrial salamanders: techniques in a study of the cow know
salamander, Plethodon punctatus. Pages 38–44 in R. Szaro, K. Severson, and D.
Patton, editors. Management of amphibians, reptiles, and small mammals in North
America. RM–GTR–166. U.S. Forest Service, Flagstaff, Arizona.
Burton, T. M. and G. E. Likens. 1979a. Salamander populations and biomass in the
Hubbard Brook Experimental Forest, New Hampshire. Copeia 1979:541–546.
Burton, T. M. and G. E. Likens. 1979b. Energy flow and nutrient cycling in salamander
populations in the Hubbard Brook Experimental Forest, New Hampshire. Ecology
56:1068–1080.
Bury, R. B. 2004. Wildfire, fuel reduction, and herpetofaunas across diverse landscape
mosaics in northwestern forests. Conservation Biology 18:968–975.
Chazal, A. C., and P. H. Niewiarowski. 1998. Responses of mole salamanders to
clearcutting: using field experiments in forest management. Ecological Applications
8:1133–1143.
Clawson, R. G., B. G. Lockaby, and R. H. Jones. 1997. Amphibian response to helicopter
harvesting in forested floodplains of low order, blackwater streams. Forest Ecology
and Management 90:225–235.
Cole, E. C., W. C. McComb, M. Newton, C. L. Chambers, and J. P. Leeming. 1997.
Response of amphibians to clearcutting, burning, and glyphosphate application in the
Oregon coast range. Journal of Wildlife Management 61:656–664.
Cromer, R. B., J. D. Lanham, and H. H. Hanlin. 2002. Herpetofaunal response to gap and
skidder-rut wetland creation in a southern bottomland hardwood forest. Forest
Science 48:407–413.
Cunningham, M., and C. Moritz. 1998. Genetic effects of forest fragmentation on a
rainforest restricted lizard (Scincidae: Gnypetoscincus queenslandiae). Biological
Conservation 83:19–30.
Degraaf, R. M., and D. D. Rudis. 1990. Herpetofaunal species composition and relative
abundance among three New England forest types. Forest Ecology and Management
32:155–165.
Degraaf, R. M., and M. Yamasaki. 2002. Effects of edge contrast on redback salamander
distribution in even-aged northern hardwoods. Forest Science 48:351–363.
deMaynadier, P. G. and M. L. Hunter, Jr. 1994. The relationship between forest
management and amphibian ecology: A review of the North American literature.
Environmental Reviews 3:230–261.
deMaynadier, P. G. and M. L. Hunter, Jr. 1998. Effects of silvicultural edges on the
distribution and abundance of amphibians in Maine. Conservation Biology 12:
340–352.
Diller, L. V., and R. L. Wallace. 1994. Distribution and habitat of Plethodon elongatus on
managed, young growth forests in north coastal California. Journal of Herpetology
28:310–318.
Dodd, C. K., Jr, and L. L. Smith. 2003. Habitat destruction and alteration: historical
trends and future prospects for amphibians. Pages 94–112 In R. D. Semlitsch, editor.
Amphibian Conservation. Smithsonian Books, Washington and London.
Driscoll, D. A., and M. K. Henderson. 2009. How many common reptile species are fire
specialists? A replicated natural experiment highlights the predictive weakness of a
fire succession model. Biological Conservation 141:460–471.
Duguay, J. P., and P. B. Wood. 2002. Salamander abundance in regenerating forest
stands on the Monongahela National Forest, West Virginia. Forest Science 48:
331–335.
Enge, K. M. and W. R. Marion. 1986. Effects of clearcutting and site preparation on
herpetofauna of a north Florida flatwoods. Forest Ecology and Management 14:177–
192.
Fitch, H. S. 1949. Study of snake populations in central California. American Midland
Naturalist 41:513–579.
Ford, W. M., M. A. Menzel, D. W. McGill, J. Laerm, and T. S. McCaye. 1999. Effects of
a community restoration on small mammals and herpetofauna in the southern
Appalachians. Forest Ecology and Management 114:233–243.
Gaines, G. D. and J. W. Creed. 2003. Forest health and restoration project. National
forests in Alabama, Bankhead National Forest Franklin, Lawrence, and Winston
Counties, Alabama. Final environmental impact statement. Management Bulletin R8–
MB 110B. 353 pp.
Gardner, T. A., J. Barlow, and C. A. Peres. 2007. Paradox, presumption and pitfalls in
conservation biology: the importance of habitat change for amphibians and reptiles.
Biological Conservation 138:166–179.
Gibbons, 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 50:653–656.
Gibbs, J. P. 1998a. Distribution of woodland amphibians along a forest fragmentation
gradient. Landscape Ecology 13:263–268.
Gibbs, J. P. 1998b. Amphibian movements in response to forest edges, roads, and
streambeds in southern New England. Journal of Wildlife Management 62:584–
589.
Goldstein, M. E., R. N. Wilkins, and T. E. Lacher, Jr. 2005. Spatiotemporal responses of
reptiles and amphibians to timber harvest treatments. Journal of Wildlife Management
69:525–539.
Graeter, G. J., B. B. Rothermel, and J. W. Gibbons. 2008. Habitat selection and
movement of pond-breeding amphibians in experimentally fragmented pine forests.
Journal of Wildlife Management 72:473–482.
Greenberg, C. H. 2000. Response of reptile and amphibian communities to canopy gaps
created by wind disturbance in the southern Appalachians. Forest Ecology and
Management 148:135–144.
Greenberg, C. H., D. G. Neary, and L. D. Harris. 1994. Effect of high-intensity wildfire
and silvicultural treatments on reptile communities in sand-pine scrub. Conservation
Biology 8:1047–1057.
Greenberg, C. H., and T. A. Waldrop. 2008. Short-term response of reptiles and
amphibians to prescribed fire and mechanical fuel reduction in a southern
Appalachian upland hardwood forest. Forest Ecology and Management 255:2883–
2893.
Grialou, J. A, S. D. West, and R. N. Wilkins. 2000. The effects of forest clearcut
harvesting and thinning on terrestrial salamanders. Journal of Wildlife Management
64:105–113.
Hanlin, H. G., F. D. Martin, L. D. Wike, and S. H. Bennett. 2000. Terrestrial activity,
abundance and species richness of amphibians in managed forests in South Carolina.
The American Midland Naturalist 143:70–83.
Harper, C. A. and D. C. Guynn, Jr. 1999. Factors affecting salamander density and
distribution within four forest types in the Southern Appalachian Mountains. Forest
Ecology and Management 114:245–252.
Harpole, D. N. and C. A. Haas. 1999. Effects of seven silvicultural treatments on
terrestrial salamanders. Forest Ecology and Management 114:349–356.
Healthy Forest Restoration Act. Act of December 3, 2003. 117 Stat., as amended; 16
U. S. C. 1887–1915.
Herbeck, L. A. and D. R. Larsen. 1999. Plethodontid salamander response to silvicultural
practices in Missouri Ozark forests. Conservation Biology 13:623–632.
Homyack, J. A., and C. A. Haas. 2009. Long-term effects of experimental forest
harvesting on abundance and reproductive demography of terrestrial salalamanders.
Biological Conservation 142:110–121.
Horn, S., J. L. Hanula, and M. D. Ulyshen. 2005. Abundance of green tree frogs and
insects in artificial canopy gaps in a bottomland hardwood forest. American Midland
Naturalist 153:321–326.
Hossack, B. R., L. A. Eby, C. G. Guscio, and P. S. Corn. 2009. Thermal characteristics of
amphibian microhabitats in a fire-disturbed landscape. Forest Ecology and
Management 258:1414–1421.
Karraker, N. E., and H. H. Welsh, Jr. 2006. Long-term impacts of even-aged timber
management on abundance and body condition of terrestrial amphibians in
northwestern California. Biological Conservation 131:132–140.
Kerby, J. L., and L. B. Kats. 1998. Modified interactions between salamander life stages
caused by wildfire-induced sedimentation. Ecology 79:740–745.
Kilpatrick, E. S., T. A. Waldrop, J. D. Lanham, C. H. Greenberg, and T. H. Contreras.
2010. Short-term effects of fuel reduction treatments on herpetofauna from the
southeastern United States. Forest Science 56:122–130.
Kirkland, G. L, Jr., H. W. Snoddy, and T. L. Amsler. 1996. Impact of fire on small
mammals and amphibians in a central Appalachian deciduous forest. American
Midland Naturalist 135:253–260
Knapp, S. M., C. A. Haas, D. N. Harpole, and R. L. Kirkpatrick. 2003. Initial effects of
clearcutting and alternative silvicultural practices on terrestrial salamander abundance.
Conservation Biology 17:752–762.
Lemckert, F. 1999. Impacts of selective logging on frogs in a forested area of northern
New South Wales. Biological Conservation 89:321–328.
Loehle, C., T. B. Wigley, P. A. Shipman, S. F. Fox, S. Rutzmoser, R. E. Thill, and M. A.
Melchiors. 2005. Herpetofaunal species richness responses to forest landscape
structure in Arkansas. Forestry Ecology and Management 209:293–308.
MacCracken, J. G. 2005. Effects of uneven-aged timber harvest on forest floor
vertebrates in the Cascade Mountains of southern Washington. Forest Ecology and
Management 208:123–135.
Marzluff, J. M., M. G. Raphael, and R. Sallabanks. 2000. Understanding the effects of
forest management on avian species. Wildlife Society Bulletin 28:1132–1143.
McKenny, H. C., W. S. Keeton, and T. M. Donovan. 2006. Effects of structural
complexity enhancement on eastern red-backed salamander (Plethodon cinereus)
populations in northern hardwood forests. Forest Ecology and Management 230:186–
196.
McLeod, R. F and J. E. Gates. 1998. Response of herpetofaunal communities to forest
cutting and burning at Chesapeake Farms, Maryland. American Midland Naturalist
139:164–177.
McWilliams, W. H. 1991. Availability of yellow pine sawtimber in Alabama. Pages 115–
119 in R. H. Jones, editor. Proceedings: Alabama’s forest resources: past, present, and
future. Auburn University.
Means, D. B., C. K. Dodd, Jr., S. A. Johnson, and J. G. Palis. 2004. Amphibians and fire
in longleaf pine ecosystems: response to Schurbon and Fauth. Conservation Biology
18:1149–1153.
Messere, M. and P. K. Ducey. 1998. Forest floor distribution of northern redback
salamanders, Plethodon cinereus, in relation to canopy gaps: first year following
selective logging. Forest Ecology and Management 107:319–324.
Moseley, K. R., S. B. Castleberry, and S. H. Schweitzer. 2003. Effects of prescribed fire
on herpetofauna in bottomland hardwood forests. Southeastern Naturalist 2:475–486.
Mushinsky, H. R. 1985. Fire and the Florida sandhill herpetofaunal community: with
special attention to responses of Cnemidophorus sexlineatus. Herpetologica 41:333–
342.
Naughton, G. P., C. B. Henderson, K. R. Foresman, and R. L. McGraw, II. 2000. Longtoed salamanders in harvested and intact douglas-fir forests of western Montana.
Ecological Applications 10:1681–1689.
Patrick, D. A., M. L. Hunter, Jr., and A. J. K. Calhoun. 2006. Effects of experimental
forestry treatments on a Maine amphibian community. Forest Ecology and
Management 234:323–332.
Pechmann, J. H. K., D. E. Scott, R. D. Semlitsch, J. P. Caldwell, L. J. Vitt, and J. W.
Gibbons. 1991. Declining amphibian populations: the problems of separating human
impacts from natural fluctuations. Science 253:892–895.
Perkins, D. W., and M. L. Hunter, Jr. 2006. Effects of riparian timber management on
amphibians in Maine. Journal of Wildlife Management 70:657-670.
Perison, D., J. Phelps, C. Pavel, and R. Kellison. 1997. The effects of timber harvest in a
South Carolina blackwater bottomland. Forest Ecology and Management 90:171–
185.
Petranka, J. W., M. E. Eldridge, and K. E. Haley. 1993. Effects of timber harvesting on
southern Appalachian salamanders. Conservation Biology 7:363–370.
Petranka, J. W. and S. S. Murray. 2001. Effectiveness of removal sampling for
determining salamander density and biomass: A case study in an Appalachian
streamside community. Journal of Herpetology 35:36–44.
Pilliod, D. S., R. B. Bury, E. J. Hyde, C. A. Pearl, and P. S. Corn. 2003. Fire and
amphibians in North America. Forest Ecology and Management 178:163–181.
Renken, R. B., W. K. Gram, D. K. Fantz, S. C. Richter, T. J. Miller, K. B. Ricke, B.
Russell, and X. Wang. 2004. Effects of forest management on amphibians and reptiles
in Missouri Ozark forests. Conservation Biology 18:174–188.
Rothermel, B. B., and T. M. Luhring. 2005. Burrow availability and dessication risk of
Mole salamanders (Ambystoma talpoideum) in harvested versus unharvested forest
stands. Journal of Herpetology 39:619–626.
Russell, K. R., D. H. Van Lear, and D. C. Guynn, Jr. 1999. Prescribed fire effects on
herpetofauna: review and management implications. Wildlife Society Bulletin
27:374–384.
Russell, K. R., T. B. Wigley, W. M. Baughman, H. G. Hanlin, and W. M. Ford. 2004.
Responses of southeastern amphibians and reptiles to forest management: A review.
Pages 319–334, in M. H. Rauscher and K. Johnsen, editors. Southern forest science:
past, present, and future. United States Department of Agriculture Southern Research
Station Gen. Tech. Rep. SRS–75.
Ruthven, D. C. III, R. T. Kazmaier, and M. W. Janis. 2008. Short-term response of
herpetofauna to various burning regimes in the south Texas plains. The Southwestern
Naturalist 53:480–487.
Ryan, T. J., T. Philippi, Y. A. Leiden, M. E. Dorcas, T. B. Wigley, and J. W. Gibbons.
2002. Monitoring herpetofauna in a managed forest landscape: effects of habitat types
and census techniques. Forest Ecology and Management 167:83–90.
Schlaepfer, M. A., and T. A. Gavin. 2001. Edge effects on lizards and frogs in tropical
forest fragments. Conservation Biology 15:1079–1080.
Schurbon, J. M. and J. E. Fauth. 2003. Effects of prescribed burning on amphibian
diversity in a southeastern U. S. national forest. Conservation Biology 17:1338–
1349.
Schweitzer, C. J. and W. Tadesse. 2004. The interaction of fire frequency and stand
density on plant community dynamics. Pages 1–14 in CREST–Center for Forest
Ecosystem Assessment (CFEA) Alabama A&M University. National Science
Foundation proposal.
Semlitsch, R. D., T. J. Ryan, K. Hamed, M. Chatfield, B. Drehman, N. Pekarek, M.
Spath, and A. Watland. 2007. Salamander abundance along road edges and within
abandoned logging roads in Appalachian forests. Conservation Biology 21:159–
167.
Smith, W. B, J. S. Vissage, D. R. Darr, and R. M. Sheffield. 2001. Forest Resources of
the United States, 1997. U.S. Department of Agriculture, Forest Service; General
Technical Report; NC-219. 190 pp.
Stoddard, M. A., and J. P. Hayes. 2005. The influence of forest management on
headwater stream amphibians at multiple spatial scales. Ecological Applications
15:811–823.
Stuart, S. N., J. S. Chanson, N. A. Cox, B. E. Young, A. S. L. Rodriques, D. L. Fischman,
and R. W. Waller. 2004. Status and trends of amphibian declines and extinctions
worldwide. Science 1103538:1–8.
Todd, B. D., and K. M. Andrews. 2007. Response of a reptile guild to forest harvesting.
Conservation Biology 22:753–761.
Vitt, L. J., T. C. S. Avila-Pires, J. P. Caldwell, and V. R. L. Oliveira. 1998. The impact of
individual tree harvesting on thermal environments of lizards in Amazonian rain
forest. Conservation Biology 12:654–664.
Wake, D. B. 1991. Declining amphibian populations. Science 253:860.
Welsh, H. H., Jr. 1990. Relictual amphibians and old-growth forests. Conservation
Biology 4:309–319.
Welsh, H. H., Jr., and L. M. Ollivier. 1998. Stream amphibians as indicators of ecosystem
stress: a case study from California’s redwoods. Ecological Applications 8:1118–
1132.
Wilgers, D. J. and E. A. Horne. 2006. Effects of different burn regimes on tallgrass
prairie herpetofaunal species diversity and community composition in the Flint Hills,
Kansas. Journal of Herpetology 40:73–84.
Wyman, R. L. 1998. Experimental assessment of salamanders as predators of detrital
food webs: effects on invertebrates, decomposition and the carbon cycle. Biodiversity
and Conservation 7:641–650.
Yager, L. Y. M. G. Hinderliter, C. D. Heise, and D. M. Epperson. 2007. Gopher tortoise
response to habitat management by prescribed burning. Journal of Wildlife
Management 71:428–434.
Young, B. E., S. N. Stuart, J. S. Chanson, N. A. Cox, and T. M. Boucher. 2004.
Disappearing jewels: the status of new world amphbians. NatureServe. Arlington,
Virginia. 53 pp.
Figure 1.2 Locations of study sites within the northwestern portion of the William B.
Bankhead National Forest, Alabama, U. S. A. Forest treatments consisted of the six
following categories: C-control, B-burn, HT-heavy thin, LT-light thin, HTB-heavy thin
and burn, and LTB-light thin and burn. Numbers after treatment abbreviations correspond
to block number.
HT3
HTB1
LT3
LT2
C3
B3
HT2
HTB2 HT1
C2
LT1
C1
LTB2
HTB3
LTB1
LTB3
B1
B2
CHAPTER 2
HABITAT RELATIONSHIPS OF REPTILES IN PINE BEETLE DISTURBED
FORESTS OF ALABAMA, U.S.A. WITH GUIDELINES FOR A MODIFIED DRIFTFENCE SAMPLING METHOD
Introduction
Examining species and habitat relationships is important for understanding
species habitat requirements and predicting the potential impacts of future habitat
changes. Some reptile species (e.g., semi-fossorial snakes) may be sensitive to habitat
alterations, whereas other species (e.g., heliothermic lizards) may respond positively to
drastic disturbances (Greenberg 1994; Vitt et al. 1998; Dodd et al. 2007; Todd and
Andrews 2008). Recent documentation of reptile declines has made it critical to
understand habitat relationships of these species (Gibbons et al. 2000). Although a
majority of the research examining herpetofaunal response to habitat disturbances have
been biased towards amphibians (Greenberg 2000), reptiles appear to respond positively
to habitat disturbances including silvicultural practices (Renken et al. 2004; Goldstein et
al. 2005; Greenberg and Waldrop 2008) and minor urbanization (Barrett and Guyer
2008). However, large-scale urbanization and habitat alteration appear to be the major
cause of widespread reptile population declines (Gibbons et al. 2000).
Habitat disturbances have the potential to affect reptile species in a variety of
manners. High intensity disturbances initially tend to favor reptile species that require
characteristics of open, early successional habitats (Mushinsky 1985; Greenberg et al.
1994a), whereas disturbances that maintain the structural integrity of the habitat (e.g.,
partial harvesting and stochastic events) may favor species that benefit from lowerintensity disturbances (Vitt et al. 1998; Greenberg 2000; Todd and Andrews 2007). A
majority of studies have examined reptile response to anthropogenic habitat disturbances
(Russell et al. 2004), but little research has examined reptile response to natural and
stochastic forest disturbances (but see Greenberg 2000). The Southern Pine Beetle
(Dendroctonus frontalis) is a forest pest that can cause major disturbances in southeastern
pine forests (Gaines and Creed 2003; Duncan and Linhoss 2005). If allowed to spread
unimpeded, this pest can have drastic effects on forest structure because it leads to the
creation of snags, large canopy gaps, and increased downed woody debris. However, no
research has evaluated reptile response to this type of disturbance and it is likely that
these disturbances greatly influence reptile populations.
An array of techniques exists to sample the overall reptile community including
drift-fence arrays, artificial cover objects, visual encounter surveys, and road-cruising
(Hutchens and DePerno, 2009a, b). It is important for researchers to evaluate sampling
requirements and employ several techniques in order to obtain a complete sample of the
overall herpetofaunal community (Ryan et al. 2002; Hutchens and DePerno 2009a). For
example, visual encounter surveys may work well for litter-dwelling herpetofauna, but
may be ineffective for fast moving, cryptic, or rare herpetofauna such as medium to
large-bodied snakes (Ryan et al., 2002). Furthermore, most reptile survey methods are not
effective for sampling larger snake species and therefore overlook an important portion of
the reptile biodiversity (Enge, 2001).
Drift-fence arrays provide an effective way to sample many reptile species (Enge
1997; Ryan et al. 2002), but larger snakes are not sampled adequately unless large box
traps are employed (Enge 2001; Burgdorf et al. 2005). Drift-fence arrays are useful for
collecting amphibian and population data and have been used in many studies including
evaluations of herpetofaunal response to forest disturbances (Greenberg 2000; Schurbon
and Fauth 2003; Renken et al. 2004). Numerous drift-fence array designs have been
employed (Corn 1994), with some designs more effective than others. Greenberg et al.
(1994b) utilized a combination of pitfall traps and box traps, but acknowledged that
funnel traps were much better for determining herpetofaunal species diversity than pitfall
traps. Intensive long-term sampling can be time consuming and cost prohibitive;
therefore passive sampling techniques, such as drift-fence surveys can serve as one
potential option to effectively sample rare and elusive species. Unfortunately, there is a
lack of published literature detailing the construction of box traps capable of capturing
larger snake species (but see Burgdorf et al. 2005).
My objectives were to (1) evaluate relationships between reptile community and
habitat characteristics in pine-hardwood forests of northwestern Alabama, U.S.A. and (2)
test the effectiveness of an alternative drift-fence trapping method to sample the reptile
species at these sites.
Materials and Methods
Study Site
My study sites were located in the William B. Bankhead National Forest (BNF)
along the southern terminus of the highly dissected portion of the Southern Cumberland
Plateau in Lawrence, Winston, and Franklin Counties of northwestern Alabama, U.S.A.
(Central Work Station, 34⁰ 20’ 38.0”N, 87⁰ 20’ 17.7”W). Bankhead National Forest is
nearly 72,000 ha in size and represents one of the largest tracts of contiguous forests in
the southeastern United States. Forests of the BNF are typically composed of upland
hardwood and pine species, such as the Chestnut Oak (Quercus prinus), Sourwood
(Oxydendron arboretum), Red Maple (Acer rubrum), Black Gum (Nyssa sylvatica),
Loblolly Pine (Pinus taeda), and Virginia Pine (Pinus virginiana). Over the last 30 years,
Loblolly Pine has been used to reforest areas of the BNF that were traditionally cleared
for agricultural purposes (Gaines and Creed 2003). The combination of overstocked
Loblolly Pine stands along with dry growing seasons have lead to widespread infestations
of the Southern Pine Beetle, causing extensive damage to many of the Loblolly Pine
stands. An estimated 7,527 hectares of pine forest have been killed by this epidemic in
the BNF (Gaines and Creed 2003) alone, a majority of which occurred during the 2001
growing season (Duncan and Linhoss 2006). Ultimately, these disturbances have caused
a large increase in coarse woody debris (CWD) throughout many forest stands of the
BNF. Throughout BNF, I randomly selected 18 forest stands based on site accessibility,
stand size (~9 ha), and similarity of past and future management plans.
The selected forest stands were generally located on upland sites that were
composed of Loblolly Pine 25-50 years of age that also possessed a considerable
hardwood component (Gaines and Creed 2003). At the time of this study, these stands
had not been recently harvested and each stand had varying levels of damage from the
Southern Pine Beetle. Other past disturbances included the clearing of hardwood stands
throughout the region for Loblolly Pine plantations during the early 1970’s (Gaines and
Creed 2003).
Reptile Sampling
I sampled reptiles using a trapping system consisting of three drift fences
(aluminum flashing) 15 m in length radiating from a central triangular box trap (see
Figure 2.1, 2.2, and 2.3) modified from Renken et al. (2004) and Bugdorf et al. (2005).
Large box traps have been proven successful for capturing reptiles, particularly medium
to large snake species (Burgdorf et al. 2005). I installed one drift-fence unit per study plot
after randomly determining the trap unit location by dividing each study plot into four
equal quadrants corresponding to the four cardinal directions. Drift fence arrays were
buried approximately 10 cm into the ground and included one large box trap at the
terminus of each drift fence (three per array) and two pitfall traps at the midpoint of each
drift fence (six per array). I designed a triangular-array system because it covers the same
area as a four-fence array system, but reduces the labor and cost of installing an
additional drift-fence and the accompanying traps. This trap arrangement presents the
best use of the entire trapping unit because a reptile can encounter the drift-fence at any
point and potentially be captured in the center or terminal box traps.
I sampled intermittently throughout March and April 2005 and began sampling
continuously from May through mid-November 2005. I resumed trapping from March
2006 and continued through June of the same year. During intermittent sampling periods,
I would open traps only for rain events, whereas during continuous sampling periods,
traps were open for a three-four day period and then closed for two to three days. I
checked traps daily when open between 0700-1400 hours to minimize animal mortality.
Across all sites I sampled a total of 584 trap nights (i.e., sites 1-6 [120 trap nights], sites
7-12 [204 trap nights], and sites 13-18 [240 trap nights]). For each animal capture, I
recorded trap type (pitfall or box trap), species, sex, life stage (juvenile, sub-adult, and
adult), snout-vent length (mm), total length (mm), mass (g), and animal viability. I
marked each animal with a plot-specific mark through toe-clips (lizards), scale clips
(snakes), and scute etching (turtles) to ensure that recaptured individuals would not be
counted in subsequent captures (Enge 1997). I released all marked individuals at a
minimum of 10 m on the side of the drift fence in which they were captured. I
acknowledge that my results are limited to reptile species that were sampled via driftfence arrays.
Habitat Sampling
I quantified habitat features using a line transect method modified from Herbeck
and Larsen (1999) and Greenberg (2000). I established three habitat sampling plots at
each trapping site, and determined plot placement a priori via a random compass bearing
(0°-360°) and distance (30-50 m) originating from the center of each trapping array. I
restricted habitat sampling to these distances to minimize the effects of habitat
disturbances that were created during trap installation. Each habitat sampling plot
consisted of two 20 m perpendicular transects placed north-south and east-west from the
habitat plot center. I used a two meter piece of 1.9 cm diameter polyvinyl chloride pipe as
a transect marker, and recorded the presence or absence of microhabitat features every
0.5 m along the transect line. Based on these data, I calculated percent coverage of litter,
bare ground, herbaceous and woody vegetation, slash, and coarse woody debris (CWD),
and rock (Table 2.1). I measured CWD length and diameter to determine CWD volume
based on the volume of a cylinder (VC = π r2 x L), and measured litter depth with a metric
ruler (measured every 2 m) and estimated percent canopy cover with a spherical
densiometer (measured every 5 m; Table 2.1). I assigned vertical forest structure values
on a scale of 1-4 by estimating the height of the dominant forest structure every 5 m and
assigned values according to the designations described in Forest Inventory and Analysis
(1998; Table 2.1).
Statistical Analysis
I examined correlations and relationships between habitat variables with principal
components analysis (PCA). After examining the biological relevance of the resulting
components, I retained one variable from highly correlated (≥ 0.90) variable sets for
subsequent analyses (Crosswhite et al. 2004). I calculated the total number of each reptile
species captured and corrected for trap nights sampled at each survey site by dividing
species counts by the total number of trap nights (1 trap night = 1 trap opened for 24 hrs),
and used canonical correspondence analysis (CCA) to examine the relationships between
the animal capture data and habitat features using CANOCO v.4.5. Canonical
correspondence analysis is a direct gradient analysis technique where the ordination
procedure is constrained by a set of a priori covariates (e.g., habitat and climate data)
predicted to affect the observed distribution of the organismal group in question (ter
Braak, 1995). Ordination techniques are essential when inter-specific and habitat
relationships are to be compared among large species groups. To improve the quality of
the CCA output, I did not include species with only one detection. Ter Braak (1995)
suggests that rare species have little influence on the analysis and can be removed if
necessary.
Results
I captured a total of 19 reptile species representing 226 individuals during 564
trap nights (Table 2.2). The most commonly captured lizard species were Little Brown
Skinks (Scincella lateralis, n=70) and Green Anoles (Anolis carolinensis, n=29), whereas
the most commonly captured snake species were Copperheads (Agkistrodon contortrix,
n=27) and Eastern Garter Snakes (Thamnophis s. sirtalis, n=10; Table 2.2). Overall,
lizards (Phrynosomatidae, Polychrotidae, and Scincidae) accounted for 65% (n=148) of
the total reptile captures, whereas snakes (Colubridae and Viperidae) accounted for 34%
(n=76) and turtles (Emydidae) accounted for <1% (n=2) of the total reptile captures
(Table 2.2). My box traps captured 66 total snakes in 564 trap nights (one trap night
equaled one trap array passively sampled for 24 consecutive hours), equaling 0.117 snake
captures per trap night.
The study stands were considerably different in terms of overall habitat
composition. Vegetative and woody coverage ranged from 3.8%-49.6% and 3.3%-31.7%,
respectively, whereas canopy cover and CWD ranged from 72.8%-90.1% and 0-491.1
m3, respectively (Table 2.3). Principal components analysis revealed five components
that explained 85% of the habitat variance (Table 2.4). Component one described a
gradient ranging from sites with high percent canopy cover to sites with less canopy
cover and greater CWD volume with greater woody, herbaceous, slash, and CWD cover,
whereas component two described a gradient ranging from sites with greater overstory
percent cover and litter depth to sites with less canopy cover and litter depth (Table 2.4).
Components three and four described gradients ranging from sites with greater
understory, overstory, and percent litter coverage to sites with less litter and tree
coverage, whereas component five described rock coverage across sites (Table 2.4).
Canonical correspondence analysis accounted for 40% (Axis 1: 16%, Axis 2:
13%, and Axis 3: 11%) of the overall variance in the species relationships and 58% (Axis
1: 24%, Axis 2: 18%, and Axis 3: 16%) of the species-habitat relationships. A distinct
habitat gradient was revealed, ranging from sites with greater canopy cover and litter
depth to sites with more woody, herbaceous, CWD, and percent slash coverage (Figure
2.4). A secondary gradient orthogonal to the first gradient ranged from sites with greater
percent CWD coverage to sites with greater percent rock coverage (Figure 2.4). Eastern
Worm Snakes (Carphophis a. amoenus) and Little Brown Skinks were associated with
sites with greater percent canopy cover and greater litter depth, whereas Copperheads,
Gray Ratsnakes (Pantherophis spiloides), Eastern Garter Snakes, and Eastern Fence
Lizards (Sceloporus undulatus) were associated with sites that had greater CWD volume
and percent CWD coverage (Figure 2.4). Interestingly, Green Anoles were located at the
ordination center indicating a broad, generalized distribution. Eastern Kingsnakes
(Lampropeltis getula nigra), Timber Rattlesnakes (Crotalus horridus), and Red
Cornsnakes (Pantherophis guttatus) were located along the periphery of the ordination
plot indicating their rarity and lack of association with any habitat characteristics. A
majority of the study plots were located along the portion of the gradient associated with
greater litter depth and canopy coverage. Fewer plots were located along the portion of
the gradient associated with greater woody, herbaceous, and CWD coverage and greater
CWD volume (Figure 2.5).
Discussion
I detected distinct relationships among several reptile species and habitat
characteristics. My data illustrated that Little Brown Skinks and Eastern Worm Snakes
were associated with greater litter depth and canopy cover, whereas Eastern Fence
Lizards, Gray Ratsnakes, and Copperheads were associated with greater CWD coverage,
CWD volume, and less canopy coverage.
Eastern Fence Lizards were associated with relatively open habitats and higher
CWD levels. This species is commonly associated with highly disturbed sites (Greenberg
et al. 1994b; Greenberg and Waldrop 2008), indicating that intense habitat disturbances
would likely benefit this species. Copperheads were also associated with open sites
possessing relatively high CWD volume. Cross and Peterson (2001) determined that
copperheads commonly utilized microhabitats that possessed structural diversity and
CWD cover. Little Brown Skinks and Eastern Worm Snakes are primarily found along
the forest floor and in this study were associated with greater litter depth and canopy
cover. Other species of litter-dwelling snakes have been detected in higher densities in
forest stands possessing greater canopy cover and litter depth (Todd and Andrews, 2008).
In addition to single species responses, CCA is advantageous because it permits analysis
of species relationships. My analysis revealed two distinct reptile groups including
species associated with greater percent litter and canopy cover (e.g., semi-fossorial
snakes and litter-dwelling lizards) and species associated with less percent canopy cover
and greater percent structural cover (e.g., heliothermic lizards and larger snake species).
Examining species data at the community level is advantageous because it may reveal
multiple species that share similar habitat associations (Crosswhite et al. 2004). In
addition, habitat management strategies can be developed from a multiple species
perspective. By understanding organismal habitat requirements from through this
perspective, forest managers can simultaneously develop conservation strategies that
benefit multiple species groups (Lindenmayer et al. 2000).
Although CCA was only able to explain 40% of the species relationships and 58%
of the species-habitat relationships, I believe that the results provide useful information
regarding reptile species and habitat relationships in disturbed forest stands. The low
variance explained is typical for this type of study because wildlife community data along
with the associated habitat relationships are often complicated by many confounding
variables and stochastic variations, which are difficult to be quantified or explained
(Palmer 1993). Interpretability of the CCA output is the most important part of this
analysis technique (Ter Braak 1995), and the gradients described in this study were
biologically relevant.
Many of the forest stands examined during this study had a history of damage
through southern pine beetle infestations. This insect pest is a normal inhabitant of
southeastern forests where it along with natural fire, play an important role in maintaining
equilibrium in pine-dominated forests (Land and Rieske 2006; Knebel and Wentworth
2007). Environmental stressors such as prolonged drought and overstocking of Pinus
species tends to increase the severity of Southern Pine Beetle infestations (Gaines and
Creed 2003; Duncan and Linhoss 2005). Although Loblolly Pine is a natural component
of southern forests, it was planted in unnaturally high densities in the BNF, easing the
spread of the Southern Pine Beetle throughout the study stands. Damage from Southern
Pine Beetle infestations usually leads to an increase of fallen logs and large canopy gaps
(Duncan and Linhoss 2005). My data suggest that these disturbances were likely
important for influencing the observed reptile community by creating large changes in the
overall forest structure. Although no published work has examined herpetofaunal
response to southern pine beetle infestations, increased snag density has been found to
influence the presence and aboreal activity of certain lizard species (James and
M’Closkey 2003), whereas the creation of canopy gaps by anthropogenic and natural
means has been found important for colonization of lizards in disturbed tropical forests
(Vitt et al. 1998). Moreover, Owens et al. (2008) determined that total reptile counts and
species richness were relatively unaffected by experimental CWD additions, suggesting
that some reptile species respond positively to the combination of both canopy gap
creation and increased structural diversity. Further examination of herpetofaunal response
to stochastic disturbances is necessary to provide forest managers with information to
guide management strategies that mimic natural disturbance patterns.
My trap design provides an effective way to sample many reptile species,
particularly larger snakes. I was unable to directly compare my trap results with those
based on other trap methods. However, straight-line drift fences used in a similar study in
northeastern Alabama captured 64 total snakes during 3030 trap nights (0.021 snakes per
trap night; Felix 2007). In addition, I captured considerably more snakes in less trap
nights than Burgdorf et al. (2005; 224 individual snakes during 13,920 trap nights; 0.016
snakes per trap night). I encourage researchers to compare the ability of my modified box
trap design to traditional straight-line drift-fences to determine the efficiency of my
design.
Bibliography
Barrett, K., and C. Guyer. 2008. Differential responses of amphibians and reptiles in
riparian and stream habitats to land use disturbances in western Georgia, USA.
Biological Conservation 141:2290–2300.
Burgdorf, S. J, D. C. Rudolph, R. N. Conner, D. Saenz, R. R. Schaefer. 2005. A
successful trap design for capturing large terrestrial snakes. Herpetological Review
36:421–424.
Corn, P. S. 1994. Straight-line drift fences and pitfall traps. In: W. R. Heyer, M. A.
Donnelly, R. W. McDiarmid, L. C. Hayek, M. S. Foster ed. Measuring and
Monitoring Biological Diversity. Standard Methods for Amphibians. Smithsonian
Institution Press: Washington and London, 109–117.
Cross C. L., and C. E. Peterson. 2001. Modeling snake microhabitat from radiotelemetry
studies using polytomous logistic regression. Journal of Herpetology 35:590–597.
Crosswhite D. L., S. F. Fox, R. E. Thill. 2004. Herpetological habitat relations in the
Ouachita mountains, Arkansas. In: Guldin, James M., tech. comp. 2004. Ouachita and
Ozark Mountains symposium: ecosystem management research. Gen. Tech. Rep.
SRS–74. Asheville, NC: U.S. Department of Agriculture, Forest Service, Southern
Research Station, 273–282.
Dodd, C. K., Jr, W. J. Barichivich, S. A. Johnson, and J. S. Staiger. 2007. Changes in a
northwestern Florida Gulf Coast herpetofaunal community over a 28-y period.
American Midland Naturalist 158:29–48.
Duncan R. S., and J. E. Linhoss. 2005. Regeneration of Virginia pine (Pinus virginiana)
following southern pine beetle (Dendroctonus frontalis) outbreak in the Sipsey
Wilderness, Alabama. Forest Ecology and Management 212:65–74.
Enge, K. M. 1997. A standardized protocol for drift-fence surveys. Florida Game and
Freshwater Commission. Technical Report No. 14:1–69.
Enge, K. M. 2001. The pitfalls of pitfall traps. Journal of Herpetology 35:467–478.
Felix, Z. I. 2007. Response of forest herpetofauna to various levels of overstory tree
retention in northern Alabama. PhD dissertation. Alabama A&M University. 165 pp.
Forest Inventory and Analysis, 1998. Field instructions for southern forest inventory.
Remeasurement of prism plots. Southern Research Station, Forest Service, U.S.
Department of Agriculture. Item 26 version of manual.
Gaines, G. D., and J. W. Creed. 2003. Forest health and restoration project. National
forests in Alabama, Bankhead National Forest Franklin, Lawrence, and Winston
Counties, Alabama. Final environmental impact statement. Management Bulletin R8 –
MB 110B.
Gibbons, 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 50:653–656.
Goldstein, M. E., R. N. Wilkins, T. E. Lacher, Jr. 2005. Spatiotemporal responses of
reptiles and amphibians to timber harvest treatments. Journal of Wildlife Management
69:525–539.
Greenberg, C. H., D. G. Neary, L. D. Harris. 1994a. Effect of high-intensity wildfire and
silvicultural treatments on reptile communities in sand-pine scrub. Conservation
Biology 8:1047–1057.
Greenberg, C. H., D. G. Neary, and L. D. Harris. 1994b. A comparison of herpetofaunal
Sampling effectiveness of pitfall, single-ended, and double-ended funnel traps used
with drift fences. Journal of Herpetology 29:319–324.
Greenberg, C. H. 2000. Response of reptile and amphibian communities to canopy gaps
created by wind disturbance in the southern Appalachians. Forest Ecology and
Management 148:135–144.
Greenberg, C. H., and T. A. Waldrop. 2008. Short-term response of reptiles and
amphibians to prescribed fire and mechanical fuel reduction in a southern Appalachian
upland hardwood forest. Forest Ecology and Management 255:2883–2893.
Herbeck, L. A., and D. R. Larsen. 1999. Plethodontid salamander response to silvicultural
practices in Missouri Ozark forests. Conservation Biology 13:623–632.
Hutchens, S. J., and C. S. DePerno. 2009a. Efficacy of sampling techniques for
determining species richness estimates of reptiles and amphibians. Wildlife Biology
15:113–122.
Hutchens, S. J., and C. S. DePerno. 2009b. Measuring species diversity to determine
land-use effects on reptile and amphibian assemblages. Amphibia–Reptilia 30:81–88.
James, S. E., and R. T. M’Closkey. 2003. Lizard microhabitat and fire fuel management.
Biological Conservation 114:293–297.
Knebel, L., and T. R. Intworth. 2007. Influence of fire and southern pine beetle on
pine-dominated forests in the Linville Gorge Wilderness, North Carolina. Castanea
72:214–225.
Land, A. D., and L. K. Rieske. 2006. Interactions among prescribed fire, herbivore
pressure and shortleaf pine (Pinus echinata) regeneration following southern pine
beetle (Dendroctonus frontalis) mortality. Forest Ecology and Management 235:260–
269.
Lindenmayer, D. B., C. R. Margules, and D. B. Botkin. 2000. Indicators of biodiversity
for ecologically sustainable forest management. Conservation Biology 14:941–950.
Mushinsky, H. R. 1985. Fire and the Florida sandhill herpetofaunal community: with
special attention to responses of Cnemidophorus sexlineatus. Herpetologica 41:333–
342.
Owens, A. K., K. R. Moseley, T. S. McCay, S. B. Castleberry, J. C. Kilgo, W. M. Ford.
2008. Amphibian and reptile community response to coarse woody debris
manipulations in upland loblolly pine (Pinus taeda) forests. Forest Ecology and
Management 256:2078–2083.
Palmer, M. W. 1993. Putting things in even better order: the advantages of canonical
correspondence analysis. Ecology 74:2215–2230.
Renken, R. B., W. K. Gram, D. K. Fantz, S. C. Richter, T. J. Miller, K. B. Ricke, B.
Russell, X. Wang. 2004. Effects of forest management on amphibians and reptiles in
Missouri Ozark forests. Conservation Biology 18:174–188.
Russell, K. R., T. B. Wigley, W. M. Baughman, H. G. Hanlin, and W. M. Ford. 2004.
Responses of southeastern amphibians and reptiles to forest management: A review.
Pages 319–334, In M. H. Rauscher and K Johnsen ed. Southern forest science: past,
present, and future. United States Department of Agriculture Southern Research
Station Gen. Tech. Rep. SRS–75.
Ryan, T. J., T. Philippi, Y. A. Lieden, M. E. Dorcas, T. B. Wigley, and J. W. Gibbons.
2002. Monitoring herpetofauna in a managed forest landscape: effects of habitat types
and census techniques. Forest Ecology and Management 167:83–90.
Schurbon, J. M., J. E. Fauth. 2003. Effects of prescribed burning on amphibian diversity
in a southeastern U.S. national forest. Conservation Biology 17:1338–1349.
ter Braak, C. F. J. 1995. Ordination. In R. H. G. Jongman, C. J. F. ter Braak, O. F. R. Van
Tongeren ed. Data Analysis in Community and Landscape Ecology. Cambridge:
Cambridge University Press 91–173.
Todd, B. D., and K. M. Andrews. 2007. Response of a Reptile Guild to Forest
Harvesting. Conservation Biology 22:753–761.
Vitt, L. J., T. C. S. Avila-Pires, J. P. Caldwell, and V. R. L. Oliveira. 1998. The impact of
individual tree harvesting on thermal environments of lizards in Amazonian rain
forests. Conservation Biology 12:654–664.
Figure 2.1 Overhead schematic of drift-fence trapping array. The central triangle
represents a large triangular box trap, while the rectangles represent box traps. The
straight lines spanning between the box traps and triangle traps represent the three 15 m
pieces of aluminum flashing (drift-fence). Pitfall traps were 19 L plastic utility buckets
buried flush with the ground surface. The small lines originating from the corners of the
box traps represent 1 m pieces of aluminum flashing used to direct the animals into the
traps.
Figure 2.2 Schematics for the construction of terminal box traps (A) and center triangle
box traps (B). The frame of each trap was constructed from pressure-treated lumber. I
used 23-gauge hardware cloth with 0.64 cm (¼”) mesh openings. I secured the main
portion of the frame (i.e., legs and main braces) using 5.1 cm (2”) galvanized screws and
smaller 2.5 cm (1”) zinc-plated screws to attach the tops, bottoms, and back portions of
the traps. The hardware cloth was attached to the trap body using 0.95 cm (3/8”) staples.
A
B
Figure 2.3 Schematic for conical trap funnels. Funnels can be formed by placing the two
pieces of the hardware cloth together. Once the funnel has been shaped, small cable ties
can be used to hold the funnel in place. If a large number of funnels are to be made, a
pattern will make constructing the funnels much easier. Figure adapted from Enge
(1997).
Percent Litter
Percent Bare Ground
Percent Herbaceous
Percent Woody
Percent Slash
Percent Rock
Percent CWD
Litter Depth
Canopy Cover
Forest Level 1
Forest Level 2
Forest Level 3
Forest Level 4
CWD volume
%_litt
%_bare
%_herb
%_woody
%_slash
%_rock
%_CWD
l_dep
can_cov
for_lev1
for_lev2
for_lev3
for_lev4
CWD_vol
Habitat Description
Presence of ground cover such as leaves or small woody debris
Absence of ground cover (e.g., exposed soil)
Any non-woody stem such as grasses, ferns, and Smilax and Vitus spp.
Any woody stems such as seedlings and large trees (living or dead); woody stems taller than one meter had to contact transects directly to be counted
Any woody debris composed of two or more stems 30 cm or higher from the ground (e.g., fallen treetops)
Presence of rocky substrate greater than 10 cm in size
Any fallen woody debris larger than 10 cm in diameter (must touch the ground somewhere along the length to be counted)
Determined by measuring depth of the substrate to the nearest 0.5 cm with a metric ruler
Estimated with a spherical densiometer as the sum percentage of open points subtracted from 100%
Percent coverage of forest levels ≤ 2m (classified as ground cover)
Percent coverage of forest levels > 2 m – ≤ 4 m (classified as understory)
Percent coverage of forest levels > 4 m – ≤ 6 m (classified as midstory)
Percent coverage of forest levels > 6 m (classified as overstory)
Calculated as volume of a cylinder for each enumerated CWD (see text)
Table 2.1 Habitat variables assessed at each survey point
point
Habitat Variable
Code
4
4
CAAM
COCO
CRHO
LAT R
NESI
PAGU
PASP
PLFA
PLLA
SCUN
SCLA
ST DE
ST OC
T ECA
T HSI
Northern Black Racer
T imber Rattlesnake
Southern Ring-necked Snake DIPU
LAGE
Worm Snake
Eastern Black Kingsnake
Red Milksnake
Northern Watersnake
Red Cornsnake
Gray Ratsnake
Common Five-lined Skink
Broad-headed Skink
Eastern Fence Lizard
Little Brown Skink
Midland Brownsnake
Northern Red-bellied Snake
Eastern Box T urtle
Eastern Gartersnake
Carphophis a. amoenus
Coluber c. constrictor
Crotalus horridus
Diadophis p. punctatus
Lam propeltis getula nigra
Lam propeltis triangulum syspila
Nerodia s. sipedon
Pantherophis guttatus
Pantherophis spiloides
Plestiodon fasciatus
Plestiodon laticeps
Sceloporus undulatus
Scincella lateralis
Storeria dekayi wrightorium
Storeria o. occipitomaculata
Terrapene c. carolina
Tham nophis s. sirtalis
T otals
4
ANCA
Green Anole
Anolis carolinensis
226
10
2
1
1
70
12
17
20
6
4
1
1
9
4
29
27
AGCO
Copperhead
Agkistrodon contortrix
100
4.4
0.9
0.4
0.4
31.0
5.3
7.5
8.8
2.7
1.8
0.4
0.4
1.8
1.8
1.8
4.0
1.8
12.8
11.9
NA
7
2
1
1
15
5
10
14
6
4
1
1
4
4
3
7
3
11
13
Number of Captures Percentage of Captures No. of Plots where a Species was Detected
Code
Common Name
Scientific Name
Table 2.2 Total reptile captures in the Bankhead National Forest, Alabama, U.S.A.
Table 2.3 Range of habitat variables (Table 2.1) derived from 18 forest stands of the
Bankhead National Forest, Alabama, U.S.A.
Habitat Code
%_litt
%_bare
%_herb
%_wood
%_rock
%_slash
%_cwd
can_cov
l_dep
for_1
for_2
for_3
for_4
cwd_vol
Range of Habitat Variables
98.3 – 100
0
6.3 – 49.6
3.3 – 31.7
0 – 0.8
0 – 2.5
0 – 4.2
72.8 – 90.1
4.0 – 7.6
25 – 100
54.2 – 95.8
35.7 – 95.8
75 – 100
0 – 491.1
Table 2.4 Principal components matrix derived from habitat variables (Table 2.1)
collected in 18 forest stands of the Bankhead National Forest, Alabama, U.S.A.
Variables
Component 1 Component 2 Component 3 Component 4 Component 5
% litt
-0.246
0.077
0.712
0.443
0.024
% herb
0.914
0.19
-0.056
0.109
-0.014
% woody
0.908
0.261
-0.049
0.241
-0.113
% rock
-0.289
0.597
-0.05
-0.037
0.694
% slash
0.773
0.328
-0.146
0.283
-0.153
% CWD
0.558
-0.248
0.317
-0.607
-0.039
can_cov
-0.812
-0.008
-0.366
-0.194
0.026
litt_dep
0.05
-0.904
-0.127
0.048
-0.055
for_1
0.661
0.11
-0.301
-0.025
0.356
for_2
-0.214
0.537
-0.447
-0.038
-0.536
for_3
-0.157
0.187
0.891
0.109
-0.117
for_4
0.257
-0.59
-0.233
0.571
0.195
cwd_vol
0.806
-0.089
0.254
-0.473
0.083
Eigenvalue
4.6
2.1
2
1.3
1
% Variance
35.4
16.5
15.3
10.1
7.7
Cumulative % Variance
35.4
51.9
67.2
77.3
85
Figure 2.4 Canonical correspondence analysis ordination plot displaying reptile species
and habitat relationships in the Bankhead National Forest, Alabama, U.S.A. Triangles
with four-lettered abbreviations represent species scientific names (Table 2.2) and
arrowed lines represent habitat variables (Table 2.1).
Figure 2.5 Canonical correspondence analysis ordination plot displaying sample plot and
habitat relationships in the Bankhead National Forest, Alabama, U.S.A. Numbered points
represent survey points and arrowed lines represent habitat variables (Table 2.1).
CHAPTER 3
AMPHIBIAN AND REPTILE RESPONSE TO THINNING AND PRESCRIBED
BURNING PRACTICES IN THE WILLIAM B. BANKHEAD NATIONAL FOREST,
ALABAMA, U.S.A.
Introduction
Habitat disturbances have the potential to affect species composition and can be
either beneficial or detrimental for a given species. Understanding the relationships
between disturbance regimes and wildlife responses is important for the conservation of
these species. Amphibians and reptiles, collectively known as herpetofauna, have high
diversity and often form a large portion of the vertebrate biomass in areas of eastern
North America (Burton and Likens 1975a, b; Petranka and Murray 2001). Many of these
species, especially in southeastern forests, occupy habitats that have a distinct disturbance
regime (Yager et al. 2007). Anthropogenic disturbances such as forest fragmentation and
conversion of historical forest types have altered disturbance regimes so severely that
historical disturbance events no longer occur with the same frequency (Turner et al.
1989). Forest disturbances such as burning and canopy removal are essential for the
maintenance of these processes in forest ecosystems and should be adapted to mimic the
effects of naturally occurring disturbance patterns (Greenberg 2000; Drever et al. 2006).
Forest management practices affect large forested areas and can vary greatly in
scale and disturbance intensity. Because these disturbances potentially affect large areas
of the landscape, there has been much controversy regarding the effects of forest
management on the flora and fauna inhabiting these areas. As herpetofauna play key roles
in forest ecosystems (Wyman 1998) along with evidence of worldwide herpetofaunal
declines (Gibbons et al. 2000; Stuart et al. 2004), there has been heightened interest in the
response of these organismal groups to forest management (Russell et al. 2004).
Numerous studies have evaluated amphibian response to clearcut harvesting and
have found that population parameters of adult amphibians (Enge and Marion 1986;
Grialou et al. 2000; Knapp et al. 2003; Karraker and Welsh 2006; Patrick et al. 2006;
Perkins et al. 2006; Homyack et al. 2009) and juvenile amphibians (Knapp et al. 2003;
Patrick et al. 2006) respond negatively to these treatments, whereas other studies have
found minimal impacts of clearcuts on adult amphibian population parameters (Chazal
and Niewiarowski 1998; Renken et al. 2004; Goldstein et al. 2005). Partial harvesting
procedures such as initial shelterwood cuts (Harpole and Haas 1999; Knapp et al. 2003)
and thinning operations (Naughton et al. 2000; Grialou et al. 2000) negatively affect
salamander populations, but appear to be species-specific. Uneven-aged management,
such as group and single tree selection either minimally affect amphibian populations
(Messere and Ducey 1998; Greenberg 2000; McKenny et al. 2006) or show speciesspecific negative impacts to some amphibian populations (Cromer et al. 2002;
MacCracken 2005). Far fewer studies have evaluated reptile response to forest
management practices in the eastern and southeastern United States compared to
amphibians. A majority of studies have found that lizard species respond positively to
tree harvesting operations (Adams et al. 1996; McLeod and Gates 1998; Renken et al.
2004; Goldstein et al. 2005; Greenberg and Waldrop 2008) or in a species-specific
manner (Greenberg et al. 1994).
Herpetofaunal responses to prescribed burning are largely unknown because most
analyses have been limited to small plots with unreplicated treatments, retrospective
study designs, and biases towards certain ecoregions (e.g., southeastern Coastal Plain;
Russell et al. 1999; Russell et al. 2004). Prescribed fire impacts on herpetofauna are
likely influenced by life stage (e.g., adult or larvae), environmental conditions, season of
burn, and whether a certain species has evolved under a fire regime (Russell et al. 1999;
Pilliod et al. 2003; Driscoll and Henderson 2008). Prescribed fire appears to have
negative short-term effects on amphibian species that inhabit areas that are not fire prone
ecosystems (Kirkland et al. 1996; Cole et al. 1997; Mcleod and Gates 1998) or areas
where the natural fire regime has been altered (Schurbon and Fauth 2003; Means et al.
2004). Other studies have found negligible impacts on amphibians (Ford et al. 1999;
Moseley et al. 2003; Greenberg and Waldrop 2008) and short-term positive effects for
certain species (Mushinsky 1985; Wilgers and Horne 2006). Reptile response to
prescribed burning appears to be either positive (Moseley et al. 2003; Wilgers and Horne
2006) or with negligible impacts (Mcleod and Gates 1998). In regions such as the
southeast, prescribed burning is essential to maintain habitat for reptiles native to longleaf
pine ecosystems (Russell 1999; Means et al. 2004, Yager et al. 2007); however our
knowledge of herpetofaunal responses to fire disturbances throughout other areas of the
United States remains unclear (Bury 2004).
Herpetofaunal studies coupled with forest disturbance provide useful information;
although the ability to draw predictable conclusions has been limited by factors such as
inadequate stand replication, lack of pre-treatment data, and short sampling periods (<2
years). Most studies of herpetofaunal response to forest management have taken a
retrospective approach and do not or are unable to sample during pre-harvest conditions
(Renken et al. 2004; Russell et al. 2004). Past studies also tend to relate organismal
abundance and population viability to forest composition, but do not examine
mechanisms responsible for causing observed population fluctuations (Figure 3.1 in text;
Marzluff et al. 2000). In addition, studies of herpetofaunal response to ecological
disturbance in the southeastern Appalachians have focused on salamander populations,
whereas reptile species have largely been ignored (Greenberg 2000).
Figure 3.1 Proposed mechanisms linking overall forest condition to herpetofaunal
viability. Bottom curved line represents the route of conventional studies failing to
identify causal mechanisms. Adapted from Marzluff et al. 2000.
Overall
Forest
Condition
Canopy coverage
Litter composition
Vegetation structure
Coarse woody debris
Forest Stratification
Air and soil temperature
Relative humidity
Light intensity
Population
Viability
I sought to evaluate herpetofaunal response to forest management practices
(thinning and prescribed burning) as part of a larger study evaluating ecosystem response
to large-scale forest restoration treatments. I took a large-scale, replicated, stand-level
approach to evaluate disturbance response of herpetofauna inhabiting these ecosystems. I
assumed that reptile population parameters (e.g., counts and species richness) would
increase after treatment and would be highest in thin-only plots, whereas amphibian
population parameters would decline most precipitously in thin*burn plots after treatment
implementation. I hypothesized that reptile population parameters would be correlated
with increased thermoregulation sites and structural diversity, whereas amphibian
population parameters would decline in highly disturbed plots due to cumulative
disturbance interactions (i.e., simultaneous reduction of litter and reduction of canopy
cover). Because I was able to identify changes in population parameters for many
species, I examined correlations between herpetofaunal population changes and
environmental characteristics to evaluate potential mechanisms responsible for
structuring upland pine-hardwood herpetofaunal communities. I assumed that the
measured habitat and climate measurements would be responsible for causing the
observed changes in amphibian and reptile population parameters.
Materials and Methods
Study Site Description
My study was centered in the northern portion of the William B. Bankhead
National Forest (BNF), located in Lawrence, Winston, and Franklin Counties, of
northwestern Alabama. Bankhead National Forest is a 72,800 ha multi-use forest located
along the highly dissected portion of the southern Cumberland Plateau (Smalley 1982;
Gaines and Creed 2003). Soils within this region are typically composed of HartsellsRock and limestone-Hector (Smalley 1982). Mixed forests of the southern Cumberland
Plateau tend to be dominated by oak-hickory forest types (McWilliams 1991) except in
areas where pines were actively planted for commercial purposes. Loblolly pine (Pinus
taeda L.) was used to re-establish forest conditions in abandoned agricultural and heavily
timbered areas (Gaines and Creed 2003). Reforestation efforts along with natural growth
have resulted in 31,600 ha of Loblolly Pine throughout BNF (Gaines and Creed 2003).
For the past decade, Southern Pine Beetle (Dendroctonus frontalis Zimmermann)
infestations have affected Loblolly Pine stands, producing large numbers of standing
dead trees and increased fuel loads, elevating the risk of damaging wildfires. Because
canopy removal and fire disturbance have been prevented in forests throughout the study
area for decades, the BNF initiated a Forest Restoration Plan to reduce wildfire risk and
promote natural forest growth through tree thinning and prescribed fire disturbance. The
BNF has not traditionally utilized prescribed fire as a management tool, but has opted to
include prescribed burning in the forest restoration plan due to administrative
recommendations. Forest restoration plans in BNF mirror regulations set forth in the
Healthy Forest Restoration Act, which authorizes advanced vegetation management
projects when specified conditions (e.g., existence of insect or disease epidemic) pose a
significant threat to ecosystem health (Healthy Forest Restoration Act 2003).
The selected forest stands were generally located on upland sites composed of
loblolly pine 25-50 years of age that also possessed a considerable hardwood component
(Gaines and Creed 2003; Schweitzer and Tadessee 2004). At the time of this study, these
stands had not been recently harvested, and each stand had varying levels of damage
from the Southern Pine Beetle. Other past disturbances included the clearing of hardwood
stands throughout the region for loblolly pine plantations during the early 1970’s (Gaines
and Creed 2003).
Experimental Design and Forest Treatments
The Bankhead National Forest implemented this research, which was designed
and developed by Alabama A&M University and the Southern Research Station of the
USDA Forest Service as part of a longterm examination of forest management impacts
on forested ecosystems in northwestern Alabama. The experiment consisted of a beforeafter, control-impact (BACI), complete block design. The process of assigning treatments
to each forest stand was not a fully random process because forest treatment designations
had to align with the longterm management goals of the BNF. For example, forest stands
assigned to prescribed burn treatments had to be located in a portion of the BNF that was
designated a burn area in the original forest restoration management plan. Forest
manipulation treatments consisted of a two by three factorial arrangement of three
thinning levels (no thin, 11 m2ha-1 residual basal area [BA], and 17 m2ha-1 residual BA)
along with two burn treatments (no burn and burn) equaling six treatments per replicate.
In addition to single thinning and burning treatments, I also evaluated the interaction of
each thinning level with a burn treatment. Each of the six treatments in this experiment
was replicated three times across the landscape (Figure 3.2), with each treatment
approximately 9 ha in size.
Due to limited resources and difficulties implementing this large scale study in a
single year, treatments were blocked temporally (i.e., year). Block 1 was treated during
the summer of 2005, whereas block 2 and block 3 were treated during the summer and
fall of 2006. All harvesting procedures were thin-from-below and were completed at two
levels of BA retention (Figure 3.2). Thinning-from-below is a harvesting procedure that
targets trees from the lower crown classes (i.e., suppressed and intermediate crown
classes) in order to provide limited resources (e.g., water and light) to dominant and codominant trees (Smith et al. 1997). All harvesting was completed by feller bunchers and
trees were harvested until the desired residual BA had been achieved. Hardwood tree
species, such as Quercus spp. and Carya spp. were preferentially retained during the
thinning as much as possible. All thinning procedures in a particular block were
completed in the same year (i.e., block one 2005 and blocks two and three 2006).
Prescribed burns consisted of low-intensity fires, which were completed during
the dormant season (January–February) when air temperatures were low and relative
humidity was high. Backing fires were initiated to ensure that prescribed burns were
limited to understory and litter layers. Prescribed burns were ignited with drip torches and
managed by the BNF U.S. Forest Service. Fires were generally low intensity and rarely
reached more than two meters in height. Once the designated area was blocked off and
ignited, fires were permitted to burn until all available fuel was consumed and the fires
diminished.
Amphibian and Reptile Sampling
I collected herpetofaunal capture data over a period of four years (2005–2008).
Due to the staggered nature of the treatments, this resulted in three total years of capture
data (one year pre-treatment; two years post-treatment). I was able to collect pretreatment data over a period of three months (April-June 2005) for block one and six
months (May 2005-August 2005 and March 2006-May 2006) for blocks two and three.
Although I sampled less during pre-treatment surveys compared to post-treatment
surveys, I constrained all analyses to seasons that were directly comparable between preand post-treatment data.
I employed a trapping method consisting of three drift fences (aluminum flashing)
15 m in length radiating 120⁰ from a central triangular box trap (Sutton et al. In press).
Trapping units also included one large box trap at the terminus of each drift fence (three
per array) and two pitfall traps at the midpoint of each drift fence (six per array). I chose
this design because large box traps have been proven successful for capturing and
sampling medium-large snake species (Burgdorf et al. 2005). I installed one drift-fence
array in each study plot by dividing each study plot to quadrants corresponding to the
four cardinal directions and randomly assigned the drift-fence array to one of these
quadrants.
After the completion of pre-treatment surveys, I removed all drift-fence arrays to
avoid damage from tree harvesting and prescribed burning procedures. To locate trap
locations after treatments, I sunk fluorescent stake whiskers (Forestry Suppliers, Jackson,
Mississippi) into the ground with large steel nails to mark the location of each box trap.
Once all forest treatments were completed, I re-installed traps in the same location where
pre-treatment surveys were completed.
I began sampling intermittently throughout March and April and began
continuous sampling by the beginning of May and ended sampling by September during
each year. During sampling periods, I opened traps by block(s) depending on weather
conditions and manpower, with the replication number and order of traps randomly
determined a priori. I checked traps daily between 0700-1400 hours (CST) to minimize
animal mortality. After recording demographic data (e.g., snout-vent length and mass), I
marked each individual with a plot-specific mark through toe-clips (lizards, anurans, and
salamanders), scale clips (snakes), and scute etching (turtles) to ensure that recaptured
individuals would not be counted in subsequent captures (Enge, 1997). I released all
marked individuals at a minimum of 10 m on the side of the drift fence in which they
were captured.
Environmental Parameters
One HOBO© (Onset Computer Corp.) datalogger was installed at each trapping
array to record air temperature, soil temperature, relative humidity, and light intensity
(Table 3.1). Dataloggers were programmed to record measurements every twelve hours
starting at 2:00 PM. Due to limitations during pre-treatment surveys, I collected climate
data May 15-July 15 during all survey years. I also installed rain gauges (Taylor
Precision Products, Oak Brook, Illinois, U.S.A.) to monitor precipitation events only
during trapping events.
Habitat Parameters
I recorded pre- and post-treatment habitat complexity and heterogeneity data via
three yearly line-transect surveys at each treatment plot. I determined plot placement a
priori via a random compass bearing (0-360°) and distance (30-50 m) originating from
the center of each trapping array. I restricted habitat surveys to these distances in order to
avoid any habitat disturbance created during trap installation. To quantify the degree of
habitat disturbance, I completed habitat surveys in the same location each year. Each
habitat survey consisted of two 20 m perpendicular transects placed north-south and east-
west from the habitat plot center. I used a two meter piece of 1.9 cm diameter polyvinyl
chloride pipe as a transect marker and recorded the presence or absence of microhabitat
variables every 0.5 m as indicated in Table 3.1. I also measured CWD volume (m3), litter
depth (%), and percent forest cover and determined vertical forest structure values as
indicated in Table 3.1. I assigned vertical forest structure values on a scale of 1-4 by
estimating the height of the dominant forest structure every 5 m and assigned values
according to the designations described in Forest Inventory and Analysis (1998; Table
3.1).
Data Analysis
To explore forest management impacts on amphibian and reptile species diversity
at multiple scales, I compared herpetofaunal alpha diversity, beta diversity, and gamma
diversity patterns. Alpha diversity represents diversity within individual sample units
(i.e., diversity at the stand level), whereas gamma diversity represents the diversity in a
collection of sample units (i.e., landscape level diversity; McCune and Grace 2002). Beta
diversity represents the amount of composition variation among sample units and is
calculated as β = γ / α, where β is the beta diversity, γ is the landscape level diversity (i.e.,
gamma diversity), and α is the average diversity in a sample unit (i.e., alpha diversity;
McCune and Grace 2002). I determined alpha and gamma level species diversity by
constructing rarefaction curves using EstimateS v. 8.2.0 (Colwell 2006). To do this, I
reconstructed the yearly sampling history for all herpetofaunal captures and amphibian
and reptile captures separately, and constructed three sets of rarefaction curves (i.e., one
pre-treatment and two post-treatment). For all species richness calculations, I chose the
Chao 2 estimator, because this method results in species accumulation curves that reach
near-maximum values with very few samples (Barrett and Guyer 2008). The Chao 2
estimator calculates species richness by adjusting for species captured only once or twice,
which is advantageous when errors associated with lack of detection are likely (Barrett
and Guyer 2008). The Chao 2 estimator is calculated as SChao2 = Sobs + Q12 / 2Q2, where
Sobs equals the total number of species detected in a given area and Q1 and Q2 equals the
number of species detected one and two times, respectively (Colwell and Coddington
1994). I also calculated species heterogeneity via the Shannon-Wiener diversity index
using EstimateS v. 8.2.0 (Colwell 2006) and Morisita’s similarity index using Ecological
Methodology v. 6.1.1 (Krebs 2003). The Shannon-Wiener index takes into account
species richness and evenness and greater values indicate greater overall diversity,
whereas Morisita’s index calculates overall species similarity between samples and
assigns a similarity value ranging from 0 (no similarity) to 1 (complete similarity; Krebs
1999).
The primary focus of this research was to explore the response of amphibians and
reptiles to forest management practices. I used mixed models (PROC MIXED) analysis
of variance (ANOVA; SAS v. 9.3) to test changes in herpetofaunal counts and
biodiversity measures (e.g., species richness, heterogeneity, and similarity) between preand post-treatment surveys among the treatments. Mixed models permit the analysis of
random effects (i.e., block) along with fixed effects (i.e., treatment), while controlling for
repeated samples (i.e., year). For individual species and species group comparisons, I
divided the total number of individuals by the total number of trap nights (one trap night
= one trap opened for 24 hours) to correct for differences in trapping effort among years.
I then multiplied the trap-night corrected count by 1000 to estimate the number of
animals captured per 1000 trap nights. Prior to analysis, herpetofauanal count and
biodiversity measures were normalized with square root and logistic transformations.
To explore relationships among amphibian and reptile community and
microhabitat and microclimate variables, I used canonical correspondence analysis
(CCA), which is a direct gradient analysis technique where the ordination procedure is
constrained by a set of a priori covariates (e.g., habitat and climate data) that are
predicted to influence the observed distribution of the organismal groups in question (ter
Braak 1995). To control for rare species effects on the ordination output, I only included
species with at least four captures (ter Braak 1995; Crosswhite et al. 2004). To select
habitat variables for the analysis, I examined relationships among habitat variables with a
correlation matrix. In cases where variables were correlated ≥ 0.80, I retained the variable
with greatest biological relevance. This process excluded soil temperature, light intensity,
relative humidity, and percent bare ground from further analyses. I constructed CCA
plots for each treatment year and compared the changes in species and habitat
relationships. Because amphibians and reptiles have different life history characteristics, I
chose to examine each group separately.
I used principal components analysis (PCA) to examine relationships among
habitat and climate parameters. Upon confirming the biological relevance of the
generated components, I examined overall changes of the habitat and climate components
between pre- and post-treatment surveys using mixed models ANOVA (PROC MIXED;
SAS v. 9.3). For all null-hypothesis testing techniques, I declared statistical significance
at an alpha level ≤ 0.05.
Results
Overall Captures
I captured 2,662 amphibians and reptiles representing 47 species during 2,862
trap nights (i.e., block 1 [672 total trap nights], block 2 [1134 total trap nights], and block
3 [1056 total trap nights]) over a four-year survey period (2005-2008). The most
commonly captured lizard and snake species were Green Anoles (Anolis carolinensis; n =
283) and Copperheads (Agistrodon contortrix; n = 178), respectively, whereas
Mississippi Slimy Salamanders (Plethodon mississippi; n = 674) and Fowler’s Toads
(Anaxyrus fowleri; n = 177) represented the most commonly captured salamander and
anuran species, respectively (Tables 3.2, 3.3, 3.4, and 3.5). Eastern Box Turtles
(Terrapene c. carolina) were the most commonly captured turtle species (n = 8). A total
of 371 individuals were recaptures, with Green Anoles being the most commonly
recaptured reptile species (146 recaptures) and the Fowler’s Toad being the most
commonly recaptured amphibians species (65 recaptures; Tables 3.2, 3.3, 3.4, and 3.5).
Species Diversity
Overall herpetofaunal and reptile alpha diversity tended to be the greatest two
years after treatment in light thin treatments and were generally the lowest in control
plots (Table 3.6). There was a significant effect of year on alpha diversity for all
herpetofauna (F2, 34 = 4.68; p = 0.016) and reptiles only (F2, 34 = 14.10; p < 0.0001; Table
3.6). Overall alpha diversity of amphibians was greatest in heavy thin plots, but I only
detected a statistically significant effect of thin*burn (F2, 34 = 3.62; p = 0.038) on
amphibian alpha diversity (Table 3.6). Gamma diversity for all herpetofauna two years
after treatment increased the greatest from pre-treatment levels in burn, light thin, and
light thin*burn plots, whereas gamma diversity for reptiles increased the greatest in burn,
light thin, and heavy thin*burn plots (Table 3.6). Gamma diversity of amphibians
increased the greatest from pre-treatment levels in light thin plots (Table 3.6). Overall
beta diversity values were similar among all treatment plots, with little change among
years (Table 3.6). Species rarefaction plots indicate that species accumulation rates were
similar for all herpetofauna (Figure 3.3), amphibians only (Figure 3.4), and reptiles only
(Figure 3.5) between all treatment years. However, more reptiles were captured in plots
that received a thinning treatment when compared to control and burn plots during the
second year post-treatment (Figure 3.5).
Overall herpetofaunal similarity generally decreased after treatment in most
treatment plots (Table 3.6), with significant year (F2, 36 = 12.87; p < 0.0001), thin*year
(F2, 36 = 5.64; p = 0.0013), and burn*thin*year (F4, 36 = 6.46; p = 0.0005) effects. Reptile
similarity indices generally increased in thinned plots after harvest (F2, 34 = 14.47; p =
0.0001), with a significant thin*year effect (F4, 34 = 3.34; p = 0.021; Table 3.6), whereas
reptile similarity indices were also high in thin*burn plots after treatment implementation
with a marginally significant effect of thin*burn*year (F4, 34 = 2.57; p = 0.056; Table
3.6). I detected an overall year effect (F2, 36 = 6.12; p = 0.005) and a marginally
significant burn*thin*year effect (F4, 36; p = 0.056) on amphibian similarity (Table 3.6).
Although herpetofaunal heterogeneity increased greatest in light thin & burn
plots, and reptile heterogeneity increased the greatest in burn and light thin plots, I was
unable to detect a treatment effect (Table 3.6). Amphibian heterogeneity was similar
among all treatment plots and treatment years, with no detectable treatment effect (Table
3.6).
Species Responses
Eastern Fence Lizard (Sceloporus undulatus) counts were greatest in thin*burn
plots two years after treatment, with an overall thin effect (F2, 36 = 9.02; p = 0.0007) and
an additional year*burn effect (F2, 36 = 4.23; p = 0.02; Figure 3.6). Green Anole counts
were greatest in thinned plots throughout the entire study with an overall thin effect (F =
16.88; p < 0.0001; Figure 3.7). Little Brown Skink counts were much higher across all
treatments during pre-treatment surveys when compared to first year post-treatment
surveys, with a marginal year effect (F2, 36 = 3.23; p = 0.051; Figure 3.8). Total
heliothermic lizard counts were greatest two years after treatment in all thinned plots and
lowest in burned stands, indicating an overall positive thin effect (F2, 34 = 20.85; p <
0.0001; Figure 3.9).
Northern Black Racer (Coluber c. constrictor) counts were greatest in thin plots
and thin & burn plots two years after treatment indicating an overall thin*burn effect (F2,
36
= 4.61; p = 0.017), with an additional thin*year interaction (F4, 36 = 2.67; p = 0.047;
Figure 3.10). Copperhead counts varied greatly among all years with no detectable
treatment effects (Figure 3.11). Total large snake counts tended to be greatest in thin-only
plots, with only a marginally significant burn effect (F1, 36 = 3.80; p = 0.059; Figure 3.12).
Mississippi Slimy Salamander counts simultaneously declined across all
treatments through all years indicating a significant year effect (F2, 36 = 4.72; p = 0.015),
with an additionally marginal negative thin*burn effect (F2, 36 = 3.16; p = 0.054; Figure
3.13). There was no detectable treatment effect on Fowler’s Toad (Figure 3.14) and true
frog (genus Lithobates) counts (Figure 3.15). However, true frog counts increased
considerably and were highest during the second year post-treatment in control plots
(Figure 3.15).
Habitat Response
Throughout three field seasons, I completed 162 total habitat surveys (three at
each plot for each year). Using PCA I was able to extract five components that accounted
for 80.6% of the overall variance (Table 3.7) in the original habitat dataset (Table 3.8).
Component one accounted for 46.8% of the overall variance and described a gradient
ranging from sites with greater canopy cover, lower temperatures, greater relative
humidity, and greater percent coverage of litter to sites with less canopy cover, greater
light intensity, higher temperatures, and greater percent bare ground coverage, whereas
component two accounted for 13.7% of the overall variance and described a gradient
ranging from sites with greater percent bare ground cover to sites with greater percent
coverage of litter, woody, and herbaceous growth (Table 3.7). Component three
accounted for 8.5% of the overall variance and described CWD coverage and volume,
whereas components four and five described 11.5% of the overall variance associated
with rock and overstory percent coverage, respectively (Table 3.7).
Habitat gradient values for principal component one were greatest in thin only and
thin*burn plots, illustrating an overall significant effect of thin (F2, 34 = 174.78; p <
0.0001) and burn (F1, 34 = 14.98; p = 0.0005), with additional effects of thin*year (F4, 34 =
32.16; p < 0.0001) and burn*year (F2, 34 = 10.71; p = 0.0002; Figure 3.16). Habitat
gradient values for principal component decreased in all treated stands during posttreatment year one and stayed low in burn only treatments during post-treatment year
two, indicating overall effects of both thin (F2, 34 = 5.93; p = 0.0062) and burn (F1, 34 =
9.42; p = 0.0042; Figure 3.17).
Species and Habitat Relationships
Canonical correspondence analysis indicated changes in species and habitat
relationships between pre-treatment and post-treatment years for amphibians (Figures
3.18, 3.19, and 3.20) and reptiles (Figures 3.21, 3.22, and 3.23). I found distinct increases
in variance explained for species relationships of amphibians during both post-treatment
years and for reptiles during the first year post-treatment compared to pre-treatment
values (Table 3.9).
Amphibian CCA diagrams for pre-treatment (Figure 3.18) and post-treatment year
one (Figure 3.19) did not reveal any obvious habitat gradients or species associations.
However, the CCA diagram for the second year post-treatment amphibian data (Figure
3.20) reveals a habitat gradient ranging from sites with greater canopy cover and greater
litter depth to sites with greater air temperature and greater percent coverage of slash and
herbaceous cover. Anuran species such as Cope’s Grey Treefrogs (Hyla chrysoscelis) and
Green Treefrogs (Hyla cinerea) were associated with greater CWD and herbaceous cover
on highly disturbed sites, whereas the Red Salamanders (Pseudotriton r. ruber) and
permanent pool-breeding frog species (e.g., Southern Leopard Frogs [Lithobates
sphenocephalus] and Northern Green Frogs [Lithobates clamitans melanota]) were
associated with greater canopy cover and litter depth (Figure 3.20). There was a distinct
separation of sites that received a thin treatment versus burn and control sites along the
canopy cover, air temperature, groundcover, and tree canopy gradient (Figure 3.20).
The CCA diagrams for reptiles illustrate that habitat gradients ranged from sites
with greater canopy cover and litter depth to sites with greater slash, CWD, and
herbaceous cover (Figures 3.21, 3.22, and 3.23). Air temperature and percent litter
coverage played a larger role in the observed habitat gradients during the post-treatment
analyses (Figures 3.22 and 3.23). Habitat gradients were longer and more well-defined
during post-treatment year one (Figure 3.22) and appeared to lose definition during posttreatment year two (Figure 3.23). Eastern Five-lined Skinks (Plestiodon fasciatus),
Broad-headed Skinks (P. laticeps), Green Anoles, and Eastern Fence Lizards were
associated with disturbed sites that possessed greater CWD, slash, and herbaceous cover
and warmer air temperatures in post-treatment CCA diagrams (Figures 3.22 and 3.23).
These species changed locations greatly within the post-treatment ordination plots when
compared to the pre-treatment analysis (Figure 3.21). Eastern Worm Snakes (Carphophis
amoenus) and Little Brown Skinks were consistently associated with sites possessing
greater percent canopy cover and greater litter depth during pre-treatment and posttreatment analyses (Figures 3.21, 3.22, and 3.23). There was a greater separation of sites
that received a thin treatment from control and burn sites in first year post-treatment CCA
diagram (Figure 3.22) when compared to the second year post-treatment CCA diagram
(Figure 3.23) along the canopy cover, air temperature, and groundcover gradient.
Discussion
When evaluating organismal response to disturbance, it is important to examine
the response at multiple levels. For this study, I evaluated the effects of forest
management practices on herpetofaunal species diversity measures, individual species
responses, and collective species and habitat responses. Overall, I did not find a
significant effect of forest treatment effect on herpetofaunal species diversity measures;
however, I did find that overall reptile species similarity was greatest in plots receiving a
thinning treatment. A majority of studies evaluating herpetofaunal response to forest
management have not found a significant effect of forest management practices on
species richness (Greenberg et al. 1994; Provencher et al. 2003; Wilgers and Horne 2006;
Greenberg and Waldrop 2008) and species diversity measures (Wilgers and Horne 2006).
However, studies have found associations between herpetofaunal species richness and
forest stand conditions (e.g., stand age and forest type; Adams et al. 1996; Ross et al.
2000; Loehle et al. 2005). Species richness measures are difficult to estimate correctly
(see Gotelli and Colwell 2001), but can be used to examine changes in vertebrate
composition throughout time (Dodd et al. 2007) and also to determine overall species
responses to landscape disturbances (Barrett and Guyer 2008). It is likely that the
relatively close proximity of the study plots to one another along with the upland site
locations explains the overall lack of differences in species richness among them.
I hypothesized that amphibians would respond negatively to forest management
practices, whereas reptiles would respond positively. My results suggested that
amphibian were not negatively impacted by thinning and prescribed burning. The overall
declines of this species throughout all treatment plots made it difficult to evaluate the
response of this species to forest management. Evaluations of salamander response to
thinning and burning interactions are few (but see Greenberg and Waldrop 2008), but it
has been documented that drastic habitat change negatively affected woodland
salamander populations (Grialou et al. 2000; Knapp et al. 2003; Karraker and Welsh
2006; Homyack et al. 2009), whereas partial harvesting procedures either negatively
affected (Knapp et al. 2003; Grialou et al. 2000) or had no impacts to woodland
salamander populations (Brooks 1999; McKenny et al. 2006; Greenberg and Waldrop
2008). I believe that climate variation among the years might have affected capture rates
for some amphibian species. Total anuran captures were much higher during the second
year post-treatment (2008) when compared to first year post-treatment captures (2007).
The 2007 field season was extremely dry and rain events were very scarce. I observed
that captures of pool-breeding amphibian species (e.g., Spotted Salamanders and
Southern Leopard Frogs) were greater in treatment plots within close proximity to
suitable amphibian breeding habitat. However, I did see a marked increase of true frogs
(genus Lithobates) in control plots during second year post-treatment surveys, suggesting
a potential association of these species with closed canopy plots.
My results indicated that reptiles generally responded positively to thinning
treatments. Lizards showed immediate responses to forest management practices with
Green Anoles, Eastern Fence Lizards, and all heliothermic lizards responding positively
to forest thinning. Although Little Ground Skinks declined in all treatment plots, it
declined most precipitously in thinned plots during the first year post-treatment,
suggesting a negative impact of forest management on this species. A majority of studies
have found that lizard species either responded positively to tree harvesting operations
(Adams et al. 1996; McLeod and Gates 1998; Renken et al. 2004; Goldstein et al. 2005;
Greenberg and Waldrop 2008) or in a species-specific manner (Greenberg et al. 1994).
Moreover, intense landscape disturbances appear to influence lizard populations. Fires
with higher intensities and frequent burning regimes appear to benefit certain lizard
species including six-lined racerunners (Mushinsky 1985; Ruthven et al. 2008) and fence
lizards (Greenberg 1994; Greenberg and Waldrop 2008). This would likely explain the
greater abundance of eastern fence lizards in thin*burn treatment plots.
In this study, large snakes generally responded positively to thinning operations.
Specifically, Eastern Black Racers increased in thinned plots, whereas total large snakes
had highest counts in thin-only plots during the second year post-treatment. Little
research exists regarding snake response to timber harvesting; however, clearcutting has
been shown to negatively impact litter-dwelling snakes (Todd and Andrews 2007). The
response of medium-large snake species to forest management has not been adequately
evaluated due to difficulties associated with sampling these species.
Changes in the physical environment (e.g., structure and thermal characteristics)
were responsible for habitat associations of several amphibian and reptile species. Cope’s
Grey Treefrogs and Green Treefrogs were associated with highly disturbed sites
possessing lower canopy cover and increased CWD cover, whereas Green Anoles,
Eastern Five-lined Skinks, Broad-headed Skinks, and Eastern Fence Lizards were
associated with increased ambient temperatures, and increased herbaceous and CWD
cover. Little brown skinks had a close association with increased litter depth and canopy
cover. Green Treefrogs and Eastern Hognose Snakes (Heterodon platirhinos) were only
detected during post-treatment surveys. Green Treefrogs have been found to colonize
sites recently disturbed by forest management practices (Cromer et al. 2002; Horn et al.
2005). All but one Eastern Hognose Snake capture (n = 7) occurred in plots that were
disturbed by forest management.
My study employed a large-scale, multi-year approach to evaluate amphibian and
reptile response to forest disturbances. My study has obvious advantages including large
plot size (~9 ha), pre-treatment data, and a factorial treatment arrangement with three
complete replications. Previous studies of herpetofaunal response to forest management
practices have taken a retrospective approach and have not compared post-treatment
results to baseline data. Russell et al. (2004) estimates that nearly 94% of all forest
management studies evaluating herpetofaunal response did not include pre-treatment
data. Without pre-treatment data it is impossible to know pre-existing organismal
distributions and therefore difficult to extrapolate post-disturbance findings. I was
generally unable to detect treatment differences until the second survey season,
suggesting that multiple survey seasons may be necessary to detect disturbance effects on
amphibians and reptiles. Previous forest management studies have examined prescribed
burning and thinning separately. My experimental design permitted us to tease apart the
effects of thinning, burning, and their associated interactions on amphibian and reptiles.
Due to anthropogenic alteration of natural disturbance regimes, prescribed
burning is currently used to simulate natural disturbance regimes in many ecosystems.
Prescribed burning is essential in habitats such as coastal longleaf pine forests (Means et
al. 2004) and xeric scrublands (Ruthven et al. 2008). In recent years, prescribed burning
has been used frequently in geographic provinces that do not have a history of frequent
fire disturbance (Ford et al. 1999; Greenberg and Waldrop 2008). If historical fire is not a
common component of an ecosystem’s disturbance regime, the organisms that inhabit
these landscapes likely do not possess adaptations to benefit from disturbances caused by
fire. Many aspects of prescribed fire (e.g., intensity, burning season, and fire return
interval) vary greatly and make it difficult to determine the longterm impacts on
organisms that inhabit these habitats. In addition, a majority of the existing research has
been limited by small treatment size, unreplicated experimental designs, and no pretreatment data (Russell et al. 2004).
I examined herpetofaunal response to low-temperature, dormant-season (i.e.,
February and March) prescribed burns. Therefore, these fires would be more likely to
negatively impact species, mainly amphibians (e.g., Plethodon salamanders and early
breeding frog species), that are active during the colder periods of the year. I did not find
any impacts (positive or negative) of prescribed burning on herpetofauna and I attribute
this to season and intensity of the burn. The prescribed burns examined during this study
were implemented during very cool months and were controlled to primarily affect litter
and forest understory layers. These fires burned at low temperatures and had little impact
on overall forest structure. Moreover, the burn treatment in block two was completed on a
cold and wet day resulting in a very sparse burn. I believe this may have confounded my
data, resulting in an inaccurate representation of herpetofaunal response to burning
practices.
Many of the forest stands examined during this study had a history of damage
through Southern Pine Beetle infestations. This insect pest is a normal inhabitant of
southeastern forests where it along with natural fire, play an important role in maintaining
equilibrium in pine-dominated forests (Knebel and Wentworth 2007; Land and Rieske
2006). In the BNF, fire is not a natural disturbance and historically forests were
dominated by hardwood tree species. The original hardwood stands were initially cleared
for agricultural purposes in the early 1900’s and a majority of these areas were
subsequently reforested with Loblolly Pine (Gaines and Creed 2003). Although Loblolly
Pine is a natural component of southern forests, it is not natural for this tree species to
occur in the high densities found within the reforested stands of BNF, which has resulted
in severe overcrowding. As a result, these Loblolly Pine stands were greatly stressed,
making them susceptible to attacks from Southern Pine Beetle infestations. Large-scale
Southern Pine Beetle infestations were common throughout areas of the Southern
Cumberland Plateau from 1999 to 2002, indicating that environmental conditions
throughout the region were highly supportive for a large-scale infestation (Duncan and
Linhoss 2005; Land and Rieske 2006).
Many of the forest stands examined during this study had a history of damage
through Southern Pine Beetle infestations. This insect pest is a normal inhabitant of
southeastern forests where it along with natural fire, play an important role in maintaining
equilibrium in pine-dominated forests (Knebel and Wentworth, 2007; Land and Rieske,
2006). Environmental stressors such as prolonged drought and overstocking of Pinus
species tends to increase the severity of Southern Pine Beetle infestations (Gaines and
Creed 2003; Duncan and Linhoss 2005). Although Loblolly Pine is a natural component
of southern forests, it was planted in unnaturally high densities in the BNF, leading to the
spread of the Southern Pine Beetle throughout the forest. Damage from Southern Pine
Beetle infestations usually leads to an increase of fallen logs and large canopy gaps
(Duncan and Linhoss 2005). My data suggests that these disturbances were likely
important for influencing the observed reptile community by creating large changes in the
overall forest structure. Although I have shown that thinning does cause overall changes
in the reptile community, I also observed that a broader diversity of reptiles inhabited
forest stands directly affected by Southern Pine Beetle disturbances during pre-treatment
surveys (Sutton et al. In Press). Control plots were similarly affected by Southern Pine
Beetle infestations and the high reptile species richness values observed in these plots
were likely due to these disturbances.
The forest disturbances evaluated during this study were implemented with the
goal to convert the Loblolly Pine dominated forests to stands dominated by hardwood
tree species. Although counts for certain reptile species increased after the
implementation of forest treatments, I am confident that the habitat heterogeneity created
through southern Pine Beetle Disturbances played a major role in determining the
observed herpetofaunal communities. It is important to prevent the spread of Southern
Pine Beetle infestations through active management strategies to avoid losing valuable
timber resources, but more research needs to be completed to understand the importance
of Southern Pine Beetle disturbances on overall herpetofaunal species diversity.
Bibliography
Adams, J. P., M. J. Lacki, and M. D. Baker. 1996. Response of herpetofauna to
silvicultural prescriptions in the Daniel Boone National Forest, Kentucky. Proc. Annu.
Conf. SEAFWA 50:312–320.
Barrett, K., and C. Guyer. 2008. Differential responses of amphibians and reptiles in
riparian and stream habitats to land use disturbances in western Georgia, USA.
Biological Conservation 141:2290–2300.
Burgdorf, S. J., D. C. Rudolph, R. N. Conner, D. Saenz, and R. R. Schaefer. 2005. A
successful trap design for capturing large terrestrial snakes. Herpetological Review
36:421–424.
Burton, T. M. and G. E. Likens. 1979a. Salamander populations and biomass in the
Hubbard Brook Experimental Forest, New Hampshire. Copeia 3:541–546.
Burton, T. M. and G. E. Likens. 1979b. Energy flow and nutrient cycling in salamander
populations in the Hubbard Brook Experimental Forest, New Hampshire. Ecology
56:1068–1080.
Bury, R. B. 2004. Wildfire, fuel reduction, and herpetofaunas across diverse landscape
mosaics in northwestern forests. Conservation Biology 18:968–975.
Brooks, R. T. 1999. Residual effects of thinning and high white-tailed deer densities on
northern redback salamanders in southern New England oak forests. Journal of
Wildlife Management 63:1172–1180.
Chazal, A. C. and P. H. Niewiarowski. 1998. Responses of mole salamanders to
clearcutting: using field experiments in forest management. Ecological Applications
8:1133–1143.
Cole, E. C., W. C. McComb, M. Newton, C. L. Chambers, and J. P. Leeming. 1997.
Response of amphibians to clearcutting, burning, and glyphosphate application in the
Oregon coast range. Journal of Wildlife Management 61:656–664.
Colwell, R. K. 2006. EstimateS: statistical estimation of species richness and shared
species from samples. Version 8. Persistent URL <purl.oclc.org/estimates>, (accessed
July 20, 2009).
Colwell, R. K., and J. A. Coddington. 1994. Estimating terrestrial biodiversity through
extrapolation. Philosophical Transactions of the Royal Society, London B 345:101–
118.
Cromer, R. B., J. D. Lanham, and H. H. Hanlin. 2002. Herpetofaunal response to gap and
skidder-rut wetland creation in a southern bottomland hardwood forest. Forest
Science 48:407–413.
Crosswhite, D. L., S. F. Fox, and R. E. Thill. 2004. Herpetological habitat relations in the
Ouachita mountains, Arkansas. Pages 273–282 in J. M. Guldin, editor. Ouachita and
Ozark Mountains symposium: ecosystem management research. Gen. Tech. Rep.
SRS–74. Asheville, NC: U.S. Department of Agriculture, Forest Service, Southern
Research Station.
Dodd, C. K. Jr., W. J. Barichivich, S. A. Johnson, and J. S. Staiger. 2007. Changes in a
northwestern Florida gulf coast herpetofaunal community over a 28-y period.
American Midland Naturalist 158:29–48.
Drever, C. R., G. Peterson, C. Messier, Y. Bergeron, and M. Flannigan. 2006. Can forest
management based on natural disturbances maintain ecological resilience? Canadian
Journal of Forest Resources 36:2285–2299.
Driscoll, D. A., and M. K. Henderson. 2008. How many common reptile species are fire
specialists? A replicated natural experiment highlights the predictive weakness of a
fire succession model. Biological Conservation 141:460-471.
Duncan, R. S., and J. E. Linhoss. 2005. Regeneration of Virginia pine (Pinus virginiana)
following southern pine beetle (Dendroctonus frontalis) outbreak in the Sipsey
Wilderness, Alabama. Forest Ecology and Management 212:65–74.
Enge, K. M. and W. R. Marion. 1986. Effects of clearcutting and site preparation on
herpetofauna of a north Florida flatwoods. Forest Ecology and Management 14:177–
192.
Ford, W. M., M. A. Menzel, D. W. McGill, J. Laerm, T. S. McCay. 1999. Effects of a
community restoration fire on small mammals and herpetofauna in the southern
Appalachians. Forest Ecology and Management 114:233–243.
Forest Inventory and Analysis. 1998. Field instructions for southern forest inventory.
Remeasurement of prism plots. Southern Research Station, Forest Service, U.S.
Department of Agriculture. Item 26 version of manual.
Gaines, G. D. and J. W. Creed. 2003. Forest health and restoration project. National
forests in Alabama, Bankhead National Forest Franklin, Lawrence, and Winston
Counties, Alabama. Final environmental impact statement. Management Bulletin R8–
MB 110B. 353 pp.
Gibbons, 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 50:653–656.
Goldstein, M. E., R. N. Wilkins, and T. E. Lacher, Jr. 2005. Spatiotemporal responses of
reptiles and amphibians to timber harvest treatments. Journal of Wildlife
Management 69:525–539.
Gotelli, N. J. and R. K. Colwell. 2001. Quantifying biodiversity: procedures and pitfalls
in the measurement and comparison of species richness. Ecology Letters 4:379–391.
Greenberg, C. H., D. G. Neary, and L. D. Harris. 1994. Effect of high-intensity wildfire
and silvicultural treatments on reptile communities in sand-pine scrub. Conservation
Biology 8:1047–1057.
Greenberg, C. H. 2000. Response of reptile and amphibian communities to canopy gaps
created by wind disturbance in the southern Appalachians. Forest Ecology and
Management 148:135–144.
Greenberg, C. H., and T. A. Waldrop. 2008. Short-term response of reptiles and
amphibians to prescribed fire and mechanical fuel reduction in a southern
Appalachian upland hardwood forest. Forest Ecology and Management 255:2883–
2893.
Grialou, J. A, S. D. West, and R. N. Wilkins. 2000. The effects of forest clearcut
harvesting and thinning on terrestrial salamanders. Journal of Wildlife Management
64:105–113.
Healthy Forest Restoration Act. Act of December 3, 2003. 117 Stat., as amended; 16
U.S.C. 1887–1915.
Homyack, J. A., and C. A. Haas. 2009. Long-term effects of experimental forest
harvesting on abundance and reproductive demography of terrestrial salamanders.
Biological Conservation 142:110–121.
Horn, S., J. L. Hanula, and M. D. Ulyshen. 2005. Abundance of green tree frogs and
insects in artificial canopy gaps in a bottomland hardwood forest. American Midland
Naturalist 153:321–326.
James, S. E., and R. T. M’Closkey. 2003. Lizard microhabitat and fire fuel management.
Biological Conservation 114:293–297.
Karraker, N. E., and H. H. Welsh, Jr. 2006. Long-term impacts of even-aged timber
management on abundance and body condition of terrestrial amphibians in
northwestern California. Biological Conservation 131:132–140.
Kirkland, G. L. Jr., H. W. Snoddy, and T. L. Amsler. 1996. Impact of fire on small
mammals and amphibians in a central Appalachian deciduous forest. American
Midland Naturalist 135:253–260.
Knapp, S. M., C. A. Haas, D. N. Harpole, and R. L. Kirkpatrick. 2003. Initial effects of
clearcutting and alternative silvicultural practices on terrestrial salamander abundance.
Conservation Biology 17:752–762.
Knebel, L., and T. R. Wentworth. 2007. Influence of fire and southern pine beetle on
pine-dominated forests in the Linville Gorge Wilderness, North Carolina. Castanea
72:214–225.
Krebs, C. J. 1999. Ecological Methodology, 2nd edition. Addison-Wesley Educational
Publishers, Inc. 620 pp.
Krebs, C. J. 2003. Ecological Methodology v.6.1.1. Programs for Ecological
Methodology, 2nd edition.
Land, A.D., and L. K. Rieske. 2006. Interactions among prescribed fire, herbivore
pressure and shortleaf pine (Pinus echinata) regeneration following southern pine
beetle (Dendroctonus frontalis) mortality. Forest Ecology and Management 235:260–
269.
Loehle, C., T. B. Wigley, P. A. Shipman, S. F. Fox, Scott Rutzmoser, R. E. Thill, and, M.
A. Melchiors. 2005. Herpetofaunal species richness responses to forest landscape
structure in Arkansas. Forest Ecology and Management 209:293–308.
MacCracken, J. G. 2005. Effects of uneven-aged timber harvest on forest floor
vertebrates in the Cascade Mountains of southern Washington. Forest Ecology and
Management 208:123-135.
Marzluff, J. M., M. G. Raphael, and R. Sallabanks. 2000. Understanding the effects of
forest management on avian species. Wildlife Society Bulletin 28:1132–1143.
McKenny, H. C., W. S. Keeton, and T. M. Donovan. 2006. Effects of structural
complexity enhancement on eastern red-backed salamander (Plethodon cinereus)
populations in northern hardwood forests. Forest Ecology and Management 230:186–
196.
McCune, B. and J. B. Grace. 2003. Analysis of ecological communities. MjM Software.
Oregon. 304 pp.
McLeod, R. F and J. E. Gates. 1998. Response of herpetofaunal communities to forest
cutting and burning at Chesapeake Farms, Maryland. American Midland Naturalist
139:164–177.
McWilliams, W. H. 1991. Availability of yellow pine sawtimber in Alabama. Pages 115–
119 in R. H. Jones, editor. Proceedings: Alabama’s forest resources: past, present, and
future. Auburn University, AL.
Means, D. B., C. K. Dodd, Jr., S. A. Johnson, and J. G. Palis. 2004. Amphibians and fire
in longleaf pine ecosystems: response to Schurbon and Fauth. Conservation Biology
18:1149–1153.
Messere, M., and P. K. Ducey. 1998. Forest floor distribution of northern redback
salamanders, Plethodon cinereus, in relation to canopy gaps: first year following
selective logging. Forest Ecology and Management 107:319–324.
Moseley, K. R., S. B. Castleberry, and S. H. Schweitzer. 2003. Effects of prescribed fire
on herpetofauna in bottomland hardwood forests. Southeastern Naturalist 2:475–
486.
Mushinsky, H. R. 1985. Fire and the Florida sandhill herpetofaunal community: with
special attention to responses of Cnemidophorus sexlineatus. Herpetologica
41:333–342.
Naughton, G. P., C. B. Henderson, K. R. Foresman, and R. L. McGraw, II. 2000. Longtoed salamanders in harvested and intact douglas-fir forests of western Montana.
Ecological Applications 10:1681–1689.
Owens, A. K., K. R. Moseley, T. S. McCay, S. B. Castleberry, J. C. Kilgo, and W. M.
Ford. 2008. Amphibian and reptile community response to coarse woody debris
manipulations in upland loblolly pine (Pinus taeda) forests. Forest Ecology and
Management 256:2078–2083.
Patrick, D. A., M. L. Hunter, Jr., and A. J. K. Calhoun. 2006. Effects of experimental
forestry treatments on a Maine amphibian community. Forest Ecology and
Management 234:323–332.
Perkins, D. W., and M. L. Hunter, Jr. 2006. Effects of riparian timber management on
amphibians in Maine. Journal of Wildlife Management 70:657–670.
Petranka, J. W. and S. S. Murray. 2001. Effectiveness of removal sampling for
determining salamander density and biomass: a case study in an Appalachian
streamside community. Journal of Herpetology 35:36–44.
Pilliod, D. S., R. B. Bury, E. J. Hyde, C. A. Pearl, and P. S. Corn. 2003. Fire and
amphibians in North America. Forest Ecology and Management 178:163–181.
Provencher, L., A. R. Litt, and D. R. Gordon. 2003. Predictors of species richness in
northwest Florida longleaf pine sandhills. Conservation Biology 17:1660–1671.
Renken, R. B., W. K. Gram, D. K. Fantz, S. C. Richter, T. J. Miller, K. B. Ricke, B.
Russell, and X. Wang. 2004. Effects of forest management on amphibians and reptiles
in Missouri Ozark forests. Conservation Biology 18:174–188.
Russell, K. R., D. H. Van Lear, and D. C. Guynn, Jr. 1999. Prescribed fire effects on
herpetofauna: review and management implications. Wildlife Society Bulletin
27:374–384.
Russell, K. R, T. B. Wigley, W. M. Baughman, H. G. Hanlin, and W. M. Ford. 2004.
Responses of southeastern amphibians and reptiles to forest management: A review.
Pages 319–334 in M. H. Rauscher and K. Johnsen, editors. Southern forest science:
past, present, and future. United States Department of Agriculture Southern Research
Station Gen. Tech. Rep. SRS–75.
Ruthven, D. C. III, R. T. Kazmaier, and M. W. Janis. 2008. Short-term response of
herpetofauna to various burning regimes in the south Texas plains. The Southwestern
Naturalist 53:480–487.
Schweitzer, C. J. and W. Tadesse. 2004. The interaction of fire frequency and stand
density on plant community dynamics. Pages 1–14 in CREST–Center for Forest
Ecosystem Assessment (CFEA) Alabama A&M University. National Science
Foundation proposal.
Schurbon, J. M. and J. E. Fauth. 2003. Effects of prescribed burning on amphibian
diversity in a southeastern U.S. national forest. Conservation Biology 17:1338–
1349.
Smalley, G. W. 1982. Classification and evaluation of forest sites on the mid-Cumberland
Plateau. U.S. Department of Agriculture Forest Service, General Technical Report S038. Southern Forest Experimental Station, New Orleans, Louisiana. 58 pp.
Smith, D. M., B. C. Larson, M. J. Kelty, and P. M. S. Ashton. 1997. The practice of
silviculture: Applied forest ecology. John Wiley and Sons Inc. 537 pp.
Stuart, S. N., J. S. Chanson, N. A. Cox, B. E. Young, A. S. L. Rodriques, D. L. Fischman,
and R. W. Waller. 2004. Status and trends of amphibian declines and extinctions
worldwide. Science 306:1783–1791.
ter Braak C. F. J. 1995. Ordination. Pages 91–173 in R. H. G. Jongman, C.J.F. ter Braak,
and O. F. R. Van Tongeren, editors. Data Analysis in Community and Landscape
Ecology. Cambridge University Press, Cambridge.
Todd, B. D., and K. M. Andrews. 2007. Response of a reptile guild to forest harvesting.
Conservation Biology 22:753–761.
Turner, M. G., R. H. Gardner, V. H. Dale, and R. V. O’Neill. 1989. Predicting the spread
of disturbance across heterogeneous landscapes. Oikos 55:121–129.
Vitt, L. J., T. C. S. Avila-Pires, J. P. Caldwell, and V. R. L. Oliveira. 1998. The impact of
individual tree harvesting on thermal environments of lizards in Amazonian rain
forests. Conservation Biology 12:654–664.
Wilgers, D. J. and E. A. Horne. 2006. Effects of different burn regimes on tallgrass
prairie herpetofaunal species diversity and community composition in the Flint Hills,
Kansas. Journal of Herpetology 40:73–84.
Wyman, R. L. 1998. Experimental assessment of salamanders as predators of detrital
food webs: effects on invertebrates, decomposition and the carbon cycle. Biodiversity
and Conservation 7:641–750.
Yager, L. Y., M. G. Hinderliter, C. D. Heise, and D. M. Epperson. 2007. Gopher tortoise
response to habitat management by prescribed burning. The Journal of Wildlife
Management 71:428–434.
Figure 3.2 Locations of study sites within the northwestern portion of the William B.
Bankhead National Forest, Alabama, U.S.A. Forest treatments consisted of the six
following categories: C-control, B-burn, HT-heavy thin, LT-light thin, HTB-heavy thin
and burn, and LTB-light thin and burn. Numbers after treatment abbreviations correspond
to block number.
HT3
HTB1
LT3
LT2
C3
B3
HT2
HTB2 HT1
C2
LT1
C1
LTB2
HTB3
LTB1
LTB3
B1
B2
Habitat Variable
Percent Litter
Percent Bare Ground
Percent Herbaceous
Percent Woody
Percent Slash
Percent Rock
Percent CWD
Litter Depth
Canopy Cover
Forest Level 1
Forest Level 2
Forest Level 3
Forest Level 4
CWD volume
Air Temperature
Soil Temperature
Light Intensity
Relative Humidity
Code
%_litt
%_bare
%_herb
%_woody
%_slash
%_rock
%_CWD
l_dep
can_cov
for_lev1
for_lev2
for_lev3
for_lev4
CWD_vol
air_temp
soil_temp
%_light
%_hum
Habitat Description
Presence of ground cover such as leaves or small woody debris
Absence of ground cover (e.g., exposed soil)
Any non-woody stem such as grasses, ferns, and Smilax and Vitus spp.
Any woody stems such as seedlings and large trees (living or dead); woody stems taller than one meter had to contact transects directly to be counted
Any woody debris composed of two or more stems 30 cm or higher from the ground (e.g., fallen treetops)
Presence of rocky substrate greater than 10 cm in size
Any fallen woody debris larger than 10 cm in diameter (must touch the ground somewhere along the length to be counted)
Determined by measuring depth of the substrate to the nearest 0.5 cm with a metric ruler
Estimated with a spherical densiometer as the sum percentage of open points subtracted from 100%
Percent coverage of forest levels ≤ 2m (classified as ground cover)
Percent coverage of forest levels > 2 m – ≤ 4 m (classified as understory)
Percent coverage of forest levels > 4 m – ≤ 6 m (classified as midstory)
Percent coverage of forest levels > 6 m (classified as overstory)
Calculated as volume of a cylinder for each enumerated CWD (see text)
Average daily air temperature (⁰C) during the month of June recorded at 2:00 PM
Average daily soil temperature (⁰C) during the month of June recorded at 2:00 PM
Average daily light intensity (%) during the month of June recorded at 2:00 PM
Average daily relative humidity (%) during the month of June recorded at 2:00 PM
Table 3.1 Habitat variables assessed at each survey point
Scieni fic Name
Agkistrodon contortrix
Am bystoma maculatum
Am bystoma opacum
Anaxyrus fowleri
Anolis carolinensis
Carphophis a. am oenus
Coluber c. constrictor
Crotalus horridus
Diadophis p. punctatus
Eurycea cirrigera
Gastrophryne carolinensis
Hyla chrysoscelis
Hyla gratiosa
Lampropeltis getula nigra
Lampropeltis triangulum syspila
Lithobates catesbeianus
Lithobates clamitans m elanota
Lithobates palustris
Lithobates sphenocephalus utricularius
Nerodia s. sipedon
Notophthalm us v. viridescens
Pantherophis guttatus
Pantherophis spiloides
Plestiodon fasciatus
Plestiodon laticeps
Plethodon dorsalis
Plethodon m ississippi
Pseudacris brachyphona
Pseudotriton r. ruber
Scaphiopus holbrookii
Sceloporus undulatus
Scincella lateralis
Storeria dekayi wrightorium
Storeria o. occipitomaculata
Terrapene c. carolina
Thamnophis s. sirtalis
Totals
Abbreviation
AGCO
AMMA
AMOP
ANFO
ANCA
CAAM
COCO
CRHO
DIPU
EUCI
GACA
HYCH
HYGR
LAGE
LAT R
LICA
LICL
LIPA
LISP
NESI
NOVI
PAGU
PASP
PLFA
PLLA
PLDO
PLMI
PSBR
PSRU
SCHO
SCUN
SCLA
ST DE
ST OC
T ECA
T HSI.
C ommon Name
Copperhead
Spotted Salamander
Marbled Salamander
Fowler's T oad
Green Anole
Worm Snake
Northern Black Racer
T imber Ratt lesnake
Southern Ring-necked Snake
Southern T wo-lined Salamander
Eastern Narrow-mouthed T oad
Cope's Gray T reefrog
Barking T reefrog
Eastern Black Kingsnake
Red Milksnake
American Bullfrog
Northern Green Frog
Pickerel Frog
Southern Leopard Frog
Northern Watersnake
Red-spotted Newt
Red Cornsnake
Gray Ratsnake
Common Five-lined Skink
Broad-headed Skink
Northern Zigzag Salamander
Mississippi Slimy Salamander
Mountain Chorus Frog
Red Salamander
Eastern Spadefoot
Eastern Fence Lizard
Little Brown Skink
Midland Brownsnake
Northern Red-bellied Snake
Eastern Box T urt le
Eastern Gart ersnake
Control Burn He avy Thin Light Thin Heavy Thin & Burn Light Thin & Burn Total
5
3
5
5
5
4
27
0
1
0
0
0
0
1
0
1
2
0
0
0
3
0
7
9
3
2
5
26
3
0
9
6
7
4
29
1
0
0
1
2
0
4
1
2
2
2
0
2
9
0
0
0
0
1
3
4
1
0
1
0
1
1
4
2
0
0
0
0
0
2
1
3
4
0
1
0
9
0
0
0
0
1
0
1
0
1
0
0
0
0
1
2
1
0
0
0
1
4
0
0
0
0
0
1
1
5
0
2
0
2
2
11
2
3
2
0
9
10
26
0
1
1
4
0
1
7
0
0
9
0
0
0
9
0
1
0
0
0
0
1
0
0
0
0
0
1
1
1
1
1
1
0
0
4
2
0
2
0
2
0
6
4
4
2
4
4
4
22
2
3
3
2
4
2
16
2
0
6
8
0
4
20
28
28
42
19
19
68
204
0
3
1
0
1
1
6
4
0
6
2
2
1
15
0
4
0
0
0
0
4
2
0
3
4
3
0
12
20
5
10
12
15
8
70
0
0
1
0
0
0
1
0
0
1
0
0
0
1
0
1
0
0
1
0
2
1
0
2
5
0
2
10
89
73
126
78
82
125
573
Recapture s
0
0
0
10
4
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
1
0
4
0
0
0
1
0
0
0
0
0
23
Table 3.2 Total pre-treatment herpetofaunal captures in forest stands managed by prescribed burning and thinning
in the William B. Bankhead National Forest, Alabama, U.S.A. (2005-2006).
Scie nific Name
Acris crepitans crepitans
Agkistrodon contortrix
Ambystom a maculatum
Ambystom a opacum
Anaxyrus fowleri
Anolis carolinensis
Carphophis a. am oenus
Cem ophora c. copei
Coluber c. constrictor
Crotalus horridus
Diadophis p. punctatus
Gastrophryne carolinensis
Heterodon platirhinos
Hyla chrysoscelis
Hyla cinerea
Hyla gratiosa
Lam propeltis getula nigra
Lam propeltis triangulum syspila
Lithobates catesbeianus
Lithobates clamitans m elanota
Lithobates palustris
Lithobates sphenocephalus utricularius
Nerodia s. sipedon
Notophthalm us v. viridescens
Pantherophis guttatus
Pantherophis spiloides
Plestiodon a. anthracinus
Plestiodon fasciatus
Plestiodon laticeps
Plethodon dorsalis
Plethodon mississippi
Pseudacris brachyphona
Pseudotriton r. ruber
Scaphiopus holbrookii
Sceloporus undulatus
Scincella lateralis
Storeria o. occipitom aculata
Terrapene c. carolina
Tham nophis s. sirtalis
Totals
Abbreviation
ACCR
AGCO
AMMA
AMOP
ANFO
ANCA
CAAM
CECO
COCO
CRHO
DIPU
GACA
HEPL
HYCH
HYCI
HYGR
LAGE
LAT R
LACA
LACL
LIPA
LISP
NESI
NOVI
PAGU
PASP
PLAN
PLFA
PLLA
PLDO
PLMI
PSBR
PSRU
SCHO
SCUN
SCLA
ST OC
TECA
THIS
C ommon Name
Eastern Cricket Frog
Copperhead
Spotted Salamander
Marbled Salamander
Fowler's T oad
Green Anole
Worm Snake
Northern Scarlet Snake
Northern Black Racer
T imber Rattlesnake
Southern Ring-necked Snake
Eastern Narrow-mouthed T oad
Eastern Hognose Snake
Cope's Gray T reefrog
Green T reefrog
Barking T reefrog
Eastern Black Kingsnake
Red Milksnake
American Bullfrog
Northern Green Frog
Pickerel Frog
Southern Leopard Frog
Northern Watersnake
Red-spotted Newt
Red Cornsnake
Gray Ratsnake
Northern Coal Skink
Common Five-lined Skink
Broad-headed Skink
Northern Zigzag Salamander
Mississippi Slimy Salamander
Mountain Chorus Frog
Red Salamander
Eastern Spadefoot
Eastern Fence Lizard
Little Brown Skink
Northern Red-bellied Snake
Eastern Box T urtle
Eastern Gartersnake
C ontrol Burn He avy Thin Light Thin Heavy Thin & Burn Light Thin & Burn
0
5
0
0
0
3
14
4
8
24
8
10
0
1
1
0
0
4
0
1
0
0
0
0
7
19
13
7
21
21
11
1
23
24
32
25
1
1
0
1
2
0
0
0
1
0
0
0
4
7
14
4
4
2
1
1
1
1
0
0
0
0
0
1
0
1
1
4
15
0
3
0
0
0
0
0
1
0
1
0
4
1
0
3
0
0
0
0
0
1
0
1
0
0
0
0
3
1
4
2
1
3
2
1
0
0
0
0
4
3
0
0
0
9
1
3
1
0
0
21
1
3
1
0
0
2
1
2
1
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
1
3
2
0
1
0
1
2
1
2
1
0
0
0
0
0
7
0
12
4
9
9
1
0
2
9
4
4
0
0
25
7
0
5
27
30
57
68
34
56
0
2
1
1
0
0
4
0
1
0
0
3
1
4
1
1
3
1
2
2
10
12
10
11
14
4
6
3
4
3
1
1
0
0
0
0
1
0
0
0
0
0
8
3
2
1
0
0
120
105
206
176
140
199
Total Re captures
0
8
0
68
0
6
0
1
38
88
70
116
0
5
0
1
1
35
0
4
0
2
1
23
0
1
0
9
0
1
0
1
0
14
0
3
0
16
0
26
0
7
0
4
0
1
0
1
1
6
0
7
0
1
7
41
11
20
0
37
13
272
0
4
0
8
1
11
15
47
5
34
0
2
0
1
0
14
946
163
Table 3.3 Total post-treatment year one herpetofaunal captures in forest stands managed by prescribed burning
and thinning in the William B. Bankhead National Forest, Alabama, U.S.A. (2006-2007).
Scie nific Name
Acris crepitans crepitans
Agkistrodon contortrix
Am bystom a m aculatum
Am bystom a opacum
Anaxyrus fowleri
Anolis carolinensis
Carphophis a. am oenus
Cemophora c. copei
Chelydra s. serpentina
Coluber c. constrictor
Crotalus horridus
Diadophis p. punctatus
Gastrophryne carolinensis
Heterodon platirhinos
Hyla chrysoscelis
Hyla cinerea
Lampropeltis getula nigra
Lampropeltis triangulum elapsoides
Lampropeltis triangulum syspila
Lithobates clam itans melanota
Lithobates palustris
Lithobates sphenocephalus utricularius
Nerodia s. sipedon
Notophthalm us v. viridescens
Opheodrys a. aestivus
Pantherophis guttatus
Pantherophis spiloides
Plestiodon a. anthracinus
Plestiodon fasciatus
Plestiodon laticeps
Plethodon dorsalis
Plethodon m ississippi
Pseudacris brachyphona
Pseudacris crucifer
Pseudotriton r. ruber
Scaphiopus holbrookii
Sceloporus undulatus
Scincella lateralis
Storeria dekayi wrightorium
Storeria o. occipitom aculata
Terrapene c. carolina
Thamnophis s. sirtalis
Virginia v. valeriae
Totals
Abbre viation
ACCR
AGCO
AMMA
AMOP
ANFO
ANCA
CAAM
CECO
CHSI
COCO
CRHO
DIPU
GACA
HEPL
HYCH
HYCI
LAGE
LAEL
LAT R
LICL
LIPA
LISP
NESI
NOVI
OPAE
PAGU
PASP
PLAN
PLFA
PLLA
PLDO
PLMI
PSBR
PSCR
PSRU
SCHO
SCUN
SCLA
ST DE
ST OC
T ECA
T HSI.
VIVA
C ommon Name
Eastern Cricket Frog
Copperhead
Spotted Salamander
Marbled Salamander
Fowler's T oad
Green Anole
Worm Snake
Northern Scarlet Snake
Eastern Snapping T urtle
Northern Black Racer
T imber Rattlesnake
Southern Ring-necked Snake
Eastern Narrow-mouthed T oad
Eastern Hognose Snake
Cope's Gray T reefrog
Green T reefrog
Eastern Black Kingsnake
Scarlet Kingsnake
Red Milksnake
Northern Green Frog
Pickerel Frog
Southern Leopard Frog
Northern Watersnake
Red-spotted Newt
Northern Rough Greensnake
Red Cornsnake
Gray Ratsnake
Northern Coal Skink
Common Five-lined Skink
Broad-headed Skink
Northern Zigzag Salamander
Mississippi Slimy Salamander
Mountain Chorus Frog
Spring Peeper
Red Salamander
Eastern Spadefoot
Eastern Fence Lizard
Little Brown Skink
Midland Brownsnake
Northern Red-bellied Snake
Eastern Box T urtle
Eastern Gartersnake
Eastern Smooth Earthsnake
C ontrol Burn He avy Thin Light Thin He avy Thin * Burn Light Thin * Burn Total Re capture s
0
2
1
0
0
0
0
3
12
10
17
18
6
10
8
73
0
0
0
0
0
2
0
2
0
0
3
0
0
0
0
3
9
23
7
2
8
9
16
58
9
3
23
31
24
23
64
113
2
0
0
3
0
2
0
7
0
1
0
0
0
0
0
1
0
0
0
0
0
1
0
1
6
2
23
14
6
13
4
64
0
1
1
1
0
0
0
3
1
2
0
0
0
0
0
3
1
6
20
1
2
5
2
35
1
2
0
0
1
0
0
4
0
0
3
2
0
2
0
7
1
0
0
1
2
0
0
4
2
5
4
10
2
3
2
26
0
1
0
0
0
0
0
1
0
0
0
1
0
1
0
2
12
2
0
1
5
15
3
35
6
4
5
4
8
5
0
32
28
2
4
3
4
11
1
52
0
0
0
0
0
1
0
1
0
0
0
0
0
1
0
1
0
2
0
1
0
0
0
3
1
0
2
2
1
0
1
6
1
2
1
1
1
1
0
7
1
0
0
0
0
0
0
1
3
4
5
8
9
7
3
36
2
1
4
8
5
9
9
29
0
0
2
0
0
1
0
3
18
20
33
35
24
38
3
168
0
1
4
3
0
2
0
10
0
0
2
1
0
0
0
3
1
1
1
1
0
2
0
6
2
15
3
9
3
3
3
35
1
7
7
9
23
23
35
70
9
4
17
9
8
14
3
61
0
0
0
0
0
1
0
1
0
1
0
0
0
0
0
1
1
0
2
1
0
0
1
4
1
2
1
2
0
1
0
7
0
0
0
1
0
0
0
1
131
126
195
183
142
206
983
158
Table 3.4 Total post-treatment year two herpetofaunal captures in forest stands managed by prescribed
burning and thinning in the William B. Bankhead National Forest, Alabama, U.S.A. (2007-2008).
Scienific Name
Agkistrodon contortrix
Anaxyrus fowleri
Anolis carolinensis
Coluber c. constrictor
Crotalus horridus
Gastrophryne carolinensis
Heterodon platirhinos
Lam propeltis getula nigra
Lithobates clamitans melanota
Pantherophis guttatus
Pantherophis spiloides
Plestiodon fasciatus
Plestiodon laticeps
Plethodon m ississippi
Pseudacris brachyphona
Scaphiopus holbrookii
Sceloporus undulatus
Scincella lateralis
Storeria dekayi wrightorium
Tantilla coronata
Terrapene c. carolina
Thamnophis s. sirtalis
Virginia v. valeriae
Totals
Abbre viation
AGCO
ANFO
ANCA
COCO
CRHO
GACA
HEPL
LAGE
LICL
PAGU
PASP
PLFA
PLLA
PLMI
PSBR
SCHO
SCUN
SCLA
ST DE
T ACO
T ECA
T HSI.
VIVA
C ommon Name
Copperhead
Fowler's T oad
Green Anole
Northern Black Racer
T imber Rattlesnake
Eastern Narrow-mouthed T oad
Eastern Hognose Snake
Eastern Black Kingsnake
Northern Green Frog
Red Cornsnake
Gray Ratsnake
Common Five-lined Skink
Broad-headed Skink
Mississippi Slimy Salamander
Mountain Chorus Frog
Eastern Spadefoot
Eastern Fence Lizard
Little Brown Skink
Midland Brownsnake
Southeastern Crowned Snake
Eastern Box T urtle
Eastern Gartersnake
Eastern Smooth Earthsnake
C ontrol Burn He avy Thin Light Thin Heavy Thin & Burn Light Thin & Burn Total Recaptures
1
3
1
3
1
1
0
10
0
0
2
2
0
1
1
5
3
0
3
9
5
5
8
25
0
1
4
4
4
3
4
16
0
0
1
0
0
1
0
2
1
3
0
1
1
1
0
7
0
0
0
1
0
0
0
1
0
0
1
1
3
1
0
6
0
1
0
1
0
1
0
3
0
0
0
0
1
0
0
1
0
0
2
0
0
0
0
2
0
0
2
0
1
2
1
5
0
0
2
1
1
3
1
7
4
3
8
3
7
5
2
30
1
1
0
0
0
1
0
3
2
1
0
0
0
1
0
4
0
1
0
1
11
5
9
18
0
0
1
4
1
1
0
7
0
0
1
0
0
1
0
2
0
1
0
0
0
0
0
1
0
0
1
0
0
0
0
1
0
1
0
1
0
0
0
2
0
0
1
0
0
0
0
1
12
16
30
32
36
33
159
26
Table 3.5 Total post-treatment year three herpetofaunal captures in forest stands managed by prescribed burning and thinning
in the William B. Bankhead National Forest, Alabama, U.S.A. (2008).
Table 3.6 Community diversity of all herpetofauna, amphibians only, and reptiles only in
managed forest stands of the William B. Bankhead National Forest, Alabama.
α–Diversity a β–Diversity b γ–Diversity c Heterogeneity d % Similarity e % Exclusive Speciesf
All He rpe tofauna
Control
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Burn
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Heavy T hin
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Light T hin
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Heavy T hin & Burn
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Light T hin & Burn
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Amphibians
Control
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Burn
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Heavy T hin
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Light T hin
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Heavy T hin & Burn
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Light T hin & Burn
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Re ptile s
Control
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Burn
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Heavv T hin
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Light T hin
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Heavy T hin & Burn
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
Light T hin & Burn
Pre-T reatment
Post-T reatment Year
Post-T reatment Year
One
T wo
12.2 ± 2.7
15.0 ± 2.7
13.7 ± 2.9
1.9
2.0
1.8
22.6
29.4
25.1
1.7 ± 0.2
2.1 ± 0.2
1.9 ± 0.2
75.0 ± 37.4
72.3 ± 1.2
35.3 ± 5.9
0
1.7 ± 1.7
0.8 ± 0.8
One
T wo
9.9 ± 1.5
15.4 ± 2.8
16.9 ± 1.1
1.9
1.7
1.7
19.3
26.1
28.0
1.6 ± 0.1
2.0 ± 0.1
2.1 ± 0.1
69.3 ± 24.2
72.3 ± 1.5
36.7 ± 4.1
3.8 ± 2.5
2.6 ± 2.6
2.4 ± 1.4
One
T wo
15.6 ± 3.0
16.7 ± 1.7
16.5 ± 4.2
1.7
1.7
1.6
26.1
27.8
27.2
2.0 ± 0.1
2.1 ± 0.1
2.3 ± 0.2
75.0 ± 26.3
70.7 ± 5.4
58.3 ± 3.3
1.9 ± 1.0
0.9 ± 0.9
0.8 ± 0.8
One
T wo
9.0 ± 1.6
11.9 ± 0.9
18.7 ± 4.2
1.8
1.8
1.4
15.9
21.9
26.9
1.6 ± 0.3
1.8 ± 0.1
2.2 ± 0.3
51.3 ± 39.6
87.3 ± 3.0
55.3 ± 16.5
0
0
0
One
T wo
13.1 ± 3.5
12.4 ± 0.8
14.5 ± 2.6
1.6
1.5
1.5
20.6
18.5
21.5
1.7 ± 0.3
1.9 ± 0.1
2.1 ± 0.2
47.7 ± 31.6
73.7 ± 9.1
76.0 ± 2.9
1.0 ± 1.0
1.7 ± 0.9
0
One
T wo
11.5 ± 0.7
15.5 ± 3.2
17.3 ± 5.3
1.8
1.5
1.7
20.2
23.6
29.5
1.5 ± 0.1
2.0 ± 0.04
2.2 ± 0.3
81.3 ± 31.4
61.3 ± 6.9
61.7 ± 5.7
1.9 ± 1.0
0.9 ± 0.9
4.0 ± 2.9
One
T wo
3.5 ± 1.3
3.4 ± 0.8
4.1 ± 1.3
1.7
3.2
2.2
6.1
10.8
8.9
0.7 ± 0.4
1.0 ± 0.2
1.1 ± 0.3
80.7 ± 0.9
83.3 ± 4.8
43.0 ± 19.0
0
0
0
One
T wo
4.9 ± 1.8
7.0 ± 1.9
4.2 ± 1.2
2.1
1.9
2.4
10.2
13.1
9.9
0.9 ± 0.3
1.3 ± 0.2
1.1 ± 0.3
69.0 ± 1.7
79.0 ± 2.3
54.7 ± 13.7
6.7 ± 3.8
3.7 ± 3.7
0
One
T wo
5.5 ± 1.0
6.4 ± 1.3
9.2 ± 2.7
2.1
2.2
1.4
11.4
14.2
13.3
1.2 ± 0.1
1.1 ± 0.1
1.5 ± 0.3
83.0 ± 4.3
73.0 ± 5.5
72.0 ± 10.8
0
0
0
One
T wo
2.6 ± 0.5
3.6 ± 0.7
6.7 ± 3.2
1.8
2.0
2.0
4.8
7.2
13.2
0.6 ± 0.1
0.6 ± 0.04
0.9 ± 0.4
62.7 ± 17.7
97.3 ± 1.3
33.0 ± 30.0
2.2 ± 2.2
0
0
One
T wo
4.1 ± 0.9
3.0 ± 0.2
5.3 ± 1.7
2.1
1.6
1.7
8.8
4.8
8.9
0.9 ± 0.2
0.8 ± 0.05
1.1 ± 0.3
62.0 ± 14.2
73.7 ± 9.3
73.7 ± 6.2
2.2 ± 2.2
1.9 ± 1.9
0
One
T wo
3.9 ± 1.1
5.4 ± 3.0
7.2 ± 3.5
2.6
2.2
1.9
10.0
11.9
13.5
0.7 ± 0.1
1.0 ± 0.4
1.1 ± 0.6
94.0 ± 2.1
75.0 ± 1.5
66.7 ± 8.7
2.2 ± 2.2
1.9 ± 1.9
3.9 ± 3.9
One
T wo
7.1 ± 0.5
9.5 ± 0.9
8.4 ± 1.3
2.1
1.8
2.0
14.8
17.1
16.5
1.3 ± 0.05
1.7 ± 0.1
1.5 ± 0.1
54.3 ± 19.2
65.0 ± 2.1
30.1 ± 4.5
0
3.2 ± 3.2
1.3 ± 1.3
One
T wo
4.0 ± 0.4
5.5 ± 0.5
10.5 ± 0.9
2.3
2.3
1.7
9.1
12.6
17.9
1.0 ± 0.1
1.2 ± 0.1
1.7 ± 0.1
31.0 ± 2.1
34.0 ± 5.6
49.7 ± 4.6
1.8 ± 1.8
1.6 ± 1.6
4.0 ± 2.3
One
T wo
8.1 ± 2.4
9.2 ± 0.6
10.0 ± 1.8
1.7
1.4
1.3
13.4
13.2
12.5
1.5 ± 0.3
1.6 ± 0.07
1.7 ± 0.1
43.0 ± 9.6
73.3 ± 2.1
72.3 ± 7.5
3.5 ± 1.8
1.6 ± 1.6
1.3 ± 1.3
One
T wo
4.6 ± 1.4
8.9 ± 1.4
11.7 ± 1.2
2.2
1.6
1.4
10.0
14.3
16.7
1.2 ± 0.3
1.6 ± 0.1
1.9 ± 0.2
53.7 ± 5.8
63.0 ± 9.5
67.0 ± 5.9
0
0
0
One
T wo
5.2 ± 1.5
9.5 ± 1.4
8.0 ± 1.1
2.2
1.2
2.0
11.5
11.8
16.0
1.3 ± 0.3
1.6 ± 0.07
1.7 ± 0.1
43.3 ± 18.3
94.0 ± 0.6
88.3 ± 2.8
0
1.6 ± 1.6
0
One
T wo
6.9 ± 1.0
8.5 ± 0.4
10.3 ± 2.1
1.7
1.2
1.5
11.9
10.2
15.1
1.4 0.1
1.6 ± 0.07
1.8 0.2
40.3 ± 13.5
83.3 ± 3.8
65.7 ± 8.4
1.8 ± 1.8
0
4.0 ± 2.3
n = 3 sites per category
a Number of species estimated to be present in each treatment plot based on the Chao 2 estimator
b Represents change in species diversity along a gradient. Determined as β = γ / α
c Number of species estimated to be present within a treatment level based on the Chao 2 estimator
d Derived using the Shannon-Wiener diversity index
e Derived using Morisita's index. Values were obtained by comparing index for a given treatment against other replicates
f Number of species not found elswhere as a percentage of the landscape total
Figure 3.3 Species accumulation curves for amphibians and reptiles, A) pre-treatment, B)
post-treatment year one, and C) post-treatment year two.
Chao 2 estimated number of species
22
A
20
18
16
14
12
10
Control
8
Burn
6
Heavy Thin
4
Light Thin
Heavy Thin * Burn
2
Light Thin * Burn
0
0
5
10
15
20
25
30
Average number of individuals
Chao 2 estimated number of species
22
B
20
18
16
14
12
10
Control
8
Burn
6
Heavy Thin
Light Thin
4
Heavy Thin * Burn
2
Light Thin * Burn
0
0
5
10
15
20
25
30
35
40
45
50
55
60
65
Chao 2 estimated number of species
Average number of individuals
22
C
20
18
16
14
12
10
Control
Burn
8
Heavy Thin
6
Light Thin
4
Heavy Thin * Burn
Light Thin * Burn
2
0
0
5
10
15
20
25
30
Average number of individuals
35
40
45
Figure 3.4 Species rarefaction curves for amphibians only, A) pre-treatment, B) posttreatment year one, and C) post-treatment year two.
Chao 2 estimated number of species
9
Control
8
A
Burn
Heavy Thin
7
Light Thin
Heavy Thin * Burn
6
Light Thin * Burn
5
4
3
2
1
0
0
5
10
15
20
Average number of individuals
Chao 2 estimated species richness
9
B
8
7
6
5
4
Control
Burn
3
Heavy Thin
2
Light Thin
Heavy Thin * Burn
1
Light Thin * Burn
0
0
5
10
15
20
25
30
35
40
Average number of individuals
Chao 2 estimated number of species
9
C
8
7
6
5
4
Control
Burn
Heavy Thin
Light Thin
Heavy Thin * Burn
Light Thin * Burn
3
2
1
0
0
5
10
15
Average number of individuals
20
Chao 2 estimated number of species
Figure 3.5 Species rarefaction curves for reptiles only, A) pre-treatment, B) posttreatment year one, and C) post-treatment year two.
12
A
10
8
6
Control
Burn
4
Heavy Thin
Light Thin
2
Heavy Thin * Burn
Light Thin * Burn
0
0
2
4
6
8
10
12
Chao 2 estimated number of species
Average number of individuals
12
B
10
8
6
Control
4
Burn
Heavy Thin
Light Thin
2
Heavy Thin * Burn
Light Thin * Burn
0
0
2
4
6
8
10
12
14
16
18
20
22
24
26
Chao 2 estimated number of species
Number of individuals
12
C
10
8
6
Control
Burn
4
Heavy Thin
Light Thin
Heavy Thin * Burn
2
Light Thin * Burn
0
0
2
4
6
8
10
12
14
16
18
Number of individuals
20
22
24
26
Figure 3.6 Total Eastern Fence Lizard (Sceloporus undulatus) captures in managed forest
stands of the William B. Bankhead National Forest (2005-2008).
Figure 3.7 Total Green Anole (Anolis carolinensis) captures in managed forest stands of
the William B. Bankhead National Forest (2005-2008).
Figure 3.8 Total Little Brown Skink captures (Scincella lateralis) in managed forest
stands of the William B. Bankhead National Forest (2005-2008).
Figure 3.9 Total heliothermic lizard captures in managed forest stands of the William B.
Bankhead National Forest (2005-2008). Little Brown Skinks and Coal Skinks were
excluded from this analysis because they do not display heliothermic behaviors (Vitt et
al. 1998). Bottom pie chart illustrates percent composition of sampled lizard species.
Figure 3.10 Total Black Racer captures (Coluber c. constrictor) in managed forest stands
of the William B. Bankhead National Forest (2005-2008).
Figure 3.11 Total Copperhead (Agkistrodon contortrix) captures in managed forest stands
of the William B. Bankhead National Forest (2005-2008).
Figure 3.12 Total large snake captures in managed forest stands of the William B.
Bankhead National Forest (2005-2008). Bottom pie chart illustrates percent composition
of sampled snake species.
Figure 3.13 Total Mississippi Slimy Salamander (Plethodon mississippi) captures in
managed forest stands of the William B. Bankhead National Forest (2005-2008).
Figure 3.14 Total Fowler’s Toad (Bufo fowleri) captures in managed forest stands of the
William B. Bankhead national Forest (2005-2008).
Figure 3.15 Total frog captures of the genus Lithobates in managed forest stands of the
William B. Bankhead National Forest (2005-2008). Bottom pie chart illustrates percent
composition of each anuran species.
Table 3.7 Principal components analysis results of 18 original habitat variables. See
Table 3.1 for habitat variable descriptions.
%_litt
%_bare
%_herb
%_wood
%_rock
%_slash
%_cwd
can_cov
l_dep
for_1
for_2
for_3
for_4
cwd_vol
air_t
soil_t
rel_hum
%_light
Eigenvalues
Percent Variance
Cumulative Variance
1
-0.66
0.64
0.52
0.63
-0.24
0.75
0.59
-0.91
-0.63
-0.08
-0.86
-0.74
-0.54
0.29
0.88
0.86
-0.79
0.95
8.42
46.78
46.78
2
0.52
-0.55
0.69
0.59
-0.45
0.01
-0.03
-0.12
0.35
0.76
-0.14
-0.24
0.15
0.07
-0.04
0.04
0.19
-0.05
2.47
13.73
60.51
Component
3
-0.36
0.33
0.00
-0.12
-0.36
-0.08
0.59
-0.07
0.15
0.25
0.13
0.06
-0.03
0.67
-0.29
-0.32
0.20
-0.07
1.53
8.49
69.00
4
-0.12
0.15
0.29
0.32
0.55
-0.24
-0.09
0.02
-0.48
0.30
0.17
0.21
-0.28
0.11
-0.13
0.08
0.22
-0.09
1.15
6.38
75.37
5
-0.28
0.27
0.09
-0.04
0.17
0.30
0.01
-0.05
-0.04
0.24
-0.10
-0.16
0.59
-0.37
-0.18
-0.10
0.17
-0.03
0.93
5.18
80.56
Figure 3.16 Relationships of PCA axis one in 18 managed forest stands of the William B.
Bankhead National Forest, Alabama.
Figure 3.17 Relationships of PCA axis two in 18 managed forest stands of the William B.
Bankhead National Forest, Alabama.
Post-Treatment Year Two
Post-Treatment Year One
Pre-Treatment
Table 3.8 Raw habitat measurements collected from managed forest stands in the
William B. Bankhead National Forest, Alabama. See Table 3.2 for habitat variable
descriptions.
Habitat Variable
%_litt
%_bare
%_herb
%_wood
%_rock
%_slash
%_CWD
can_cov
l_dep
for_1
for_2
for_3
for_4
cwd_vol
air_t
soil_t
rel_hum
%_light
%_litt
%_bare
%_herb
%_wood
%_rock
%_slash
%_CWD
can_cov
l_dep
for_1
for_2
for_3
for_4
cwd_vol
air_t
soil_t
rel_hum
%_light
%_litt
%_bare
%_herb
%_wood
%_rock
%_slash
%_CWD
can_cov
l_dep
for_1
for_2
for_3
for_4
cwd_vol
air_t
soil_t
rel_hum
%_light
Control
99.0 ± 1.1
0.0
9.3 ± 3.2
5.1 ± 1.9
0.0
0.0
2.3 ± 1.7
88.1 ± 1.8
6.9 ± 0.4
68.1 ± 13.4
73.6 ± 13.4
54.2 ± 19.1
100 ± 0.0
76.9 ± 58.3
25.4 ± 0.2
18.8 ± 0.5
63.6 ± 2.0
111.8 ± 29.5
99.5 ± 0.5
0.0
11.2 ± 8.4
6.8 ± 1.9
0.0
0.0
1.4 ± 0.5
91.0 ± 1.8
7.4 ± 0.6
43.1 ± 28.4
83.3 ± 14.4
43.1 ± 6.4
94.4 ± 9.6
36.5 ± 28.6
29.5 ± 0.5
19.5 ± 0.6
39.6 ± 7.6
106.0 ± 30.0
98.8 ± 1.5
0.0
7.0 ± 3.7
9.6 ± 1.9
0.1 ± 0.2
0.1 ± 0.2
1.2 ± 0.4
93.6 ± 1.0
7.0 ± 0.5
54.1 ± 25.3
72.2 ± 12.0
50.0 ± 22.0
100 ± 0.0
49.8 ± 43.7
28.2 ± 0.12
20.2 ± 0.6
55.8 ± 4.0
111.2 ± 25.6
Burn
99.5 ± 0.2
0.0
13.5 ± 11.4
7.1 ± 3.7
0.3 ± 0.5
0.0
0.7 ± 0.7
88.4 ± 2.5
4.7 ± 0.8
76.4 ± 16.8
84.7 ± 9.6
72.2 ± 17.3
89.5 ± 13.0
10.2 ± 10.5
25.3 ± 0.2
18.9 ± 0.1
64.6 ± 1.4
76.2 ± 20.9
97.5 ± 1.8
1.5 ± 1.3
14.6 ± 8.7
8.6 ± 6.2
0.3 ± 0.5
0.0
1.0 ± 0.9
90.4 ± 3.8
3.7 ± 1.3
29.2 ± 4.1
87.5 ± 21.7
84.7 ± 16.8
84.7 ± 9.6
18.0 ± 16.6
29.3 ± 0.9
21.4 ± 1.9
40.7 ± 8.5
121.3 ± 68.6
97.6 ± 2.9
1.4 ± 2.4
12.7 ± 12.4
12.1 ± 5.0
0.8 ± 0.8
0.0
0.9 ± 0.8
92.0 ± 0.4
4.3 ± 1.4
43.1 ± 12.7
68.1 ± 12.7
59.7 ± 6.4
93.1 ± 8.7
15.6 ± 21.3
28.7 ± 0.9
22.1 ± 3.2
48.5 ± 11.2
116.0 ± 25.1
Light Thin
99.5 ± 0.6
0.0
20.7 ± 25.1
15.7 ± 14.0
0.0
0.8 ± 1.4
1.7 ± 1.1
80.4 ± 6.6
5.6 ± 0.5
62.5 ± 30.0
73.6 ± 6.4
73.6 ± 15.8
100 ± 0.0
97.3 ± 122.7
25.5 ± 0.3
19.2 ± 0.7
64.5 ± 1.6
71.7 ± 40.4
97.6 ± 2.0
1.1 ± 0.7
20.8 ± 14.6
17.5 ± 8.5
0.0
3.5 ± 1.7
3.0 ± 0.2
62.3 ± 2.4
5.1 ± 0.3
34.7 ± 9.6
33.3 ± 7.2
26.4 ± 4.8
66.7 ± 26.0
109.5 ± 72.0
34.8 ± 2.1
24.5 ± 1.1
30.8 ± 4.6
594.8 ± 127.2
98.7 ± 1.1
0.1 ± 0.2
31.7 ± 17.5
23.2 ± 12.9
0.0
1.9 ± 1.1
2.1 ± 1.5
68.7 ± 1.3
4.8 ± 0.4
75.0 ± 18.2
29.2 ± 11.0
19.4 ± 6.4
83.3 ± 7.2
96.4 ± 103.7
34.1 ± 1.5
24.8 ± 1.1
34.4 ± 5.4
566.2 ± 49.4
Heavy Thin
99.2 ± 0.9
0.0
26.8 ± 2.8
15.3 ± 7.2
0.0
0.8 ± 0.8
1.7 ± 1.1
83.9 ± 1.0
6.6 ± 0.8
80.6 ± 12.0
68.1 ± 20.6
40.3 ± 4.8
93.1 ± 8.7
152.4 ± 131.5
25.5 ± 0.5
19.2 ± 0.5
64.0 ± 3.0
73.4 ± 36.7
98.2 ± 0.7
0.4 ± 0.7
23.2 ± 3.8
20.7 ± 8.2
0.0
2.9 ± 1.7
2.2 ± 1.6
64.2 ± 5.6
4.8 ± 0.8
52.8 ± 19.7
22.2 ± 6.4
22.2 ± 8.7
73.6 ± 20.6
56.9 ± 69.4
35.3 ± 1.2
23.7 ± 0.6
40.8 ± 21.3
669.7 ± 13.7
98.5 ± 0.5
0.5 ± 0.5
36.8 ± 9.5
33.9 ± 13.1
0.0
1.1 ± 1.3
2.5 ± 1.4
69.8 ± 2.2
5.5 ± 0.9
73.6 ± 16.8
31.9 ± 2.4
12.5 ± 8.3
77.8 ± 12.0
82.9 ± 77.0
34.2 ± 3.7
27.0 ± 5.1
37.5 ± 8.7
528.1 ± 174.3
Light Thin*Burn
99.7 ± 0.5
0.0
22.5 ± 15.3
14.9 ± 7.3
0.0
0.1 ± 0.2
2.4 ± 2.4
83.1 ± 7.1
6.9 ± 0.7
75.0 ± 43.3
79.2 ± 11.0
69.4 ± 35.4
84.7 ± 10.5
194.6 ± 259.0
25.6 ± 0.5
18.8 ± 0.5
62.4 ± 2.4
91.7 ± 48.1
90.8 ± 3.6
7.5 ± 2.9
23.2 ± 10.4
19.4 ± 5.3
0.0
4.3 ± 3.4
3.1 ± 1.7
50.9 ± 6.5
3.2 ± 0.8
52.8 ± 10.5
23.6 ± 15.8
20.8 ± 15.8
75.0 ± 19.1
58.8 ± 31.2
32.6 ± 1.2
25.2 ± 1.5
31.1 ± 2.6
770.0 ± 248.7
94.1 ± 4.2
4.1 ± 3.5
34.3 ± 13.6
30.1 ± 14.4
0.0
2.4 ± 2.1
3.3 ± 1.1
63.4 ± 7.3
3.7 ± 0.5
77.8 ± 6.4
18.1 ± 14.6
19.4 ± 12.7
90.3 ± 2.4
55.9 ± 33.4
34.8 ± 1.7
25.6 ± 2.7
35.5 ± 6.9
718.6 ± 90.5
Heavy Thin*Burn
99.4 ± 1.0
0.0
8.4 ± 5.6
8.0 ± 2.5
0.0
0.0
1.0 ± 0.3
82.2 ± 4.8
7.2 ± 0.4
83.3 ± 16.7
72.2 ± 12.7
76.4 ± 33.7
98.6 ± 2.4
41.9 ± 29.6
25.5 ± 0.2
19.0 ± 0.3
62.5 ± 1.3
78.8 ± 32.1
91.6 ± 5.4
6.1 ± 3.4
15.7 ± 7.1
15.4 ± 5.9
0.0
5.3 ± 4.0
3.2 ± 1.9
58.9 ± 2.0
3.7 ± 0.8
37.5 ± 11.0
19.4 ± 23.7
22.2 ± 21.0
66.7 ± 36.1
137.4 ± 113.2
35.3 ± 0.8
27.6 ± 1.6
28.9 ± 0.5
792.9 ± 6.4
98.0 ± 1.1
1.1 ± 0.5
36.4 ± 12.8
25.4 ± 7.6
0.0
3.3 ± 2.6
2.2 ± 1.3
66.9 ± 10.0
3.8 ± 0.4
68.1 ± 6.4
16.7 ± 7.2
19.4 ± 19.7
87.5 ± 11.0
95.9 ± 79.9
33.3 ± 2.2
25.3 ± 1.4
37.1 ± 7.2
665.9 ± 13.6
Figure 3.18 Canonical correspondence analysis ordination plot displaying pre-treatment
A) amphibian and habitat (see Table 3.2 for definitions) relationships and B) plot and
habitat relationships. In diagram A, species are designated with four-lettered
abbreviations (see Figures 3.4, 3.5, 3.6, and 3.7). In diagram B, treatment plots are
designated as follows: C- control, B- burn, HT- heavy thin, LT- light thin, HTB- heavy
thin and burn, and LTB- light thin and burn. Habitat relationships are indicated by
arrowed lines.
Figure 3.19 Canonical correspondence analysis ordination plot displaying post-treatment
year one results for A) amphibian and habitat (see Table 3-2 for definitions) relationships
and B) plot and habitat relationships. In diagram A, species are designated with fourlettered abbreviations (see Figures 3.3, 3.4, 3.5, and 3.6). In diagram B, treatment plots
are designated as follows: C- control, B- burn, HT- heavy thin, LT- light thin, HTBheavy thin and burn, and LTB- light thin and burn. Habitat relationships are indicated by
arrowed lines.
Figure 3.20 Canonical correspondence analysis ordination plot displaying post-treatment
year two results for A) amphibian and habitat (see Table 3.2 for definitions) relationships
and B) plot and habitat relationships. In diagram A, species are designated with fourlettered abbreviations (see Figures 3,3, 3.4, 3.5, and 3.6). In diagram B, treatment plots
are designated as follows: C- control, B- burn, HT- heavy thin, LT- light thin, HTBheavy thin and burn, and LTB- light thin and burn. Habitat relationships are indicated by
arrowed lines.
Figure 3.21 Canonical correspondence analysis ordination plot displaying pre-treatment
results for A) reptile and habitat relationships and B) plot and habitat (see Table 3.2 for
definitions) relationships. In diagram A, species are designated with four-lettered
abbreviations (Figures 3.3, 3.4, 3.5, and 3.6). In diagram B treatment plots are designated
as follows: C- control, B- burn, HT- heavy thin, LT- light thin, HTB- heavy thin and
burn, and LTB- light thin and burn. Habitat relationships are indicated by arrowed lines.
Figure 3.22 Canonical correspondence analysis ordination plot displaying post-treatment
year one results for A) amphibian and habitat (see Table 3.2 for definitions) relationships
and B) plot and habitat relationships. In diagram A, species are designated with fourlettered abbreviations (Figures 3.3, 3.4, 3.5, and 3.6). In diagram B, treatment plots are
designated as follows: C- control, B- burn, HT- heavy thin, LT- light thin, HTB- heavy
thin and burn, and LTB- light thin and burn. Habitat relationships are indicated by
arrowed lines.
Figure 3.23 Canonical correspondence analysis ordination plot displaying post-treatment
year two results for A) reptile and habitat relationships and B) plot and habitat (see Table
3.2 for definitions) relationships. In diagram A, species are designated with four-lettered
abbreviations (Figures 3.3, 3.4, 3.5, and 3.6). In diagram B, treatment plots are designated
as follows: C- control, B- burn, HT- heavy thin, LT- light thin, HTB- heavy thin and
burn, and LTB- light thin and burn. Habitat relationships are indicated by arrowed lines.
Total Inertia
Total Percent Species Variance
Total Percent Species-Environment Variance
p–value
1.32
53
65.2
0.79
1.19
57.1
67.2
0.50
1.13
59.4
73.0
0.19
1.26
47.5
58.9
0.04
0.84
51.5
70.0
0.31
0.85
48.3
65.1
0.72
Table 3.10 Overall canonical correspondence analysis results describing relationships between herpetofauna and habitat
variables in managed stands of the William B. Bankhead National Forest, Alabama.
Amphibians
Reptiles
Output Results
Pre Post Year One Post Year Two Pre Post Year One Post Year Two
Axis One
Eigenvalue
0.258
0.247
0.308
0.240
0.208
0.175
Percent Species Variance
19.6
20.7
27.3
19.1
24.8
20.6
Species-Environment Variance
24.1
24.4
33.5
23.6
33.7
27.8
Axis Two
Eigenvalue
0.222
0.236
0.214
0.189
0.144
0.139
Percent Species Variance
16.9
19.9
18.9
15.0
17.2
16.4
Species-Environment Variance
20.8
23.4
23.3
18.7
23.4
22
Axis Three
Eigenvalue
0.216
0.197
0.149
0.168
0.080
0.096
Percent Species Variance
16.5
16.5
13.2
13.4
9.5
11.3
Species-Environment Variance
20.3
19.4
16.2
16.6
12.9
15.3
CHAPTER 4
LIZARD SPECIES HABITAT AND CLIMATE RELATIONSHIPS IN MANAGED
SOUTHEASTERN FORESTS USING INFORMATION-THEORETIC APPROACHES
Introduction
The increasing evidence of worldwide biodiversity declines demands that
researchers and land managers better understand the impacts of anthropogenic
disturbances on wildlife communities (Myers 1996; Cincotta et al. 2000; Sodhi et al.
2004; Stuart et al. 2004). Traditionally many studies examine some measure of
biodiversity at the beginning of the study and compare it to an end value to draw
conclusions of the impact of the disturbance. Although these studies provide important
management information, it is also necessary for land managers to be aware of the
mechanisms responsible for causing the changes in community structure (Marzluff et al.
2000). Oftentimes managers must consider the response of multiple species groups and
develop management strategies that benefit some species groups while minimizing
impacts to others. By understanding the mechanisms behind these responses, managers
will be able to adapt monitoring methods and management guidelines to mirror biological
requirements of the target species (Lindenmeyer et al. 2000; Lindenmayer et al. 2006;
Lindenmayer and Franklin 2009). I have applied information-theoretic approaches to
evaluate a series of habitat use hypotheses that best explain lizard community responses
to prescribed burning and thinning. The design of my study has permitted us to evaluate
specific habitat use hypotheses and overall response to forest management along with
changes in these patterns throughout three field seasons.
Throughout the past three decades, the standard of forest management on
federally managed lands has shifted from production forests to the recognition of the
importance of ecosystem change associated with anthropogenic alterations (Long 2009).
Prior to this, natural disturbance patterns within forest ecosystems have been greatly
altered (e.g., fire frequency and sustainable harvesting procedures), which has led to the
break-down of dynamic processes necessary for ecosystem health (Elliot et al. 1999;
Lafon et al. 2005; Land and Rieske 2006). Moreover, anthropogenic disturbance patterns
can greatly affect landscape heterogeneity, which in turn can greatly influence natural
disturbance patterns (Turner et al. 1989). Forest disturbances such as burning and canopy
removal are essential for the maintenance of these processes in forest ecosystems and
should be adapted to mimic the effects of naturally occurring disturbance patterns
(Drever et al. 2006).
Sustainable forestry, which aims to meet the needs of forest resources while
maintaining the ecosystem integrity to conserve biodiversity, has been adopted as the
current paradigm of forest management (Gustafson et al. 2007). With nearly 39 million
hectares (19%) of all timberland managed by the National Forest Service (Smith et al.
2001), there is a heightened need to evaluate forest management effects on forest
ecosystems. Forest structure may be greatly altered through forest management practices
and it is likely that multiple species may display differential responses to these
disturbances. Vertebrate distribution and overall species diversity patterns can be linked
to a multitude of factors (e.g., predation, competition, and environmental gradients), but
landscape disturbances function as an overarching regulator of these factors (Petraitis et
al. 1989).
I chose to examine lizard response to forest management, because 1) reptiles tend
to demonstrate species-specific responses to habitat alteration (Greenberg et al. 1994; Vitt
et al. 1998; Barrett and Guyer 2008); 2) many reptile species that were once common
have become increasingly rare (Gibbons et al. 2000); and 3) outside of temperature
relationships, very little is understood regarding lizard and habitat relationships (Smith
and Ballinger 2001). A majority of forest management studies tend to compare only the
overall organismal response in relation to forest stand condition and do not examine the
mechanisms (e.g., habitat change) that are responsible for causing changes in the
organismal community (Marzluff et al. 2000). I examined the relationships between
forest structural changes and reptile community response in managed forest stands. I
hypothesized that the lizards examined in this study would express species-specific
responses to prescribed burning and thinning. For example, basking lizard species would
be associated with increase gradients and would increase in thinned plots, whereas
species associated with the forest floor would be negatively affected due to litter layer
disturbances.
Methods
Study Site Description
My study was centered in the northern portion of the William B. Bankhead
National Forest (BNF), located in Lawrence, Winston, and Franklin Counties, of
northwestern Alabama. Bankhead National Forest is a 72,800 ha multi-use forest located
along the highly dissected portion of the southern Cumberland Plateau (Smalley 1982;
Gaines and Creed 2003). Soils within this region are typically composed of HartsellsRock and limestone-Hector (Smalley 1982). Mixed forests of the southern Cumberland
Plateau tend to be dominated by oak-hickory forest types (McWilliams 1991) except in
areas where pines were actively planted for commercial purposes. Loblolly Pine (Pinus
taeda L.) was used to re-establish forest conditions in abandoned agricultural and heavily
timbered areas (Gaines and Creed 2003). Reforestation efforts along with natural growth
have resulted in 31,600 ha of loblolly pine throughout BNF (Gaines and Creed 2003). For
the past decade, Southern Pine Beetle (Dendroctonus frontalis Zimmermann) infestations
have affected Loblolly Pine stands, producing large numbers of standing dead trees and
increased fuel loads, elevating the risk of damaging wildfires. Because canopy removal
and fire disturbance have been prevented in forests throughout the study area for decades,
the BNF initiated a Forest Restoration Plan to reduce wildfire risk and promote natural
forest growth through tree thinning and prescribed fire disturbance. The BNF has not
traditionally utilized prescribed fire as a management tool, but has opted to include
prescribed burning in the forest restoration plan due to administrative recommendations.
Forest restoration plans in BNF mirror regulations set forth in the Healthy Forest
Restoration Act, which authorizes advanced vegetation management projects when
specified conditions (existence of insect or disease epidemic) pose a significant threat to
ecosystem health (Healthy Forest Restoration Act 2003).
The selected forest stands were generally located on upland sites composed of
loblolly pine 25-50 years of age that also possessed a considerable hardwood component
(Gaines and Creed 2003; Schweitzer and Tadessee 2004). At the time of this study, these
stands had not been recently harvested, and each stand had varying levels of damage
from the Southern Pine Beetle. Other past disturbances included the clearing of hardwood
stands throughout the region for Loblolly Pine plantations during the early 1970’s
(Gaines and Creed 2003).
Experimental Design and Forest Treatments
The Bankhead National Forest implemented the research design developed by
Alabama A&M University and Southern Research Station of USDA Forest Service as
part of a longterm examination of forest management impacts on forested ecosystems in
northwestern Alabama. The experiment consisted of a before-after, control-impact
(BACI), complete block design. The process of assigning treatments to each forest stand
was not a fully random process because forest treatment designations had to align with
the longterm management goals of the BNF. For example, forest stands assigned to
prescribed burn treatments had to be located in a portion of BNF that was designated a
burn area in the original forest restoration management plan. Forest manipulation
treatments consisted of a two by three factorial arrangement of three thinning levels (no
thin, 11 m2ha-1 residual basal area [BA], and 17 m2ha-1 residual BA) along with two burn
treatments (no burn and burn) equaling six treatments per replicate. In addition to single
thinning and burning treatments, I also evaluated the interaction of each thinning level
with a burn treatment. Each of the six treatments in this experiment was replicated three
times across the landscape (Figure 4.1), with each treatment approximately 9 ha in size.
Due to limited resources and difficulties implementing this large scale study in a
single year, treatments were blocked temporally (i.e., year). Block 1 was treated during
the summer of 2005, while block 2 and block 3 were treated during the summer and fall
of 2006. All harvesting procedures were thin-from-below and were completed at two
levels of BA retention. Thinning-from-below is a harvesting procedure that targets trees
from the lower crown classes in order to provide limited resources (e.g., water and light)
to dominant and co-dominant trees (Smith et al. 1997). All harvesting was completed by
feller bunchers and trees were harvested until the desired residual BA had been achieved.
Hardwood tree species, such as Quercus spp. and Carya spp. were preferentially retained
during the thinning as much as possible. All thinning procedures in a particular block
were completed in the same year (i.e., block one 2005 and blocks two and three 2006).
Prescribed burns consisted of low-intensity fires, which were completed during
the dormant season (January-February) when air temperatures were low and relative
humidity was high. Backing fires were initiated to ensure that prescribed burns were
limited to understory and litter layers. Prescribed burns were ignited with drip torches and
managed by the BNF U.S. Forest Service. Fires were generally low intensity and rarely
reached more than two meters in height. Once the designated area was blocked off and
ignited, fires were permitted to burn until all available fuel was consumed and the fires
diminished.
Amphibian and Reptile Sampling
I collected lizard capture data over a period of four years (2005–2008). Due to the
staggered nature of the treatments, this resulted in three total years of lizard capture data
(one year pre-treatment; two years post-treatment). I was able to collect pre-treatment
data over a period of three months (April-June 2005) for block one and six months (May
2005-August 2005 and March 2006-May 2006) for blocks two and three. Although
sampling was greater in during post-treatment surveys than pre-treatment surveys, I
constrained all analyses to trapping periods that were directly comparable between preand post-treatment data. To correct for differences in sampling effort between years, I
divided count data by the number of trap nights and multiplied this data by 1000 to
standardize data across all years.
I employed a trapping method consisting of three drift fences (aluminum flashing)
15 m in length radiating 120⁰ from a central triangular box trap (Sutton et al. In press).
Trapping units also included one large box trap at the terminus of each drift fence (three
per array) and two pitfall traps at the midpoint of each drift fence (six per array). I
installed one drift-fence array in each study plot and surveyed the same location
throughout the study. I determined trap-unit placement by dividing each study plot into
quadrants corresponding to the four cardinal directions and randomly assigned the driftfence array to one of these quadrants. After the completion of pre-treatment surveys, I
removed all drift-fence arrays to avoid damage from tree harvesting and prescribed
burning procedures. Once forest treatments were completed, I re-installed all traps in the
same location where pre-treatment surveys were completed.
I began sampling intermittently throughout March and April and began
continuous sampling by the beginning of May and ended sampling by September during
each year. During sampling periods, I opened traps one-three block(s) at a time
depending on weather conditions and manpower, with the replication number and order
of traps randomly determined a priori. I checked traps daily between 0700-1400 hours to
minimize lizard mortality. After collecting demographic data, I marked each lizard with a
treatment-specific toe-clip to assure that recaptured lizards were not counted. I released
all captured lizards on the opposite side of the drift fence in which they were captured.
Environmental Covariates
One HOBO© (Onset Computer Corp.) datalogger was installed at each trapping
array to record air temperature, soil temperature, relative humidity, and light intensity
(Table 4.1). Dataloggers were programmed to record measurements every twelve hours
starting at 2:00 PM. I used climate information from May 15-June 15 during pre- and
post-treatment surveys. My climate data were limited to this one-month period due to the
short pre-treatment sampling period for block one.
Habitat Parameters
I recorded pre- and post-treatment habitat complexity and heterogeneity data via
yearly line transect surveys at each treatment plot and completed three habitat surveys at
each trapping array during each treatment year. I determined plot placement a priori via a
random compass bearing (0-360°) and distance (30-50 m) originating from the center of
each trapping array. I restricted habitat surveys to these distances in order to avoid any
habitat disturbance created during trap installation. In order to quantify the degree of
habitat disturbance, I completed habitat surveys in approximately the same location for
each year. Each habitat survey consisted of two 20 m perpendicular transects placed
north-south and east-west from the habitat plot center. I used a two meter piece of 1.9 cm
diameter polyvinyl chloride pipe as a transect marker and recorded the presence or
absence of a suite of microhabitat variables (Table 4.1). Basal area was determined at the
stand level as an average of five 0.08 hectare plots (Table 4.1).
To evaluate the overall impact of forest disturbances, I ranked forest treatment
treatments as a continuous variable (i.e., control [1], burn [2], light thin [3], heavy thin
[4], light thin*burn [5], and heavy thin*burn [6]) to represent increased disturbance
severity (i.e., overall impact of the disturbance on the ecosystem; Pickett and White 1985;
Table 4.1). Although forest treatments did not represent an even, continuous gradient,
each progressively higher value corresponded with increased removal of vegetation (i.e.,
tree basal area and surface vegetation; Table 4.1).
Data Analysis
To characterize vegetation and environmental characteristics of the study plots, I
conducted a principal components analysis on the habitat survey data. I analyzed the preand post-treatment habitat data simultaneously to compare the resulting components
among treatment years.
The major focus of this study was to evaluate the influence of forest disturbances
and habitat change on southeastern lizard species. I used an information-theoretic (IT)
approach to evaluate lizard response to forest disturbances. Habitat models were created
based on my biological knowledge of the study organisms in addition to data supplied in
the published literature (Table 4.2). After examining correlation matrices, I only included
variables that had a < 0.80 correlation. I originally wanted to included basal area and
canopy cover into the habitat models, but found that these variables were highly
correlated with air temperature. I included air temperature in the a priori model sets due
to support from the published literature (Pianka and Vitt 2003). For each of the five
habitat hypotheses I evaluated the importance of treatment and the treatment*year
interaction equaling at total of 13 habitat models for each species. I evaluated each of the
habitat models for Eastern Fence Lizards (Sceloporus undulatus), Green Anoles (Anolis
carolinensis), large skinks, Little Brown Skinks (Scincella lateralis), and total
heliothermic lizards. Large skinks included Eastern Five-lined Skinks (Plestiodon
fasciatus) and Broad-headed Skinks (Plestiodon laticeps). Although Broad-headed
Skinks are known to use aboreal habitat more so than Five-lined Skinks, I chose to group
these two species together based on habitat use similarities (Croshaw 2008; Winne and
Gibbons 2008). Because Little Brown Skinks are relatively small and do not actively
bask and Coal Skinks (Plestiodon anthracinus) inhabit shady and mesic sites, I did not
include these lizards in total heliothermic lizard analyses (Vitt et al. 1998). I was unable
to evaluate habitat use hypotheses for Coal Skinks due to low captures.
I used Akaike’s Information Criterion (AIC) to evaluate candidate habitat models
and “select” the most parsimonious model that best explained counts of each lizard
species. For this analysis, I calculated AICc, which is necessary if the ratio of n (sample
size) / K (total number of model parameters) is less than 40 (Burnham and Anderson
2002). To detect species-specific responses, I evaluated the same 13 habitat models for
each lizard species. I used linear mixed models (SPSS v.15.0) to create maximumlikelihood estimates for each of the proposed habitat models. In the overall model I
included block as a random effect, the environmental covariates, treatment gradient, and
the treatment*year interaction as fixed effects, and year as the repeated measure. I
calculated AICc values using the formula: AICc = -2 log (L (Θ | y)) + 2K + (2K (K + 1)) /
(n – K – 1), where L (Θ | y) is equal to the maximum log-likelihood estimate derived from
the regression results, K is the total number of parameters used in each candidate model,
and n represents sample size (Burnham and Anderson 2002). Next, I calculated a ∆AICc
value for each model by subtracting each subsequent AICc value from the smallest AICc
value of the best model (Burnham and Anderson 2002). Akaike weights (ωi) were
calculated from ∆AICc values to evaluate the evidence that a particular hypothesis was
the most parsimonious model. I examined evidence ratios to evaluate the degree of
difference between the highest-supported models and calculated model-averaged
parameter estimates to reduce bias and incorporate model selection uncertainty for
models with evidence ratios <2.7 (Burnham and Anderson, 2002). I calculated parameter
estimates for candidate models with greatest support and used Akaike’s weights along
with parameter estimates to calculate standard errors and 95% confidence intervals
(Burnham and Anderson, 2002).
Results
Habitat Characteristics
Using PCA, I was able to extract five biologically relevant components that
explained 81.4% of the variance (component 1 [48.5%], component 2 [13.9%],
component 3 [8.3%], component 4 [6.1%], and component 5 [4.6%]) in the climate and
habitat dataset (Table 4.3). Component one described a dichotomy ranging from sites
with greater canopy cover, percent humidity, basal area, and litter depth to sites with
greater temperatures, light intensity, slash cover, and herbaceous growth, whereas
component two described a gradient ranging from sites with greater litter, herbaceous,
and woody vegetation cover, to sites with greater bare ground cover (Figure 4.2).
Components three and four described relationships in CWD and rock cover, respectively,
whereas component five described a gradient of overstory tree cover.
Pre-treatment habitat structure was similar among study plots and changed
drastically after forest treatments were implemented (Figure 4.2). Treatment plots
differed greatly in terms of habitat structure directly after treatment, but during the
second year, I found that litter coverage increased considerably in plots receiving a burn
treatment compared to first year post-treatment habitat results (Figure 4.2).
Lizard Captures
I captured 718 total lizards representing six species during 2862 trap nights (i.e.,
block 1 [672 total trap nights], block 2 [1134 total trap nights], and block 3 [1056 total
trap nights]) over a four-year survey period (2005-2008). The most commonly captured
lizard species were Green Anoles (n=261), Little Brown Skinks (n=165), and Eastern
Fence Lizards (n=129). Five-lined Skinks (n=97), Broad-headed Skinks (n=64), and Coal
Skinks (n=2) accounted for the remaining 23% of the total lizard captures. As stated in
the methods, lizard counts were corrected for sampling differences between years (Table
4.4).
Lizard Habitat Selection Models
Mixed model regression results indicate that the sampled lizard species were
associated with distinct habitat attributes and were differentially impacted by the forest
disturbances. Based on Akaike’s weight (ωi), the most parsimonious habitat model for
Eastern Fence Lizards contained percent CWD, CWD volume, and treatment*year
variables (Table 4.5). The highest ranking habitat models for Green Anoles both included
percent woody cover and air temperature along with treatment gradient and the
treatment*year interaction (Table 4.6). The model composed of percent CWD and CWD
volume was the most parsimonious model for large skinks, whereas the second highest
supported model was composed of the CWD variables and treatment effects (Table 4.7).
The highest supported models for Little Brown Skinks included percent forest level 3 and
litter cover, whereas lower supported models included percent woody and litter cover
and the treatment*year interaction (Table 4.8). The global model offered best support for
total heliothermic lizards, indicating that a broad set of habitat and climate characteristics
best described the distribution of these species, whereas air temperature and percent
woody and litter cover were included in models with lower support (Table 4.9).
Model-averaged parameter estimates indicate that, 1) Eastern Fence Lizards were
positively influenced by increased percent CWD cover and increasing disturbance; 2)
Green Anoles were positively associated with percent woody vegetation and increased air
temperatures; 3) large skinks were positively associated with percent CWD cover; 4)
Little Brown Skinks were not strongly associated with any of the habitat models, and 5)
total heliothermic lizards were positively associated with the disturbance gradient and
increased woody vegetation cover and air temperature (Table 4.10).
Count patterns for Eastern Fence Lizards, Green Anoles, and total heliothermic
lizards were highly correlated with several environmental variables. Eastern Fence
Lizards were positively correlated with percent CWD, but were more strongly correlated
during both post-treatment years (Figure 4.3). Green Anoles were strongly correlated
with air temperature throughout all years, but counts were highest in thin-only plots
during second year post-treatment surveys (Figure 4.4). Large skink counts were more
highly correlated with percent CWD and were relatively unaffected by forest treatments
(Figure 4.5), wheras Little Brown Skink counts had a weak positive correlation with
percent litter cover throughout all years and also a weak negative correlation with forest
level 3 cover during pre-treatment surveys (Figure 4.6). Total heliothermic lizards were
strongly correlated with air temperature the second year post-treatment, whereas there
was a consistent relationship with percent woody cover throughout all treatment years
(Figure 4.7).
Forest Treatment Impacts
Parameter estimates indicated that the forest treatments were important for
determining lizard response (Table 4.10). Eastern Fence Lizards were considerably
higher in thin*burn treatments during second year surveys (Figure 4.3), whereas Green
Anoles were consistently higher in all plots that were disturbed by thinning and reached
their highest counts during the second year surveys in thin-only plots (Figure 4.4).
Although parameter estimates were low for treatment*year interactions for Little Brown
Skinks, they showed an initial negative response in treated plots during the first posttreatment year, but appeared to be recovering from the disturbances during the second
year (Figure 4.6). Interestingly, Little Brown Skinks declined steadily in control plots
throughout all seasons (Figure 4.6). Total heliothermic lizard counts increased during
both post-treatment years, but were highest in thin*burn plots during the second year
post-treatment (Figure 4.7).
Discussion
The need to evaluate organismal response to forest disturbances has required the
implementation of large, replicated field studies. Unfortunately, many of these studies
lack a pre-treatment component, which may limit the overall inference of the results.
Over 90% of the published research analyzing herpetofaunal response to forest
disturbances lacks a pre-treatment component (Russell et al. 2004). Because there is a
great deal of variation in wildlife datasets, it is advantageous to have pre-treatment data
to provide baseline conditions to compare the overall community response after
disturbance. By having baseline data, I was able to establish that southeastern lizard
communities change drastically following the implementation of forest management
disturbances.
My analysis revealed species-specific responses of southeastern lizard species to
prescribed burning and thinning. Lizards of the genus Sceloporus have been found
associated with disturbed, open habitats possessing an availability of CWD cover
(Greenberg et al., 1994; Angert et al. 2002; James and M’Closkey 2003), and it is likely
that physical attributes (e.g., open canopy, availability of CWD, and bare ground patches)
of thin*burn sites were more suited for Eastern Fence Lizards. I found that large skinks
were commonly found in sites with an abundance of woody debris. Both Eastern Five-
lined Skinks and Broad-headed Skinks are commonly found at sites with an abundance of
woody debris (Hecnar and M’Closkey 1998; Croshaw 2008; Winne and Gibbons 2008)
and it appears these species are relatively unaffected by forest management. The
relatively high correlation of Green Anoles and total heliothermic lizards with increased
air temperature is quite interesting because the thermal landscape has been found to be
more important and complex to lizards than the physical landscape (Pianka and Vitt,
2003) and oftentimes thermal gradients dictate overall habitat use for many lizard species
(Du et al. 2006). Little Brown Skinks are commonly within forest litter layers, and my
findings indicated that this species had an initial negative response to forest management,
but recovered to pre-treatment counts by the second year post-treatment. Greenberg et al.
(1994) similarly found that Little Brown Skinks were more abundant in unmanaged,
control sites. The steady decline of Little Brown Skinks in control plots may be due to a
widespread weather event (e.g., hard freeze) during spring 2007. Although the decline of
this species was more precipitous during the first year post-treatment, species counts
recovered to near pre-treatment levels in many canopy disturbed plots. It is necessary to
understand organismal responses to large-scale disturbances in order to develop
comprehensive management plans that may lead to the longterm conservation of forest
biodiversity. Habitat models have not been evaluated for most reptile species and by
examining species habitat use patterns, researchers can gain a broad perspective of
species conservation.
Beyond species responses, it is imperative to understand the impacts of different
disturbances on the biological processes that operate within an ecosystem (Sousa 1984).
Understanding overall ecological response is essential because post-disturbance dynamics
are important for the ecosystem to maintain function of biological processes (Keitt 2008).
Stream siltation from forest disturbances has been found to alter nutrient cycling
dynamics during salmon runs (Tiegs et al. 2008) and predator-prey interactions in
amphibian communities (Kerby and Kats 1998), whereas repeated storm disturbance has
been found to alter dispersal rates of island-dwelling turtle populations (Dodd et al.
2006). Lizard populations are vital for maintaining trophic level dynamics (Spiller and
Schoener 1988; Spiller and Schoener 2007) on isolated island systems and these
dynamics are susceptible to major change following disturbance events (Schoener et al.
2004; Spiller and Schoener 2007). In some regions of the world, lizard communities
replace the traditional ecological roles of snakes and carnivorous mammals as the
primary consumers (Pianka 1973). A wide variety of lizard species were captured during
this study and lizards are usually very abundant in southeastern forests. However, the
ecological roles of lizards in forested ecosystems are poorly understood, and in order to
understand the longterm impacts of forest disturbances on these communities and the
associated ecological processes, it is paramount to first determine what ecological roles
lizards play in these ecosystems. Future research should not only examine lizard and
habitat relationships, but should also consider additional mechanisms including the
impacts of fluctuating lizard populations on arthropod population dynamics.
Information-theoretic approaches have obvious advantages to traditional
statistical methods (Burnham and Anderson 2002), but the major benefit is the ability to
evaluate multiple a priori hypotheses. Assuming that the hypotheses are well-founded
and biologically relevant, IT approaches such as AIC can be very beneficial. I developed
five habitat hypotheses that I considered adequate to gain insight into lizard responses to
forest management. By testing all species by the same hypotheses I was able to reveal
species-specific responses that resulted in a species gradient that ranged from species
associated with relatively undisturbed sites to species that were associated with
increasingly disturbed sites (see Figure 4.8). Overall, the temperature gradient was highly
correlated with total heliothermic lizard counts and appears to be one of the major
determining factors for species that rely on thermoregulation.
Many studies have found that reptiles respond positively to habitat disturbance
(Russell et al. 2004; Greenberg and Waldrop 2008) or are associated with disturbed
habitats (Barrett and Guyer 2008). However, the type and severity of the disturbance is
important to consider because the major factor influencing reptile declines is
anthropogenic habitat alteration (Gibbons et al. 2000). Although some lizard species
adapt quite readily to human-altered landscapes, many species are sensitive to habitat
alteration and other anthropogenic disturbances (Hecnar and M’Closkey 1998; Germaine
and Wakeling 2001; Pianka and Vitt 2003). Because forest management practices alter
relatively large areas, there has been much interest in the effects of these disturbances on
the overall ecosystem. In this study, Green Anoles and total heliothermic lizards
responded positively to thinning, whereas Eastern Fence Lizards responded positively to
thin*burn forest treatments. Although certain habitat and climate characteristics were
supported in the top ranked habitat models for these species, I found that treatment
variables had high support in many of these models. This indicates that the overall
treatments and the associated habitat changes were most important for influencing lizard
distributions. My results illustrate the importance of evaluating forest disturbance
response of multiple organisms in order to develop management and conservation
strategies that benefit as many species as possible.
Bibliography
Angert, A. L., D. Hutchison, D. Glossip, and J. B. Losos. 2002. Microhabitat use and
thermal biology of the collared lizard (Crotaphytus collaris collaris) and the fence
lizard (Sceloporus undulatus hyacinthinus) in Missouri glades. Journal of
Herpetology 36:23–29.
Barrett, K., and C. Guyer. 2008. Differential responses of amphibians and reptiles in
riparian and stream habitats to land use disturbances in western Georgia, USA.
Biological Conservation 141:2290–2300.
Cincotta, R. P., J. Wisnewski, and R. Engleman. Human population in the
biodiversity hotspots. Nature 404:990–992.
Croshaw, D. A. 2008. Broadhead skink, Eumeces (Plestiodon) laticeps. Pages 315–317 in
J. B. Jensen, C. D. Camp, W. Gibbons, and M. J. Elliot, editors. Amphibians and
reptiles of Georgia. The University of Georgia Press, Athens, Georgia.
Dodd, C. K. Jr., A. Ozgul, and M. K. Oli. 2006. The influence of disturbance events on
survival and dispersal rates of Florida box turtles. Ecological Applications 16:1936–
1944.
Drever, C. R., G. Peterson, C. Messier, Y. Bergeron, and M. Flannigan. 2006. Can forest
management based on natural disturbances maintain ecological resilience? Canadian
Journal of Forest Resources 36:2285–2299.
Du, W. G., L. Shou, and J. Y. Shen. 2006. Habitat selection in two sympatric Chinese
skinks, Eumeces elegans and Sphenomorphus indicus: do thermal preferences really
matter? Canadian Journal of Zoology 84:1300–1306.
Elliott, K. J., R. L. Hendrick, A. E. Major, J. M. Vose, W. T. Swank. 1999. Vegetation
dynamics after a prescribed fire in the southern Appalachians. Forest Ecology and
Management 114:199–213.
Gaines, G.D., and J. W. Creed. 2003. Forest health and restoration project. National
forests in Alabama, Bankhead National Forest Franklin, Lawrence, and Winston
Counties, Alabama. Final environmental impact statement. Management Bulletin R8–
MB 110B. 353 pp.
Germaine, S. S., and B. F. Wakeling. 2001. Lizard species distributions and habitat
occupation along an urban gradient in Tucson, Arizona, USA. Biological
Conservation 97:229–237.
Gibbons, 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: 653–656.
Greenberg, C.H., D. G. Neary, and L. D. Harris. 1994. Effect of high-intensity wildfire
and silvicultural treatments on reptile communities in sand-pine scrub. Conservation
Biology 8:1047–1057.
Greenberg, C. H. 2000. Response of reptile and amphibian communities to canopy gaps
created by wind disturbance in the southern Appalachians. Forest Ecology and
Management 148:135–144.
Greenberg, C. H., and T. A. Waldrop. 2008. Short-term response of reptiles and
amphibians to prescribed fire and mechanical fuel reduction in a southern
Appalachian upland hardwood forest. Forest Ecology and Management 255:2883–
2893.
Gustafson, E. J., D. E. Lytle, R. Swaty, and C. Loehle, C. 2007. Simulating the
cumulative effects of multiple forest management strategies on landscape measures
of forest sustainability. Landscape Ecology 22:141–156.
Healthy Forest Restoration Act. Act of December 3, 2003. 117 Stat., as amended; 16
U. S. C. 1887–1915.
Hecnar, S. J., and R. T. M’Closkey. 1998. Effects of human disturbance on five-lined
skink, Eumeces fasciatus, abundance and distribution. Biological Conservation 85:
213–222.
James, S.E., and R. T. M’Closkey. 2003. Lizard microhabitat and fire fuel management.
Biological Conservation 114:293–297.
Keitt, T. H. 2008. Coherent ecological dynamics induced by large-scale disturbance.
Nature 454:331–334.
Kerby, J. L., and L. B. Kats. 1998. Modified interactions between salamander life stages
caused by wildfire induced sedimentation. Ecology 79:740–745.
Lafon, C.W., J. A. Hoss, and H. D Grission-Mayey. 2005. The contemporary fire regime
of the central Appalachian mountains and its relation to climate. Physical Geography
26:126–146.
Land, A. D., and L. K. Rieske. 2006. Interactions among prescribed fire, herbivore
pressure and shortleaf pine (Pinus echinata) regeneration following southern pine
beetle (Dendroctonus frontalis) mortality. Forest Ecology and Management. 235:260–
269.
Lindenmayer, D. B., C. R. Margules, and D. B. Botkin. 2000. Indicators of biodiversity
for ecologically sustainable forest management. Conservation Biology 14:941–950.
Lindenmayer , D. B., J. F. Franklin, and J. Fischer. 2006. General management principles
and a checklist of strategies to guide forest biodiversity conservation. Biological
Conservation 131:433–445.
Lindenmayer, D. B., and G. E. Likens. 2009. Adaptive monitoring: a new paradigm
for long-term research and monitoring. Trends in Ecology and Evolution 24:482–486.
Long, J. N. 2009. Emulating natural disturbance regimes as a basis for forest
management: a North American view. Forest Ecology and Management 257:1868–
1873.
Marzluff, J.M., M. G. Raphael, R. Sallabanks. 2000. Understanding the effects of forest
management on avian species. Wildlife Society Bulletin 28:1132–1143.
Mushinsky, H. R. 1985. Fire and the Florida sandhill herpetofaunal community: with
special attention to responses of Cnemidophorus sexlineatus. Herpetologica 41:333–
342.
Myers, M. 1996. Environmental services of biodiversity. Proceedings of the National
Academy of Sciences 93:2764–2769.
Owens, A. K., K. R. Moseley, T. S. McCay, S. B. Castleberry, J. C. Kilgo, and W. M.
Ford. 2008. Amphibian and reptile community response to coarse woody debris
manipulations in upland loblolly pine (Pinus taeda) forests. Forest Ecology and
Management 256:2078–2083.
Petraitis, P. S., R. E. Latham, R. A. Niesenbaum. 1989. The maintenance of species
diversity by disturbance. The Quarterly Review of Biology 64:393–418.
Pianka, E. R. 1973. The structure of lizard communities. Annual Review of Ecology and
Systematics 4:53–74.
Pianka, E. R., and L. J. Vitt. 2003. Lizards: windows to the evolution of diversity.
University of California press, Berkley, California. 333 pp.
Ross, B., T. Fredericksen, E. Ross, W. Hoffman, M. L. Morrison, J. Beyea, M. B. Lester,
B. N. Johnson, and N. J. Fredericksen. 2000. Relative abundance and species richness
of herpetofauna in forest stands in Pennsylvania. Forest Science 46:139–146.
Russell, K. R, T. B. Wigley, W. M. Baughman, H. G. Hanlin, and W. M. Ford. 2004.
Pages 319–334 in M. H. Rauscher and K. Johnsen, editors. Responses of southeastern
amphibians and reptiles to forest management: a review. Southern forest science: past,
present, and future. United States Department of Agriculture Southern Research
Station, General Technical Report SRS–75.
Ruthven, D. C., III, R. T. Kazmaier, and M. W. Janis. 2008. Short-term response of
herpetofauna to various burning regimes in the south Texas plains. The Southwestern
Naturalist 53:480–487.
Schoener, T. W., D. A. Spiller, and J. B. Losos. 2004. Variable ecological effects of
hurricanes: the importance of seasonal timing for survival of lizards on Bahamian
islands. Proceedings of the National Academy of Sciences 101:177–181.
Schweitzer, C. J. and W. Tadesse. 2004. The interaction of fire frequency and stand
density on plant community dynamics. Pages 1–14 in CREST–Center for Forest
Ecosystem Assessment (CFEA) Alabama A&M University. National Science
Foundation proposal.
Smalley, G. W. 1982. Classification and evaluation of forest sites on the mid-Cumberland
Plateau. U.S. Department of Agriculture Forest Service, General Technical Report
S0–38. Southern Forest Experimental Station, New Orleans, Louisiana. 58 pp.
Smith, D. M., B. C. Larson, M. J. Kelty, and P. M. S. Ashton. 1997. The practice of
silviculture: Applied forest ecology. John Wiley and Sons Inc. 537 pp.
Smith, G. R., and R. E. Ballinger. 2001. The ecological consequences of habitat and
microhabitat use in lizards: a review.
(http://www.contemporaryherpetology.org/ch/2001/3/.)
Smith, W. B., J. S. Vissage, D. R. Darr, and R. M. Sheffield. 2001. Forest Resources of
the United States, 1997. U.S. Department of Agriculture, Forest Service. General
Technical Report, NC-219. 190 pp.
Sodhi, N. S., L. P. Koh, B. W. Brook, and P. K. L. Ng. 2004. Southeast Asian
biodiversity: an impending disaster. Trends in Ecology and Evolution 19:654–660.
Sousa, W. P. 1984. The role of disturbance in natural communities. Annual Review of
Ecology and Systematics 15:353–391.
Spiller, D. A., and T. W. Schoener. 1988. An experimental study of the effect of lizards
on web-spider communities. Ecological Monographs 58:58–77.
Spiller, D. A., and T. W. Schoener. 2007. Alteration of island food-web dynamics
following major disturbances by hurricanes. Ecology 88:37–41.
Tiegs, S. D., D. T. Chaloner, P. Levi, J. Rüegg, J. L. Tank and G. A. Lamberti. 2008.
Timber harvest transforms ecological roles of salmon in southeast Alaska rain forest
streams. Ecological Applications 18:4–11.
Todd, B.D., and K. M. Andrews. 2007. Response of a reptile guild to forest harvesting.
Conservation Biology 22:753–761.
Turner, M. G., R. H. Gardner, V. H. Dale, and R. V. O’Neill. 1989. Predicting the spread
of disturbance across heterogeneous landscapes. Oikos 55:121–129.
Vitt, L. J., T. C. S. Avila-Pires, J. P. Caldwell, and V. R. L. Oliveira. 1998. The impact of
individual tree harvesting on thermal environments of lizards in Amazonian rain
forests. Conservation Biology 12:654–664.
White, P. S., and S. T. A. Pickett. 1985. Natural disturbance and patch dynamics: an
introduction. Pages 1–9 in S. T. A. Pickett and P. S. White, editors. Ecology of natural
disturbance patch dynamics. Academic Press, Orlando, Florida.
Winne, C. T., and W. Gibbons. 2008. Five-lined skink, Eumeces (Plestiodon) fasciatus.
Pages 308–311 in J. B. Jensen, C. D. Camp, W. Gibbons, and M. J. Elliot, editors.
Amphibians and reptiles of Georgia. The University of Georgia Press, Athens,
Georgia.
Figure 4.1 Locations of study sites within the northwestern portion of the William B.
Bankhead National Forest, Alabama, U.S.A. Forest treatments consisted of the six
following categories: C-control, B-burn, HT-heavy thin, LT-light thin, HTB-heavy thin
and burn, and LTB-light thin and burn. Numbers after treatment abbreviations correspond
to block number.
HT3
HTB1
LT3
LT2
C3
B3
HT2
HTB2 HT1
C2
LT1
C1
LTB2
HTB3
LTB1
LTB3
B1
B2
Bas_area Cross-sectional area of trees within a 0.08 hectare plot. Overall plot value (m /ha) taken as average of five 0.08 hectare plots determined by measuring DBH of all trees 15 cm and larger.
Treat
Forest treatments ranked to represent increased basal area and vegetation removal (i.e., 1- Control, 2-Burn, 3-Light Thin, 4- Heavy Thin, 5-Light Thin*Burn, and 6-Heavy Thin*Burn)
Basal Area
Treatment Rank
2
Calculated as volume of a cylinder (m ) for each enumerated CWD (see text).
Determined by measuring depth of the substrate to the nearest 0.5 cm with a metric ruler measured at every 2 m.
Estimated with a spherical densiometer as the sum percentage of open points subtracted from 100% measured at every 5 m.
Percent coverage of forest levels ≤ 2m (classified as ground cover) measured at every 5 m.
Percent coverage of forest levels > 2 m – ≤ 4 m (classified as understory) measured at every 5 m.
Percent coverage of forest levels > 4 m – ≤ 6 m (classified as midstory) measured at every 5 m.
Percent coverage of forest levels > 6 m (classified as overstory) measured at every 5 m.
Average daily air temperature (⁰C) during the month of June recorded at 2:00 PM
Average daily soil temperature (⁰C) during the month of June recorded at 2:00 PM
Average daily light intensity (%) during the month of June recorded at 2:00 PM
Average daily relative humidity (%) during the month of June recorded at 2:00 PM
3
CWD_vol
L_dep
Can_cov
For_lev1
For_lev2
For_lev3
For_lev4
Air_temp
Soil_temp
Light
Hum
CWD volume
Litter Depth
Canopy Cover
Forest Level 1
Forest Level 2
Forest Level 3
Forest Level 4
Air Temperature
Soil Temperature
Light Intensity
Relative Humidity
Habitat Description
Presence (%) of ground cover such as leaves or small woody debris measured at every 0.5 m.
Absence (%) of ground cover (e.g., exposed soil) measured at every 0.5 m.
Presence of non-woody stems (%) such as grasses, ferns, and Smilax and Vitus sp. measured at every 0.5 m.
Presence of any woody stems (%) such as seedlings and large trees (living or dead) measured at every 0.5 m.; woody stems taller than one meter had to contact transects directly to be counted
Presence of rocky substrate (%) greater than 10 cm in size measured at every 0.5 m.
Presence of any fallen woody debris larger than 10 cm in diameter (must touch the ground somewhere along the length to be counted) measured at every 0.5 m.
Presence of any woody debris (%) composed of two or more stems 30 cm or higher from the ground (e.g., fallen treetops) measured at every 0.5 m.
Code
%_litt
%_bare
%_herb
%_woody
%_rock
%_CWD
%_slash
Habitat Variable
Percent Litter
Percent Bare Ground
Percent Herbaceous
Percent Woody
Percent Rock
Percent CWD
Percent Slash
Table 4.1 Habitat and environmental characteristics measured at each study plot in the William B. Bankhead National
Forest before and after thinning and burning disturbances (2005-2008).
Model Terms
%_Litt + %_Woody + Treat
%_CWD + CWD_Vol + Treat
Air_Temp + %_Woody + Treat
For_3 + %_Litt + Treat
%_Litt + %_Woody + %_CWD + CWD_vol +
Air_Temp + For_3 + Treat
Justification
Variability of ground cover will differentially affect each lizard species (Mushinsky, 1985; Greenburg, 1994; Ruthven, et al. 2008)
Increased structural diversity may provide perch and cover sites for lizards (James and M'Closkey, 2003; Owens, et al. 2008)
Increased air temperature and habitat heterogeneity will lizards (Vitt et al. 1998; Pianka and Vitt 2003)
Canopy gaps and litter cover variation will benefit each lizard species (Vitt et al. 1988; Greenburg 2000)
Global model
Table 4.2 A-priori hypotheses to evaluate habitat use and forest management impacts on southeastern lizard species in the
Bankhead National Forest (2005–2008).
Table 4.3 Raw habitat and climate values for variables used in PCA and AIC analyses to
develop habitat models for southeastern lizard species in the William B. Bankhead
National forest before and after prescribed burning and thinning (2005-2008).
Habitat Variable
%_ litt
Treatment Year
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
Control
99.0 ± 1.1
99.5 ± 0.5
98.8 ± 1.5
Burn
99.5 ± 0.2
97.5 ± 1.8
97.6 ± 2.9
Light Thin
99.2 ± 0.9
98.2 ± 0.7
98.5 ± 0.5
Heavy Thin
99.5 ± 0.6
97.6 ± 2.0
98.7 ± 1.1
%_bare
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
0.0
0.0
0.0
0.0
1.5 ± 1.3
1.4 ± 2.4
0.0
0.4 ± 0.7
0.5 ± 0.5
0.0
1.1 ± 0.7
0.1 ± 0.2
0.0
6.1 ± 3.4
1.1 ± 0.5
0.0
7.5 ± 2.9
4.1 ± 3.5
%_herb
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
9.3 ± 3.2
11.2 ± 8.4
7.0 ± 3.7
13.5 ± 11.4
14.6 ± 8.7
12.7 ± 12.4
26.8 ± 2.8
23.2 ± 3.8
36.8 ± 9.5
20.7 ± 25.1
20.8 ± 14.6
31.7 ± 17.5
8.4 ± 5.6
15.7 ± 7.1
36.4 ± 12.8
22.5 ± 15.3
23.2 ± 10.4
34.3 ± 13.6
%_wood
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
5.1 ± 1.9
6.8 ± 1.9
9.6 ± 1.9
7.1 ± 3.7
8.6 ± 6.2
12.1 ± 5.0
15.3 ± 7.2
20.7 ± 8.2
33.9 ± 13.1
15.7 ± 14.0
17.5 ± 8.5
23.2 ± 12.9
8.0 ± 2.5
15.4 ± 5.9
25.4 ± 7.6
14.9 ± 7.3
19.4 ± 5.3
30.1 ± 14.4
%_rock
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
0.0
0.0
0.1 ± 0.2
0.3 ± 0.5
0.3 ± 0.5
0.8 ± 0.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
%_cwd
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
2.3 ± 1.7
1.4 ± 0.5
1.2 ± 0.4
0.7 ± 0.7
1.0 ± 0.9
0.9 ± 0.8
1.7 ± 1.1
2.2 ± 1.6
2.5 ± 1.4
1.7 ± 1.1
3.0 ± 0.2
2.1 ± 1.5
1.0 ± 0.3
3.2 ± 1.9
2.2 ± 1.3
2.4 ± 2.4
3.1 ± 1.7
3.3 ± 1.1
%_slash
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
0.0
0.0
0.1 ± 0.2
0.0
0.0
0.0
0.8 ± 0.8
2.9 ± 1.7
1.1 ± 1.3
0.8 ± 1.4
3.5 ± 1.7
1.9 ± 1.1
0.0
5.3 ± 4.0
3.3 ± 2.6
0.1 ± 0.2
4.3 ± 3.4
2.4 ± 2.1
cwd_vol
Pre-Treatment
76.9 ± 58.3 10.2 ± 10.5 152.4 ± 131.5 97.3 ± 122.7
Post-Treatment Year One 36.5 ± 28.6 18.0 ± 16.6 56.9 ± 69.4 109.5 ± 72.0
Post-Treatment Year Two 49.8 ± 43.7 15.6 ± 21.3 82.9 ± 77.0 96.4 ± 103.7
41.9 ± 29.6
137.4 ± 113.2
95.9 ± 79.9
194.6 ± 259.0
58.8 ± 31.2
55.9 ± 33.4
l_depth
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
6.9 ± 0.4
7.4 ± 0.6
7.0 ± 0.5
4.7 ± 0.8
3.7 ± 1.3
4.3 ± 1.4
6.6 ± 0.8
4.8 ± 0.8
5.5 ± 0.9
5.6 ± 0.5
5.1 ± 0.3
4.8 ± 0.4
7.2 ± 0.4
3.7 ± 0.8
3.8 ± 0.4
6.9 ± 0.7
3.2 ± 0.8
3.7 ± 0.5
can_cov
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
88.1 ± 1.8
91.0 ± 1.8
93.6 ± 1.0
88.4 ± 2.5
90.4 ± 3.8
92.0 ± 0.4
83.9 ± 1.0
64.2 ± 5.6
69.8 ± 2.2
80.4 ± 6.6
62.3 ± 2.4
68.7 ± 1.3
82.2 ± 4.8
58.9 ± 2.0
66.9 ± 10.0
83.1 ± 7.1
50.9 ± 6.5
63.4 ± 7.3
for_1
Pre-Treatment
68.1 ± 13.4 76.4 ± 16.8
Post-Treatment Year One 43.1 ± 28.4 29.2 ± 4.1
Post-Treatment Year Two 54.1 ± 25.3 43.1 ± 12.7
80.6 ± 12.0
52.8 ± 19.7
73.6 ± 16.8
62.5 ± 30.0
34.7 ± 9.6
75.0 ± 18.2
83.3 ± 16.7
37.5 ± 11.0
68.1 ± 6.4
75.0 ± 43.3
52.8 ± 10.5
77.8 ± 6.4
for_2
Pre-Treatment
73.6 ± 13.4 84.7 ± 9.6
Post-Treatment Year One 83.3 ± 14.4 87.5 ± 21.7
Post-Treatment Year Two 72.2 ± 12.0 68.1 ± 12.7
68.1 ± 20.6
22.2 ± 6.4
31.9 ± 2.4
73.6 ± 6.4
33.3 ± 7.2
29.2 ± 11.0
72.2 ± 12.7
19.4 ± 23.7
16.7 ± 7.2
79.2 ± 11.0
23.6 ± 15.8
18.1 ± 14.6
for_3
Pre-Treatment
54.2 ± 19.1 72.2 ± 17.3
Post-Treatment Year One 43.1 ± 6.4 84.7 ± 16.8
Post-Treatment Year Two 50.0 ± 22.0 59.7 ± 6.4
40.3 ± 4.8
22.2 ± 8.7
12.5 ± 8.3
73.6 ± 15.8
26.4 ± 4.8
19.4 ± 6.4
76.4 ± 33.7
22.2 ± 21.0
19.4 ± 19.7
69.4 ± 35.4
20.8 ± 15.8
19.4 ± 12.7
for_4
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
100 ± 0.0
94.4 ± 9.6
100 ± 0.0
89.5 ± 13.0
84.7 ± 9.6
93.1 ± 8.7
93.1 ± 8.7
73.6 ± 20.6
77.8 ± 12.0
100 ± 0.0
66.7 ± 26.0
83.3 ± 7.2
98.6 ± 2.4
66.7 ± 36.1
87.5 ± 11.0
84.7 ± 10.5
75.0 ± 19.1
90.3 ± 2.4
air_temp
Pre-Treatment
25.4 ± 0.2
Post-Treatment Year One 29.5 ± 0.5
Post-Treatment Year Two 28.2 ± 0.12
25.3 ± 0.2
29.3 ± 0.9
28.7 ± 0.9
25.5 ± 0.5
35.3 ± 1.2
34.2 ± 3.7
25.5 ± 0.3
34.8 ± 2.1
34.1 ± 1.5
25.5 ± 0.2
35.3 ± 0.8
33.3 ± 2.2
25.6 ± 0.5
32.6 ± 1.2
34.8 ± 1.7
soil_temp
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
18.9 ± 0.1
21.4 ± 1.9
22.1 ± 3.2
19.2 ± 0.5
23.7 ± 0.6
27.0 ± 5.1
19.2 ± 0.7
24.5 ± 1.1
24.8 ± 1.1
19.0 ± 0.3
27.6 ± 1.6
25.3 ± 1.4
18.8 ± 0.5
25.2 ± 1.5
25.6 ± 2.7
light_int
Pre-Treatment
111.8 ± 29.5 76.2 ± 20.9 73.4 ± 36.7
71.7 ± 40.4
Post-Treatment Year One 106.0 ± 30.0 121.3 ± 68.6 669.7 ± 13.7 594.8 ± 127.2
Post-Treatment Year Two 111.2 ± 25.6 116.0 ± 25.1 528.1 ± 174.3 566.2 ± 49.4
78.8 ± 32.1
792.9 ± 6.4
665.9 ± 13.6
91.7 ± 48.1
770.0 ± 248.7
718.6 ± 90.5
%_hum
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
63.6 ± 2.0
39.6 ± 7.6
55.8 ± 4.0
64.6 ± 1.4
40.7 ± 8.5
48.5 ± 11.2
64.0 ± 3.0
40.8 ± 21.3
37.5 ± 8.7
64.5 ± 1.6
30.8 ± 4.6
34.4 ± 5.4
62.5 ± 1.3
28.9 ± 0.5
37.1 ± 7.2
62.4 ± 2.4
31.1 ± 2.6
35.5 ± 6.9
basal_ar
Pre-Treatment
Post-Treatment
28.9 ± 0.2
30.3 ± 0.2
27.6 ± 3.2
29.5 ± 2.9
27.9 ± 1.4
14.6 ± 1.2
29.3 ± 2.6
11.9 ± 0.9
29.1 ± 3.0
13.8 ± 0.4
26.8 ± 1.3
10.9 ± 0.8
18.8 ± 0.5
19.5 ± 0.6
20.2 ± 0.6
Light Thin*Burn Heavy Thin*Burn
99.4 ± 1.0
99.7 ± 0.5
91.6 ± 5.4
90.8 ± 3.6
98.0 ± 1.1
94.1 ± 4.2
Figure 4.2 Multivariate ordination of habitat and climate measurements to characterize
study plots using PCA. Each plot corresponds to a different year, A) pre-treatment, B)
post-treatment year one, and C) post-treatment year two.
Table 4.4 Total lizard captures in 18 forest stands of the William B. Bankhead National
Forest (2005-2008). Results have been standardized by dividing total captures by total
number of trap nights and multiplied by 1000 to correct for trap night differences.
Species and Treatment Year
Sceloporus undulatus
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
Control
Burn
Heavy Thin
Light Thin
Heavy Thin*Burn Light Thin*Burn
2.8 ± 2.8
1.9 ± 1.9
0.7 ± 0.7
0.0
2.0 ± 1.0
5.6 ± 2.7
8.3 ± 8.3
9.1 ± 4.1
5.6 ± 1.5
6.5 ± 6.5
10.0 ± 5.6
7.7 ± 3.2
5.6 ± 2.8
9.2 ± 0.4
21.4 ± 6.5
0.0
10.2 ± 0.6
18.3 ± 3.8
Anolis carolinensis
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
4.4 ± 2.4
9.2 ± 4.8
10.1 ± 7.1
0.0
0.0
2.1 ± 1.3
15.5 ± 5.8
22.2 ± 6.9
23.9 ± 14.5
8.6 ± 6.3
22.5 ± 3.4
32.1 ± 16.5
13.9 ± 7.3
30.8 ± 7.6
21.1 ± 3.9
6.3 ± 4.4
29.1 ± 5.6
24.3 ± 14.6
Plestiodon sp.
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
11.4 ± 3.4
7.5 ± 2.0
4.1 ± 0.1
11.6 ± 2.6
0.0
3.5 ± 1.9
10.2 ± 3.6
13.6 ± 3.6
8.2 ± 2.0
7.4 ± 1.7
11.9 ± 6.8
12.5 ± 3.5
15.3 ± 7.7
12.6 ± 3.7
13.1 ± 4.8
9.7 ± 5.0
13.0 ± 5.4
11.6 ± 4.0
Plestiodon anthracinus
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
0.0
0.8 ± 0.8
1.4 ± 1.4
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
15.1 ± 7.6
5.3 ± 1.1
15.6 ± 4.6
20.3 ± 11.9
3.1 ± 1.9
8.2 ± 2.0
21.6 ± 14.8
4.2 ± 2.9
6.9 ± 1.5
12.7 ± 6.2
2.8 ± 0.3
13.7 ± 5.5
11.6 ± 2.6 34.1 ± 12.8
2.8 ± 0.3 44.8 ± 8.2
11.2 ± 5.3 38.3 ± 15.5
25.3 ± 8.5
44.4 ± 8.6
52.3 ± 16.1
34.7 ± 17.7
52.6 ± 11.5
55.6 ± 13.9
18.8 ± 3.2
47.5 ± 12.4
54.3 ± 12.6
a
Scincella lateralis
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
31.4 ± 11.5 7.2 ± 4.9
13.4 ± 3.1 3.1 ± 3.1
7.8 ± 4.4
4.8 ± 3.8
b
Total Lizards
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
a
18.5 ± 5.7
18.6 ± 4.4
15.0 ± 6.0
Counts include P. lateralis and P. fasciatus
b
Counts include S. undulatus , A. carolinensis , P. lateralis , and P. fasciatus
e
d
c
b
a
Model
%_CWD + CWD_vol + Treat*Year
%_CWD + CWD_vol + Treat + Treat*Year
%_Litt + %_Woody
%_Litt + %_Woody + Treat*Year
Air_Temp + %_Woody + Treat*Year
%_Litt + %_Woody + Treat + Treat*Year
For_3 + %_Litt + Treat*Year
Air_Temp + %_Woody
Air_Temp + %_Woody + Treat + Treat*Year
%_CWD + CWD_vol
For_3 + %_Litt + Treat + Treat*Year
Global
For_3 + %_Litt
a
335.53
334.52
342.94
341.15
343.02
341.14
344.33
347.61
342.94
349.55
344.33
331.62
354.30
–2 log likelihood
b
9
10
8
9
9
10
9
8
10
8
10
14
8
K
c
AICc
357.62
359.64
362.14
363.24
365.11
366.26
366.42
366.81
368.06
368.75
369.45
370.39
373.50
Akaike's weight value indicates relative weight of each model. Higher values indicate models with better support.
The difference between the best supported AIC model and each candidate model.
Akaike's information criterion adjusted for small samples.
Number of parameters in each model.
Value derived from regression output.
Species
Eastern Fence Lizard
d
∆AICc
0.00
2.02
4.52
5.62
7.49
8.64
8.80
9.19
10.44
11.13
11.83
12.77
15.88
e
ωi
0.62
0.23
0.07
0.04
0.01
0.01
0.01
0.01
0.00
0.00
0.00
0.00
0.00
Table 4.5 Model selection results evaluating habitat use and response of Eastern Fence Lizards to prescribed burning
and thinning in the William B. Bankhead National Forest (2005-2008).
e
d
c
b
a
Model
Air_Temp + %_Woody + Treat + Treat*Year
Air_Temp + %_Woody
Air_Temp + %_Woody + Treat*Year
%_CWD + CWD_vol + Treat*Year
%_Litt + %_Woody
%_Litt + %_Woody + Treat*Year
Global
%_CWD + CWD_vol + Treat + Treat*Year
%_Litt + %_Woody + Treat + Treat*Year
For_3 + %_Litt + Treat*Year
For_3 + %_Litt + Treat + Treat*Year
For_3 + %_Litt
%_CWD + CWD_vol
a
399.83
407.40
406.93
408.99
413.48
411.35
395.07
408.95
410.93
415.03
414.18
423.29
426.38
–2 log likelihood
b
10
8
9
9
8
9
14
10
10
9
10
8
8
K
c
AICc
424.95
426.60
429.02
431.08
432.68
433.44
433.84
434.07
436.05
437.12
439.30
442.49
445.58
Akaike's weight value indicates relative weight of each model. Higher values indicate models with better support.
The difference between the best supported AIC model and each candidate model.
Akaike's information criterion adjusted for small samples.
Number of parameters in each model.
Value derived from regression output.
Species
Green Anole
d
∆AICc
0.00
1.65
4.07
6.13
7.73
8.49
8.89
9.12
11.10
12.17
14.35
17.54
20.63
Table 4.6 Model selection results evaluating habitat use and response of Green Anoles to prescribed burning and
thinning in the William B. Bankhead National Forest (2005-2008).
e
ωi
0.60
0.26
0.08
0.03
0.01
0.01
0.01
0.01
0.00
0.00
0.00
0.00
0.00
e
d
c
b
a
a
K
8
10
9
10
8
10
14
10
9
8
8
9
9
b
c
d
AICc
∆AICc
360.15
0.00
362.16
2.01
362.44
2.29
369.94
9.79
370.21
10.06
372.03
11.88
372.26
12.11
372.74
12.59
372.93
12.78
373.29
13.14
373.47
13.32
375.92
15.77
375.98
15.83
Akaike's weight value indicates relative weight of each model. Higher values indicate models with better support.
340.95
337.04
340.35
344.82
351.01
346.91
333.49
347.62
350.84
354.09
354.27
353.83
353.89
–2 log likelihood
The difference between the best supported AIC model and each candidate model.
Akaike's information criterion adjusted for small samples.
Number of parameters in each model.
Value derived from regression output.
Model
%_CWD + CWD_vol
%_CWD + CWD_vol + Treat + Treat*Year
%_CWD + CWD_vol + Treat*Year
%_Litt + %_Woody + Treat + Treat*Year
%_Litt + %_Woody
For_3 + %_Litt + Treat + Treat*Year
Global
Air_Temp + %_Woody + Treat + Treat*Year
%_Litt + %_Woody + Treat*Year
For_3 + %_Litter
Air_Temp + %_Woody
Air_Temp + %_Woody + Treat*Year
For_3 + %_Litt + Treat*Year
* Analysis included P. fasciatus and P. lateralis .
Species
Large Skinks*
e
ωi
0.58
0.21
0.19
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Table 4.7 Model selection results evaluating habitat use and response of large skinks to prescribed burning and thinning
in the William B. Bankhead National Forest (2005-2008).
e
d
c
b
a
Model
For_3 + %_Litt
%_Litt + %_Woody
Air_Temp + %_Woody
%_Litt + %_Woody + Treat*Year
For_3 + %_Litt + Treat*Year
%_CWD + CWD_vol
Air_Temp + %_Woody + Treat*Year
%_Litt + %_Woody + Treat + Treat*Year
For_3 + %_Litt + Treat + Treat*Year
%_CWD + CWD_vol + Treat*Year
Air_Temp + %_Woody + Treat + Treat*Year
%_CWD + CWD_vol + Treat + Treat*Year
Global
a
387.43
387.71
389.21
387.13
387.21
390.29
389.21
386.51
386.60
390.28
387.41
387.97
384.64
–2 log likelihood
b
8
8
8
9
9
8
9
10
10
9
10
10
14
K
c
AICc
406.63
406.91
408.41
409.22
409.30
409.49
411.30
411.63
411.72
412.37
412.53
413.09
423.41
Akaike's weight value indicates relative weight of each model. Higher values indicate models with better support.
The difference between the best supported AIC model and each candidate model.
Akaike's information criterion adjusted for small samples.
Number of parameters in each model.
Value derived from regression output.
Species
Little Brown Skink
d
∆AICc
0.00
0.28
1.78
2.59
2.67
2.86
4.67
5.00
5.09
5.74
5.90
6.46
16.78
e
ωi
0.29
0.25
0.12
0.08
0.08
0.07
0.03
0.02
0.02
0.02
0.02
0.01
0.00
Table 4.8 Model selection results evaluating habitat use and response of Little Brown Skinks to prescribed burning and thinning
in the William B. Bankhead National Forest (2005-2008).
e
d
c
b
a
b
14
8
10
9
10
9
8
9
10
10
9
8
8
K
c
AICc
461.88
462.43
462.85
464.93
465.11
466.45
466.70
468.88
469.45
480.18
481.18
481.64
485.73
Akaike's weight value indicates relative weight of each model. Higher values indicate models with better support.
The difference between the best supported AIC model and each candidate model.
Akaike's information criterion adjusted for small samples.
Number of parameters in each model.
Value derived from regression output.
–2 log likelihood
Species
Model
Total Heliothermic Lizards* Global
423.11
%_Litt + %_Woody
443.23
Air_Temp + %_Woody + Treat + Treat*Year
437.73
%_Litt + %_Woody + Treat*Year
442.84
%_Litt + %_Woody + Treat + Treat*Year
439.99
%_CWD + CWD_vol + Treat*Year
444.36
Air_Temp + %_Woody
447.50
Air_Temp + %_Woody + Treat*Year
446.79
%_CWD + CWD_vol + Treat + Treat*Year
444.33
For_3 + %_Litt + Treat + Treat*Year
455.06
For_3 + %_Litt + Treat*Year
459.09
%_CWD + CWD_vol
462.44
For_3 + %_Litt
466.53
* Analysis included S. undulatus , A. carolinensis , P. fasciatus, and P. lateralis .
a
d
∆AICc
0.00
0.55
0.97
3.05
3.23
4.57
4.82
7.00
7.57
18.30
19.30
19.76
23.85
e
ωi
0.33
0.25
0.20
0.07
0.07
0.03
0.03
0.01
0.01
0.00
0.00
0.00
0.00
Table 4.9 Model selection results evaluating habitat use and response of total heliothermic lizards to prescribed burning
and thinning in the William B. Bankhead National Forest (2005-2008).
Table 4.10 Model-averaged parameter estimates, standard errors, and 95% confidence
intervals for habitat, climate, and forest treatment variables in the AIC candidate models
for each lizard species group in William B. Bankhead National Forest, Alabama U.S.A.
before and after implementation of thinning and prescribed burning treatments (20052008).
Species
Eastern Fence Lizard
Variable
%_CWD
CWD_Vol
Treat
Treat*year
β ± S.E.
95% Confidence Interval
1.42 ± 0.68
0.09
2.75
0.00 ± 0.01
–0.02
0.02
–0.71 ± 0.70
–2.13
0.70
0.84 ± 0.21
0.43
1.25
Green Anole
%_Woody
Air_temp
Treat
Treat*Year
0.41 ± 0.17
2.02 ± 0.55
3.45 ± 1.23
–1.24 ± 0.73
0.08
0.94
0.96
–2.72
0.74
3.10
5.93
0.24
Large Skinks
%_CWD
CWD_Vol
Treat
Treat*year
2.22 ± 0.67
0.02 ± 0.01
1.23 ± 0.70
–0.02 ± 0.09
0.91
–0.04
–0.14
–0.19
3.53
0.04
2.60
0.15
Little Brown Skink
%_Litter
%_Woody
Air_temp
For_3
Treat*year
0.45 ± 0.30
0.01 ± 0.06
–0.52 ± 0.34
–0.01 ± 0.01
0.04 ± 0.08
–0.51
–0.03
–1.21
–0.10
–0.12
1.05
0.05
0.18
0.06
0.20
%_Litter
%_Woody
%_CWD
CWD_Vol
Air_temp
For_3
Treat
Treat*year
–1.19 ± 1.13
1.02 ± 0.27
3.19 ± 2.01
0.03 ± 0.03
1.48 ± 0.92
–0.01 ± 0.08
3.00 ± 3.14
–1.42 ± 0.90
–3.40
0.49
–0.85
–0.02
–0.32
–0.17
–0.90
–3.18
1.02
1.55
7.22
0.08
3.28
0.15
5.56
0.34
a
b
Total Heliothermic Lizards
a
Analysis included P. fasciatus and P. lateralis
b
Analysis included S. undulatus , A. carolinensis , P. fasciatus , and P. lateralis
Figure 4.3 Univariate regressions comparing Eastern Fence Lizard counts with variables
from the highest supported models evaluating habitat use and treatment response in forest
stands treated with prescribed burning and thinning in the William B. Bankhead National
Forest (2005-2008).
Figure 4.4 Univariate regressions comparing Green Anole counts with variables from the
highest supported models evaluating habitat use and treatment response in forest stands
treated with prescribed burning and thinning in the William B. Bankhead National Forest
(2005-2008).
Figure 4.5 Univariate regressions comparing large skink counts with variables from the
highest supported models evaluating habitat use and treatment response in forest stands
treated with prescribed burning and thinning in the William B. Bankhead National Forest
(2005-2008).
Figure 4.6 Univariate regressions comparing Little Brown Skink counts with variables
from the highest supported models evaluating habitat use and treatment response in forest
stands treated with prescribed burning and thinning in the William B. Bankhead National
Forest (2005-2008).
Figure 4.7 Univariate regressions comparing total heliothermic lizard counts with
variables from the highest supported models evaluating habitat use and treatment
response in forest stands treated with prescribed burning and thinning in the William B.
Bankhead National Forest (2005-2008).
Figure 4.8 Lizard species gradient as revealed by prescribed burning and thinning
treatments in the William B. Bankhead National Forest (2005-2008). Species are as
follows: A) Coal Skink (Plestiodon anthracinus), B) Little Brown Skink (Scincella
lateralis), C) Eastern Five–lined Skink (top, Plestiodon fasciatus) and Broad–headed
Skink (bottom, Plestiodon laticeps), D) Green Anole (Anolis carolinensis), and Eastern
Fence Lizard (Sceloporus undulatus)
CHAPTER 5
SNAKE RADIOTELEMETRY RESEARCH IN EASTERN NORTH AMERICA:
WHAT ARE WE PUBLISHING AND HOW CAN WE IMPROVE FUTURE
CONTRIBUTIONS?
Introduction
Eastern North America is recognized as an area greatly altered by anthropogenic
disturbances. Alterations including forest fragmentation (Saunders 1991; Murcia 1995;
Woodroffe and Ginsberg 1998), road systems (Row et al. 2007; Andrews et al. 2008),
forest homogenization (Rooney et al. 2004; Vellend et al. 2007), urbanization (McKinney
2002; Olden et al. 2006), and forest management (Marzluff et al. 2000; Russell et al.
2004) all have widely ranging biodiversity impacts and are common disturbances
throughout this region (Bailey et al. 2006; Mitchell et al. 2006a). Although biodiversity is
not as high in this region compared to other regions of the world, the high rate of habitat
alteration along with a wide diversity of habitat types makes this region an area of
conservation interest (Bailey et al. 2006; Mitchell et al. 2006a). Within this region,
geographic provinces such as the southeastern Coastal Plain have a high degree of
organismal endemicity with many of these species requiring distinct disturbance regimes
(Mitchell et al. 2006b).
In order to develop longterm conservation strategies for species of interest, it is
essential to understand the biological and ecological requirements for these species. This
task is much easier for sedentary organisms, but for mobile vertebrates, such as snakes, it
can be difficult to pinpoint basic biological characteristics, such as hibernation sites and
breeding behaviors (Reinert 1992). Transect and trap surveys provide good information
for monitoring changes in species abundance patterns (Ryan et al. 2002), but it is often
difficult to collect detailed biological information using these methods. In addition, many
snakes are difficult to study because of restricted ranges and problems with detection
which can be due to secretive behavior, small size, and / or cryptic coloration.
Radiotelemetry permits precise determination of snake movements (Beaupre and
Duvall 1998), and no other method provides the opportunity to collect detailed ecological
information from study individuals over long time periods (Reinert 1992). Fitch and
Shirer (1971) completed the first and one of the largest snake radiotelemetry studies with
67 individuals from eight species. Since this publication, great improvements have been
made in the construction of transmitter packages, transmitter implantation methods,
tracking equipment, and data analysis techniques.
Radiotelemetry has proven to be an indispensible technique to study snake
ecology and behavior. It has provided valuable insight regarding life history attributes
such as nesting sites (Plummer 1990b; Cunnington and Cebek 2005) and snake mating
behavior (Brown and Weatherhead 1999; Jellen et al. 2007). Radiotelemetry has proven
to be essential for spatial ecology studies (Johnson 2000; Dodd and Barichivich 2007;
Durbian et al. 2008) and has yielded remarkable results regarding snake habitat use
(Reinert and Zappalorti 1988b; Blouin-Demers and Weatherhead 2001b; Row and
Blouin-Demers 2006b, c). Other applications such as snake translocation programs
(Reinert and Rupert 1999; Plummer and Mills 2000; King et al. 2004) and snake foraging
studies (Reinert et al. 1984; Clark 2006a, b) are nearly impossible to complete and
evaluate without the use of radiotelemetry. Radiotelemetry has yielded important data
regarding snake ecology and behavior and it is likely that continued use of this technique
will contribute to the longterm conservation of many snake species (Cooke 2008).
Radiotelemetry has proven to be an invaluable tool to study various aspects of
snake ecology; however, I observed that many aspects of snake radiotelemetry have not
been explored. Therefore I found it necessary to evaluate trends of snake radiotelemetry
publications in Eastern North America. My purpose is not to provide an extensive review
of snake radiotelemetry results, but rather evaluate trends of snake radiotelemetry
research in eastern North America, similar to reviews published by Marzluff et al. (2000)
and Fazey et al. (2005), with the goal of providing recommendations for future studies
that utilize this technique.
Materials and Methods
Evaluation Criteria
I reviewed peer-reviewed articles on snake radiotelemetry research conducted in
eastern North America in states and provinces east of and adjacent to the Mississippi
River. I chose this geographic area because of the large amount of available literature and
the authors’ experience with snake biology in this region. I used major search engines
(e.g., BioOne, JSTOR, Biological Abstracts) along with Google Scholar to compile a
complete review of snake radiotelemetry publications in eastern North America. I made
every attempt to include all studies published in the primary literature, and only included
peer-reviewed articles due to difficulties locating studies published as technical reports.
Furthermore, I wanted an unbiased representation of what is being published in the
primary literature. I excluded thermoregulation telemetry studies unless findings were
extended to discuss the influence of thermal relationships on habitat use and spatial use
patterns. I acknowledge that thermal patterns are very important in snake ecology studies,
but this literature has been adequately evaluated elsewhere (please see, Peterson et al.
1993). I classified each study based on the following criteria: 1) study type (e.g., natural
history, spatial ecology, habitat use, translocation, foraging ecology, disturbance, genetic
analyses, overwintering ecology, and techniques); 2) snake genus; 3) journal title; 4) year
of publication; and 5) geographic location of study sites.
I classified study type (i.e., category 1) based on the major subject(s) of each
snake radiotelemetry article. My evaluation criteria were as follows: 1) natural historystudies that examined snake life history attributes (e.g., nesting site location and mating
behavior), which usually included more anecdotal findings; 2) spatial ecology-studies
that reported snake spatial data (e.g., home range estimates and movement parameters; 3)
habitat use-studies that used statistical analyses to evaluate snake habitat use; 4)
translocation-studies that evaluated the effectiveness of snake translocation efforts (e.g.,
augmentation, introduction, and repatriation); 5) foraging ecology-studies that examined
snake foraging behaviors; 6) disturbance-studies that evaluated snake response to
anthropogenic disturbances; 7) molecular analyses-studies that coupled molecular
analyses (e.g., genetic relatedness) with radiotelemetry; 8) overwintering biology-studies
that examined snake overwintering sites and behaviors; and 9) techniques-studies that
either provided new data relating to snake radiotelemetry techniques (e.g., transmitter
implantation and anesthesia methodology) or evaluations of existing techniques.
Category 1 (i.e., study type) and category 3 (i.e., journal title) included all
radiotelemetry applications; category 1 was not mutually exclusive as an article could
have been classified under more than one category (e.g., spatial ecology and habitat use).
If there was only one article per journal title, I included the journal title in the
miscellaneous journal column. Category 2 (i.e., snake genus), category 4 (i.e., year of
publication), and category 5 (i.e., geographic distribution) comparisons excluded all
technique studies and only included studies that used radiotelemetry to examine
ecological aspects for a particular snake species. Technique studies were excluded from
these analyses because several of these studies were not specific to a particular snake
genus and / or were not associated with a specific geographic location therefore making it
difficult to include them in the analysis.
To illustrate the geographic distribution of the studies, I created a map of eastern
North America and layered GPS coordinates of each publication onto the map using
ArcGIS v.9.3. I obtained GPS coordinates from each publication or when coordinates
were not provided I used the search command within Google Earth to approximate study
site locations based on verbal site descriptions. In order to examine the distribution of
radiotelemetry studies, I separated eastern North America into the following four regions:
1) Canada-all studies published in Canadian provinces; 2) northeastern United States-all
states east of Indiana and north of Kentucky and Virginia; 3) southeastern United Statesall states south of Missouri, Indiana, and West Virginia; 4) midwestern United States-all
states west of Ohio and north of Kentucky and Arkansas.
Results
General Publication Trends
I reviewed 112 peer-reviewed articles dealing with various aspects of snake
radiotelemetry from 1963 to 2010 within eastern North America (Table 5.1). These
articles were published in 32 journal titles with 26% of all articles published in the
Journal of Herpetology (Figure 5.1). Another 34% were published in Copeia,
Herpetological Review, and Herpetologica combined. Miscellaneous journals comprised
approximately 13% of all articles (Figure 5.1). Most studies focused on spatial ecology
(31%), habitat use (26%), natural history (13%), and radiotelemetry techniques (10%;
Figure 5.2). Other topics including translocation, hibernation, disturbance response,
foraging, and genetic analyses composed the remaining 20% (Figure 5.2). Nearly 55% of
all studies, excluding technique studies, used radiotelemetry to examine various
ecological aspects of the following three genera: Crotalus (21%), Pantherophis (16%),
and Sistrurus (15%; Figure 5.3). The remaining radiotelemetry studies (48%) examined
the seven following snake genera: Nerodia, Pituophis, Coluber, Agkistrodon,
Lampropeltis, Heterodon, and Opheodrys (Figure 5.3). Only one study was conducted for
each of the genera Masticophis, and Thamnophis (Figure 5.3). The majority of studies
(73%) were published between 2000 and 2008 (Figure 5.4). One year in particular (2006)
accounted for 21% of the published studies (Figure 5.4).
Studies were slightly skewed towards the southeastern (29%) and northeastern
United States (27%), with many studies isolated at a few locations (Figure 5.5). Fewer
studies were published in Canadian provinces (23%) and midwest (21%; Figure 5.5).
Although greater than one-quarter of the studies occurred in the southeast, studies were
very spotty in Mississippi, Alabama, and Georgia. Three study locations, Berks County
Pennsylvania (n=6), Ocean County New Jersey (n=5), and Queen’s University Biological
Station (n=17) Ontario accounted for 29% of all published studies (Figure 5.5).
Spatial Ecology
A total of 41 studies calculated snake home-range estimates in eastern North
America. Of these studies, 39% were completed for crotalid species, while 61% were
completed for colubrid species. Many publications (46%) solely estimated home-range
size with linear estimation techniques (e.g., minimum convex polygons). Fewer studies
(10%) used utilization distribution estimation techniques (e.g., kernel density and
harmonic mean estimators) to create home range estimations, whereas 44% applied both
linear estimation techniques and probability based techniques to estimate home range
size. Nearly 76% of spatial ecology publications provided little or no rationale behind the
selection of home range estimation technique(s). In addition, many home range estimates
for a particular species were available only from one study site. I found that home range
size varied greatly for the Eastern Massassauga Rattlesnake (Sistrurus c. catenatus) and
Ratsnake (Pantherophis sp.) depending on geographical location and year of study (Table
5.2). Half of the studies for S. c. catenatus reported home range estimates on the smaller
end of the spectrum (males: < 8 hectares (ha); females: < 3 ha), whereas the remaining
estimates were much larger (males: > 25 ha; females:3–41 ha; Table 5.2). Only one of the
studies for Pantherophis sp. produced considerably higher home range estimates (males:
25 ha; females: 12–17 ha) when compared to the other published studies (males < 10 ha;
females < 9 ha; Table 5.2).
Habitat Use and Analysis
A total of 37 habitat use studies were completed in eastern North America of
which 47% were on crotalid species and the remaining 53% were on colubrid species.
Among these, 43% examined macrohabitat characteristics and 38% examined
microhabitat characteristics only, whereas 19% investigated both as part of a multi-scale
habitat analysis. Habitat use studies incorporated a wide array of statistical techniques.
Multivariate analysis of variance (MANOVA; 32%) and univariate statistical techniques
(35%) were the most commonly used analyses for habitat use studies. Discriminant
function analysis (DFA) and multiple logistic regression were used in 22% and 16% of
the studies, respectively. The application of information theoretic approaches such as
Akaike’s Information Criterion (AIC) for developing and assessing snake habitat use
models was limited (11%) among the reviewed studies.
Natural History
Of the seventeen natural history studies, 41% reported data regarding snake
mating behavior and reproductive ecology along with characteristics of snake nesting
sites. The remaining 53% of the natural history studies examined predator and prey
interactions, snake mortality rates, and scent trailing between conspecifics.
Techniques
Fifteen studies focused on methodological aspects, mainly including transmitter
attachment methods and snake response to anesthetic substances. Technical applications
(e.g., transmitter construction methods and selection of proper receiver packages) and
techniques for analyzing radiotelemetry data techniques have been evaluated by other
researchers (see Reinert 1992; Ujvari and Korsos 2000) and will not be evaluated further
in this review. Of the studies included in this review, 33% provided descriptions of
transmitter attachment methods, whereas 27% evaluated the effect of anesthetic
substances on various snake species. Twenty-seven percent of the published studies
evaluated the effects of radiotelemetry on snakes, whereas 13% included miscellaneous
aspects of snake radiotelemetry, such as the ability to detect cryptic snake species and
methods to monitor small snake species (i.e., harmonic direction finder).
Disturbance Response
Seven studies used radiotelemetry to evaluate snake response to anthropogenic
disturbance. Of these studies, three examined the response of S. c. catenatus to smallscale anthropogenic disturbances, whereas other studies used radiotelemetry to examine
impacts of existing road systems on resident snake populations. Snake response to largescale disturbances was limited to one study that evaluated the response of S. c. catenatus
to prescribed burning and vegetation management methods.
Hibernation Biology
Eight studies used radiotelemetry to assess aspects of snake overwintering sites.
Of these studies, 75% used radiotelemetry to examine micro-and macrohabitat features
with the goal of predicting overwintering sites and to identify active overwintering sites
for rare or sensitive snake species. An additional 25% examined relationships between
genetic relatedness and spatial movements between neighboring hibernacula.
Translocation
Five studies evaluated the success of translocation programs in which
radiotelemetry was used in three studies to assess snake response to augmentation efforts
(release of individuals into an area currently inhabited by the respective species: Reinert
1991). Introduction (release of individuals into an area not previously inhabited by the
respective species: Reinert 1991) and repatriation (intentional release of animal into an
area no longer inhabited by the respective species: Reinert 1991) efforts made up the
remaining 40% of translocation studies.
Foraging Ecology
Four studies used radiotelemetry to examine snake foraging behavior. Three of
these studies examined foraging events and post-feeding behaviors for the Timber
Rattlesnake (Crotalus horridus), whereas the remaining study related habitat use to
feeding condition for the Eastern Ratsnake (Pantherophis alleghaniensis).
Molecular Analyses
Four studies coupled radiotelemetry with genetic analyses. Two of these studies
examined genetic relatedness of C. horridus among communal hibernacula, whereas the
other two examined relationships of reproductive success and mate selection for the
Common Watersnake (Nerodia s. sipedon) and seasonal fluctuations in hormone levels of
C. horridus in response to reproductive and environmental stressors.
Discussion
Overall review of snake radiotelemetry results
The use of radiotelemetry in snake research has provided important insight into
snake ecology and behavior. Snake biologists have used radiotelemetry to examine a
wide variety of snake species in eastern North America. However, over half of the studies
(55%) concentrated on representatives of three genera (i.e., Crotalus, Sistrurus, and
Pantherophis). Crotalus and Sistrurus might be commonly used because both genera
possess species that are declining throughout their respective ranges and the conservation
status has made it necessary to understand factors leading to declines of these species.
Additionally, biological aspects (e.g., sit-and-wait predation strategy and site fidelity)
make these two genera ideal subjects for evaluating hypotheses about habitat and space
use (Beaupre and Duvall 1998). Because many aspects of snake ecology remain
unexplored for common species, it is important to expand radiotelemetry applications to
include a broader range of species.
The prevalence of spatial ecology and habitat use studies is not surprising due to
the relative ease with which location data can be acquired during a radiotelemetry studies.
Home range estimation (e.g., activity area and area use estimation techniques) and habitat
use techniques (e.g., use-availability studies) are well-developed and have yielded
important information for many snake species (Reinert 1992; Reinert 1993). Spatial and
habitat use data provide information that is crucial for managing wild snake populations.
In light of the evidence of reptile declines (Gibbons et al. 2000), it is surprising that more
studies have not evaluated snake response to anthropogenic and stochastic disturbances.
Future studies would benefit by extending study objectives beyond spatial ecology and
habitat use studies because these additional data are necessary for the longterm
conservation for many snake species. Seigel (1993) suggests that when technique,
question, and study animal are well matched, researchers have the potential to make
important discoveries regarding snake ecology.
The majority of snake radiotelemetry publications were published in the Journal
of Herpetology, Copeia, and Herpetologica, which is not surprising considering the
herpetological scope of these journals. Radiotelemetry represents one of the most
important tools for herpetologists to examine multiple aspects of ecology, such as mating
and nesting biology, predator / prey interactions, and interactions between spatial and
habitat use data. Researchers should consider coupling multiple ecological aspects in
order to increase opportunities to publish research results in broad ecological journals.
Almost three quarters of the reviewed studies were published between 2000 and
2010, though nearly one third of the studies published during this period were published
in 2006. Major improvements have been made in the construction and availability of
transmitter and receiver packages over the past decade. These improvements have made
it possible to not only monitor snakes for multiple seasons, but also examine juvenile and
smaller snake species (Plummer 1990a, b; Jellen and Kowalski 2007).
A greater number of studies were completed in the southeastern and northeastern
United States. Nearly one-third of the published studies were completed at three study
sites, reflecting the importance of longterm study sites for understanding snake ecology
and behavior. Without these sites, a large portion of the published snake radiotelemetry
data would not exist. Although 26% of the reviewed radiotelemetry studies were in the
southeast, there are limited studies in the states of Mississippi, Alabama, Georgia, and
Florida. This is startling because the southeast possesses the greatest snake species
richness in eastern North America and many snake species are either uncommon or of
conservation concern (Bailey et al. 2006).
Understanding Factors that Influence Snake Movements
Many factors influence individual snake movement patterns and overall spatial
ecology (Macartney et al. 1988) including thermal environment, prey availability (Clark
2006a), reproductive condition (Plummer and Mills 2000; Blouin-Demers and
Weatherhead 2002b; Gerald et al. 2006b; Marshall 2006; Waldron 2006a, b), denning
sites (Brown et al. 1982), refuge sites (Himes 2000; Dodd and Barichavich 2007), and
site-specific habitat characteristics (Durner and Gates 1993; Gerald et al. 2006a; Wund et
al. 2007). Spatial ecology studies represented the most common snake radiotelemetry
study. Researchers should consider coupling spatial data with additional data (e.g.,
habitat data) in order to better understand the nature of space use patterns. For example
Bushar et al. (1998) examined correlations between spatial patterns, hibernation site
location, and genetic relationships of C. horridus and discovered that snakes using the
same hibernacula were more closely related than snakes from nearby hibernation sites.
Due to the close proximity of basking sites to hibernation sites, females were more likely
to breed with males from the same hibernation sites. By coupling spatial data with habitat
use data, researchers can better understand landscape-scale habitat relationships (Moore
and Gillingham 2006; Row and Blouin-Demers 2006c).
Evaluating Snake Habitat Selection Patterns
Nearly 26% of the published radiotelemetry studies examined snake habitat
relationships. Many studies (35%) used univariate statistical methods to examine snake
habitat selection patterns. The sole use of univariate techniques makes it difficult to
consider habitat parameters that may be important for snake habitat selection.
Researchers should consider using multivariate techniques (e.g., DFA) in order to better
estimate the realized niche for study species (Reinert 1984a, b). The application of
information theoretic approaches to model selection, such as AIC (Burnham and
Anderson 2002), is still limited among the reviewed studies in developing and assessing
snake habitat use models. By using information-theoretic approaches researchers can use
their biological knowledge of the study organism to simultaneously evaluate multiple
habitat use models (Burnham and Anderson 2002). These approaches could also be
extended to include disturbance gradients and individual attributes (e.g., body size;
Hyslop et al. 2009) to evaluate the importance of these factors on spatial ecology, habitat
use, and mortality rates. The number of studies that evaluated habitat use patterns at
multiple scales was limited (20%). Factors affecting snake habitat selection may be scale
dependent, which may lead to different habitat selection patterns at multiple scales (Row
and Blouin-Demers 2006). For species inhabiting landscapes with uncommon habitat
attributes, multiple scale analyses may reveal habitat selection patterns that would be
missed if only one scale was examined (Waldron et al. 2008). To truly understand snake
habitat requirements, it is necessary to understand the relationship(s) between micro- and
macrohabitat use (see Harvey and Weatherhead 2006a). By incorporating within and
outside home range habitat use and landscape scale habitat use, researchers can
hierarchically evaluate snake habitat use patterns.
I found that space-use patterns can vary greatly for species examined at multiple
geographic locations. Reinert (1993) suggests that although many snake species have
characteristic space and habitat use patterns, these parameters are likely to differ across
geographic areas. Site-specific factors such as hibernation, basking, and breeding sites are
likely to influence snake movements and it is important to consider these factors when
developing conservation plans for species with broad geographic (e.g., Timber
Rattlesnakes) or spotty (e.g., Eastern Massassauga Rattlesnake) distribution patterns.
Researchers must be able to not only identify site-specific factors, but also integrate these
factors into long-term species conservation plans.
Recommendations
Upon reviewing the published literature I formulated several recommendations
for future snake radiotelemetry studies. First, studies should examine species that are not
adequately represented. Three snake species (C. horridus, S. c. catenatus, and P.
alleghaniensis) have commonly been examined in radiotelemetry studies, whereas
radiotelemetry applications are limited for species such as the Eastern Kingsnake
(Lampropeltis g. getula) and Southern Hognose Snake (Heterodon simus) even though
these species are experiencing population declines in parts of their range (Gibbons et al.
2000; Winne et al. 2007). Studies should also strive to include common species, because
information such as the location of nesting sites, clutch size, and home range estimates
are unknown for many common snake species.
Secondly, it would be useful to examine single species at multiple sites and scales
throughout their geographic range to account for site-specific differences in habitat use
and spatial ecology. For sensitive species or species with a broad geographic distribution,
employing radiotelemetry studies at multiple sites may provide data that are essential for
the conservation of a species on a regional scale or reveal distinct geographic habitat use
and spatial differences, respectively (see Reinert 1993). To facilitate home range
comparisons between studies and study sites, we recommend that researchers include
more than one home range estimation technique. Although minimum convex polygon
(MCP) estimates provide biased home range estimates, researchers should include these
estimates to permit comparisons among current and previous studies (Marshall et al.
2006). I also recommend Horne and Garton’s (2006) approach of using information
theoretic approaches to assess model fit of various probability distributions of the overall
radiotelemetry dataset to estimate home range size from the best fit model. Through this
approach, utilization distributions can be estimated at the individual level based on the
nature of the movements and the fit of the data. Researchers should also explore Horne et
al. (2008) to couple spatial patterns and habitat use data through information-theoretic
approaches.
It is apparent from this review that many aspects of snake radiotelemetry remain
unexplored. I found that a majority of snake radiotelemetry studies examined spatial
ecology and habitat use. I believe that radiotelemetry could be used to study other aspects
of snake ecology such as response to environmental disturbances. Although these
environmental changes affect snake populations in differing degrees, habitat alteration
and destruction represent the greatest challenges for snake conservation (Gibbons et al.
2000). Future radiotelemetry studies of snake response to anthropogenic disturbance
should include large-scale, replicated experimental designs to best provide
recommendations to conserve snake species inhabiting these disturbed landscapes.
Few studies have used radiotelemetry to study juveniles (but see Cobb et al. 2005;
Blouin-Demers et al. 2007; Jellen and Kowalski 2007) and many aspects of juvenile
snake biology and natural history remain unexplored. It is likely that juvenile snakes have
different spatial and habitat requirements than adult conspecifics (Jellen and Kowalski
2007). The greatest problem of radiotracking juvenile snakes is finding an effective
method to attach transmitters. Currently the only available transmitter attachment
methods are external attachment (Cobb et al. 2005; Jellen and Kowalski 2007) or surgical
implantation (Reinert and Cundall 1992). Snakes with externally attached transmitters
run the risk of becoming stuck in small spaces and transmitter packages may come
unglued. Surgical implantation is limited to snakes with a coelomic cavity large enough
to accommodate a transmitter (Blouin-Demers et al. 2007). Englesoft et al. (1999) have
developed innovative techniques (i.e., harmonic direction finder) for tracking and
monitoring juvenile snake and small snake species. Unfortunately, this method does not
permit individual snake identification and has a short detection range (Boyarski et al.
2007).
Radiotelemetry is an indispensable method to determine many aspects of snake
ecology. Including telemetry research into broad ecological programs will provide data
from free-ranging animals that otherwise may be impossible to collect (Cooke et al.
2004). This review highlights the major aspects of snake radiotelemetry in eastern North
America and provides recommendations that will improve future snake radiotelemetry
research. Basic life history information is unknown for many snake species and
radiotelemetry provides the most promising methodology to obtain these data.
Bibliography
Aird, S. D. 1986. Methoxyflurane anesthesia in Crotalus: comparisons with other gas
anesthetics. Herpteological Review 17:82–84.
Anderson, C. D., and M. Talcott. 2006. Clinical practice versus field surgery: a
discussion of the regulations and logistics of implanting radiotransmitters in snakes.
Wildlife Society Bulletin 34:1470–1471.
Anderson, C. A. 2010. Effects of movement and mating patterns on gene flow among
overwintering hibernacula of the timber rattlesnake (Crotalus horridus). Copeia
2010:54–61.
Andrews, K. M., J. W. Gibbons, and D. M. Jochimsen. 2008. Ecological effects of roads
on amphibians and reptiles: a literature review. Pages 121–143 in J. C. Mitchell, R. E.
Jung Brown, and B. Bartholomew, editors. Urban Herpetology. Herpetological
Conservation, Number 3. Salt Lake City, Utah: Society for the Study of Amphibians
and Reptiles.
Bailey, M. A., J. N. Holmes, K. A. Buhlmann, and J. C. Mitchell. 2006. Habitat
management guidelines for amphibians and reptiles of the southeastern United States.
Partners in amphibian and reptile conservation technical publication HMG-2,
Montgomery, Alabama. 88 pp.
Baxley, D. L., and C. P. Walls. 2009. Black pine snake (Pituophis melanoleucus lodingi):
spatial ecology and associations between habitat use and prey dynamics. Journal of
Herpetology 43:284–293.
Beaupre, S. J., and D. J. Duvall. 1998. Integrative biology of rattlesnakes. Contributions
to biology and evolution. Bioscience 48:531–538.
Bell, S. L. M., T. B. Herman, and R. J. Wassersug. 2007. Ecology of Thamnophis
sauritus (eastern ribbon snake) at the northern limit of its range. Northeastern
Naturalist 14:279–292.
Bennett, R. A. 1991. A review of anesthesia and chemical restraint in reptiles. Journal
of Zoo and Wildlife Medicine 22:282–303.
Blouin-Demers, G., and P. J. Weatherhead. 2001a. An experimental test of the link
between foraging, habitat selection and thermoregulation in black rat snakes Elaphe
obsoleta obsoleta. Journal of Animal Ecology 70:1006–1013.
Blouin-Demers, G., and P. J. Weatherhead. 2001b. Habitat use by black rat snakes
(Elaphe obsoleta obsoleta) in fragmented forests. Ecology 82:2882–2896.
Blouin-Demers, G., and P. J. Weatherhead. 2002a. Habitat-specific behavioural
thermoregulation by black rat snakes (Elaphe obsoleta obsoleta). Oikos 97:59–68.
Blouin-Demers, G., and P. J. Weatherhead. 2002b. Implications of movement patterns for
gene flow in black rat snakes (Elaphe obsoleta). Canadian Journal of Zoology
80:1162–1172.
Blouin-Demers, G., K. A. Prior, and P. J. Weatherhead. 2000. Patterns of variation
in spring emergence by black rat snakes (Elaphe obsoleta obsoleta). Herpetologica
56:175–188.
Blouin-Demers, G., P. J. Weatherhead, and J. R. Row. 2004. Phenotypic
consequences of nest-site selection in black rat snakes (Elaphe obsoleta). Canadian
Journal of Zoology 82:449–456.
Blouin-Demers, G., L. P. G. Bjorgan, and P. J. Weatherhead. 2007. Changes in
habitat use and movement patterns with body size in black ratsnakes (Elaphe
obsoleta). Herpetologica 63:421–429.
Blouin-Demers, G., P. J. Weatherhead, C. M. Shilton, C. E. Parent, and G. P.
Brown. 2000. Use of inhalant anesthetics in three snake species. Contemporary
Herpetology 4. (http://www.contemporaryherpetology.org/ch/2000/4/.)
Boyarski, V. L., G. H. Rodda, and J. A. Savidge. 2007. Evaluation of harmonic
direction-finding systems for detecting locomotor activity. Journal of Wildlife
Management 71:1704–1707.
Brown, W. S., D. W. Pyle, K. R. Greene, and J. B. Friedlaender. 1982. Movements
and temperature relationships of timber rattlesnakes (Crotalus horridus) in
northeastern New York. Journal of Herpetology 16:151–161.
Brown, G. P., and P. J. Weatherhead. 1999. Female distribution affects mate
searching and sexual selection in male northern water snakes (Nerodia sipedon).
Behavioral Ecology and Sociobiology 47:9–16.
Browning, D. M., S. J. Beaupre, and L. Duncan. 2005. Using partitioned Mahalanobis
D2(K) to formulate a GIS-based model of timber rattlesnake hibernacula. Journal of
Wildlife Management 69:33–44.
Burger, J., and R. T. Zappalorti. 1988. Habitat use in free-ranging pine snakes,
Pituophis melanoleucus, in New Jersey pine barrens. Herpetologica 44:48–55.
Burger, J., and R. T. Zappalorti. 1989. Habitat use by pine snakes (Pituophis
melanoleucus) in the New Jersey Pine Barrens: individual and sexual variation.
Journal of Herpetology 23:68–73.
Burnham, K. P, and D. R. Anderson. 2002. Model selection and multimodel inference.
A practical information-theoretic approach, 2nd edition, Springer New York. 488 pp.
Bushar, L. M., H. K. Reinert, and L. Gelbert. 1998. Genetic variation and gene flow
within and between local populations of the timber rattlesnake, Crotalus horridus.
Copeia 2:411–422.
Camper, J. D. 2009. Core terrestrial habitat around wetlands: contributions from the
spatial ecology of the redbelly watersnake (Nerodia erythrogaster erythrogaster).
Copeia 3:556–562.
Carfagno, G. L. F., and P. J. Weatherhead. 2006. Intraspecific and interspecific
variation in use of forest-edge habitat by snakes. Canadian Journal of Zoology
84:1440–1452.
Carfagno, G. L. F., and P. J. Weatherhead. 2008. Energetics and space use: intraspecific
and interspecific comparisons of movements and home ranges of two Colubrid snakes.
Journal of Animal Ecology 77:416–424.
Carfagno, G. L. F., and P. J. Weatherhead. 2009. Ratsnake response to bottomland
flooding: implications for avian nest predation. Acta Herpetologica 4:191–194.
Carfagno, G. L. F., E. J. Heske, and P. J. Weatherhead. 2006. Does mammalian
prey abundance explain forest-edge use by snakes? Ecoscience 13:293–297.
Charland, M. B. 1991. Anesthesia and transmitter implantation effects on gravid garter
snakes (Thamnophis sirtalis and T. elegans). Herpetological Review 22:46–47.
Clark, R. W. 2005. Pursuit-deterrent communication between prey animals and timber
rattlesnakes (Crotalus horridus): the response of snakes to harassment displays.
Behavioral Ecology and Sociobiology 59:258–261.
Clark, R. W. 2006a. Fixed videography to study predation behavior of an ambush
foraging snake, Crotalus horridus. Copeia 2006:181–187.
Clark, R. W. 2006b. Post-strike behavior of timber rattlesnakes (Crotalus horridus)
during natural predation events. Ethology 112:1089–1094.
Cobb, V. A., J. J. Green, T. Worrall, J. Pruett, and B. Glorioso. 2005. Initial den
location behavior in a litter of neonate Crotalus horridus (timber rattlesnakes).
Southeastern Naturalist 4:723–730.
Cooke, S. J. 2008. Biotelemetry and biologging in endangered species research and
animal conservation: relevance to regional, national, and IUCN Red List threat
assessments. Endangered Species Research 4:165-185.
Cooke, S. J., S. G. Hinch, M. Wikelski, R. D. Andrews, L. J. Kuchel, T. G. Wolcott,
and P. J. Butler. 2004. Biotelemetry: a mechanistic approach to ecology. Trends
in Ecology and Evolution 19:334–343.
Cross, C. L., and C. E. Peterson. 2001. Modeling snake microhabitat from
radiotelemetry studies using polytomous logistic regression. Journal of Herpetology
35:590–597.
Cunnington, G. M. and J. E. Cebek. 2005. Mating and nesting behavior of the eastern
hognose snake (Heterodon platirhinos) in the northern portion of its range. American
Midland Naturalist 154:474–478.
Dodd, C. K., Jr., and W. J. Barichivich. 2007. Movements of large snakes
(Drymarchon, Masticophis) in north-central Florida. Florida Scientist 70:83–93.
Durbian, F. E. 2006. Effects of mowing and summer burning on the massasauga
(Sistrurus catenatus). American Midland Naturalist 155:329–334.
Durbian, F. E., R. S. King, T. Crabill, H. Lambert-Doherty, and R. A. Seigel.
2008. Massasauga home range patterns in the Midwest. Journal of Wildlife
Management 72:754–759.
Durner, G. M., and J. E. Gates. 1993. Spatial ecology of black rat snakes on Remington
Farms, Maryland. Journal of Wildlife Management 57:812-826.
Englesoft, C. E., K. Ovaska, and N. Honkanen. 1999. The harmonic direction finder:
a new method for tracking movements of small snakes. Herpetological Review
30:84–87.
Fazey, I., J. Fischer, and D. B. Lindenmayer. 2005. What do conservation biologists
publish? Biological Conservation 124:63–73.
Fitch, H. S., and H. W. Shirer. 1971. A radiotelemetric study of spatial relationships in
some common snakes. Copeia 1971:118–128.
Galligan, J. H., and W. A. Dunson. 1979. Biology and status of timber rattlesnake
(Crotalus horridus) populations in Pennsylvania. Biological Conservation 15:13–58.
Gerald, G. W., M. A. Bailey, and J. N. Holmes. 2006a. Habitat utilization of Pituophis
melanoleucus melanoleucus (Northern Pinesnake) on Arnold Air Force Base in
middle Tennessee. Southeastern Naturalist 5:253–264.
Gerald, G. W., M. A. Bailey, and J. N. Holmes. 2006b. Movements and activity range
sizes of northern pinesnakes (Pituophis melanoleucus melanoleucus) in Middle
Tennessee. Journal of Herpetology 40:503–510.
Gibbons, 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 50:653–656.
Hackenbrock, C. R., and M. Finster. 1963. Fluothane: a rapid and safe inhalation
anesthetic for poisonous snakes. Copeia 1963:440–441.
Harlow, P., and R. Shine. 1988. Voluntary ingestion of radiotransmitters by snakes in
the field. Herpetological Review 19:57.
Harvey, D. S. 2005. Detectability of a large-bodied snake (Sistrurus c. catenatus) by
time-constrained searching. Herpetological Review 36:413–415.
Harvey, D. S., and P. J. Weatherhead. 2006a. A test of the hierarchical model of
habitat selection using eastern massasauga rattlesnakes (Sistrurus c. catenatus).
Biological Conservation 130:206–216.
Harvey, D. S., and P. J. Weatherhead. 2006b. Hibernation site selection by eastern
massasauga rattlesnakes (Sistrurus catenatus catenatus) near their northern range
limit. Journal of Herpetology 40:66–73.
Himes, J. G. 2000. Burrowing ecology of the rare and elusive Louisiana pine snake,
Pituophis ruthveni (Serpentes: Colubridae). Amphibia-Reptilia 22:91–101.
Himes, J. G., L. M. Hardy, D. C. Rudolph, and S. J. Burgdorf. 2002. Growth rates and
mortality of the Louisiana pine snake (Pituophis ruthveni). Journal of Herpetology
36:683–687.
Himes, J. G., L. M. Hardy, D. C. Rudolph, and S. J. Burgdorf. 2006. Movement patterns
and habitat selection by native and repatriated Louisiana pine snakes (Pituophis
ruthveni): implications for conservation. Herpetological Natural History 9:103–
116.
Horne, J. S., and E. O. Garton. 2006. Selecting the best home range model: an
information-theoretic approach. Ecology 87:1146–1152.
Horne, J. S., E. O. Garton, and J. L. Rachlowa. 2008. A synoptic model of animal
space use: simultaneous estimation of home range, habitat selection, and inter/intraspecific relationships. Ecological Modelling 214:338–348.
Hoss, S. K., C. Guyer, L. L. Smith, and G. W. Schuett. 2010. Multiscale influences of
landscape compositon and configuration on the spatial ecology of eastern diamondback rattlesnakes (Crotalus adamanteus). Journal of Herpetology 1:110–123.
Hyslop, N. L., R. J. Cooper, and J. M. Meyers. 2009a. Seasonal shifts in shelter and
microhabitat use of Drymarchon couperi (eastern indigo snake) in Georgia. Copeia
2009:458–464.
Hyslop, N. L., J. M. Meyers, R. J. Cooper, and T. M. Norton. 2009b. Survival of radioimplanted Drymarchon couperi (eastern indigo snake) in relation to body size and sex.
Herpetologica 65:199–206.
Jellen, B. C., and M. J. Kowalski. 2007. Movement and growth of neonate eastern
massasaugas (Sistrurus c. catenatus). Copeia 2007:994–1000.
Jellen, B. C., D. B. Shepard, M. J. Dreslik, and C. A. Phillips. 2007. Male movement
and body size affect mate acquisition in the eastern massasauga (Sistrurus catenatus).
Journal of Herpetology 41:451–457.
Johnson, G. 2000. Spatial ecology of the eastern massasauga (Sistrurus c. catenatus) in a
New York peatland. Journal of Herpetology 34:186–192.
Kapfer, J. M., J. R. Coggins, and R. Hay. 2008a. Estimates of population size,
measurements of sex ratios, and reported mortality rates for bullsnakes (Pituophis
catenifer sayi) at a site in the upper Midwestern United States. Journal of Herpetology
42:265–269.
Kapfer, J. M., J. R. Coggins, and R. Hay. 2008b. Spatial ecology and habitat selection of
bullsnakes (Pituophis catenifer sayi) at the northern periphery of their geographic
range. Copeia 2008:815–826.
Kapfer, J. M., M. J. Pauers, D. M. Reineke, J. R. Coggins, and R. Hay. 2008.
Environmental, behavioral, and habitat variables influencing body temperature in
radio-tagged bullsnakes, Pituophis catenifer sayi. Journal of Thermal Biology 33:174–
179.
Keller, W. L., and E. J. Heske. 2000. Habitat use by three species of snakes at the
Middle Fork Fish and Wildlife Area, Illinois. Journal of Herpetology 34:558–564.
King, R., C. Berg, and B. Hay. 2004. A repatriation study of the eastern massasauga
(Sistrurus c. catenatus) in Wisconsin. Herpetologica 60:429–437.
Kingsbury, B.A., and C.J. Coppola. 2000. Hibernacula of the copperbelly water snake
(Nerodia erythrogaster neglecta) in southern Indiana and Kentucky. Journal of
Herpetology 34:294–298.
Kjoss, V. A., and J. A. Litvaitis. 2001. Community structure of snakes in a humandominated landscape. Biological Conservation 98:285–292.
Lagory, K. E., L. J. Walston, C. Goulet, R. A. V. Lonkhuyzen, S. Najjar, and C.
Andrews. 2009. An examination of scale-dependent resource use by eastern hognose
snakes in southcentral New Hampshire. Journal of Wildlife Management 73:1387–
1393.
Lutterschmidt, W. I., and H. K. Reinert. 1990. The effect of ingested transmitters
upon the temperature preference of the northern water snake, Nerodia s. sipedon.
Herpetologica 46:39–42.
Lutterschmidt, W. I., D. I. Lutterschmidt, R. T. Mason, and H. K. Reinert. 2009. Seasonal
variation in hormonal responses of timber rattlesnakes (Crotalus horridus) to
reproductive and environmental stressors. Journal of Comparative Physiology
179:747–757.
Macartney, J. M., P. T. Gregory, and K. W. Larsen. 1988. A tabular survey of data
on movements and home ranges of snakes. Journal of Herpetology 22:61–73.
Marshall, J. C., J. V. Manning, and B. A. Kingsbury. 2006. Movement and
macrohabitat selection of the eastern massasauga in a fen habitat. Herpetologica
62:141–150.
Martin, D. L. 1984. An instance of sexual defense in the cottonmouth, Agkistrodon
piscivorus. Copeia 1984:772–774.
Marzluff, J. M, M. G. Raphael, and R. Sallbanks. 2000. Understanding the effects of
forest management on avian species. Wildlife Society Bulletin 28:1132–1143.
McKinney, M. L. 2002. Urbanization, biodiversity, and conservation. Bioscience
52:883–890.
Mitchell, J. C., A.R. Breisch, and K.A. Buhlmann. 2006a. Habitat management
guidelines for amphibians and reptiles of the northeastern United States. Partners in
amphibian and reptile conservation, Technical Publication HMG-3, Montgomery,
Alabama. 108 pp.
Mitchell, R. J., J. K. Hiers, J. J. O’Brien, S. B. Jack, and R. T. Engstrom. 2006b.
Silviculture that sustains: the nexus between silviculture, frequent prescribed fire, and
conservation of biodiversity in longleaf pine forests of the southeastern United States.
Canadian Journal of Forest Resources 36:2724–2736.
Moore, J. A., and J. C. Gillingham. 2006. Spatial ecology and multi-scale habitat
selection by a threatened rattlesnake: the eastern massassauga (Sistrurus catenatus
catenatus). Copeia 2006:742–751.
Mullin, S. J., W. H. N. Gutzke, G. D. Zenitsky, and R. J. Cooper. 2000. Home ranges
of rat snakes (Colubridae: Elaphe) in different habitats. Herpetological Review
31:20–22.
Murcia, C. 1995. Edge effects in fragmented forests: implications for conservation.
Trends in Ecology and Evolution 10:58–62.
Olden, J. N., N. L. Poff, M. L. McKinney. 2006. Forecasting faunal and floral
homogenization associated with human population geography in North America.
Biological Conservation 127:261–271.
Parent, C., and P. J. Weatherhead. 2000. Behavioral and life history responses of
eastern massasauga rattlesnakes (Sistrurus catenatus catenatus) to human disturbance.
Oecologica 125:170–178.
Pattishall, A., and D. Cundall. 2008. Spatial biology of northern watersnakes
(Nerodia sipedon) living along an urban stream. Copeia 2008:752–762.
Pattishall, A., and D. Cundall. 2009. Habitat use by synurbic waternsakes (Nerodia
sipedon). Herpetologica 65:183–198.
Peterson, C. R, A. R. Gibson, and M. E. Dorcas. 1993. Snake thermal ecology: the
causes and consequences of body-temperature variation. Pages 241–314 in R. A.
Seigel and J. T. Collins, editors. Snakes: Ecology and Behavior, McGraw-Hill
Publishing Company, New York.
Plummer, M. V. 1990a. High predation on green snakes, Opheodrys aestivus. Journal of
Herpetology 24:327–328.
Plummer, M. V. 1990b. Nesting movements, nesting behavior, and nest sites of green
snakes (Opheodrys aestivus) revealed by radiotelemetry. Herpetologica 46:190–
195.
Plummer, M. V., and J. D. Congdon. 1994. Radiotelemetric study of activity and
movements of racers (Coluber constrictor) associated with a Carolina bay in South
Carolina. Copeia 1994:20–26.
Plummer, M. V., and N. E. Mills. 1996. Observations on trailing and mating behaviors
in hognose snakes (Heterodon platirhinos). Journal of Herpetology 30:80–82.
Plummer, M. V., and N. E. Mills. 2000. Spatial ecology and survivorship of resident and
translocated hognose snakes (Heterodon platirhinos). Journal of Herpetology
34:565–575.
Prior, K. A. and P. J. Weatherhead. 1994. Response of free-ranging eastern
massasauga rattlesnakes to human disturbance. Journal of Herpetology 28:255–
257.
Prior, K. A. and P. J. Weatherhead. 1996. Habitat features of black rat snake hibernacula
in Ontario. Journal of Herpetology 30:211–218.
Reinert, H. K. 1984a. Habitat separation between sympatric snake populations. Ecology
65:478–486.
Reinert, H. K. 1984b. Habitat variation within sympatric snake populations. Ecology
65:1673–1682.
Reinert, H. K. 1991. Translocation as a conservation strategy for amphibians and reptiles:
some comments, concerns, and observations. Herpetologica 47:357–363.
Reinert, H. K. 1992. Radiotelemetric field studies of pitvipers: data acquisition and
analysis. Pages 185–197 in J. A. Campbell and E. D. Brodie, editors. The Biology of
Pitvipers, Selva, Tyler, Texas.
Reinert, H. K. 1993. Habitat selection in snakes. Pages 201–240 in R. A. Seigel and J. T.
Collins, editors. Snakes: Ecology and Behavior. McGraw-Hill Publishing Company,
New York.
Reinert, H. K., and D. Cundall. 1982. An improved surgical implantation method for
radiotracking snakes. Copeia 1982:702–705.
Reinert, H. K., and W. R. Kodrich. 1982. Movements and habitat utilization by the
massasauga, Sistrurus catenatus catenatus. Journal of Herpetology 16:162–171.
Reinert, H. K., and R. R. Rupert. 1999. Impacts of translocation on behavior and
survival of timber rattlesnakes, Crotalus horridus. Journal of Herpetology 33:45–
61.
Reinert, H. K., and R. T. Zappalorti. 1988a. Field observation of the association of
adult and neonatal timber rattlesnakes, Crotalus horridus, with possible evidence for
conspecific scent trailing. Copeia 1988:1057–1059.
Reinert, H. K., and R. T. Zappalorti. 1988b. Timber rattlesnakes (Crotalus horridus) of
the Pine Barrens: their movement patterns and habitat preference. Copeia
1988:964–978.
Reinert, H. K., D. Cundall, and L.M. Bushar. 1984. Foraging behavior of the timber
rattlesnake, Crotalus horridus. Copeia 1984:976–981.
Richardson, M. L., P. J. Weatherhead, and J. D. Brawn. 2006. Habitat use and
activity of prairie kingsnakes (Lampropeltis calligaster calligaster) in Illinois. Journal
of Herpetology 40:423–428.
Roe, J. H., B. A. Kingsbury, and N. R. Herbert. 2003. Wetland and upland use patterns
in semi-aquatic snakes: implications for wetland conservation. Wetlands 23:1003–
1014.
Roe, J. H., B. A. Kingsbury, and N. R. Herbert. 2004. Comparative water snake ecology:
conservation of mobile animals that use temporary dynamic resources. Biological
Conservation 118:79–89.
Roe, J. H., J. Gibson, and B. A. Kingsbury. 2006. Beyond the wetland border: estimating
the impact of roads for two species of water snakes. Biological Conservation 130:161–
168.
Rooney, T. P., S. W. Wiegmann, D. A. Rogers, and D. M. Waller. Biotic impoverishment
and homogenization in unfragmented forest understory communities. Conservation
Biology 18:787–798.
Roth, T. C., and B. D. Greene. 2006. Movement patterns and home range use of the
northern watersnake (Nerodia sipedon). Copeia 2006:544–551.
Row, J. R., and G. Blouin-Demers. 2006a. Kernels are not accurate estimators of homerange size for herpetofauna. Copeia 2006:797–802.
Row, J. R., and G. Blouin-Demers. 2006b. Thermal quality influences effectiveness of
thermoregulation, habitat use, and behaviour in milk snakes. Oecologia 148:1–11.
Row, J. R., and G. Blouin-Demers. 2006c. Thermal quality influences habitat selection at
multiple spatial scales in milksnakes. Ecoscience 13:443–450.
Row, J. R., G. Blouin-Demers, and P. J. Weatherhead. 2007. Demographic effects of
road mortality in black ratsnakes (Elaphe obsoleta). Biological Conservation
137:117–124.
Rudolph, D. C., S. J. Burgdorf, R. R. Schaefer, R. N. Conner, and R. T. Zappalorti.
1998. Snake mortality associated with late season radio-transmitter implantation.
Herpetological Review 29:155–156.
Rudolph, D. C., R. R. Schaefer, S. J. Burgdorf, M. Duran, and R. N. Conner. 2007.
Pine snake (Pituophis ruthveni and Pituophis melanoleucus lodingi) hibernacula.
Journal of Herpetology 41:560–565.
Russell, K. R., T. B. Wigley, W. M. Baughman, H. G. Hanlin, and W. M. Ford. 2004.
Responses of southeastern amphibians and reptiles to forest management: a review.
Pages 319–334 in M. H. Rauscher and K. Johnsen, editors. Southern forest science:
past, present, and future. United States Department of Agriculture Southern Research
Station Gen. Tech. Rep. SRS–75.
Ryan, T. J., T. Philippi, Y. A. Lieden, M. E. Dorcas, T. B. Wigley, J. W. Gibbons. 2002.
Monitoring herpetofauna in a managed forest landscape: effects of habitat types and
census techniques. Forest Ecology and Management 167:83–90.
Saunders, D. A., R. J. Hobbs, and C. R. Margules. 1991. Biological consequences of
ecosystem fragmentation: a review. Conservation Biology 5:18–32.
Seigel, R. A. 1993. Summary: future research on snakes, or how to combat “lizard
envy.” Pages 395–402 in R. A. Seigel and J. T. Collins, editors. Snakes: Ecology and
Behavior, McGraw-Hill Publishing Company, New York.
Shephard, D. B., A. R. Kuhns, M. J. Dreslik, and C. A. Phillips. 2008. Roads as
barriers to animal movement in fragmented landscapes. Animal Conservation 11:288–
296.
Smith, C. F., G. W. Schuett, R. L. Earley, and K. Schwenk. 2009. The spatial and
reproductive ecology of the copperhead (Agkistrodon contortrix) at the northeastern
extreme of its range. Herpetological Monongraphs 23:45–73.
Steen, D. A., and L. L. Smith. 2009. Eastern kingsnake (Lampropeltis getula getula)
home ranges exhibit limited overlap. Southeastern Naturalist 8:553–558.
Tiebout, H. M., and J. R. Cary. 1987. Dynamic spatial ecology of the water snake,
Nerodia sipedon. Copeia 1987:1–18.
Timmerman, W. W. 1995. Home range, habitat use, and behavior of the eastern
diamondback rattlesnake (Crotalus adamanteus) on the Ordway Preserve. Bulletin of
the Florida Museum of Natural History 38:127–158.
Ujvari, B., and Z. Korsos. 2000. Use of radiotelemetry on snakes: a review. Acta
Zoologica Academiae Scientarium Hungaricae 46:115–146.
Vellend, M, K. Verheyen, K. M. Flinn, H. Jacquemyn, A. Kolb, H. V. Calster, G.
Peterken, B. J. Graae, J. Bellemare, O. Honnay, J. Brunet, M. Wulf, F. Gerhardt, and
M. Hermy. 2007. Homogenization of forest plant communities and weakening of
species-environment relationships via agricultural land use. Journal of Ecology
95:565–573.
Waldron, J. L., J. D. Lanham, and S. H. Bennett. 2006a. Using behaviorally-based
seasons to investigate canebrake rattlesnake (Crotalus horridus) movement patterns
and habitat selection. Herpetologica 62:389–398.
Waldron, J. L., S. H. Bennett, S. M. Welch, M. E. Dorcas, J. D. Lanham, and W.
Kalinowsky. 2006b. Habitat specificity and home-range size as attributes of species
vulnerability to extinction: a case study using sympatric rattlesnakes. Animal
Conservation 9:414–420.
Waldron, J. L., S. M. Welch, and S. H. Bennett. 2008. Vegetation structure and the
habitat specificity of a declining North American reptile: a remnant of former
landscapes. Biological Conservation 141:2477–2482.
Weatherhead, P. J., and F. W. Anderka. 1984. An improved radio transmitter and
implantation technique for snakes. Journal of Herpetology 18:264–269.
Weatherhead, P. J., and M. B. Charland. 1985. Habitat selection in an Ontario
population of the snake, Elaphe obsoleta. Journal of Herpetology 19:12–19.
Weatherhead, P. J., and D. J. Hoysak. 1989. Spatial and activity patterns of black rat
snakes (Elaphe obsoleta) from radiotelemetry and recapture data. Canadian Journal of
Zoology 67:463–468.
Weatherhead, P. J., and K. A. Prior. 1992. Preliminary observations of habitat use and
movements of the eastern massasauga rattlesnake (Sistrurus c. catenatus). Journal of
Herpetology 26:447-452.
Weatherhead, P. J., and G. Blouin-Demers. 2004. Long-term effects of radiotelemetry on
black ratsnakes. Wildlife Society Bulletin 32:900–906.
Weatherhead, P. J., M. R. Prosser, H. L. Gibbs, and G. P. Brown. 2002. Male
reproductive success and sexual selection in northern water snakes determined by
microsatellite DNA analysis. Behavioral Ecology 13:808–815.
Winne, C. T., J. D. Willson, B. D. Todd, K. M. Andrews, J. W. Gibbons. 2007.
Enigmatic decline of a protected population of eastern kingsnakes (Lampropeltis
getula), in South Carolina. Copeia 2007:507–519.
Woodroffe, R., and J. R. Ginsberg. 1998. Edge effects and the extinction of populations
inside protected areas. Science 280:2126–2128.
Wund, M. A., M. E. Torocco, R. T. Zappalorti, and H. K. Reinert. 2007. Activity
ranges and habitat use of Lampropeltis getula getula (eastern kingsnakes).
Northeastern Naturalist 14:343–360.
Table 5.1 Snake radiotelemetry publications in eastern North America. Abbreviations are
as follows: A-adult, SA-subadult, J-juvenile, M-males, NG-non-gravid female, G-gravid
female. Dashed lines indicate data was not available, while NA signifies that the category
was not applicable for the study. An asterisk indicates a snake species not present in
eastern North America. These studies were included because the methodology is
applicable to all snake species.
Study
Category
Snake
Species
Spatial
Ecology
Agkistrodon c. contortrix
35
M-20; NG-12;
G-3
3years
Smith et al. (2009)
Coluber c. constrictor
7
M-1; F-6
3 mos.
Plummer and Congdon (1994)
5
M-4; F-1
4.5 months
(1)-15
(2)- 18
M-7; F-8
M-9; M-9
4 years
8
M-5; F-3
3.5 years
C. c. constrictor (1) &
Pantherophis alleghaniensis
(2)
Crotalus adamanteus
Sample
Size (n)
Gender
Composition
Study
Duration
Citation
Kjoss and Litvaitis (2001)
Carfagno and Weatherhead (2008)
Timmerman (1995)
11 mos
Hoss et al. (2010)
10
M-4; F-6
(1)-21
(2)-18
(1)-M-6;
NG-13; G-2
(2)-M-8; NG9; G-1
7 years
Waldron et al. (2006b)
6
M-4; G-2
6 mos.
Galligan and Dunson (1979)
5
M-1; NG-4;
G-1
3 mos
Brown et al. (1982)
15
M-5: F-10
~3 years
Reinert and Zappalorti (1988b)
30
M-14; NG-9;
G-7
4 years
Reinert and Rupert (1999)
17
M-8; NG-9
3 years
Waldron et al. (2006a)
23
M-12; F-11
5 years
Anderson (2010)
(1)- 1
(2)- 4
(1) M-1
(2) M-2; F-2
1 year
Dodd and Barichivich (2007)
16
-----
5 mos
Plummer and Mills (2000)
8
M-1; F-7
2 years
Lagory (2009)
Lampropeltis c. calligaster
10
M-6; F-4
1 year
Richardson et al. (2006)
Lampropeltis g. getula
9
M-5; F-4
3 years
Wund et al. (2007)
12
M-10; F-2
1.5 yrs
Steen and Smith (2009)
Lampropeltis t. triangulum
10
M-10
2 years
Row and B.-Demers (2006a)
Nerodia erythrogaster
16
M-8; F-8
5 years
Camper (2009)
(1)-15
(2)-13
(1)-M-8; F-7
(2)-M-4; F-9
2 years
Roe et al. (2003)
C. adamanteus (1) &
Crotalus horridus (2)
C. horridus
Drymarchon couperi (1) &
Masticophis f. flagellum (2)
Heterodon platirhinos
Nerodia erythrogaster
neglecta (1) & Nerodia s.
sipedon (2)
Table 5.1 cont.
Study
Category
Spatial
Ecology
Snake
Species
Sample
Size (n)
Gender
Composition
Study
Duration
Nerodia erythrogaster
neglecta (1) & Nerodia s.
sipedon (2)
(1)- 15
(2)-13
(1) M-8; F-7
(2) M-4; F-9
2 years
Roe et al. (2004)
10
F-10
2 years
Tiebout and Cary (1987)
39
-----
3 years
Brown and Weatherhead (1999)
18
M-9; F9
~4 months
50
M-8; NG-8;
G-29
3 years
11
M-7; F-4
1.5 years
32
M-18; F-14
2 years
82
M-29; F-53
3.5 years
A-35
J-10
-----
3 years
B.-Demers et al. (2007)
Pantherophis spiloides
8
M-6; F-2
7 mos.
Mullin et al. (2000)
Pituophis catenifer sayi
27
M-15; F-12
3 years
Kapfer et al. (2008b)
Pituophis m. melanoleucus
14
M-9; F-5
2 years
Gerald et al. (2006b)
Pituophis m. lodingi
8
M-4; F-4
1.5 years
A-12
SA-9
A (M-7; F-5)
SA (M-2; F-7)
3 years
Himes et al. (2006)
25
-----
5 mos.
Reinert and Kodrich (1982)
12
M-8; F-4
1.25 years
15
M-11; NG-2;
G-2
3 years
Johnson (2000)
A-7
SA-16
A (M-4; F-3)
SA (M-11; F5)
5-6 mos
King et al. (2004)
26
M-9; NG-9;
G-8
3 years
Marshall et al. (2006)
16
M-5; NG-1:
G-10
5.5 months
Moore and Gillingham (2006)
12
-----
~ 2 months
Jellen and Kowalski (2007)
87
M-29; NG-8;
G-22;
J-28
11 years
13
M-2; F-11
6 mos.
Bell et al. (2007)
(1)- 5
(2)- 5
(1 ) M-5; F-5
(2) M-5; F-5
5 years
Cross and Petersen (2001)
N. s. sipedon
P. alleghaniensis
Pituophis ruthveni
Sistrurus c. catenatus
Sistrurus c. catenatus
Thamnophis s. sauritus
Habitat Use
Agkistrodon c. contortrix (1)
& A. piscivorous (2)
Citation
Roth and Greene (2006)
Pattishall and Cundall (2008)
Weatherhead and Hoysak (1989)
Durner and Gates (1993)
B.-Demers and Weatherhead (2002b)
Baxley and Qualls (2009)
Weatherhead and Prior (1992)
Durbian et al. (2008)
Table 5.1 cont.
Study
Category
Habitat Use
Snake
Species
Sample
Size (n)
Gender
Composition
Study
Duration
A. c. contortrix (1) &
C. horridus (2)
(1)-20
(2)-21
(1)-M-8; F-12
(2)-M-12; F-9
3 years
Reinert (1984a)
(1)-20
(2)-21
(1)-M-8; F-12
(2)-M-12; F-9
3 years
Reinert (1984b)
8
M-5; F-3
4 years
Timmerman (1995)
21
M-6; F-15
3 years
Waldron et al. (2008)
10
M-4; F-6
11 mos
Hoss et al. (2010)
(1)-21
(2)-18
(1)-M-6;
NG-13; G-2
(2)-M-8;
NG-9; G-1
7 years
Waldron et al. (2006b)
C. adamanteus
C. adamanteus (1) & C.
horridus (2)
C. horridus
Citation
15
M-5; NG-3;
G-7
3 years
Reinert and Zappalorti (1988b)
17
M-8; NG-9
3 years
Waldron et al. (2006a)
(1)- 15
(2)- 18
(1) M-7; F-8
(2) M-9; F-9
3 years
Carfagno and Weatherhead (2006)
(1)- 11
(2)- 13
(1) M- 6; F-5
(2) M- 7; F- 6
5 mos.
Carfagno et al. (2006)
(1)-4
(2)-11
(3)-5
-----
2 years
Keller and Heske (2000)
Drymarchon couperi
32
M-19; F-13
~2.5 years
Hyslop et al. (2009a)
H. platirhinos
8
M-1; F-7
2 years
Lagory et al. (2009)
L. c. calligaster
10
M-6; F-4
1 year
Richardson et al. (2006)
L. g. getula
9
M-5; NG-4
2 years
Wund et al. (2007)
L. t. triangulum
25
M-17; NG-8
2 years
Row and B.-Demers (2006b)
25
M-17; NG-8
2 years
Row and B.-Demers (2006c)
N. erythrogaster
16
M-8; F-8
5 years
Camper (2009)
N. sipedon
50
M-8; NG-8;
G-29
3 years
Pattishall and Cundall (2009)
P. alleghaniensis
7
M-4; F-3
4 mos.
Weatherhead and Charland (1985)
53
M-17; F-36
3 years
B.-Demers and Weatherhead (2001b)
23
M-9; F-14
2 years
B.-Demers and Weatherhead (2002a)
A-35
J-10
-----
3 years
B.-Demers et al. (2007)
27
M-15; F-12
3 years
Kapfer et al. (2008b)
27
M-15; F-12
3 years
Kapfer et al. (2008)
C. c. constrictor (1) & P.
alleghaniensis (2)
C. c. constrictor (1) E. o.
obsoleta (2) & Elaphe
vulpina (3)
P. c. sayi
Table 5.1 cont.
Study
Category
Habitat Use
Snake
Species
Sample
Size (n)
Gender
Composition
Study
Duration
P. m. melanoleucus
10
M-5; F-5
2 years
Burger and Zappalorti (1988)
P. m. melanoleucus
10
M-5; F-5
2 years
Burger and Zappalorti (1989)
9
M-6; F-3
2 years
Gerald et al. (2006a)
A-12
SA-9
A (M-7; F-5)
SA (M-2; F-7)
3 years
Himes et al. (2006)
25
-----
5 mos.
Reinert and Kodrich (1982)
12
M-8; F-4
1.25 years
A-7
SA-16
A (M-4; F-3)
SA (M-11;
F-5)
5-6 mos
King et al. (2004)
34
M-19; NG-6;
G-9
2 years
Harvey and Weatherhead (2006a)
26
M-9; NG-9;
G-8
3 years
Marshall et al. (2006)
16
M-5; NG-1:
G-10
5.5 months
Contia tenuis*
A-5
J-4
-----
8 mos.
Crotalus sp.
NA
NA
NA
Aird (1986)
C. horridus
16
M-8; F-8
NA
Hackenbrock and Finster (1963)
C. horridus & A. c. contortrix
38
-----
NA
Reinert and Cundall (1982)
C. horridus & P. ruthveni
65
NA
NA
Rudolph et al. (1998)
N. s. sipedon
5
M-5
NA
Lutterschmidt and Reinert (1990)
P. alleghaniensis
7
NA
NA
Weatherhead and Anderka (1984)
-----
-----
6 years
P. alleghaniensis (1) N. s.
sipedon (2) & S. c. catenatus
(3)
(1)-61
(2)-70
(3)-36
(1)----(2)----(3)-M-12;
NG-9;
G-10
NA
B.-Demers et al. (2000)
Pseudechis porphyriacus* &
Pseudonaja textilis*
14
-----
NA
Harlow and Shine (1988)
S. c. catenatus
7
M-5; NG-2
2 years
Thamnophis sp.
7
G-7
NA
Charland (1991)
various reptile species
NA
NA
NA
Bennett (1991)
various snake species
NA
NA
NA
Ujvari and Korsos (2000)
NA
NA
NA
NA
Anderson and Talcott (2006)
Agkistrodon contortrix
35
M-20; NG-12;
G-3
3 years
P. ruthveni
S. c. catenatus
Techniques
Natural
History
Citation
Weatherhead and Prior (1992)
Moore and Gillingham (2006)
Englesoft et al. (1999)
Weatherhead and B.-Demers (2004)
Harvey (2005)
Smith et al. (2010)
Table 5.1 cont.
Study
Category
Natural
History
Snake
Species
Sample
Size (n)
Gender
Composition
Study
Duration
Agkistrodon piscivorus
NA
NA
NA
Martin (1984)
C. horridus
NA
NA
NA
Reinert and Zappalorti (1988a)
C. horridus
17
M-6; F-11
2 years
Clark (2005)
A-1
J-4
A-(G-1); J-(4)
5 mos.
Cobb et al. (2005)
D. couperi
32
M-19; F-13
~2.5 years
H. platirhinos
NA
NA
NA
Plummer and Mills (1996)
NA
NA
NA
Cunnington and Cebek (2005)
N. s. sipedon
39
-----
3 years
Opheodrys aestivus
9
G-9
-----
Plummer (1990a)
9
G-9
-----
Plummer (1990b)
64
NG-38; G-26
6 years
B.-Demers et al. (2004)
22
NA
2 years
Carfagno and Weatherhead (2009)
A-12
SA-9
A-(M-7;
F-5)
SA-(M-2;
F-7)
3 years
Himes (2000)
14
M-7; F-7
5 years
Himes et al. (2002)
Pituophis catenifer sayi
59
-----
3 years
Kapfer et al. (2008a)
S. c. catenatus
38
M-17; F-21
3 years
Jellen et al. (2007)
C. c. constrictor
5
M-4; F-1
1.5 years
Kjoss and Litvaitis (2001)
105
M-25; NG-52;
G-28
26 years
Row et al. (2007)
(1)-15
(2)-13
(1) M-8; F-7
(2) M-4; F-9
2 years
Roe et al. (2006)
12
M-4; F-8
-----
Prior and Weatherhead (1994)
30
M-11; NG-9;
G-10
2 years
Parent and Weatherhead (2000)
14
M-8; F-6
3-4 mos.
40
M-21; NG10; G-9
3 years
Shepard et al. (2008)
28
-----
14 years
Bushar et al. (1998)
-----
-----
-----
23
M-12; F-11
5 years
P. alleghaniensis
P. ruthveni
Disturbance
Response
E. o. obsolete
N. e. neglecta (1) & N. s.
sipedon (2)
S. c. catenatus
Hibernation
Biology
C. horridus
Citation
Hyslop et al (2009b)
Brown and Weatherhead (1999)
Durbian (2006)
Browning et al. (2005)
Anderson (2010)
Table 5.1 cont.
Study
Category
Snake
Species
Gender
Composition
Study
Duration
18
M-7; F-11
3 years
Kingsbury and Coppola (2000)
-----
-----
-----
Prior and Weatherhead (1996)
-----
-----
-----
B.-Demers et al. (2000)
(1)- 7
(2)- 5
-----
2 years
Rudolph et al. (2007)
S. c. catenatus
32
-----
2 years
Harvey and Weatherhead (2006b)
C. horridus
4
M-2; G-2
~1 month
30
M-14; NG-9;
G-7
4 years
Reinert and Rupert (1999)
16
-----
5 mos
Plummer and Mills (2000)
P. ruthveni
A-12
SA-9
A (M-7; F-5)
SA (M-2; F-7)
3 years
Himes et al. (2006)
S. c. catenatus
A-7
SA-16
A (M-4; F-3)
M-11; F-5
5-6 mos
King et al. (2004)
21
-----
3 years
Reinert et al. (1984)
17
M-6; F-11
2 years
Clark (2006a)
17
M-6; F-11
2 years
Clark (2006b)
P. alleghaniensis
53
M-17; F-36
3 years
B.-Demers and Weatherhead (2001a)
C. horridus
28
-----
14 years
Bushar et al. (1998)
-----
-----
-----
23
M-12; F-11
5 years
19
M-19
-----
N. e. neglecta
P. alleghaniensis
Hibernation
Biology
Translocation
P. ruthveni (1) & P. m.
melanoleucus (2)
H. platirhinos
Foraging
Ecology
Molecular
Analyses
C. horridus
N. s. sipedon
Sample
Size (n)
Citation
Galligan and Dunson (1979)
Lutterschmidt et al. (2009)
Anderson (2010)
Weatherhead et al. (2002)
Figure 5.1 Distribution of studies that examined aspects of snake radiotelemetry in
eastern North America published during the period 1963-2010, among peer-reviewed
scientific journals, N=112.
Figure 5.2 Distribution of study type for snake radiotelemetry publications in eastern
North America published during the period 1963–2010, N=112 (total number of studies
equals 142 because studies may contain multiple study categories). Includes all studies.
Figure 5.3 Distribution of snake species examined with radiotelemetry in eastern North
America published during the period 1979–2010, N=97 (total number equals 105 because
multiple species were examined in some studies. Does not include methodological
studies.
Figure 5.4 Distribution of snake radiotelemetry studies by year in eastern North America
published during the period 1979-2010, N=97. Does not include methodological studies.
Figure 5.5 Geographic distribution of snake radiotelemetry studies published in eastern
North America during the period 1979-2010, N=97. Does not include methodological
studies.
Table 5.2 Comparison of home range size for snake species tracked at different sites and
independent studies at the same site. Abbreviations are as follows: M-male, NG-nongravid female, and G-gravid female.
Species
Sample Size (N)
MCP estimate (ha)
Crotalus horridus
Reinert and Zappalorti (1988b)
Reinert and Rupert (1999)
Anderson (2010)
M–7; NG–6; G–7
M–8; NG–7; G–4
M–11; NG–10 ; G–11
M–48.6 ± 47.6; N–17.2 ± 15.1; G–9.9 ± 7.0
M–59.9 ± 34.5; N–41.9 ± 49.4; G–5.0 ± 5.4
M–115.7 ± 136.4; NG–11.8 ± 4.6; G–7.3 ± 5.3
Sistrurus c. catenatus
Reinert and Kodrich (1982)
Weatherhead and Prior (1992)
Johnson (2000)
Marshall et al (2006)
Moore and Gillingham (2006)
Durbian et al. (2008)
All-25
M–8; F–3
M–11; NG–2; G–2
M–8; NG–9; G–8
M–5; NG–1; G–10
M–29; NG–8; G–22
All–1.0 ± 1.4
M–33.3 ± 24.8; F–3.8 ± 1.4
M–27.8 ± 16.0; NG–41.4 ± 4.8; G–2.0 ± 1.2
M–7.3 ± 4.3; NG–3.4 ± 2.0; G–1.4 ± 1.4
M–1.6 ± 0.7; F–1.2 ± 0.4
M–38.3 ± 19.7; N–3.6 ± 0.9; G–5.1 ± 1.1
Pituophis sp.
Gerald et al. (2006)
Baxley and Qualls (2010)
M–10 ; F–5
M–3 ; F–2
M–85.3±57.9 ; F–67.8 ± 71.1
M–223.4 ± 156.2 ; F–184.4 ± 17.9
Pantherophis sp.
Weatherhead and Hoysak (1989)
Durner and Gates (1993)
Mullin et al. (2000)
B.-Demers and Weatherhead (2002b)
Carfagno and Weatherhead (2008)
M–8; F–6
M–18; F–14
M–6; F–2
M–14; NG–29; G–16
M–9; F–9
M–3.9 ± 1.7; F–1.2 ± 0.7
M–9.7± 1.4; F–9.2 ± 1.5
M–4.7 ± 5.0; F–3.3 ± 2.8
M-25.5 ±7.6; N-11.5 ± 2.5; G-17.3 ± 5.6
M–6.2 ± 0.9; F–3.5 ± 0.5
CHAPTER 6
MULTI-SCALE HABITAT SELECTION PATTERNS OF THE COPPERHEAD
(AGKISTRODON CONTORTRIX) IN DISTURBED SOUTHEASTERN PINEHARDWOOD FORESTS
Introduction
Increasing evidence of worldwide biodiversity declines (Myers 1996; Cincotta et
al. 2000; Sodhi et al. 2004) and its links to habitat destruction have made it essential to
understand reptile responses to anthropogenic disturbances. Although most of the
attention has been focused on global amphibian declines (Stuart et al. 2004), evidence
suggests that reptiles are declining at equally rapid rates (Gibbons et al. 2000). Habitat
destruction through anthropogenic means (Garber and Burger 1995; Germaine and
Wakeling 2000) and alteration of natural disturbance regimes have been cited as major
factors leading to these declines. Because forest management practices generally disturb
large areas of previously intact forest, there is a heightened interest in organismal
response to these disturbances (Gram et al. 2001; Provencher et al. 2003). Although
herpetofaunal response to forest management has been well documented (Russell et al.
2004), most studies have evaluated lizard response (Greenberg et al. 1994; Greenberg and
Waldrop 2008) and little published work exists regarding snake response to forest
management (but see Todd and Andrews 2007). This is most likely due to either
inadequate sampling methods (i.e., traps are too small; Enge 2001) or problems
associated with detecting cryptic and secretive snake species during visual surveys (Ryan
et al. 2002).
Understanding the environmental factors that influence species distribution
patterns is essential for longterm species conservation. The overall combination of biotic
and abiotic factors that regulate these patterns represents an organism’s fundamental
niche (Hutchinson 1957). Species are often distributed in areas with a broad range of
habitats and species select discrete units of the overall habitat that possesses the
necessary niche space. Habitat selection refers to an animal’s conscious perception of the
surrounding landscape and their ability to limit activity to particular subunits of the
available environment (Reinert 1993); these patterns can be revealed by comparing
aspects of the selected habitat against features within the available habitat (Reinert 1993;
Harvey and Weatherhead 2006).
Organisms may exhibit different habitat selection patterns at different spatial
scales (Wiens et al. 1987; Orians and Wittenberger 1991; Compton et al. 2002; Harvey
and Weatherhead 2006). Habitat selection patterns in snakes are believed to proceed in a
hierarchical fashion where habitat selection is initially determined by internal
physiological factors (e.g., digestive state and reproductive condition; Reinert 1993).
Through this hierarchical approach, snakes will select the most suitable habitat at the
landscape scale (e.g., deciduous forest, field edge, grassland) and then within these
subunits, snakes will select the most suitable microhabitat features (Harvey and
Weatherhead 2006; Row and Blouin-Demers 2006) that correspond with small-scale
biological interactions (e.g., temperature preferences, prey odors, and intraspecific
relationships; Reinert 1993). Examining snake habitat use patterns in a hierarchical
approach may have important conservation implications. For example, a species may
show strong selection for certain macrohabitats, whereas another species may not be
selective at the macrohabitat level, but select similar microhabitats dispersed throughout
the larger landscape (Harvey and Weatherhead 2006; Moore and Gillingham 2006;
Waldron et al. 2008; Hoss et al. 2010). Therefore, it is important to understand the
influence of spatial scale on habitat selection patterns when developing longterm
conservation strategies.
To obtain a more complete understanding of snake ecology, it is essential to
understand space use patterns. For example Bushar et al. (1998) examined correlations
between spatial patterns, hibernation site location, and genetic relationships of C.
horridus and discovered that snakes using the same hibernacula were more closely
related than snakes from nearby hibernation sites. Due to the close proximity of basking
sites to hibernation sites, females were more likely to breed with males from the same
hibernation sites. Many snake habitat selection studies do not include spatial data, and
coupling spatial data with habitat use data with provide a more complete picture of
habitat use patterns and will greatly increase the inference of snake ecology studies (see
Richardson et al. 2006; Moore and Gillingham 2006).
By altering environmental conditions of the immediate and surrounding habitat,
large-scale disturbances may alter snake spatial use and habitat selection patterns.
Although studies of snake habitat selection and spatial ecology are abundant, detailed
studies of how these processes and individual snakes are affected by disturbance are
lacking (but see Kjoss and Litvaitis 2000 and Durbian 2006). I evaluated habitat selection
patterns at multiple scales and home-range size of Copperheads (Agkistrodon contortrix)
to identify the environmental processes that regulate snake habitat use patterns in
disturbed landscapes. As snakes rely on thermoregulation to complete physiological
process (e.g., digestion), I hypothesized that habitat selection patterns and home-range
sizes would be different in canopy disturbed forests when compared to stands with an
intact canopy.
Materials and Methods
Study Organism
The Copperhead (Agkistrodon contortrix) is a widely distributed pit-viper species
ranging from southern Texas through southern Georgia with a northern distribution
extending into southern Connecticut. This relatively common species is composed of five
recognized sub-species that intergrade extensively throughout the range. Copperheads
were chosen for this study because 1) they are common inhabitants of eastern and
southeastern forests (Fitch 1960), 2) they are sit-and-wait predators and are likely to
demonstrate clear habitat selection patterns, 3) cited home range values of 3-10 ha (Fitch
and Shirer 1971; study completed in Kansas) coincide well with stand size in this study;
however, current data indicates that home-range size for these snakes is larger than
previously thought (Smith et al. 2009; study completed in Connecticut), and 4) habitat
use and spatial relationships of copperheads are lacking in the Southeast.
Study Site Description
My study was centered in the northern portion of the William B. Bankhead
National Forest (BNF), located in Lawrence, Winston, and Franklin Counties, of
northwestern Alabama. Bankhead National Forest is a 72,800 ha multi-use forest located
along the highly dissected portion of the southern Cumberland Plateau (Smalley 1982;
Gaines and Creed 2003). Soils within this region are typically composed of HartsellsRock and limestone-Hector (Smalley 1982). Mixed forests of the southern Cumberland
Plateau tend to be dominated by oak-hickory forest types (McWilliams 1991) except in
areas where pines were actively planted for commercial purposes. Loblolly pine (Pinus
taeda L.) was used to re-establish forest conditions in abandoned agricultural and heavily
timbered areas (Gaines and Creed 2003). Reforestation efforts along with natural growth
have resulted in 31,600 ha of Loblolly Pine throughout BNF (Gaines and Creed 2003).
For the past decade, southern pine beetle (Dendroctonus frontalis Zimmermann)
infestations have affected Loblolly Pine stands, producing large numbers of standing
dead trees and increased fuel loads, elevating the risk of damaging wildfires. Because
canopy removal and fire disturbance have been prevented in forests throughout the study
area for decades, the BNF initiated a forest restoration plan to reduce wildfire risk and
promote natural forest growth through tree thinning and prescribed fire disturbance. The
BNF has not traditionally utilized prescribed fire as a management tool, but has opted to
include prescribed burning in the forest restoration plan due to administrative
recommendations. Forest restoration plans in BNF mirror regulations set forth in the
Healthy Forest Restoration Act, which authorizes advanced vegetation management
projects when specified conditions (existence of insect or disease epidemic) pose a
significant threat to ecosystem health (Healthy Forest Restoration Act 2003).
The selected forest stands were generally located on upland sites composed of
loblolly pine 25-50 years of age that also possessed a considerable hardwood component
(Gaines and Creed 2003; Schweitzer and Tadessee 2004). Previous to this study, these
stands had not been recently harvested, and each stand had varying levels of damage
from the Southern Pine Beetle. Other past disturbances included the clearing of hardwood
stands throughout the region for Loblolly Pine plantations during the early 1970’s
(Gaines and Creed 2003).
Snake Capture and Surgery Procedures
I radiotracked Copperheads in forest stands that had been disturbed by various
levels of thinning and prescribed burning (Figure 6.1). Thinning and prescribed burning
operations were generally completed 1-2 years before snakes were monitored. Thinning
was generally completed during the summer months, whereas prescribed burns were
completed during the dormant season (February-March). I captured Copperheads in
modified drift-fence trap arrays and during incidental surveys in the designated research
plots. I determined snake gender by using cloacal probes and assessed female
reproductive condition, when possible, during surgery (Figure 6.2) and by field
observations (i.e., sedentary behavior [Reinert 1984a]) throughout the active season.
Male snakes that were larger than 180 g and female snakes that were larger than 150 g
were implanted intraperitoneally with a radiotransmitter (males: model SI-2 [9 grams],
females: model SB-2 [5 grams]; Holohil systems ltd., Carp, Ontario) following the
procedures in Reinert and Cundall (1982). I sterilized all surgery equipment in an
autoclave and maintained sanitary conditions during surgery by wearing sterile latex
gloves and by disinfecting all surfaces with 90% ethanol. I anesthetized snakes with
isoflurane (MWI, Meridian, Idaho) and did not begin surgical procedures until snakes
were unresponsive to touch. During the entire surgical period I restrained each snake in
an acrylic snake tube to prevent incidental snake bites, and used the tubes to maintain
anesthesia by placing an anesthesia-soaked cotton ball in the end of the snake tube. I
followed traditional transmitter implantation methods (i.e., Reinert and Cundall 1982)
except I used a self-retaining retractor (Codman surgical instruments, Raynham, MA) to
keep the snake’s body cavity open (Figure 6.3) and a small diameter piece (0.23 mm
outside diameter by 305 mm) of aluminum tubing to ease the placement of the transmitter
wire between the skin and ribcage of the snake (Figure 6.4). I used a 3-0 dissolvable
suture to close the inner body cavity and outer skin and used Nexaband (Abbott
Laboratories, Chicago, IL) to ensure the suture site stayed closed. Upon completion of
surgery, the implanted transmitter was located approximately two inches up from the
cloaca (Figure 6-5). Snakes were maintained in observation for two days and were
released at their original capture site.
Radiotelemetry Procedures
As snakes were released, I relocated them every 5-7 days throughout the active
season (April-November) in 2006-2008 using a Merlin 12 (Custom Electronics, Urbana,
IL) and TRX-48S (Wildlife Materials, Carbondale, IL) receiver equipped with a three-
prong yagi antenna. At each location I recorded locality information using a handheld
GPS (Garmin Etrex, Olathe, KS) at ≤ 6 m accuracy.
Home-range Estimation
I determined home-ranges for snakes that were followed throughout a majority of
the active season and tracked to the hibernacula. I used the Animal Space Use v.1.3
(ASU) program to evaluate home-range patterns through an information-theoretic
approach (Horne and Garton 2006). Through this approach I was able to establish
unbiased home-range estimates by examining the overall movement patterns that best
explain the space use of the individual. Most area-usage techniques have limitations
(Marshall et al. 2006) and it is important to estimate home-range use patterns based on
the fit of the data (Horne and Garton 2006). Potential home-range estimators evaluated by
ASU include the exponential power, one-mode bivariate normal, two-mode bivariate
circle mix, two-mode bivariate normal mix, fixed kernel, and adaptive kernel home-range
techniques. In cases where adaptive and fixed kernels both received high support (i.e., < 1
CVC point apart [similar to ∆AIC value]), I chose the fixed kernel over the adaptive
kernel to avoid inflating home-range estimates. For each of these analyses I determined a
95%, 50%, and core area home-range estimate. I used the Animal Space Use for Arc-GIS
v.1 extension (Carpenter 2009) to import the ASU output into Arc-GIS v.9.3 to view and
delineate the area for each home-range utilization distribution. I used Hawth’s Tools
v.3.27 (Beyer 2004) for Arc-GIS to determine minimum convex polygons (MCP) for
each snake. I compared differences in home-range estimates between male and gravid
female snakes and between males monitored in closed canopy and open canopy plots
using an independent samples t-test. Statistical significance was determined at an alpha
level of 0.05.
Macrohabitat Analysis
I evaluated copperhead macrohabitat selection patterns at two scales. To identify
the area of available habitat, I buffered each snake location by the average 95%
utilization distribution area (males-134.3 m; gravid females-84.8 m) and core utilization
distribution area (males-46.1 m; females-28.5 m) using Arc-GIS 9.3. I grouped the
buffered areas into one layer and used Hawth’s Tools v.3.27 for Arc-GIS (Beyer 2004) to
generate random points (21-33 points for each snake) within each buffered home-range to
represent available habitat. The number of randomly created points was identical to the
number of used points for each snake. I identified major habitat features (Table 6.1) at
used and available points using aerial photographs (2006 and 2008 data) of the study site.
I only examined macrohabitat selection patterns for snakes included in home-range
calculations. I used compositional analysis (Aebischer et al. 1993; Richardson et al.
2006) to examine habitat selection patterns at each home-range scale for males, gravid
females, and all snakes together. Because I only tracked one non-gravid female, I
excluded this snake from the analysis.
Microhabitat Analysis
I assessed habitat use patterns through a use-availability approach, whereby used
locations were compared directly to a paired random unused location (Compton et al.
2002; Harvey and Weatherhead 2006). Random locations were established by
determining a random bearing (0⁰-360⁰) and distance (1-50 m) determined directly from
each snake location. At each snake and random location I completed a detailed habitat
analysis as described in Table 6.2 similar to that in Reinert (1984a, b) and Harvey and
Weatherhead (2006). I placed a 1 m2 grid around each location and used a 10 X 10 square
grid to determine percent cover of a suite of variables within the 1 m2 area (see Table
6.2). I measured climatic variables directly after locating snakes and returned later to
measure the remaining habitat variables to avoid disturbance to the snakes. I began
measuring habitat plot data at the beginning of May and ended these measurements by
the end of August for each year. I completed a habitat plot only when a snake moved at
least 1 m from a previous location or when a snake’s location could be visually
confirmed or directly pinpointed. I did not perform a habitat analysis when snakes were
pinpointed to an underground location or when snakes were located at duplicate
locations.
I used an information-theoretic approach to evaluate micro-habitat use patterns of
male and female copperheads and for male snakes only in closed canopy and open
canopy forest stands. Because prescribed burns showed little impact to the forest canopy,
I grouped snakes from these plots with those from control plots (Control BA-30.3 ± 0.3
m2/ha; Burn BA-29.5 ± 2.9 m2/ha). Similarly, snakes from thinned plots were grouped
together because the amount of canopy removal was similar in these treatments (Heavy
Thin BA-11.4 ± 0.6 m2/ha; Light Thin BA-14.2 ± 0.6 m2/ha). I only used male snake
habitat data to evaluate habitat disturbance impacts because I did not have sufficient
numbers of female snakes in control plots. I developed habitat models based on variables
supported in habitat use studies of other pit-viper species (Reinert 1984a, b; Cross and
Peterson 2001; Harvey and Weatherhead 2006) and from my biological knowledge of the
study organism. I used paired logistic regression (PROC LOGISTIC; SAS v.9.3) to
compare used locations versus unused locations and create maximum-likelihood
estimates. I used Akaike’s Information Criterion adjusted for small sample sizes (AICc;
Burnham and Anderson 2002) to evaluate the candidate habitat models and “select” the
most parsimonious model that best explained Copperhead habitat selection. I calculated
AICc values using the formula: AICc = -2 log (L (Θ | y)) + 2K + (2K (K + 1)) / (n – K – 1),
where L (Θ | y) is equal to the maximum log-likelihood value derived from logistic
regression results, K is the total number of parameters used in each candidate model, and
n represents sample size (Burnham and Anderson 2002). Next, I calculated a ∆AICc
value for each model by subtracting each subsequent AICc value from the smallest AICc
value of the best model (Burnham and Anderson 2002). Akaike weights (ωi) were
calculated from ∆AICc values to evaluate the evidence that a particular hypothesis was
the most parsimonious model. I examined evidence ratios to evaluate the degree of
difference between the highest-supported models and calculated model-averaged
parameter estimates to reduce bias and incorporate model selection uncertainty for
models with evidence ratios <2.7 (Burnham and Anderson, 2002). I calculated parameter
estimates for candidate models with greatest support and used Akaike’s weights along
with parameter estimates to calculate standard errors and unconditional 95% confidence
intervals (Burnham and Anderson, 2002).
Results
I recorded 766 locations from 30 individual Copperheads over the course of this
study (Table 6.3). The number of observations per snake ranged from 6 (5 weeks) to 53
(13 months over 1.5 years; Table 6.3). Five snakes were depredated during this study,
whereas four snakes were lost due to transmitter failure and one snake died one month
after surgery due to surgery complications (Table 6.3). I sampled microhabitat features at
432 (males-378; gravid females-54) snake locations and at 432 random locations.
Home-range Analysis
I was able to obtain enough data to calculate home-ranges for 22 total snakes (16
males, 5 gravid females, and 1 non-gravid female). The fixed kernel home-range
procedure was identified as the best home-range estimation technique for 11 snakes,
whereas the adaptive kernel was selected as the best estimation procedure for five snakes
(Table 6.4). The two-mode bivariate normal mix, two-mode bivariate circle, and onemode bivariate normal procedures were each identified twice as the best home-range
estimation technique (Table 6.4).
I detected differences in home-range size between male and gravid female snakes.
Males had larger home-range sizes than gravid females for 95% utilization distributions
(males 17.8 ± 2.3 ha; females 7.1 ± 1.8 ha; t19 = 2.49, p = 0.023), 50% utilization
distributions (males 3.9 ± 0.6 ha; females 1.0 ± 0.4 ha; t19 = 2.687, p = 0.012), core area
utilization distributions (males 2.1 ± 0.3 ha; females 0.8 ± 0.3 ha; t19 = 2.687, p = 0.015),
and 100% minimum convex polygons (males 12.0 ± 1.9 ha; 4.1 ± 1.1 ha; t18.8 = 3.590, p =
0.002; Table 6.4). The one non-gravid female had home-range estimates larger than
gravid females and more similar to males in this study (Table 6.4). In addition to having
smaller core area estimates, gravid females usually had one centralized activity area,
whereas males had more than one activity center (Figure 6.6). I was unable to detect an
effect of canopy removal on male home-range size for 95% utilization distributions
(closed canopy 16.7 ± 5.7 ha; open canopy 18.1 ± 2.6 ha; t14 = 0.268, p = 0.793), 50%
utilization distributions (closed canopy 3.1 ± 0.9 ha; open canopy 4.2 ± 0.7 ha; t14 =
0.817, p = 0.427), core area utilization distributions (closed canopy 2.2 ± 0.3 ha; open
canopy 1.8 ± 0.5 ha; t14 = 0.687, p = 0.503), and 100% minimum convex polygons
(closed canopy 12.2 ± 2.2 ha; open canopy 11.7 ± 3.9 ha; t14 = 0.087, p = 0.932; Table
6.4).
Macrohabitat Selection
I was able to detect slight differences in male macrohabitat selection at the larger
macrohabitat scale. Males used a variety of habitats, but selected hard edges and southern
pine beetle stands greater than the availability of these habitats (Table 6.5). I was unable
to detect differences in macrohabitat selection for male snakes at the smaller
macrohabitat scale; however, the gradient of preferred habitats was very similar to the
larger scale. Gravid females, and male and gravid females combined, did not select
habitats differenty from random at both macrohabitat scales (Table 6.5). Although gravid
females did not select macrohabitats different than what was available, I observed that
three of the gravid females selected cut stands, whereas the other two females selected
forested habitat with canopy gaps created through windthrow of Quercus species (Figure
6.7).
Microhabitat Selection
The variables AT_amb, RH_amb, and T_soil were highly correlated with other
variables and were not included in further analyses. Microhabitat models for males
(Table 6.6) and gravid females (Table 6.7) were similar and both included %_CWD,
L_Depth, %_Herb, Log_vol, and Bas_area. The top supported model for gravid females
was more complex and also included For_3 and Can_cov, which were both lower in
selected sites compared to random sites (Table 6.8). Gravid females inhabited sites with
greater available %_CWD than male snakes and greater log_vol than male snakes (Table
6.8). Male snakes inhabited sites with greater L_Depth, greater %_Herb, and lower
Bas_area estimates. Parameter estimates along with unconditional 95% confidence
intervals indicate that %_CWD and L_Depth were both important for discriminating
between selected and random sites, whereas Log_vol and Bas_area did not provide good
separation between these sites (Table 6.9).
Microhabitat models for male snakes in closed canopy (Table 6.10) and open
canopy plots (Table 6.11) indicate that %_CWD, L_Depth, and %_Herb were both
important for differentiating between selected and random sites. The variable %_Slash
was also supported in the top model in closed canopy sites and was supported in the
second most parsimonious model in open canopy sites. Overall, %_CWD and Bas_area
were greater in closed canopy selected sites, whereas %_Slash, L_Depth, and %_Herb
were greater in open canopy selected sites (Table 6.12). Similar to male and female
comparisons, parameter estimates along with 95% confidence intervals indicate that
%_CWD and L_depth were most important for differentiating between selected and
random sites for male snakes in closed and open canopy plots (Table 6.9).
Discussion
It is important to understand spatial and habitat use patterns for organisms
inhabiting disturbed landscapes. I was able to use a novel approach to evaluate space use
patterns in Copperheads. By examining the home-range data through an informationtheoretic approach I was able to generate home-range estimates based on the fit of the
data rather than choosing a technique based on its use in the published literature. Using a
home-range estimation technique that actually fits the data is important because the
solution is biologically founded rather than forced to fit the dataset (Horne and Garton
2006). The estimated home-range values in this study were larger than those determined
by Fitch and Shirer (1971; 3-9 ha [MCP]). Fitch and Shirer’s estimates likely
underestimate the true home-range because they force-fed the snakes with transmitters,
which have been found to initiate post-feeding behaviors (e.g. preference for warm
locations and sedentary behavior; Lutterschmidt and Reinert 1990). Copperhead homerange estimates from a northern population ranged from 0.6-44.5 ha (Smith et al. 2009),
which are comparable to home-range sizes estimated in this study. However, several of
these home-range estimates were much larger than the maximum size estimated in my
study. Snake home-range patterns are likely influenced by the distance each snake has to
travel in order to gain access to resources (e.g., breeding sites and hibernacula). Further
work is necessary to understand the factors that cause home-range size variations for the
same species at different geographic locations.
Male Copperhead home-ranges were larger than gravid females indicating space
use differences between these individuals. In addition, the sole non-gravid female in this
study had space use patterns similar to male snakes. Home-range patterns in pit-vipers are
highly dependent on reproductive condition, where males and non-gravid females have
larger home-ranges than gravid females (Johnson 2000; Marshall et al. 2006; Waldron et
al. 2006). Gravid females generally made a series of large movements during the early
spring and settled on a permanent location usually sometime in June and remained at this
location until late August or early September. Although the non-gravid female had one
activity center like the gravid females, this snake tended to make a series of large
movements to and from the core area and never settled on one location. Similarly, males
made a series of continuous movements throughout the active season and usually had
multiple areas of core spatial use. Male snakes tended to be very mobile and rarely settled
on site for more than two locations.
I was able to detect macrohabitat selection patterns for male Copperheads at the
95% utilization distribution scale. Males selected hard edge and Southern Pine Beetle
habitats greater than what was available, whereas they selected pine plantations and cut
forests considerably less than what was available. A majority of the selected hard edge
habitat occurred at the interface between forest and open fields (i.e., cultivated game
fields). Snakes often use forest / field edges due to increased thermoregulation
opportunities (Blouin-Demers and Weatherhead 2001; Carfagno and Weatherhead 2006;
Carfagno et al. 2006; Row and Blouin-Demers 2006) and cover sites that are formed due
to disturbances at the field/edge interface (Waldron et al. 2006). Similarly, male snakes
likely selected Southern Pine Beetle disturbed habitats due to the increased availability of
cover (i.e., large fallen logs) and thermoregulation sites (i.e., large canopy gaps). Largescale Southern Pine Beetle infestations usually kill a large number of affected trees,
thereby increasing the abundance of fallen logs and large canopy gaps (Duncan and
Linhoss 2005). Snakes may have also selected Southern Pine Beetle habitats due to
increased prey density, as increased CWD cover has been shown to benefit small
mammal communities (Loeb 1999; McKay and Komoroski 2004). Although habitat
selection was not evident for gravid females at the landscape level, these snakes
concentrated activity in areas with canopy disturbances. Three of these snakes had core
activity areas in harvested stands, whereas two snakes had core activity areas in stands
with canopy gaps caused by windthrow of large Quercus species. As I did not have pretreatment data for these snakes, it is difficult to know whether these snakes used these
areas in prior seasons. Gravid female snakes have been found to select open
macrohabitats with greater thermoregulation opportunities (Harvey and Weatherhead
2006; Marshall et al. 2006; Crane and Greene 2008), which are important for these
snakes during embryogenesis (Crane and Greene 2008; Foster et al. 2009). The sample
size of gravid females was low during this study, which undoubtedly influenced the
macrohabitat selection results. However, as thermally optimal sites are not a limiting
factor in the southeast, it is not surprising that gravid females were able to find optimal
thermoregulation sites in macrohabitats other than cut forest. I did not include the one
non-gravid female into habitat analyses due to low sample size, but space use patterns
along with personal observations indicate that the non-gravid female was more similar to
males in their space and habitat use patterns than the gravid females.
Microhabitat comparisons indicated that both male and gravid female snakes
selected sites possessing greater litter depth and CWD cover. Male copperheads in
northern populations have been found to select microhabitats with abundant CWD cover
(Cross and Peterson 2001), whereas gravid females tend to select sites with an open
canopy and abundant rock cover (Reinert 1984a). Although male and gravid female
Copperheads selected microsites with similar habitat attributes in this study, I suggest
they selected these habitats for different reasons. Male Copperheads likely selected these
sites as optimal prey ambush sites, whereas gravid females selected more open sites with
woody cover for optimal thermoregulatory opportunities and protection from potential
predators. Male and non-gravid Timber Rattlesnakes (Crotalus horridus) have been
found to use logs and other downed woody debris as prey ambush sites (Reinert et al.
1984a). Moreover, forest small mammals use these logs as runways and sit-and-wait
predators can capitalize on the increased chance of capturing prey at these sites (Reinert
et al. 1984). The proclivity of gravid pit-viper species for open habitats is well
documented and these individuals select these habitats for protection from predators and
increased thermoregulation opportunities (Reinert 1984a; Marshall et al. 2006).
I was unable to detect impacts of canopy removal on space use or habitat selection
patterns for male Copperheads. I originally hypothesized that snakes in harvested stands
would have smaller home-ranges due to increased cover sites and thermoregulation
opportunities. Most snakes that were originally captured in harvested stands did not
consistently use the harvested habitats and used a wide variety of the surrounding
landscape. When male snakes moved into cut stands, they often used large slash piles as
cover sites. Managers should continue to leave behind slash and logs as cover for snakes
and other wildlife, as a wide array of forest dwelling species have been found to use slash
and other down woody debris as cover sites (Hassinger 1989). Although I was unable to
test microhabitat differences for gravid females in control and canopy-removed sites,
these snakes appeared to show a preference for cut stands. Inference from this study is
quite limited because I was unable to track the same individuals for a full active season
before and after the implementation of the forest treatments. In future studies, researchers
should work along with forest managers to ensure that snake movement and habitat
selection data can be collected before, during, and after forest management practices.
In this study, Copperheads did not demonstrate strong patterns of hierarchical
habitat selection. Although males selected Southern Pine Beetle and hard edge habitats
different than random, I believe these habitats were mainly selected due the availability
of suitable microhabitat. Gravid females did not show a strong tendency for one
macrohabitat type, but tended to seek out sites with suitable cover and thermoregulation
sites. Both males and gravid females appear to select microhabitats based on increased
cover in the form of CWD and litter depth, whereas gravid females additionally sought
out sites with optimal thermoregulation opportunities. Similarly, Harvey and
Weatherhead (2006) found that Eastern Massassauga Rattlesnakes (Sistrurus c.
catenatus) did not show strong habitat selection at the macrohabitat level, but selected
optimal microhabitat sites within each macrohabitat type. Conversely, Waldron et al.
(2008) found that Eastern Diamondback Rattlesnakes (Crotalus adamanteus) exhibited
strong selection for certain macrohabitats and then selected suitable microhabitats within
these macrohabitats. As Copperheads inhabit a broad range of habitat types, it is not
surprising that this species selects optimal microhabitat sites across a broad range of
macrohabitat types. In order to better examine habitat selection processes, future studies
need to include an adequate number of individuals from all reproductive classes across a
broad range of habitat types.
Bibliography
Aebischer, N. J., P. A. Robertson, and R. E. Kenward. 1993. Compositional analysis of
habitat use from animal radio-tracking data. Ecology 74:1313–1325.
Beyer, H. L. 2004. Hawth's Analysis Tools for ArcGIS. Available at
(http://www.spatialecology.com/htools.)
Blouin-Demers, G., and P. J. Weatherhead. 2001. Habitat use by black rat snakes (Elaphe
obsoleta obsoleta) in fragmented habitats. Ecology 82:2882–2896.
Burnham, K. P. and Anderson, D. R. 2002. Model selection and multimodel inference: a
practical information-theoretic approach, 2nd edition. Springer Press, New York. 488
pp.
Bushar, L. M., H. K. Reinert, and L. Gelbert. 1998. Genetic variation and gene flow
within and between local populations of the timber rattlesnake, Crotalus horridus.
Copeia 1998:411–422.
Carfagno, G. L. F., and P. J. Weatherhead. 2006. Intraspecific and interspecific variation
in use of forest-edge habitat by snakes. Canadian Journal of Zoology 84:1440–1452.
Carfagno, G. L. F., E. J. Heske, and P. J. Weatherhead. 2006. Does mammalian prey
abundance explain forest-edge use by snakes? Ecoscience 13:293–297.
Carpenter, J. 2009. Animal Space Use for ArcGIS v.1. Georgia Coastal Ecosystems
LTER File Archive, University of Georgia, Athens, Georgia.
(http://gcelter.marsci.uga.edu/public/app/resource_details.asp?id=355).
Cincotta, R. P., J. Wisnewski, and R. Engleman. Human population in the biodiversity
hotspots. Nature 404:990–992.
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.
Crane, A. L., and B. D. Greene. 2008. The effect of reproductive condition on
thermoregulation in female Agkistrodon piscivorus near the northwestern range limit.
Herpetologica 64:156–167
Cross, C. L., and C. E. Peterson. 2001. Modeling snake microhabitat from
radiotelemetry studies using polytomous logistic regression. Journal of Herpetology
35:590–597.
Duncan, R. S., and J. E. Linhoss. 2005. Regeneration of Virginia pine (Pinus virginiana)
following southern pine beetle (Dendroctonus frontalis) outbreak in the Sipsey
Wilderness, Alabama. Forest Ecology and Management 212:65–74.
Durbian, F. E. 2006. Effects of mowing and summer burning on the massasauga
(Sistrurus catenatus). American Midland Naturalist 155:329–334.
Fitch, H. S. 1960. Autecology of the copperhead. University of Kansas Publications
Museum of Natural History 13:85–288.
Fitch, H. S., and H. W. Shirer. 1971. A radiotelemetric study of spatial relationships in
some common snakes. Copeia 1971:118–128.
Foster, M. S., K. M. Bissell, H. Campa III, and T. M. Harrison. 2009. The influence of
reproductive status on thermal ecology and vegetation use of female eastern
massassauga rattlesnakes (Sistrurus catenatus catenatus) in southwestern Michigan.
Herpetological Conservation and Biology 4:48–54.
Garber, S., and J. Burger. 1995. A 20-yr study documenting the relationship between
turtle decline and human recreation. Ecological Applications 5:1151–1162.
Gaines, G.D., and J. W. Creed. 2003. Forest health and restoration project. National
forests in Alabama, Bankhead National Forest Franklin, Lawrence, and Winston
Counties, Alabama. Final environmental impact statement. Management Bulletin R8–
MB 110B. 353 pp.
Germaine, S. S., and B. F. Wakeling. 2001. Lizard species distributions and habitat
occupation along an urban gradient in Tucson, Arizona, USA. Biological
Conservation 97:229–237.
Gram, W. K., V. L. Sork, R. J. Marquis, R. B. Renken, R. L. Clawson, J. Faaborg, Debra
K. Fantz, J. L. Corff, J. Lill, and P. A. Porneluzi. 2001. Evaluating the effects of
ecosystem management: a case study in a Missouri Ozark forest. Ecological
Applications 11:1667–1669.
Greenberg, C.H., D. G. Neary, and L. D. Harris. 1994. Effect of high-intensity wildfire
and silvicultural treatments on reptile communities in sand-pine scrub. Conservation
Biology 8:1047–1057.
Greenberg, C. H., and T. A. Waldrop. 2008. Short-term response of reptiles and
amphibians to prescribed fire and mechanical fuel reduction in a southern
Appalachian upland hardwood forest. Forest Ecology and Management 255:2883–
2893.
Harvey, D. S., and P. J. Weatherhead. 2006. A test of the hierarchical model of
habitat selection using eastern massasauga rattlesnakes (Sistrurus c. catenatus).
Biological Conservation 130:206–216.
Hassinger, J. S. 1989. Dead wood for wildlife. Pages 171–184 in J. C. Finley and and M.
C. Brittingham, editors. Timber management and its effects on wildlife. Pennsylvania
State University, University Park, PA.
Healthy Forest Restoration Act. Act of December 3, 2003. 117 Stat., as amended; 16
U.S.C. 1887–1915.
Horne, J. S., and E. O. Garton. 2006. Selecting the best home range model: an
information-theoretic approach. Ecology 87:1146–1152.
Hoss, S. K., C. Guyer, L. L. Smith, and G. W. Schuett. 2010. Multiscale influences of
landscape composition and configuration on the spatial ecology of Eastern
Diamondback Rattlesnakes (Crotalus adamanteus). Journal of Herpetology
44:110–123.
Hutchinson, G. E. 1957. Concluding remarks. Cold Spring Harbor Symposia on
Quantitative Biology 22:415–427.
Johnson, G. 2000. Spatial ecology of the eastern massasauga (Sistrurus c. catenatus) in a
New York peatland. Journal of Herpetology 34:186–192.
Kjoss, V. A., and J. A. Litvaitis. 2001. Community structure of snakes in a humandominated landscape. Biological Conservation 98:285–292.
Loeb, S. C. 1999. Responses of small mammals to coarse woody debris. Journal of
Mammalogy 80:460–471.
Lutterschmidt, W. I., and H. K. Reinert. 1990. The effect of ingested transmitters
upon the temperature preference of the northern water snake, Nerodia s. sipedon.
Herpetologica 46:39–42.
Marshall, J. C., Jr., J. V. Manning, and B. A. Kingsbury. 2006. Movement and
macrohabitat selection of the eastern massasauga in a fen habitat. Herpetologica
62:141–150.
McKay, T. S., and M. J. Komoroski. 2004. Demographic responses of shrews to removal
of coarse woody debris in a managed pine forest. Forest Ecology and Management
189:387–395.
McWilliams, W. H. 1991. Availability of yellow pine sawtimber in Alabama. Pages 115–
119 in R. H. Jones, editor. Proceedings: Alabama’s forest resources: past, present, and
future. Auburn University, AL. May 29–30, 1991.
Moore, J. A., and J. C. Gillingham. 2006. Spatial ecology and multi-scale habitat
selection by a threatened rattlesnake: the eastern massassauga (Sistrurus catenatus
catenatus). Copeia 2006:742–751.
Myers, M. 1996. Environmental services of biodiversity. Proceedings of the National
Academy of Sciences 93:2764–2769.
Orians, G. H., and J. F. Wittenberger. 1991. Spatial and temporal scales in habitat
selection. The American Naturalist 137:29–49.
Provencher, L., A. R. Litt, and D. R. Gordon. 2003. Predictors of species richness in
northwest Florida longleaf pine sandhills. Conservation Biology 17:1660–1671.
Reinert, H. K., and D. Cundall. 1982. An improved surgical implantation method for
radio-tracking snakes. Copeia 1982:702–705
Reinert, H. K. 1984a. Habitat separation between sympatric snake populations. Ecology
65:478–486.
Reinert, H. K. 1984b. Habitat variation within sympatric snake populations. Ecology
65:1673–1682
Reinert, H. K. 1993. Habitat selection in snakes. Pages 201–240 in R. A. Seigel and J. T.
Collins, editors. Snakes: Ecology and Behavior. McGraw-Hill Publishing Company,
New York.
Reinert, H. K., D. Cundall, and L.M. Bushar. 1984. Foraging behavior of the timber
rattlesnake, Crotalus horridus. Copeia 1984:976–981.
Richarson, M. L., P. J. Weatherhead, and J. D. Brawn. 2006. Habitat use and activity of
prairie kingsnakes (Lampropeltis calligaster calligaster) in Illinois. Journal of
Herpetology 40:423–428.
Row, J. R., and G. Blouin–Demers. 2006. Thermal quality influences habitat selection at
multiple spatial scales in milksnakes. Ecoscience 13:443–450.
Russell, K. R., T. B. Wigley, W. M. Baughman, H. G. Hanlin, and W. M. Ford. 2004.
Responses of southeastern amphibians and reptiles to forest management: a review.
Pages 319–334 in M. H. Rauscher and K. Johnsen, editors. Southern forest science:
past, present, and future. United States Department of Agriculture Southern Research
Station Gen. Tech. Rep. SRS–75.
Ryan, T. J., T. Philippi, Y. A. Leiden, M. E. Dorcas, T. B. Wigley, and J. W. Gibbons.
2002. Monitoring herpetofauna in a managed forest landscape: effects of habitat types
and census techniques. Forest Ecology and Management 167:83–90.
Schweitzer, C. J. and W. Tadesse. 2004. The interaction of fire frequency and stand
density on plant community dynamics. Pages 1–14 in CREST–Center for Forest
Ecosystem Assessment (CFEA) Alabama A&M University. National Science
Foundation proposal.
Smalley, G. W. 1982. Classification and evaluation of forest sites on the mid-Cumberland
Plateau. U.S. Department of Agriculture Forest Service, General Technical Report
S0–38. Southern Forest Experimental Station, New Orleans, Louisiana. 58 pp.
Smith, D. M., B. C. Larson, M. J. Kelty, and P. M. S. Ashton. 1997. The practice of
silviculture: applied forest ecology. John Wiley and Sons Inc. 537 pp.
Smith, C. F., G. W. Schuett, R. L. Earley, and K. Schwenk. 2009. The spatial and
reproductive ecology of the copperhead (Agkistrodon contortrix) at the northeastern
extreme of its range. Herpetological Monographs 23:45–73.
Sodhi, N. S., L. P. Koh, B. W. Brook, and P. K. L. Ng. 2004. Southeast Asian
biodiversity: an impending disaster. Trends in Ecology and Evolution 19:654–660.
Stuart, S. N., J. S. Chanson, N. A. Cox, B. E. Young, A. S. L. Rodriques, D. L. Fischman,
and R. W. Waller. 2004. Status and trends of amphibian declines and extinctions
worldwide. Science 306:1783–1791.
Todd, B.D., and K. M. Andrews. 2007. Response of a reptile guild to forest harvesting.
Conservation Biology 22:753–761.
Waldron, J. L., S. H. Bennett, S. M. Welch, M. E. Dorcas, J. D. Lanham, and W.
Kalinowsky. 2006. Habitat specificity and home-range size as attributes of species
vulnerability to extinction: a case study using sympatric rattlesnakes. Animal
Conservation 9:414–420.
Waldron, J. L., S. M. Welch, and S. H. Bennett. 2008. Vegetation structure and the
habitat specificity of a declining North American reptile: a remnant of former
landscapes. Biological Conservation 141:2477–2482.
Wiens, J. A., J. T. Rotenberry, and B. V. Horne. 1987. Habitat occupancy patterns of
North American shrubsteppe birds: the effect of spatial scale. Oikos 48:132–147.
Figure 6.1 Locations of study sites within the northwestern portion of the William B.
Bankhead National Forest, Alabama, U.S.A. Forest treatments consisted of the six
following categories: C-control, B-burn, HT-heavy thin, LT-light thin, HTB-heavy thin
and burn, and LTB-light thin and burn. Numbers after treatment abbreviations correspond
to block number.
HT3
HTB1
LT3
LT2
C3
B3
HT2
HTB2 HT1
C2
LT1
C1
LTB2
HTB3
LTB1
LTB3
B1
B2
Figure 6.2 Gravid female Copperhead as indicated by the presence of a developing
embryo. (Photo by Scott White, DVM).
Figure 6.3 Self-retaining retractor used to hold open the peritoneal cavity during surgery.
(Photo by Scott White, DVM).
Figure 6.4 Small diameter aluminum tubing used to thread transmitter wire in-between
the outer skin and rib-cage. (Photo by Scott White, DVM).
Figure 6.5 Implanted radiotransmitter inside of adult Copperhead. (X-ray by Scott White,
DVM).
a
Description
Forests composed of a broad mixture of deciduous tree species, with interspersed evergreen tree species.
Planted pine forests usually dominated by unthinned Viginia Pine (Pinus virginiana ) or Loblolly Pine (P. taeda ).
Forest stands that have been thinned through silvicultural means.
Forest stands disturbed by infestations of the Southern Pine Beetle (Dendroctonus frontalis ).
An abrupt interface between forest and an open habitat. Mainly included roads and fields. Hard edges extended 15 m on either side of the interface (Blouin-Demers and Weatherhead (2001).
The interface between intact forest and thinned forest. Soft edges extended 15 m on either side of the interface (Blouin-Demers and Weatherhead 2001).
Habitats within the study area were characterized by combining aerial photographs along with Southern Pine Beetle forest stand shapefiles supplied by the USDA Forest Service.
Habitat Typea
Mixed Deciduous Forest
Pine Plantation
Cut Forest
Southern Pine Beetle
Hard Edge
Soft Edge
Table 6.1 Macrohabitat types used by Copperheads (Agkistrodon contortrix) in the William B. Bankhead National Forest,
Alabama, U.S.A.
%_Litt
%_Bare
%_Herb
%_Wood
%_CWD
%_Slash
%_Rock
For_1
For_2
For_3
For_4
L_depth
Can_cov
No_ws
T_ws
AT_snake
AT_amb
RH_snake
RH_amb
T_soil
Number of Woody Stems
Tallest Woody Stem
Air Temperature Snake
Air Temperature Ambient
Relative Humidity Snake
Relative Humidity Ambient
Soil Temperature
Bas_area
Per_pine
Per_hwd
Per_snag
Over_d
Over_dbh
Und_d
Und_dbh
N_logd
Log_vol
Rock_d
Shrub_d
Shrub_dia
Basal Area
Percent Pine
Percent Hardwood
Percent Snag
Nearest Overstory Tree Distance
Nearest Overstory Tree DBH
Nearest Understory Tree Distance
Nearest Understory Tree DBH
Nearest Log Distance
Log Volume
Nearest Rock Distance
Nearest Shrub Distance
Nearest Shrub Diameter
Variables assessed outside of 1 m2 grid
Abbreviation
Variables assessed inside of 1 m grid
Percent Litter
Percent Bare
Percent Herbaceous
Percent Woody
Percent CWD
Percent Slash
Percent Rock
Forest Level 1
Forest Level 2
Forest Level 3
Forest Level 4
Litter Depth
Canopy Cover
2
Habitat Variable
2
2
Distance (cm) of nearest rock ≥ 10 cm in diameter to plot center. Rock could not be within 1 m grid.
Distance (cm) of nearest shrub species to plot center. Shrub could not be over 2 m in height to be counted.
Diameter (cm) of nearest shrub species to plot center. Shrub could not be over 2 m in height to be counted.
2
Volume (m ) of nearest log ≥ 10 cm in diameter to plot center. Determined as volume of a cylinder: π r x length
3
Distance (cm) of nearest log ≥ 10 cm in diameter to plot center. Log could not be within 1 m grid.
2
Cross-sectional area (m /ha) of living overstory trees surrounding each sampled plot. Determined with a 10 power prism.
Composition (%) of overstory pine tree species. Limited to trees tallied in basal area estimate.
Composition (%) of overstory deciduous tree species. Limited to trees tallied in basal area estimate.
Composition (%) of snags. Limited to trees that would be tallied in basal area estimate.
Distance (m) of nearest overstory tree species ≥ 2.4 cm DBH.
DBH (cm) of nearest tree ≥2.4 cm DBH
Distance (m) of nearest understory tree species <2.4 cm DBH.
DBH (cm) of nearest tree <2.4 cm DBH
Tallest woody stem (cm) within the 1 m grid.
Air temperature (⁰C) taken within 10 cm of the snake.
Air temperature (⁰C) taken at breast height.
Relative humidity (%) taken within 10 cm of the snake.
Relative humidity (%) taken at breast height.
Soil temperature (⁰C) taken within 10 cm of the snake.
2
Number of woody stems within the 1 m2 grid.
Forest litter cover (%) within a 10 cm x 10 cm grid. Variable had to occupy one grid cell to be counted.
Bare ground cover (%) within a 10 cm x 10 cm grid. Variable had to occupy one grid cell to be counted.
Vegetative cover (%) within a 10 cm x 10 cm grid. Variable had to occupy one grid cell to be counted.
Woody stem cover (%) within a 10 cm x 10 cm grid. Variable had to occupy one grid cell to be counted.
Cwd cover (%) within a 10 cm x 10 cm grid. Variable had to occupy one grid cell to be counted.
Slash cover (%) within a 10 cm x 10 cm grid. Variable had to occupy one grid cell to be counted.
Rock cover (%) within a 10 cm x 10 cm grid. Variable had to occupy one grid cell to be counted.
Average vegetative ground tree cover (%) taken at each of four points along the grid.
Average midstory tree cover (%) taken at each of four points along the grid.
Average sub-canopy tree cover (%) taken at each of four points along the grid.
Average canopy tree cover (%) taken at each of four points along the grid.
Average litter depth (cm) taken at each of four points along the grid.
Average canopy cover (%) take at each of four points along the grid.
Description
Table 6.2 Microhabitat variables assessed at each used and random site for the Copperhead (Agkistrodon
contortrix) in the William B. Bankhead National Forest, Alabama, U.S.A.
a
Gender
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
F(G)
F(NG)
F(G)
F(G)
F(Unk)
F(G)
F(Unk)
Treatment Type Number of Locations
Thin
27
Thin & burn
30
Burn
29
Control
14
Thin & Burn
28
Thin
29
Thin
28
Thin
29
Thin and Burn
42
Thin and Burn
37
Thin and Burn
40
Control
15
Thin and Burn
6
Thin
9
Thin and Burn
33
Thin
20
Thin
45
Thin and Burn
27
Control
35
Thin
10
Control/Burn
53
Control
31
Control
13
Thin and Burn
21
Thin and Burn
21
Thin and Burn
30
Thin
27
Control
8
Thin
21
Control/Burn
8
Tracking period
8/18/2006 – 9/4/2007
5/6/2008 – 11/10/2008
8/21/2006 – 8/27/2007
5/9/2008 – 8/1/2008
8/23/2006 – 9/4/2007
5/9/2008 – 11/10/2008
8/23/06 – 9/4/2007
5/16/2008 – 11/10/2008
5/8/2007 – 9/29/2008
5/17/2007 – 9/2/2008
5/19/2007 – 9/26/2008
6/3/2007 – 9/22/2008
6/9/2007 – 7/17/2007
6/23/2007 – 8/16/2007
7/23/2007 – 9/29/2008
7/23/2007 – 7/9/2008
5/17/2006 – 8/28/2007
5/23/2008 – 11/10/2008
7/7/2006 – 9/2/2007
7/27/2006 – 9/21/2006
4/22/2006 – 9/11/2007
4/29/2006 – 10/27/2006
8/10/2006 – 6/14/2007
5/20/2007 – 11/2/2007
4/30/2008 – 9/2/2008
5/9/2008 – 11/10/2008
5/31/2008 – 11/10/2008
4/29/2006 – 6/4/2006
6/29/2008 – 11/10/2008
5/27/2006 – 7/7/2006
M- male, F- female, G-gravid, NF-nongravid, and Unk-unknown reproductive condition.
*Individuals used for home-range estimations.
†Habitat data from these individuals was used to evaluate microhabitat use.
a
Snake Number
133*†
133J*†
134*†
134B†
135*†
135C*†
136*†
136T*†
137*†
423*†
424*†
426†
427
428†
429*†
430†
831*†
833S*†
834*†
835†
836*†
837*†
840†
425*†
425*
503*†
504*†
832
834B*†
838
Final Result
Transmitter removed
Transmitter left in snake
Transmitter removed
Depredated (unknown predator)
Transmitter removed
Transmitter left in snake
Transmitter removed
Transmitter left in snake
Transmitter removed
Transmitter battery died
Transmitter removed
Depredated (unknown predator)
Depredated (unknown predator)
Depredated (unknown predator)
Transmitter removed
Depredated (unknown predator)
Transmitter left in snake
Transmitter left in snake
Transmitter removed
Depredated (Black Kingsnake)
Transmitter removed
Transmitter removed the following spring; snake unhealthy
Transmitter battery died
NA
Transmitter battery died
Transmitter left in snake
Transmitter left in snake
Died (incision site opened)
Transmitter left in snake
Transmitter battery died
Table 6.3 Summary of total Copperheads (Agkistrodon contortrix) monitored during a
radiotelemetry study in the William B. Bankhead National Forest, Alabama, U.S.A. (2006-2008).
Gender
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
F(G)
F(NG)
F(G)
F(G)
F(G)
F(G)
Treatment Type
Thin
Thin & Burn
Burn
Thin & Burn
Thin
Thin
Thin
Thin & Burn
Thin & Burn
Thin & Burn
Thin & Burn
Thin
Thin & Burn
Control
Control/Burn
Control
Thin & Burn
Thin & Burn
Thin & Burn
Thin
Thin
Thin
Number of Locations
27
30
29
28
29
28
29
26
22
25
26
32
27
28
33
31
21
21
30
27
27
21
Evaluated using the Animal Space Use program (Horne and Garton 2006).
Evaluated using Hawth's Tools v.3.27 for Arc–GIS (Beyer 2004).
b
a
Snake Number
133
133J
134
135
135C
136
136T
137
423
424
429
831
833S
834
836
837
425
425
503
504
833
834B
Homerange Method
Fixed Kernel
Fixed Kernel
Fixed Kernel
Fixed Kernel
Fixed Kernel
Two-mode Bivariate Normal Mix
Fixed Kernel
Two-mode Bivariate Circle
Fixed Kernel
One-mode Bivariate Normal
Adaptive Kernel
Adaptive Kernel
One-mode Bivariate Normal
Fixed Kernel
Fixed Kernel
Adaptive Kernel
Fixed Kernel
Two-mode Bivariate Circle
Adaptive Kernel
Adaptive Kernel
Two-mode Bivariate Normal Mix
Fixed Kernel
a
95% Utilization Distribution (ha)
18.7
16.8
6.6
19.8
43.7
17.1
20.0
13.2
15.6
8.3
9.2
20.1
15.0
19.7
8.9
31.4
11.3
29.7
9.3
8.4
5.5
0.8
b
50% Utilization Distribution (ha) Core Utilization Distribution (ha) MCP (ha)
3.9
2.1
23.8
4.1
2.3
7.7
1.3
0.8
4.6
4.5
2.3
14.9
10.9
5.4
28.9
4.2
2.3
17.6
4.4
2.4
8.5
2.3
1.7
8.5
3.9
1.9
7.0
1.9
1.1
3.8
1.9
1.1
3.9
4.6
2.3
10.3
3.5
1.9
9.9
3.9
2.1
12.9
2.1
1.1
7.2
5.2
3.2
22.0
2.7
1.6
3.2
1.7
1.5
12.7
0.9
0.8
5.0
1.0
1.0
7.9
0.5
0.3
3.2
0.1
0.1
1.4
Table 6.4 Home-range estimates for Copperheads (Agkistrodon contortrix) in the William B. Bankhead National Forest,
Alabama, U.S.A. (2006-2008).
Figure 6.6 Utilization distributions along with minimum convex polygons for a male and
gravid female Copperhead (Agkistrodon contortrix) in the William B. Bankhead National
Forest, Alabama, U.S.A. Note the double and single activity centers for the male and
female snake, respectively.
b
4.4/6.3
5.8/7.9
0.0/1.0
0.0/1.9
4.4/9.0
5.8/11.1
16.1/15.8
11.7/13.8
30.2/22.1
30.2/13.7
16.1/16.6
11.7/17.5
11.5/12.0
13.2/11.0
5.9/15.2
5.9/3.2
11.5/6.7
13.2/7.8
10.1/4.2
13.0/4.8
0.7/2.2
0.7/2.2
10.1/4.0
13.0/4.5
2.9/3.2 HE SE PP SP DF CF
1.6/2.4 HE SP PP SE CF DF
7.2/5.9 CF DF SE HE PF SP
7.2/6.7 SE CF PP DF SP HE
2.9/6.0 HE SP SE CF DF PP
1.6/2.2 HE SP DF SE PP CF
a
0.78
0.44
0.21
0.38
0.69
0.39
Λ
0.90
2.76
0.92
0.41
1.44
3.40
Statistics
F
c
b
a
Represents area within average core utilization distribution.
Represents area within average 95% utilization distribution.
Wilk's lambda.
See Table (6.1) for descriptions of macrohabitat types and the text for details on the delineation of available habitat. Habitats that share a common underline were equally preferred.
54.3/56.7
54.8/58.2
55.9/51.9
c
Core Area Utilization Disttribution
All Snakes
21
Male Snakes
16
Gravid Females
5
54.3/59.5
54.8/55.9
% Use / % Available
Preferred↔Avoided
Mixed Deciduous Pine Plantation Cut Forest Southern Pine Beetle Hard Edge Soft Edge
(MD
(PP)
(CF)
(SP)
(HE)
(SE)
55.9/70.8
21
16
N
5
Gravid Females
95% Area Utilization Distribution
All Snakes
Male Snakes
Category
Table 6.5 Macrohabitat selection, availability, and preference by Copperheads (Agkistrodon contortrix) in the William B.
Bankhead National Forest (2006-2008).
0.51
0.07
0.64
0.81
0.26
0.04
p
Figure 6.7 Utilization distributions along with minimum convex polygons for six female
Copperheads (Agkistrodon contortrix) in the William B. Bankhead National Forest.
Gravid females (A–E); non-gravid female (F). Gravid females B, C, and E selected
thinned forest stands, whereas gravid females A and D selected forest stands with natural
canopy gaps.
A
B
C
D
E
F
e
d
c
b
a
a
196.14
192.32
195.69
189.42
206.91
207.04
206.95
227.24
227.23
232.54
257.67
261.31
260.61
434.77
434.61
488.96
527.2
–2 Log Likelihood
b
6
9
8
11
4
4
5
3
4
4
3
2
3
2
3
3
3
K
Akaike's weight value indicates relative weight of each model. Higher values indicate models with better support.
The difference between the best supported AIC model and each candidate model.
Akaike's information criterion adjusted for small samples.
Value derived from paired logistic regression output.
Number of parameters in each model.
Model
% CWD + L_Depth + Log_vol + % Herb + Bas_area
% CWD + L_Depth + Log_vol + % Herb + Bas_area + For_3 + Can_cov + AT_snake
% CWD + L_Depth + Log_vol + % Herb + Bas_area + For_3 + Can_cov
Global
% CWD + L_Depth + Bas_area
% CWD + % Herb + L_Depth
% CWD + L_Depth + Log_vol + % Herb
% CWD + L_Depth
% CWD + L_Depth + Log_vol
Bas_area + Can_cov + L_Depth
Can_cov + L_Depth
L_depth
Over_d + L_Depth
% CWD
Log_vol + % CWD
AT_snake + % Herb
Can_cov + rock_d
c
AICc
208.34
210.75
212.03
212.05
215.00
215.13
217.09
233.30
235.32
240.63
263.73
265.34
266.67
438.80
440.67
495.02
533.26
d
∆AICc
0.00
2.41
3.69
3.71
6.67
6.80
8.75
24.96
26.99
32.30
55.39
57.00
58.33
230.46
232.33
286.68
324.92
e
ωi
0.59
0.18
0.09
0.09
0.02
0.02
0.01
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Table 6.6 Microhabitat models for male Copperheads (Agkistrodon contortrix) in the William B. Bankhead National
Forest, Alabama, U.S.A. Models were evaluated using Akaike’s information criterion.
e
d
c
b
a
18.76
18.52
29.62
30.14
26.16
29.22
17.47
34.37
35.23
33.66
34.64
42.46
48.49
52.03
51.48
67.08
70.01
–2 Log Likelihood
a
b
8
9
4
4
6
5
11
3
3
4
4
3
2
2
3
3
3
K
Akaike's weight value indicates relative weight of each model. Higher values indicate models with better support.
The difference between the best supported AIC model and each candidate model.
Akaike's information criterion adjusted for small samples.
Number of parameters in each model.
Value derived from paired logistic regression output.
Model
% CWD + L_Depth + Log_vol + % Herb + Bas_area + For_3 + Can_cov
% CWD + L_Depth + Log_vol + % Herb + Bas_area + For_3 + Can_cov + AT_snake
% CWD + L_Depth + Bas_area
% CWD + % Herb + L_Depth
% CWD + L_Depth + Log_vol + % Herb + Bas_area
% CWD + L_Depth + Log_vol + % Herb
Global
% CWD + L_Depth
Over_d + L_Depth
% CWD + L_Depth + Log_vol
Bas_area + Can_cov + L_Depth
Can_cov + L_Depth
L_depth
% CWD
Log_vol + % CWD
Can_cov + rock_d
AT_snake + % Herb
c
AICc
35.10
36.95
37.71
38.23
38.36
39.36
40.10
40.43
41.29
41.75
42.73
48.52
52.52
56.06
57.54
73.14
76.07
d
∆AICc
0.00
1.85
2.61
3.13
3.26
4.26
5.00
5.33
6.19
6.65
7.63
13.42
17.42
20.96
22.44
38.04
40.97
e
ωi
0.41
0.16
0.11
0.09
0.08
0.05
0.03
0.03
0.02
0.01
0.01
0.00
0.00
0.00
0.00
0.00
0.00
Table 6.7 Microhabitat models for female Copperheads (Agkistrodon contortrix) in the William B. Bankhead National
Forest, Alabama, U.S.A. Models were evaluated using Akaike’s information criterion.
Table 6.8 Average habitat values at used and random sites for variables supported in top
ranked microhabitat models for A) male and B) female Copperheads (Agkistrodon
contortrix) in the William B. Bankhead National Forest, Alabama, U.S.A. (2006-2008).
A
Habitat Variable
% CWD
L_Depth
% Herb
Log_vol
Bas_area
Used Locations
8.0 ± 0.6
7.4 ± 0.1
35.2 ± 1.3
192.2 ± 171.4
17.7 ± 0.4
Random Locations
2.1 ± 0.3
4.4 ± 0.1
27.5 ± 1.1
181.9 ± 163.7
19.6 ± 0.4
Used Locations
10.5 ± 2.0
6.1 ± 0.4
30.0 ± 2.7
355.7 ± 84.8
18.2 ± 0.8
82.7 ± 1.6
13.9 ± 3.0
Random Locations
2.03 ± 0.61
3.99 ± 0.26
24.44 ± 2.80
241.98 ± 50.93
20.1 ± 0.9
84.4 ± 1.6
27.5 ± 4.6
B
Habitat Variable
% CWD
L_Depth
% Herb
Log_vol
Bas_area
Can_cov
For_3
Table 6.9 Model-averaged parameter estimates, standard errors, and 95% confidence
intervals for Copperhead (Agkistrodon contortrix) microhabitat models in the William B.
Bankhead National Forest, Alabama, U.S.A. (2006-2008).
Male A. contortrix
Model Term β ± S.E.
95% Confidence Interval
% CWD
0.11 ± 0.02
0.08
0.14
L_depth
0.77 ± 0.09
0.59
0.95
Log_vol
0.00 ± 0.01
–0.02
0.02
% Herb
0.02 ± 0.01
0.00
0.04
Bas_area
–0.07 ± 0.02
–0.10
–0.03
Female A. contortrix % CWD
L_depth
Log_vol
% Herb
Bas_area
For_3
Can_cov
AT_snake
0.18 ± 0.09
1.34 ± 0.54
0.00 ± 0.00
0.03 ± 0.03
–0.10 ± 0.10
–0.08 ± 0.05
–0.19 ± 0.11
–0.23 ± 0.49
0.00
0.28
0.00
–0.03
–0.30
–0.18
–0.41
–1.19
0.36
2.40
0.00
0.09
0.10
0.02
0.03
0.73
Male Closed Canopy % CWD
% Slash
L_depth
% Herb
Bas_area
Can_cov
0.14 ± 0.04
0.19 ± 0.22
0.42 ± 0.09
0.01 ± 0.01
–0.05 ± 0.05
–0.01 ± 0.02
0.06
–0.21
0.24
–0.01
–0.15
–0.05
0.22
0.65
0.60
0.03
0.05
0.03
Male Open Canopy
0.08 ± 0.03
0.06 ± 0.22
1.08 ± 0.17
0.02 ± 0.01
–0.02 ± 0.01
0.02
–0.37
0.75
0.00
0.00
0.14
0.49
1.41
0.04
–0.04
% CWD
% Slash
L_depth
% Herb
Bas_area
e
d
c
b
a
a
97.36
101.91
97.07
95.87
104.95
108.11
95.62
113.32
129.91
131.01
135.56
139.31
143.30
144.91
148.95
189.60
190.37
–2 Log Likelihood
b
6
4
7
8
5
4
10
4
4
4
3
2
3
3
2
3
3
K
Akaike's weight value indicates relative weight of each model. Higher values indicate models with better support.
The difference between the best supported AIC model and each candidate model.
Akaike's information criterion adjusted for small samples.
Number of parameters in each model.
Value derived from paired logistic regression output.
Model
% CWD + % Slash + L_Depth + % Herb + Bas_area
% CWD + L_Depth + Bas_area
% CWD + % Slash + L_Depth + % Herb + Bas_area + Can_cov
% CWD + % Slash + L_Depth + % Herb + Bas_area + Can_cov + AT_snake
% CWD + % Slash + L_Depth + % Herb
% CWD + L_depth + % Herb
Global
% CWD + % Slash + L_Depth
Bas_area + Can_cov + L_depth
% CWD + % Slash + % Herb
% CWD + % Slash
% CWD
Can_cov + L_depth
Over_d + L_depth
L_depth
AT_snake + % Herb
Can_cov + For_3
c
AICc
109.59
110.02
111.37
112.26
115.11
116.22
116.22
121.43
138.02
139.12
141.62
143.34
149.36
150.97
152.98
195.66
196.43
∆AICc
0.00
0.43
1.79
2.67
5.52
6.63
6.63
11.84
28.43
29.53
32.04
33.76
39.78
41.39
43.40
86.08
86.85
d
Table 6.10 Microhabitat models for male Copperheads (Agkistrodon contortrix) in closed canopy stands of the
William B. Bankhead National Forest, Alabama, U.S.A. (2006-2008).
e
ωi
0.38
0.31
0.16
0.10
0.02
0.01
0.01
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
e
d
c
b
a
a
91.56
90.57
89.10
93.27
88.74
95.13
87.95
84.63
99.10
105.57
104.83
105.50
222.15
262.45
289.93
306.62
329.15
–2 Log Likelihood
b
4
5
6
4
7
4
8
10
4
2
3
3
4
3
2
3
3
K
Akaike's weight value indicates relative weight of each model. Higher values indicate models with better support.
The difference between the best supported AIC model and each candidate model.
Akaike's information criterion adjusted for small samples.
Number of parameters in each model.
Value derived from paired logistic regression output.
Model
% CWD + L_depth + % Herb
% CWD + % Slash + L_Depth + % Herb
% CWD + % Slash + L_Depth + % Herb + Bas_area
% CWD + L_Depth + Bas_area
% CWD + % Slash + L_Depth + % Herb + Bas_area + Can_cov
% CWD + % Slash + L_Depth
% CWD + % Slash + L_Depth + % Herb + Bas_area + Can_cov + AT_snake
Global
Bas_area + Can_cov + L_depth
L_depth
Over_d + L_depth
Can_cov + L_depth
% CWD + % Slash + % Herb
% CWD + % Slash
% CWD
AT_snake + % Herb
Can_cov + For_3
c
AICc
99.67
100.73
101.33
101.38
103.04
103.24
104.34
105.23
107.21
109.60
110.89
111.56
230.26
268.51
293.96
312.68
335.21
d
∆AICc
0.00
1.06
1.66
1.71
3.38
3.57
4.67
5.56
7.54
9.93
11.23
11.90
130.59
168.85
194.29
213.02
235.55
e
ωi
0.33
0.20
0.15
0.14
0.06
0.06
0.03
0.02
0.01
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Table 6.11 Microhabitat models for male Copperheads (Agkistrodon contortrix) in open canopy stands of the William
B. Bankhead National Forest, Alabama, U.S.A. (2006-2008).
Table 6.12 Average habitat values at used and random sites for variables supported in top
ranked microhabitat models male Copperheads (Agkistrodon contortrix) in A) closed
canopy stands and B) open canopy stands of the William B. Bankhead National Forest,
Alabama, U.S.A. (2006-2008).
A
Habitat Variable
% CWD
% Slash
L_Depth
% Herb
Bas_area
Can_cov
Used Locations
10.1 ± 1.0
0.4 ± 0.2
6.9 ± 0.2
28.3 ± 1.9
20.4 ± 0.7
85.1 ± 14.0
Random Locations
2.0 ± 0.4
0.03 ± 0.03
4.8 ± 0.2
25.2 ± 1.8
22.9 ± 0.8
82.3 ± 10.0
Used Locations
6.8 ± 0.7
4.0 ± 1.0
7.6 ± 0.2
39.3 ± 1.7
16.1 ± 0.5
Random Locations
2.2 ± 0.4
0.02 ± 0.02
4.2 ± 0.1
29.0 ± 1.4
17.7 ± 0.5
B
Habitat Variable
% CWD
% Slash
L_Depth
% Herb
Bas_area
CHAPTER 7
THE USE OF ARTIFICIAL COVER OBJECTS TO DETERMINE HERPETOFAUNAL
RESPONSE TO FOREST DISTURBANCES
Introduction
Understanding organismal responses to anthropogenic disturbances is essential for
longterm species conservation. Due to recent reports of worldwide amphibian and reptile
declines, it is essential to understand the impacts of stochastic and anthropogenic
disturbances on these species (Gibbons et al. 2000; Stuart 2004). Because forest
management practices alter large areas of the landscape, these disturbances are often
implicated as a contributing factor in these declines (Gibbons et al. 2000). Amphibians
and reptiles collectively known as herpetofauna can reach high densities in eastern forests
(Burton and Likens 1979a; Petranka and Murray 2001) and have been shown to play
essential roles in predator / prey relationships (Fitch 1949; Burton and Likens 1979b) and
nutrient cycling (Wyman 1998), so it is important to understand the potential impacts of
forest disturbances on these species and the associated ecological processes.
A majority of herpetofaunal forest management studies have been biased towards
the examination of herpetofaunal response to clearcut harvesting (Ash 1988; Ash 1997;
Duguay and Wood 2002; Goldstein 2005), whereas little research has examined
herpetofaunal response to forest thinning (but see Brooks 1999; Grialou 2000; Naughton
et al. 2000). Few studies have examined herpetofaunal response to prescribed burning,
but prescribed fire appears to have negligible impacts on amphibians (Ford et al. 1999;
Moseley et al. 2003; Greenberg and Waldrop 2008) and short-term positive effects for
certain species (Mushinsky 1985; Wilgers and Horne 2006; Hossack et al. 2009), whereas
reptile response to prescribed burning appears to be either positive (Moseley et al. 2003;
Wilgers and Horne 2006; Kilpatrick et al. 2010) or negligible (McLeod and Gates 1998).
A majority of studies have also focused on salamander response (Messere and
Ducey 1998; Harpole and Haas 1999; Herbeck and Larsen 1999; Knapp et al. 2003;
McKenny et al. 2006), whereas reptiles have largely been ignored (Greenberg 2000). Of
all herpetofaunal groups, studies of fossorial snakes are lacking. Besides recent studies
examining ecological and natural history aspects of these species (Willson and Dorcas
2004; Todd et al. 2008a, b), very little has been published regarding the response of these
species to stochastic or anthropogenic disturbances (but see Todd and Andrews 2007).
Drift-fence sampling has proven very effective for sampling amphibians and
reptiles, but not all species are sampled effectively using this methodology (Hutchens and
DePerno 2009). Small, fossorial snakes and salamanders can be difficult to survey with
pitfall traps (Willson and Dorcas 2004) and other methods should be used to monitor
these species. Coverboards serve as a way to target these fossorial species and have been
used to determine salamander response to silvicultural practices (Brooks 1999; Harpole
and Hass 1999) and forest edges (DeGraaf and Yamasaki 2002; Semlitsch 2007).
The focus of this study is to evaluate the response of fossorial herpetofauna to
prescribed burning and thinning using artificial cover objects. I predict that plethodontid
salamanders will respond negatively to forest stands that have been treated with
prescribed burning and thinning, whereas most reptiles will significantly increase in
thinned stands.
Materials and Methods
Study Site
My study was centered in the northern portion of the William B. Bankhead
National Forest (BNF), located in Lawrence, Winston, and Franklin Counties, of
northwestern Alabama. Bankhead National Forest is a 72,800 ha multi-use forest located
along the highly dissected portion of the southern Cumberland Plateau (Smalley 1982;
Gaines and Creed 2003). Soils within this region are typically composed of HartsellsRock and limestone-Hector (Smalley 1982). Mixed forests of the southern Cumberland
Plateau tend to be dominated by oak-hickory forest types (McWilliams 1991) except in
areas where pines were actively planted for commercial purposes. Loblolly Pine (Pinus
taeda L.) was used to re-establish forest conditions in abandoned agricultural and heavily
timbered areas (Gaines and Creed 2003). Reforestation efforts along with natural growth
have resulted in 31,600 ha of Loblolly Pine throughout BNF (Gaines and Creed 2003).
For the past decade, Southern Pine Beetle (Dendroctonus frontalis Zimmermann)
infestations have affected Loblolly Pine stands, producing large numbers of standing
dead trees and increased fuel loads, elevating the risk of damaging wildfires. Because
canopy removal and fire disturbance have been prevented in forests throughout the study
area for decades, the BNF initiated a forest restoration plan to reduce wildfire risk and
promote natural forest growth through tree thinning and prescribed fire disturbance. The
BNF has not traditionally utilized prescribed fire as a management tool, but has opted to
include prescribed burning in the forest restoration plan due to administrative
recommendations. Forest restoration plans in BNF mirror regulations set forth in the
Healthy Forest Restoration Act, which authorizes advanced vegetation management
projects when specified conditions (existence of insect or disease epidemic) pose a
significant threat to ecosystem health (Healthy Forest Restoration Act 2003).
The selected forest stands were generally located on upland sites composed of
Loblolly Pine 25-50 years of age that also possessed a considerable hardwood component
(Gaines and Creed 2003; Schweitzer and Tadessee 2004). At the time of this study, these
stands had not been recently harvested, and each stand had varying levels of damage
from the Southern Pine Beetle. Other past disturbances included the clearing of hardwood
stands throughout the region for loblolly pine plantations during the early 1970’s (Gaines
and Creed 2003).
Forest Treatments
The Bankhead National Forest implemented this research designed and developed
by Alabama A&M University and the Southern Research Station of the USDA Forest
Service as part of a longterm examination of forest management impacts on forested
ecosystems in northwestern Alabama. The experiment consisted of a before-after,
control-impact (BACI), complete block design. The process of assigning treatments to
each forest stand was not a fully random process because forest treatment designations
had to align with the longterm management goals of the BNF. For example, forest stands
assigned to prescribed burn treatments had to be located in a portion of BNF that was
designated a burn area in the original forest restoration management plan. Forest
manipulation treatments consisted of a two by three factorial arrangement of three
thinning levels (no thin, 11 m2 ha-1 residual basal area [BA], and 17 m2 ha-1 residual BA)
along with two burn treatments (no burn and burn) equaling six treatments per replicate.
In addition to single thinning and burning treatments, I also evaluated the interaction of
each thinning level with a burn treatment. Each of the six treatments in this experiment
was replicated three times across the landscape (Figure 7.1), with each treatment
approximately 9 ha in size.
Due to limited resources and difficulties implementing this large scale study in a
single year, treatments were blocked temporally (i.e., year). Block 1 was treated during
the summer of 2005, while block 2 and block 3 were treated during the summer and fall
of 2006. All harvesting procedures were thin-from-below and were completed at two
levels of BA retention. Thinning-from-below is a harvesting procedure that targets trees
from the lower crown classes in order to provide limited resources (e.g., water and light)
to dominant and co-dominant trees (Smith et al. 1997). All harvesting was completed by
feller bunchers and trees were harvested until the desired residual BA had been achieved.
Hardwood tree species, such as Quercus spp. and Carya spp. were preferentially retained
during the thinning as much as possible. All thinning procedures in a particular block
were completed in the same year (i.e., block one 2005 and blocks two and three 2006).
Prescribed burns consisted of low-intensity fires, which were completed during
the dormant season (February-March) when air temperatures were low and relative
humidity was high. Backing fires were initiated to ensure that prescribed burns were
limited to understory and litter layers. Prescribed burns were ignited with drip torches and
managed by the BNF U.S. Forest Service. Fires were generally low intensity and rarely
reached more than two meters in height. Once the designated area was blocked off and
ignited, fires were permitted to burn until all available fuel was consumed and the fires
diminished.
Climate and Habitat Parameters
One HOBO© (Onset Computer Corp.) datalogger was installed at each trapping
array to record air temperature, soil temperature, relative humidity, and light intensity
(Table 7.1). Dataloggers were programmed to record measurements every twelve hours
starting at 2:00 PM. I used climate information from May 15-June 15 during pre- and
post-treatment surveys. Climate data were limited to this one-month period due to the
short pre-treatment sampling period for block one.
I recorded pre- and post-treatment habitat complexity and heterogeneity data via
yearly line transect surveys at each treatment plot. I determined plot placement a priori
via a random compass bearing (0-360°) and distance (30-50 m) originating from the
center point of all six coverboard arrays. In order to quantify the degree of habitat
disturbance, I completed habitat surveys in approximately the same location for each
year. Each habitat survey consisted of two 20 m perpendicular transects placed northsouth and east-west from the habitat plot center. I used a two meter piece of 1.9 cm
diameter polyvinyl chloride pipe as a transect marker and recorded the presence or
absence of a suite of microhabitat variables (Table 7.1). Basal area was determined at the
stand level as an average of five 0.08 hectare plots (Table 7.1).
Herpetofaunal Sampling
I used artificial cover objects (i.e., coverboards) to evaluate herpetofaunal
response to forest management. Coverboards are wooden cover objects used to estimate
amphibian and reptile population parameters (Fellers and Drost 1994; Dodd 2003). With
this method, amphibians and reptiles are sampled by turning over the coverboards and
capturing the animals underneath. This technique is useful because it permits a researcher
to standardize the amount and area of cover objects sampled across multiple study plots
(Fellers and Drost 1994).
In each stand I alternated each of three small board and large board (61.0 cm x
121.9 cm) arrays that were cut from (243.8 cm x 121.9 cm x 1.3 cm) OSB boards and
placed each board array 50 m from the trap array (Figure 7.2). Each large-board array
was composed of four boards (12 per treatment), whereas each small board array was
composed of 20 boards (60 per treatment). In 2005 I used three-layer plywood to for
board material, but found these boards warped considerably resulting in drying under the
boards. Therefore OSB boards were used for all successive surveys. I placed each board
array 50 m from a centralized drift-fence array. Boards were placed in the study plots in
early November and were allowed to weather for approximately 2.5 months before
surveys. I began weekly surveys each year in mid-January and continued until the first
week in June, and ended surveys in early summer because the surfaces under the boards
were usually dry at this time. Boards were originally placed into treatment plots to obtain
pre-treatment data and were removed before the forest treatments were initiated. After the
forest stands had been treated, I placed the boards back in the same place where pretreatment surveys were completed. I did not begin collecting post-treatment data until
spring 2007. This was done because prescribed burns were not completed until the spring
after the implementation of the harvesting treatments. Therefore, boards were not placed
back in the stands until the fall after implementation of the forest treatments. I was able to
collect one season of pre-treatment data and two seasons of post-treatment data.
I recorded snout-vent length (mm) and total length (mm) with a dial caliper and
mass (g) with a digital scale, along with the type of coverboard (i.e., large or small) for
each animal capture. Captured animals received a treatment and subplot specific mark to
denote recaptures by toe-clipping (salamanders and lizards), scale-clipping (snakes), and
scute-etching (turtles). After marking, each animal was released under the board it was
originally captured.
Statistical Analysis
To characterize vegetation and environmental characteristics of the study plots, I
conducted a principal components analysis on the habitat survey data. I analyzed the preand post-treatment habitat data simultaneously to compare the resulting components
among treatment years.
To standardize amphibian and reptile count data, I divided the total number of
captures by the number of surveys during each year and multiplied these values by 100. I
used mixed models ANOVA (PROC MIXED; SAS v.9.3) to test for differences among
forest treatment types and years. For each of these models, treatment represented the
fixed effects, block represented the random effect, and year represented the repeated
measure. I normalized all amphibian and reptile capture data using square root and
logistic transformations. For all analyses, I declared significance at an alpha level ≤ 0.05.
Results
Habitat Response
Using PCA, I was able to extract five biologically relevant components that
explained 81.4% of the variance (component 1 [48.5%], component 2 [13.9%],
component 3 [8.3%], component 4 [6.1%], and component 5 [4.6%]) in the orginal
climate and habitat dataset (Table 7.2). Component one described a dichotomy ranging
from sites with greater canopy cover, percent humidity, basal area, and litter depth to
sites with greater temperatures, light intensity, slash cover, and herbaceous growth,
whereas component two described a gradient ranging from sites with greater litter,
herbaceous, and woody vegetation cover, to sites with greater bare ground cover (Figure
7.3). Components three and four described relationships in CWD and rock cover,
respectively, whereas component five described a gradient of overstory tree cover.
Pre-treatment habitat structure was similar among study plots and changed
drastically after forest treatments were implemented (Figure 7.3). Treatment plots
differed greatly in terms of habitat structure directly after treatment, but during the
second year, I found that litter coverage increased considerably in burned stands when
compared to the first year post-treatment habitat data (Figure 7.3).
Herpetofaunal Response
I captured 1074 individual amphibians and reptiles representing 14 species during
173 sampling days (i.e., 73,664 board days; Table 7.3). Each treatment stand was
sampled approximately 20 times during each year, except for block one pre-treatment
surveys in 2005, which were only sampled eight times throughout the sampling season
due to a short pre-treatment sampling period.
Southern Zig-zag Salamanders (Plethodon ventralis) represented the most
commonly captured amphibian species with 665 total captures, whereas Ringneck Snakes
(Diadophis punctatus) were the most commonly captured reptile species with 58 total
captures (Table 7.3). I recorded 859 total recaptures, with Southern Zig-zag Salamanders
and Eastern Worm Snakes (Carphophis a. amoenus) being the most commonly
recaptured species with, 382 and 23 recaptures, respectively (Table 7.3).
Of the four most commonly detected species, I captured 458 individuals using
large boards and 574 individuals using small boards. The majority of this variation
between board size was due to Southern Zig-zag Salamander captures (Figure 7.4), but
this difference was not statistically significant. Captures of Mississippi Slimy
Salamanders (Plethodon mississippi), Ringnecked Snakes, and Eastern Worm Snakes
were consistent across large and small coverboards and was not statistically different
(Figure 7.4).
I was unable to detect any impacts of forest management on plethodontid
salamanders. Counts of Southern Zig-zag Salamanders were significantly greater in
thinned stands (F2, 34 = 16.99; p < 0.0001) stands and lowest in control stands throughout
the entire study period and varied little across treatment stands throughout all years
(Figure 7.5). Similarly, Mississippi Slimy Salamanders (Plethodon mississippi) were
unaffected by forest management treatments. Although counts were greatest in burn
stands across all years, counts declined considerably across all treatment stands during
the first year post-treatment indicating a highly significant year effect (F2, 34 = 8.44; p =
0.001; Figure 7.6).
I detected a significant thin*burn*year effect (F4, 36 = 2.70; p = 0.045) on Ringnecked Snakes, which increased considerably in thin*burn stands during post-treatment
year two (Figure 7.7). Similarly, I detected a significant thin*burn*year effect (F4, 36 =
4.50; p = 0.005) on total fossorial snakes, which also increased in thin*burn stands during
post-treatment year two (Figure 7.8).
Size class distributions of Ring-necked Snakes were different among closed and
open canopy types. Average SVL of Ring-necked Snakes in closed canopy stands (mean
SVL = 249.5 mm) was larger (F2, 58 = 5.38; p = 0.007) than snakes captured in open
canopy stands (post-treatment year one mean SVL = 198.4 mm; post-treatment year two
mean SVL = 180.6 mm; Figure 7.9). Although average SVL’s for snakes in open canopy
stands differed between years, these differences were not significant (Figure 7.9).
Discussion
Past studies examining wildlife responses to forest management have been limited
in overall inference due to inadequate experimental designs, lack of pre-treatment data,
low replication number, and short survey periods (Marzluff et al. 2000; Russell et al.
2004). My study alleviated many of these problems, but most importantly I was able to
obtain one-year of pre-treatment survey data. Having access to this data greatly increased
the level of inference of this study. Researchers should work alongside forest managers
during the project planning process to communicate the importance of having access to
pre-treatment data.
My analyses revealed species-specific responses of herpetofauna to various levels
of thinning and prescribed burning. My results indicate that these forest treatments have
relatively little negative impacts on species inhabiting mixed pine-hardwood forests in
northern Alabama. This is particularly important because many mixed-hardwood stands
throughout the region are managed with thinning to reduce the spread of the southern
pine beetle. I found that neither Southern Zig-zag Salamanders nor Mississippi Slimy
Salamanders were negatively impacted by these treatments. Past studies of salamander
response to forest management have found that salamanders generally respond negatively
to clearcutting (Grialou 2000; Knapp et al. 2003; Karraker and Welsh 2006; Homyack
and Haas 2009), but have shown either negative responses (Grialou et al. 2000; Naughton
et al. 2000) or negligible responses (Brooks 1999) to thinning operations. I predicted that
increased air temperatures in thinned sites would cause declines in resident salamander
populations. However, plethodontid salamanders in northern Alabama are active in the
cooler and wetter months of the year (January-April), which may have reduced the
impacts of timber harvesting on salamanders in this study.
Fossorial snake counts (mainly Ring-necked Snakes and Eastern Worm Snakes)
increased considerably in thin*burn stands after the implementation of forest treatments.
Little research has evaluated fossorial snake response to forest management, but species
within this guild appear to be negatively impacted by forest management disturbances
that cause large reductions of litter and coarse woody debris cover (Todd and Andrews
2007). Because thermoregulation is important for many reptile species (Pianka and Vitt
2003), fossorial snake counts may have increased in thin*burn plots due to the more open
nature and greater availability of thermoregulation sites in these stands. Additionally, the
increased use of coverboards by these snakes during post-treatment surveys may have
been due to the reduction in certain habitat features (i.e., reduced litter depth and litter
cover). There was an increase in CWD cover in these sites, with a concomitant decrease
in overall CWD volume. The higher CWD volume recorded during pre-treatment surveys
was due to treefall caused by Southern Pine Beetle damage and the reduction in CWD
volume after treatment occurred during harvesting as feller bunchers smashed these
larger logs. Although the importance of CWD for some reptile species is known (James
and M’Closkey 1998; Angert et al. 2002), it is unknown how CWD volume and state of
decay is linked to reptile population parameters. I also found that average SVL for Ringnecked Snakes was smaller than snakes in closed canopy sites. The transition to smaller
size classes along with concomitant increases in snake counts suggests that juvenile
recruitment is greater in thin*burn stands. Size-class differences have also been found for
Ring-necked Snakes in forest stands disturbed by prescribed burning (Wilgers and Horne
2006). Although this does provide some support that thinning and burning increase
juvenile fossorial snake recruitment, it is unknown whether this result is due to sizedependent habitat use differences caused by canopy disturbance.
Coverboards are a useful passive sampling technique for evaluating herpetofaunal
occupancy patterns. However, the validity of using coverboards to determine salamander
population parameters has been questioned due to biases towards larger salamander size
classes (Houze and Chandler 2002; Dodd 2003; Marsh and Goicochea 2003). I did not
see biases towards larger salamander size classes, but found that juvenile Mississippi
Slimy Salamanders were captured more frequently under coverboards than adult
conspecifics. My observations also indicated that adult Mississippi Slimy Salamanders
were more commonly captured under coverboards that had weathered for one full season.
Conversely, all size classes of Southern Zig-zag Salamanders readily used coverboards
one to two months after placement in study plots. These observations indicate the
importance of understanding potential biases associated with using coverboards to
evaluate salamander population parameters. No published research comparing fossorial
snake occupancy patterns of artificial cover objects versus natural cover objects exists,
but salamanders are more likely to use various natural cover objects (Richmond and
Trombulak 2009) when compared to artificial cover objects (Houze and Chandler 2002).
There are conflicting results of whether coverboard surveys provide better estimates of
species occupancy patterns when compared to visual encounter surveys (Ryan et al.
2002; Hutchens and DePerno 2009). Because visual encounter surveys were not used in
this study, my inference is limited to disturbed sites that have been sampled with artificial
cover objects.
Coverboards are useful for detecting many species of herpetofauna; however,
because many of these species are not detected equally, there are errors associated with
lack of detection. Imperfect detection can be due to numerous factors, such as
environmental conditions, surveyor experience, and behavioral characteristics of the
target species (Mazerolle et al. 2007). Studies that do not estimate detection rates are
greatly limited because there are errors associated with species that are not detected at
constant rates throughout the study period (Dodd and Dorazio 2004). Therefore my study
may be limited because the changes in amphibian and reptile counts seen in this study
may be partially due to errors in detection or detection error associated with coverboard
used (Hyde and Simons 2001) and not the silvicultural treatments. Although the
availability of pre-treatment data does offer some confidence as to the influence of the
forest treatments on herpetofaunal response, future studies should integrate detection
probabilities to correct for errors of imperfect detection.
Bibliography
Angert, A. L., D. Hutchison, D. Glossip, and J. B. Losos. 2002. Microhabitat use and
thermal biology of the collared lizard (Crotaphytus collaris collaris) and the fence
lizard (Sceloporus undulatus hyacinthinus) in Missouri glades. Journal of
Herpetology 36:23–29.
Ash, A. N. 1988. Disappearance of salamanders from clearcut plots. The Journal of the
Elisha Mitchell Scientific Society 104:116–122.
Ash, A. N. 1997. Disappearance and return of Plethodontid salamanders to clearcut plots
in the southern Blue Ridge Mountains. Conservation Biology 11:983–989.
Burton, T. M. and G. E. Likens. 1979a. Salamander populations and biomass in the
Hubbard Brook Experimental Forest, New Hampshire. Copeia 1979:541–546.
Burton, T. M. and G. E. Likens. 1979b. Energy flow and nutrient cycling in salamander
populations in the Hubbard Brook Experimental Forest, New Hampshire. Ecology
56:1068–1080.
Degraaf, R. M., and M. Yamasaki. 2002. Effects of edge contrast on redback salamander
distribution in even-aged northern hardwoods. Forest Science 48:351–363.
Dodd, C. K. Jr. 2003. Monitoring amphibians in Great Smoky Mountains National
Park. U. S. Department of the Interior, U. S. Geological Survey. Circular 1258:1–
118.
Dodd, C. K. Jr., and R. M. Dorazio. 2004. Using counts to simultaneously estimate
abundance and detection probabilities in a salamander community. Herpetologica
60:468–478.
Duguay, J. P., and P. B. Wood. 2002. Salamander abundance in regenerating forest
stands on the Monongahela National Forest, West Virginia. Forest Science 48:
331–335.
Fellers, G. M., and C. A. Drost. 1994. Sampling with artificial cover. Pages 146–150 in
W. R. Heyer, M. A. Donnelly, R. W. McDiarmid, L. C. Hayek, and M. S. Foster,
editors. Measuring and monitoring biological diversity. Standard methods for
amphibians. Smithsonian Institute Press, Washington and London.
Fitch, H. S. 1949. Study of snake populations in central California. American Midland
Naturalist 41:513–579.
Ford, W. M., M. A. Menzel, D. W. McGill, J. Laerm, and T. S. McCaye. 1999. Effects of
a community restoration on small mammals and herpetofauna in the southern
Appalachians. Forest Ecology and Management 114:233–243.
Gaines, G. D. and J. W. Creed. 2003. Forest health and restoration project. National
forests in Alabama, Bankhead National Forest Franklin, Lawrence, and Winston
Counties, Alabama. Final environmental impact statement. Management Bulletin R8–
MB 110B. 353 pp.
Gibbons, 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 50:653–656.
Goldstein, M. E., R. N. Wilkins, and T. E. Lacher, Jr. 2005. Spatiotemporal responses of
reptiles and amphibians to timber harvest treatments. Journal of Wildlife Management
69:525–539.
Greenberg, C. H. 2000. Response of reptile and amphibian communities to canopy gaps
created by wind disturbance in the southern Appalachians. Forest Ecology and
Management 148:135–144.
Greenberg, C. H., and T. A. Waldrop. 2008. Short-term response of reptiles and
amphibians to prescribed fire and mechanical fuel reduction in a southern
Appalachian upland hardwood forest. Forest Ecology and Management 255:2883–
2893.
Grialou, J. A, S. D. West, and R. N. Wilkins. 2000. The effects of forest clearcut
harvesting and thinning on terrestrial salamanders. Journal of Wildlife Management
64:105–113.
Harpole, D. N. and C. A. Haas. 1999. Effects of seven silvicultural treatments on
terrestrial salamanders. Forest Ecology and Management 114:349–356.
Healthy Forest Restoration Act. Act of December 3, 2003. 117 Stat., as amended; 16
U.S.C. 1887–1915.
Herbeck, L. A. and D. R. Larsen. 1999. Plethodontid salamander response to silvicultural
practices in Missouri Ozark forests. Conservation Biology 13:623–632.
Homyack, J. A., and C. A. Haas. 2009. Long-term effects of experimental forest
harvesting on abundance and reproductive demography of terrestrial salalamanders.
Biological Conservation 142:110–121.
Hossack, B. R., L. A. Eby, C. G. Guscio, and P. S. Corn. 2009. Thermal characteristics of
amphibian microhabitats in a fire-disturbed landscape. Forest Ecology and
Management 258:1414–1421.
Houze, C. M. Jr., and C. R. Chandler. 2002. Evaluation of coverboards for sampling
terrestrial salamanders in south Georgia. Journal of Herpetology 36:75–81.
Hutchens, S. J., and C. S. DePerno. 2009. Efficacy of sampling techniques for
determining species richness estimates of reptiles and amphibians. Wildlife Biology
15:113–122.
Hyde, E. J., and T. R. Simons. 2001. Sampling plethodontid salamanders: sources of
variability. Journal of Wildlife Management 65:624–632.
James, S.E., and R. T. M’Closkey. 2003. Lizard microhabitat and fire fuel management.
Biological Conservation 114:293–297.
Karraker, N. E., and H. H. Welsh, Jr. 2006. Long-term impacts of even-aged timber
management on abundance and body condition of terrestrial amphibians in
northwestern California. Biological Conservation 131:132–140.
Kilpatrick, E. S., T. A. Waldrop, J. D. Lanham, C. H. Greenberg, and T. H. Contreras.
2010. Short-term effects of fuel reduction treatments on herpetofauna from the
southeastern United States. Forest Science 56:122–130.
Knapp, S. M., C. A. Haas, D. N. Harpole, and R. L. Kirkpatrick. 2003. Initial effects of
clearcutting and alternative silvicultural practices on terrestrial salamander abundance.
Conservation Biology 17:752–762.
Marsh, D. M., and M. A. Goicochea. 2003. Monitoring terrestrial salamanders: biases
caused by intense sampling and choice of cover objects. Journal of Herpetology
37:460–466.
Marzluff, J. M., M. G. Raphael, and R. Sallabanks. 2000. Understanding the effects of
forest management on avian species. Wildlife Society Bulletin 28:1132–1143.
Mazerolle, M. J., L. L. Bailey, W. L. Kendall, J. A. Royle, S. J. Converse, and J. D.
Nichols. 2007. Making great leaps forward: accounting for detectability in
herpetological field studies. Journal of Herpetology 41:672–689.
McKenny, H. C., W. S. Keeton, and T. M. Donovan. 2006. Effects of structural
complexity enhancement on eastern red-backed salamander (Plethodon cinereus)
populations in northern hardwood forests. Forest Ecology and Management 230:186–
196.
McLeod, R. F and J. E. Gates. 1998. Response of herpetofaunal communities to forest
cutting and burning at Chesapeake Farms, Maryland. American Midland Naturalist
139:164–177.
McWilliams, W. H. 1991. Availability of yellow pine sawtimber in Alabama. Pages 115–
119 in R.H. Jones, editor. Proceedings: Alabama’s forest resources: past, present, and
future. Auburn University, AL.
Messere, M. and P. K. Ducey. 1998. Forest floor distribution of northern redback
salamanders, Plethodon cinereus, in relation to canopy gaps: first year following
selective logging. Forest Ecology and Management 107:319–324.
Moseley, K. R., S. B. Castleberry, and S. H. Schweitzer. 2003. Effects of prescribed fire
on herpetofauna in bottomland hardwood forests. Southeastern Naturalist 2:475–
486.
Mushinsky, H. R. 1985. Fire and the Florida sandhill herpetofaunal community: with
special attention to responses of Cnemidophorus sexlineatus. Herpetologica 41:333–
342.
Petranka, J. W. and S. S. Murray. 2001. Effectiveness of removal sampling for
determining salamander density and biomass: A case study in an Appalachian
streamside community. Journal of Herpetology 35:36–44.
Pianka, E. R., and L. J. Vitt. 2003. Lizards: windows to the evolution of diversity.
University of California press, Berkley, California. 333 pp.
Richmond, L. S., and S. C. Trombulak. 2009. Distribution of red-backed salamander
(Plethodon cinereus) with respect to cover object characteristics in the green
mountains of Vermont. Northeastern Naturalist 16:13–26.
Russell, K. R., T. B. Wigley, W. M. Baughman, H. G. Hanlin, and W. M. Ford. 2004.
Responses of southeastern amphibians and reptiles to forest management: a review.
Pages 319–334 in M. H. Rauscher and K. Johnsen, editors. Southern forest science:
past, present, and future. United States Department of Agriculture Southern Research
Station Gen. Tech. Rep. SRS–75.
Ryan, T. J., T. Philippi, Y. A. Leiden, M. E. Dorcas, T. B. Wigley, and J. W. Gibbons.
2002. Monitoring herpetofauna in a managed forest landscape: effects of habitat types
and census techniques. Forest Ecology and Management 167:83–90.
Schweitzer, C. J. and W. Tadesse. 2004. The interaction of fire frequency and stand
density on plant community dynamics. Pages 1–14 in CREST–Center for Forest
Ecosystem Assessment (CFEA) Alabama A&M University. National Science
Foundation proposal.
Semlitsch, R. D., T. J. Ryan, K. Hamed, M. Chatfield, B. Drehman, N. Pekarek, M.
Spath, and A. Watland. Salamander abundance along road edges and within
abandoned logging roads in Appalachian forests. Conservation Biology 21:159–
167.
Smalley, G. W. 1982. Classification and evaluation of forest sites on the mid-Cumberland
Plateau. U.S. Department of Agriculture Forest Service, General Technical Report
S0–38. Southern Forest Experimental Station, New Orleans, Louisiana. 58 pp.
Smith, D. M., B. C. Larson, M. J. Kelty, and P. M. S. Ashton. 1997. The practice of
silviculture: Applied forest ecology. John Wiley and Sons Inc. 537 pp.
Stuart, S. N., J. S. Chanson, N. A. Cox, B. E. Young, A. S. L. Rodriques, D. L. Fischman,
and R. W. Waller. 2004. Status and trends of amphibian declines and extinctions
worldwide. Science 1103538:1–8.
Todd, B. D., and K. M. Andrews. 2007. Response of a reptile guild to forest harvesting.
Conservation Biology 22:753–761.
Todd, B. D., J. D. Willson, C. T. Winne, and J. W. Gibbons. 2008a. Aspects of the
ecology of the earth snakes (Virginia valeriae and V. striatula) in the upper Coastal
Plain. Southeastern Naturalist 7:349–358.
Todd, B. D., J. D. Willson, C. T. Winne, R. D. Semlitsch, and J. W. Gibbons. 2008b.
Ecology of the southeastern crowned snake, Tantilla coronata. Copeia 2008:388–
394.
Wilgers, D. J. and E. A. Horne. 2006. Effects of different burn regimes on tallgrass
prairie herpetofaunal species diversity and community composition in the Flint Hills,
Kansas. Journal of Herpetology 40:73–84.
Willson, J. D., and M. E. Dorcas. 2004. Aspects of the ecology of small fossorial snakes
in the western Piedmont of North Carolina. Southeastern Naturalist 3:1–12.
Wyman, R. L. 1998. Experimental assessment of salamanders as predators of detrital
food webs: effects on invertebrates, decomposition and the carbon cycle. Biodiversity
and Conservation 7:641–650.
Figure 7.1 Locations of study sites within the northwestern portion of the William B.
Bankhead National Forest, Alabama, U.S.A. Forest treatments consisted of the six
following categories: C-control, B-burn, HT-heavy thin, LT-light thin, HTB-heavy thin
and burn, and LTB-light thin and burn. Numbers after treatment abbreviations correspond
to block number.
HT3
HTB1
LT3
LT2
C3
HT2
HTB2
HT1
C2
LT1
C1
LTB2 HTB3
B3
LTB1
LTB3
B1
B2
Bas_area Cross-sectional area of trees within a 0.08 hectare plot. Overall plot value (m /ha) taken as average of five 0.08 hectare plots determined by measuring DBH of all trees 15 cm and larger.
Basal Area
2
Calculated as volume of a cylinder (m ) for each enumerated CWD (see text).
Determined by measuring depth of the substrate to the nearest 0.5 cm with a metric ruler measured at every 2 m.
Estimated with a spherical densiometer as the sum percentage of open points subtracted from 100% measured at every 5 m.
Percent coverage of forest levels ≤ 2m (classified as ground cover) measured at every 5 m.
Percent coverage of forest levels > 2 m – ≤ 4 m (classified as understory) measured at every 5 m.
Percent coverage of forest levels > 4 m – ≤ 6 m (classified as midstory) measured at every 5 m.
Percent coverage of forest levels > 6 m (classified as overstory) measured at every 5 m.
Average daily air temperature (⁰C) during the month of June recorded at 2:00 PM
Average daily soil temperature (⁰C) during the month of June recorded at 2:00 PM
Average daily light intensity (%) during the month of June recorded at 2:00 PM
Average daily relative humidity (%) during the month of June recorded at 2:00 PM
3
CWD_vol
L_dep
Can_cov
For_lev1
For_lev2
For_lev3
For_lev4
Air_temp
Soil_temp
Light
Hum
CWD volume
Litter Depth
Canopy Cover
Forest Level 1
Forest Level 2
Forest Level 3
Forest Level 4
Air Temperature
Soil Temperature
Light Intensity
Relative Humidity
Habitat Description
Presence (%) of ground cover such as leaves or small woody debris measured at every 0.5 m.
Absence (%) of ground cover (e.g., exposed soil) measured at every 0.5 m.
Presence of non-woody stems (%) such as grasses, ferns, and Smilax and Vitus sp. measured at every 0.5 m.
Presence of any woody stems (%) such as seedlings and large trees (living or dead) measured at every 0.5 m.; woody stems taller than one meter had to contact transects directly to be counted
Presence of rocky substrate (%) greater than 10 cm in size measured at every 0.5 m.
Presence of any fallen woody debris larger than 10 cm in diameter (must touch the ground somewhere along the length to be counted) measured at every 0.5 m.
Presence of any woody debris (%) composed of two or more stems 30 cm or higher from the ground (e.g., fallen treetops) measured at every 0.5 m.
Code
%_litt
%_bare
%_herb
%_woody
%_rock
%_CWD
%_slash
Habitat Variable
Percent Litter
Percent Bare Ground
Percent Herbaceous
Percent Woody
Percent Rock
Percent CWD
Percent Slash
Table 7.1 Habitat and environmental characteristics measured at each study plot in the William B. Bankhead National
Forest before and after thinning and burning disturbances.
Figure 7.2 Overhead schematic of coverboard sampling arrays. Center of the diagram
represents a drift-fence trapping array. Boards were placed 50 m from the trap array.
Table 7.2 Raw habitat and climate values derived from 18 forest stands in the William B.
Bankhead National forest before and after prescribed burning and thinning (2005-2009).
Habitat Variable
%_ litt
Treatment Year
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
Control
99.0 ± 1.1
99.5 ± 0.5
98.8 ± 1.5
Burn
99.5 ± 0.2
97.5 ± 1.8
97.6 ± 2.9
Light Thin
99.2 ± 0.9
98.2 ± 0.7
98.5 ± 0.5
Heavy Thin
99.5 ± 0.6
97.6 ± 2.0
98.7 ± 1.1
%_bare
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
0.0
0.0
0.0
0.0
1.5 ± 1.3
1.4 ± 2.4
0.0
0.4 ± 0.7
0.5 ± 0.5
0.0
1.1 ± 0.7
0.1 ± 0.2
0.0
6.1 ± 3.4
1.1 ± 0.5
0.0
7.5 ± 2.9
4.1 ± 3.5
%_herb
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
9.3 ± 3.2
11.2 ± 8.4
7.0 ± 3.7
13.5 ± 11.4
14.6 ± 8.7
12.7 ± 12.4
26.8 ± 2.8
23.2 ± 3.8
36.8 ± 9.5
20.7 ± 25.1
20.8 ± 14.6
31.7 ± 17.5
8.4 ± 5.6
15.7 ± 7.1
36.4 ± 12.8
22.5 ± 15.3
23.2 ± 10.4
34.3 ± 13.6
%_wood
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
5.1 ± 1.9
6.8 ± 1.9
9.6 ± 1.9
7.1 ± 3.7
8.6 ± 6.2
12.1 ± 5.0
15.3 ± 7.2
20.7 ± 8.2
33.9 ± 13.1
15.7 ± 14.0
17.5 ± 8.5
23.2 ± 12.9
8.0 ± 2.5
15.4 ± 5.9
25.4 ± 7.6
14.9 ± 7.3
19.4 ± 5.3
30.1 ± 14.4
%_rock
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
0.0
0.0
0.1 ± 0.2
0.3 ± 0.5
0.3 ± 0.5
0.8 ± 0.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
%_cwd
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
2.3 ± 1.7
1.4 ± 0.5
1.2 ± 0.4
0.7 ± 0.7
1.0 ± 0.9
0.9 ± 0.8
1.7 ± 1.1
2.2 ± 1.6
2.5 ± 1.4
1.7 ± 1.1
3.0 ± 0.2
2.1 ± 1.5
1.0 ± 0.3
3.2 ± 1.9
2.2 ± 1.3
2.4 ± 2.4
3.1 ± 1.7
3.3 ± 1.1
%_slash
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
0.0
0.0
0.1 ± 0.2
0.0
0.0
0.0
0.8 ± 0.8
2.9 ± 1.7
1.1 ± 1.3
0.8 ± 1.4
3.5 ± 1.7
1.9 ± 1.1
0.0
5.3 ± 4.0
3.3 ± 2.6
0.1 ± 0.2
4.3 ± 3.4
2.4 ± 2.1
cwd_vol
Pre-Treatment
76.9 ± 58.3 10.2 ± 10.5 152.4 ± 131.5 97.3 ± 122.7
Post-Treatment Year One 36.5 ± 28.6 18.0 ± 16.6 56.9 ± 69.4 109.5 ± 72.0
Post-Treatment Year Two 49.8 ± 43.7 15.6 ± 21.3 82.9 ± 77.0 96.4 ± 103.7
41.9 ± 29.6
137.4 ± 113.2
95.9 ± 79.9
194.6 ± 259.0
58.8 ± 31.2
55.9 ± 33.4
l_depth
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
6.9 ± 0.4
7.4 ± 0.6
7.0 ± 0.5
4.7 ± 0.8
3.7 ± 1.3
4.3 ± 1.4
6.6 ± 0.8
4.8 ± 0.8
5.5 ± 0.9
5.6 ± 0.5
5.1 ± 0.3
4.8 ± 0.4
7.2 ± 0.4
3.7 ± 0.8
3.8 ± 0.4
6.9 ± 0.7
3.2 ± 0.8
3.7 ± 0.5
can_cov
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
88.1 ± 1.8
91.0 ± 1.8
93.6 ± 1.0
88.4 ± 2.5
90.4 ± 3.8
92.0 ± 0.4
83.9 ± 1.0
64.2 ± 5.6
69.8 ± 2.2
80.4 ± 6.6
62.3 ± 2.4
68.7 ± 1.3
82.2 ± 4.8
58.9 ± 2.0
66.9 ± 10.0
83.1 ± 7.1
50.9 ± 6.5
63.4 ± 7.3
for_1
Pre-Treatment
68.1 ± 13.4 76.4 ± 16.8
Post-Treatment Year One 43.1 ± 28.4 29.2 ± 4.1
Post-Treatment Year Two 54.1 ± 25.3 43.1 ± 12.7
80.6 ± 12.0
52.8 ± 19.7
73.6 ± 16.8
62.5 ± 30.0
34.7 ± 9.6
75.0 ± 18.2
83.3 ± 16.7
37.5 ± 11.0
68.1 ± 6.4
75.0 ± 43.3
52.8 ± 10.5
77.8 ± 6.4
for_2
Pre-Treatment
73.6 ± 13.4 84.7 ± 9.6
Post-Treatment Year One 83.3 ± 14.4 87.5 ± 21.7
Post-Treatment Year Two 72.2 ± 12.0 68.1 ± 12.7
68.1 ± 20.6
22.2 ± 6.4
31.9 ± 2.4
73.6 ± 6.4
33.3 ± 7.2
29.2 ± 11.0
72.2 ± 12.7
19.4 ± 23.7
16.7 ± 7.2
79.2 ± 11.0
23.6 ± 15.8
18.1 ± 14.6
for_3
Pre-Treatment
54.2 ± 19.1 72.2 ± 17.3
Post-Treatment Year One 43.1 ± 6.4 84.7 ± 16.8
Post-Treatment Year Two 50.0 ± 22.0 59.7 ± 6.4
40.3 ± 4.8
22.2 ± 8.7
12.5 ± 8.3
73.6 ± 15.8
26.4 ± 4.8
19.4 ± 6.4
76.4 ± 33.7
22.2 ± 21.0
19.4 ± 19.7
69.4 ± 35.4
20.8 ± 15.8
19.4 ± 12.7
for_4
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
100 ± 0.0
94.4 ± 9.6
100 ± 0.0
89.5 ± 13.0
84.7 ± 9.6
93.1 ± 8.7
93.1 ± 8.7
73.6 ± 20.6
77.8 ± 12.0
100 ± 0.0
66.7 ± 26.0
83.3 ± 7.2
98.6 ± 2.4
66.7 ± 36.1
87.5 ± 11.0
84.7 ± 10.5
75.0 ± 19.1
90.3 ± 2.4
air_temp
Pre-Treatment
25.4 ± 0.2
Post-Treatment Year One 29.5 ± 0.5
Post-Treatment Year Two 28.2 ± 0.12
25.3 ± 0.2
29.3 ± 0.9
28.7 ± 0.9
25.5 ± 0.5
35.3 ± 1.2
34.2 ± 3.7
25.5 ± 0.3
34.8 ± 2.1
34.1 ± 1.5
25.5 ± 0.2
35.3 ± 0.8
33.3 ± 2.2
25.6 ± 0.5
32.6 ± 1.2
34.8 ± 1.7
soil_temp
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
18.9 ± 0.1
21.4 ± 1.9
22.1 ± 3.2
19.2 ± 0.5
23.7 ± 0.6
27.0 ± 5.1
19.2 ± 0.7
24.5 ± 1.1
24.8 ± 1.1
19.0 ± 0.3
27.6 ± 1.6
25.3 ± 1.4
18.8 ± 0.5
25.2 ± 1.5
25.6 ± 2.7
light_int
Pre-Treatment
111.8 ± 29.5 76.2 ± 20.9 73.4 ± 36.7 71.7 ± 40.4
Post-Treatment Year One 106.0 ± 30.0 121.3 ± 68.6 669.7 ± 13.7 594.8 ± 127.2
Post-Treatment Year Two 111.2 ± 25.6 116.0 ± 25.1 528.1 ± 174.3 566.2 ± 49.4
78.8 ± 32.1
792.9 ± 6.4
665.9 ± 13.6
91.7 ± 48.1
770.0 ± 248.7
718.6 ± 90.5
%_hum
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
63.6 ± 2.0
39.6 ± 7.6
55.8 ± 4.0
64.6 ± 1.4
40.7 ± 8.5
48.5 ± 11.2
64.0 ± 3.0
40.8 ± 21.3
37.5 ± 8.7
64.5 ± 1.6
30.8 ± 4.6
34.4 ± 5.4
62.5 ± 1.3
28.9 ± 0.5
37.1 ± 7.2
62.4 ± 2.4
31.1 ± 2.6
35.5 ± 6.9
basal_ar
Pre-Treatment
Post-Treatment
28.9 ± 0.2
30.3 ± 0.2
27.6 ± 3.2
29.5 ± 2.9
27.9 ± 1.4
14.6 ± 1.2
29.3 ± 2.6
11.9 ± 0.9
29.1 ± 3.0
13.8 ± 0.4
26.8 ± 1.3
10.9 ± 0.8
18.8 ± 0.5
19.5 ± 0.6
20.2 ± 0.6
Light Thin*Burn Heavy Thin*Burn
99.4 ± 1.0
99.7 ± 0.5
91.6 ± 5.4
90.8 ± 3.6
98.0 ± 1.1
94.1 ± 4.2
Figure 7.3 Multivariate ordination of habitat and climate measurements to characterize
study plots using PCA. Each plot corresponds to a different year, A) pre-treatment, B)
post-treatment year one, and C) post-treatment year two.
Table 7.3 Total amphibian and reptiles captures in disturbed pine-hardwood forests of the
William B. Bankhead National Forest, Alabama, U. S. A. (2005-2009).
Scientific Name
Common Name
Control Burn
Heavy Thin
Light Thin
Heavy Thin*Burn
Light Thin*Burn
Total
Recaptures
Pre-Treatment
Anolis carolinensis
Carphophis a. amoenus
Diadophis punctatus
Plethodon ventralis
Plethodon mississippi
Scincella lateralis
Green Anole
Eastern Worm Snake
Ringneck Snake
Southern Zig-zag Salamander
Mississippi Slimy Salamander
Little Brown Skink
0
0
0
13
11
0
2
2
1
14
24
1
0
0
4
60
17
1
0
3
4
47
16
1
0
1
2
41
14
1
0
0
0
14
29
0
2
6
11
189
111
4
1
3
1
95
48
0
Post-Treatment Year One
Ambystoma opacum
Carphophis a. amoenus
Diadophis punctatus
Plestiodon fasciatus
Plestiodon lateralis
Plethodon dorsalis
Plethodon mississippi
Pseudotriton r. ruber
Scincella lateralis
Tantilla coronata
Terrapene c. carolina
Virginia v. valeriae
Marbled Salamander
Eastern Worm Snake
Ringneck Snake
Eastern Five-lined Skink
Broad-headed Skink
Southern Zig-zag Salamander
Mississippi Slimy Salamander
Red Salamander
Little Brown Skink
Southeastern Crowned Snake
Eastern Box Turtle
Smooth Earthsnake
0
0
0
0
0
5
6
1
4
0
0
0
0
0
1
0
0
11
13
0
0
0
0
0
1
3
2
0
0
85
5
0
4
0
0
1
0
2
1
0
0
60
12
0
0
1
0
0
0
0
7
0
1
58
7
0
2
0
0
0
0
3
10
2
0
15
6
0
1
0
1
0
1
8
21
2
1
234
49
1
11
1
1
1
1
3
4
1
0
119
20
3
0
2
0
0
Post-Treatment Year Two
Anolis carolinensis
Carphophis a. amoenus
Diadophis punctatus
Plestiodon fasciatus
Plethodon dorsalis
Plethodon mississippi
Scincella lateralis
Storeria dekayi wrightorium
Virginia v. valeriae
Green Anole
Eastern Worm Snake
Ringneck Snake
Eastern Five-lined Skink
Southern Zig-zag Salamander
Mississippi Slimy Salamander
Little Brown Skink
Midland Brown Snake
Smooth Earthsnake
0
0
0
0
10
17
0
0
0
0
0
2
0
16
45
3
0
0
0
6
5
2
88
10
2
1
1
1
7
1
0
69
5
1
0
0
0
1
5
2
47
15
2
1
0
1
6
13
1
12
21
1
0
0
2
20
26
5
242
113
9
2
1
0
17
16
2
168
148
1
0
1
Figure 7.4 Comparison of amphibian and reptile captures using different coverboard sizes
for the most commonly captured species in the William B. Bankhead National Forest,
Alabama, U. S. A. (2005-2009).
Figure 7.5 Response of Southern Zig-zag Salamanders (Plethodon ventralis) to
prescribed burning and thinning in the William B. Bankhead National Forest, Alabama,
U. S. A. (2005-2009).
Figure 7.6 Response of Mississippi Slimy Salamanders (Plethodon mississippi) to
prescribed burning and thinning in the William B. Bankhead National Forest, Alabama,
U. S. A. (2005-2009).
Corrected Count * 100 Survey Days
Figure 7.7 Response of Ring-neck Snakes (Diadophis punctatus) to prescribed burning
and thinning in the William B. Bankhead National Forest, Alabama, U. S. A. (20052009). Spaces with an X indicate study plots where no individuals were captured.
30
25
20
15
10
5
0
X X X
Control
X
Burn
Heavy Thin
Light Thin
Heavy
Thin*Burn
Light
Thin*Burn
Treatment
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
Figure 7.8 Response of fossorial snakes to prescribed burning and thinning in the William
B. Bankhead National Forest, Alabama, U. S. A. (2005-2006). Bottom pie chart indicates
the percentage of total captures each species represented. Please see Table 7-3 for total
species captures Spaces with an X indicate study plots where no individuals were
captured.
Corrected Count * 100 Survey Days
45
40
35
30
25
20
15
10
5
X X X
X
0
Control
Burn
Heavy Thin
Light Thin
Heavy
Thin*Burn
Light
Thin*Burn
Treatment
Pre-Treatment
Post-Treatment Year One
Post-Treatment Year Two
Diadophis punctatus
Carphophis a. amoenus
Storeria dekayi wrightorium
Virginia v. valeriae
Tantilla coronata
Figure 7.9 Average SVL of Ring-necked Snakes (Diadophis punctatus) in forest stands
disturbed by prescribed burning and thinning in the William B. Bankhead National
Forest, Alabama, U. S. A. (2005-2009).
VITA
William B. Sutton, son of James Sutton and Rochelle Sutton was born April 8th
1980 in Philippi, West Virginia. In August 1998, he attended Wheeling Jesuit University,
Wheeling, West Virginia, and received a Bachelors of Science in Biology in May 2002.
In August 2002, he attended Marshall University, Huntington, West Virginia, and
received a Masters of Science in Biological Sciences in August 2004. He entered
graduate school at Alabama A&M University, Normal, Alabama, in January 2005 where
he has been working towards a Ph.D. focusing on forest ecology and disturbance
response of forest herpetofauna.