The University of Southern Mississippi The Aquila Digital Community
Dissertations
Fall 12-2007 THE BLACK PINESNAKE: SPATIAL ECOLOGY, PREY DYNAMICS, PHYLOGENETIC ASSESSMENT AND HABITAT MODELING Danna Leah Baxley University of Southern Mississippi
Follow this and additional works at: https://aquila.usm.edu/dissertations Part of the Animal Sciences Commons, Biodiversity Commons, Biology Commons, and the Terrestrial and Aquatic Ecology Commons
Recommended Citation Baxley, Danna Leah, "THE BLACK PINESNAKE: SPATIAL ECOLOGY, PREY DYNAMICS, PHYLOGENETIC ASSESSMENT AND HABITAT MODELING" (2007). Dissertations. 1298. https://aquila.usm.edu/dissertations/1298
This Dissertation is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Dissertations by an authorized administrator of The Aquila Digital Community. For more information, please contact [email protected]. The University of Southern Mississippi
THE BLACK PINESNAKE: SPATIAL ECOLOGY, PREY DYNAMICS,
PHYLOGENETIC ASSESSMENT AND HABITAT MODELING
by
Danna Leah Baxley
A Dissertation Submitted to the Graduate Studies Office of The University of Southern Mississippi in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
Approved:
December 2007
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. COPYRIGHT BY
DANNA LEAH BAXLEY
2007
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The University of Southern Mississippi
SPATIAL ECOLOGY, PREY DYNAMICS, HABITAT MODELING, RESOURCE
SELECTION, AND PHYLOGENETIC ASSESSMENT OF THE BLACK PINESNAKE
by
Danna Leah Baxley
Abstract of a Dissertation Submitted to the Graduate Studies Office of The University of Southern Mississippi in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
December 2007
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT
SPATIAL ECOLOGY, PREY DYNAMICS, HABITAT MODELING, RESOURCE
SELECTION, AND PHYLOGENETIC ASSESSMENT OF THE BLACK PINESNAKE
by Danna Leah Baxley
December 2007
As human-caused species losses approach unprecedented levels, informed
biological studies at the landscape, community, organismal, and genetic levels become
increasingly important. The six chapters of this dissertation target the black Pinesnake,
Pituophis melanoleucus lodingi, an imperiled specialist of the longleaf pine ecosystem, to
answer biological questions on all four of these levels. At the landscape level, we sought
to identify areas where black Pinesnake populations persist as well as the landscape-level
variables driving presence of this secretive taxon. At the community level, we assessed
habitat associations of reptile and amphibian communities within the longleaf pine
ecosystem, and identified habitat parameters that appear to drive community
composition. On a finer scale, we conducted a radio-telemetry study to examine spatial
ecology of black Pinesnakes, and to test several hypotheses regarding prey dynamics and
resource selection of these snakes. Lastly, we used molecular markers (mitochondrial
DNA, i.e. mtDNA) to assess the intraclade relationships of eastern Pituophis, which
includes P. m. lodingi, P. m. mugitus, and P. m. melanoleucus, in an attempt to add to
existing phylogenetic knowledge of these eastern congeners.
ii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Results of our landscape and community-level studies demonstrate the importance
of large forested tracts of longleaf pine as well as the importance of frequent fire and low
stand densities (low basal areas) to black Pinesnake persistence and restoration. Our
radio-telemetry study indicates that these snakes travel great distances, inhabit large
home ranges, and frequently cross and suffer mortality from roads. Core home ranges of
P. m. lodingi contained significantly more hispid cotton rats than peripheral areas,
indicating a correlation between prey availability and movement for these snakes.
Results of molecular analysis show a lack of sub-species monophyly within eastern
Pituophis, and little geographic structure; however, since there are high levels of
morphological and ecological distinction between these three sub-species, future work
using alternative molecular markers may clarify these within-clade relationships. By
using multiple scales and divergent methods of study, we hope to provide land managers
and biologists with the information necessary to conserve and restore P. m. lodingi
populations.
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGMENTS
Numerous people have been invaluable to the completion of the research involved
in this dissertation. My deepest thoughts of appreciation go to Drs. Carl Qualls, Brian
Kreiser, Jake Schaefer, and Michael Davis, members of my graduate committee, and to
Linda LaClaire of U.S. Fish and Wildlife Service. Each member of my committee has
played a major part in the progress and completion of my research. Dr. Jake Schaefer
provided statistical guidance, conceptual ideas, help with manuscript revisions, and
access to multivariate software. Dr. Micheal Davis assisted in plant identification, and
facilitated thought in plant-animal interactions, which resulted in my second dissertation
chapter. Dr. Brian Kreiser provided manuscript reviews, constructive criticism, and
countless hours of instruction and assistance in the genetics laboratory. I am grateful to
Dr. Carl Qualls, my major professor, for diligently seeking and attaining project funding,
for editing and contributing to countless manuscripts, abstracts, and papers, and for his
availability and openness as a mentor.
This dissertation would not have been possible without funding from the
Mississippi Department of Wildlife, Fisheries, and Parks, U. S. Forest Service, Gopher
Tortoise Council, The Nature Conservancy, The Greater Cincinnati Herpetological
Society, and The University of Sothem Mississippi Biological Sciences Department.
Special thanks goes to Sandra Kilpatrick of the U.S. Forest Service for her interest and
involvement, to Dr. Thomas Ricks who conducted all transmitter implantation surgeries,
and to Mike Sisson, Krista Noel, Josh Ennen, Sam Holcomb, Shea Hammond, Wyatt
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rochelle, and Will Selman for their steadfast assistance in the field and lab. The GIS
habitat modeling portion of this dissertation would not have been possible without the
assistance, patience, and commitment of Greg Lipps. I would also like to express
appreciation to Tom Mann and Dr. Robert Jones of the Mississippi Museum of Natural
Science for providing scientific collection permits as well as their opinions and advice.
Numerous biologists contributed tissue samples to our genetics project and, although I
will not list them all here, I would like to thank them all for their gracious donations of
time and tissue to our project. Lastly, I would like to thank my parents, Jody and Danny
Baxley, my sister Tracy Baxley, and Mike Sisson for their unwavering support and
encouragement throughout my graduate career.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS
ABSTRACT...... ii
ACKNOWLEDGMENTS...... iv
LIST OF TABLES...... viii
LIST OF ILLUSTRATIONS...... x
CHAPTER
I. GENERAL INTRODUCTION...... 1
II. HABITAT ASSOCIATIONS OF REPTILE AND AMPHIBIAN COMMUNITIES IN LONGLEAF PINE HABITATS OF SOUTH MISSISSIPPI...... 7
Abstract Introduction Materials and Methods Results Discussion
III. THE BLACK PINESNAKE {PITUOPHIS MELANOLEUCUS LODINGI): SPATIAL ECOLOGY AND ASSOCIATIONS BETWEEN HABITAT USE AND PREY DYNAMICS...... 29
Abstract Introduction Materials and Methods Results Discussion
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. IV. ASSESSING SECOND AND THIRD ORDER HABITAT SELECTION BY BLACK PINESNAKES {PITUOPHIS MELANOLEUCUS LODINGI): A BOOLEAN MODEL USING GEOGRAPHIC INFORMATION SYSTEMS...... 48
Abstract Introduction Materials and Methods Results Discussion
V. RAPID POPULATION EXPANSION AND RECENT DIVERGENCE OF EASTERN PITUOPHIS: P. M. LODINGI, P. M. MUGITUS, AND P. M. MELANOLEUCUS...... 69
Abstract Introduction Materials and Methods Results Discussion
VI. GENERAL CONCLUSION AND RECCOMENDATIONS...... 82
APPENDIXES...... 85
LITERATURE CITED...... 96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES
Table
2.1 Property type, number of years surveyed, species richness, and habitat type for sites included in habitat analysis ...... 13
2.2 Reptiles and amphibians surveyed using drift fences and active searching and percentage of sites where each species was surveyed from 2004-2006 ...... 15
2.3 Summary of habitat variables quantified at 20 sites surveyed for reptiles and amphibians in south Mississippi...... 19
2.4 PCA loadings of habitat variables from 20 sites surveyed in 2004 ...... 21
2.5 Canonical correspondence analysis (CCA) final scores for 34 reptile and amphibian species...... 23
3.1 Tracking and home range data for five Black Pinesnakes radio-tracked during 2005-2006...... 38
3.2 Means, standard deviations, test statistics, and p-values for habitat variables collected from mammal transects within core versus outer home ranges of Black Pinesnakes ...... 43
4.1 Mean percent composition, standard deviation, test statistic, and significance level for each variable tested for compositional differences between real and simulated home ranges ...... 59
4.2 Mean, test statistic, and p-values of Wilcoxon Sign-Rank tests comparing minimum distances of radio-tracked Black Pinesnakes and minimum distances of random simulated points to environmental variables ...... 60
4.3 Input parameters for Boolean model ...... 61
4.4 Summary of habitat variables quantified (mean + standard deviation) at 131 sites of recent (post-1987) and historic P. m. lodingi occurrence ...... 64
5.1 Mitochondrial NADH sequences of Pituophis and Lampropeltis obtained from GenBank ...... 74
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5.2 Number and location of Pituophis samples used in this study ...... 75
5.3 Haplotype diversity (h), nucleotide diversity, range of pair-wise distance values (d), number of samples (s), and number of unique haplotypes (u) for Pituophis using ND4 and combined ND2 and ND4 ...... 78
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF ILLUSTRATIONS
Figure
2.1 Sites surveyed for reptiles and amphibians 2004-2006 ...... 10
2.2 Linear regression between reptile diversity and canopy cover for all sites surveyed 2004-2006...... 18
2.3 Principal components analysis (PCA) of habitat variables for all sites surveyed in 2004...... 22
3.1 Locations of radio-tracked Black Pinesnakes in South Mississippi ...... 33
3.2 Radio-telemetric locations of eight Black Pinesnakes tracked 2005-2006 ...... 39
3.3 Mean monthly movement distances of Black Pinesnakes in South Mississippi ...40
3.4 Species and number of small mammal individuals captured in core versus outer home ranges of Black Pinesnakes ...... 42
4.1 Boolean model representing areas of optimal suitability (Boolean score = 6), for Black Pinesnakes ...... 62
4.2 Percent occurrence records plotted against percent reduction in land area for Boolean model ...... 63
5.1 Distribution of eastern Pituophis, adapted from Conant and Collins, 1991 ...... 72
5.2 Strict consensus of 15 most parsimonious trees from ND4 data set. Maximum Parsimony, Maximum Likelihood, and Bayesian support values are indicated along branches ...... 77
5.3 Mismatch distribution of pair-wise sequence differences using NADH dehydrogenase subunit 4 (ND4) ...... 79
x
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1
CHAPTERI
GENERAL INTRODUCTION
The federal Endangered Species Act (ESA), established in 1973, is the current
legislative effort to safeguard and protect the biodiversity within the United States of
America. Section 2 of this Act states, “Congress finds and declares that various species
of fish, wildlife, and plants in the United States have been rendered extinct as a
consequence of economic growth and development untempered by adequate concern and
conservation.” As a result, “the United States has pledged itself as a sovereign state in
the international community to conserve to the extent practicable the various species of
fish or wildlife and plants facing extinction.” Despite efforts of the ESA, trends of
accelerated loss and declines of species have been recognized for nearly all taxonomic
groups, including reptiles and amphibians both within the United States and globally
(Dirzo and Raven, 2003). Assessing population trends of reptiles and amphibians proves
problematic for several reasons. Long-term population monitoring for herpetofauna is
resource intensive, requires significant amounts of time, and is rarely implemented
(Gibbons et al., 2000). Furthermore, it is difficult to separate natural population
fluctuations from population declines, especially for amphibians (Pechmann et ah, 1991;
Blaustein et ah, 1994; Green, 2003). As a result of these factors, reptiles and amphibians
often invoke conservation concern and attain candidate or protected ESA status only after
populations suffer reductions so severe that recovery is difficult or improbable. For
example, the Mississippi Gopher Frog (Rana sevosa) was reduced to one breeding
population before attaining ESA protection (Richter and Seigel, 2002) and several reptile
species, such as the Eastern Indigo Snake (Drymarchon corais couperi) and the Southern
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2
Hog-nosed Snake (Heterodon simus), quietly disappeared from Mississippi before
biologists noticed or acted upon declines (Tuberville et al., 2000).
Small, isolated populations, especially those that require unique or sensitive
habitats, are particularly susceptible to population declines and extinction. The longleaf
pine ecosystem, a landscape that once almost entirely characterized the southeastern
United States and has been severely diminished, is home to approximately 72 reptile and
amphibian species. Of these, about one third are classified as specialists and are found
only within the historic range of longleaf pine forests (Guyer and Bailey, 1993).
Countless factors have contributed to the decline and degradation of the longleaf pine
ecosystem. The main causes for habitat loss include fire suppression, removal of pine
stumps for the naval stores industry, invasion of non-native species, and the conversion
of longleaf pine forests to urban, silvicultural, or agricultural areas (Frost, 1993; Outcalt,
1997; Varner and Kush, 2004).
Even in conserved areas, such as DeSoto National Forest, Mississippi, longleaf
pine forests suffer ongoing quality reduction because prescribed bums predominately
occur during the winter to ameliorate issues of liability and smoke control. Historically,
lightning-caused fires burned longleaf pine forests during the growing season. These
types of fires bum at higher temperatures and are more effective at eradicating competing
midstory and shrub vegetation (Glitzenstein et al., 1995). Winter bums have also been
implicated in amphibian declines, though research on this topic has been highly
controversial (Schurbon and Fauth, 2003; Means et al., 2004).
In addition to creating logistical problems with prescribed burning, habitat
fragmentation is emerging as a serious threat to remaining tracts of longleaf pine forest.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3
The size, growth, and resource demands of the human population are the most important
agents of change in spatial patterns of biodiversity (Vitousek et al., 1997; Sala et al.,
2000). Although global estimates of human population predict declines in U.S.
population by 2050 (Cohen, 2003), U.S. trends of migration to warmer and more
temperate areas of the country will probably continue. As human population growth and
resulting development in the southeast increases, many isolated tracts of longleaf pine
forest will become functionally extinct because prescribed fire will no longer be feasible.
Without natural or prescribed fire, the longleaf ecosystem quickly loses the successional
battle and hardwood species become dominant. In the absence of nearby longleaf
habitats containing metapopulations, to which amphibians and reptiles are capable of
dispersing, the reptile and amphibian populations within these fragments may become
extinct (McCullough, 1996; Marsh and Trenham, 2001).
The efforts of the ESA to manage habitat use on both public and private lands and
to mitigate effects of human development and urbanization upon threatened and
endangered taxa have been widely criticized (Tear et al., 1993; Schemske et al., 1994;
Foin et al., 1998). This act has been criticized for exhibiting a taxonomic bias in favor of
animals over plants, vertebrates over invertebrates, and birds and mammals over fish and
herpetofauna (Tear et ah, 1995), although the main criticism of the ESA is the lack of
encouragement within this statute for proactive actions that may preclude the need to list
species as threatened or endangered (Noss et ah, 1997). In an attempt to amend some of
the problems associated with this reactive approach to conservation, over 275 U.S.
species are listed as “candidates” for federal ESA listing; this means not enough
information is currently available to make informed decisions about listing these species,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. but population declines are suspected. Among these federal candidate species is the
black Pinesnake ( Pituophis melanoleucus lodingi), a longleaf pine specialist that is also
state-listed in Mississippi with the status of “endangered.”
Until recently, very little was known concerning the natural history, spatial
ecology and current distribution of the Black Pinesnake; much of the available literature
concerns captive specimens or laboratory trials (Clibum, 1957; Clibum, 1962; Clibum,
1976). Although the historic range of P. m. lodingi spanned from western Florida to
extreme eastern Louisiana (Crain and Clibum, 1971; Conant and Collins, 1991), this sub
species is thought to be extirpated from Louisiana, and has not been reported west of the
Pearl River for nearly forty years. Approximately 61% of historical Black Pinesnake
populations are thought to be extirpated or are in decline (Duran, 1998b). Aside from
one unpublished radio-telemetry study conducted within optimal habitat (Duran, 1998a)
and comparison of prey handling techniques between P. m. lodingi and P. ruthveni
(Rudolph et al., 2002), there is a paucity of P. m. lodingi research by which to inform
candidate ESA status.
In addition to the general lack of ecological data for Black Pinesnakes, there is
also very little available data concerning the genetic relationships among eastern
Pituophis. Currently one of fifteen recognized sub-species of Pituophis, this snake is
geographically isolated from all other members of the genus; consequently, some
researchers support elevation of this sub-species to full species status (Duran, 1998b).
Debates regarding the relationships among eastern and western Pituophis are
characterized by several taxonomic hypotheses (see Rodriguez-Robles and Jesus-
Escobar, 2000 for a complete review). Most recently, Burbrink and Lawson (2007)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5
suggested inclusion into the Pituophis genus of all members of the clade formed by the
most recent common ancestor of Pantherophis guttatus. Another recent systematic
analysis of new world gopher, bull, and Pinesnakes indicates that the eastern Pinesnake
clade (P. m. lodingi , P. m. mugitus, and P. m. melanoleucus) and the western Pinesnake
clade (P. c. affinis, P. c. annectens, P. c. bimaris, P. c. catenifer , P. c. deserticola, P. c.
sayi, and P. c. vertebralis) represent divergent lineages and should be recognized as two
distinct species (Rodriguez-Robles and Jesus-Escobar, 2000). Although Rodriguez-
Robles and Jesus-Escobar (2000) found monophyly of the eastern Pinesnake clade to be
well supported, within-clade relationships for this group were characterized by low
bootstrap values, and were relatively unresolved. Aside from the Rodriguez-Robles and
Jesus-Escobar study, there is little information concerning molecular systematics of the
eastern Pinesnake clade as a whole, and the Black Pinesnake in particular. For instance,
total molecular systematic work on the Black Pinesnake consists of mitochondrial DNA
analysis from two black Pinesnakes obtained via the pet trade, where snakes are often
hybridized or back-crossed resulting in uncertain lineages.
Informed biological studies at the landscape, community, organismal, and genetic
levels become increasingly important as P. m. lodingi populations continue to decline.
The objective of this dissertation was to use a variety of research scales to expand the
base of available knowledge regarding P. m. lodingi to provide land managers and
biologists with decision-making tools. At the landscape level, we sought to identify areas
where black Pinesnake populations persist as well as the landscape-level variables
driving presence of this secretive taxon. At the community level, we assessed habitat
associations of reptile and amphibian communities within the longleaf pine ecosystem,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6
and used multivariate analysis to identify habitat parameters that appear to drive
community composition. On a finer scale, we conducted a radio-telemetry study to
examine spatial ecology of Black Pinesnakes, and to test several hypotheses regarding
prey dynamics and spatial ecology of these snakes (e.g. quantify 3rd order resource
selection within home ranges). Lastly, we used molecular markers (mtDNA) to assess
the intraclade relationships of the eastern Pituophis clade, which includes P. m. lodingi,
P. m. mugitus, and P. m. melanoleucus, in an attempt to add to existing phylogenetic
knowledge of these eastern congeners. By using multiple scales and divergent methods of
study, we hope to provide land managers and biologists with the information necessary to
conserve and restore P. m. lodingi populations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7
CHAPTER II
HABITAT ASSOCIATIONS OF REPTILE AND AMPHIBIAN COMMUNITIES IN
LONGLEAF PINE HABITATS OF SOUTH MISSISSIPPI
Abstract.—Habitat associations of reptiles and amphibians native to longleaf pine forests
in southern Mississippi are scantly studied and poorly understood. From 2004 to 2006,
we surveyed the herpetofauna of twenty-four longleaf pine communities in twelve
counties in south Mississippi. For each site, we quantified herpetofaunal diversity and
relative abundance and a suite of habitat variables to address the following objectives: (1)
determine what levels of habitat heterogeneity exist in longleaf pine forests in south
Mississippi; (2) determine if there are differences in reptile and amphibian community
composition among these sites; and (3) if habitat-faunal differences exist among sites,
identify what habitat variables are driving these community differences. Multivariate
analysis identified three distinct longleaf pine habitat types, differing primarily in soil
composition and percent canopy cover of trees. Canonical correspondence analysis
indicated that canopy cover, basal area, percent grass in the understory, and soil
composition (percent sand, silt, and clay) were the predominant variables explaining
community composition at these sites. Many species exhibited associations with some or
all of these habitat variables. The significant influence of these habitat variables,
especially basal area and canopy cover, upon herpetofaunal communities in south
Mississippi indicates the importance of incorporating decreased stand density into
management practices for longleaf pine habitat.
Key Words.—amphibian; canopy cover; community; habitat associations; herpetofauna;
longleaf pine, Mississippi; reptile;
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8
Introduction
The longleaf pine (Pinus palustris) ecosystem, a landscape that once almost entirely
characterized the southeastern coastal plain of the United States, has been severely
diminished. The reduction of longleaf pine forest to less than five percent of its historic
range is largely due to fire suppression and the conversion of longleaf pine forests to
agricultural, silvicultural, or human use areas (Frost 1993; Outcalt 1997; Varner and
Kush 2004). Due to the elusive, fossorial, and often cryptic nature of many reptile and
amphibian species, the community assemblages and habitat associations of herpetofauna
native to longleaf pine forests in Mississippi are poorly known. Guyer and Bailey (1993)
described the unusually diverse herpetofauna of longleaf forests and estimated that
approximately 72 reptile and amphibian species are associated with this habitat; while
one third of these species are found only within the historic longleaf pine range and are
classified as specialists. Although studies describing reptile and amphibian communities
in upland longleaf pine habitats are relatively common (Tuberville et al., 2005; Smith et
al., 2006), most studies addressing habitat-faunal relationships within the longleaf pine
ecosystem have targeted Florida and South Carolina (Dodd and Franz, 1995; Yale et al.,
1999; Litt et al., 2001).
Southern Mississippi contains a relatively under-studied sector of longleaf pine
forest and, to our knowledge, there are no published studies concerning the habitat
associations of herpetofaunal communities within this area. In an attempt to better define
the faunal-habitat relationships of these communities, we sought to address the following
three objectives: (1) determine what levels of habitat heterogeneity exist in longleaf pine
forests in south Mississippi; (2) determine if there are differences in reptile and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 9
amphibian community composition among these sites; and (3) if habitat-faunal
differences exist among sites, identify what habitat variables are driving these
differences. As proposed by Dodd and Franz (1993), we also sought to add to the base of
information concerning entire herpetofaunal communities, including common species,
such that monitoring protocols may ultimately be developed and implemented in South
Mississippi. By better defining the faunal-habitat relationships of these communities, the
natural history and ecology of many longleaf pine resident and specialist species may be
better understood.
M a t e r ia l s a n d M e t h o d s
Study Sites.—The search for suitable longleaf pine habitat was primarily restricted
to the southern, coastal plain region of Mississippi. Habitat targeted in this study was
characterized by sandy soils, an overstory of longleaf pine trees, a fire-maintained,
herbaceous, grassy understory, and a fire-suppressed mid-story. In total, twenty-four
sites with suitable habitat were identified and surveyed in twelve counties (Figure 2.1).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 10
Sites 1 i
17,18 v,- -
Figure 2.1 Sites surveyed for reptiles and amphibians 2004-2006. Shaded areas represent public land (National Forest or Wildlife Management Area).
Of these sites, twenty-two were on public land and two were on private land. Seventeen
of the public sites were within the DeSoto National Forest (six within the Chickasawhay
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11
district and eleven within the DeSoto District), we surveyed one site on the Mississippi
Sandhill Crane National Wildlife Refuge, three sites within the Marion County Wildlife
Management Area, and one site owned by the Hancock County School Board.
Additionally, several private landowners in both Hancock County and George County
allowed their lands to be surveyed.
Survey techniques.—To address habitat variation as well as habitat/community
associations of reptiles and amphibians in Mississippi, our project was divided into two
parts. In 2004, all identified sites were surveyed using standard herpetological sampling
techniques using a combination of drift fences with pitfall and funnel traps (Gibbons and
Semlitsch, 1981; Campbell and Christman, 1982; Fitch, 1987), manual searching, cover
boards, and road surveys. Due to time and budget constraints, a subset of the sites
surveyed in 2004 were selected for further surveys in 2005 and 2006, and three sites were
surveyed only in 2005 (see Table 2.1 for number of survey years for each site). In 2004
each drift fence (100 ft long by 3 ft high) was equipped with four funnel traps, one at
each end on each side, and four pitfall traps, one near each end on each side. To sample
multiple sites within a local region simultaneously, twelve or more drift fence/trap arrays
were deployed at all times. As a result of low capture rates for species of interest with
cover boards and pit-fall traps in 2004, we did not employ these techniques in 2005 or
2006. The number of drift fences erected at each site was determined by the area of
suitable habitat present at each site. Each year, sites were surveyed for at least 14
consecutive days in the spring (March 15- May 27) and 14 consecutive days in the
summer (May 28-July 29).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 12
All arrays were checked at least once per day, and were checked twice per day
during hot conditions. All captured animals were removed from traps and released at the
site of capture each time traps were checked. For each animal captured, the location,
date, and species were recorded. During the first week of July 2004, 2005, and 2006,
habitat data was collected at each site where sampling occurred. The following
characteristics of each site were quantified and recorded: soil composition, dominant
vegetation species, percent canopy cover of trees, percent mid-story coverage, percent
shrub coverage, basal area of trees, percentage cover of grasses, forbs, and shrubs in
understory, ground-cover percentage of litter and bare soil, and recency of fire. A
spherical densiometer (Forest Densiometers model-C) was used to measure canopy
cover. At each drift fence array, four densiometer readings were taken to estimate
percent canopy cover facing north, south, east, and west of the array; these readings were
averaged to estimate overall cover. A Cruz-All basiometer (model 59793) was used to
measure factor-10 basal area. Measurements for both canopy cover and basal area were
taken from the middle of each 100’ drift fence array. We used a one-meter quadrat (four
replicates per drift fence) to estimate average percentage of grasses, forbs, shrubs, litter
and bare soil comprising the groundcover, and we visually estimated the percentage of
mid-story coverage and percentage of shrubs in the mid-story. To increase measurement
consistency, all habitat parameters were measured by one individual (D. B.). Lastly, we
used the soil hydrometer method (Sheldrick and Wang, 1993) to determine percentage
content of sand, silt, and clay for all soil samples.
Data Analysis .— To assess degree of habitat heterogeneity among survey sites,
we used principal components analysis (PCA) coupled with analysis of similarity
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 13
(ANOSIM) using the Bray Curtis similarity index. For this analysis, we included all sites
surveyed in 2004 (twenty sites). To explicitly assess habitat associations of reptiles and
amphibians, we employed canonical correspondence analysis (CCA) using all sites (and
all years) surveyed.
Table 2.1. Site numbers corresponding to map (Figure 2.1), property type, number of years surveyed, species richness, and habitat type for sites included in the PCA habitat analysis (O = open canopied group, M = moderate canopy closure group, and C = closed canopy group).
Site # County Property Type Years Surveyed So. Rich Hab
1 Jones DeSoto National Forest 3 20 C 2 Wayne DeSoto National Forest 2 16 - 3 Wayne DeSoto National Forest 3 20 C 4 Wayne DeSoto National Forest 3 24 C 5 Wayne DeSoto National Forest 3 29 c 6 Wayne DeSoto National Forest 1 8 o 7 Greene DeSoto National Forest 1 17 M
8 Marion Marion Co. Wildlife Management Area 1 10 -
9 Marion Marion Co. Wildlife Management Area 1 21 -
10 Marion Marion Co. Wildlife Management Area 1 15 - 11 Forrest DeSoto National Forest 1 11 c 12 Forrest DeSoto National Forest 1 14 M 13 Forrest DeSoto National Forest 1 15 O 14 Perry DeSoto National Forest 32 M 15 Forrest DeSoto National Forest 1 7 C 16 Pearl River DeSoto National Forest 1 10 C 17 Perry DeSoto National Forest 1 13 M 18 Perry DeSoto National Forest 1 15 O 19 George Private 1 11 M 20 Harrison DeSoto National Forest 1 17 M 21 Harrison DeSoto National Forest 1 8 M 22 Hancock Private 1 19 O 23 Hancock Private 1 16 M 24 Jackson Sandhill Crane National Wildlife Refuge 1 21 O
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Since habitat parameters for each site did not remain constant across years (substantial
changes in habitat resulted from prescribed fire and Hurricane Katrina), we considered
each survey year for a given site a different replicate. All community data were
standardized to adjust for small disparities in number of trap days for each site (resulting
from differences in suitable habitat and number of drift fence arrays per site). We did not
include arboreal snakes (Opheodrys aestivus and Elaphe obsoleta), frogs o f the genus
Hyla, or extremely rare reptiles and amphibians (total number of captures less than five
individuals) in our community analyses. We excluded arboreal snakes and tree frogs
from our analysis because these groups are not effectively sampled using drift fences
with funnel and pitfall traps (Dodd, 1991), and we sought to minimize sampling bias to
the greatest extent possible. Scaling options for CCA were as follows: axis scores
centered and standardized to unit variance, axes scaled to optimize representation of sites,
and we chose to use Monte Carlo permutation tests (Manly, 1991) with 9,999 iterations to
test two null hypotheses: that there is no relationship between the species and habitat
matrices, and that there are no species-environment correlations. To test the null
hypothesis that slope of reptile diversity plotted against percent canopy cover is equal to
zero, we conducted a one-way ANOVA using JMP (Version 6. SAS Institute., Cary, NC,
1989-2005); alpha was 0.05 for all tests. All multivariate analyses were conducted using
PRIMER 5.2.9 (Primer-E Ltd, Roborough, Plymouth, United Kingdom) and PcOrd 5
(MJM Software Design, Gleneden Beach, Oregon, United States).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 15
R e s u l t s
Over three field seasons, 3,464 trap days (one drift fence array open for one 24-
hour period = one trap day) resulted in 2,192 captures (Table 2.2), representing 54
species (20 amphibian and 34 reptile).
Table 2.2. Reptiles and amphibians surveyed using drift fences and active searching and percentage of sites where each species was surveyed from 2004-2006. Abbreviations in parentheses are used in Table 2.5.
Scientific Name Common Name # Surveyed % o f Sites
Turtles Gopherus polyphemus (Gophpoly) Gopher Tortoise 68 66.7 Terrapene Carolina (Terracar) Eastern Box Turtle 13 33.3
Lizards Anolis carolinensis (Anolcar) Green Anole 37 66.7 Aspidocelis sexlineatus (Aspsex) Six-lined Racerunner 31 37.5 Eumeces fasciatus (Eumefasc) Common Five-lined Skink 11 33.3 Eumeces inexpectatus (Eumeinex) Southeastern Five-lined Skink 53 66.7 Eumeces laticeps Broad-headed Skink 2 8.3 Ophisaurus attenuatus Slender Glass Lizard 1 4.2 Ophisaurus ventralis Eastern Glass Lizard 2 8.3 Sceloporus undulatus (Scelopun) Eastern Fence Lizard 160 88 Scincella lateralis (Scinclat) Ground Skink 362 92
Snakes Agkistrodon contortrix (Agkiscon) Copperhead 7 29.2 Agkistrodon piscivorus (Agkistpi) Cottonmouth 5 20.8 Cemophora coccinea (Cemcocc) Scarlet Snake 63 58.3 Coluber c. priapus (Coicon) Black Racer 158 87.5 Crotalus adamanteus (Crotadam) Eastern Diamondback Rattlesnake 6 25.0 Diadophis punctatus Ring-necked Snake 3 12.5 Elaphe guttata (Elaphgut) Com Snake 9 25.0 Elaphe obsoleta Gray Rat Snake 10 16.6 Heterodon platirhinos (Hetplati) Eastern Hog-nosed Snake 19 37.5 Lampropeltis getula (Lampgetu) Eastern King Snake 18 33.3 Lampropeltis triangulum (Lamptria) Scarlet King Snake 24 33.3 Masticophis flagellum (Mastflag) Coachwhip 52 45.8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 16
Table 2.2. (cont.) Reptiles and amphibians surveyed using drift fences and active searching and percentage of sites where each species was surveyed from 2004-2006.
Scientific Name Common Name # Surveyed % of ______Sites
Snakes Micrurus fulvius Coral Snake 3 8.3 Nerodia erythrogasterflavigaster Yellow-bellied Water Snake 2 8.3 Opheodrys aestivus Rough Green Snake 3 12.5 Pituophis melanoleucus (Pitmel) Pine Snake 10 20.8 Regina rigida Crayfish Snake 3 12.5 Sistrurus miliarius (Sistrmil) Pygmy Rattlesnake 8 29.2 Storeria dekayi (Stordeka) Dekay’s Snake 8 16.7 Storeria occipitomaculata Red-bellied Snake 1 4.2 Tantilla coronata (Tantcoro) Southeastern Crowned Snake 13 54.2 Thamnophis sauritus (Thamsaur) Eastern Ribbon Snake 16 33.3 Thamnophis sirtalis (Thamsirt) Common Garter Snake 19 79.2
Salamanders Eurycea guttolineata (Eurycgut) Three-Lined Salamander 8 25.0 Eurycea quadridigitata Dwarf Salamander 2 8.3 Notopthalmus viridescens Eastern Newt 1 4.2 Plethodon Mississippi (Plethmis) Mississippi Slimy Salamander 22 37.5
Frogs Acris gryllus (Accgryl) Southern Cricket Frog 22 37.5 Rufo terrestris (Bufter) Southern Toad 244 87.5 Bufo quercicus (Bufquerc) Oak Toad 16 4.2 Gastrophryne carolinensis (Gascar) Eastern Narrow-mouthed Toad 557 91.7 Hyla chrysoscelis/versicolor Gray Treefrog 2 8.3 Hyla cinerea Green Treefrog 2 8.3 Hyla femoralis Pinewoods Treefrog 20 29.2 Hyla gratiosa Barking Treefrog 8 12.5 Hyla squirella Squirrel Treefrog 2 8.3 Pseudacris crucifer Spring Peeper 2 8.3 Rana catesbeiana (Rancat) Bull Frog 6 20.8 Rana c. clamitans (Ranclam) Bronze Frog 51 41.7 Rana grylio Pig Frog 1 4.2 Rana palustris Pickerel Frog 1 4.2 Rana sphenocephala (Ransphen) Southern Leopard Frog 7 29.2 Scaphiopus holbrooki (Scaphhol) Eastern Spade foot Toad 154 25.0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Average trap mortality was 3.3% and red imported fire ants ( Solenopsis invicta ) were
present in 71% of these mortality events. There were significant yearly differences in
reptile and amphibian communities sampled (ANOSEM Global R = 0.316, P = 0.008);
lowest numbers of both species and individuals were sampled in the severe drought year
of 2006, though yearly differences were not significant (relative abundance x2 = 4.82, df
= 2, P = 0.089; diversity (S) x2 = 4.07, df = 2, P = 0.13). It is of note that we surveyed
few true longleaf pine specialists; furthermore, we failed to detect Heterodon simus,
Drymarchon covais couperi, Rana sevosa or Pseudacris ornata. A one-way ANOVA
test (F-ratio = 4.92, df = 1,33, P = 0.03) indicated a significant negative linear
relationship between percent canopy cover and reptile diversity (Figure 2.2). Relative
abundances of reptiles and amphibians surveyed are listed in Table 2.2. Specific locality
data and further survey data for sites surveyed in 2004 and 2005 are catalogued with the
Mississippi Museum of Natural Science (Smith and Qualls, 2004; Baxley and Qualls,
2005).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 18
—
Q_
O r~ i Q Average % Canopy Cover
Figure 2.2 Linear regression indicating a significant negative linear relationship between reptile diversity and canopy cover for all sites surveyed 2004-2006.
For our habitat analysis of twenty sites in eleven counties, the first three PCA
axes explained 71.5% of the habitat variation between sites. Percent canopy coverage
and soil composition (percent sand silt and clay in the soil) were the factors with highest
loadings on the first two axes, explaining most of the variation in the model, while basal
area, and percentage of shrub growth and bare soil in the understory explained most of
the variation on the third axis (Table 2.2). Three distinct habitat-group clusters resulted
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 19
from the PCA (Figure 2.3). We identified three habitat-group clusters from our PCA, and
subsequently defined each group as separate factors within ANOSIM. Monte Carlo tests
indicated that these three groups represent significantly different site types (Global R =
0.452, P = 0.001). Means and standard deviations for habitat variables for these three
groups are displayed in Table 2.4.
Table 2.3. Summary of habitat variables quantified (mean ± standard deviation) at 20 sites surveyed for reptiles and amphibians in south Mississippi. Groups represent different site types identified by PCA and ANOVA (see Fig. 2.3).
Habitat variable Group 1 Group 2 Group 3 Closed Canopy Moderate Canopy Open Canopy
% Sand 61.7+11.9 83.3 ±12.6 80.2 ± 11.6
% Clay 6.2+ 2.2 3.4 ±2.2 3.9 ±2.4
% Silt 32.1 ± 10.9 13.2 ±10.5 15.9 ±9.8
% Forb 3.9 ±3.9 1.8 ±1.1 3.8 ±4.3
% Shrub 22.3 + 14.0 33.1 ± 15.3 25.8 ±25.8
% Litter 66.7 ±26.7 78.2 ±16.3 57.0 ± 15.5
% Bare ground 16.4 ± 17.1 12.1 ± 12.9 17.9 ± 14.9
Basal area m2 110 ±27.7 69.0 ±9.9 24.0 ± 8.9
% Canopy Cover 67.1 ± 12.1 45.0 ±5.9 23.6 ±8.8
Bum recency (months) 17.6 ± 18.0 28.0 ±8.8 28.8 ± 15.5
% Grass 15.9 ±11.0 11.7 ± 12.8 21.7 ± 15.6
% Midstory coverage 17.0^30.0 18.0 ± 12.2 28.2 ±25.8
% Shrubs in midstory 48.2 ±22.1 68.0 ± 16.2 48.9 ±27.1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 20
The first three axes of the CCA explained 22.9% of the variance within the site
and species matrices. Soil composition (percent sand and silt in the soil) accounted for
most of the variation explained on the first axis, percent grass in the understory and
percent sand accounted for most of the species-habitat variation explained by axis two,
and canopy cover and basal area explained most of the variation on the third axis. Monte
Carlo permutation tests revealed significant structure (i.e. P = proportion of randomized
runs with eigenvalues greater than or equal to the observed eigenvalue) in our habitat and
community data for the second and third axes, but not for axis one (axis one P = 0.12,
axis two P = 0.02, axis three P = 0.0006). Although species-environment correlations
were statistically significant only for axis three (axis one P = 0.22, axis two P = 0.49, and
axis three P = 0.002), these correlations were extremely high for both observed (real data
mean correlation = 0.90) and simulated data (Monte Carlo mean correlation = 0.85).
When sites are plotted in ordination space according to habitat parameters, the majority
of sites cluster near the middle of the habitat ordination, although there are several sites
that are apparent outliers, characterized by habitat dissimilar to most of the other sites.
When species are plotted in ordination space according to habitat parameters, several
species are correlated with one or more habitat parameters (See Table 2.4 for loadings).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 21
Table 2.4. PCA loadings of habitat variables from 20 sites surveyed in 2004. The first three axes of the PCA explain 71.5% of habitat variance. The major axis loading for each habitat variable is displayed in bold.
Variable PCI PC2 PC3
% midstory -0.25 -0.20 -0.12
% shrub coverage -0.14 0.15 0.46
% sand -0.41 0.25 -0.09
% clay 0.40 -0.12 0.07
% silt 0.40 -0.27 0.08
Ave. forb 0.28 0.16 0.34
Ave. Litter -0.30 -0.33 0.21
Ave Bare Soil 0.12 0.26 0.58
Ave Shrub -0.10 0.02 0.59
Basal Area 0.12 0.22 -0.49
% Canopy Cover 0.02 -0.54 -0.07
Bum recency -0.34 -0.08 0.01
Ave Grass 0.36 0.17 0.09
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 22
I
% Canopy Cover
Basal Area
cn O CL
-2 --
-3 -■
-4 -■
3 2 0 1 2 4 f> 4 Percent Silt + Percent Sand Percent Clay
PC1
Figure 2.3. Principal components analysis (PCA) of habitat variables for all sites surveyed in 2004. Triangles represent closed canopied sites, inverse triangles represent sites with moderate canopy coverage, and squares represent sites with relatively open canopies. Numbers correspond to site numbers used in both Figure 2.1 and Table 2.1.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2.5. Canonical correspondence analysis (CCA) final scores for 34 reptiles and amphibian species. See Ter Braak (1986) for detailed explanation of CCA output. Species abbreviations are given in Table 2.2. The major axis loading for each species displayed in bold.
Species Axis 1 Axis 2 Axis 3
Accgryl 0.14 0.62 -0.94 Agkiscon 2.47 -0.10 0.29 Agkistpi -1.36 2.23 1.93 Anolcar 0.83 0.45 -0.52 Aspsex -2.63 -1.08 0.23 Bufter -0.39 0.04 -0.12 Bufquerc -5.23 5.35 6.80 Cemcocc 0.50 -0.31 0.75 Coicon -0.03 0.56 -0.01 Crotadam 1.32 0.53 0.84 Elaphgut -0.92 0.88 0.98 Eumefasc -0.98 1.99 0.28 Eumeinex 0.08 -0.10 0.97 Eurycgut 0.12 -1.00 -1.96 Gascar -0.60 0.29 -0.28 Gophpoly 0.18 -2.33 0.42 Hetplati 0.61 -0.26 -1.17 Lampgetu -1.00 -1.82 -0.24 Lamptria 1.88 0.01 0.13 Mastflag -0.27 -2.18 1.04 Pitmel 0.95 -1.54 0.63 Plethmis -0.92 1.03 -2.98 Rancat -0.77 1.58 -0.92 Ranclam -0.28 0.53 -1.90 Ransphen -1.12 0.21 0.30 Scaphhol -2.60 -1.78 -1.27 Scelopun -0.10 0.32 0.47 Scinclat 0.96 0.40 -0.50 Sistrmil -0.68 -0.02 0.44 Stordeka 1.34 -0.53 3.98 Tantcoro 0.48 1.44 -0.25 Terracar 1.69 -0.16 2.16 Thamsirt 1.35 -0.06 0.65 Thamsaur 2.11 1.05 0.01
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 24
For example, the Oak Toad ( Bufo quercicus) is positively correlated with high
percent grass in the understory and extremely low percent canopy cover. Several species,
such as the Six-lined Racerunner (Aspidoscelis sexlineatus), Spadefoot Toad ( Scaphiopus
holbrooki), and Eastern Coachwhip (Masticophis flagellum) are positively associated
with sandy soils and open canopies. Conversely, several species seem to be associated
with sites characterized by more closed canopies and less sand in the soil: Eastern Ribbon
Snake ( Thamnophis sauritus), Copperhead (Agkistrodon contortrix), Scarlet Kingsnake
(Lampropeltis triangulum), Garter Snake {Thamnophis sirtalis ), and Dekay’s Snake
{Storeria dekayi). As expected, many common/generalist species plot near the center of
the ordination, and are neither positively or negatively correlated with specific habitat
parameters: Southern Toad {Bufo ter res tr is), Fence Lizard {Sceloporus undulatus),
Narrowmouth Toad {Gastrophryne carolinensis), Black Racer {Coluber constrictor
priapus), Southern Cricket Frog {Acris gryllus), and Bronze Frog {Rana clamitans
clamitans).
D is c u s s io n
Defining and understanding regional habitat associations of reptiles and
amphibians is a multi-dimensional and complicated facet of community ecology. With
so much stochasticity inherent to natural systems, questions at the community level are
difficult to address, and results of such studies can be equally difficult to interpret. When
interpreting the results of this study, we assumed that any given species present in upland
habitat in south Mississippi has an equal capture probability across sites; therefore, the
likelihood of capturing this species over a standardized amount of trap days should be
equal for any given site. Variation in weather patterns and landscape variables surely
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 25
affected our study; despite these factors, we believe that the duration and intensity of our
sampling allows us to compare relative abundances of reptiles and amphibians across
sites, and correlate habitat variables to these relative abundances.
The first objective of this study was to determine levels of heterogeneity within
historical upland longleaf pine habitat in south Mississippi, and it is apparent from our
ANOSIM that substantial variation exists. In contrast to the densely planted silvicultural
forests typifying much of the gulf coast plain, old-growth longleaf pine forests are
characterized by relatively open canopies and low basal areas, and were described by
Bartram (1791) as “savannah” habitat. This frequently burned, open-canopied longleaf
habitat described by Bartram is characterized by high percentages of grasses and forbs in
the groundcover, whcih this translates into some of the richest plant biodiversity in North
America: up to thirty species per square meter (Frost et al., 1986). As most forest stands
in south Mississippi are in some stage of forest regeneration, it follows that basal area and
percent canopy cover are two factors explaining much of the variation between surveyed
stands.
Our second objective was to correlate reptile and amphibian community
differences with the observed differences in longleaf pine habitat. Although we did see
significant structure in our CCA, indicating that our measured habitat variables explained
some of the variation in observed herpetofuanal communities, most of the community
variation observed in our study was not explained in the CCA model. The missing
variation in our model is likely due to both historical effects and landscape level effects
(habitat fragmentation, proximity to roads, previous forestry practices such as intensive
site preparation, etc.).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 26
The highly significant structure (P = 0.0006) and species-environment correlation
(P = 0.0020) on axis three of the CCA (where most of the variation is explained by
canopy cover and basal area) indicate that percent canopy cover and basal area are two
factors contributing greatly to faunal-community relationships in longleaf pine forests of
South Mississippi. Canopy cover is widely recognized as a key factor determining forest
microhabitat, plant growth and survival, and wildlife habitat within the forest (Jennings et
al., 1999). Furthermore, fire is a key factor in maintaining the open canopies of longleaf
pine savannahs in the southeast (Komarek, 1965). Without fire, hardwood encroachment
ensues which results in canopy closure and, as less sunlight filters to the forest floor, loss
of herbaceous grandcover (Waldrop et ah, 1992). This herbaceous groundcover is
essential for growth, development, and reproduction of the gopher tortoise (Mushinsky
and McCoy, 1994), which has been portrayed as an ecosystem engineer (Jones et ah,
1994) of the longleaf pine ecosystem. Although recommended percent canopy cover for
gopher tortoises is between thirty and fifty percent (Wilson et al, 1997), our CCA did not
positively associate gopher tortoises with open canopied forests. This anomaly is likely
because most active gopher tortoise burrows within our study sites were encountered
along roads or powerlines. These edge areas are characterized by more open canopies
and grassier understories than the interior areas containing our drift fences, where we
collected habitat parameters. To further stress the importance of open canopies to gopher
tortoises, this animal has been shown to abandon its burrow primarily in association with
changes in overstory structure (increased canopy cover) which result in shading of active
burrows (Aresco and Guyer, 1999).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Our research supports the findings of several previous studies regarding the
importance of canopy cover to herpetofaunal community structure in the southeast. For
example, Meshaka and Layne (2002) described openings with sparse or no shrub cover as
critical factors allowing the persistence of xeric specialists on a long unbumed site in
Florida. In addition, a study of herpetological diversity within young growth forests of
South Carolina found a negative correlation between reptile diversity and upland canopy
closure (Russell et al., 2002). Lastly, Dodd and Franz (1995) found snake communities
of high pine habitats in north-central Florida to be more diverse and evenly distributed
than snake communities sampled in closed xeric hammock habitat.
Many previous habitat association studies targeting herpetofaunal communities
specifically address community differences in response to management practices; for
example, Litt et. al. (2001) found higher capture rates of both Aspidoscelis sexlineatus,
and Sceloporus undulatus in burned habitat plots when compared to control or herbicide
treated plots. Although we did not find notable habitat associations for either of these
species when examining recency of prescribed bum, we did find Aspidoscelis sexlineatus
to be associated with open habitats. This is not surprising given that teiid lizards are
known to have the highest rates of oxidative metabolism among reptiles (Asplund, 1974),
and may require open habitats to achieve their preferred high body temperatures.
The results of this study and previous studies indicating the significant influence
of basal area and canopy cover upon longleaf pine communities have important
management and conservation implications. In addition to the gopher tortoise, many
other longleaf pine residents and specialists have undergone similar evolutionary
pressures within this ecosystem; consequently, lowering stand densities and basal areas
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 28
will likely benefit many members of this imperiled community. In the aftermath of
Hurricane Katrina, development pressures have increased, and continue to increase, in
coastal Mississippi. As remaining tracts of longleaf pine forests become more and more
fragmented, and forested land continues to be sub-divided and developed, the long-term
persistence of many native longleaf pine species relies heavily on management practices
on public lands. To conserve the wealth of biodiversity represented within the longleaf
pine ecosystem, it is important to recognize the benefits to herpetofauna of lowered stand
densities, and the marked effect these management decisions have on community
structure.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 29
CHAPTER III
THE BLACK PINESNAKE (PITUOPHIS MELANOLEUCUS LODINGI): SPATIAL
ECOLOGY AND ASSOCIATIONS BETWEEN HABITAT USE AND PREY
DYNAMICS
A b s t r a c t .— As remaining longleaf pine forests become increasingly
fragmented, wildlife management becomes a daunting task, especially for rare and
relatively unstudied taxa such as the Black Pinesnake (P. m. lodingi). Appropriate
management for P. m. lodingi depends on our understanding of the ecology of this sub
species; consequently, there are two main objectives of this study: to address questions of
spatial ecology for Black Pinesnakes on a range-wide scale, and to employ small
mammal trapping within home ranges of telemetered snakes to investigate the
relationship between prey density and spatial ecology of these snakes. P. m. lodingi
exhibited large home ranges (Minimum Convex Polygon (MCP) home ranges were 92 -
396 ha), frequently crossed roads, and average (non-zero) movement distance for these
snakes was 338 meters per location event. Core home ranges of telemetered pinesnakes
were characterized by significantly greater abundances of hispid cotton rats ( Sigmodon
hispidus), and significantly higher total captures of hispid cotton rats and cotton mice
(Peromyscus gossypinus). By better defining the relationship between spatial ecology
and prey dynamics for P. m. lodingi, land managers will be better equipped to manage
and conserve remaining Black Pinesnake populations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 30
The longleaf pine savanna ecosystem once stretched along the coastal plain
landscape, nearly unbroken, from southwestern Virginia to eastern Texas (Bartram,
1791). As of 2000, approximately 841,800 hectares of naturally regenerated longleaf
pine stands remain; this figure represents less than 3% of the historic range (Ware et al.,
1993; Frost, 2006). A myriad of factors have contributed to the decline of this fire-
dependent ecosystem including: decimation from logging and the naval stores industry,
conversion into agricultural or silvicultural areas, fire-suppression, invasive species, and,
more recently, habitat fragmentation and development (Outcalt, 1997; Varner and Kush,
2004; Frost, 2006). The historic and continued decline of longleaf pine forests poses a
threat to approximately 72 resident reptile and amphibian species and 30 specialist
species: reptiles and amphibians whose ranges fall completely within the historic range of
longleaf pine forests (Guyer and Bailey, 1993; Means, 2006). The Southern Hognose
Snake (Heterodon simus), Flatwoods Salamander ( Ambystoma cingulatum), and Gopher
Tortoise ( Gopherus polyphemus) are just a few examples of highly imperiled longleaf
pine specialists characterized by declining populations (McCoy and Mushinsky, 1992;
Means et al., 1996; Gibbons et al., 2000; Tuberville et al., 2000). Although it is difficult
to distinguish natural population fluctuations from population declines, many populations
of longleaf pine taxa, such as the Black Pinesnake ( Pituophis melanoleucus lodingi) and
the Mimic Glass Lizard (Ophisaurus mimicus) are suspected to be in decline, but
insufficient historical and natural history data exists to determine geographic extent and
severity of these declines. With the precipitous decline in area of this habitat type,
concern has arisen regarding the dearth of natural history knowledge available for many
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 31
longleaf pine specialists that are apparently declining, especially those that are cryptic or
occur at naturally low densities.
Until recently, very little was known concerning the natural history and spatial
ecology of the Black Pinesnake, a large colubrid longleaf pine specialist whose historic
range spanned from far western Florida to extreme eastern Louisiana (Conant and
Collins, 1991; Paul Moler pers. comm.). Currently, this sub-species is thought to be
extirpated from Louisiana and has not been reported west of the Pearl River for nearly
forty years. C. M. Duran’s (unpubl. data) radio-telemetry study during the late 1990’s,
conducted within a managed gopher tortoise reserve on Camp Shelby (an Army National
Guard training facility, DeSoto National Forest), elucidated habitat preferences and
movement patterns of this taxon and dispelled the belief that Black Pinesnakes primarily
use gopher tortoise burrows as refugia. The average minimum convex polygon (MCP)
home range in Duran’s study (unpubl. data) was 47.5 hectares, and he found that Black
Pinesnakes use pine stumpholes almost exclusively for refugia. Since the only field
research addressing Black Pinesnake ecology and natural history occurred on a managed
gopher tortoise preserve, on habitat far superior to most habitat remaining within the
range of Black Pinesnakes, our first objective was to address questions of spatial ecology
on a range-wide scale, especially in areas characterized by fragmented or sub-optimal
habitat.
As remaining longleaf pine forests become increasingly fragmented, wildlife
management becomes a daunting task, especially for relatively unstudied taxa such as the
Black Pinesnake. Since Black Pinesnakes move long distances (C. M. Duran unpubl.
data) and often fall victim to road mortality (Olsen and Lee, 2007; Baxley and Qualls,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 32
this study), our second objective was to employ small mammal trapping within home
ranges of telemetered Black Pinesnakes to investigate the relationship between prey
density and spatial ecology of these snakes. We hypothesized that core home ranges of
Black Pinesnakes would have greater abundances of small mammals and greater total
mammal biomass than areas within home ranges that were seldom used. By examining
these relationships, we sought to indirectly test one of the basic tenets of ideal free
distribution theory (Fretwell and Lucas, 1970; Fretwell, 1972) which states that predator
distribution should be proportionally tied to prey distributions. If our small mammal
community experiment supports this theory, we should find that black pine snakes spend
more time in areas characterized by “better” (i.e. higher abundance, greater biomass)
small mammal communities. By better defining the spatial ecology of Black Pinesnakes
and the relationship between spatial ecology, small mammal communities, and habitat
use, we will be better able to manage and conserve this unique longleaf pine specialist.
M a t e r ia l s a n d M e t h o d s
Study Area.—Between April 2005 and December 2006, we conducted radio
telemetry within six counties in South Mississippi: George, Harrison, Jones, Marion,
Perry, and Wayne (Figure 3.1). Four of the six study sites were within DeSoto National
Forest; the remaining two sites, Marion County and George County, are within the
Marion County Wildlife Management Area and on private land respectively. Overstory
vegetation at all sites was dominated by either longleaf pine (Pinus palustris), or a
combination of longleaf pine and slash pine {Pinus elliotti) or loblolly pine (Pinus taeda).
All sites have a history of prescribed fire and have been burned at least once within the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 33
past five years. Level of fragmentation varied among sites; for example, the two sites
within the Chickasawhay District of DeSoto National Forest (Jones and Wayne Counties)
were fragmented only by forest service roads, while all of the remaining sites were
bisected by both paved roads and private in-holdings.
Figure 3.1. Locations of radio-tracked Black Pinesnakes in South Mississippi.
Survey and Radio-telemetry Techniques.—Black Pinesnakes were captured using
common herpetological sampling techniques including drift fences with funnel traps
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 34
(Gibbons and Semlitsch, 1981; Campbell and Christman, 1982; Fitch, 1987), manual
searching, and road surveys. Each drift fence (30.5 m long by 1.2 m high) was equipped
with four funnel traps, one at each end on each side. To sample multiple sites within a
local region simultaneously, twelve or more drift fence/trap arrays were deployed at all
times between 15 March and 1 July 2004, 2005 and 2006.
All Black Pinesnakes captured were transported to the University of Southern
Mississippi where we assessed feasibility of radio-transmitter implantation for each
individual (transmitter < 5% of snake mass). Temperature-sensitive radio transmitters
(Holohil Systems, Ltd. SI-2 and AI-2 transmitters) were surgically implanted within the
coelomic cavity of suitably large Black Pinesnakes following the procedures of Reinert
and Cundall (1982); furthermore, we did not perform surgery on snakes captured between
August and February, to avoid late-season surgery-related mortalities (Rudolph et al.,
1998).
In an attempt to increase post-surgury survival, all snakes were released at the
point of capture after being held in the laboratory for a maximum of five days (Hardy and
Greene, 1999). After release, snakes were located every three days during the spring and
summer, and once per week during the fall and winter using a Telonics telemetry receiver
and hand-held antenna. Microhabitat occupied, activity status, body temperature, road
crossing occurrences, and behavioral observations were recorded for each location
episode. Tracked individuals were located at different times of day, to assess the full
range of activity patterns, and care was taken not to disturb the animals when locating
them during active periods (which may bias microhabitat, activity, or behavioral
observations). All snakes were tracked until mortality ensued or until transmitters began
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 35
to fail. When transmitters began to fail, we captured the snake, removed the transmitter,
and released the individual at the point of capture.
Small Mammal Trapping .—From 2 June through 10 July 2006 small mammal
trapping, using Sherman traps, was conducted within the home ranges of four telemetered
Black Pinesnakes. ArcGIS was used to randomly select point locations within the core
home range of each snake (inner 10% useage area) as well as the outer home range of
each snake (outer 10% usage area) to serve as transect center-points. Within each home
range, two sets of two parallel transects (40 total Sherman traps) were simultaneously
employed within the core and outer home range of each snake; transect direction was
east-west for all replicates. We aimed to reduce temporal bias in our analysis of mammal
communities by using this paired survey technique. Transects were used to sample small
mammals because transect arrangements have been shown to yield more total captures,
more individual captures, and more species than grid arrangements (Pearson and
Ruggiero, 2003). Sherman traps were place 10 meters apart, and large (7.62cm x 8.89cm
x 22.86cm) and small (5.08cm x 6.35cm x 16.51cm) Sherman traps were equally
distributed between transects and placed such that no two small traps and no two large
traps were adjacent. We chose to space traps ten meters apart because this is the most
frequent distance for small mammal trap spacing reported in the literature (Bowman et
al., 2001). Traps were closed during the heat of the day, opened and baited with oatmeal
at dusk, and checked shortly after dawn on the following day. Species, total length, and
mass were recorded for each captured mammal, and all individuals were marked via a
unique spot-shaving pattern. Processing time varied from five to fifteen minutes;
subsequently, mammals were released at the point of capture.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 36
Habitat Data.— During the first week of July 2006, habitat data was collected at
the center point for each small mammal transect. The following characteristics were
quantified and recorded for each habitat sampling locality: dominant overstory plant
species, dominant understory plant species, percent canopy cover of trees, percent mid
story coverage, percent over-story shrub coverage, basal area of trees, percentage cover
of grasses, forbs, and shrubs in understory, ground-cover percentage of litter and bare
soil, and recency of fire. A spherical densiometer (Forest Densiometers model-C) was
used to measure canopy cover. At the center-point of each transect, four densiometer
readings were taken to estimate percent canopy cover facing north, south, east, and west;
these readings were averaged to estimate overall cover. A Cruz-All basiometer (model
59793) was used to measure factor-10 basal area. We used a one-meter quadrat (four
replicates per transect) to estimate average percentage of grasses, forbs, shrubs, litter and
bare soil comprising the groundcover, and we visually estimated the percentage of mid
story coverage and percentage of shrubs in the mid-story. To increase measurement
consistency, one individual (D. B.) measured all habitat parameters.
Analysis .—We used Hawth’s Analysis Tools for ArcGIS (Beyer, 2004) to
calculate minimum convex polygon (MCP) and fixed kernel home ranges for telemetered
Black Pinesnakes. For fixed kernel home ranges, we adjusted the smoothing factor until
the area of the 95% kernel equaled the area of the MCP, as recommended by Row and
Blouin-Demers (2006). As a result of low sample sizes and high mortality of Black
Pinesnakes, we did not statistically compare differences in home ranges and movement
activity between male and female individuals. Although given the paucity of published
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 37
information concerning this taxon, we feel that a descriptive report of these values is
warranted.
Since we hypothesized that biomass and abundance of small mammals would be
greater in the inner versus the outer home ranges of telemetered Black Pinesnakes, we
used one-tailed Wilcoxon signed-rank tests (alpha = 0.05) to assess these differences.
For our purposes, abundance was defined as mean number of captures per species per
day, and we used only non-recaptured mammals in our analysis (i.e., each mammal was
counted only once, even if re-captured several times). We also used a non-parametric
Wilcoxon test (two-tailed) to assess differences in total mammal captures between large
and small Sherman traps
To assess differences in habitat between core and outer small mammal transects,
we conducted non-parametric Wilcoxon signed-rank tests or parametric matched-pairs
tests (depending upon whether habitat parameters met assumptions of normality and
equal variances). To decrease the probability of Type I error, we applied a Bonferroni
correction (Miller, 1981) resulting in an alpha level of 0.005. We used JMP version 5.1
(2003) for all statistical analyses.
R e s u l t s
Survey results.—Between 2004 and 2006, we acquired twenty-three new Black
Pinesnake records. Using drift fences and funnel traps, we captured six Black Pinesnakes
over the course of 4,088 trap days (1 per 688 trap days). Five Black Pinesnakes were
hand-captured, and road cruising resulted in twelve new records (five live and seven dead
on road, DOR, Black Pinesnakes). For all methods and all years, May and June were the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 38
two most productive months for detecting Black Pinesnakes, presumably as a result of
breeding activity.
Radiotelemetry.—After implanting radio-transmitters into eight suitably large
adults, we tracked snakes between April 2005 and December 2006 for a total of 463
location events (Table 3.1).
Table 3.1. Tracking and home range data for five Black Pinesnakes (Pituophis melanoleucus lodingi) radio-tracked during 2005 and 2006.
#Days # Location Kernel (ha) Snake ID Sex Tracked Events MCP(ha) 50% 90%
1 F 401 101 196.99 27.9 142.4
2 M 136 29 182.61 45.2 145.9
3 M 381 85 395.98 62.1 289.2
4 M 561 114 91.68 18.4 64.3
5 F 487 118 171.72 24.2 116.4
Roads did not pose a barrier to movement; 18.6% of the time, Black Pinesnakes
crossed roads between location events, and two of eight Black Pinesnakes fitted with
radio-transmitters were found dead on paved roads. Black Pinesnakes predominantly
used rotted pine stump holes for refuge, and seldomly frequented abandoned gopher
tortoise burrows and other types of refugia (Figure 3.2). Telemetered individuals were
found above ground only 29.6% of the time. In our study, snakes were seen fleeing into
refugia 5.8% of the time, and snakes exhibited a decrease in body temperature during
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 39
13.8% of all location attempts, indicating a change in body position within the
environment (e.g. snake fled from observer). When all location events are included in
analysis, average movement distance per location event is 214.46 m.
Blackberry thicket Non-stump hole Unidentifiable hole c o Brush Pile oC3 j Armadillo Burrow Open # of Location Events Figure 3.2. Radio-telemetric locations of 8 Black Pinesnakes, Pituophis melanoleucus lodingi, tracked during 2005 and 2006. When only non-zero movement distances are included in analysis, average movement distance per location event jumps to 337.98 meters; snakes exhibited greatest movement activity during the month of June (Figure 3.3). It is of note that snakes moved shorter distances during the severe drought of 2006 when compared to 2005 (Figure 3.3). Black Pinesnakes did not appear to exhibit true hibernation behavior and were seen above-ground, or found to have moved to different refugia, on warm days Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 40 throughout all months of the year. The longest movement over a three day period by any snake was 2,050 meters. Snakes appeared to exhibit some degree of fidelity to stump- holes; 6.5% of the time snakes moved to a previously occupied stump-hole. Five snakes tracked for more than 135 days exhibited large minimum convex polygon (MCP) home 2 c 400 = 300 S 200 IE 100 loco coidlo inm cn iococo cococo coco to cococo co oo ooo ooo ooo ooo ooo ooo o rS® OXDCD Month Figure 3.3. Mean monthly movement distances (meters) of Black Pinesnakes, Pituophis melanoleucus lodingi, in South Mississippi. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 41 ranges; estimates ranged from 92 to 396 hectares, much larger than the 47.5 hectare (average MCP home range) previously reported for Black Pinesnakes on Camp Shelby (C. M. Duran unpubl. data). Black Pinesnakes in this study exhibited extremely high mortality rates; of eight snakes implanted with transmitters, only two survived to the end of the study. Two snakes were found dead on paved roads, one snake was apparently predated, one snake died from transmitter migration, and the remaining two snakes were found dead, unpredated, above-ground, during the height of the severe drought of 2006. Both snakes found dead during the drought were emaciated, and necropsy of these two individuals revealed no transmitter complications. One of these snakes, a female, crawled to a seasonally flooded bottomland area (which was dry at the time of the drought) and remained there for nearly a month before she eventually died. Small Mammal Trapping .—Over the course of 1440 trap days, small mammal trapping resulted in 179 captures of 64 individuals representing six species (Figure 3.4). Wilcoxon signed rank tests (one-tailed) indicated a greater abundance of hispid cotton rats (Paired T-ratio = 2.25, p = 0.022) in core home ranges as well as higher total captures of hispid cotton rats and cotton mice in core home ranges of telemetered Pinesnakes (Paired T-ratio= -2.76, p = 0.005). Total biomass was higher in the core home range, though this difference was not statistically significant (Paired T-ratio = -1.77,/) = 0.059). Furthermore, significantly more mammals were captured in large Sherman traps (x2 = 7.38, p = 0.007) when compared to small Sherman traps. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 42 Peromyxus gossypnus Neotoma floridana vi Sigmodon hispid us znI Ochrotomjs nuttalli Didelphis marsupialis Blarina bre vicauda 1------1------l------1------1------| Core Home Range 0 5 10 15 2 0 2 5 3 0 Number of Individuals H f Outer H ome Range Figure 3.4. Species and number of small mammal individuals captured in core versus outer home ranges of Black Pinesnakes. Habitat Analysis .—Small mammal transects within core home ranges of telemetered Black Pinesnakes were characterized by significantly less canopy cover, and less forbs, litter, and bare soil in the groundcover (Table 3.2). Transects within core home ranges had higher percentages of grass comprising the groundcover, had been burned more recently, and had lower percentages of midstory and shrub coverage, though these differences were not statistically significant. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 43 Table 3.2. Means, standard deviations, test statistics, and p-values for habitat variables collected from mammal transects within core (I) versus outer (O) home ranges of black Pinesnakes (matched pair analysis). Habitat variables followed by an asterisk did not meet parametric assumptions and were analyzed using Wilcoxon Matched Pairs Signed Rank tests. Mean SD Variable I 0 I O Test Statistic P % Canopy Cover 41.0 67.2 17.2 19.5 4.27 0.0007 Bum recency (mo) 22.5 40.5 9.9 13.3 1.88 0.16 % Shrubs in midstory 48.8 70.0 22.1 14.1 3.83 0.03 % Midstory coverage 14.3 40.0 9.4 32.4 1.83 0.16 % Forbs* 5.5 0.1 5.5 0.3 -45.5 0.0001 % Grass* 11.8 4.8 15.3 7.1 -21.5 0.14 % Litter* 88.8 98.1 8.5 2.5 43.5 0.001 % Bare soil* 4.7 0.6 5.2 1.4 -39.0 0.0001 % Shrubs 29.8 45.4 19.1 23.9 1.97 0.07 DISCUSSION The results of our study highlight many similarities and a few striking differences between the Black Pinesnake and its eastern and central congeners, the Northern Pinesnake ( Pituophis melanoleucus melanoleucus), Louisiana Pinesnake {Pituophis ruthveni) and Florida Pinesnake {Pituophis melanoleucus mugitus). Average movement distances and minimum convex polygon home ranges of Black Pinesnakes in this study are larger than, though still comparable to, those previously reported for eastern and central members of the genus Pituophis (Franz, 1986; Himes and Hardy, 2002; Gerald et Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 44 al., 2006); all of these snakes are known to inhabit large home ranges and move long distances. Percent of location events where snakes returned to a previously occupied retreat for Black Pinesnakes (6.5%) was similar to values recorded for Northern Pinesnakes (7.2%) (Gerald et al., 2006); this potentially reflects both the importance of these retreat sites, and existence of some level of fidelity to refugia for both of these sub species. Perhaps one of the most notable ecological difference between the Black Pinesnake and its eastern and central congeners is that the black pine is not known to excavate nests, hibemacula, or pocket gopher burrows (Burger et al., 1988; Burger and Zappalorti, 1991; Himes, 2000; Franz, 2001). Furthermore, the Black Pinesnake inhabits a unique ecological niche because, unlike the Louisiana Pinesnake and the Florida Pinesnake, the Black Pinesnake is not sympatric with pocket gophers (Geomys breviceps shares its range with the Louisiana Pinesnake and Geomys pinetis shares its range with the Florida Pinesnake) (Himes, 2000; Franz, 2001). Given the absence of pocket gophers within the range of the Black Pinesnake, alternative food sources, hibemacula, and refugia become increasingly important. As most old-growth longleaf pine stumps were removed and utilized in the Naval Stores industry, and gopher tortoise burrows are no longer prominent features of the landscape, the continued removal of longleaf pine stumps during intensive site preparation and salvage logging operations poses a serious threat to this taxon. It is very possible that presence and distribution of pine stump holes may be a feature that limits within-range abundance of Black Pinesnakes. The disturbingly high mortality of Black Pinesnakes in this study and Louisiana Pinesnakes in a previous study (19 mortalities of 38 telemetered snakes, Himes et al., Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2002) suggests that population declines may be due in part to adult mortality in excess of annual recruitment. Given the high frequency with which Black Pinesnakes cross, and suffer mortality from, roads, the boom in post-Hurricane Katrina development of the Mississippi Gulf Coast does not bode well for already-fragmented Black Pinesnake populations. Furthermore, the slow-moving nature of Pinesnakes increases their susceptibility to vehicular mortality, especially in areas where intentional killing prevails because of the negative attitudes toward all snake species. Two of the snakes tracked in this study died during the severe drought of 2006. Although we cannot conclusively say that these mortality events are a result of drought-related stress, we can recognize the potential for drought-related population fluctuations, a phenomenon that is completely unstudied in upland snakes. The results of our paired small mammal experiment fit within the context of ideal free distribution theory, and hold important management and conservation implications. Corroborating our hypothesis, core home ranges of four telemetered Black Pinesnakes had higher mammal abundance, especially of hispid cotton rats, and greater mammal biomass than little-used areas along the periphery of these snake’s home ranges. Although there have been no previous studies using telemetry and small mammal trapping to assess potentially prey-driven habitat use within the genus Pituophis, a recent study examining the relationship between edge habitat use and small mammal abundance in Elaphe obsoleta and Coluber constrictor found no evidence that the preferential use of edge habitats by both snakes was a consequence of small mammal abundance (Carfagno et al., 2006). Carfagno et al. (2006) further postulated that prey distribution is less important than thermoregulatory reasons for generalist predators, especially in higher Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 46 latitudes. This hypothesis is supported by findings that Elaphe in Ontario, Canada prefer edge habitats regardless of the fact that these habitats do not contain higher abundances of small mammals (Blouin-Demers and Weatherhead, 2001). In contrast, Madsen and Shine (1996) and Heard et al. (2004) found evidence for prey-driven habitat use of Australian pythons ( Liasis fuscus and Morelia spilota metcalfei). It is impossible to separate causation from correlation regarding the preferential use of open, herbaceous habitats, which also contain higher abundances of hispid cotton rats, by Black Pinesnakes. Both taxa have been previously shown to be denizens of open, grassy areas. Our findings of higher percentages of forbs within core home range mammal transects (P = 0.0001) is not surprising given the heavy reliance of Hispid cotton rats on herbaceous ground-cover for refugia and food (Goertz, 1964; Kincaid and Camereon, 1982; Duran, 1994). Further, bluestem grass ( Schizachyrium spp.), the historically dominant grass of western longleaf pine ecosystems, is thought to be a major food item for this taxon (Miller, 1999). Duran (1998) found that Black Pinesnakes significantly avoided areas of high canopy closure, and were found more often in areas with high percentages of herbaceous vegetation. Regardless of whether prey abundance drives (or is correlated with) habitat use by these snakes, our findings hold important management implications, and indicate that habitat preferences of P. m. lodingi and Sigmodon hispidus tend strongly toward the historic longleaf pine ecosystem, and away from the current status quo of densely planted silvicultural land. Although efforts to increase the use of growing season prescribed burning on public land have increased over the past few years, dormant season burning of longleaf pine forests in Mississippi is still the norm. In the aftermath of Hurricane Katrina, problems characteristic of growing Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 47 season bums (liability, smoke, etc.) have been exacerbated due to the resulting fuel loads from the storm. By prioritizing management resulting in open, grassy, forb-rich, longleaf pine ecosystems (growing season bums, lower planting densities) populations of Black Pinesnakes, and other longleaf pine specialists will likely benefit. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 48 CHAPTER IV ASSESSING SECOND AND THIRD ORDER HABITAT SELECTION BY BLACK PINESNAKES {PITUOPHIS MELANOLEUCUS LODINGI): A BOOLEAN MODEL USING GEOGRAPHIC INFORMATION SYSTEMS Abstract.—With the historic and continued fragmentation and degradation of longleaf pine forests, an increasing need emerges to identify remaining Black Pinesnake populations and understand species-environment correlations such that conservation efforts for this taxon are both directed and appropriate. This study has three main objectives: 1) To use radio-telemetry data in conjunction with GIS modeling to assess second and third order resource selection (habitat suitability model) for Black Pinesnakes west of the Mobile River; 2) to use occurrence data (historic and recent) to validate our habitat suitability model; and 3) to quantify on-the-ground habitat characteristics of Black Pinesnake locality records and evaluate differences in environmental variables between historic and recent occurrence records, potentially as a barometer of habitat loss. On the landscape level, Black Pinesnakes require large tracts of evergreen forests (historically Pinus palustris) in upland areas lacking cultivated crops, pasture/hay fields, development, and roads. Within home ranges, Black Pinesnakes require some degree of patchiness and habitat heterogeneity; snakes were found closer to scrub/shrub habitat, open areas, and deciduous forests than expected. Our Boolean model reduced the area deemed suitable by 84.3%, while accurately categorizing 75% of recent Black Pinesnake occurrences (post-1987), demonstrating the ability to develop robust models with very small sample sizes. We found significant habitat differences between sites of recent and historic P. m. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 49 lodingi occurrence, which corroborate the findings of our Boolean model. Recent sites of occurrence were, on average, 597 m from areas with a model score of “optimal,” while sites with no recent records were, on average, 1582 m from areas with an “optimal” score; furthermore, recent sites were characterized by significantly less canopy cover, lower basal area, less midstory cover, greater percentages of grass, bare soil, and forbs in the groundcover, less shrubs and litter in the groundcover and a more recent bum history than sites with no recent records. These data suggest that areas containing historic Black Pinesnake populations are simultaneously becoming unsuitable on both the landscape and microhabitat levels. The Boolean model approach we utilized is easy to understand and update, and can be used to develop testable hypotheses of changes to Black Pinesnake populations resulting from changes in the landscape. INRODUCTION Recent advances in Geographic Information Systems (GIS) technology have provided conservation biologists with much-needed tools to assess habitat selection and clarify range distributions of vertebrates. Prior to the widespread adoption of GIS, vertebrate distribution maps and indices of habitat selection were most often constructed using historic locality records, which are inherently biased and unreliable (Freitag et al., 1998; Reddy and Davalos, 2003). The problems of biased locality data and false absences are exacerbated for snakes and other taxa that are found in inaccessible habitats, are difficult to detect, or are characterized by cryptic behaviors (Rodda, 1993; Kery, 2002). Over the past two decades, GIS has been widely used to predict occurrences of or generate species-environment suitability models for birds and large mammals (e.g. Dubuc Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 50 et al., 1990; Clark et al., 1993; Knick and Dyer, 1997; Corsi et al., 1999). In contrast, herpetofauna are under-represented in the GIS-modeling literature, presumably because reptiles and amphibians typically move less and inhabit smaller home ranges than large mammals and birds, thus increasing the difficulty of landscape-level habitat modeling. Despite the potential difficulty of applying these techniques to herpetofauna, species- specific modeling for reptiles and amphibians is becoming more common (Lipps, 2005; Browning et al., 2005; Dayton and Fitzgerald, 2006; Rouse, 2006); however, most of these studies address questions at fine scales (e.g. timber rattlesnake hibemacula within Madison County Wildlife Management Area; terrestrial salamander abundance within Hoosier National Forest). Herpetological modeling studies on a broad scale are feasible only for species that have large home ranges, move long distances, or have extremely specialized species- environment associations. For example, Baskaran et al. (2005) utilized the known association of Gopher Tortoises ( Gopherus polyphemus) with specific soil and land cover types to create a GIS model which successfully predicted the occurrence of Gopherus polyphemus burrows over a five-county region in Georgia. Although most North American snake species do not occupy sufficiently large home ranges on which to focus broad-scale GIS modeling efforts, one notable exception is the Black Pinesnake (.Pituophis melanoleucus lodingi). P. m. lodingi move long distances, inhabit large (up to 396 ha) home ranges (Duran, 1998; Baxley and Qualls, in prep), and is considered one of the most imperiled snakes in North America. Currently a federal candidate for Endangered Species Act listing, the Black Pinesnake once ranged from western Florida to eastern Louisiana along the Gulf Coast Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 51 Plain, but is now extirpated from Louisiana and exhibiting trends of decline in Mississippi and Alabama. Guyer and Bailey (1993) list the Black Pinesnake as a specialist of the longleaf pine ecosystem, an ecosystem that has been reduced to less than five percent of its historic range due to fire suppression and conversion of these forests to agricultural, silvicultural, or human use areas (Frost 1993; Outcalt 1997; Varner and Kush 2004). With the historic and continued fragmentation and degradation of longleaf pine forests, an increasing need emerges to identify remaining Black Pinesnake populations and understand species-environment correlations such that conservation efforts for this taxon are both directed and appropriate. This study has three main objectives: 1) To use radio-telemetry data in conjunction with GIS modeling to assess second and third order resource selection, and create a habitat suitability model for Black Pinesnakes west of the Mobile River; 2) to use occurrence data (historic and recent) to validate our habitat suitability model; and 3) to quantify on-the-ground habitat characteristics of Black Pinesnake locality records and evaluate differences in environmental variables between historic and recent occurrence records, potentially as a barometer of habitat loss. In this study, we follow the selection hierarchy presented by Johnson (1980) that defines second order selection as the home range of an individual and third order selection as the usage of different habitat components within the home range of an individual. With the overall efforts to conserve species in the United States viewed as opportunistic and uncoordinated (Ando et al., 1998), we hope the results of this study will provide tools to organize and implement both directed and appropriate conservation actions aimed at restoring Black Pinesnake populations. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 52 M a t e r ia l s a n d M e t h o d s Radio-telemetry Between March and July 2004-2006, we surveyed for Black Pinesnakes in South Mississippi using standard herpetological techniques including drift fences with funnel traps (Gibbons and Semlitsch, 1981; Campbell and Christman, 1982; Fitch 1987), manual searching, and road surveys. We implanted radio-transmitters into and subsequently radio-tracked eight individuals (see Baxley and Qualls, in prep for further methods). From April, 2005 through December, 2006, snakes were located every three days during the spring and summer, and once per week during the fall and winter. Home ranges of radio-tracked individuals fell within six counties in South Mississippi: George, Harrison, Jones, Marion, Perry, and Wayne. Due to high mortality, only five of the eight radio tracked Black Pinesnakes were tracked for sufficient periods (> 130 days) to warrant inclusion in the habitat model. Selection o f Environmental Variables Several different approaches have been used to determine the species- environment requirements of vertebrates for the purpose of discriminating between environmental variables for use in habitat models. Deductive approaches use a-priori species requirements and expert opinion to select habitat parameters of importance, while inductive approaches derive an “ecological signature” from the characterization and statistical testing of animal location data (Corsi et al., 2000). As model performance may be reduced by the inclusion of expert opinion (Clevenger et al., 2002; Seone, et al., Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 53 2005) we chose an inductive approach, using data collected in the field to drive the model development, instead of relying on a-priori assumptions of suitable habitat. To assess second order resource selection by Black Pinesnakes, we first generated random home ranges in ArcGIS (Version 9. Redlands, CA: ESRI, Inc; 2004) using Hawth’s Tools (Beyer, H. L. 2004. Hawth’s Analysis Tools for ArcGIS. Available at http://www.spatialecology.com/htools) such that each random home range was equal in area to the average minimum convex polygon (MCP) home range of telemetered Black Pinesnakes (radius = 812.3 m). We chose to use randomly generated home ranges to represent habitat availability because landscape proportions are not equal to proportions of habitats in random home ranges; consequently, inclusion of the whole landscape as “available” results in high type I error rates (Katnik and Wielgus, 2005). After generating random home ranges in ArcGIS 9, we used land use / land cover (LULC) layers (30 meter pixels, USGS National Land Cover Dataset, 2001, Homer, et al. 2004), of fifteen environmental variables as well as a roads layer obtained from the U. S. Census Bureau (2000; Tiger/line file) as the starting point for environmental variable selection (Table 4.1). After using ArcGIS 9 to calculate percent composition, within each real and simulated home range, for each variable, we employed Analysis of Similarity (ANOSIM, a Monte Carlo re-sampling technique) to test the null hypothesis that the compositional proportion of each variable is equal across simulated and real home ranges. The Black Pinesnake is an upland longleaf pine specialist, and there is no doubt that well drained soils are important to Black Pinesnake distribution. Surprisingly, soil analysis of sites for which recent (2004-2007) black Pinesnake records exist revealed large variances in soil composition (range of percent sand 5-95%, range of percent silt 1- Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 54 58%, range of percent clay 0.5-37%; soil hydrometer method Sheldrick and Wang, 1993). This high variance in soil composition leads us to believe that habitat structure may trump soils (as the limiting factor for Black Pinesnakes) in many fragmented or marginal areas whereP. m. lodingi populations have managed to persist. In order to err on the side of suitable habitat overestimation instead of potentially eliminating areas of suitable habitat characterized by marginal soils, we chose not to incorporate a soils layer into our model. Furthermore, unlike its congeners, P. m. lodingi is not known to excavate burrows for nesting or hibemacula (like P.ruthveni, and P. m. melanoleucus; Burger et al, 1988; Burger and Zappalorti, 1991; Himes, 2000), nor does it feed on Geomys sp., pocket gophers, (like P. ruthveni and P. m. mugitus; Himes, 2000; Franz, 2001); these life history traits serve to reduce the probability of strict “priority” (McDearman, 2005) soil association for P. m. lodingi. To identify variables of importance for third order habitat selection, we used Euclidean distance analysis following the methods of Conner et al. (2003). Within each home range, we generated 1000 random points. For each of these random points, and for each radio-telemetry point, we measured mimiumum distances to each land use/land cover feature. Habitat variables that were not present in any home range, and were, on average, greater than 800 m from telemetry points (e.g. open water made up 0% of total home range area for all snakes; mean distance from telemetry points was 1883 m) were eliminated from the distance analysis, because selection at this level is considered 2nd order selection. We used a Hotelling’s T-test to test the null hypothesis that the resulting vector of mean distances did not differ from a vector of 1 ’s; if the vector of mean distances differs from a vector of l ’s, habitat selection exists (see Conner et al., 2003 for Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 55 a complete review of this method). After using MANOVA to establish overall habitat selection, we tested each of the remaining variables for statistical significance using Wilcoxon Matched-Pairs Signed Rank tests. By using matched-pairs tests, and removing among-home range variation, we were confident that we were assessing within-home range differences in resource selection/use by snakes. Habitat Modeling and Validation After identifying the environmental associations of Black Pinesnakes on the landscape level, we constructed a Boolean overlay model incorporating the six habitat variables found to be significant in our 2nd order analysis. To establish model parameters for each significant habitat variable, we used the minimum and maximum values observed for each variable in home ranges of radio-tracked Black Pinesnakes. Using the focal statistics function of ArcGIS, each 30 m raster cell was assigned a predicted suitability score based on the number of model parameters satisfied within a radius of 812.3 m. Areas that fulfilled all six model input parameters (Table 4.3) were given a rank of “6,” areas fulfilling five of six parameters were given a score of “5” and so forth. Methods of Boolean overlay are widely used in GIS-based approaches of multi- criteria-decision-making (Hopkins, 1977; Tomlin, 1990; Berry, 1993; Malczewski, 1996), and have proven to be simple yet effective tools for questions at this scale. One major criticism of habitat suitability models is the general absence of efforts to validate and ground-truth models after completion. To determine the quality of our model, we compiled all known locality data (154 records) for P. m. lodingi, and overlaid these records on our Boolean model. As our model output is essentially a prediction of the suitability of a raster cell as the center of a hypothetical circular home range, occurrence Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 56 records were considered correctly classified by the model when they were <813 m from areas of predicted suitable habitat (Boolean score of “6”). We also assessed model performance by constructing a plot of percent occurrence records versus percent reduction in land area to determine what Boolean ranking should serve as a threshold for “suitability” (i.e. the point where number of occurrence records is maximized while percent land area is minimized). We then visited all locality records west of the Mobile River (barring those that were inaccessible due to military road closures or gated, private property) to evaluate on-the-ground habitat and examine microhabitat differences between recent and historic records. By assessing microhabitat variables at these P. m. lodingi occurrence sites, we sought to explain some of the missing variation in our model (i.e. investigate if low quality forests are present at sites where historic, but not recent, records exist for Black Pinesnakes). When validating our model, we used twenty years (1987) as a cut-off for “recent” records. Although we realize there are likely some under sampled areas where Black Pinesnakes persist, but have not been seen post-1987, we felt that twenty years was a conservative (albeit subjective) cut-off point by which we could be relatively confident in the existence of present Black Pinesnake populations. To assess habitat on a small-scale for each P. m. lodingi occurrence, we quantified and recorded the following characteristics of each site: dominant vegetation species, percent canopy cover of trees, percent mid-story coverage, percent shrub coverage, basal area of trees, percentage cover of grasses, forbs, and shrubs in understory, ground-cover percentage of litter and bare soil, and recency of fire. We used a spherical densiometer (Forest Densiometers model-C) to measure canopy cover. At each site, four densiometer readings were taken to estimate percent canopy cover facing north, south, east, and west Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 57 of the locality point; these readings were averaged to estimate overall cover. A Cruz-All basiometer (model 59793) was used to measure factor-10 basal area. We used a one- meter quadrat (four replicates per site) to estimate average percentage of grasses, forbs, shrubs, litter and bare soil comprising the groundcover, and we visually estimated the percentage of mid-story coverage and percentage of shrubs in the mid-story. To increase measurement consistency, one individual (D. B.) measured all habitat parameters. We used non-parametric Wilcoxon tests (alpha = 0.05) to examine differences between habitat variables at recent (post-1987) and historic sites of P. m. lodingi occurrence. Sites with no viable habitat (i.e. 15 sites that were completely urban) were not included in analysis. We used JMP (Version 6. SAS Institute, Cary, NC, 1989-2005) and PRIMER 5.2.9 (Primer-E Ltd, Roborough, Plymouth, United Kingdom) for statistical analyses. R e s u l t s 2nd and 3rd Order Selection In our second order analysis, ANOSIM revealed six habitat variables that differed significantly between real and simulated home ranges on the landscape level: percent evergreen forest, percent pasture/hay, percent crops, percent open water, percent developed, and density of “A3” (secondary) roads (Table 4.1). Home ranges of Black Pinesnakes were characterized by less open water, pasture/hay, cropland, and roads, and more evergreen forest than random home ranges. For third order analysis, MANOVA indicated highly significant overall selection in all four tests (P<0.0001 for Wilk’s Lambda, Pillai’s Trace, Hotelling Lawley, and Roy’s Max Root tests). Distance analysis to evaluate 3rd order habitat selection by Black Pinesnakes resulted in significant selection or avoidance of seven variables (Table 4.2); we eliminated open water, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 58 developed (medium), developed (low), developed (high), barren, roads, and cultivated crops from analysis since these variables represented 0% total area of all home ranges. P. m. lodingi exhibited selection “for” shrub/scrub, deciduous forest, and developed (open) areas within their home ranges and selection “against” woody wetland, pasture/hay, mixed forest, and emergent wetland. Although 3rd order selection for evergreen forests was not statistically significant, radio-tracked Black Pinesnakes were found closer to this habitat class than any other (mean distance = 13.1m). The only other LULC where Black Pinesnakes were found, on average, within 150 meters was scrub/shrub (mean distance = 141.7 m); mean distances of radio-tracked Black Pinesnakes to all other LULC classes averaged 3,236 meters. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 59 Table 4.1. Mean percent composition, standard deviation, test statistic, and significance level for each variable tested for compositional differences between real (R) and simulated (S) home ranges. Means and standard deviations were calculated from the entire data set, while, statistical values reflect random 2 sub-sampling (ANOSIM). Instead of percent composition, we used density (m/km ) for the roads data. Variable Mean SD Test Statistic P R S R S (Global R) Evergreen forest 68.5 31.4 9.5 22.2 0.52 0.008 Open water 0 1.7 0 3.9 0.48 0.048 Developed (open) 2.8 5.2 1.1 5.9 0.35 0.056 Developed (low) 0 1.4 0 4.0 0.54 0.048 Developed (med) 0 0.5 0 1.9 0 1.000 Developed (high) 0 0.3 0 2.6 0 1.000 Barren 0 0.4 0 1.4 0.24 0.167 Deciduous forest 0.6 5.6 1.0 10.5 -0.06 0.444 Mixed forest 9.3 11.1 5.2 10.6 -0.11 0.873 Shrub/scrub 9.1 11.8 8.3 9.1 -0.11 0.738 Grass/herbaceous 1.0 1.3 2.0 3.5 -0.09 0.841 Pasture/hay 0.3 7.5 0.7 10.3 0.38 0.032 Cultivated crops 0 1.9 0 4.0 0.94 0.008 Woody wetland 8.2 19.0 7.6 21.7 -0.084 0.841 Emergent wetland 0.1 1.1 0.3 5.7 -0.06 1.000 Roads (secondary) 0 97.7 0 242.4 0.31 0.050 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 60 4.2. Mean, test statistic, and p-values of Wilcoxon Sign-Rank tests comparing Euclidean distances of radio-tracked Black Pinesnakes (R) and Euclidean distances of random simulated points (S) to land use/land cover variables. Variable Mean Distance (m) Test Statistic P R S Evergreen forest 13.1 15.4 -639.5 0.44 Developed (open) 252.7 360.8 -16000 0.0001 Deciduous forest 3520.6 3588.2 -7760.5 0.004 Mixed forest 199 135.9 22443.5 0.0001 Shrub/scrub 141.7 253.8 -19448 0.0001 Grass/herbaceous 1991.6 2002.5 3404 0.207 Pasture/hay 1093.3 1044.2 6315 0.017 Woody wetland 477.2 352.3 -21129 0.0001 Emergent wetland 2222.5 2069.6 16021 0.0001 Boolean Model Within ArcGIS 9, we constructed model parameters (Table 4.3) using the results of our 2nd order analysis (landscape-level selection for evergreen forests and against cultivated crops, pasture/hay, roads, developed areas, and open water). Each raster cell within the study area was reclassified according to the number of model parameters met Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 61 within a 812.3 m radius, from a score of 0 (meeting none of the model parameters) to 6 (meeting all of the model parameters). Table 4.3. Input parameters for Boolean model Variable Minimum % Maximum % Evergreen forest 57.17 no upper limit Open water 0.00 0.00 Cultivated crops 0.00 0.00 Pasture/hay 0.00 1.48 Developed areas 1.78 4.14 Roads 0.00 0.00 Validation We overlaid a total of 154 historic and recent P. m. lodingi locality points on our Boolean model. Although only 15.7% of the geographic area within our model was < 813 meters from a pixel with the optimal Boolean score of six, 75% of “recent” black Pinesnake records (84 of 112) were located <813 m from areas with an optimal suitability score of six. In effect, the model reduced the area deemed suitable by 84.3%, while accurately Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 62 *v , *Y> 'f'* . < f V, > J\ ■I * Mf\ * * ■ - i . * ■v *.: y s ■ ; ‘.' V ’> -—“ '•’4-•. ■ •*,, ’ -V *.< i'S; *4' > I M*& *£•' . ' r a ,------4 * V % ' *V V r I Ii ■1 s .*.• •« ?> « * ** ^.-y j • * .. . ^ c / f'U ■> ( r .. . ■ o c. { v,\ -* t '.> ■ '4 j t ',. ° J i' "4 r ) i 0 25 50 100 Kilometers 1 i i i 1 i i i I Legend H ♦ <813 m eters from predicted suitable m Predicted Suitable Habitat it >813 m eters from predicted suitable Figure 4.1. Boolean model representing areas of predicted suitable habitat (Boolean score = 6), for Black Pinesnake. Seventy-five percent of occurrence records were located <813 m of suitable habitat (dark diamonds). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 63 100.0 9 0 .0 8 0 .0 - 7 0 .0 6 0 .0 I ■ % Occurrence records!; 5 0 .0 I - A % Reduction in Area ; 4 0 .0 3 0 .0 20.0 10.0 0.0 Boolean score threshold for habitat suitability Figure 4.2. Percent reduction of total area classified as suitable habitat and percent of occurrence records correctly classified as being on suitable habitat are graphed against the corresponding Boolean score thresholds. A Boolean score of 6 indicates predicted suitable habitat, while a value of zero indicates furthest from suitable. The graph indicates the trade-off that occurs between the reduction in total area classified as suitable and the accuracy in classifying occurrence records. categorizing 75% of recent test points. Mean distance of “recent” (post-1987) points of Black Pinesnake occurrence to optimally suitable habitat (Boolean score = 6) was 597 m, while mean distance of “historic” records areas with a score of six was 1582 m. In contrast, random points within the landscape had a mean distance of 2710 m from habitat with a Boolean score of six. While our model best performs when areas <813 m of cells scoring six are considered suitable habitat, a plot of percent occurrence records versus percent reduction in land area finds a Boolean rank of “4” is the threshold for suitability if distance is not taken into account. Only 35.5% of the study area received a Boolean score >4, while 71.9% of all occurrence records were found in these areas (Figure 4.2); targeting these areas will optimize the probability of finding new Black Pinesnake populations, while minimizing area to be searched. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 64 Habitat Data Sites of historic P. m. lodingi occurrence with no known recent (post-1987) records differed from those with known recent occurrences. When comparing recent and historic sites, recent sites were characterized by significantly less canopy cover, lower basal area, less midstory cover, greater percentages of grass, bare soil, and forbs in the groundcover, less shrubs and litter in the groundcover and a more recent bum history (Table 4. 4). The only habitat parameter that did not differ between recent and historic sites was percent shrubs in the midstory. Table 4.4. Summary of habitat variables quantified (mean ± standard deviation) at 131 sites of recent (post-1987) and historic Black Pinesnake occurrence. Variable Mean ± SD Test Statistic P Pre-1987 Post-1987 Chi2 Canopy Cover (%) 63.3 ±27.6 60.0 ±86.0 10.89 0.001 Bum recency (mo) 73.4 ±46.7 45.5 ±44.5 13.76 0.0002 Basal area (10F) 10.0 ±6.1 6.3 ±3.7 13.56 0.0002 % Shrubs in midstory36.8 ±32.8 30.6 ±29.7 1.20 0.27 % Midstory coverage 49.2 ± 37.0 28.4 ±28.7 8.08 0.0045 % Grass 8.2 ±9.0 21.4 ± 21.7 17.3 0.0001 % Forbs 3.7 ±5.5 6.2 ±8.0 4.48 0.0344 % Shrubs 39.8 ±25.8 27.8 ± 16.5 5.65 0.0175 % Litter 92.3 ± 13.1 73.8 ±27.0 21.62 0.0001 % Bare soil 5.2 ± 12.3 17.0 ±20.6 16.83 0.0001 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 65 D is c u s s io n Habitat structure and resource availability are two of the most important factors that affect movement patterns in snakes (Gregory et al., 1987). In this study, resource selection by Black Pinesnakes was evident both within home ranges of snakes (3rd order) and on the landscape level (2nd order). On the landscape level, Black Pinesnakes require large tracts of evergreen forests (historically Pinus palustris) in upland areas lacking cultivated crops, pasture/hay fields, development, and roads. With the size, growth, and resource demands of the human population acting as the most important agents of change in spatial patterns of biodiversity (Vitousek et al., 1997; Sala et al., 2000), it is no surprise that our model predicted P. m. lodingi occurrence in areas where wide-spread human disturbance is not yet evident on the landscape level. Although global estimates of human population predict declines in U.S. population by 2050 (Cohen, 2003), U.S. trends of migration to warmer and more temperate areas of the country will probably continue. As human population growth and resulting development in the southeast increases, many isolated tracts of longleaf pine forest will become functionally extinct because prescribed fire will no longer be feasible. This impending fragmentation coupled with the notion that large home range size increases extinction vulnerability (Woodroffe and Ginsberg, 1998) adds to the urgency and importance of conserving and managing large tracts of relatively undisturbed land, especially areas that buffer existing Black Pinesnake populations. Within home ranges, Black Pinesnakes require some degree of patchiness and habitat heterogeneity; snakes were found closer to scrub/shrub habitat, open areas, and deciduous forests than expected. At first glance, the within-home range preference for Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. deciduous forest patches seems anomalous, given the fidelity of P. m. lodingi to longleaf pine forests. Upon closer inspection of this finding, we see that mean distance from radio-telemetry points to deciduous forest is over 3,500 m; consequently, the statistical preference for this habitat type is probably caused by one snake spending substantial time in a disturbed, cut-over area characterized by deciduous trees, which are often the only trees remaining after longleaf pine forests are harvested. Thermal quality has been shown to influence habitat use in other snakes (Huey et al., 1989; Row and Blouin-Demers, 1996; Shine and Fitzgerald, 1996; Blouin-Demers and Weatherhead, 2002; Row and Blouin-Demers, 2006b), so it is likely that these open/patchy areas provide thermoregulatory opportunities for Black Pinesnakes that are not present in more densely forested, homogenous areas. It is also of note that the Northern Pinesnake (Pituophis melanoleucus melanoleucus) is positively associated with open areas, both in Tennessee and in New Jersey (Gerald et al., 2006b; Burger et al., 1988). These open, patchy areas likely take on greater importance for the Black Pinesnake in the current landscape due to the artificially high planting density and basal area of present-day longleaf pine forests in comparison to the historic landscape (Landers et al., 1995). Areas containing historic Black Pinesnake populations are simultaneously becoming unsuitable on both the landscape and microhabitat levels. Our Boolean model results coupled with the results of on-the-ground habitat analysis give congruent support to the notion that reduced black Pinesnake populations are primarily due to habitat fragmentation and habitat alteration. When compared to recent Black Pinesnake occurrence records, historic (pre-1987) sites of Black Pinesnake occurrence are consistently farther from areas deemed “suitable” by our model; furthermore, these areas Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 67 are typified by qualities indicative of high planting densities, and fire suppression (closed canopies, high percentages of leaf litter, low percentages of grass etc.). Given the status of Black Pinesnakes as longleaf pine specialists, the importance of open-canopied forests and dense, herbaceous ground-cover can not be over-emphasized. Moreover, the selection “for” open/patchy areas within home ranges is not so surprising, given the structural patchiness and high stand variability typical of old-growth longleaf pine forests (Noel et al., 1998). In a recent study, Waldron et al. (2006) postulated that habitat specificity to the imperiled longleaf pine ecosystem may be a significant contributor to the decline of the Eastern Diamondback Rattlesnake; we suspect that this is also true for the Black Pinesnake. In terms of model performance, we accurately classified 75% of recent black Pinesnake occurrence points by using independent home range data to construct model input parameters. Boolean Models often overestimate the amount of suitable habitat (Type 2 error; Standora, 2002), and this can be difficult to detect when only presence data is available to evaluate the model results. However, our model appeared to perform well in this category also, with <16% of the study area predicted suitable in the best performing model. A paradox of working with endangered and rare species is that sample sizes large enough to develop robust estimates of habitat composition are rarely available. With few exceptions, habitat modeling is rarely attempted with sample sizes as low as those in the present study (five home ranges). The success of this model serves as an excellent example of the utility of GIS modeling amidst the constraints of studying rare taxa. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 68 Two other strengths of our modeling approach in terms of its application to applied conservation and management are deserving of comment. First, our Boolean approach utilizes an easily comprehended methodology requiring the end user to have no expertise in GIS or habitat modeling. Roux et al. (2006) argue that unobstructed knowledge flow between science and management is one of the great challenges for sustainable ecosystem management and overly complex models may not be integrated into policy-making if they are perceived by managers as merely an “intellectual curiosity.” Second, our model can be easily updated as additional data on Black Pinesnakes or updated digital data such as LULC and roads become available. Even without new data, hypothetical data could be utilized to make testable hypotheses concerning the effects of different land management or conservation plans, or to predict the impact that future landscape changes may have on populations of the Black Pinesnake. This forecasting of the anthropogenic effects on patterns of biodiversity is seen as one of the most important uses of habitat models (Guisan and Thuiller, 2005) and our model lays the framework for additional study in this area. In summary, we believe the current model provides both land managers and biologists with a much-needed tool to assist in directed conservation efforts and further our knowledge of the biology of the Black Pinesnake. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTERV RAPID POPULATION EXPANSION AND RECENT DIVERGENCE OF EASTERN PITUOPHIS: P. M. LOD1NGI, P. M. MELANOLEUCUS, AND P. M. MUGITUS A b s t r a c t .—Systematics of the genus Pituophis have been described by several phylogenetic hypotheses, and the intraclade relationships of eastern Pituophis: P. m. melanoleucus, P. m. lodingi, and P. m. mugitus are particularly unstudied and unresolved This study has three main objectives: 1) to expand upon Rodriguez-Robles and Jesus- Escobar’s phylogenetic study through additional sampling of wild caught individuals of the three easternPituophis sub-species from across their ranges; 2) to employ an additional mitochondrial gene to see if we can obtain better phylogenetic resolution of the three sub-species; and 3) to use molecular data to make inferences about demographic history. Phylogenetic analysis using Maximum Parsimony, Maximum Likelihood, and Bayesian methods did not recover Pituophis sub-species as monophyletic groups. Analysis of pair-wise sequence differences resulted in a distribution not significantly different from the rapid expansion model (P = 0.57), and we estimate rapid population expansion as occurring between 63,125 and 1,518,174 years BP. Given the high degree of morphological and ecological dissimilarities between P. m. mugitus, P. m. lodingi, and P. m. melanoleucus, future genetic work using microsatellite loci will likely reveal increased geographic structure within this clade. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 70 Introduction Relationships within the transcontinental genus Pituophis have long been debated and are characterized by several taxonomic hypotheses (see Rodriguez-Robles and Jesus- Escobar, 2000 for a complete review). Recent systematic analysis of new world gopher, bull, and Pinesnakes indicated that the eastern Pinesnake clade (P. m. lodingi, P. m. mugitus, and P. m. melanoleucus) and the western Pinesnake clade (P. c. affinis, P. c. annectens, P. c. bimaris, P. c. catenifer, P. c. deserticola, P. c. sayi, P. c. vertebralis) represent divergent lineages and should be recognized as two distinct species (Rodriguez- Robles and Jesus-Escobar, 2000). In their study, monophyly of the eastern Pinesnake clade was well supported (MP bootstrap value = 100%); however, within-clade relationships for this group were characterized by low bootstrap values, and were relatively unresolved. Particularly unresolved in the Rodriguez-Robles and Jesus- Escobar study (2000) was the monophyly of the black Pinesnake {Pituophis melanoleucus lodingi) group (MP Bootstrap value = 56%). Aside from this study, to our knowledge there is no other information concerning molecular systematics of the eastern Pinesnake clade as a whole, and the black Pinesnake in particular. Non-molecular work, a phenetic analysis (using morphological variables) of eight North American subspecies of Pituophis conducted by Reichling (1995), found that canonical coefficients for eastern Pinesnakes clustered independently from those of both western and Louisiana Pinesnakes. Although Reichling did not test differences within the eastern Pinesnake clade for significance, he divided the eastern Pinesnakes into four Operational Taxonomic Units (OTUs), and found several morphological variables that, when averaged for each OTU, differed between groups. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 71 The interest in resolving the phylogenetic relationships within Pituophis is attributable to several factors: the trans-continental distribution of this genus, the unique and charismatic natural history of many currently recognized sub-species of the genus, and the growing conservation concern for several members of the Pituophis clade. Eastern Pinesnakes (P. m. lodingi , P. m. mugitus, and P. m. melanoleucus) are rare and secretive taxa differing in range distribution (Fig. 5.1), morphology, and ecological niche (Franz, 1986; Burger et al., 1988; Burger and Zappalorti, 1991; Conant and Collins, 1991; Himes, 2000; Franz, 2001; Himes and Hardy, 2002; Gerald et al., 2006). P. m. lodingi is currently a federal candidate for Endangered Species Act listing, and several states have acknowledged declining eastern Pituophis populations. For example, P. m. lodingi is listed as imperiled in Mississippi (S2 designation), critically imperiled in Louisiana (SI), and of high conservation concern in Alabama. Additionally, P. m. melanoleucus is listed as threatened in New Jersey, and P. m. mugitus is listed as rare in Florida (S3 designation). As habitat of the eastern Pinesnakes continues to be fragmented and degraded, populations continue to decline, and an increasing need emerges for resolution of the phylogenetic relationships among these snakes. Despite ongoing debates concerning the validity of the sub-species rank (see Burbrink et al., 2000 for a review) and perpetual discordance among scientists regarding species concepts (Mayden, 1997; Coyne and Orr, 2004), we feel that attempts to clarify and expand upon knowledge of molecular variation are warranted, especially for taxa of conservation concern. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 72 ^ / j - P. raffcvenf - P. m. lo d in g i ■ P. 771. meianoieucus Area of hybridization P. m. mugitus W l * a / Figure 5.1. Distribution of eastern Pituophis, adapted from Conant and Collins, 1991. This study has three main objectives: 1) to expand upon Rodriguez-Robles and Jesus- Escobar’s study through additional sampling of wild caught individuals of the three eastern Pituophis sub-species from across their ranges; 2) to employ an additional mitochondrial gene to see if we can obtain better phylogenetic resolution of the three sub species; and 3) to use molecular data to make inferences about demographic history. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 73 M a t e r ia l s a n d M e t h o d s Molecular techniques.—Total genomic DNA was extracted from scale clips or sloughed skins of Pituophis using the Qiagen Dneasy extraction kit (QIAGEN Inc., Valencia, CA). The polymerase chain reaction was then used to amplify 855 bp of the NADH dehydrogenase 4 (ND4) gene using two sets of primers developed specifically for Pituophis based on sequences from GenBank (5’-CTAAAACTGGGRGGATACGG-3’ and 5 ’-CTATTATCCTTTAGTAGAGGG-3 ’ and 5 ’ -TC AATTTT ACT A A AACT GGG-3 ’ and 5 ’ -GAGTT AGC AGGTCTT ATGGC-3 ’). We also amplified 517 bp of the NADH dehydrogenase 2 gene (ND2) using primers (H3518 and H5382) from deQueiroz et al. (2002), one primer (L4437b) from Kumazawa et al (1996) and one primer (16Sb) from Palumbi (1996). PCR cycling conditions consisted of an initial 1 min denaturing step at 95°C followed by 30 cycles of 1 min at 95°C, 1 min at 50°C and 3 min at 72°C. A final elongation step of 7 min at 72°C completed the cycle. PCR products were cleaned using the ExoSAP-IT system (USB Co., Cleveland, OH) and then used as template in a cycle sequencing reaction with ABI BigDye™ Terminator cycle sequencing kit (Perkin Elmer Applied Biosystems, Foster City, CA). All sequencing reactions were sephadex cleaned (Princeton Separations, Adelphia, NJ) prior to gel runs by the Iowa State University DNA Sequencing and Synthesis Facility. Phylogenetic Analysis. — We obtained sequence data from Rodriguez-Robles and Jesus-Escobar’s systematic study from GenBank (Table 5.1). We aligned GenBank Sequences to our Pituophis sequences using Sequencher v. 4.1 (GeneCodes Co., Madison, WI); see Table 5.2 for all samples used in this study. In order to compare our Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 74 sequences to those obtained via GenBank, we truncated ND4 sequences to achieve a length of 808 base pairs. Table 5.1. Mitochondrial NADH sequences of Pituophis and Lampropeltis obtained from GenBank. N represents the number of sequences for each species Species N GenBank Species N GenBank accession # accession # L. getula 1 AF138759 P. m. melanoleucus 3 AF138770 AF141116- AF141117 P. m. mugitus 2 AF138769- P. m. lodingi 2 AF141114- AF141115 AF141118 P. eaten ifer 1 AF141106 We inferred phylogenetic relationships with PAUP* v.4.0bl0 (Swofford, 2002) using the methods of maximum parsimony (MP) and maximum likelihood (ML), and a Bayesian inference of the phylogeny was performed with Mr. Bayes v. 3.1 (Ronquist and Huelsenbeck, 2003). For ML analyses, we inferred a suitable model of sequence evolution using MODEL-TEST 3.06 (Posada and Crandall, 1998). Lampropeltis getula was used as an outgroup in our ND4 phylogenetic analyses because this was one of the outgroups used by Rodriguez-Robles and Jesus-Escobar, and Lampropeltis has been previously been shown to be closely related to Pituophis (Rodriguez-Robles and Jesus- Escobar, 1999). For analysis of our combined ND4 and ND2 data set (which excludes Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 75 sequences used in Rodriguez-Robles and Jesus-Escobar, 2000), we used one wild-caught Pituophis catenifer as an outgroup. Table 5.2. Number and location of Pituophis samples used in this study. Taxon # Samples State P. m. lodingi 26 Mississippi P. m. mugitus 3 Georgia P. m. mugitus 2 Florida P. m. melanoleucus 2 North Carolina P. m. melanoleucus 9 New Jersey P. m. melanoleucus 2 Alabama P. catenifer 2 California Phylogenetic support was assessed through bootstrapping (Felsenstein, 1985) with 1,000 rounds of resampling for the MP and ML analyses and through posterior probabilities for the Bayesian analysis. To assess geographic structure, we used 1315 bp of combined ND2 and ND4 sequences (from 36 individuals) and 808 bp of ND4 sequence (from 43 individuals) to construct mismatch distributions, conduct Analysis of Molecular Variance (AMOVA), and estimate haplotype diversity (h) and nucleotide diversity (%) using Arlequin version 2.001 (Schneider et al., 2000). We used models of Rogers and Harpending (1992) to determine the probability of rapid population expansion in eastern Pituophis and to estimate the time since expansion. For the mismatch distribution, we used 3.5 years as the average generation time for Pituophis (Fitch, 1970), and we selected Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 76 1,4%/MY as the mutation rate, a conservative value similar to previously reported herpetofaunal mutation rates for ND4 and cytB (Carranza et al., 2006; Castoe et al., 2007). R e s u l t s All phylogenetic methods resulted in the recovery of basically the same topology (Figure 5.2). We found little difference between analysis of ND4 sequences alone and analysis of the combined ND2 and ND4 data set; consequently, we display only the results of the ND4 analysis since these are directly comparable to the previous analysis of Rodriguez-Robles and Jesus-Escobar (2000). Analysis of 808 bp of sequence from the ND4 gene for 44 individuals of Pituophis melanoleucus resulted in fifteen unique haplotypes, and maximum divergence between haplotypes was 14 bp, while analysis of 1315 bp of combined ND4 and ND2 sequence resulted in twelve unique haplotypes and maximum divergence between haplotypes was 7 bp (see Table 5.3 for h and n). While the eastern Pituophis clade formed a strongly supported monophyletic group, sub-species within this clade were not recovered as monophyletic; furthermore, one of the two weakly supported clades within eastern Pituophis included members of all three sub species. There are considerable differences between bootstrap values obtained in Rodriguez and Escobar’s analysis when compared to the re-analysis with added samples of eastern Pituophis (Fig. 5.2). Although both analyses indicated very high support for monophyly of the eastern Pinesnake clade (bootstrap value = 100% for both analyses), coherent subspecies groups deteriorate with the addition of samples. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 77 P. m. lodingi (1) i P.m. fcdiRgf(l)* 55/ 54/73 Pm . lodingi (Id) P. m. mugitus (1) P. m. lodingi (1) * P. m. lodingi (3) P. m. melanoleucus (8) * P. m. melanoleucus (2) P. «.lodingi (4) P.m. mugitus(1) P. m. melanoleucus (1) * P. m, melanoleucus (1) * 80/92/78 P. m. mugitus (2) * 100/ 98/100 P. m. mugitus (1) * P. catenifer (1) ------p. catenifer(I)* Lampropeltis getula (1) ’ 0.01 substitutions/sito Figure 5.2. Maximum liklihood tree from the ND4 data set. Support values are indicated along the branches in the following order: MP bootstrap, ML bootstrap, and Bayesian posterior probability. Number of sequences for each taxa are indicated in parentheses. * indicates a group contains sequence(s) from GenBank. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 78 Table 5.3. Haplotype diversity (h), nucleotide diversity, range of pair-wise distance values (d), number of samples (s), and number of unique haplotypes (u) for Pituophis using ND4 and combined ND2 and ND4 datasets. 71 u Total ND4 0.80 +0.05 0.003+0.002 2.25+1.26 43 15 ND4 & 0.81±0.06 0.002+ 0.001 2.15+1.23 35 12 ND2 By Subspecies P. m. lodingi ND4 0.60+0.10 0.001±0.001 1.06+0.73 26 N D4& 0.62+0.11 0.001+ 0.001 1.57+0.97 23 ND2 P. m. melanoleucus ND4 0.56+0.15 0.002+ 0.001 1.64+1.04 12 ND4& 0.56+0.17 0.001+ 0.001 1.39+0.94 ND2 P. m. mugitus ND4 0.90+0.16 0.007+0.005 5.80+3.34 ND4& 1.00+0.27 0.004+0.003 4.67+3.13 ND2 While Rodriguez-Robles and Jesus-Escobar found 82% support for monophyly of all P. m. melanoleucus, this relationship is completely unresolved when additional eastern Pituophis haplotypes are included in the analysis. Little geographic structuring is evident for eastern Pituophis; furthermore, haplotypes from the northern portion of the range Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 79 appear to be embedded within the matrix of southern haplotypes. AMOVA indicated that sequence variation was much higher within populations (63% ND4 only; 67% using combined ND2 and ND4 data set) than between populations (37% ND4 only; 33% using combined data set). Upon analysis of pair-wise sequence differences, we found that our distribution was not significantly different from the rapid expansion model (P = 0.54 for combined ND2 and ND4; P = 0.57 for ND4 only), and we estimate rapid population expansion as occurring between 63,125 and 1,518,174 years BP (Fig. 5.3). Mismatch Distribution 300 250 200 I------Observedj 150 I______Si m ula ted^ | 100 50 0 0/ 0- Figure 5.3. Distribution of pair-wise sequence differences (using 808 bp ND4). Estimate of rapid population expansion: 63,125 - 1,135,938 years BP. Estimate of rapid population expansion for combined ND4 and ND2 data set (1315 bp, not shown) 74,729 -1,518,174 years BP. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 80 D is c u s s io n The first objective of this study was to re-assess Rodriguez-Robles and Jesus- Escobar’s phylogenetic hypothesis for eastern Pituophis; unexpectedly, expanded geographic sampling failed to recover monophyletic groups representing the three eastern Pituophis sub-species. The lack of mitochondrial evidence for geographic structure among eastern Pituophis coupled with support for the rapid population expansion model suggests relatively recent divergence of these three taxa. Given our estimates of recent population expansion (between 63,125 and 1,518,174 years BP), it is likely that the distribution of Pituophis populations was limited to southern North America during the last Pleistocene glacial advance, and underwent subsequent rapid population expansion from the southeastern United States. Many vertebrate taxa have displayed similar phylogeographic patterns of vicariant separation during glacial periods, followed by dispersal to contemporary ranges (Jaarola and Tegelstrom, 1995; Bematchez and Wilson, 1998; Taberlet et al., 1998; Hewitt, 1999). An anomalous outcome of this and Rodriguez and Escobar’s (2000) study, perhaps lending support to the hypothesis of post-Pleistocene rapid expansion from the Southeastern United states, is the P. m. mugitus individual that appears to be basal to the eastern Pituophis clade (Fig. 5.2), differing from all other eastern Pituophis individuals by 11-14 bp. Given the recent divergence of eastern Pituophis, we must keep in mind potential effects of incomplete lineage sorting (Avise, 2000), which may obscure our understanding of the phylogenetic history of this clade. Although our data suggest that these three sub-species do not represent reciprocally monophyletic groups, by no means should this be viewed as evidence to treat eastern Pituophis as one management unit. There are many ecological and behavioral Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 81 differences among the eastern Pinesnakes; consequently, we feel that these three forms of Pituophis inhabit distinct ecological and evolutionary niches, and may represent incipient species. Oftentimes adaptive differences between populations are not partitioned in a monophyletic manner, and since the process of speciation produces polyphyletic relationships, which eventually progress to paraphyletic, and ultimately monophyletic relationships, Crandall et al. (2000) argues that using criteria of reciprocal monophyly to define management units for conservation is overly stringent. Furthermore, ecological differentiation may provide better evidence than molecular data for divergence relevant to population persistence and conservation (Legge et al., 1996; Crandall et al., 2000). Reciprocal monophyly is unlikely to be found in vertebrate populations characterized by high levels of gene flow (Crandall et al., 2000; Avise, 2004). Given the large home ranges of eastern Pituophis, it is possible that hybridization between subspecies has occurred with sufficient frequency to blur mitochondrial distinction between these groups; however, range reductions among all three taxa will likely reduce contact between sub-species thereby making it imperative to conserve and restore populations of all three eastern Pituophis. In efforts to preserve biodiversity, conservation biologists must attempt to maximize evolutionary potential through the protection of populations on separate evolutionary trajectories (O’Brien and Mayr, 1991; Hey et al., 2003; Vignieri et al., 2006), even if these populations are not reciprocally monophyletic. Considering the morphological and ecological dissimilarities between P. m. mugitus, P. m. lodingi, and P. m. melanoleucus, we recommend future genetic work using microsatellite loci to further investigate geographic structure and levels of gene flow within this clade. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 82 CHAPTER VI GENERAL CONCLUSION AND RECOMMENDATIONS The broad objective of this dissertation was to utilize multiple research tools (field surveys, radio-telemetry, GIS technology, molecular techniques) to ultimately provide land managers and biologists with information necessary to conserve and restore P. m. lodingi populations. Although this project is predominantly species-specific in focus, experimental biologists have given recent attention to the efficacy of merging single species research with ecosystem-oriented studies to achieve conservation goals amidst limited resources (Lindenmayer et al., 2007). Given the quantity of vertebrate and invertebrate species of current conservation concern, it is understandable that species- specific monitoring and management for all of these taxa is cost-prohibitive. Fortunately, species-specific research often identifies unique requirements of an organism which can easily be incorporated into ecosystem management plans. For example, recent evidence indicating the importance of rotting pine stumps to individual longleaf pine specialists (e.g. Black Pinesnakes, Eastern Diamondback Rattlesnakes) has resulted in new management trends aimed at preserving pine stumps, thus benefiting multiple taxa. We hope that the information presented here regarding black Pinesnake habitat preferences and spatial ecology is taken into consideration during future habitat management decisions. Our research supports several management practices that we believe will be beneficial to Black Pinesnakes, and surely other longleaf pine specialists, including: planting at lowered stand densities to reduce canopy cover (< 60%), substituting growing-season prescribed fires for the conventionally used winter bums, and taking precautions during logging activities such that pine stumps remain in-situ. Of Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. these recommendations, the use of growing season bums is especially important to Black Pinesnakes for two main reasons: 1) growing season bums result in increased herbaceous vegetation and forbs which are required by prey (Sigmodon hispidus), and 2) these fires typically bum in an uneven manner resulting in low to moderate levels of landscape patchiness. Our GIS model and distance analysis of third order habitat selection demonstrates that low levels of patchiness and heterogeneity are “preferred” by Black Pinesnakes. Unless major effort is invested into preventing the development and urbanization of private lands containing P. m. lodingi populations (i.e. via cooperative landowner incentives), Black Pinesnakes will likely exist only on public lands within the next half- century. With this in mind, biologists and land managers need to acknowledge the growing evidence (Bernardino and Dalrymple 1992; Hels and Buchwald, 2001; Smith and Dodd, 2003; Andrews and Gibbons, 2005; Aresco, 2005; Steen et al., 2006) detailing the negative impacts of roads on herpetofauna. Since P. m. lodingi inhabit large home ranges and move long distances, it is no surprise that road mortality (25% in this study) is high. As developed areas encroach on public lands, and National Forests and Wildlife Management Areas continue to be un-buffered and fragmented by new, paved roads, it is possible that, even on protected land, road mortality for Black Pinesnakes may be in excess of annual recruitment. Consequently, there is an urgent need for the scientific community to convey the importance of and foster research into new, innovative, cost- effective underpass designs such that human growth and ecosystem health are not mutually exclusive. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 84 Although we found no evidence for reciprocal monophyly of eastern Pituophis using mitochondrial genes, this by no means provides evidence to support treating eastern Pituophis as a single management unit. All three eastern Pituophis have suffered historic range reductions, and areas of potential gene flow between sub-species are shrinking due to habitat fragmentation and local population extinction. With this in mind, future genetic work employing additional molecular markers (microsatellites) will better reveal Pituophis genetic diversity and how this diversity is distributed between and among Pituophis populations. Further analysis of genetic diversity within the eastern Pinesnake clade will serve to better inform future conservation decisions, especially in an era of endangered species mitigation. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 85 APPENDIX A Staftsn No. so be Cnsditad Permit femtie. UNITED STATES DEPA -.TVIENT OF THE 1'NTERiCR 43615 04-0 ils FiS'H AND WILDLIFE SERVICE S»JKV*f~V. D ate Mississippi Sandhill Crane National Wildlife Refuge 03-02-2004 period of Use (inclusive) SPECIAL USE PERMIT From March 15 . 2004 T o J u ly 30 .2 0 0 4 permittee Name I Permitis Danna Smith (PI), Sam Holcomb, Cart Qualls I Department of Biological Sciences University of Southern Mississippi i Hattiesburg, MS 1304/237-1941, danna.*mith®usm.edu Purpose (specify in detail privilege requested, or units of products involved) To conduct the "Survey to Determine the Conservation Status of toe Mimic Glass Lizard in Mississippi", During a 3-wee!; period, permittees may survey for glass lizards at designated areas on the refuge.A combination of drift fences, pitfall and funnel traps, as wait as manual sea'ching, and road surveys witl be techniques used. Once traps are set up, permittees wish to survey daily for 14-21 consecutiva days. Captured fcards wiil fee removed from tfaps; they may be measured and mproductuve status assessed. One voucher specimen may i Reded Lorn We refuge. Prior to release, each lizard may be marked with dorsal paint mark and PIT tag for iderrtificafkm. Habitat data i he collected. j Description (specify unit m e n b e ia c jm e lw * « n 4 i»ui«t», w other rSssgrstabfe -Ui-aignWions;- Amourtt of fee if not a fixed payment, specify rate and unit of charge: {" Payment Exempt -Justification: * T Foil Payment i p merit -Balance of payments to be made as follows: Reeor f /merits Special Conditions , ; Permittees will closely coordinate the location and timing of their surveys with refuge biologist to reduce chance of disturbance to rwsSop ! . r i as t nuit c i ltd tim ng of surveys must be OK’d by refuge biologist (During the crane nesting season Mar 01- Jun 30, a I are* i8,u..« j — l - d to disturbance unless arranged]. Permitees will stop in at refuge office to check in at beginning of the 21-day - ri « -»»s wi I sgn out keys for refuge gate at office, Permittees will needa 4WD vehicle (preferably with winch'' to r Of ua-» vv ’"Vm fleas w.ll keep this permit with them or ;n their vehicle white on the refuge. A report on sumey fmd-’-g w u' -'-voted to the refuge biologist. m p t m l :s tMMKi i f f* » O.S. fak snd vv lr» «- t~ rrovftT: 3t>on£ -fi r®gervatksns. sxprsssod or STipfed h©.w., ^ o «lO*««««,*, y^F / ; ^ s « m u ttse / / 7 3'7" -yy'~-C'9^0? / Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 86 APPENDIX B Authorization iD: FS-27O!W5s02S9j Contact ID: gmb no . gsOd-cosz Expiration Dais: Use Code: U.S. DEPARTMENT OF AGRICULTURE Forest Service TEMPORARY SPECIAL - USE PERMIT (FSH 2709.11, sec. 54.6} AUTHORITY: Act of June 4,1897 University of Southern Mississippi hereinafter called the Holder, is hereby authorized to use, subject to the terms and conditions of this permit. National Forest System land identified within the unit area and described as 36 sites within T2N.R9W.S3S: T2N.Rf2W.S3: T1N.R3VV.S2: T1N.R10W.S22: TIN R12W.S?3:T1S.R12W.S12: TiS.R11W.S17&20: T1S.R12W.S19: T1S.R14W.S14: T3S.R9W.S20: T4S.R11W.S27&28: T6S R11W.SS. T55 R11W.S14 as shown on the attached Maps. This authorization cavers approximately2 acres. The holder is authorized to conduct the following activities and/ or install the following temporary improvements on the permitted area. Conduct surveys for the mimic glass lizard, including construction of temporary drift fences with pitfali and fur.nei traps (see attached Exhibit 1 - Research Proposal), TERMS AND CONDITIONS I. Use under this permit shall begin on 3/15/2004 and end on 3/14/2005. The permit shall not be extended. 2 The fee for this use is S2_. tt shall be paid in advance and is not refundable 3 The holder shall conduct the authorized activities according to the attached approved plans and specifications, Exhibits 1 & 2 . 4. The holder shall not install any improvements not specifically identified and approved above. 5. No soil, frees, or other vegetation may be destroyed or removed from National Forest System lands without specific prior written permission from the authonzed officer. 6 The holder shall comply with all Federal, State, county, and municipal laws, ordinances, and regulations which are applicable to the area or operations covered by this permit 7 The holder shall maintain the improvements and premises to standards of repair, orderliness, neatness, sanitation, and safety acceptable to the authorized officer. The holder dial! fully repair and bear the expense for alt damage, other than ordinary wear and tear, to National Forest System lands, roads and trails caused by the holder's activities. 8 Tire holder has the responsibility of inspecting the use area and adjoining areas for dangerous trees, hanging limbs, and other evidence of hazardous conditions which would pose a risk, of injury to individuals. After securing permission from the authorized officer, the holder shall remove such hazards. S. The holder shall be liable for any damage suffered by the United States resulting from or related to use of this permit, including damages to National Forest resources and costs of fire suppression. 10, The holder shall hold harmless the United States from any liability from damage to life or property arising from the holder’s occupancy or use of National Forest lands under this permit. I I . The holder agrees to permit the free and unrestricted access to and upon the premises at ail times for al! lawful and proper purposes not inconsistent with the intent of the permit or with the reasonable exercise and enjoyment by the holder of the privileges thereof. 12, This permit is subject to all valid existing rights and claims outstanding in third parties. 13. This penult may be revoked upon breach of any of the conditions herein or at the discretion of the authorized officer, Upon expiration or revocation of this permit, the holder shall immediately remove all improvements except those owned by the United States, and shall restore the site within 15 days, unless otherwise agreed upon in writing. If the holder fails to remove the improvements, they shall become the Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 87 property of the Unites States, tut teat will not relieve the holder of liability for the cost of their removal and restoration of the site, 14. This permit Is 8 license for the use of federally owned land. It does not grant any interest in real property This permit Is not transferable. The holder shatl not enter into any agreements with third parties for occupancy of the authorized premises and improvements, 16, Appeal of any provisions of this permit or any requirements thereof shall be subject to tee appeal regulations at 36 OFR, Subpart C. or revisions thereof. 16. This permit is accepted subject to the conditions set forth herein, conditions 1-17 and Exhibits JL&J. attached to and made a part of this permit, 17, The above clauses shall control if they conflict with additional clauses or provisions I have read and understand the terms and conditions and agree to abide by them, HOLDER U. S. DEPARTMENT OF AGRICULTURF University of Southern Mississippi Forest Service By By: •/f- Address:! 118 CoBweDrive #6013 HatBesbura. MS 39406-0001 Name dudith Henry THie:Dlsthct Render (Authorized Officer) Phone #: (601) 266-6906 Date: -iy~j Date. is lh«P e c w u fc WsvketfopAciaf TOporm »a«»3vtr^»«»gp nee*n^cjfBw?^fs*J fS*n S**wn tor«*. t tm s> * m » m , e to ty w m *»T «mP o m * **a, (*ftor«t Fanw!4»i An* f *m* Ad, H im tti Umv*9 Act MMh* 7#ro Panr-fc Act Act Of 3- ‘iNWwtvwa .Act fattooM itom USA* ms 7f*ia Act Aefof NovwnNw 1§, tS73, ArdtHtogkal feaiaspgt* P*otaelion Aest. «r*2 Alass* MMprt^ Hwniwrt iK-xta C »hi^r*t?afT Act P* $*cr«t*y 4 A^rieiture to a«u* $w**erfMf»r* 'or tfw us* *w oocuperey «f ?**¥«■*» fg ^ a Syswrt tank. *>• S*pr*t«y of A ^roitsrrj nojtafiavs St. X O f tUSS’t. Pm Sb43pw1 B, »«st«5A prtj ^2#)*»slm FrwKterB efV»ann*B not* fwrnponete swap** Of W5«3^ op»f*i»A aod’of rrv«?w»Mfi» t*Lir>r««gxfie« «#<«?; 8fttf*ntvwoapri ns^fef.Mt* *«*■*»<« me? tefcAteAeHttrgeaa rsppm •^s. <**% *Mi t*r»a»r.flfan. at**** Wn«waSoh. iM or«r saw*# fnfemtfEA oqtMW. 1>& fer mv®wsi *-**.«**». *»«**<$ cwwjne «** Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a p p e n d ix c HANCOCK S O t a . J gave the University of Southern Mississippi herpetology lab permission to conduei^i'foctclogtcal surveys for the mimic glass lizard f Qphteaurus minucus) on . my private land or 16 Section land 5 fence Date 8f &-CV y HANCOCK SCHOOL DISTRICT 17304 HWY. 603 KILN, MS 3SS58 x--" Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 89 APPENDIX D THE UNIVERSITY OF SOUTHERN MISSISSIPPI June 15. 2004 Dr, Carl Qualls Uepaifnctu of B iology I nnerMt} of Southern Mississippi Ucttk -.butg. MS 39406 Ri F \( Cc p ro v e d U H00SW Dear Dr, Quails, V.>m protocol entit.ed " Ophis-aurns Project” has, been reviewed and approved by the University of southern \!t >Ms»inpi Institutional Animal Care and Use Committee. The project is approved for the period from March 15. 2004 until September 30. 2007. If there are any substantive changes in the protocol, you are required to submit an addendum to the original proposal, which must be approved by the IACUC. ’1 he 1 \('l U gives assurance that the proposal complies with the public health policy on humane c.itv and use of laboratory animals and all applicable provisions of the Animals Welfare Act. cd Sheree Watson, Ph.D. Interim Chair. IACUC DEPARTMENT OF PSYCHOLOGY Sox it CD • Mscuesbuts,itriesburj MS ■ ’9406-Mi T.orse 16041 ioiv4177 • Fax (oOlS 266-551' •w.usaiedu Hattiesburg * L<>«g Beacf * Q a-a n Springs * BilsM * .bhn V. Sct'tinisS[xicc Center Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX E AfcO ’■« !» fR!> * s n i p LUCtDALE MISS. February 1!, 2004 * * Oatma Smith 404 M orth 2 151 Avenue Hattiesburg. MS 39401 Dear Danna. Please find enclosed maps for three different tracts as well as a plat map. The tracts are in Lucedale, Mississippi (George County) and are easily accessible. To find the tracts, all you have to do is drive on Highway 49 South to Highway 9£ East. You will stay or. Highway 98 for about 50 to 60 miles mid all three tracts are located on Highway 98. When you come to a new Chevron cm your right- hand side (south side of 98), you are on the tract labeled "Bush Dirt Pit". Please note where I drew in the Chevron station on the stand and plat maps. All tracts are under a hunting lease and therefore I need to know in advance when you will be present. Please call me with any questions. Sincerely, fSjUAMM Sarah Watkins Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 91 APPENDIX F 13 March 2004 :;'::;:,ADX1 X1 STRATIVS SCIENTIFIC ' LL M PERMIT :-0 WHOM I T WAV ...COKCBSN: Per-'.seiort is granted, to; Danna S m ith■ Ur:rvebSityi'df '■ Southern M issi 82 sIpp 1 Dept&vof B iological Science's lI?8,sCol lege Drive # SQiS m t t i e s b u r g , MS 324 On --OOOl, assisted by Dr. Carl Quails; Krista Noel, Thomas Mohrmar-, and Samuel Holcomb, no bbiltect amphibians, reptiles, and ama.11 mammals while undertaking a survey for :a&h±saurus (glass Lizard) species oh vei1-drained to excessiveiy wall-drained ■ &oi I s in s o u th e r n M i s s i s s i p p i . Collection hods may include cover boards, drift fences with pitfall traps and funnel traps, ana nand collection during foot surveys. Because- drift fence pitfall trap .’ill typically be placed in areas suitable for tortoises:, cam 'Utt oe taxen to avoid incidental captur f i or sec. Bucket traps should be fitted y>.th a cylinder or wire mesh of m l enough diameter to exclude juven-le e ’ j‘f■:**.& but large- enough to permit r i h - glass Iisards and smaller, lizards. anct snakes. Bucket traps should be tat o prevent over-heat ing o f captives and the buckets should have some sort of cover and moisture source on the bottom, buckets must also have holes in the bottom to allow drainage of rainwater. This habitat, type also supports the federally and state endangered Mississippi, g o p h e r f r o g {Rana se ro sa ), the state enaanqerea, federal candidate black pine sn a k e (PLtuopkis lodlngi: the. fed* r» 1 . state endangered gopher co r y o iw e iGovhorus polypnea *)Mi , t ) ed, state endangered mdlgo snak e i Drym&jrchGn c o ra ls} . a? * t southern hognose snake 'Hoterodor: si.vus} . It is ur.Iikv I t \ pecies except the tortoise w ill be encountered, and as stresses above, trapping apparatus should bo designed i.e. minimize the likelihood of incidental capture of tortoises. However, should any of these species be 1 ncxderta 3.ly captured within trapping gear or ericounfeied while, conducting surveys on foot, brief handling is authoritea to secure voucher photographs adequate to permit unequivocal species deterrninat,on. Turing such handling permittees are sc be regarded as Agents of the State. Collection gear left unattended in the field must be properly identi.fie-d. with the exception of incidental capture as described above, this permit does not. authorize the taking of any threatened or endangered species (state or federal, list attached?. r.'f.-crdi'y S.iiuni: X’i^-moinn f>:V-^VD: &!«SV? » MC.M O’;. IS i m fM 553 * PHW E f-OI ::f~1C 0; fAX St 1 ^-7 1 2 7 • Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 92 Because most of tac proposed collection nit kg w-‘i be on the Desoto Natinr.jQ Forest, and because of the possibility of incidental capture of federally listed specie a vithir. trap arrays and/or fur dncumtrcation purposes during foot surveys of trapping sites, your collect, i one iroy require a permit from Lbs National Forest Service. The state permit s invalid until such time as you obtain the *J.S. ForesL Secvice Permit I if necessary), and it ie aLtached hereto and a copy forwarded to ths M ississippi xuaeurr. c i Natural Science (ATTH; Scientific Collection Ferm.it Ocaruttee) . This permit cr a copy of this permit must be in possession of tbs permittee Is whan o r g a n is e s a r e ta k e n o r pm;i;K:;r.;e.d u n d e r t h e c o n d itio n s o f t h e p e r m i t . A complete report of collections must be sent to the Mlnniasippi Museum of Natural Science, 2148 Riverside Dr., Jackson, MS 19232-l.wi (Attn; Scientific Collection Ferrr.it Review Committee) within 15 days of permit expiration. Collection reports should list Laxa collected, number of individuals of each, exact collection locality and d»?,e of collection. locality information muct include me county of collection, and if. is preferred that precise locality information be provided in I a t.itude/longitude (GPS) or in the township, range, a n d s e c ti o n 1TR8 ; system. If the TKS system is used, precise location within a section Bhould be indicated (e.g.; KV?4 of 5E4 of Sec 11). if ponsiblc. A single ramhr.r ssccinm of Pphisaurus mimicus (the Tiicmic glaeo liiardi nr.d •other non listed species may be retained from, each population (trapping sitoB ir. close proximity should be considered a single srte for voucher purposes) All other Bpecimens w ill be narked and re.eased alive at. t-.hr point of capture, Voucher specimens w ill be deposited at the M ississippi Museum of Natural Science. A cany of publications, survey reports, and other printed materials produced as a result of this sol 1 ect.i on shov'-d be 3ont to the Mississippi Kuaeum o£ Natural Science (Att.o: Scientific Collection Permit Review Committee;- , r'-!.r r->- ;-nf --rn;- f r -i-“x .; f on 11 M arch 2 DOS m :tm , e-ouservation hioicgy section CC Cary Bcrrv. Manager, Rr.g-.on. 2 and Pegian 2 Diane Tyrone, DcSoto National forest Mjae-iui Enc lo s u r c Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 93 APPENDIX G United States Forest National Forests 654 W. Frontage RdL, P.O. Box 243 Ip Defiartnictit of Service In Mississippi Wiggins, MS 3957*? Agriculture ...... liy^^?2:rrf'Y60{/02S-2Si0 File Code: 2670 Dale: February 16,20G5 Route To: Subject: Black Pine Snake Research T»: University o f Southern Mississippi Department o f Biological Sciences Attn; Danna Smith ! 18 College Dr. #5018 Hattiesburg, MS 39406-0001 This letter is your authorization to conduct research on the conservation biology and ecology of the Mack pine snake in Mississippi We have burned the site on February 12 and you can install the fences immediately. Please follow the guidelines in the permit from the Mississippi Museum o f Natural Science. I look forward to getting a copy of your research results. 'JUDITH HENRY District Ranger De Soto Ranger District C aring Tar the Land and Serving People Pratted 30 KeeveJrcs Pnoe* o Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 94 APPENDIX H KOf,T?L?:ri Apr'iNisT'iTivi sctettitk collect ::.;!-! It; VAt CTHCc.KN: eel is gra-Tsd v.t: Southern 1\ i to.-; losipp i.Mp v,. e f B jolagisai Sc^ncc.s MS C ciie# Drive # iai'B :Uvt.M o:.; vp MS Mi-M-f ft M :,-T,;: ;.fi» £■ , C« » QU-Sii"; P lS V i NoML, TOCOO': HC-V. r C? 7: . W i -M O O .. A S fj , .--ro M .s s ls s i.p p .f. A lth ou gh th e fm;..-; o f b ls c k p in e momo rnmo;rah w;.,., v rtixP M'-'- M-:otc; Km tonal Kntosf' In Kayne arm -Miry CoaAften. to r bo Mart on :'.'.:o.o v _•.• W.i. I-.if :: ••• IMnacovonf: Aron tr. Hsrtnn County M ono!ftonal upon rorrmoMo-n of Mot omaqor f sony Birin qe.M, ono on Loco Paoktno Ccopony land Mor-dif i.onn i open pmoum;: on to/ the landov-nerf tn loorcro Cn .snty, b!.ao> pine snakes incidents •:, v u.o nmM msmMare ray be .included wi b m ?: the etody protcool, reopee fcecw t C o :. o-cti-.-n nafhodn ;;oiy inoitdo oovof boardv, Sheroon live Irene, dvirf. Moon •■ at v.n .funnel fceye, and tend collection durino foot, survey-. lee.nor drift 'Mr:-v farmo.i. trap -mays wl.H fypiceii y oo pCaced ir. areas suitable ter for' o ;.t?.t, ony t reuse b e otvo: to avoid .MoidnnTai oepto.ro of juvenile tns'ocl o n e . f ore-- : ■• r e r o told .f-r noau-M to preves* coe.r-heet..; v.o of captives and onoulo ca ere:.!- J 0 o f ficien t.'. y rrom onb'y m t ioasf once per day, one nore cider, -e tor conn.i.:;on.- • to emut.re M e t c e p f o i e d a n i o n M do .not d e c o d e n e b y n r e r ec. o.r ev e m e a t o t:. •Mcroro; 0 ...e;:- a n d o r. ir. o r r o t on C o : s n a k e m t u a i a v-u.ll b e i r- ae"* t to o Loci vnitk m ;.;;fre.rod '; J-.t- c c:ne r e . o ' e o t p i n e e r e o o r a M I h a v e PIT f a c e t r t p i a n t o i. or:.- Pu.itad.:/ CtiOT irtrii ;-C.d'-o M M r ; f o ;v,.rru .;e 1 ;.y irtplnrced ’with, a f crepe: e ;o; r r - t.onto: : . ■•;•; O l d : ; t r atiortitr . Tvansciitter implaniration should fc© done according to acc^ptoc prococol (contact Dr. Craxg Rudolph: (S36> 56^-7981) by 3 veterinarian or an individual experienced with this procedure. Surgery-.related pmbloms should be immediately reported to Tom Mann or Dr. Bob Jones at the Mississippi Museum cf Science ({501> 354-6367, extensions 116 and 113. r e s p e c t i v e l y ) . n rs 1 ;.,v - t v v t ..:., i f p r: i: t I t i t ,. ' f:t .:.. i ;; * • f ,t -;. '. ri ". -t i / ili-1. fr^M-v t: t -•' f cay bv wc-ti t\ qt:; :;.i, ptii.ttt.r, 1 • -Vi't'tb t .: t t t i 't f tbCiv;Vti y .1 f i V'.:? " t ft'ty , Phis h :s;v:. t At. eb -P btpp.t.ttb f P5 fudv.:-;.M / bPP P t to tv;, h-': t q “ r pr'i ";. ss'.Mt:; /ty ., pppiuv- f .Atti s t’.vs.b , ": t; f e-ttrt. 11 y f pr-.at e:;t p, « tb‘.v tr. .M 'pevec yt'pf t- • t t r t - t t .' tot t r u t p.1" .’yyt.f .it's:::; , f.tt M tvrM ;y t f. r. vt tv.-:' ---.A, ?f a:.' r-tti M.".;:; v f : ft' . t / t ; .the;- t i t . , t t a thu :P a:.t tvidunyv-r.'-tc sourV;#-rr. t t y b s t :. .-MMt'tM:. ;. .'•..'. tu-bM 7 f lb b;'.ti y t.haL u:q, o f : oxc^ p t t'.ho t b t t 1.1 .. • vtto Oo t"o.:r.: e.t'-d, and .-s sf ?:-t st^d M tv t, f r-.vp.triq optirat os t- Pt-j.': o to tvs .:. at. I to oirO.-t;-o: tte 1 i to f ; f yoJ o i ,t-.a ..dvoval to:.:-;, a :. o o f f o r t o la a a . Uowa-’O;;, t aa f a aoy to fhto^ sp~o.:,po >>r- top;dootM.iy vapopr-;.o oi't’otrappiA O bear or «ACbo;t*o;roi vobilt to.ittort; ourvoyo f tot , bri-rf h -'tio q is A vtfiiitt;: tc to o.oo: v— too- : : : .t o e r t j t: •; Joqoa' o to ro-rroo aoo y ; ;. vooo ; t - t t o ia s tio f o i r:o o o ' : an . fo r io q o.oo. - -.0 M. op p o o -o .tto o b to'o t o b-; r t p r c t f a s P:;-;.a..a tf t :•<; B ta to . it-oori oy; Mv-i/ar A'f.'.'.a.vt’p’’? o i O.-P-.oi Mof * P 6 P .MbM.ob * PH0« ‘ fo POOob FM 69i ^-?52T * o.-iooono;: Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 95 Cellection ^ a r ^ eft anattended in the field j o l s be properly identified With 5.h« exr;yp-iori c-f capture of clack pir.e ftimicns *nd incident* 1 c a p tu re cjI ether listed aperies os described above, this permit. docs not authorize* the taking of *iry threatened or endangered species (.^tate or federal, lis t attached! , most of the proposed collection ail.** will be on the UsSoto Natrona} torest, and became of the possibility cf incidental capture of federally 'isted .‘rpneie? within trap arrays and/or for docur^entat-.ion purposes This permit or a copy of this permit must be in possession of the permittee (x when organisms are taken or possessed under Uv- conditions of the permit. A complete report or col' not •ons must be z>~nt to the Mississippi ^4u3eum or Natural Su.ier.cn, 2.14S Riverside Or., JotAscn, MS 3$202-1R53 (Attnt Scientific Collection iermit Review Committee) within '5 days of permit vxpiratior.. Collection reports should list Lax*. collected, number of individuals of each, exact collection locality and date ol' collection. Locality information muxt include the county of collection, and if is preferred that precis* locality information be provided ir. latitL.c.e/3unqitude (GPS) or U\ the township, tvsnqe, and s e c tio n (TR5) system - i f th e TRS system u scd r p r e c is e i:>»:a<;ion w ith in a section fthrnlc be indicated (e.g.: NV4 cf Si4 of Sec ll)j if possible. Voucher specimens of Aon-ilsl.ijd species may ne ret.*ined from each pcpnl$tior. (trapping sites in close proximity should b« considered a single site for-vcicber purp::s«r;; . All other specimens will be marked (which m^y include insertion of pit tags; and released alive at the poir.i of rapture. Vauihcr specimens will be rlep c situ d a t th e M is s is s ip p i Museum o f n a tu r a l Scior.ee. A cc-ipy of publications, survay reports, and other printed m aterials produced sk a result of this collection ahouLfi bn r,*nt to the Kissl-s^lppi Xuaeur. of Natural Science IAttns SciecLific Collection Eerrnil. Review Committee:-. 3 Haxch 200*-, a f V Ulrtcr., zcti.*;c'rvotion biciogy sc rtim <:c Gary b e rry , [O n ag er, P.r*:j F.ncl<35ure Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LITERATURE CITED Ando, A., J. Camm, S. Polasky, and A. Solow. 1998. Species distributions, land values, and efficient conservation. Science 279:2126-2128. Andrews, K. M., and J. W. Gibbons. 2005. How do highways influence snake movement? Behavioral responses to roads and vehicles. Copeia 4:772-782. Aresco, M. J. 2005. The effect of sex-specific terrestrial movements and roads on the sex ratio of freshwater turtles. Biological Conservation 123:37-44. Aresco, M. J., and C. Guyer. 1999. Burrow abandonment by gopher tortoises in slash pine plantations of the Conecuh National Forest. Journal of Wildlife Management 63:26-35. Asplund, K. K. 1974. Body size and habitat utilization in whiptail lizards (Cnemidophorus). Copeia 1974:695-703. Avise, J. C. 2000. Phylogeography: The History and Formation of Species. Harvard University Press, Cambridge, Massachusetts 464 p. . 2004. Molecular Markers, Natural History and Evolution. Sinauer Associates Inc., Sunderland, Massachusetts. 511 p. Bartram, W. 1791. Travels through North Carolina and South Carolina, Georgia, East and West Florida. Dover Publications, Inc., New York. Baskaran, L. M., V. H. Dale, R. A. Efroymson, and W. Birkhead. 2005. Habitat modeling within a regional context: An example using gopher tortoise. The American Midland Naturalist 2:335-351. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 97 Baxley, D. and C. Qualls. 2005. Conservation and ecology of the black Pinesnake ( Pituophis melanoleucus lodingi) in South Mississippi. Report to the Mississippi Department of Wildlife, Fisheries, and Parks. Jackson, Mississippi, USA. 38 pp. Bernardino, Jr., F. S., and G. H. Dalrymple. 1992. Seasonal activity and road mortality of the Pa-hay-okee wetlands of the Everglades National Park, USA. Biological Conservation 61:71-75. Bematchez, L., and C. C. Wilson. 1998. Comparative phylogeography of Nearctic and Palearctic fishes. Molecular Ecology 7:431-452. Berry, J. K. 1993. Beyond Mapping-Concepts, Algorithms, and Issues in GIS. GIS World Books, Fort Collins, Colorado. Blaustein, A.R., D. B. Wake, and W. P. Sousa. 1994. Amphibian declines: judging stability, persistence, and susceptibility of populations to local and global extinctions. Conservation Biology 8:60-71. Blouin-Demers, G., and P. J. Weatherhead. 2001. Habitat use by black rat snakes ( Elaphe obsoleta obsoleta) in fragmented forest. Ecology 82:2992-2896. ------. 2002. Habitat-specific behavioral thermoregulation by black rat snakes {Elaphe obsoleta obsoleta). Oikos 97:59-68. Bowman, J., C. V. Corkum, and G. J. Forbes. 2001. Spatial scales of trapping in small- mammal research. Canadian Field Naturalist 115:472-475. Browning, D. M., S. J. Beaupre, and L. Duncan. 2005. Using partitioned Mahalanobis D2 to formulate a GIS-based model of timber rattlesnake hibemacula. Journal of Wildlife Management 69:33-44. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 98 Burbrink, F. T., and R. Lawson. 2007. How and when did old world ratsnakes disperse into the new world? Molecular Phylogenetics and Evolution 43:173-189. Burbrink, F. T., R. L. Lawson, and J. B. Slowinski. 2000. Mitochondrial DNA phylogeography of the polytypic North American rat snake ( Elaphe obsoleta): A critique of the subspecies concept. Evolution 54:2107-2118. Burger, J., and R. T. Zappalorti. 1991. Nesting behavior of Pinesnakes ( Pituophis melanoleucus melanoleucus) in the New Jersey Pine Barrens. Journal of Herpetology 25:152-160. Burger, J., R. T. Zappalorti, M. Gochfeld, W. I. Boarman, M. Caffrey, V. Doig, S. D. Garber, B. Lauro, M. Mikovsky, C. Safina, and J. Saliva. 1988. Hibemacula and summer den sites of Pinesnakes {Pituophis melanoleucus) in the New Jersey Pine Barrens. Journal of Herpetology 22:425-433. Campbell, H. W., and S. P. Christman. 1982. Field techniques for herpetofaunal community analysis. In N. Scott, Jr. (ed.), Herpetological Communities, pp. 193- 200. U.S. Fish and Wildlife Service, Wildlife Research Report 13. Carfagno, G. L., E. J. Heske, and P. J. Weatherhead. 2006. Does mammalian prey abundance explain forest-edge use by snakes? Ecoscience 13:293-297. Carranza, S., E. N. Arnold, and J. M. Pleguezuelos. 2006. Phylogeny, biogeography, and evolution of two Mediterranean snakes, Malpolon monspessulanus and Hemorrhois hippocrepis (Squamata, Colubridae), using mtDNA sequences. Molecular Phylogenetics and Evolution 40:532-546. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Castoe, T. A., C. L. Spencer, and C. L. Parkinson. 2007. Phylogeographic structure and historical demography of the western diamondback rattlesnake ( Crotalus atrox): A perspective on North American desert biogeography. Molecular Phylogenetics and Evolution 42:193-212. Clark, J. D., J. E. Dunn, and K. G. Smith. 1993. A multivariate model of female black bear habitat use for geographic information systems. Journal of Wildlife Management 57:519-526. Clevenger, A. P., J. Wierzchowski, B. Chruszcz, and K. Gunson. 2002. GIS-generated, expert-based models for identifying wildlife habitat linkages and planning mitigation passages. Conservation Biology 16:503-514. Clibum, J. W. 1957. Behavior of a captive black Pinesnake. Herpetologica 13:66. . 1962. Further notes on the behavior of a captive black Pinesnake {Pituophis melanoleucus lodingi Blanchard). Herpetologica 18:34-37. ------. 1976. Observations of ecdysis in the black Pinesnake Pituophis melanoleucus lodingi (Reptilia, Serpentes, Colubridae). Journal of Herpetology 10:299-301. Cohen, J. E. 2003. Human population: The next half century. Science 5648:1172-1175. Conant, R., and J. T. Collins. 1991. A Field Guide to Amphibians and Reptiles of Eastern and Central North America. Houghton Mifflin Co. Boston. Conner, M. L., M. D. Smith and L. W. Burger. 2003. A comparison of distance-based and classification-based analyses of habitat use. Ecology 84:526-531. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100 Corsi, F., J. De Leeuw, and A. Skidmore. 2000. Modeling species distribution with GIS. In L. Boitani and T. K. Fuller (eds.), Research Techniques in Animal Ecology: Controversies and Consequences, pp. 389-434. Columbia University Press, New York. Corsi, F., E. Dupre, and L. Boitani. 1999. A large-scale model of wolf distribution in Italy for conservation planning. Conservation Biology 13:1-11. Coyne, J. A., and H. A. Orr. 2004. Speciation. Sinauer Associates, Sunderland, Massachusetts, USA. Crain, J. L. and J. W. Clibum. 1971. Pituophis melanoleucus lodingi from the Western part of its range. The Southwestern Naturalist 15:496-497. Crandall, K. A., O. R. P. Bininda-Emonds, G. M. Mace, and R. K. Wayne. 2000. Considering evolutionary processes in conservation biology. Trends in Ecology and Evolution 15:290-295 Dayton, G. H., and L. A. Fitzgerald. 2006. Habitat suitability models for desert amphibians. Biological Conservation 132:40-49. De Queiroz, A., R. Lawson, and J. A. Lemos-Espinal. 2002. Phylogenetic relationships of North American garter snakes ( Thamnophis ) based on four mitochondrial genes: How much DNA sequence is enough? Molecular Phylogenetics and Evolution 22:315-329. Dirzo, R., and P. H. Raven. 2003. Global state of biodiversity and loss. Annual Review of Energy and the Environment 28:137-167. Dodd, C. K. 1991. Drift fence-associated sampling bias of amphibians at a Florida sandhills temporary pond. Journal of Herpetology 25:296-301. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 101 Dodd, C. K. and R. Franz. 1993. The need for status information on common herpetofaunal species. Herpetological Review 24:47-49. . 1995. Seasonal abundance and habitat use of selected snakes trapped in xeric and mesic communities of north-central Florida. Bulletin of the Florida Museum of Natural Flistory 38:43-67. Dubuc, L. J., W. B. Krohn, and R. B. Owen Jr. 1990. Predicting occurrence of river otters by habitat on Mount Desert Island, Maine. Journal of Wildlife Management 54:594-599. Duran, C. M. 1994. Ecology of the Hispid Cotton Rat (Sigmodon hispidus ) With Comments on the Effects of Artificial Population Manipulation on an Eastern Texas Small Mammal Community. Unpubl. M.S. Thesis, Stephen F. Austin State University Nacogdoches, Texas. ,1998a. Radio-telemetric study of the Black Pinesnake {Pituophis melanoleucus lodingi) on the Camp Shelby Training Site. Final Report. Misssissippi Natural Heritage Program and the Mississippi National Guard. . 1998b. Status of the black Pinesnake (Pituophis melanoleucus lodingi). Final Report to the United States Fish and Wildlife Service . 32 pp. Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39:783-791. Fitch, H. S. 1970. Reproductive cycles of lizards and snakes. Miscellaneous Publications of the Museum of Natural History, University of Kansas 52:1-247. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 102 . 1987. Collecting and life-history techniques. In R. Seigel, J. Collins and S. Novak (eds.), Snakes: Ecology and Evolutionary Biology, pp. 143-181. McGraw- Hill, New York. Foin, T. C., S. P. D. Riley, A. L. Pawley, D. R. Ayres, T. M. Carlsen, P. J. Hodum, and P. V. Switzer. 1998. Improving recovery planning for threatened and endangered species. Bioscience 48:177-184. Franz, R. 1986. The Florida gopher frog and the Florida pinesnake as burrow associates of the gopher tortoise in northern Florida. In D. R. Jackson and R. J. Bryant (eds.), The Gopher Tortoise and Its Community, pp. 16-20. Florida State Museum, Gainesville. ------. 2001. Pituophis melanoleucus mugitus. Digging behavior. Herpetological Review 32:109. Freitag, S., C. Hobson, H. C. Briggs, and A. S. Van Jaarsveld. 1998. Testing for potential survey bias: the effect of roads, urban areas and nature reserves on a southern African mammal data set. Animal Conservation 1:119-127. Fretwell, S. D. 1972. Populations in a Seasonal Environment. Princeton University press, Princeton, New Jersey, USA. Fretwell, S. D., and H. J. Lucus Jr. 1970. On territorial behaviour and other factors influencing habitat distribution in birds. Acta Biotheoretica 19:16-36. Frost, C.C. 1993. Four centuries of changing landscape patterns in the longleaf pine ecosystem. Proceedings of the Tall Timbers Fire Ecology Conference 18:17-43. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 103 . 2006. History and future of the longleaf pine ecosystem. In S. Jose, E. Jokela, and D. Miller (eds.), The longleaf pine ecosystem, ecology, silviculture, and restoration, pp. 9-48. Springer, New York. Frost, C. C., J. Walker, and R. K. Peet. 1986. Fire-dependent savannas and prairies of the Southeast: original extent, preservation status and management problems. Pp. 348- 357 In Wilderness and natural areas in the eastern United States: a management challenge. Kulhavy, D. L., and R. N. Conner (Eds.). Center for Allied Studies, School of Forestry, Stephen F. Austin University, Nacogdoches, Texas. Gerald, G. W., M. A. Bailey, and J. N. Holmes. 2006a. Movements and activity range sizes of Northern Pinesnakes ( Pituophis melanoleucus melanoleucus) in Middle Tennessee. Journal of Herpetology 40:503-510. . 2006b. Habitat utilization of Pituophis melanoleucus melanoleucus (Northern Pinesnakes) on Arnold Air Force Base in middle Tennessee. Southeastern Naturalist 5:253-264. 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 deja vu amphibians. Bioscience 50:656-666. Gibbons, J. W., and R. D. Semlitsch. 1981. Terrestrial drift fences with pitfall traps: an effective technique for quantitative sampling of animal populations. Brimleyana 7:1-16. Glitzenstein, J. S., W. J. Platt, and D. R. Streng. 1995. Effects of fire regime and habitat on tree dynamics in North Florida longleaf pine savannahs. Ecological Monographs 4:441-476. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Goertz, J. W. 1964. The influence of habitat quality upon density of cotton rat populations. Ecological Monographs 34:359-381. Green, D. M. 2003. The ecology of extinction: population fluctuation and decline in amphibians. Biological Conservation 111:331-343. Gregory, P. T., J. M. Mccartney, and K. W. Larsen. 1987. Spatial patterns and movements. In R. Seigel, J. Collins and S. Novak (eds.). Snakes: Ecology and Evolutionary Biology, pp. 366-396. McGraw-Hill, New York. Guisan, A. and W. Thuiller. 2005. Predicting species distribution: offering more than simple habitat models. Ecology Letters 8:993-1009. Guyer, C., and M. A. Bailey. 1993 Amphibians and reptiles of longleaf pine communities. In S. Hermann (ed.), The longleaf pine ecosystem: ecology, restoration and management, pp. 130-158. Tall Timbers Research Station, Tallahassee, Florida. Hardy, D. L., and H. W. Greene. 1999. Surgery on rattlesnakes in the field for implantation of transmitters. Sonoran Herpetologist 12:26-28. Heard, G. W., D. Black, and P. Robertson. 2004. Habitat use by the inland carpet python (Morelia spilota metcalfev. Pythonidae): Seasonal relationships with habitat structure and prey distribution in a rural landscape. Austral Ecology 29:446-460. Hels, T., and E. Buchwald. 2001. The effect of road kills on amphibian populations. Biological Conservation 99:331-340. Hewitt, G. M. 1999. Post-glacial re-colonization of European biota. Biological Journal of the Linnean Society 68:87-112. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hey, J., R. S. Waples, M. L. Arnold, R. K. Butlin, and R. G. Harrison. 2003. Understanding and confronting species uncertainty in biology and conservation. Trends in Ecology and Evolution 18:597. Himes, J. G. 2000. Burrowing ecology of the rare and elusive Louisiana Pinesnake, Pituophis ruthveni (Serpentes: Colubridae). Amphibia-Reptilia 22:91-101. Himes, J. G. and L. M. Hardy. 2002. Movement patterns and habitat selection by native and repatriated Louisiana Pinesnakes (Pituophis ruthveni): implications for conservation. Herpetological Natural History 9:103-116. Himes, J. G., L. M. Hardy, D. C. Rudolph, and S. J. Burgdorf. 2002. Growth rates and mortality of the Louisiana Pinesnake. Journal of Herpetology 36:683-687. Homer, C., C. Huang, L. Yang, B. Wylie, and M. Coan. 2004. Development of a 2001 national land cover database for the United States. Photogrammetric Engineering and Remote Sensing 70:829-840. Hopkins, C. D. 1977. Methods f or generating land suitability maps: a comparative evaluation. Journal of the American Institute of Planners 43:386-400. Huey, R. B., C. R. Peterson, S. J. Arnold, and W. P. Porter. 1989. Hot rocks and not-so- hot rocks: retreat site selection by garter snakes and its thermal consequences. Ecology 70:931-944. Jaarola, M., and H. Tegelstrom. 1995. Colonization history of north European field voles (Microtus agrstis) revealed by mitochondrial DNA. Molecular Ecology 4:299- 310. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 106 Jennings, S. B., N. D. Brown, and D. Sheil. 1999. Assessing forest canopies and understory illumination: canopy closure, canopy cover and other measures. Forestry 72:59-74. JMP Version 5.1. SAS Institute Inc., Cary, North Carolina, USA. Johnson, D. 1980. The comparison of usege and availability measurements for evaluating resource preference. Ecology 61:65-71. Jones, C. G., J. H. Lawton, and M. Shachak. 1994. Organisms as ecosystem engineers. Oikos 69:373-386. Katnik, D. D., and R. B. Wielgus. 2005. Landscape proportions versus monte carlo simulated home ranges for estimating habitat availability. Journal of Wildlife Management 69:20-32. Kery, M. 2002. Inferring the absence of a species- a case study of snakes. Journal of Wildlife Management 66:330-338. Kincaid, B. W. and G. N. Cameron. 1982. Effects of species removal on resource utilization in a Texas rodent community. Journal of Mammology 63:229-235. Knick, S. T. and D. L. Dyer. 1997. Distribution of black-tailed jackrabbit habitat determined by GIS in southwestern Idaho. Journal of Wildlife Management 6:75- 85. Komarek, E. V., 1965. Effects of fire on temperate forests and related ecosystems: Southeastern United States. Pp. 251-277 In Fire and Ecosystems. Kozlowski, T. T. and E. E. Ahlgren (Eds.). Academic Press, New York, New York, USA. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 107 Kumazawa, Y., H. Ota, M. Nishida, and T. Ozawa. 1996. Gene rearrangements in snake mitochondrial genomes: Highly concerted evolution of control-region-like sequences duplicated and inserted into a tRNA gene cluster. Molecular Biology and Evolution 13:1242-1254. Landers, J. L., D. H. van Lear, and W. D. Boyer. 1995. The longleaf pine forests of the southeast: requiem or renaissance? Journal of Forestry 93:39-44. Legge, J. T., R. Roush, R. Desalle, A. P. Vogler, and B. May. 1996. Criteria for establishing evolutionarily significant units in Cryan’s Buckmoth. Conservation Biology 10:85-98. Lindenmayer, D. B., J. Fischer, A. Felton, R. Montague-Drake, A. D. Manning, D. Simberloff, K. Youngentob, D. Saunders, D. Wilson, A. M. Fleton, C. Blackmore, A. Lowe, S. Bond, N. Munro, and C. P. Elliott. 2007. The complementarity of single-species and ecosystem-oriented research in conservation research. Oikos 116:1220-1226. Lipps, G. 2005. A framework for predicting the occurrence of rare amphibians: A case study with the green salamander. M.S. Thesis. Bowling Green State University, Bowling Green, Ohio. Litt, A. R., L. Provencher, G. W. Tanner, and R. Franz. 2001. Herpetofaunal responses to restoration treatments of longleaf pine sandhills in Florida. Restoration Ecology 9:462-474. Madsen, T., and R. Shine. 1996. Seasonal migration of predators and prey: Pythons and rats in tropical Australia. Ecology 77:149-156. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 108 Malczewski, J. 1996. A GIS-based approach to multiple criteria group decision making. International Journal of Geographic Information Systems 8:955-971. Manly, B.F.J. 1991. Randomization and Monte Carlo Methods in Biology. Chapman and Hall, New York, New York, USA. Marsh, D. M. and P. C. Trenham. 2001. Metapopulation dynamics and amphibian conservation. Conservation Biology 15:40-49. Mayden, R. L. 1997. A hierarchy of species concepts: the denouement in the saga of the species problem. In M. F. Claridge (ed.), Species: the units of biodiversity, pp. 381-424, Chapman and Hall, London. McCoy, E. D., and H. R. Mushinsky. 1992. Studying a species in decline: Changes in populations of the gopher tortoise on federal lands in Florida. Florida Scientist 55:116-124. McCullough, D. R., editor. 1996. Metapopoulations and Wildlife Conservation. Island Press, Washington, D. C., U.S.A. McDearman, W. 2005. Gopher Tortoise (Gopherus polyphemus) soil classification for the federally listed range. U.S. Fish and Wildlife Service, Jackson, Mississippi. Means, D. B. 2006. Vertebrate faunal diversity of longleaf pine ecosystems. In S. Jose, E. Jokela, and D. Miller (eds.), The longleaf pine ecosystem, ecology silviculture, and restoration, pp. 157-213. Springer, New York. Means, D. B., C. K. Dodd, S. A. Johnson, and J. G. Palis. 2004. Amphibians and fire in longleaf pine ecosystems: Response to Schurbon and Fauth. Conservation Biology 4:1149-1153. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Means, D. B., J. G. Palis, and M. Baggert. 1996. Effects of slash pine silviculture on a Florida population of flatwoods salamander. Conservation Biology 10:426-437. Meshaka, W. E., and J. N. Layne. 2002. Herpetofauna of a long-unbumed sandhill habitat in South-Central Florida. Florida Scientist 65:35-50 Miller, J. H. and K. V. Miller. 1999. Forest Plants of the Southeast and Their Wildlife Uses. Southern Weed Science Society, Champaign, Illinois. Miller, R. G., Jr. 1981. Simultaneous Statistical Inference. McGraw Hill, N.Y. Mushinsky, H. R., and E. D. McCoy. 1994. Comparison of gopher tortoise populations on islands and on the mainland in Florida. Pps. 37-49 In Biology of North American tortoises. Bury, R. B. and D. J. Germano (Eds.). U.S. National Biological Survey, Fish and Wildlife Research 13, Washington D.C. USA. Noel, J. M., W. J. Platt, and E. B. Moser. 1998. Structural characteristics of old- and second-growth stands of longleaf pine (Pinus palustris) in the Gulf Coastal region of the U.S.A. Conservation Biology 12:533-548. Noss, R. F., M. A. O’Connell, and D. D. Murphy. 1997. The Science of Conservation Planning. Island Press: Washington D.C., U. S. A. O’Brien, S. J. and E. Mayr. 1991. Bureaucratic mischief- recognizing endangered species and subspecies. Science 251:1187-1188. Olsen, M., and J. Lee. 2007. Pituophis melanoleucus lodingi. Tree Climbing. Herpetological Review 38:210. Outcalt, K. W. 1997. Status of the longleaf pine forests of the West Gulf Coastal Plain. Texas Journal of Science 49:5-12. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 110 Palumbi, S. R. 1996. Nucleic acids II: the polymerase chain reaction. In D. M. Hillis, C. Moritz, and B. K. Mable (eds.), Molecular Systematics, pp. 205-320, Sinauer, Sunderland, Massachusetts. Pearson, D. E. and L. F. Ruggiero. 2003. Transect versus grid trapping arrangements for sampling small-mammal communities. Wildlife Society Bulletin 31:454-459. 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 problem of separating human impacts from natural fluctuations. Science 23:892-895. Posada, D., and K. A. Crandall. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14:817-818. Reddy, S., and L. M. Davalos. 2003. Geographic sampling bias and its Implications for conservation priorities in Africa. Journal of Biogeography 30:1719-1727. Reichling, S. B. 1995. The taxonomic status of the Louisiana Pinesnake, Pituophis melanoleucus ruthveni (Serpentes: Colubridae). Journal of Herpetology 29: 186-198. Reinert, H. K., and D. Cundall. 1982. An improved surgical implantation method for radio-tracking snakes. Copeia 3:702-705. Richter, S. C. and R. A. Seigel. 2002. Annual variation in the population ecology of the endangered gopher frog, Rana sevosa Goin and Netting. Copeia 4:962-972. Rodda, G. 1993. Where’s Waldo (and the snakes)? Herpetological Review 24:44-45. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 111 Rodriguez-Robles, J. A., and J. M. De Jesus-Escobar. 1999. Molecular systematics of New World lampropeltinine snakes (Colubridae): Implications for biogeography and evolution of food habits. Biological Journal of the Linnean Society 68:355- 385. ______. 2000. Molecular systematics of new world gopher, bull, and pinesnakes {Pituophis. Colubridae), a transcontinental species complex. Molecular Phylogenetics and Evolution 14:35-50. Rogers, A. R., and H. Harpending. 1992. Population growth makes waves in the distribution of pairwise genetic differences. Molecular Biology and Evolution 9:552-569. Ronquist, F., and J. P. Huelsenbeck. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 12:1572-1574. Roux, D. J., K. H. Rogers, H. C. Biggs, P. J. Ashton and A. Sergeant. 2006. Bridging the science-management divide: moving from unidirectional knowledge transfer to knowledge interfacing and sharing. Ecology and Society 11(1): 4. Rouse, Jeremy. 2006. Spatial ecology of Sistrurus catenatus catenatus and Heterodon platirhinos in a rock-barren landscape. M.S. thesis. University of Guelph, Canada. Row, J. R., and G. Blouin-Demers. 2006. Kernels are not accurate estimators of home- range size for herpetofauna. Copeia: 797-802. ______. 2006b. Thermal quality influences effectiveness of thermoregulation, habitat use, and behavior in milk snakes. Oecologia 148:1-11. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rudolph, D. C., S. J. Burgdorf, R. N. Conner, C. S. Collins, D. Saenz, R. R. Schaefer, T. Trees, C. M. Duran, M. Ealy, and J. G. Himes. 2002. Prey handling and diet of Louisiana Pinesnakes {Pituophis ruthveni) and black Pinesnakes {Pituophis melanoleucus lodingi), with comparisons to other selected colubrid snakes. Herpetological Natural History 9:57-62. Rudolph, D. C., Burgdorf, S. J., Schaefer, R. R, Conner, R. N., and Zappalorti, R. T. 1998. Snake mortality associated with late season radio-transmitter implantation. Herpetological Review 29:155-156. Russell, K. R., D. C. Guynn, and H. G. Hanlin. 2002. Importance of small isolated wetlands for herpetofaunal diversity in managed, young growth forests in the Coastal Plain of South Carolina. Forest Ecology and Management 163:43-59. Sala, O. E., F. S. Chapin, J. J. Armesto, E. Berlow, J. Bloomfield, J. Dirzo, E. Huber- Sanwald, L. F. Huenneke, R. B. Jackson, A. Kinzig, R. Leemans, D. M. Lodge, H. A. Mooney, M. Oesterheld, N. L. Poff, M. T. Sykes, B. H. Walker, M. Walker, and D. H. Wall. 2000. Global biodiversity scenarios for the year 2100. Science 287:1770-1774. Schemske, D. W., B. C. Husband, H. M. Ruckelshaus, C. Goodwillie, M. Parker, and J. G G. Bishop. 1994. Evaluating approaches to the conservation of rare and endangered plants. Ecology 75:584-606. Schneider, S., D. Roessli, and L. Excoffier. 2000. ARLEQUIN. A software for population genetics data analysis, version 2.0. Genetics and Biometry Laboratory, Department of Anthropology, University of Geneva, Geneva. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 113 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. Seone, J., J. Bustamante, and R. Diaz-Delgado. 2005. Effect of Expert Opinion on the Predictive Ability of Environmental Models of Bird Distribution. Conservation Biology 19:512-522. Sheldrick, B.H. and C. Wang. 1993. Particle size distribution. Pp. 499-557 In Soil Sampling and Methods of Analysis. Carter, M. R. (Ed.). Lewis Publishers, Ann Arbor, Michigan, USA. Shine, R., and M. Fitzgerald. 1996. Large snakes in a mosaic rural landscape: the ecology of carpet pythons Morelia spilota (serpents: pythonidae) in coastal eastern Australia. Biological Conservation 76:113-122. Smith, D. and C. Qualls. 2004. Herpetofaunal survey of longleaf pine forests in South Mississippi to determine the status and distribution of the mimic glass lizard and black Pinesnake. Report to the Mississippi Department of Wildlife, Fisheries, and Parks. Jackson, Mississippi, USA. 46 pp. Smith, L. L., and C. K. Dodd, Jr. 2003. Wildlife mortality on U.S. Highway 441 across Paynes Prairie, Alachua County, Florida. Florida Scientist 66:128-140. Smith, L. L., D. A. Steen, J. M. Stober, M. C. Freeman, S. W. Golladay, L. M. Conner and J.Cochrane. 2006. The vertebrate fauna of Ichauway, Baker County, Georgia. Southeastern Naturalist 4:599-620. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 114 Standora, M. 2002. Landscape level GIS modeling of Eastern Massasauga Rattlesnake (,Sistrurus catenatus catenatus) habitat in Michigan. M.S. Thesis. Indiana-Purdue University, Ft. Wayne, Indiana. Steen, D. A., M. J. Aresco, S. G. Beike, B. W. Compton, E. P. Condon, C. K. Dodd, H. Forrester, J. W. Gibbons, J. L. Greene, G. Johnson, T. A. Langden, M. J. Oldham, D. N. Oxier, R. A. Saumere, F. W. Schueler, J. M. Sleeman, L. L. Smith, J. K. Tucker, and J. P. Gibbs. 2006. Relative vulnerability of female turtles to road mortality. Animal Conservation 9:269-273. Swofford, D. L. 2002. PAUP* 4.0: Phylogenetic Analysis Using Parsimony (*and other methods). Version 4.b.l0. Sinauer Associates, Sunderland, Massachusetts. Taberlet, P., L. Fumagalli, A. Wust— Saucy, and J. Cosson. 1998. Comparative phylogeography and postglacial colonization routes in Europe. Molecular Ecology 7:453-464. Tear, T. H., J. M. Scott, P. H. Hayward, and B. Griffith. 1993. Status and prospects for success of the Endangered Species Act: a look at recovery plans. Science 262:976-977. ______. 1995. Recovery plans and the endangered species act: are criticisms supported by data? Conservation Biology 9:182-195. Ter Braak, C. J. F. 1986. Cononical correspondence analysis: a new eigenvector technique for multivariate direct gradient analysis. Ecology 67:1167-1179. Tomlin, C. D. 1990. Geographic Information Sytems and Cartographic Modeling. Prentice Hall College Division, Englewood Cliffs, New Jersey. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Tuberville, T. D., J. R. Bodie, J. B. Jensen, L. LaClaire, and J. W. Gibbons. 2000. Apparent decline of the southern hog-nosed snake, (Heterodon simus). Journal of the Elisha Mitchell Scienteific Society 116:19-40. Tuberville, T.D., J.D. Wilson, M.E. Dorcas, and J.W. Gibbons. 2005. Herpetofaunal species richness of southeastern national parks. Southeastern Naturalist 4:537- 569. Varner, J. M. and J. S. Kush. 2004. Remnant old-growth longleaf pine (Pinus palustris Mill.) savannas and forests of the southeastern USA: Status and threats. Natural Areas Journal 24:141-149. Vignieri, S. N., E. M. Hallerman, B. J. Bergstrom, D. J. Hafner, A. P. Martin, P. Devers, P. Grobler, and N. Hitt. 2006. Mistaken view of taxonomic validity undermines conservation of an evolutionarily distinct mouse: a response to Ramey et al. (2005). Animal Conservation 9:237-243. Vitousek, P. M., H. A. Mooney, J. Lubchenco, and J. M. Melillo. 1997. Human domination of Earth’s ecosystems. Science 277:494-499. 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. Waldrop, T. A., D. L. White, and S. M. Jones. 1992. Fire regimes for pine-grassland communities in the southeastern United States. Forest Ecology and Management 47:195-210. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ware, S., C. Frost, and P. Doerr. 1993. Southern mixed hardwood forest: The former longleaf pine forest. In W. Martin, S. Boyce, and A. Echtemacht (eds.), Biodiversity of the Southeastern United States, pp. 447-493. John Wiley & Sons, New York. Wilson, D. S., H. R. Mushinsky, and R. A. Fischer. 1997. Species profile: Gopher tortoise (Gopherus polyphemus) on military installations in the southeastern United States. U.S. Army Corps of Engineers Technical Report (SERDP-97-10). 36 pp. Woodroffe, R., and J. R. Ginsberg. 1998. Edge effects and the extinction of populations inside protected areas. Science 280:2126-2128. Wright, A. H., and Wright, A. A. 1957. Handbook of snakes of the United States and Canada, Vol. II, Comstock Publishing Associates, Ithaca, New York. Yale, A. L., M. E. Dorcas and J. W. Gibbons. 1999. Herpetofaunal diversity in coastal Plain communities of South Carolina. The Journal of the Elisha Mitchell Scientific Society 115:270-280. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.