1 SPATIAL ECOLOGY OF THE GOPHER TORTOISE ( Gopherus polyphemus ) IN COASTAL SAND DUNE HABITAT: BURROW SITE SELECTION, HOME RANGE AND SEASONAL ACTIVITY PATTERNS By ANTHONY YIN KUN LAU A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2011
2 2011 Anthony Yin Kun La u
3 To the tortoises
4 ACKNOWLEDGMENTS This study would not be possible without the support of a large group of people. I am indebted to my advisors Ken Dodd and Ray Carthy for guidance, mentorship, and constructive and helpful criticism throughout this study. I also thank my two remaining committee members Jerry Johnston and Kenne th Krysko for their invaluable input and support. I thank Joan Berish with the Florida Fish and Wildlife Conservation Commission (FWC) for inspiring my interest in sand dune o sharing her expertise with me through interesting discussions, and lendin g equipment for trapping tortoises I am grateful to the Gopher Tortoise Council for providing financial support for this study. I thank the staff at Guana Tolomato Matanzas National Estuarine R esearch Reserve (GTMNERR) particularly Jo s e ph Burgess, Emily Montgomery, Randy Altman, and Scott Eastman for providing access to study sites GIS and tortoise data, and logistical support throughout the study period. I am indebted to the many volunteers that have helped me in the field under the scorching Florida sun, particularly Alice Chow, Stephen Harris, Robert Lara, David Markus, Ryo Sakurai, Irina Skinner, Chad Teal, Nick Tolopka, and Travis Thomas. Special thank s go es to Cody Godwin and Eric Suarez, who helped me track tortoises when I was unable to do so. I am thankful for fellow Wildlife Ecology and Conservation ( WEC ) graduate students and building 1 5 0 office mates particularly Willandia Didier Chaves, Jackson Frechette Santiago Espinosa, Fang Yuan Hua Gloria Lentijo, Mauricio Nunez Regueiro, Eduardo Andres Silva Rodriguez Jose Soto and Marianela Velilla for
5 friendship and statistical assistance. I thank Jason Fidorra Theresa Walters and Na talie Williams for providing helpful comments on my proposal and earlier drafts of this thesis. I thank the faculty and staff of the Department of Wildlife Ecology and Conservation, who provided a great learning environment and helped me throughout the process of obtaini ng funding, meeting deadli nes, and staying out of trouble I thank Akiko Arai and my dog Piper for their companionship and mental support throughout this study. Last but not least I am thankful to my family for allowing me to follow my passions and pursue a career in t ortoise hunting and cooter (turtle) chasing This research was conducted under FWC permit LSSC 10 0005 and University of Florida Animal Research Committee permit 015 09WEC.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 ABSTRACT ................................ ................................ ................................ ................... 10 C H APTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 Habitat Selection ................................ ................................ ................................ ..... 12 Hierarchy in Habitat Selection ................................ ................................ .......... 13 Habitat Selection in Reptiles ................................ ................................ ............. 14 Fossorial Reptiles ................................ ................................ ................................ ... 16 Status, Ecology, and Conservation of Gopher Tortoises ................................ ........ 17 Project Objectives ................................ ................................ ................................ ... 19 Description of the Habitat and Study Site ................................ ............................... 20 2 FACTORS INFLUENCING BURROW SITE SELECTION OF GOPHER TORTOISEs IN COASTAL SAND DUNE H ABITAT ................................ ............... 32 Background ................................ ................................ ................................ ............. 32 Influences of Anthropogenic Disturbances ................................ ....................... 34 Habitat Selection of Gopher Tortoises ................................ .............................. 35 Methods ................................ ................................ ................................ .................. 37 Study Area ................................ ................................ ................................ ........ 37 Burrow Survey and Available Burrow Sites ................................ ...................... 38 Explanatory Variables ................................ ................................ ...................... 38 Data Analysis ................................ ................................ ................................ .......... 40 Burrow Site Se lection ................................ ................................ ....................... 40 Model Selection ................................ ................................ ................................ 40 Results ................................ ................................ ................................ .................... 41 Discussion ................................ ................................ ................................ .............. 42 3 HOME RANGE, SEASONAL ACTIVITY AND MOVEMENT PATTERNS OF GOPHER TORTOISE IN COASTAL SAND DUNE HABITAT ................................ 55 Background ................................ ................................ ................................ ............. 55 Methods ................................ ................................ ................................ .................. 58 Study Area ................................ ................................ ................................ ........ 58 Radio Telemetry ................................ ................................ ............................... 59 Data Analysi s ................................ ................................ ................................ .......... 60
7 Results ................................ ................................ ................................ .................... 60 Movement Distances ................................ ................................ ........................ 61 Burrow Use and Activity Patterns ................................ ................................ ..... 61 Annual Home Range ................................ ................................ ........................ 62 Dispersal Pattern ................................ ................................ .............................. 62 Discussion ................................ ................................ ................................ .............. 62 4 CONCLUSIONS ................................ ................................ ................................ ..... 71 Status of Sand Dune Tortoises ................................ ................................ ............... 71 Effects of Human Activities ................................ ................................ ..................... 73 Sea Level Rise ................................ ................................ ................................ ........ 74 Management Implications ................................ ................................ ....................... 75 Coastal Habitat Management ................................ ................................ ........... 75 Feasibility As Relocation Sites ................................ ................................ ......... 76 Public Education ................................ ................................ ............................... 77 Future Research ................................ ................................ ................................ ..... 78 APPENDIX A HABITAT CHARACTERISTICS (BURROW AND RANDOM LOCATIONS) OF ADULT GOPHER TORTOISES IN COASTAL SAND DUNES, GUANA TOLOMAT O MATANZAS NATIONAL ESTUARINE RESEARCH RESERVE, SOUTH PONTE VEDRA BEACH, FLORIDA, USA (2010 2011) ............................ 83 B PARAMETRIC CO RRELATION MATRIX FOR VARIABLES CONSIDERED IN BURROW SITE SELECTION MODEL ................. 84 LIST OF REFERENCES ................................ ................................ ............................... 85 BIOGRAPHIC AL SKETCH ................................ ................................ ............................ 94
8 LIST OF TABLES Table page 2 1 Potential factors influencing burrow site selection ................................ .............. 48 2 2 Burrow site selection models ................................ ................................ .............. 49 2 3 Relative importance of burrow site selection factors ................................ ........... 50 3 1 Mean annual home ran ge (ha) of adult Gopher Tortoises ................................ .. 66
9 LIST OF FIGURES Figure page 1 1 Map of Gopher Tortoise study site ................................ ................................ ...... 24 1 2 Aerial photographs of a portion of the northern coastal sand dune ................... 25 1 3 Representative topographic profile and land cover maps ................................ ... 26 1 4 Aerial photographs of a portion of the southern coastal sand dune (1/4) ........... 27 1 5 Aerial photographs of a portion of the southern coastal sand dune (2/4) ........... 28 1 6 Aerial photographs of a portion of the southern coastal sa nd dune (3/4) ........... 29 1 7 Aerial photographs of a portion of the southern coastal sand dune (4/4) ........... 30 1 8 A female Gopher Tortoise ( Gopherus polyphemus ) foraging ............................. 31 2 1 Probability of burrow site selection as a function of soil resistance .................... 51 2 2 Probability of burrow site selection as a function of distance to edge ................. 52 2 3 Probability of burrow site selection as a function of % herbaceous cover .......... 53 2 4 Probability of burrow site selection as a function of n o. of tortoise burrows ........ 54 3 1 A Gopher Tortoise with a radio transmitter ................................ ......................... 67 3 2 Seasonal movement patterns (mean distance traveled per move + SE) ............ 68 3 3 Seasonal movement patterns (mean number of burrows used + SE) ................ 69 3 4 Mean annual ho me range ( MCP ) + SE of adult Gopher Tortoises ..................... 70 4 1 Size class distribution of Gopher Tortoise burrows ................................ ............. 80 4 2 Monthly beach visitor counts (January 2009 December 2010) ........................ 81 4 3 The Florida coastline aft er a sea level rise of 3 meters ................................ ...... 82
10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the R equirements for the Degree of Master of Science SPATIAL ECOLOGY OF THE GOPHER TORTOISE ( Gopherus polyphemus ) IN COASTAL SAND DUNE HABITAT: BURROW SITE SELECTION, HOME RANGE AND SEASONAL ACTIVITY PATTERNS By Anthony Yin Kun Lau December 2011 Chair: C Kenneth Dodd, Jr. Cochair: Raymond Carthy Major: Wildlife Ecology and Conservation The Gopher T ortoise ( Gopherus polyphemus ) has been suggested to be a keystone species in southeastern upland sandhill ecosystems, and its habitat requirements have been wel l documented. Relatively few studies have been conducted on populations that occur in coastal sand dunes Due to their close proximately to the ocean and highly fragmented habitat, coastal populations of Gopher T ortoise s are affected by unique factors that are not observed in upland populations. Ideal Gopher T ortoise habitats in upland s consist of well drained soil, open canopy, and abundant herbaceous ground vegetation. In this s tudy, burrow site selection of Gopher T ortoises in a coastal sand dune site w as quantitative ly modeled and significant biological, environmental, and anthropogenic factors that may influence burrow site selection at fine and coarse spatial scales were identified. Land cover type, distance to edge, soil resistance, percentage herbaceous cover, slope angle, and number of tortoise burrows have significant influence s on burrow site selection probability
11 Radiotelemetry techniques were used to characterize habitat use movement and seasonal activity patterns, and area l requirement s of adult Gopher T ortoises. Twenty tortoises were fitted with radio transmitters and tracked for up to 12 months. Home range size, distances moved, number of burrows used, and duration of winter inactivity were determined and used to compare with previous studies conducted in upland habitats. Home range size was determined using minimum convex polygon (MCP) and fixed kernel methods. Sand dune Gopher Tortoises exhibited significantly smaller home ranges and used fewer number of burrows than upland tortoises Dispersal away from sand dune habitats was not observed probably because of the proximity of a heavily used highway paralleling the coast Conservation needs and management implications for sand dune Gopher Tortoise s are discussed. Coastal sand dune s may serve as short term alternative recipient site s for Gopher Tortoise s affected by human development.
12 CHAPTER 1 INTRODUCTION Habitat selection across multiple spatial and temporal scales is a central topic in wildlife ecology. Extensive research has been conducted on a variety of taxa, and t he insight gained from these studies has dramatical ly improved our understanding o f spatial and resource requirements of many species. Consequently, conservation planning for habitat preservation and restoration has bee n improved for many imperiled taxa and ecosystems. R elatively few quantitative studies have been conducted on habitat selection, as opposed to simple habitat use, of burrowing terrestrial reptiles largely because of their cryptic behavior Conservation pr iorities for these species include a better understanding of : 1) the process of habitat selection, 2) the integral roles of these species in their ecosystems and 3) anthropogenic influences on habitat selection and movement patterns. Further, there is a need to effectively communicat e an appreciation of these roles to the public and to resource managers Habitat S election The relationship between animals and their environments h as been a subject of int erest to biologists for centurie s, and the associati on is a fundamental question in ecology and conservation (Martin 1998, Groom et al. 2006). Animals, unlike plants, have the ability to actively choose their habitat. In addition to climatic conditions and availabili ty of food and breeding sites, animals fr equently are able to recognize distinct features and actively select optimal habitat ; this process is termed habitat selection Some general models have b een developed to demonstrate this process : ideal free distribution ( Fretwell and Lucas 1970) ; despotic distribution (Fretwell 1972, Van Horne
13 1983) ; ideal preemptive distribution (Pulliam a nd Danielson 1991) Habitat selection ha s considerable influences on survivorship and reproductive success of a species (Thomas and Taylor 1990 Block and Brennan 1993) Many migratory species, such as pelagic mammals and song birds, require di stinct habitats for overwintering, nursing, feeding, and mating ; whereas most pond breeding amphibians depend on both aquatic and terrestrial habitats (Semlitsch 2000) Understandi ng what habitat features are important for wildlife in different life stages will allow conservationists to protect, improve, and restore habitats. Hierarchy in Habitat S election Some of the earlier literature described habitat selection as a two staged pr ocess (Svardson 1949 Hilden 1965) In the first stage, individuals select among different environments using general features over a broad landscape scale. In the second stage, individuals respond to more subtle and specific habitat features within the se lected landscape. Habitat selection may be influenced by a wide array of factors or features such as visual habitat cues, habitat structure, bioenergetics, and the presence of competitors, predators, and prey. The complicated, multivariate nature of the ha bitat selection process has given rise to the concept of the n dimensional niche (Hutchinson 1957). H abitat selection of nonbreeding migratory birds has been suggested to be a multi stage hierarchical decision making process (Hutto 1985) Johnson (1980) pr oposed a similar scheme that consists of four orders of selection. First order selection, at the broadest scale, represents the selection of the geographic or physical ran ge of a species restricted by climatic conditions and geographic continents. Second order selection represents the home range of an animal within its geographic range. Third
14 choose s to hu nt, forage, rest, mate or nest. Fourth order selection, at the finest scale, represents specific microhabitat features (i.e. a particular perch or a log) within a particular site (i.e. feeding ground or nesting ground) determined in third order selection In vertebrates, habitat selection studies often focus on second order (i.e. home range/territory) and third order selection (i.e. nesting ground/feeding ground) because manipulation and replication at the scale in which these selection s operate is more feasible than those at larger scales The hierarchical nature of habitat selection has also been an emphasis in lan dscape ecology, where the scale of ecological processes such as habitat selection and habitat use are explicitly considered (Lima and Zolln er 1996). Since then, h abitat selection has been studied in many taxa acr oss different ecosystems at multiple spatial scales. These studies have improved understanding and knowledge of habitat requirements, species interactions, home range use and migrator y behavior of many understanding of habitat selection processes undoubtedly contributes to wildlife Habitat Selectio n in R eptile s As noted by Rouse et al. (2011), habitat selection in reptiles may be influenced by many factors, such as foraging mode, mating system and reproductive mode. Reptile s because of their ectothermic terrestrial lifestyle, may have different ha bitat se le ction process es and habitat requirement s than similar si zed endothermic bird s or mammal s Most re ptiles are incapable of regulating their body temperature (T b ) metabolically and rely on basking to increase their T b ; as such, appropriate basking locations play an
15 important role in reptile habitat selection. Burrowing is another important aspect of reptile body temperature regulation Because many lizards, snakes, and t ortoise s live in underground burrows, which serve as shelter s from predators and site s for thermoregulation the sub surface physical characters of habitats (e.g., soil friability and moisture content) become important components of habitat use Few studies have focused on fine scale habitat selection by burrowing reptiles due to the fact that many reptiles especially squamates, are fossorial or semi fossorial Further, the paucity of studies is part ially due to the difficulty of observing free living reptiles under natural conditions a nd finding their n ests, burrows or underground retreats (Burger and Gochfeld 1991 L p ez et al. 199 8 Dodd and Barichivich 2007). Habitat selection of tortoises has been studied by many authors using radiotelemetry techniques Kazmaier et al. (2001) studied Texas Tortoises ( Gopherus berlandieri ) in semiarid shrubland s and found that Texas Tortoises exhibited broad scale selectio n for home ranges and avoided areas with high (riparian) and low (old field) canopy cover. In Mediterranean semiarid shrublands, Gre ek Tortoises ( Testudo graeca ) select habitats with simplified vegetation structures (re colonization shrubland s and small non irrigated fields) at the landscape scale and areas with high and complex vegetation cover at the home range scale (Anadon et al. 2 006). Guzman and Stevenson (2008) studied Yellow footed Tortoises ( Chelonoidis denticulata ) in the Peruvian Amazonian forest and found that C. denticulata have strong preference for open canopy swamp in floodplain forest and bamboo or grass dominated vege tation in terra firma forest The results from these studies suggest that habitat selection of tortoise s is mostly influence d by canopy cover and vegetation structure.
16 The effects of intra and inter species interaction and fine scale resource distribution on habitat selection of subterranean reptiles are seldom exa mined. A more thorough understanding o f the role of these factors (temperature, species interactions, and resources distribution) on habitat selection of reptiles is needed in order to efficiently conserve these species. Fossorial R eptiles Fossorial reptiles are an important component of many ecosystems. For example, t ortoises of the genus Gopherus excavate extensive underground burrows that provide shelter for many vertebrate and invertebrate spe cies. More than three hundred species of animals have been documented using Gopher Tortoises ( G. polyphemus ) burrows (Young and Goff 1939 Jackson and Milstrey 1989, Kent et al. 1997) S ome of these animals are obligate burrow commensals found nowhere else (Mushinsky et al. 2006). Due to the importance of Gopher Tortoise burrows to the biodiversity of upland ecosystems, Gopher Tortoises are viewed as a keystone species and a n ecosystem engineer (Eisenburg 1983 Mushinsky et al. 2006). Some snakes (e.g. Masticophis spp., Pituophis spp., Crotalus spp., and Dryma r chon spp.) are apex predators in their respective ecosystems and the decline in these species may indirectly alter the pop ulation and community structure of their prey and ecosystems Additionally fossorial reptiles recycle micronutrients to the top soil layer in the process of digging and burrowing and may affect above ground vegetation commun ity structure (Abaturov 1972). Fossorial reptiles face a number of threats globally For example, Gopher T ortoise populations in the s outheastern United States have declined due to rapid development in the regio n, human harvesting for food, and habitat degradation caused by fire
17 suppression (Mushinsky et al. 2006). Consequently populations of associated burrow commensals such as the Gopher F rog ( Lithobates capito ), Florida M ouse ( Podomys floridanus ) and E astern Indigo S nake ( D. couperi ) have declined In addition, these species are generally neither charismatic nor popular among the general public making it difficult to ga rner public support for conservation projects of fossorial or semi fossorial reptiles For example, Kellert (1980) found that the general public is less willing to modify energy projects for the sake of prot ecting E astern I ndigo S nake s ( D co u peri ) com pared with charismatic animals such as P umas ( Puma concolor ) or Bald E agle s ( Haliaeetus leucocephalus ). i op Status, Ecology and Conservation of Gopher T ortoise s The Gopher T ortoise Gopherus polyphemus (Daudin 1802), is the only t estudinid that occurs in eastern North America (east of the Mississippi River) and is one of the most studied tortoise species in the world The Gopher T ortoise is so named because of its impressive burrowing ability; Gopher T ortoise burrows on average are 4.5 m in length and 2 m in depth (Hansen 1963). This long lived, modera t e sized herbivorous tortoise usually occurs in habitats with well drained, sandy soils with open canopy that support herbaceous plant growth. It ranges from extreme southeastern South Carolina, south through peninsular Florida and west through southern Georgia, Alabama, Mississippi and southeastern Louisiana (Ernst and L ovich 2009) The Gopher T ortoise is emblematic of many southeastern fire adapted ecosystems as it often thrives in well managed, freq uently burned upland ecosystems Some of the most notable places that support large populations of Gopher Tortoise s are Ocala National Forest and Merritt Island National Wildlife Refuge in Florida. In addition to upland pine forest and sandy
18 scrubs, Gopher Tortoise s also can be found in coastal sand dunes and on barrier islands off the coast of Florida and Georgia. P o pulation s of Gopher Tortoise s appear to be declining range wide mainly outside of protected areas ( Hermann et al. 2002, Smith et al. 2006). Populations are affected by a number of threats Habitat destruction and degradation, habitat fragmentation, and up per respi ra tory tract disease (URTD) have been cited as the greatest threats to Gopher Tortoise survival ( Mushinsky et al. 2006 ). In Florida, tortoise habitats are constantly being replaced by housing developments, shopping centers, pine plantations and agriculture. M ore roads are being built to support the ever growing human population. Consequently, tortoise habitats and populations are becoming more fragmented as roads act as effective barrier s to tortoise movements. In the past, tortoises have bee n slaughtered for food and some populations in northwestern Florida may have been hunted to exti rpation ( J. E. Berish, personal communication). Human predation on Gopher Tortoise s is still taking place in part s of its range despite its protected status (Mu shinsky et al. 2006, T. M. Thomas, personal communication). The Gopher Tortoise role as an ecosystem engineer and its long lifespan (up to 50 70 years) has inspired many ecologists to study their ecology Several long term studies have been conducted to determine population density and abundance size class structure, sex ratio, survival and fecundity ( Landers et al. 1980, Landers et al. 1982, Auffenb e rg and Franz 1982 Diemer 1992b ) Burrowing, grazing, and mound building by Gopher To rtoise s have been found to increase vegetative species richness
19 and promote environmental heterogeneity in an oak pine sandhi ll forest (Kaczor and Hartnett 1990). T he Gopher Tortoise is a protected species throughout its range and is listed as International Union for Conservation of Nature (IUCN) Red List of Threatened Species (IUCN 2011) On the federal level, the U.S. Fish and Wildlife Service listed populations west of Mobile and Tombigbee Rivers in Ala bama Mississippi and Louisiana ed under provisions of the Endangered Species Act of 1973, as amended Currently, populations in the eastern part of its range are being considered to be listed as (USFWS 2011) In Florida where this species has been historical ly found in all 67 counties, it is also listed as T he Florida Fish and Wildlife Conservation Commission specifically prepared a management plan (FWC 2007) to guide conservation and research efforts on this species and its habitat s in Florida. Project Objectives To date, few studies have incorporated anthropogenic factors such as human disturbances and anthropogenic habitat modifications. Further, e ven fewer have considered factors at multiple spatial scales (Row and Blouin Demers 2006) Many biological, environmental, and anthropogenic variables have been shown to affect habitat selection of wildlife, but the impact may not be detectable in all spatial scales. Thus, there is a need to integrate multiple factors and spatial scales in to hab itat selection analyses. I therefore conducted a study on the spatial ecology of one of North s in a relatively under studied habitat in order to identify biological, environmental, and anthropogenic factors that may influence Gopher Tortoise burrow sites selection at fine and coarse spatial scales I quantified the habitat
20 characteristics of occupied versus randomly selected habitats as well as the effects of a priori selected factors on burrow site selection. Finally, I u sed r adio telemetry to determine home range s movement s and seasonal activity patterns of adult Gopher Tortoises. Measures of movement and spati al dispersion (Rouse et al. 2011) allo w ed me to access the extent of current occupancy and to help evaluate this and other coasta l dune ecosystems as relocation sites for tortoises moved from lands slated for development. Description of the Habitat and Study Site The field study was conducted between 15 January 2010 and 7 May 2011 at the coastal sand dune s within Guana Tolomato Matanzas National Estuarine Research Reserve (GTMNERR, 30.126111 81.348056 ) on the northeastern coast of Florida, St. Johns County (Figure 1 1 ) The northern section of GTMNERR includes some of the only pristine, undeveloped dunes on the eastern coast of Florida. The ca. 78 ha study area (7.1 km long and 90 145 m wide) is composed mostly of beach dune a nd coastal strand habitat (FNAI 2010). Beach dune is a n herbaceous community built by a number of coastal specialist grasse s such as Sea O ats ( Uniola paniculata ) and B itter P anicgrass ( Panicum amarrum ) whose stems trap the sand grain blown off the beach and form the foundation of the dune. Other plants commonly seen in the beach dune are Railroad V ine ( Ipomoea pescapra e spp. brasiliensis ) Seaco ast M arshelder ( Iva imbricata ), Saltmeadow C ordgrass ( Spatina patens ), Beach Morning G lory ( Ipomoea imperati ), Indian B lanket ( Gaillardia pulchella ), Bull T histle ( Cirsium horridulum ) and P rickly P ears ( Opuntia spp.). Coastal strand is the woody plant comm unity l ocated further inland and directly behind the beach dune It is dominated by dense S aw P almetto ( Serenoa repens ) and dwarfed Cabbage P alm ( Sabal palmetto ) on the
21 seaward edge, and Red B ay ( Persea borbonia ), Red C edar ( Juniperus virginiana ), Y aupon Holly ( Ilex vomitoria ), and Tough B ully ( Sideroxylon tenax ) further inland. Habitat heterogeneity and vegetation community succession in this site is driven mainly by water, wind and occasional ly fire In the beach dune, plants on the foredune are constan tly exposed to salt spray and sand burial from onshore wind blowing from the Atlantic O cean, while plants on the upper dune are subjected to the same stresses in addition to occasional inundation by storm tides and destruction by strong waves. The plants i n the beach dune are ada pted to withstand these stresses. In the coastal strand, salt spray blown off the ocean maintains a low, even canopy of hardwood. The natural fire frequency in this community is unknown. However, the major vegetative component ( Saw P almetto) is highly flammable and fires typically spread rapidly in t he coastal strand when started. The study area is bordered by a high use road (Florida State Road A1A) to the west, the Atlantic Ocean to the east, and private housing developments to the north and south. It includes some of the only pristine and undeveloped dunes on the eastern coast of Florida. The beach is accessible by humans through three elevated public beach accesses, located at 0.66 km, 1.76 km, and 5.46 km north of the southern mo st point of the study site, respectively. In addition to the se three boardwalks, there are foot paths that dissect the beach du ne at various locations. These paths are believed to be created by frequent foot traffic of beachgoers before the public beach access es were made. Public access to the beach dune is restricted to protect the vegetation community of the coastal beach dune.
22 The study area can be further divid ed into two sections by width, topographic profile and vegetation community structure (Figure 1 3 ) The northern section (15.7 ha) consists of the area north of the northern most public beach access (Figure 1 2 ) whereas the southern section (62.3 ha) con sist s of the remaining area (Figure 1 4, 1 5 1 6 and 1 7 ) The overall width of the northern section (min = 124 m, max = 145 m) is wider than the southern section (min = 90 m, max = 122 m). The northern section has the same number of blow outs and unoffi cial paths (n = 4) as the southern section despite being significantly shorter The over all topographic profile in the southern section is steeper than the northern section, and the vegetative community appear s to be less heterogeneous. In addition to nat u ral disturbances, the study site has experienced large scale anthrop ogenic disturbances in the past In the early 19 th century, the Ponte Vedra Beach area, then known as Mineral C ity, w as mined for rutile and ilmenite, which contain titanium and zirconium respectively Other minerals such as zircon, aluminum silicate, and monazite have also been mined The mining lasted until the early 1930s when the state government pressured the mining company to stop mining operations because of beach destruction ( R. E. Moore, personal communication). During the late 1940s, free range cattle w ere a problem faced by the Community Associa tion of Ponte Vedra Beach, who install ed electric fence s to keep the cattle o ff of the beach. T hese activitie s may have had long lasting effects on the vegetation community structure of the sand dune, creating large areas of openings. The re are two soil types found in the study site: the Fripp series and the Beach es s eries. The Fripp series comprises deep, rapidly permeable, well drained soils that
23 formed in thick sediments near beaches. The Beaches series consists of shallow, well drained, moderately permeable soils that formed in residuum from fined grained, metamorphic sandstone (U SDA 2010) The Fripp series occur s predominantly in the coastal strand and in the upper part of the beach dune, w hereas the Beaches series is restrict ed mostly to the foredune and the adjoining beach Gopher Tortoise s are presen t at the site in varying densities Based on burrow survey s conducted by GTMNERR staff from 2005 2007 and m y self during the present study the overall density of tortoise burrows in dunes range s from 0.64 burrows /ha to 3.05 burrows /ha. B urrow density of the northern section (12.17 burrows/ha) is much higher than the southern section (0.75 burrows/ha), probably due to the varying width, topographic profile, and vegetative community structure between the two sections (Figure 1 3 ) Many other animals use tortoise burrows in the sand d une s as temporary shelter s or permanent home s ; during the course of this study, E astern C oachwhip ( Masticophis flagellum n = 3 ), Marsh R abbit ( Sylvilagus palustris n = 2 ), and Southern T oad ( Anaxyrus terrestris n = 1 ) were observed using t ortoise burrows. Footprints of Eastern W oodrat s ( Neotoma floridana ) also have been observed in front of a tortoise burrow. Di stinctive foraging trails created by Gopher Tortoise s during their feeding forays can be found near tortoise burrows (Figure 1 8 ) Gopher Tortoise s may in fact be an important stressor on fine scale heterogeneity of the herbaceous plant community in the sand dune. As one of the major herbivores in this coastal community, tortoises may also be an important seed disperser.
24 Figure 1 1. Map of Gopher Tortoise study site in Guana Tolomato Matanzas National Estuarine Research Reserve (GTMNERR) and adjacent counties. Florida, USA.
25 Figure 1 2 Aerial photograph s of a portion of the northern coastal sand dune study site for Gopher Tortoises in Guana Tolomato Matan z as National Estuarine Research Reserve, South Ponte Vedra Beach, Florida, USA.
26 Figure 1 3 Representative topographic profile and land cover map s of the northern (L) and southern (R) sections of the coastal sand dune study site for Gopher Tortoises in Guana Tolomato Matan z as National Estuarine Research Reserve, South Ponte Vedra Beach, Florida, USA.
27 Figure 1 4 Aerial photograph s of a portion of the southern coastal san d dune (1/4) study site for Gopher Tortoises in Guana Tolomato Matan z as National Estuarine Research Reserve, South Ponte Vedra Beach, Florida, USA.
28 Figure 1 5 Aerial photograph s of a portion of the southern coastal sand dune (2/4) study site for Gopher Tortoises in Guana Tolomato Matan z as National Estuarine Research Reserve, South Ponte Vedra Beach, Florida, USA.
29 Figure 1 6 Aerial photograph s of a portion of the southern coastal sand dune (3/4) study site for Gopher Tortoises in Guana Tolomato Matan z as National Estuarine Research Reserve, South Ponte Vedra Beach, Florida, USA.
30 Figure 1 7 Aerial photograph s of a portion of the southern coastal sand dune (4/4) study site for Gopher Tortoises in Guana Tolomato Matan z as National Estuarine Research Reserve, South Ponte Vedra Beach, Florida, USA.
31 Figure 1 8 A female Gopher Tortoise ( Gopherus polyphemus ) foraging at the coastal sand dune study site in Guana Tolomato Matanzas National Estuarine Research Rese rve, South Ponte Vedra Beach, Florida USA. (Photo courtesy of the author)
32 CHAPTER 2 FACTORS INFLUENCING BURROW SITE SELECTIO N OF GOPHER TORTOISE S IN COASTAL SAND DUNE HA BITAT Background Habitat selection is a fundamental concept in ecology and has direct applications to conservation. Disproportionate use of habitat and/or resources presumably improve s the fitness and survivo rship of a species (Thomas and Taylor 1990, Block and Brennan 199 3). For decades, habitat selection of wildlife has been studied extensively in numerous species across multiple spatial and temporal scales. The insights gained from these studies have dramatically i mproved our understanding of spatial and resource require ments of many species. Consequently, c onservation planning for habitat preservation and restoration has been improved for many imperiled taxa and ecosystems. R elatively few quantitative studies have been conducted on habitat selection of fossorial reptiles a vertebrate group that play s important ecological roles in their respective ecosystems A better understanding of the process of habitat selection of these animals is needed in order to design conservation plans and manage protect ed areas. Reptile s because of their ectothermic lifestyle, may have vastly different habitat se le ction process es and habitat requirement s than a similar sized endothermic bird or mammal. Most r eptiles are incapable of regulating their body temperature (T b ) metabolically and rely on basking to increase their T b The use of burrows is another important aspect of reptile body temperature regulation because it allows sp ecies to control temperature by responding to changes in burrow te mperature with depth through direct contact w ith the substrate M any lizards, snakes, and turtles live in underground burrows, which serve as shelter s from predators as well as retreats from harsh surface
33 temperatures or environmental conditions, such as drought Few studies have focused on fine scal e habitat selection in burrowing reptiles, despite the fact that many reptiles, especially squamates, are fossorial or semi fossorial This is due partially to the difficulty of observing them under natural conditions and finding their nests or burrows (Burger and Gochfeld 1991 Lopez et al. 1998 Dodd and Barichivich, 2007). Habitat selection of tortoises has been studied by many authors using radio telemetry Kazmaier et al. (2001) studied Texas Tortoises ( Gopherus berlandieri ) in semiarid s hrublands and found that Texas Tortoises exhibited broad scale selection for home ranges and avoided areas with high (riparian) and low (old field) canopy cover. In Mediterranean semiarid shrublands, Greek Tortoises ( Testudo graeca ) select habitats with si mplified vegetation structures (re colonization shrubland and small non irrigated fields) at the landscape scale and areas with high and complex vegetation cover at the hom e range scale (Anadon et al. 2006). Guzman and Stevenson (2008) studied Yellow foote d Tortoises ( Chelonoidis denticulata ) in the Peruvian Amazonian forest and found that C. denticulata have strong preference for open canopy swamp in floodplain forest and bamboo or grass dominated vegetation in terra firma forest. The results from these s tudies suggest that habitat selection of tortoises is mostly influenced by canopy cover and vegetation structures. Despite these examples the effects of intra and inters pecies interaction and fine scale resource distribution on habitat selection of subte rranean reptiles are seldom examined. A more thorough understanding on the role of these factors (temperature, species interactions, and resources distribution) on habitat selection of reptiles is needed in order to efficiently conserve these species.
34 Inf luences of Anthropogenic Disturbances Environmental and biological facto rs have significant influences on habitat selection of wildlife at multiple spatial scales. Anthropogenic factors are also considered to be influential in highly disturbed area, but s eldom have the influences of all three factors been studied simultaneously. Several studies (Burger and Gochfeld 1991, Mahan and Yahner 1996, Peters and Otis 2007, Gallant et al. 2009) have attempted to identify these factors and their relative importance. Anthropogenic disturbances could influence habitat selection in a variety of ways. Gallant et al. (2009) studied habitat selection of River O tters ( Lontra canade n sis ) under different land use regimes (protected vs. disturbed riparian habitats) and found that environmental factors are more influential than anthropogenic factors in describing habitat use. They suggested that anthropogenic disturbances indirectly affect habitat selection of river otters by causing loss of particular habitat features (e.g., beaver ponds) that are relevant to river otter habitat requirements. Peters and Otis (2007) did not detect evidence for negative effects of anthropogenic disturbances (level of boat disturbance) on roost site selection in shore birds, and noted that it is difficult to demonstrate impacts o n shorebird fitness as a result of anthropogenic disturbance. Anthropogenic disturbances may affect certain aspects of species behavi or such as social information ( Berry 1986, Citta and Lindberg 2007, Betts et al. 2008) that are difficult for investigators to detect. Many authors (Baldwin e t al. 2006, Graeter et al. 2008, Blomquist and Hunter 2009) have investigated the effects of anthr opogenic disturbances such as roads and timber management practices on amphibian habitat selection and found that moisture is the most important driver in habitat selection of amphibians at multiple spatial scales
35 and anthropogenic disturbances (e.g. clea r cuts) have different effects on habitat sele ction depending on the species. In addition to Peters and Otis (2007) other studies have investigated the effects of anthropogeni c disturbances on wildlife in coastal system s Bird et al. (2004) studied the effect of light pollution on the foraging behavior of beach mice and conclu ded that artificial lights reduce the foraging frequency and efficiency of beach mice and deserve greater consideration in conservation planning and management. Witherington and Martin (2000) studied the effects of light pollution on hatchling sea turtles and found that artificial lighting disorients hatchling sea turtles from reaching the ocean. Coastal areas worldwide are under rapid human encroachm ent and coastal ecosystems are under tremendous pressure. In addition to light pollution, wildlife in coastal ecosystems a re subjected to invasive exotic plants, feral/domestic predators, habitat degradation, loss and fragmentation, sand mining, and global climate change is inducing a rising sea level. Coastal ecosystems (marshes, mangroves, reefs and dunes) worldwide are important to human s because they provide essential ecosystem services such as storm buffering, fisheries production and enhance d water qu ality (Barbier et al. 2008). The inhabitants of these fragile ecosystems should be protected to ensure their long term persistence and proper functioning of these ecosystem services. Habitat Selection of Gopher Tortoises The Gopher Tortoise Gopherus polyp hemus is a moderately sized, foss or ial tortoise that occurs in a variety of habitats with well drained, sandy soil in southeastern North America. The species is threatened by habitat loss and fra gmentation throughout its range and the decline of this species may consequently affect many wildlife species that share its habitat.
36 The tortoise burrow is an important feature in many ecosystems where Gopher Tortoise s occur. The burrow and its associated apron are important for a tortoise because they a re used for resting, basking, courting, shelter from predators, nest deposition, and a s the center of feeding activities The Gopher Tortoise burrow is used as a shelter by many other animals, some of which (i.e. Florida M ouse, Podomys floridanus ) even mo dify it by excavat ing a separate entrance called a chimney near the main entrance to the Gopher Tortoise burrow. The tortoise burrow provide s shelter to other animals during cold, drought and fire, and some animals are obligate commensals that can only be found in tortoise burrows Thus the distribution and survivorship of these burrow commensals is directly dependent upon the habitat modifications made by the Gopher Tortoise The Gopher Tortoise occur s in habitats with well drained sandy soils and abunda nt low growing herbaceous cover, but the exact habitat features and mechanism s that trigger a tortoise to excavate a burrow are poorly understood. Hatchling and young tortoises use active or inactive adult burrows near the nest from which they hatched ( Smi th 1992, J. E Berish, personal communication), al though some also constru ct small burrows (Butler et al. 19 95, Pike 200 6) Adult tortoises exhibit high burrow fidelity and may use the same burrows over multiple years. Since tortoises spend most of their time in a burrow, previous studies on habitat selection of Gopher Tortoise s have focused on the distribution, density, ratio of active to inactive, and habitat characteristics of the burrows ( Boglioli et al. 200 0 Jones and Dorr 2004, Baskaran et al. 2006) A better understanding of the burrow site selection process will
37 allow wildlife managers to protect and improve high quality areas that will benefit tortoises. In this study, I used a quantitative modeling approach to identify factors that influence b urrow site selection of Gopher Tortoise s, based on a set of a prior i hypotheses that incorporate d various combinations of environmental, biological, and anthropogenic factors measured at microhabitat and home range spatial scales. The result s of this study can help wildlife manager s identify critical areas in which tortoises are li kely to settle and to focus conservation effort s on these areas. Methods Study Area This study was conducted in Guana Tolomato Matanzas National Estuarine Research Reserve (GTMNE RR ; 30.126111, 81.348056 ) located on the northeastern coast of Florida, St. Johns County. The ca. 78 ha study area (7.1 km long ; 90 145 m wide) is composed of mostly beach dune and coastal strand habitat (FNAI, 2010) Habitat heterogeneity and vegetation community succession in this site is driven mainly by wind. The study site is bordered by a high use road (Florida State Road A1A) to the West, the Atlantic Ocean to the e ast, and private residences to the n orth and s ou th It includes some of the only pristine and undeveloped dunes on the eastern coast of Florida. The beach is accessible by human s through three boardwalks, located at 0.66 km, 1.76 km, and 5.46 km north of the southe rn most point of the study site. In ad dition to these three boardwal ks, there are numerous foot paths that dissect the beach du ne at various locations. These paths are believed to have be en created by frequent foot traffic of beachgoers before the public beach access es were made. Public access to the
38 beach dune is restricted in order to protect the vegetation commu nity of the coastal beach dune. Burrow Survey and Available Burrow Sites B urrow surveys were conducted from January to May 2010. The entire coastal sand dune was searched systematically to locate and categorize Gopher Tortoise burrow s following methods described by Diemer ( 1992 b ). The latitude longitu d e location of each burrow was recorded ( 3 m) using a handheld GPS unit ( Garmin GPSMAP 60 CSx; Garmin International Inc, Ola the, K ansas ). Each burrow was classif ied as active, inactive, or abandoned, according to criteria described by Auffenberg and Franz (1982). Of 238 burrows located, 100 active burrows were randomly selected using a random number generator to determine facto rs influencing burrow site selection. For comparison with randomly selected burrow sites, an equal number of random points (n = 100) within the study site were generated as available habitat using the Geospatial Modelling Environment extension in ArcGIS (Spatial Ecology LLC., 2010). Explanatory Variables To determine factors that may influence burrow site selection of Gopher Tortoise s three groups of explanatory variables ( environmental factors, biological factors, and anthropogenic factors) were measured in t wo spatial scales : m icro habitat scale (1 m radius) and h ome range scale (20 m radius) from the burrow opening The 1 m radius is assumed to be an appr opriate scale for microhabitat habitat selection because tortoises are thought to have good v ision. The 20 m radius is assumed to be an appropriate scale for home range scale selection because it is the radius of the mean annual home range s of Gopher Tortoise s in the sand dune (Lau, unpublished data; Chapter 3 of this
39 thesis). Measurements were ma de at both burrow and random loc ations from May to August 2010. Five environmental factors were measured : s oil resistance, elevation, slope and s lope angle were measured at the microhabitat scale. Soil type was measured at the home range scale. Soil resistance was measured to the closest kg per cm 2 using a handheld pocket soil penetrometer ( Geotest Instrument E280; Geotest Instrument Corp. Evanston, Illinois ) with a 2.54 cm diameter adapter foot. Elevation was measured to the closest f t (0.3048 m) us ing a LI DAR generated 1 foot contour map in a GIS software (ESRI ArcMap 9.3 Redland, California ) Slope was measured as percentage change in elevation within 1 m using a meter stick and a level. Slope angle is the orientation of the respective slope and w as measured to the closest degree using a compass. Soil type is defined as the dominant soil type of each location, following the USDA Soil Survey Geographic Database classifications ( USDA 2010). Nine biological factors were measured : c anopy cover, % bare ground, % herbaceous cover, % grass, % scrub and vines and % litter were measured at the microhabitat scale. Distance to edge, land cover type, and number of tortoise burrows were measured at the home range scale. Canopy cover i s define d as the presence or absence of canopy forming plants (> 2 m in height) at each location. Percentage bare ground, % herbaceous cover, % grass, % scrub and vines and % litter coverage were visually quantified to the closest % in a 1 m x 1 m quadrat at each location. At burrow locations, the soil deposit (apron ) excavated by the tortoise during its burrowing activity was not included in the 1 m 2 quadrat. Distance to edge is defined as the shortest distance in meter s from each location to a different land cove r type (i.e. between
40 coastal scrub and sand dune habitat) following Williams et al. (2002) Distance to edge was measured on aerial images of the study site using GIS software. Land cover type is defined as the dominant land cover type of each location following the Florida Natural Area Inventory classifications (FNAI 2010) Number of tortoise burrows is the count of Gopher Tortoise burrows (adults and subadults) located within a 20 m radius of each location. Two anthropogenic factors were measured. Dist ance to road is the perpendicula r distance of each location to the closest road. Distance to beach access is the perpendicular distance of each location to the closest beach access. Both distances were measured in meter s on high resolution aerial images of the study site using GIS software. Data Analysis Burrow Site S election To identify factors that influence burrow site selection, I used logistic regression models to predict the probability of burrow site selection as a function of factors measured in both spatial scales. The response variable is whether or not a habitat was selected (burrow locations) or available (random locations). I used the generalized linear model with binomial link function in SAS (Statistical Applications Systems, Cary, N orth C arolina ) to estimate regression coefficients with the logistic models. Model S election To address potential multi parametric correlation matrix was created to identify factors that we re strongly correlated (r > |0.6|). Factors that we re strongly correlated were excluded from the final models Distance to edge is strongly correlated with both elevation and distance to road. Distance to road and
41 elevation were both dropped from the models because the road is parallel to the elevation gradient and these two variables provide d essentially the same informat ion as distance to edge. Percentage cover of woody scrubs and vines is strongly correlated with % coverage of bare ground. The latter was dropped because it is essentially the reciprocal of the former. Nine logistic models containing different combinations of factors were created : 8 based on a prior i hypotheses and one global model that include d all factors. The best approximating model was selected Information Criterion corrected for small sample sizes (AICc) (Burnham and Anderson 2002 Anderson 2008 ) Wald Chi square test s were used to test for significance of factors A factor wa s considered to be significant if p < 0.05. The relative importance of factors was determined by summing Akaike weight of the models containing said factors. The factor with the highest summed Akaike weight wa s considered to be the most important (Burnham and Anderson 2001) Results Gopher Tortoise burrow site selection is influenced by both environmental and biological factors at microhabitat and home range spatial scales. The environmental + biological model was the best and most parsimonious model and t he global model was within 2 AICc unit s and therefore considered to be a statistica l equivalent of the best model (T able 2 2) The two best model s contain all the same factors except distance to beach access (an anthropogenic factor) and have a combined model weight of 0.997 Six of twelve factors are indicating they are good predi ctors for burrow site selection (Table 2 3). Of the continuous variables, soil resistance, distance to edge, percent herbaceous cover, and number of tortoises burrows are statistically significant. While holding all other factors constant, burrow site
42 sele ction probability decreases as soil resistance increase s 0.93 0.35, 2 = 7.05, p = 0.0079) (Figure 2 1). Burrow site selection probability also decreases as 0.07 0.02, Wald 2 = 7.71, p = 0.0055) (Figure 2 2). In contrast, burrow site selection probability increases as percentage herbaceous cover 2 = 5.09 p = 0.0241 ), while other factors are being held at constant (Figure 2 3). Finally, burrow site selection probability increases as 2 = 14.64, p = 0.0001) (Figure 2 4). Land cover type and distance to edge were the most important factors in determining burrow site selection followed by percent grass cover, percent woody scrub and vine cover, percent herbaceous cover, percent litter cover, and presence or absence of canopy cover (Table 2 3). Land cover type was the only significant categorical variable. Burrow site selection probability is higher for coastal sand dune habitat than for coastal strand Discussion Gopher Tortoise burrow site selection in a coastal sand dune site was quantitatively modeled with a combination of environmental biological and anthropogenic factors at microhabi tat and home range spatial scales. Out of the eight a priori model s best fitting and the second most complex model c ontaining all but one of the explanatory variables ( factors). The high number of factors in the best fitting model indicated that most factors considered have at least some effect s on the burrow site selection of Gopher Tortoise s and that the selection process itself is complex and involves multiple spatial scales. I n effect burrow site selection of Gopher Tortoise s in coastal sand dune was related to land cover type,
43 soil resistance, percentage herbaceous ground cover, distance to edge, number of tortoise burrow s near by, and the angle of the slope. Land cover type a home range scale factor, is the most important and significant factor indicated by the combined model weight and Wald Chi square value, respectively This is not surprising as there are only two available land cover types (beach dune and coastal strand) and the majority of Gopher Tortoise burrows were located in the beach dune. The beach dune land cover type has an open canopy and well drained soil which promote s the growth of herbaceous plants and allows for sunny nesting sites. Previous studies ( Boglioli et al. 2000, Jones and Dorr 2004).suggest that canopy closure is one of the main factors negatively influencing the distribution of tortoise burrows The coastal strand, due to canopy closure, provides little forage for Gopher Tortoise s (combined mean percent cover for grass and herbaceous plants = 7.29%), and few burrows (4%) are found within canopy closure Most burrows in the coastal strand are locate d very close to the edge (mean distance to edge = 1.5 m). Gopher Tortoise s and their burrows are rarely observed in the coastal strand partially due to the low relative detectability of burrows in that habitat. Most of the coastal strand is inaccessible to humans, and some tortoises may use it as a shelter from potential predators (i.e humans). Whe n approached, tortoises would escape towards the coastal strand when a nearby burrow is absent. Distance to edge is another significant home range scale factor that influence s burrow site selection Gopher Tortoise s are more likely to dig burrows near the edge between beach dune and coastal strand. The edge typically occurs in the climax of the upper dune and is parallel to the elevation gradient. As elevation and distance to dune
44 are strongly correlated only one factor was included in the original a priori hypothese s and the corresponding predictive models E levation may also be an important factor influencing burrow site selection. The lowest elevation for a burrow site found in the study site was 3.1 meters above sea level wh ereas the lowest for a r andom site was 0 meters above sea level. Hansen (1963) found that the mean depth of Gopher Tortoise s burrows is 2 m. Therefore, elevation may be a limiting factor on burrow construction in coastal sand dune because of a need to prevent salt water intrusion into the burrow Burrow site selection probability increases as the percentage of herbaceous vegetation cover measured in the microhabitat scale increases. Gopher Tortoise s mostly forage within 50 m of their burrow, and most foraging activities are concen trated on areas with high herbaceous ground vegetation. The result s of this study s uggest that th e amount of herbaceous vegetation cover could be a visual cue that Gopher Tortoise s use to select suitable burrow site. Gopher Tortoise s appear to select burrow sites with softer soil, but the exact mechanism they use to determine soil resistance is unknown. At the coastal sand dune site, the soil in the fore dune mostly consists of coarse mollusk shell fragments, whereas the soil in the upper dune consists of finer, smaller grain ed sandy soils. Gopherus polyphemus are prolific burrowers with morphological adaptations that improve burrowing efficiency (Auffenberg 1966), and it is possi ble that these adaptations may help the tortoise differentiate soil textur es. The n umber of tortoise burrows within a 20 m radius of each location cor relates positivity to the probability of burrow site selection. This result is likely an autocorrelation. Since Gopher Tortoise s excavate multiple burrows and use them
45 concurrently, and burrows are usually the center of their home ranges, it is not surprising that there are more than one burrow within the radius of a Gopher Tortoise home range. The r es ults of this study are consistent with the Gopher Tortoise habitat s election literature. Jones and Dorr (2004) using a similar approach, found that Gopher Tortoise occurrence in an intensively managed p ine plantation is influenced by soil types total and mid story canopy coverage, and edaphic and vegetative conditions. A ctive burrow occurrence was related positively to increasing mid story canopy closure and certain soil types. Baskaran et al (2006) used a similar approach to develop a model that predict s Gopher Tortoise habitat in a five county region in Georgia. The mo del they developed predicted Gopher Tortoise habitat based on burrow associations with land cover type, soil type, topography and hydrology. Land cover types, soil types, and distances to streams and roads were the most significant predictors in their mode l. Although t he best model did not includ e any anthropogenic factors, the significance of distance to beach access in the global model which is a statistical equivalent of the best model, approaches significan ce 2 = 3.71, p = 0.054 ) and ha s 0.00121 0.00063 ) indicating that burrow site selection probability increase s farther away from the beach access. Based on the at this study site burrow density tends to be higher in area s farth er away from the beach access, and most human activities occur within <200 m from the beach access. The only other anthropogenic factor measured in this study was distance to road, a factor that was dropped because of high correlations with distance to edg e. Distance to road was found to be a useful predictor of the occurrence of Gopher
46 Tortoise burrow s in another study (Baskaran et al. 2006) however Pruner (2010) studied nest site selection of the Snowy Plover ( Charadrius alexandrinus ) and found that the presence of symbolic fencing increased probability of nest site selection. Anthropogenic disturbances in the form of motorized vehicles, human activities, or introduced predators should not be overlooked, as our knowledge of impacts to wildlife associated with human disturbance is limited. The effect of sex on habitat selection was not examined in this study as it was logistically difficult to determine the sex of the tortoise that excavated each burrow. Nonetheless, sex has been shown to have an effect on habitat selection as Graeter et al (2008) found that female Marbled Salamanders ( Ambystoma opacum ) selected clear cut forest more frequently than males. Male Gopher Tortoises may prefer to excavate burrows in areas with high female d ensity, whereas females may prefer to do so in areas with adequate nesting habitat. The study design I employed can easily be replicated to study habitat selection of other animals with conspicuous habitat features (such as nest or burrows) or radio tagged animals. The results of this study demonstrate that f ine scale factors that relate directly to the natural history of Gopher Tortoise s such as percent herbaceous cover and soil resistance have significant influence s on habitat selection, and Gopher Tortoise s may use these visual and physical cues to select burrow site s More research should be conducted to understand these mechanisms. Research c ould address whether species richness affect s burrow site selection whether tortoises are able to recogni ze specific vegetative species and choose their burrow sites accordingly and whether tortoises can differentiate between well drained soil and excessively drained
47 soil for burrow selection, much as nesting sea turtles and diamondback terrapins have been shown to do in similar habitats (Burger and Montevecchi 1975, Garmestani et al. 2000, Wood and Bjourndal 2000)
48 Table 2 1. Potential factors influencing burrow site selection for Gopher Tortoise s in coastal sand dunes of Guana Tolomato Matanzas National E stuarine Research Reserve, South Ponte Vedra Beach, Florida, USA, 2010. Variable Spatial scale Environmental factors Soil resistance m icrohabitat Elevation m icrohabitat Slope m icrohabitat Slope angle m icrohabitat Soil type home range Biological factors % bare ground microhabitat % herbaceous cover microhabitat % grass microhabitat % woody scrub and vines microhabitat % litter microhabitat Canopy cover microhabitat Distance to edge home range Land cover type home range Number of tortoise burrows home range Anthropogenic factors Distance to road home range Distance to beach access home range = categorical variables
49 Table 2 2. Burrow site selection models based on a priori hypotheses for Gopher Tortoise s (n = 198) in the coastal sand dunes of Guana Tolomato Matanzas National Estuarine Research Reserve, South Ponte Vedra Beach, Florida, USA. The best m odels ( ) are cho sen corrected for small sample sizes (AICc), AICc is based on 2 log likelihood, which is the value of the maximized log likelihood function of the model factors given the data set, the number of factors (K), and Cc is the AICc differences relative to the smallest AICc in the model set. Model Factors K 2 LL AICc w environmental + biological SR SL SA HE GR SV LI CA SO LC TB DE 13 125.14 153.13 0.00 0.5378 g lobal SR SL SA HE GR SV LI CA SO LC TB DE DA 14 123.14 153.45 0.32 0.4588 biological + anthropogenic HE GR SV LI CA LC TB DE DA 10 142.88 164.06 10.93 0.0023 biological only HE GR SV LI CA LC TB DE 9 147.55 166.51 13.38 0.0007 home range only SO LC TB DE DA 6 155.10 167.55 14.42 0.0004 environmental +anthropogenic SR SL SA SO DA 7 179.76 194.36 41.23 <0.0001 microhabitat only SR SL SA HE GR SC LI CA 9 202.19 221.16 68.03 <0.0001 environmental only SR SL SA SO 5 222.84 233.15 80.02 <0.0001 anthropogenic only DA 2 258.72 262.78 109.65 <0.0001 SR = soil resistance, SL = soil slope, SA = slope angle, HE = % herbaceous cover, GR = % grass cover, SV = % woody scrubs and vines cover, LI = % litter cover, CA = presence or absence of canopy (>2 m) cover, SO = soil type, LC = land cover type, TB = number of tortoise burrows within 17 m radius, DE = distance to the nearest edge, DA = distance to the nearest beach access.
50 Table 2 3. Relative importance o f burrow site selection factors, parameter coefficient estimates ( ) for Gopher Tortoi se s (n = 198) in the coastal sand dunes of Guana Tolomato Matanzas National Estuarine Research Reserve, South Ponte Vedra Beach, Florida, USA, ranked by the sum of model weight. The most important factor ( ) has the highest sum of model weight. ues environmental + biological Factors ** Sum of model weight (SE) P LC 1.0000 1.91 ( 0.40) <0.0001 *** DE 1.0000 0.07 ( 0.03) 0.0055 ** HE 0.9996 0.04 ( 0.02) 0.0241* GR 0.9996 0.02 ( 0.01) 0.0939 SV 0.9996 0.02 ( 0.01) 0.0539 LI 0.9996 0.03 ( 0.02) 0.1231 CA 0.9996 0.32 ( 0. 34) 0.3403 TB 0.9993 0.54 ( 0.14) 0.0001 *** SR 0.9973 0.93 ( 0.35) 0.0079 ** SO 0.9971 0.71 ( 0.38) 0.0659 SL 0.9967 0.97 ( 1.14) 0.3966 SA 0.9967 0.01 ( 0.01) 0.0121 DA 0.4615 ** SR = soil resistance, SL = soil slope, SA = slope angle, HE = % herbaceous cover, GR = % grass cover, SV = % woody scrubs and vines cover, LI = % litter cover, CA = presence or absence of canopy (>2 m) cover, SO = soil type, LC = land cover type, TB = number of tortoise burrows within 17 m radius, DE = distance to the nearest edge, DA = distance to the nearest beach access. 0.0005
51 Figure 2 1. Probability of burrow site selection a s a function of soil resistance from the most parsimonious model for Gopher Tortoise s in the coastal sand dunes of Guana Tolomato Matanzas National Estuarine Research Reserve South Ponte Vedra Beach, Florida, USA, 2010. Solid line represents parameter coe 0.0 0.2 0.4 0.6 0.8 1.0 0 1 2 3 4 5 Probability of Selection Soil Resistance
52 Figure 2 2. Probability of burrow site selection as a function of distance to edge (m) ( 1 SE) from the most parsimonious model for Gopher Tortoise s in the coastal sand dunes of Guana Tolomato Matanzas Nati onal Estuarine Research Reserve South Ponte Vedra Beach, Florida, USA, 2010. Solid line represents 0.0 0.2 0.4 0.6 0.8 1.0 0 10 20 30 40 50 Probability of Selection Distance to Edge (m)
53 Figure 2 3. Probability of burrow site selection as a function of percent h erbaceous cover ( 1 SE) from the most parsimonious model for Gopher Tortoise s in the coastal sand dunes of Guana Tolomato Matanzas Nati onal Estuarine Research Reserve South Ponte Vedra Beach, Florida, USA, 2010. Solid line represents parameter coefficien 0.0 0.2 0.4 0.6 0.8 1.0 0 20 40 60 80 100 Probability of Selection Herbaceous Cover (%)
54 Figure 2 4. Probability of b urrow site selection as a function of number of tortoise burrows ( 1 SE) from the most parsimonious model for Gopher Tortoise s in the coastal sand dunes of Guana Tolomato Matanzas National Estuarine Research Reserve South Ponte Vedra Beach, Florida, USA, 2010. Solid line 0.0 0.2 0.4 0.6 0.8 1.0 0 2 4 6 8 10 Probability of Selection Number of tortoise burrows
55 CHAPTER 3 HOME RANGE, SEASONAL ACTIVITY AND MOVEMENT PATTERNS OF GOPHER TORTOISE IN COASTAL SAND DUNE HABITAT Background The concept of home range first proposed by Burt (1943), is a fundamental concept in spatial ecology. The home range of an animal is the area that encompasses all the resources the animal requires to survive and reproduce. Several general principles o f home ranges of vertebrates have been proposed. McNab (1963) first demonstrated that mammal home range size is a linear function of body size, and noted that carnivores have larger home ranges than similar sized herb ivores. Since then, multiple formulas have been generated to empirically estimate home range sizes of mammals based on body size and diet (Harestad and Bunnell 1979). Birds, due to their high basal metabolic rate comparable to some mammals, exhibit a simil ar relationship. Mace and Harvey (1983) found the same body size home range size relationship in passerine birds and rodents, and noted that birds have larger home range sizes than similar sized rodents with similar diets, possibly due to their higher ener gy requirements Additionally, the se authors found that home range size of granivores are consistently larger than that of other herbivores. In solitary territorial carnivorous mammals, female home ranges are determined by the distribution and abundance o f food, wh ereas male home ranges, during the mating season are largely determined by the distribution of conspecif ic females (Erlinge and Sandell 1986). Home range size of predator and prey density appears to be negatively correlated in female B obcats ( Lyn x rufus ) (Litvaitis et al. 1986) and Canadian L ynx ( L canadensis ) (Ward and Krebs 1985). Home range sizes of males during the mating season are expected to be larger than estimates based on energy requirements, since a
56 hat of several females to increase the probability of mating (Sandell 1989). In reptiles, home range sizes are generally associated with the distribution of one or more resources (e.g. food, shelter mates, thermoregulation sits ) and positive correlation s between home range size and body size have been observed in many species. Sex, reproductive state, density, and food availability are some of the major factors that influence home range size in reptiles (Zug et al. 2001) Gregory et al. (1987) suggested that habitat structure and resource availability are the most important and obvious factors that affect home ranges and movement patterns of snakes. For example, snakes in the Temperate Zone require separate overwintering sites and feeding areas that occur in widely separated habitats, and long distance seasonal migrations between these habitats have been observed by many authors ( Larsen 1987, Madsen and Shine 1996 Wilson et al. 2006) Hailey and Coulson (1996) studied home range and daily movem ent distance of two African tortoises ( Leopard Tortoise, Geochelone pardalis and Kinixys spekii ) and found that G. pardalis exhibited proportionally larger home ranges than K. spekii The authors observed frequent movements by G. pardalis to an area with sodium rich soils and argued that Leopard tortoises have exceptionally large home range because of the spatial di stribution of sodium rich soils rather than energetic requirements. The home range of the same organisms also may d iffer according to h abitat type and quality (Diemer 1992a Smith et al. 1997 ). Gopher Tortoise s ( Gopherus polyphemus ) occur in a variety of habitats throughout Florida. Although the majority of ecological studies have been conducted in sandhill
57 (longleaf pine turkey oak wiregrass community) or pine plantations that were once sandhill relatively few studies have focused on populations that occur in coastal ecosystems (Kushlan and Maz zotti 1984, Pike 2006 Waddle et al. 2006 ) and information on the spatial ecology of this species in these ecosystems is lacking Gopher Tortoise s that live in coastal sand dunes may be challenged by threats such as beach erosion, exposure to high salinity, urban predators, and other natural disa sters such as hurricanes and catastrophic wildfire in addition to the typical habitat loss and degradation experienced by in land populations. Florida is surrounded by coastal sand dunes that have the potential to suppo rt considerable populations of Gopher Tortoise s (A u ffenberg and Franz 1982 ) As h ighly fossorial herbivore s Gopher Tortoise s spend up to 98% of their time inside their burrows and only leave to forag e, court mat e, or nest ( Johnston 1996, Ernst and Lovich 2009) Individual tortoise s m ay use multiple burrows and int r a specific burrow sharing is not uncommon. A typical home range of a Gopher Tortoise is small and centers around it s burrow s and includ es the surrounding foraging areas. Gopher Tortoise s typically feed within 50 m of the burrow (Au ffenberg and Franz 1982). M ales typically have larger home range s that overlap with the home range of several females (Auffenberg and Iverson 1979). Home range size has been hypothesized to be an indicator of Gopher Tortoise habitat quality (Diemer 1992, Mushinsky and McCoy 1994). Other authors have found that the amount of herbaceous ground cover is negatively correlated with home range size (Auffenberg and Iverson 1979 Mushinsky and Gibson 1991). A number of home range studies have been conducted in san dhill and pin e plantations (McRae et al 1981
58 Diemer 1992a Smith 1992 Smith et al. 1997 Backus 2003 Eubanks et al. 2003 Yager et al. 2007 ) with varying results, but the general consensus is that Gopher Tortoise home range varies greatly both spatiall y and temporally. Nevertheless, home range size continues to be recognized as an important predictor of habitat quality. H ome range size and shape of sand dune Gopher Tortoise s may be very different compared with those occupying uplands. In terms of shape, sand dune Gopher Tortoise s likely have linear home ranges because of the linear distribution of habitat along beach dunes. Therefore, this suggests that the overall size of coastal home ranges might be either smaller (due to beach related disturbances and subsequent effects on activity) or larger (depending on forage distribution and quality) than Gopher Tortoise home ranges in in land habitats. In this study, radio telemetry was used to determine the movement, seasonal activity patterns, and annual home range of Gopher Tortoise s at a coastal sand dune site. Habitat requirement s of Gopher Tortoise s in this understudied habitat was inferred and the result s of this study c ould help wildlife managers design better conservation plans to protect Gopher To rtoises Methods Study Area This study was conducted in Guana Tolomato Matanzas National Estuarine Research Reserve (GTM NERR) located on the northeastern coast of Florida, St. Johns County. The northern section of GTM NERR includes some of the only pristine undeveloped dunes o n the eastern coast of Florida. All Gopher Tortoise s used for radio telemetry were captured in the 15.7 ha plot in the northernmost section of the study site, north of the North Beach Access
59 Radio T elemetry The coastal sand dune was surveyed for Gopher Tortoise burrows from January through March 2010. Gopher Tortoise s were captured using bucket pit fall traps (Burke and Cox 1988) and opportunistically by hand from April 2010 to May 2010 Upon capture, standard measurements (carapace length and width, plastron length, plastron concavity, mass, sex) were taken with calipers to the nearest mm and a digital scale to the nearest 10 0 g. Males and females were distinguished visually by secondary sexual characteristics (Landers et al. 1980). Each tortoise was assigned a unique number and permanently marked ( Cagle 1 939 ). A total of 20 tortoises (11 females and 9 males) were fitted with radio transmitters (Wildlife Materials, Inc., SOPR 2380). Each radio transmitte r weigh ed less than 5% of the b ody mass of the t ortoise and was glued onto the posterior edge of the carapace using quick bonding epoxy putty stick ( West Marine, Watsonville, CA ) (Figure 3 1) Tortoises were released immediately at their site of capture following measurements and transmitter s attachment Tortoises w ere located once every 2 to 3 days from May to August 2010 and once every 7 10 days from September 2010 to May 2011 using a 3 element Yagi Uda antenna and a TRX 1000WR receiver (Wildlife Mater ial, Inc.) Geographic locations were recorded with a handheld GPS with a n error of +/ 3m (Garmin International Inc, Olathe, K S ), along with t ime of day, weather, and specific activity ( e.g. walking, foraging, resting in burrow ) of the tortoise. During t he inactive season from December to February sticks were placed in front of over wintering burrows to determine the length of the inactive period and to confirm survival during the winter (Diemer 1992a ).
60 Data Analysis I examine d the movement patterns of adult Gopher Tortoise s during the en tire study period by determining the mean distance moved between successive locations, the mean annual home range area, the number of moves, and the total number of burrows used following the methods of Eubanks et al ( 2003). Distances moved were calculated as the straight line distance between successive burrows or non burrow locations. Mean distance moved w as measured per move within a calendar month (i.e 100 m / 4 moves = 25 m/move). Home range area ( 100% minimum co nvex polygon [MCP] and 95% fixed kernel ) were calculated using Biotas version 2.0a 3.8 (Ecological Software Solutions LLC ). The s moothing parameter in 95% kernel estimate s w as selected by least squares cross validation ( Seaman and Powell 1996). A Shapiro Wilk test for normality indicated that the values of annual home range area s (MCP) and mean distance s moved between successive locations were non normal ly distributed. All other data were normally distributed. Mean number of burrows used and mean number of moves by males and females were compared using one way ANOVAs. The n on parametric Kruskal Wallis rank sum test was used to compare mean distance moved between successive locations and the mean annual home range areas ( MCP ) by males and females. Alpha le vel for all analyses was 0.05 and all means were reported with standard errors unless otherwise specified All statistical analyses were conducted in R version 12 (R Development Core Team 2010). Results From May 2010 to May 2011, 20 Gopher Tortoise s were tracked 1,120 times (mean = 56 times, range = 46 72, S.E = 1.19 ). Tortoises on average were tracked for a period of 325 days (range = 257 3 58, S. E = 4.70 ).
61 Movement D istances For t he entire study period, female tortoises on average traveled a distance of 39.2 m per move (n = 11, range = 5.5 156.9, S. E = 13 3 ) wh ereas males tr aveled 47.0 m per move (n= 9, range = 24.6 83.4, S. E = 7.68) Mean distance moved wa s not significantly different between males and fem ales (Kruskal Wallis 2 = 2.92, df = 1, p = 0.08). Both sexes were mostly inactive from January to February and traveled fewer than 4 m per move Movements were observed as early as March and as late as December. In terms of mean distance traveled per move by month, b oth sexes e xhibited similar pattern s but males traveled significantly more distance per move in April and August (Figure 3 2 ) Burrow U se and A ctivity P atterns Of 1,120 locations, tortoises were observed above ground only 26 times (2.32%), of which 19.2% were foraging, 11.5% were basking, and 69.2% were walking. Over the entire study period, f emales on average used 6 burrows (n = 11, range = 2 12, S.E. = 0.89), wh ereas males used 11 burrows (n = 9, range = 5 14, S.E. = 1.09). Males used significantly more b urrows than females over the course of the year (F 1, 18 = 11.43, p = 0.003). In terms of number of moves during the year f emales on average moved 14 times (n = 11, range = 3 28, S.E. = 2.52), wh ereas males moved 28 times (n = 9, range = 6 38, S.E. = 3 .10 ) Males moved more often t han females (F 1, 18 = 13.21, p = 0.002) Both sexes exhibited similar seasonal pattern s in number of burrow s used, but males used almost twice as many burrows as females per month from July to August (Figure 3 2).
62 Annual Home R ange Over the entire study period, adult Gopher Tortoise s had an annual mean home range of 0.37 ha (MCP, n = 20, S. E = 0.14 ) and 0.25 ha (kernel, n = 20, S. E = 0. 02 ) (Table 3 1). Home range medians did not differ between sex for MCP estimates (Males: 0. 32 ha, n = 9, S.E = 0.06; Females: 0.42 ha, n = 11, S.E = 0.26 ; Krusk al Wallis 2 = 2.19, df = 1, p = 0.1385). For fixed kernel estimates, males had significantly larger mean annual home range area s than females (Males: 0.31 ha, n = 9, S.E. = 0.03; Females: 0.21 ha, n = 11, S.E. = 0.03 ; F 1, 18 = 5.50, p = 0.03 ) (Table 3 1) Dispersal P attern During the 13 month study period, all but one radio tagged tortoise stayed within the 78 ha study area, and most movements were confined within the 15.7 ha sub plot where these tortoises were originally captured One male (#T18) used two burrows located near a beach front house outside of the study site boundary. Most movements were from north to south or vice versa rather th an from east to west The longest recorded movement was 650 m over a 12 day period by a female tortoise ( C arapace length = 26.8 cm). Discussion In this study c onsiderable amount s of individual variation exist amon g annual home range areas distance moved, number of burrow s used, and number of moves. Overall, Gopher Tortoise s in coastal sand dune s have smaller home range s than conspecifics found in other habitat types (Table 3 4; Pike 2006) The home range of male tortoises appears to be la rger than that of females, and males used more burrows than females. These results are consistent with those reported in the literature (McRae
63 et al. 1981 Diemer 1992 a Eubanks et al. 2003). Gopher Tortoise s tend to move short distances. The results of t his study suggest that dispersal by Gopher Tortoise s out of sand dune s is not common. During the course of this study, all radio tagged tortoise s stay ed within the coastal dune area, but 2 unmarked tortoises (1 juvenile and 1 subadult male ) were found dead on heavily traveled State Road AIA, apparently killed by passing vehicles while attempting to cross the road. The area west of the study site consists of mostly coastal strand, coastal scrub and maritime hammock. Tortoise burrows have been observed in the coastal scrub and maritime hammock but a thorough burrow survey has not been conducted, as the area has not been burned recently and it is physically imposs ible for humans to penetrate without removing obstructive plants. T he tortoises in the beach sand d une are essentially trapped within their habitat, forming an isolated population. The dispersal ability of tortoise s is limited and it is unlikely that many tortoises can disperse out of the sand dune without being hit by a vehicle. The m inimum convex poly gon is one of the most widely used methods for home range estimation, but it tends to overestimate home ranges by including outlying locations that are not used frequently. On the other hand, the kernel method tends to focus on major activity centers and i gnores outlying locations, which may underestimate home range in species that spend a majority of time in fixed locations, such as the Gopher Tortoise In this study, mean annual ho me ranges estimated by MCP for all tortoise and females only we re larger th an those estimated by the kernel method (48% and 1 01%, respectively; Table 3 1). As mentioned, MCP tends to overestimate home range, and
64 the mean annual home ranges of females estimate d by MCP was possibly inflated due to a female tortoises that had an exceptionally large home range (2.95 ha) due to a single long distance movement in July (650 m). If that tortoise was removed, the mean annual MCP home rang e for females becomes 0.16 ha. T his study demonstrated that the home range of sand dune Gopher Tortoise s is smaller than that of upland conspecifics. Gopher Tortoise s in coastal sand dune at GTMNERR have far smaller home ranges and use d fewer burrows than Gopher Tortoise s at Kennedy Spac e Center, a coastal scrub site (Smith et al. 1997). McRa e et al (1981), Diemer (1992b), Smith et al. (1997), Eubanks et al (2003), and Yager et al (2007) estimated home range of adult Gopher Tortoise s using similar methods, and reported male mean annual home ranges that are 0.45, 2.92, 4.86, 2.39, 5.02, and 3.01 times larger respectively, than e stimates from the present study (Figure 3 4). Sand dune tortoises also appear to frequent fewer burrows th a n upland tortoises, possibly due to the distribution of forag e plants. Based on the results of this study (Chapter 2), Gopher Tortoise burrow selection is related to percen t herbaceous cover. By excavating burrows near their primary food source, tortoises need not travel far for food. Based on radio telemetr y data, burrow sharing among conspecifics of the same or opposite sex is common at GTMNERR and was observed multiple times (n = 40). As many as four radio tagged tortoises have been observed sharing a single burrow simultaneously and two tortoises overwin tered in the same burrow Diemer (1992a) and Epperson and Heise (2003) reported similar findings at n orth ern Florida and s outhern Mississippi site s respectively and indicat ed burrow sharing in Gopher Tortoise s is not a
65 rare phenomenon. Burrow sharing by Gopher Tortoises also has been observed commonly on Egmont Key, Florida (C.K. Dodd, Jr., personal communication). There are few accounts of courtship and mating behaviors of Gopher Tortoise s, which typically occur in the spring to autumn in Florida (Auffen berg 1966, Ernst and Lovich 2009 C. K. Dodd, Jr., personal communication ). At this study site, courtship and mating have been observed as late as October. The burrow defense behavior described by Diemer (1992a) and Dodd (personal communication) was observ ed when a mating pair was interrupted by the author. T he female immediate ly retreated into the burrow and the male blocked the entrance by turning itself sideway s
66 Table 3 1. Mean annual home range (ha) of adult Gopher Tortoise s in coastal sand dunes, Guana Tolomato Matanzas National Estuarine Research Reserve, South Ponte Vedra Beach, Florida, USA (2010 2011), using 100% minimum convex polygon (MCP) and 95% fixed kernel methods. Values after removing a outlier tortoise are in parentheses. n MCP HR SE Range Kernel HR SE Range All tortoises 20 0.37 0.14 <0.01 2.94 0.25 0.02 0.09 0.52 Males 9 0.32 0.06 0.13 0.63 0.31 0.03 0.23 0.52 Females 11 0.42 (0.16) 0. 25 (0.05) <0.01 2.94 0.21 0.03 0.09 0.41
67 Figure 3 1. A Gopher Tortoise with a radio transmitter (Wildlife Materials, Inc., SOPR 2380) affix ed to its carapace A unique number was painted on each individual to allow temporar y identification from a distance. (Photo courtesy of author)
68 Figure 3 2 Seasonal movement patterns (mean distan ce traveled per move + SE ) of radio tagged adult Gopher Tortoise s (n = 20) in coastal sand dunes, Guana Tolomato Matanzas National Estuarine Research Reserve, South Ponte Vedra Beach, Florida, USA, May 2010 May 2011. Solid bars represent females and empt y bar s are males. 0 20 40 60 80 100 120 140 M J J A S O N D J F M A M Distance Moved (m) Months
69 Figure 3 3 Seasonal movement patterns (mean number of burrows used + SE) of radio tagged adult Gopher Tortoise s (n = 20) in coastal sand dunes, Guana Tolomato Matanzas National Estuarine Research Reserve, South Ponte Vedra Beach, Florida, USA, May 2010 May 2011. Solid bars are females and empty bar s are males. 0 1 2 3 4 5 6 M J J A S O N D J F M A M No. of Burrows Months
70 Figure 3 4. Mean annual home range (minimum convex polygons) + SE of adult Gopher Tortoise s in previous and present studies Solid bars are females and empty bar s are males. 0.0 0.5 1.0 1.5 2.0 2.5 Present study Yager et al. Control 2007 Yager et al. Burned 2007 Smith et al. 1997 McRae et al. 1981 Eubanks et al. 2003 Diemer 1992a Mean Annual Home Range (ha)
71 CHAPTER 4 CONCLUSIONS The Gopher Tortoise management plan prepared by FWC (2007) suggested that site specific studies on topics such as seasonal activity, home range size, fecundity, and adult sex ratios may be needed in certain understudied habitats. The results presented in this study may serve as an initial assessment of Gopher Tortoise pop ulations in coastal sand dunes. Status of Sand Dune Tortoises As mentioned in Chapter 1, the density and distribution of Gopher Tortoise at the GTMNERR site is highly patchy The majority of the burrows were found in the northern section, which indicates relatively high habitat quality in the area. The southern secti on contain ed much fewer burrows and represents marginal habitat. Burrow locat ions in the southern section were farther apart from each other and most likely a result of the relatively low tortoise density. The vegetation structure and relative openness of the habitat inevitably have effects on burrow site selection of Gopher Tortoise in GTM NERR, as presented in Chapter 2, but the relationships between those variables and tortoise density is unclear. Habitat characteristics of random points did not differ si gnificantly between the northern and southern sections (all significant value s (p) >0.05). The disparity in tortoise density at the GTMNERR sand dunes appears to be consistent over time, as the same trend was observed in a burrow survey conducted in 2007 ( GTMNERR, unpub. data) During the course of this study, few hatchlings or juveniles (n = 4) were encountered at the GTMNERR site. Three were observed in the beach dune an d one was a road kill found on State Road A1A. Based on burrow surveys conducted by th e
72 author, juvenile burrows were few, and the population appears to be skewed towards larger adults (Figure 4 1 ). The lack of juvenile burrows may not reflect the true population structure because of low relative detectability of juveniles and their burrows (Pike 2006). Juveniles are also known to use adult burrows (Diemer 1992a). Nonethe less, the unimoda l distribution of burrow widths should not be completely dismissed, as a burrow survey conducted in similar coastal habitat in Cape Sabal, Florida using sim ilar metho do logy ha s produced a bimodal distribution that indicates high recruitment (Waddle et al. 2006) The shortline Gopher Tortoise at GTMNERR should continue to be monitored to assess its long term status as it represent s one of the largest pop ulati ons within GTMNERR and in n ortheastern Florida. The coastal population at GTMNERR may have reached its carrying capacity, meaning that it cannot support more tortoises. Carrying capacity is positively correlated with habitat quality, but a high quality habitat may be below its carrying capacity if an external factor such as human exploitation reduces population size. Density can also be a misleading indica tor of habitat quality if the habitat is indeed an ecological trap (Van Horne 1983). Little is known about t he population status (i.e. stable, declining, or increasing) of Gopher Tortoises in other coastal d une system s in Florida C oastal beach dunes are found throughout most of the state, but few pristine beach dunes remain undeveloped or unaltered by human activities On the Atlantic coast, the largest populations of Gopher Tortoises can be found along Cape Canaveral National Sea Shore and Merritt Islan d National Wildlife Refuge (Kennedy Space Center), where a large amount of land has been under protection since the 1960s (Breininger et al. 1991 Breininger et al. 1994 Smith et al. 1997 Pike 2006) Outside of Kennedy Space Center, Gopher
73 Tortoise popul ations probably exist in small, isolated habitat pockets in between beach front properties similar to the present study site. The southernmost population of this species, at Cape Sabal, Monroe County ha s experienced a steep decline in recent years (19 90 t o 2001). Waddle et al. (2006) reported a 76% decline of active burrows at that site possibly due to the reduction of habitat quality and tropical storms. Sch wartz et al (2005) visited the same site in 2001 and suggested the population may have be en extirpated by natural causes (i.e. H urricane Gabrielle). Large tropical storms have been suggested to be an important cause of mor t ality in coastal populations of Gopher Tortoises (Kushlan and Mazzotti 1984), and are most likely responsible for the uneve n vegetation structure and tortoise burrow density found between the northern and southern sections at the GTMNERR site. Effects of H uman A ctivities Human activities may be the great est deterministic t h reat to sand dun e Gopher Tortoise s. During the one yea r study period, three road killed Gopher Tortoise s were observed (2 adult males, 1 juvenile) Since the coastal sand dunes study site is essentially a closed habitat surrounded by housing developments, the Atlantic O cean and State Road A1A, tortoises are confined to a narrow strip of habitat and have no place to disperse. The only vi able option for dispersal is to move west toward s the coastal scrub /maritime hammock habitat. However, this move is extremely risky as the traffic on State Road A1A increases substantially during the tortoise active s eason (March November) because of the increase in number of beach us ers (Figure 4 2 ) In addition to direct mortality caused by road traffic, sand dune tortoises may also be affected by human activities on the beach and by introduced species or human subsidized predators (e.g., raccoons) South Ponte Vedra Beach is a highly used
74 recreational beach throughout the year. There are three official dune overpasses and several other unofficial trails that allow easy access throughout the beach. Access to the dunes is prohibited by the reserve, but human activities are apparent based on the amount of trash and litter observed in the dunes. From Chapter 2 of this thesis, distance to beach access has bee n shown to be a significant factor that negatively affect burrow site selection probability. Further, human activities may have indirect effects on tortoise behavior that are not easily observable (e.g. reduce foraging effort; discourage dispersal) Wild tortoises are generally nervous when they encounter humans and terminate whatever activity they are engaged in and run for the closest shelter (e.g. burrows). Additionally, non native species may negatively affect Gopher Tortoise s. Feral cats have be en observed in the dunes and they may prey upon hatchling and juvenile tortoises. The exotic C a ctus M oth ( Cactoblastis cactorum ) ha s infested Prickly P ear ( Opuntia spp.) in the dunes and i s causing localized mortality of the plant, which is a favored torto ise food source and an important source of fresh water for sand dune Gopher Tortoise s Sea Level R ise Global climate change and the associated sea level rise could become a major threat to the continued survival of sand dune Gopher Tortoise s Global sea level rise projections by 2100 var y greatly depending on sources and the climate scenarios used in the models The United Nations Intergover nmental Panel on Climate change Assessment Report, one of the most cited sources of information on sea level rise, p redicted a 0. 18 m to 0.59 m rise in the 21 st century (IPCC 2007). However, Allison et al. (2009) argued that sea level is likely to rise as much as 2 m by 2100 as the IPCC models did not take into account the dynamical processes and the resulting contribution
75 to global sea level rise by Greenland and Antarctic ice sheets. Furthermore, both IPCC (2007) and Ovenpeck and Weiss (2009) warned that global warming in this century may be sufficient to produc e a 4 to 6 m sea level rise in future centuries Under such scenario, the majority of the present Florida coastal area wi ll be under water (Figure 4 3 ). As sea level continues to rise, m ost of the fore dune will be gradually washed out and the dune will m ove inland. S and dune wildlife, including Gopher T ortoise s, will have to m ove inland as well. However, State Road A1A will likely act as a barrier to the natural migration process. In the future, wildlife manager s may have to monitor coastal wi ldlife migration and implement facilities that would aid migration (i.e. corridors or culverts Dodd et al. 2004 ) as the coastal habitat starts to shift inland Manual trans location may also be an option for focal species with limited dispersal ability such as the Gopher Tortoise s. Management I mplicat ions Coastal Habitat Management In terms of habitat management in the sand dune, certain procedures intended to make the beach more accessible to recreational users such as raking seaweed off of the beach or building pedestrian paths should be avoided as it may destabilize the natural pl ant community successions. Seaweeds are important nutrient source for some of the pioneer species on the fore dune. In areas with strong onshore winds, pedestrian paths that are perpendicular to the beach can promote blowouts and allow a wave of sand to bu ry existing stable dune vegetation; thus pedestrian paths should be built to form an angle less than 90.
76 Feasibility As Relocation Sites Although the use of reintroduction as a conservation technique has been a controversial subject ( Burke 1991 Dodd and Seigel 1991 Reinert 1991 ), it is used by the FWC to mitigate populations of Gopher Tortoise affected by human development. One of the main objectives of the Gopher Tortoise management plan is to increase the number of relocations to suitable, man aged, and protected habitat in stead of unprotected areas (FWC 2007). However, relocation should be treated as a last resort, as mismanaged past relocations have led to mixed gene pool s (Schwartz and Karl 2005) disease transmission and outright failure Th e coastal sand dune of GTMNERR with ample a mount of herbaceous cover and suitable soil s for burrowing, appears to be a high quality habitat for Gopher Tortoise s. The lack of frequent fire in this system simplifies the management requirements often needed in other Gopher Tortoise habitat (e.g. prescribed fire in upland pine forest and scrub). However, in its current state, the GTMNERR site particularly the northern section, appear s to be occupired at maximum carrying capacity by Gopher Tortoises and therefore should not be considered as a recipient site. The exact causes of local extirpation of the once abundant coastal populations of Gopher Tortoise is unclear. Disappearance could be a result of stochastic event s such as large tropical storms an d fire or deterministic threats such as overexploitation for food, habitat loss rela ted to development, disease or a combination of many causes Wildlife managers need to consider the viability of reintroducing Gopher Tortoises through detailed individua l habitat assessment and experimental releases before committing to an y large scale reintroduction. Priorities should be given to sites that have
77 support ed Gopher Tortoises historically and still po ss ess the optimum habitat characteristics th at will suppor t viable population s M any of the undeveloped coasta l area s in Florida are protected and may potentially serve as inexpensive alternative recipient sites. Management costs for Gopher Tortoises in coastal sand dunes are considerably cheaper than in upland habitats, since prescribed fires and other fire related habitat alteration (i.e. rol ler chopping) are not necessary. The success rate of Gopher Tortoise relocation is potentially affected by the time of introduction, habitat similarity between donor site and recipient site, and sex ratios of tortoises (FWC 2007). Temporary enclosures (i.e ., soft release) have been used at many recipient sites to increase site fidelity (Lohoefener and Lohmeier 1986, Tuberville et al. 2005). Precautions should also be taken to prevent introducing tortoises infected by upper tract respiratory disease (UTRD) t o the recipient site. Public Education Perhaps the most important task for Gopher Tortoise management in the coastal sand dunes is to properly educate the public about the existence of these animals and their important ecological roles. Based on conversati ons with some frequent beach use rs many of them did not know Gopher Tortoise s can be found in the dunes, and some people have even attempted to release tortoises into the ocean as they have mistaken Gopher Tortoise s for sea turtles. Information kiosk s con taining relevant information on Gopher Tortoise ecology and threats should be added to the existing arrays of educational materials at the GTMNERR public beach access in order to increase public awareness.
78 Wildlife crossing sign s should be installed at the highway to warn passing motorist s about crossing wildlife. During the course of this study, i n addition to Gopher Tortoise s, many other wildlife species have been observed dead on the highway, including the B lack R acer ( Coluber constrictor n = 2), E astern D iamondback R attlesnake ( Crotalus adamanteus n = 1), R at S nake ( P a n t h erophis spp., n = 3), Striped Mud T urtle ( Kinosternon baurii n = 1), Marsh Rabbit ( S y lvilagus palustris n = 3), countless Raccoon ( Procyon lotor ), Virginia Opossum ( Didelphis virginian us ), Nine banded A rma dillo ( Dasypus novemcinctus ) frogs, and numerous passerine birds. Future Research Future research on sand dune Gopher Tortoises should focus on quantifying population level parameters such as gr owth, recruitment, and survival rates, historic and contemporary gene flow patterns between adjacent areas, and viability of coastal sand dunes as long term r ecipient sites for restocking. Such data are especially needed at populations affected by human ac tivities and at sites where tortoises have been relocated as mitigation for ongoing development. In order to fully understand the value of coastal dunes as recipient sites, data on natural carry ing capacity of this habitat need to be collected. Unfortunate ly, few coastal sand dunes in Florida remain ed undeveloped. The GTMNERR study site is one of the few remaining pristine coastal dunes in Florida that support a viable Gopher Tortoise population, and data gathered from this study, such as burrow density and habitat requirements, should serve as baseline data that inform s management decisions on whether or not a coastal sand dune could support viabl e Gopher Tortoise populations. The habitat selection model developed in this study, with modifications such as incorporating fine scale bioclimatic factors can be used to model the habitat suitability
7 9 of the remaining coastal sand dunes throughout Florida to identify potential sites for tortoise relocati on (Baskaran et al 2006). Finally little is known about the population genetic structure of Gopher Tortoise populations that are isolated by habitat fragmentation, such as the one found in GTMNERR, or insular populations, such as the populations on Egmont Key, Florida. Gopher Tortoises in these areas may have experienced population bottlenecks due to lack of gene flow. The GTMNERR site would be a suitable site to study gene flow of Gopher Tortoises across a h uman altered landscape.
80 Figure 4 1. Size class distribution of Gopher Tortoise burrows (n = 106 ) at the coastal sand dune study site at Guana Tolomato Matanzas National Estuarine Research Reserve, South Ponte Vedra Beach, Florida, USA. 0 5 10 15 20 25 30 35 8 12 16 20 24 28 32 36 40 44 48 52 Frequency Burrow width (cm)
81 Figure 4 2. Monthly beach visitor counts (January 2009 December 2010) at Guana Tolomato Matanzas National Es tuarine Research Reserve, South Ponte Vedra Beach, Florida, USA. 0 1000 2000 3000 4000 5000 6000 7000 J09 F09 M09 A09 M09 J09 J09 A09 S09 O09 N09 D09 J10 F10 M10 A10 M10 J10 J10 A10 S10 O10 N10 D10 Visitors
82 Figure 4 3 The Florida coastline after a sea level rise of 3 meters. Re printed by permission f rom Wilber Samantha. 2011. Florida under water: Global warming and rising sea levels. Creative Loafing Tampa. Available from cltampa.com/dailyloaf/archives/2011/05/31/florida under water global warming and rising sea levels/ (A ccessed October 2011).
83 APPENDIX A HABITAT CHARACTERIST ICS (BURROW AND RAND OM LOCATIONS) OF ADU LT GOPHER TOR TOISES IN COASTAL SAND DUNES, GUANA TOLOMATO MATAN ZAS NATIONAL ESTUARI NE RESEARCH RESERVE, SOUTH PONTE VEDRA BEACH, FLORIDA, USA (2010 2011)
84 APPENDIX B PARAMETRIC CORRELATI ON MATRIX FOR VARIAB LES CONSIDERED IN BU RROW SITE SELECTION MODEL SR SA SL BA HE SV LI GR EL DE DR DA TB SR 1.00 0.09 0.35 0.25 0.08 0.30 0.02 0.08 0.50 0.47 0.45 0.12 0.24 SA 1.00 0.29 0.12 0.19 0.24 0.14 0.11 0.09 0.07 0.05 0.10 0.01 SL 1.00 0.02 0.31 0.10 0.15 0.30 0.26 0.36 0.24 0.09 0.03 BA 1.00 0.08 0.62 0.11 0.15 0.52 0.45 0.52 0.11 0.18 HE 1.00 0.24 0.39 0.30 0.15 0.23 0.01 0.13 0.14 SV 1.00 0.16 0.45 0.41 0.36 0.53 0.17 0.06 LI 1.00 0.26 0.07 0.12 0.03 0.05 0.25 GR 1.00 0.12 0.10 0.01 0.19 0.04 EL 1.00 0.80 0.67 0.10 0.29 DE 1.00 0.74 0.12 0.29 DR 1.00 0.05 0.21 DA 1.00 0.28 TB 1.00 SR = soil resistance, SL = soil slope, SA = slope angle, HE = % herbaceous cover, GR = % grass cover, SV = % woody scrubs and vines cover, LI = % litter cover, CA = presence or absence of canopy (>2 m) cover, SO = soil type, LC = land cover type, TB = number of tortoise burrows within 17 m radius, DE = distance to the nearest edge, DA = dis tance to the nearest beach access.
85 LIST OF REFERENCES Abaturov, B. D. 1972. The role of burrowing animals in the transport of mineral substances in the soil. Pedobiologia 12 :261 266. Allison, I., N. L. Bindoff, R. A. Bindschadler, P. M. Cox, N. de Noblet, M. H. England, J. E. Francis, N. Gruber, A. M. Haywood, D. J. Karoly, G. Kaser, C. Le Qur, T. M. Lenton, M. E. Mann, B. I. McNeil, A. J. Pitman, S. Rahmstorf, E. Rignot, H. J. Schellnhuber, S. H. Schneider, S. C. Sherwood, R. C. J. Somerville, K. Steffen E. J. Steig, M. Visbeck, and A. J. Weaver. 2009. The Copenhagen diagnosis 2009: Updating the world on the latest climate science. The University of New South Wales Climate Change Research Centre (CCRC), Sydney, Australia. Anderson, D. R. 2008. Model bas ed inference in life sciences: A primer on evidence. Springer Science+Business Media, LLC., New York, New York USA Anadn, J. D., A. Gimnez I. Perez, M. Martin, and M. A. Esteve. 2006. Habitat selection by the spur thighed tortoise Testudo graeca in a multisuccessional landscape: implications for habitat management. Biodiversity and Conservation 15:2287 2299. Auffenberg, W. 1966. On the courtship of Gopherus polyphemus Herpetologica 22:113 117. Auffenberg, W., and R. Franz. 1982. The status and distri bution of the Gopher Tortoise ( Gopherus polyphemus ). Pages 99 126 in Bury, R. B., editor. North American Tortoises: Conservation and Ecology. U.S. Fish & Wildlife Service, Wildlife Research Report 12. Auffenberg, W., and J. B. Iverson. 1979. Demography of terrestrial turtles. Pages 541 569 in M. Harless and H. Morlock, editors. Turtles: perspectives and research. John Wiley & Sons, New York, New York. USA. Backus, L. K. 2003. The home range and response of Gopher Tortoises in central Florida to habitat manipulation by p rescribed burning. M.S. Thesis. University of Central Florida, Orlando, Florida USA Baldwin, R. F., A. J. K. Calhoun, and P. G. deMaynadier. 2006. Conservation planning for amphibian species with complex habitat requir ements: a case study using movements and habitat selection of the wood frog Rana sylvatica Journal of Herpetology 40:442 453. Barbier, E. B., E W. Koch, B. R. Silliman, S. D. Hacker, E. Wolanski, J. Primavera, E. F. Granek, S. Polasky, S. Aswani, L. A. C ramer, D. M. Stoms, C. J. Kennedy, D. Bael, C. V. Kappel, G. M. E. Perillo, and D. J. Reed. Coastal ecosystem based management with nonlinear ecological functions and values. Science 319:321 323.
86 Baskaran, L. M., V. H. Dale, and R. A. Efroymson. 2006. Habi tat modeling within a regional context: An example using gopher tortoise. American Midland Naturalist 155:335 351. Berry, K. H. 1986. Desert tortoise ( Gopherus agassizii ) relocations: implications of social behavior and movements. Herpetologica 42:113 125. Betts, M. G., A. S. Hadley, N. Rodenhouse, and J. J. Nocera. 2008. Social information trumps vegetation structure in breeding site selection by a migrant songbird. Proceedings of the Royal Society, Biological Science 275:2257 2263. Bird, L. B., L. C. Bra nch, and D. L. Miller. 2004. Effects of coastal lighting on foraging behavior of beach mice. Conservation Biology 18 :1435 1439. Block, W. M., and L. A. Brennan. 1993. The habitat concept in ornithology: Theory and applications. Current Ornithology 11: 35 91 Blomquist, S. M., and M. L. Hunter, Jr. 2009. A multi scale assessment of habitat selection and movement patterns by northern leopard frogs ( Lithobates [ Rana ] pipiens ) in a managed forest. Herpetological Conservation and Biology 4:142 160. Boglioli, M. D ., W. K. Michener, and C. Guyer. 2000. Habitat selection and modification by the Gopher Tortoise, Gopherus polyphemus in Georgia longleaf pine forest. Chelonian Conservation and Biology 3:699 705. Breininger, D. R., P. A. Schmalzer, and C. R. Hinkle. 1991 Estimating occupancy of Gopher Tortoise ( Gopherus polyphemus ) burrows in coastal scrub and slash pine flatwoods. Journal of Herpetology 25:317 321. Breininger, D. R., P. A. Schmalzer, and C. R. Hinkle. 1994. Gopher Tortoise ( Gopherus polyphemus ) density in coastal scrub and slash pine flatwoods in Florida. Journal of Herpetology 28:60 65. Burger, J., and M. Gochfeld. 1991. Burrow site selection by black iguana ( Ctenosaura similis ) at Palo Verde, Costa Rica. Journal of Herpetology 25:430 435. Bu rger, J., and W. A. Montevecchi. 1975. Nest site selection in the terrapin Malaclemys terrapin Copeia 1975:113 119. Burnham, K. P., and D. R. Anderson. 2001. Kullback Leibler information as a basis for strong inference in ecological studies. Wildlife Rese arch 28:111 119. Burnham, K. P., and D. R. Anderson. 2002. Model selection and multimodel inferences: a practical information theoretic approach. 2nd edition. Springer Verlag, New York, New York USA Burke, R. L. 1991. Relocations, repatriations, and tran slocations of amphibians and reptiles: Taking a broader view. Hepetologica 47:350 357.
87 Burke, R. L., and J. Cox. 1988. Evaluation and review of field techniques used to study and manage Gopher Tortoises. Pages 205 215 in Szaro, R. C., K. E. Severson, and D R. Patton, editors. Management of Amphibian, Reptiles, and Small Mammals in North America, 19 21 July 1988. US Department of Agriculture Forest Service, General Technical Report RM 166, Flagstaff, Arizona USA Burt, W. H. 1943. Territoriality and home range concepts as applied to mammals. Journal of Mammalogy 24:346 352. Butler, J. A., R. D. Bowman, T. W. Hull, and S. Sowell. 1995. Movements and home range of hatchling and yearling Gopher Tortoises, Gopherus polyphemus Chelonian Conservation and Biolog y 1:173 180. Cagle, F. R. 1939. A system of marking turtles for future identification. Copeia 1939:170 173. Citta, J. J., and M. S. Lindberg. 2007. Nest site selection of passerines: Effects of geographic scale and public and personal information. Ecology 88:2034 2046. Daudin, F. M. 1802. Histoire naturelle, gnrale et particulire, des reptiles. F. Dufart, Paris. Diemer, J. E. 1992a. Home range and movements of the tortoise Gopherus polyphemus in northern Florida. Journal of Herpetology 26:158 165. Diemer J. E. 1992b. Demography of the tortoise Gopherus polyphemus in northern Florida. Journal of Herpetology 26:281 289. Dodd, C. K. Jr. 1987. Status, conservation and management. Pages 478 513 in R. A. Seigel, J. T. Collins, and S. S. Novak, editors. Snakes: Ecology and Evolutionary Biology. Macmillan Publishing Co., New York, New York USA Dodd, C. K. Jr., and W. J. Barichivich. 2007. Movements of large snakes ( Drymarchon Masticophis ) in North Central Florida. Florida Scientist 70:83 94. Dodd, C. K. Jr., and R. A. Seigel. 1991. Relocation, repatriation, and translocation of amphibians and reptiles: Are they conservation strategies that work? Herpetologica 47:336 350. Dodd, C. K. Jr., W. J. Barichivich, and L. L. Smith. 2004. Effectiveness of a barrier wall and culverts in reducing wildlife mortality on a heavily traveled highway in Florida. Biological Conservation 118:619 631. Eisenberg, J. F. 1983. The Gopher Tortoise as a keystone species. Pages 1 4 in R. J. Bryant. and R. Franz, editors Proceedings of the 4th Annual Meeting of the Gopher Tortoise Council Florida State Museum, Gainesville, Florida USA
88 Epperson, D. M., and C. D. Heise. 2003. Nesting and hatchling ecology of Gopher Tortoises ( Gopherus polyphemus ) in southern Mississippi Journal of Herpetology 37:315 324. Erlinge, S., and M. Sandell. 1986. Seasonal changes in the social organization of male stoats, Mustela erminea : An effect of shift between two decisive resources. Oikos 47:57 62. Ernst, C. H., and J. E. Lovich. 2009. T urtles of the United States and Canada, 2nd edition. The Johns Hopkins University Press, Baltimore, Maryland USA Eubanks, J. O., W. K. Michener, and C. Guyer. 2003. Patterns of movement and burrow use in a population of Gopher Tortoises ( Gopherus polyph emus ). Herpetologica 59:311 321. FNAI (Florida Natural Areas Inventory) 2010. FNAI Guide to the natural communities of Florida: 2010 Edition. Florida Natural A reas Inventory. Available from w ww.fnai.org/natcom_accounts.cfm/ ( A ccessed September 2011). Fr etwell, S. D. 1972. Populations in a seasonal environment. Princeton University Press, Princeton, New Jersey, USA. Fretwell, S. D., and H. L. Lucas, Jr. 1970. On territorial behavior and other factors influencing habitat distribution in birds. Acta Biotheo retica 19:16 36. FWC (Florida Fish and Wildlife Conservation Commission). 2007. Gopher Tortoise Management Plan. Florida Fish and W ildlife Conservation Commission, Tallahassee, Florida USA Gallant, D., L. Vasseur, M. Dumond, E. Tremblay, and C. H. Berube 2009. Habitat selection by river otters ( Lontra canadensis ) under contrasting land use regimes. Canadian Journal of Zoology 87:422 432. Garmestani, A. S., H. F. Percival, K. M. Portier, and K. G. Rice. 2000. Nest site selection by the loggerhead sea tur of Herpetology 34:504 510. Graeter, G. J., B. B. Rothermel, and J. W. Gibbons. 2008. Habitat selection and movement of pond breeding amphibians in experimentally fragmented pine forests. Journal of Wildlife Ma nagement 72:473 482. Gregory, P. T., J. M. Macartney, and K W. Larsen. 1987. Spatial patterns and movements. Pages 366 395 in R. A. Seigel, J. T. Collins, and S. S. Novak, editors. Snakes: Ecology and Evolutionary Biology. Macmillan Publishing Co., New York, New York USA Groom, M. J., G. K. Meffe, and C. R. Carroll. 2006. Principles of Conservation Biology. Sinauer Associates, Inc., Sunderland, Massachusetts USA
89 Guzman, A., and P. R. Stevenson. 2008. Seed dispersal, habitat selection and movement pat terns in the Amazonian tortoise, Geochelone denticulata Amphibia Reptilia 29:463 472. Hailey A., and I. M. Coulson. 1996. Differential scaling of home range area to daily movement distance in two African tortoises. Canadian Journal of Zoology 74:97 102. H ansen, K. 1963. The burrow of the Gopher Tortoise. Quarterly Journal of the Florida Academy of Sciences 26:353 360. Harestad, A. S., and F. L. Bunnell. 1979. Home range and body weight a re evaluation. Ecology 60:389 402. Hermann, S. M., C. Guyer J. H. Waddle, and M. G. Nelms. 2002. Sampling on private property to evaluate population status and effects of land use practices on the Gopher Tortoise, Gopherus polyphemus Biological Conservation 108:289 298. Hilden, O. 1965. Habitat selection in bird s. Annales Zoologici Fennici 2:53 75. Hutchinson, G. E. 1957. Concluding remarks. Cold Spring Harbor Symposium on Quantitative Biology 2:415 427. Hutto, R. L. 1985. Habitat selection by nonbreeding, migratory land birds. Pages 455 476 in Cody, M. L., edito r. Habitat selection in birds. Academic Press, New York, New York USA IPCC (Intergovernmental Panel on Climate Change) 2007. Climate change 2007: the physical science basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergo vernmental Panel on Climate Change. S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, M. Tignor, and H. L. Miller, editors. Cambridge University Press, Cambridge, UK IUCN (International Union for Conservation of Nature and Natural Resourc es ) 2011. IUCN Red List of Threatened Species. Version 2011.1 Avail able from www.iucnredlist.org/ (A ccessed September 2011). Jackson, D. R. and E. G. Milstrey. 1989. The fauna of gopher tortoise burrows. Pages 86 98 in Gopher Tortoise Relocation Symposium Proceedings, Nongame Wildlife Program Technical Report 5. Florida Game and Fresh Water Fish Commission, Tallahassee, Florida USA Johnson, D. H. 1980. The com parison of usage and availability measurements for evaluating resource preference. Ecology 61:65 71. Johnston, G. R. 1996. Thermal ecology of the gopher tortoise ( Gopherus polyphemus ) in south central Florida. Ph. D. Thesis. University of Miami, Coral Gable s, Florida USA
90 Jones, J. C., and B. Dorr. 2004. Habitat associations of Gopher Tortoise burrows on industrial timberlands. Wildlife Society Bulletin 32:456 464. Kaczor, S. A., and D. C. Hartnett. 1990. Gopher Tortoise ( Gopherus polyphemus ) effects on soi ls and vegetation in a Florida sandhill community. American Midland Naturalist 123:100 111. Kazmaier, R. T., E. C. Hellgren, and D. C. Ruthven, III. 2001. Habitat selection by the Texas Tortoise in a managed thornscrub ecosystem. Journal of Wildlife Manage ment 65:653 660. Kellert, S. 1980. Public attitudes toward critical wildlife and natural habitat issues. U. S. Government Print Office, Washington, D.C. USA. Kent, D. M., M. A. Langston and D. W. Hanf. 1997. Observations of vertebrates associated with G opher Tortoise burrows in Orange County, Florida. Florida Scientist 60:197 201. Kushlan, J. A., and F. J. Mazzotti. 1984. Environmental effects on a coastal population of Gopher Tortoises. Journal of Herpetology 18:231 239. Landers, J. L., J. A. Garner, an d W. A. McRae. 1980. Reproduction of Gopher Tortoises ( Gopherus polyphemus ) in southwestern Georgia. Herpetologica 36:353 361. Landers, J. L., W. A. McRae, and J. A. Garner. 1982. Growth and maturity of the Gopher Tortoise in southwestern Georgia. Bulletin of Florida State Museum, Biological Sciences 27:810 110. Larsen, K. W. 1987. Movements and behavior of migratory garter snakes, Thamnophis sirtalis Canadian Journal of Zoology 65:2241 2247. Lima, S. L., and P. A. Zollner. 1996. Towards a behavioral ecol ogy of ecological landscape. Trends in Ecology and Evolution 11:131 135. Litvaitis, J. A., J. A. Sherburne, and J. A. Bissonette. 1986. Bobcat habitat use and home range size in relation to prey density. Journal of Wildlife Management 50:110 117. Lohoefen er, R., and L. Lohmeier. 1986. Experiments with Gopher Tortoise ( Gopherus polyphemus ) relocation in southern Mississippi. Herpetological Review 17:37, 39 40. 1998. Soil temperature, rock selection, and the thermal eco logy of the amphibaenian reptile Blanus cinereus Canadian Journal of Zoology 76:673 679. Mace, G. M., and P. H. Harvey. 1983. Energetic constraints on home range size. American Naturalist 121:120 132.
91 Madsen, T., and R. Shine. 1996. Seasonal migration o f predators and prey A study of pythons and rats in tropical Australia. Ecology 77:149 156. Mahan, C. G., R. H. Yahner. 1996. Effects of forest fragmentation on burrow site selection by the Eastern Chipmunk ( Tamias striatus ). American Midland Naturalist 136:352 357. Martin, T. E. 1998. Are microhabitat preferences of coexisting species under selection and adaptive? Ecology 79: 656 670. McNab, B. K. 1963. Bioenergetics and the determination of home range size. American Naturalist 97:133 141. McRae, W. A., J. L. Landers, and J. A. Garner. 1981. Movement patterns and home range of the Gopher Tortoise. American Midland Naturalist 106:165 179. Mushinsky, H. R., and D. J. Gibson, 1991. The influence of fire periodicity on habitat structure. Pages 237 259 in Bell, S. S., E. D. McCoy, and H. R. Mushinsky, editors. Habitat Structure: The Physical Arrangement of Objects in Space. Chapma n and Hall Ltd., London, UK Mushinsky, H. R., and E. D. McCoy. 1994. Comparison of Gopher Tortoise populations on islands and o n the mainland in Florida. Herpetologica 36:353 361. Mushinsky, H. R., E. D. McCoy, J. E. Berish, R. E. Ashton, Jr., And D. S. Wilson. 2006. Gopherus polyphemus Gopher Tortoise. Chelonian Research Monographs 3:350 375. Ovenpeck, J, T., and J. L. Weiss. 2 009. Projection of future sea level becoming more dire. Proceedings of the National Academy of Sciences, USA 106(51):21461 21462. Peters, K. A., and D. L. Otis. 2007. Shorebird roost site selection at two temporal scales: is human disturbance a factor? Jou rnal of Applied Ecology 44:196 209. Pike, D. A. 2006. Movement patterns, habitat use, and growth of hatchling tortoises, Gopherus polyphemus Copeia 2006:68 76. Pruner, R. A. 2010. Conservation and management of the Snowy Plover along the Florida gulf co ast: Habitat selection and the conseque nt reproductive performance. M. S. Thesis. University of Florida, Gainesville, Florida USA Pulliam, R. H., and B. J. Danielson. 1991. Sources, sinks, and habitat selection: A landscape perspective on population dynam ics. American Naturalist 137:S50 S66. R Development Core Team. 2010. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Ava ilable from www.R project.org/ (A ccessed June 2011).
92 Reinert, H. K. 1 991. Translocation as a conservation strategy for amphibians and reptiles: Some comments, concerns, and observations. Herpetologica 47:357 363. Rouse, J. D., R. J. Willson, R. Black and R. J. Brooks. 2011. Movement and spatial dispersion of Sistrurus cate natus and Heterodon platirhinos : Implications for interactions with roads. Copeia 2011:443 456. Row, J. R., and G. Blouin Demers. 2006. Thermal quality influences habitat selection at multiple spatial scales in milksnakes. EcoScience 13:443 450. Sandell, M 1989. The mating tactics and spacing patterns of solitary carnivores. Pages 164 182, in J. L. Gittleman, editor. Carnivore behavior, ecology, and evolution. Cornell University Press, Ithaca, New York USA Schwartz, T. S., and S. A. Karl. 2005. Populati on and conservation genetics of the gopher tortoise ( Gopherus polyphemus ). Conservation Genetics 6:917 928. Seaman, D. E., and R. A. Powell. 1996. An evaluation of the accuracy of kernel density estimators for home range analysis. Ecology 77:2075 2085. Sem litsch, R. D. 2000. Principles for management of aquatic breeding amphibians. Journal of Wildlife Management 64:615 631. Smith, L. L. 1992. Nesting ecology, female home range and activity patterns, and hatchling survivorship in the Gopher Tortoise ( Gopherus polyphemus ). M.S. Thesis, University of Florida, Gainesville, Florida USA Smith, R. B., D. R. Breininger, and V. L. Larson. 1997. Home range characteristics of radiotagged Gopher Tortoises on Kennedy Space Center, Florida. Chelonian Conservation and Biology 2:358 362. Smith, L. L., T. D. Tuberville, and R. A. Seigel. 2006. Workshop on the ecology, status, and management of the Gopher Tortoise ( Gopherus polyphemus ), Joseph W. Jones Ecological Research Center, 16 17 January 2003: Final results and recommendations. Chelonian Conservation and Biology 5:326 330. Svardson, G. 1949. Competition and habitat selection in birds. Oikos 1:157 74. Thomas, D.L., and E. J. Taylor. 1990. Study designs and tests for comparing resource use and availability. Journ al of Wildlife Management 54: 322 330. Tuberville, T. D., E. E. Clark, K. A. Buhlmann, and J. W. Gibbons. 2005. Translocation as a conservation tool: site fidelity and movement of repatriated Gopher Tortoises ( Gopherus polyphemus ). Animal Conservation 8:349 358.
93 USDA (U.S. Department of Agriculture). 2010. Soil Survey Geographic Database (SSURGO). United States Department of Agriculture, Natural Resources Conservation Service. Available from soils.usda.gov/survey/geography/ssurgo/ (accessed December 2010). USFWS (U.S. Fish and Wildlife Service). 2011. Species profile for Gopher tortoise ( Gopherus polyphemus ). Available from ecos.fws.gov/ecos/indexPublic.do ( A ccessed September 2011). Van Horne, B. 1983. Density as a misleading indicator of habitat quality. Jo urnal of Wildlife Management 47:893 901. Waddle, J. H., F. J. Mazzotti, and K. G. Rice. 2006. Changes in abundance of Gopher Tortoise burrows at Cape Sable, Florida. Southeastern Naturalist 5:277 284. Ward, R. M. P., and C. J. Krebs. 1985. Behavioral resp onses of lynx to declining snowshoe hare abundance. Canadian Journal of Zoology 63:2817 2824. Wilber, S. A. 2011. Florida under water: Global warming and rising sea levels. Creative Loafing Tampa. Available from cltampa.com/dailyloaf/archives/2011/05/31/f lorida under water global warming and rising sea levels/ ( A ccessed October 2011). Williams, K., J. Nichols, and M. Conroy. 2002. Analysis and management of animal populations. Academic Press, San Diego, California USA Wilson, J. D., C. T. Winne, M. E. Dorcas, and J. W. Gibbons. 2006. Post drought responses of semi aquatic snakes inhabiting an isolated wetland: Insights on different strategies for persistence in a dynamic habitat. Wetlands 26:1071 1078. Witherington, B. E., and R. E. Martin. 2000. Unders tanding, accessing, and resolving light pollution problems on sea turtle nesting beaches. Technical report TR 2, Florida Marine Research Institute, Florida Fish and Wildlife Conservation Commission, Tallahassee, Florida USA Wood, D. W., and K. A. Bjornda l. 2000. Relation of temperature, moisture, salinity, and slope to nest site selection in Loggerhead Sea Turtles. Copeia 2000:119 128. Yager, L. Y., M. G. Hinderliter, C. D. Heise, and D. M. Epperson. 2007. Gopher Tortoise response to habitat management by prescribed burning. Journal of Wildlife Management 71:428 434. Young, F.N., and C.C. Goff. 1939. An annotated list of the arthropods found in the burrows of the Florida Gopher Tortoise, Gopherus polyphemus (Daudin). Florida Entomologist 22:53 62. Zug, G. Z., L J. Vitt, and J. P. Caldwell. 2001. Herpetology: An introductory biology of amphibians and reptiles. Academic Press San Diego, California USA
94 BIOGRAPHICAL SKETCH Anthony Yin Kun Lau grew up in Hong Kong, China, and moved to the United State s in 2005 to pursue a un iversity education He graduated with an A.A. in z oology from Santa Fe College in 2007 and a B.S. in wildlife ecology and c onservation with honors from the University of Florida in 2009. He has been fascinated by reptiles and amphibian s since childhood after watching Sir David Attenborough spectacular nature documentary series. research interests include landscape and molecular ecology of reptiles and amphibians, conservation of threatened and endangered Asian chelo nians, and evolutionary biology of the order Testudines. In 20 11 h e received a Master of Science in wildlife ecology and conservation from the University of Florida under the mentorship of Drs. C Kenneth Dodd Jr. and Raymond Carthy.