ERROR LOADING HTML FROM SOURCE (http://ncf.sobek.ufl.edu//design/skins/UFDC/html/header_item.html)

Temporal patterns of burrow use by gopher tortoises (Gophexus Polyphemus) at the Ordway-Swisher Biological Station

Permanent Link: http://ncf.sobek.ufl.edu/NCFE004508/00001

Material Information

Title: Temporal patterns of burrow use by gopher tortoises (Gophexus Polyphemus) at the Ordway-Swisher Biological Station
Physical Description: Book
Language: English
Creator: Hayes, Forest
Publisher: New College of Florida
Place of Publication: Sarasota, Fla.
Creation Date: 2011
Publication Date: 2011

Subjects

Subjects / Keywords: Gopher Tortoise
Longleaf Pine
Gopherus Polyphemus
Prescribed Fire
Keystone Species
Genre: bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The gopher tortoise (Gopherus polyphemus) is prominent in sandhill communities of the southeastern United States and creates burrows that are used for shelter or habitat by hundreds of species. Gopher tortoises are difficult to monitor due to their reclusive nature. Infrared-triggered digital cameras allow for comprehensive monitoring of gopher tortoise burrow activity. During the summers of 2009 and 2010, six infrared-triggered camera traps were used to monitor gopher tortoise activity at burrow entrances. These cameras were used to monitor gopher tortoise burrows in order to ascertain their temporal activity patterns. In the 2010 study cameras were split between habitats treated with fire and those which were unburned for a period greater than one year. Gopher tortoise activity was almost exclusively diurnal, following a unimodal pattern. Most observations at the burrows were of gopher tortoises, followed by Florida mice and gopher frogs. The results of this study are consistent with previous studies, but further research is needed.
Statement of Responsibility: by Forest Hayes
Thesis: Thesis (B.A.) -- New College of Florida, 2011
Electronic Access: RESTRICTED TO NCF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE
Bibliography: Includes bibliographical references.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The New College of Florida, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Local: Faculty Sponsor: McCord, Elzie

Record Information

Source Institution: New College of Florida
Holding Location: New College of Florida
Rights Management: Applicable rights reserved.
Classification: local - S.T. 2011 H41
System ID: NCFE004508:00001

Permanent Link: http://ncf.sobek.ufl.edu/NCFE004508/00001

Material Information

Title: Temporal patterns of burrow use by gopher tortoises (Gophexus Polyphemus) at the Ordway-Swisher Biological Station
Physical Description: Book
Language: English
Creator: Hayes, Forest
Publisher: New College of Florida
Place of Publication: Sarasota, Fla.
Creation Date: 2011
Publication Date: 2011

Subjects

Subjects / Keywords: Gopher Tortoise
Longleaf Pine
Gopherus Polyphemus
Prescribed Fire
Keystone Species
Genre: bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The gopher tortoise (Gopherus polyphemus) is prominent in sandhill communities of the southeastern United States and creates burrows that are used for shelter or habitat by hundreds of species. Gopher tortoises are difficult to monitor due to their reclusive nature. Infrared-triggered digital cameras allow for comprehensive monitoring of gopher tortoise burrow activity. During the summers of 2009 and 2010, six infrared-triggered camera traps were used to monitor gopher tortoise activity at burrow entrances. These cameras were used to monitor gopher tortoise burrows in order to ascertain their temporal activity patterns. In the 2010 study cameras were split between habitats treated with fire and those which were unburned for a period greater than one year. Gopher tortoise activity was almost exclusively diurnal, following a unimodal pattern. Most observations at the burrows were of gopher tortoises, followed by Florida mice and gopher frogs. The results of this study are consistent with previous studies, but further research is needed.
Statement of Responsibility: by Forest Hayes
Thesis: Thesis (B.A.) -- New College of Florida, 2011
Electronic Access: RESTRICTED TO NCF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE
Bibliography: Includes bibliographical references.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The New College of Florida, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Local: Faculty Sponsor: McCord, Elzie

Record Information

Source Institution: New College of Florida
Holding Location: New College of Florida
Rights Management: Applicable rights reserved.
Classification: local - S.T. 2011 H41
System ID: NCFE004508:00001


This item is only available as the following downloads:


Full Text

PAGE 1

TEMPORAL PATTERNS OF BURROW USE BY GOPHER TORTOISES ( GOPHERUS POLYPHEMUS ) AT THE ORDWAY SWISHER BIOLOGICAL STATION BY FOREST P. HAYES A THESIS Submitted to the Division of Natural Sciences New College of Florida in partial ful fillment of the requirements for the degree Bachelor of Arts Under the sponsorship of Dr. Elzie McCord, Jr. Sarasota, Florida May, 2011

PAGE 2

ii Acknowledgment s I would like to thank my father, Dr. John P. Hayes, for instilling an appreciation and intere st in natu re from a young age and for his continued support and encouragement on all endeavors academic or otherwise. I am grate f ul to my thesis sponsor, Dr. Elzie McCord for his support and effort throughout the project and writing process. Special th anks to thesis committee member Dr. Meg Lowman, who has provided guidance and support throughout my undergraduate career at New College. Thank you to Dr. Sandra Gilchrist and Dr. Diana Weber as supportive committee members. I am grateful to Steve Coates for his support and providing access to a study site. I would like to thank Danielle Fasig for her support throughout the project and revision process. Finally, I would like to extend my thanks to the Ordway Swisher Biological Station for making this proj ect possible.

PAGE 3

iii Table of Contents Acknowledgments ii List of Figures iv List of Tables vi Abstract vii Chapter 1: Introduction 1 Literature Cited 1 4 Chapter 2: Temporal Patterns of Burrow Use by Gopher Tortoises 1 8 Abstract 28 Introduction 20 Material s and Methods 2 6 Results 3 3 Discussion 37 Literature cited 4 1 Chapter 3: Effects of Prescribed Fire on Temporal Activity of Gopher Tortoises 4 3 Abstract 4 3 Introduction 4 5 Methods 49 Results 58 Discussion 63 Literature Cited 6 7 Chapter 4: Conclusion 69 A ppendix 1 : Map of Ordway Swisher Bi ological Station Research Units 72 Appendix 2 : Research Protocol 73

PAGE 4

iv List of Figures Figure 1.1. Healthy undisturbed longleaf pine habitat in Alabama, image captured in 1902. Photograph from Frost 2006. ................................................................ 2Figure 1.2. Healthy longleaf pine habitat at th e Ordway-Swisher Bi ological Station. Image courtesy of John P. Hayes. ................................................................... 5Figure 1.3. Some of the less commonly obser ved species interacting with gopher tortoise burrows at the Ordway-Swisher Biological Stati on. A. An opossum investigates a burrow apron. B. A Fl orida Pine Snake is seen exiting the gopher tortoise burrow. C. A raccoon investigates the burrow entrance. ..... 7Figure 2.1. A gopher tortoise is seen enteri ng a burrow at the Ordway-Swisher Biological Station. ......................................................................................... 21Figure 2.2. A. A gopher frog sits at the entran ce of a gopher tortoise burrow. B. Two Florida mice are visible da shing into a gopher tortoise burrow. Both images were captured at the Ordway-Swi sher Biological Station. ............................ 24Figure 2.3. A. An inactive gopher tortoise burrow at the OSBS. B. An active gopher burrow at the OSBS. ...................................................................................... 29Figure 2.4. Calibration of the infrared trail monitors is checked using a one inch piece of PVC pipe. ...................................................................................................... 30Figure 2.5. Active infrared trail monitors are enclos ed in zip-lock bags and staked to the ground to prevent movement; meanwhile th e triggers are ch ecked using a one inch piece of PVC at OSBS. .......................................................................... 32Figure 2.6. A digital camera in a w eatherproof box is set up to capture pictures of activity at the entrance of a gopher tortoise burrow. ..................................... 32Figure 2.7. Number of gopher tortoi se observations sandhill pine habitats relative to the time of day observations occurred. ................................................................ 35Figure 2.8. Number of observati ons of different lengths of time (minutes) gopher tortoises spent in the burrow between exits. .................................................. 35Figure 2.9. Number of observati ons of different lengths of time (minutes) gopher tortoises spent out of the burrow be fore returning.time (Figure 2.11). ......... 35Figure 2.10. Number of observations of gopher fr ogs in relation to time of day. .......... 36Figure 2.11. Number of observations of class In secta in relation to time of day. .......... 36Figure 2.12. Number of observations of Florida mice in relation to time of day. .......... 37Figure 3.1. A gopher tortoise burrow in traditi onal longleaf pine habitat undergoing restoration at the Ordway-Swi sher Biological Station. ................................. 46Figure 3.2. A drip torch is used to spread fire in a longleaf pine habitat at the OrdwaySwisher Biological Station. Photogra ph courtesy of John P. Hayes. ........... 51Figure 3.3. Active infrared trail monitors are enclos ed in zip-lock bags and staked to the ground to prevent movement; meanwhile th e triggers are ch ecked using a one inch piece of PVC at OSBS. .......................................................................... 54

PAGE 5

v Figure 3.4. A 1 m2 grid is used to survey ground c over at a recently burned unit. ......... 56 Figure 3.5. Ground cover percentages are recorded using a 1 m2 grid at the OrdwaySwisher Biological Station. Photogra ph courtesy of Danielle Fasig ............ 56 Figure 3.6. Percent ground and vegetation cove r averaged for burned (blue) and unburned (red) habitats. Error bars repr esent 95% confidence intervals. .... 60 Figure 3.7. Number of gopher tortoi se observations in burned habitats relative to the time of day observations occurred. ................................................................ 62Figure 3.8. Number of gopher tortoise observations in unburned ha bitats relative to the time of day observations occurred. ................................................................ 62Figure 3.9. A gopher tortoise in a burned unit ba sks in the evening sun. Image captured at 5:21 h:m, at the Ordway-Swi sher Biological Station. ............................... 65

PAGE 6

vi List of Tables Table 2.1. Frequency of vertebrate species observed during the study over the summer of 2009 at the Ordway Swisher Biological Station. ................................ ....... 33 Table 3.1. Frequency of vertebrate species obs erved during the study in both burned and unburned habitats during the summer of 2010 at the Ordway Swisher Biological Station. ................................ ................................ ........................... 59 Table 3.2. Statistical values for gopher tortoise activity. ................................ ................ 59 T able 3.3. Statistical values for ground cover variables ................................ ................. 61

PAGE 7

vii TEMPORAL PATTERNS OF BURROW USE BY GOPHER TORTOISES ( GOPHERUS POLYPHEMUS ) AT THE ORDWAY SWISHER BIOLOGICAL STATION Forest Hayes New College of Florida, 2011 ABS TRACT The gopher tortoise ( Gopherus polyphemus ) is prominent in sandhill communities of t he southeastern United States and create s burrows that are used for shelter or habitat by hundreds of species. Gopher tortoises are difficult to monitor due to the ir reclusive nature. I nfrared triggered digital cameras allow for comprehensive monitoring of g opher tortoise burrow activity. During the summers of 2009 and 2010, six infrared triggered camera traps were used to monitor gopher tortoise activity at burrow entrances. These cameras were used to monitor gopher tortoise burrows in order to ascertain the ir temporal activity patterns In the 2010 study c ameras were split between habitats treated with fire and those which were unburned for a period greater than one year Gopher tortoise activity was almost exclusively diurnal following a unimodal pattern. Most observations at the burrows were of gopher tortoises, followed by Florida mice and gopher frogs. The results of this study are consistent with previous studies, but further research is needed. Dr. Elzie McCord, Jr. Division of Natural Sciences

PAGE 8

1 Chapter 1 Introduction Gopher tortoise ( Gopherus polyphemus Daudin) burrows were once a common landscape fixture in longleaf pine ( Pinus palustris Mill. ) habitats of the southeastern United States (Figure 1.1) (Breininger et al. 1994). The burrows are home to many species other than the gopher tortoise and provide shelter for hundreds more (Witz et al. 1991). The dramatic decline of sandhill habitats du e to aggressive development, agriculture, and fire suppression regimes has left only a fraction of these forests intact (Frost 2006). Compounding factors have caused gopher tortoise populations to decline by close to 80 % c over the past several decades ( McCoy et al. 2006). Due to slow maturation and low reproductive success, gopher tortoise populations will be slow to recover even with additional conservation efforts (Epperson and Heise 2003). Ongoing research strives to provide more information about g opher tortoises and their activity in order to better preserve the habitat they depend on and the animals that are dependent on their burrows. Gopher tortoises have an average carapace length of 25 cm and weigh approximately 4 kg T hey are naturally long lived with lifespans of approximately 40 60 years (Landers et al. 1980). Female gopher tortoises are generally larger and heavier than males (Jodice et al. 2006, McRae et al. 1981). Gopher tortoise diet s consist of a wide range of plants including the fol iage of grass like plants, flesh fruit, and legumes, and are considered to be in between specialist and generalist feeders (Birkhead et al. 2005, MacDonald and Mushinksy 1988). Gopher tortoises also consume non plant material

PAGE 9

2 Figure 1.1. Healthy undistu rbed longleaf pine habitat in Alabama, image captured in 1902. Photograph from Frost 2006.

PAGE 10

3 such as charcoal and insects, which are thought to provide additional nutrients (MacDonald and Mushinsky 1988). Consumption of seeds is thought to aid in plant di spersal, as seeds remain viable after ingestion (Brikhead et al. 2005). Wiregrass ( Astrida stricta Michx. ) accounts for a large percentage of gopher tortoise diets in habitats where it occurs. Adult tortoises are better able to digest it and ; therefore consume more wiregrass than juveniles (MacDonald and Mushinksy 1988). On average gopher tortoises reach maturity at 10 15 years of age ; however the age can vary by region. Females usually lay one clutch of eggs per year with an average clutch size of aro und 6 (Diemer 1986). Larger males tend to have greater reproductive success, while larger females are better able mates (Moon et al. 2006). Eggs are often laid in the aprons of burrows, although they may also be laid farther from the entrance (Butler and Hull 1996, Pike and Seigel 2006). Survival rates of hatchlings are low, but are variable depending on region and local predators (Epperson and Heise 2003, Pike and Seigel 2006). Fire ants ( Solenopsis invicta Westwood) an invasive exotic species, now co ver the entire range of gopher tortoises and are thought to contribute to high hatchling mortality rates (Wetterer and Moore 2005 ) Age of gopher tortoises is strongly correlated with size and can be used to easily identify younger individuals (Moon et al. 2006). Juveniles are at a higher risk of predation due in part to their in ability to escape predators such as raccoons, as successfully as adults (Wetterer and Moore 2005). Juvenile diets are similar to adults, but differ slightly due to a lesser abilit y to digest some species plant including wiregrass (Mushinksy et al. 2003, MacDonald and Mushinksy 1988). The duration of time juvenile

PAGE 11

4 gopher tortoises spend out of the burrow is also shorter than that of adults, possibly due to smaller stomachs (MacDona ld and Mushinksy 1988). Longleaf pine ( Pinus palustris Mill.) ecosystems are characterized by widely spaced trees a sparse midstory of turkey oak (Figure 1.1 and 1.2) ( Quercus laevis Walter), and a dense understory of wiregrass ( Astrida stricta ) and othe r herbaceous plants (Frost 2006, Drew et al. 1998). Historically the gopher tortoise is associated with the distribution of longleaf pine forests and sandhill ecosystems throughout the United States (USFWS 1990). Sandhill ecosystems are found throughout the southeastern United States but have declined greatly due to increased agriculture, development and fire suppression (Frost 2006). The gopher tortoise requires a habitat with well drained, loose soil, such as sandhill habitats (McRae et al. 1981). A lthough many are not representative of the typical habitat of the gopher tortoise, every county in Florida is home to gopher tortoises even in habitats that are not well drained (Breinnger et al. 1994). Fire plays many important roles in sandhill habita ts and prevents the succession to hardwood forests. The importance of fire in these ecosystems is readily apparent through the examination of areas in which it has been suppressed. Regular occurrences of fire every 1 10 years prevents the growth of an oa k midstory, leading to the eventual succession of the habitat (Drew et al. 1998). Fires occurring in long unburned areas are fueled by increased deposits of detritus, humus and the oak midstory, leading to canopy fires in longleaf pines (Frost 2006). Reg ular fires also promote the proliferation of wiregrass and other understory herbaceous plants that characterize longl eaf pine habitats

PAGE 12

5 Figure 1.2. Healthy longleaf pine habitat at the Ordwa y Swisher Biological Station. Image courtesy of John P. Hayes. Figure 1.2. Healthy longleaf pine habitat at the Ordway Swisher Biological Station. Image courtesy of John P. Hayes. Figure 1.2. Healthy longleaf pine habitat at the Ordway Swisher Biolog ical Station. Image courtesy of John P. Hayes. Figure 1.2. Healthy longleaf pine habitat at the Ordway Swisher Biological Station. Image courtesy of John P. Hayes.

PAGE 13

6 (Varner et al. 2005). Fires may also temporarily eliminate insects from an area, allowing plants to rec over (Knight and Holt 2005). Gopher tortoises are considered to be a keystone species (Eubanks et al. 2003). A keystone species is defined as a species that has a disproportionate impact on its environment in proportion to its relative abundance (Paine 19 69, Paine 1995). Species which are reliant on gopher tortoise burrows also face population declines, or extinction from an area, with a loss of gopher tortoise populations. The gopher tortoise is also an ecosystem engineer. An ecosystem engineer is an o rganism that changes the availability of resources for other species by causing physical state changes in biotic or abiotic materials (Jones et al. 1994). Gopher tortoises have large impact s on their surrounding ecosystems through the creation of burrows which alter the flora and provide habitat and shelter for many species (Figure 1.3) Through this process they modify, create, and maintain habitats. Active gopher tortoise burrows lead to higher levels of plant biodiversity directly surrounding the burr ows (Kaczor and Harnett 1990). Gopher tortoises also encourage the growth and proliferation of wiregrass (Kaczor and Harnett 1990). Wiregrass contributes to the cool, quick burning, less destructive fires that are necessary for the preservation of longle af pine habitats (Walters et al. 1994). The decline in the gopher tortoise population is a particularly large concern because of the number of vertebrates and invertebrates that are dependent on the burrows (Alexy et al. 2003). Over 360 other species are dependent on gopher tortoise burrows for refuge or habitat (Eubanks et al. 2003). Gopher tortoise burrows are especially important as refuges during fires from heat and flames (Franz 1984, Russel et al. 1999).

PAGE 14

7 Figure 1.3 Some of the less commonly obse rved species interacting with gopher tortoise burrows at the Ordway Swisher Biological Station. A. An opossum investigates a burrow apron. B. A Florida Pine Snake is seen exiting the gopher tortoise burrow. C. A raccoon investigates the burrow entranc e.

PAGE 15

8 The Florida mouse ( Podomys floridanus Chapman) a species of special concern, is restricted to Florida xeric upland habitats and dwells solely in gopher tortoise burrows (Layne and Jackson 1994), closely linking its survival to that of the gopher torto ise. The Florida mouse makes small lateral tunnels, which are known as chimneys, branching from the main burrow. They also use the burrows as habitat, protection from predators, and a source of food. Densities of active gopher tortoise burrows are closel y correlated with population densities of the Florida mouse (Layne and Jackson 1994). Gopher frogs ( Rana capito LeConte) are also dependent on gopher tortoise burrows as habitat and for migration routes during mating season (Franz 1984). Gopher frogs reuse the same burrows over multiple years on their migration routes (Witz et al 1991). Many other species, including the Florida pine snake ( Pituophis melanoleucus mugitus Barbour) and black racer ( Coluber constrictor L. ) are closely associated with and often found inside gopher tortoise burrows (Eubanks et al. 2003). Because many species are dependent on habitat created by the gopher tortoise, the impact of a declining population is magnified. T hree major reasons that are attributed to the decline in gopher tortoise populations are human predation, habitat degradation through fire suppression and habitat destruction. By the mid % decline in gopher tortoise populations in Florida over the course of a century (Diemer 1986). Upper respiratory tract infections are thought to contribu te to gopher tortoise mortality and may prevent recovery of gopher tortoise populations (McLaghlin et al. 2000). Invasive species such as the fire ant pose a threat to juvenile gop her tortoises

PAGE 16

9 (Pike and Seigel 2006). Box turtles respond to fire ants by withdrawing into their carapace but are unable to seal them out, leaving them vulnerable to the arrival of more ants. It has been suggested that the same could be an issue for adult gopher tortoises (Welterer and Moore 2005). Fire suppression in sandhill ecosystems leads to succession of longleaf pine habitats to hardwood forests, reducing the desirability of these habitats for gopher tortoises (Yeger et al. 2007, Jones and Dorr 200 4). The succession of longleaf pine habitats leads to denser canopy cover (Glitzenstein et al. 1995). Denser canopies are associated with increased burrow abandonment by gopher tortoises and lower overall populations (Aresco and Guyer 1999). Gopher tort oise burrow densities increase closer to roads in an apparent effort to decrease canopy cover near the burrow entrance. They also adopt an elongated range due to traveling a long the road side (Diemer 1992b Baskaran et al. 2006). The multitude of threats facing gopher tortoise populations combined with staggering losses of habitat, contribute to the difficulty in restoring populations to their original levels. Gopher t ortoise s are slow to mature and have low reproductive success which further exacerbate s the issue of population restoration (Waddle et al. 2006). The long life span of gopher tortoises and high visibility within their habitats makes the loss of populations harder to detect over shorter periods of time. Gopher tortoise populations have since been recognized as species of special concern and have received various types of federal and state protection status. Gopher tortoises have been listed as threatened in the state of Florida since 1999. The gopher tortoise is currently federally listed as a threatened species in Alabama, Louisiana and Mississippi (McCoy et al. 2006). The

PAGE 17

10 gopher tortoise is listed as a Species of Special Concern in Florida (Diemer, 1992 A). Even with federal listings and measures taken to protect tortoise populations, their numbers have continued to decline over the past decade (McCoy et al. 2006). Numerous methods have been developed as a means of studying gopher tortoises and their habitats. Estimates of abundance have received special attention due to the crucial r ole gopher tortoises play in sandhill ecosystems. Using transects to estimate abundance has returned overestimates as well as underestimates in different habitats (Nomani et al. 2008). Manual observation of gopher tortoises for abundance estimates is cap ital intensive, requiring large amounts of time to yield accurate results (Birkhead et al. 2005). Many methods of trapping and monitoring gopher tortoises have been used to study their behavior; each has a unique set of advantages and disadvantages. Bucke t traps may be used at the entrance of gopher tortoise burrows in order to estimate occupancy (Breininger et al. 1991). Gopher tortoises may also be pulled out of b urrows using hooks (Diemer 1992a ). One of the most accurate methods of examining gopher to rtoise occupancy and commensals involves complete excavation of burrows using a tractor although this practice has been replaced by the use of burrow cameras more recently (Witz et al. 1991). The disadvantage of these methods is the additional stress pla ced on the animal through capture of the individual s or even excavation of the burrow. Although manual observation is less stressful on the animals, it is time and cost intensive when properly conducted (Nomani et al. 2008). Burrow cameras provide a uni que view into gopher tortoise burrows and may be used to assess gopher tortoise presence (Breininger et al. 1991). The disadvantage of burrow cameras is the disturbance

PAGE 18

11 they cause when tortoises enter the burrow and the inability to accurately depict use of gopher tortoise burrows by short term visitors, particularly mammals. Monitoring of gopher tortoise movement and activity has traditionally been done by surveys conducted manually or by attaching radio transmitters to tortoise carapaces (Eubanks et al. 2003). Radio transmitters provide accurate, detailed depictions of gopher tortoise movement (Eubanks et al. 2003). The process of installing transmitters is invasive as physical capture of the tortoise is required, which may cause additional stress on th e animal (Epperson and Heise 2003). Radio transmitters also fall off over time limiting the potential duration of the study (Eubanks et al. 2003). Although it is possible to identify movement patterns of gopher tortoise populations through manual observ ation (McRae et al. 1981) it is generally not possible to document every movement or use of the burrow by nocturnal commensals. Sticks may be placed in front of gopher tortoise burrows as a crude method of checking for burrow occupancy but may misreprese nt activity due to the presence of commensals (Breininger et al. 1991). A final approach that has only recently been applied to the monitoring of gopher tortoise activity is the use of infrared triggered cameras (Alexy et al. 2003). Although numerous stud ies have used infrared triggered cameras to monitor large mammals, few studies have used the technology to study reptiles. Infrared triggered cameras have the benefit of comprehensively monitoring both diurnal and nocturnal activity at the entrance of gop her tortoise burrows. Images captured by the cameras can allow for further description of gopher tortoise activity, as the direction of movement is usually discernable. Infrared triggered cameras also present a less invasive approach than burrow cameras or any methods which require tortoise capture tortoise. Infrared triggers predate

PAGE 19

12 digital cameras but have become much more practical since the proliferation of digital SLR cameras. Digital cameras allow for many times more pictures than film cameras, ar e more economical, and facilitate immediate data analysis. Adult gopher tortoise burrows are on average 4.7 m long and 2.0 m deep, declining at an angle of 29 o (Witz et al. 1991). The shape of the burrow strongly relates to size of the tortoise, with a wi dth roughly equal to the carapace length of the tortoise, allowing it to turn around at any point in the burrow (Doonan and Stout 1994). Gopher tortoises do not usually travel far from their burrows and about 95 % of feeding activity occurs within 30 m of the burrows (Alexy et al. 2003). Gopher tortoises may travel farther in order to find mates but generally stay within one kilometer of their burrow (McRae et al. 1981). Courting and mating generally occurs just outside of the females burrow (Butler and H ull 1996). Female gopher tortoises have been observed to use fewer burrows in comparison to male tortoises in the same habitat, where mean burrow use was 5.2 and 10.0 respectively (Eubanks 2003). Gopher tortoise use of burrows is also thought to be stro ngly related to time of year (McRae et al. 1981). Gopher tortoises use the most burrows in mating season during the spring and summer and the fewest burrows during the winter when the cold weather leads to inactivity (Diemer 1992 b ). Gopher tortoises have been observed to abandon burrows at an average rate of 22% per year (Aresco and Guyer 1999). Burrow abandonment is often associated with changes in the forest canopy or other factors that reduce the desirability of the habitat.

PAGE 20

13 Burrow densities are high er in open habitats and have lower rates of abandonment (Aresco and Guyer 1999). Activity of gopher tortoises been correlated with ambient temperature (Alexy et al. 2003). During periods of higher temperature, gopher tortoises may adopt a bimodal activity pattern in which there are two peak times of activity throughout the day, to avoid extreme temperatures (McRae et al. 1981). The remainder of the year gopher tortoises adopt a unimodal pattern with a single peak time of activity (McRae et al. 1981). Activity drops greatly during the winter and observations of gopher tortoises outside the burrow dramatically decrease ( Diemer 1992b ). T hreatened gopher tortoises have received recognition at state and federal levels. The following chapters present beha vioral information on gopher tortoises and their commensals using infrared triggered digital cameras. Much of gopher tortoise biology is still poorly understood due to their reclusive nature and difficulty of comprehensive monitoring Gopher tortoises ar e associated closely with habitats that burn regularly, but little is known about the fire ecology of gopher tortoises This study seeks to examine the temporal activity patterns of gopher tortoises and their burrow commensals.

PAGE 21

14 Literature Cited Alexy, K. J., K.J. Brunjes, J.W. Gasset and K.V. Miller. 2003. Continuous remote monitoring of gopher tortoise burrows use. Wildlife Society Bulletin 31(4): 1240 1243. Aresco, M.J. and C. Guyer. 1999. Burrow abandonment by gopher tortoises in slash pine plantations of the Conecuh National Forest. The Journal of Wildlife Management 63(1): 26 35. Baskaran, L.M., V.H. Dale, R.A. Efroymson, and W. Birkhead. 2006. Habitat modeling within a regional context: an example using gopher tortoise. American Midland Naturalist 1 55:335 351. Birkhead, R.D., C. Guyer, S.M. Hermann, and W.K. Michener. 2005. Patterns of folivory and seed ingestion by gopher tortoises ( Gopherus polyphemus ) in a southeastern pine savanna. American Midland Naturalist 154: 143 151. 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 forests. Journal of Herpetology 25(3): 317 321. Breininger, D.R., P.A., Schmalzer, and C.R. Hinkle. 1994. Gopher tort oise ( Gopherus polyphemus ) densities in costal scrub and slash pine flatwoods in Florida. Journal of Herpetology 28: 60 65. Butler, J.A. and T.W. Hull. 1996. Reproduction of the tortoise, Gopherus polyphemus in Northeastern Florida. Journal of Herpetolog y 30(1): 14 18. Diemer, J.E. 1986. The ecology and management of the gopher tortoise in the southeastern United States. Herpetologica 42: 125 33. Diemer, J.E. 1992a Demography of the tortoise Gopherus polyphemus in Northern Florida. Journal of Herpetolo gy 26(3): 281 289. Diemer, J.E. 1992b Home range and movements of the tortoise Gopherus polyphemus in Northern Florida. Journal of Herpetology 26(2): 158 165. Doonan T.J., and I.J. Stout. 1994. Effects of gopher tortoise ( Gopherus polyphemus ) body size on burrow structure. American Midland Naturalist. 2: 273 280. Drew, M.B., L.K. Kirkman, and A.K. Gholson Jr. 1998. The vascular flora of Ichauway Baker County, Georgia: a remnant longleaf pine/wiregrass ecosystem. Castanea 63(1): 1 24.

PAGE 22

15 Epperson, D.H., an d C.D. Heise. 2003. Nesting and hatchling ecology of the gopher tortoises ( Gipherus polyphemus ) in Southern Mississippi. Journal of Herpetology 27(2): 315 324. Eubanks, J.O., W.K. Michener, and C. Guyer. 2003. Patterns of movement and burrow use in a popu lation of gopher tortoises ( Gopherus polyphemus ). Herpetologica 59(3): 311 321. Franz, R. 1986. The Florida gopher frog and the Florida pine snake as burrow associates of the gopher tortoise in Northern Florida. Pp. 16 20 in D.R. Jackson and R.J. Bryant (eds.). The gopher tortoise and its community. Proceedings from the 5 th annual meeting of gopher tortoise council, Florida State Museum, Gainesville. Frost, C. 2006. History and future of the longleaf pine ecosystem. Pages 9 48 in S. Jose, E. J. Jokel a, and D. L. Miller, editors. The longleaf pine ecosystem. Springer, New York, New York, USA. Glitzenstein, J.S., W.T. Platt, and D.R. Streng. 1995. Effects of fire regime and habitat on tree dynamics in North Florida longleaf pine savannas. Ecological So ciety of America 65(4): 441 476. 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. Jodice, P.G., D.M. Epperson, and G.H. Visser. 2006. Daily energy expenditure in free ranging gopher tortoises. American Society of Ichthyologists and Herpetologists 2006(1): 129 136. Jones C.G., J.H. Lawton, and M. Shachak. 1994. Organisms as Ecosystem Engineers. Oikos 69: 373 386. Jones, J.C. and B. Dorr. 2004. Habitat associations of gopher tortoise burrows on industrial timberland. Wildlife Society Bulletin 32(2): 456 464. Kaczor, S.A, and D.C. Hartnett. 1990. Gopher tortoise ( Gopherus polyphemus ) effects on soils and vegetation in a Florida sandhill community. American Midland Naturalist 123(1): 110 111. Knight, T.M., and R.D. Holt. Fire generates spatial gradients in herbivory: an example from a Florida sandhill ecosystem. Ecology 86( 3) 587 593. Landers, J.L., J.A. Garner, and W.A. McRae. 1980. Reproduction of Gopher tortoises ( Gopherus polyphemus ) in southwestern Georgia. Herpetologica. 34: 353 361.

PAGE 23

16 Layne, J.N. and R.J. Jackson. 1994. Burrow use by the Florida mouse ( Podomys florida nus ) in South Central Florida. American Midland Naturalist 131(1): 17 23. MacDonald, L.A., and H.R. Mushinsky. 1988. Foraging ecology of the gopher tortoise, Gopherus polyphemus in sandhill habitat. Herpetologica 44(3): 345 353. McCoy, E.D., H.R. Mushin sky, and J. Lindzey. 2006. Declines of the gopher tortoise on protected lands. Biological Conservation 128: 120 127. McLaughlin, G.S., E.R. Jacobson, D.R. Brown, C.E. McKenna, I.M. Schumacher, H.P. Adams, M.B. Brown, and P.A. Klein. 2000. Pathology of up per respiratory tract disease of gopher tortoises in Florida. Journal of Wildlife Diseases 36(2) 272 283. McRae, W.A., J.L. Landers, and J.A. Garner. 1981. Movement patterns and home range of the gopher tortoise. American Midland Naturalist. 106(1): 165 179. Moon, J.C., E.D. McCoy, H.R. Mushinsky, S.A. Karl. 2006. Multiple paternity and breeding system in the topher tortoise, Gopherus polyphemus Journal of Heredity 97(2): 150 157. Mushinsky, H.R., T.A. Stilson, and E.S. McCoy. Diet and dietary preferen ce of the juvenile gopher tortoise ( Gopherus polyphemus ). Herpetologica 59(4) 475 483. Nomani, S.Z., R.R. Carthy, and M.K. Oli. 2008. Comparison of methods for estimating abundance of gopher tortoises. Applied Herpetology 5: 13 31. Paine, R.T. 1969. A no te on trophic complexity and community stability. American Midland Naturalist 103: 91 93. Paine, R.T. 1995. A conversation on refining the concept of keystone species. Conservation Biology 9: 962 964. Pike, D.A., and R.A. Seigel. 2006. Variation in hatch ling tortoise survivorship at three geographical localities. Herpetologica 62(2) 125 131. Russell, K.R., D.H. Van Lear, and D.C. Guynn Jr. 1999. Prescribed fire effects on herpetofauna: review and management implications. Wildlife Society Bulletin 27(2): 374 384. U.S. Fish and Wildlife Service. 1990. Gopher tortoise ( Gopherus polyphemus ) recovery plan. Prepared by Wendell A. Neal, U.S. Fish and Wildlife Service, Jackson, MI. 28 pp.

PAGE 24

17 Varner J.M., D.R. Gordon, F.P. Putz, and J.K. Hiers. 2005. Restoring fire to long unburned Pinus palustris ecosystems: novel fire effects and consequences for long unburned ecosystems. Restoration Ecology 13(3): 536 544. Waddle, J.H., F.J. Mazzotti, and K.G. Rice. 2006. Changes in abundance of gopher tortoise burrows Cape S able, Florida. Walters, T.W., D.S. Decker Walters, and D.R. Gordon. 1994. Restoration considerations for wiregrass ( Aristida stricta ): allozymic diversity of populations. Conservation Biology 8(2): 581 585. Wetterer, J.K. and J.A. Moore. 2005. Red impor ted fire ants (Hymenoptera : Formicidae ) at gopher tortoise (Testudines: Gopherus polyphemus ) burrows. Florida Entomologist 88(4): 349 354. Witz, B.W., D.S. Wilson, and M.D. Palmer. 1991. Distribution of Gopherus polyphemus and its vertebrate symbionts in t hree burrow categories. American Midland Naturalist 126(1): 152 158. Yager, Y.L., 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(2): 428 434.

PAGE 25

18 Chapter 2 Temporal Patterns of Burrow Use by Gopher Tortoises Abstract Our understanding of the habitat ecology of the gopher tortoise ( Gopherus polyphemus Daudin) has increased substantially over the past several decades, however, the influences of habitat m anipulation, fire, and habitat quality on the behavioral ecology of tortoises remains poorly understood. Moreover, the temporal patterns of burrow use are poorly understood, as constant surveillance has previously been impractical. Use of infra red triggered cameras combined with digital photographic technology has proved a useful tool for comprehensive monitoring of gopher tortoise burrows, providing an opportunity to examine and assess the activity of gopher tortoises during all hours. Nine bur rows were periodically monitored from June to August for 58 days total Infrared camera traps were placed in front of burrows that were visually determined to have been recently active. The infrared trigger mechanism was placed so that individuals using t he burrow would break the beam when entering or exiting the burrow. The 6 infrared triggers used were rotated between burrows on a biweekly basis. Time stamps embedded in the images were recorded and analyzed in order to examine the temporal patterns of gopher tortoises. Eight species of vertebrates were photographed interacting with burrows; the most commonly observed species were gopher tortoises, gopher frogs ( Rana capitoaesopus LeConte ) and Florida mice ( Podomys floridanus Chapman) in this order. The majority of

PAGE 26

19 photos captured were of gopher tortoises. Gopher tortoises were active almost exclusive ly during daylight hours, with less than 90% of gopher tortoise observations falling between 0700h and 1800h. The mean length of time spent out of the b urrow was 24.67 minutes and mean length of time between exits during the day was 64.64 minutes.

PAGE 27

20 Introduction Gopher tortoises ( Gopherus polyphemus Daudin) play an important role in their habitats and considered to be a keystone species as well as an ecos ystem engineer (Figure 2.1) (Eubanks et al. 2003, Jones et al. 1994). Gopher tortoises are recognized as an important species due largely to their chiastic burrows. More than 300 species are dependent on gopher tortoise burrows for either refuge or habit at (Eubanks et al. 2003). Gopher tortoises are take many years to reach reproductive maturity, generally reaching it at around 15 20 years of age. Clutches are typically laid once per year, containing an average of five eggs and, of those that hatch, sur vival rates are low (Epperson and Heise 2003). The combination of slow maturity, low reproductive rates, and low survival of hatchlings contributes to slow gopher tortoise population growth and recovery. Therefore, they have difficulty recovering from lo ng term stressors on their environment. The multitude of species dependent on gopher tortoises and their burrows necessitates special attention to gopher tortoise populations. The decline in gopher tortoise populations by as much as 80 percent over the p ast century further demonstrates the importance of studying this species ( Diemer 1986). A typical gopher tortoise burrow spirals downward at an angle of 29 o extending for 4.7 meters, eventually reaching a depth of 2 meters (Witz et al. 1991). Gopher tor toise burrow use varies according to gender and time of year (Eubanks 2003, McRae et al. 1981). The majority of gopher tortoise activity is d ue to thermoregulation near the burrow entrance, in addition 95% of foraging activity has been observed within a 3 0 meter radius from the burrow entrance. Gopher tortoise burrow abandonment has been observed at slightly under 25 %of burrows per year, though it is known to vary according

PAGE 28

21 Figure 2.1 A gopher tortoise is seen entering a burrow at the Ordway Swisher Biologic al Station.

PAGE 29

22 to the quality of habitat and has been positively correlated with increased canopy coverage (Aresco and Guyer 1999). Gopher tortoises prefer habitats with loose and well drained soils and traditionally have been associated with the distribution of longleaf pine ( Pinus palustris Mill.) habitats t hroughout the United States (McR ae e t al. 1981, USFWS 1990). The role of gopher tortoises as ecosystem engineers plays an important part in the maintenance of sandhill habitats. Gopher tortoises have an impact on understory composition through their consumption of vegetation and seeds, dep osition of feces, and creation of open areas (Kaczor and Hartnett 1990). Alteration of the understory leads to the increased proliferation of wiregrass. Wiregrass understories play a crucial role in the spread of fire in sandhill ecosystems. Wiregrass b urns quickly and spreads fire easily; wiregrass fires burn at a lower temperature and for shorter durations than the understories of xeric hammocks (Walters et al. 1994). Shorter, cooler fires also lead to decreased tree mortality and less wildlife stress (Glitzenstein et al. 1995, Varner et al. 2005). Most importantly, the burrows created by gopher tortoises are depended upon by many different species ( Eubanks et al. 2003 ). During fires, gopher tortoise burrows provide refuge to hundreds of species, whi ch enter for protection from both the fire and the heat (Franz 1984, Russel et al. 1999). Numerous other animals, including gopher frogs ( Rana capito LeConte) and many insects, are dependent on gopher tortoise burrows as habitat (Witz et al. 1991, Franz 1 984, Layne and Jackson 1994). Many species of insects have been recorded around the entrance and apron of the burrows, although their attraction to burrows and

PAGE 30

23 their position in fire dominated habits is still poorly understood (Knight and Holt 2005, Izhak i et al. 2003). Gopher frogs (Figure 2.2 A) are also heavily dependent on gopher tortoise burrows for habitat and frequent many burrows, especially during migratory mating behavior in the fall (Franz 1984). Activity of gopher frogs at gopher tortoise bur rows does not appear to be dictated by temporal factors W hen Gopher frogs migrate to breeding ponds each year they cover substantial distances by moving from burrow to burrow. Upon returning and on subsequent trips in following years, gopher frogs will make use of the same tortoise burrows to reach the ponds (Franz 1984). Habitats with low gopher tortoise burrow abandonment rates are preferable due to the reuse of the same burrows year after year (Witz et al. 1991). In addition, the Florida mouse (Fig ure 2.2 B) is highly dependent on gopher tortoise burrows for habitats and creates chimneys off of gopher tortoise burrows (Layne and Jackson 1994). In contrast to gopher tortoise activity, Florida mice are almost exclusively nocturnal (Alexy et al. 2003, McRae et al. 1981). Habitats with densities of active gopher tortoise burrows are closely associated with Florida mice populations (Layne and Jackson 1994). Pine snakes ( Pituophis melanoleucus mugitus Barbour) (Figure 1.3 B) use gopher tortoise burrows as refuge from adverse conditions above ground such as fires, and to avoid predators (Franz 1984, Witz et al. 1991). The snakes are very fossorial, spending the majority of time underground, generally in burrows created by other animals. One

PAGE 31

24 study at th e Ordway Swisher Biological Station found pine snakes use of gopher tortoise burrows to account for just fewer than 25% of burrows used (Franz 1984). Figure 2.2. A. A gopher frog sits at the entrance of a gopher tortoise burrow. B. Two Florida mice are visible dashing into a gopher tortoise burrow. Both images were captured at the Ordway Swisher Biological Station.

PAGE 32

25 Given the demonstrated importance of gopher tortoises and their burrows within these ecosystems, it is clear that it is important to stud y and understand their behaviors. From ongoing research we have begun to better understand their behaviors and activity patterns. Gopher tortoises are primarily diurnal with a unimodal pattern of activity (Alexy et al. 2003, McRae et al. 1981). However, during periods of extremely high temperatures they have been observed to adopt a bimodal pattern of activity. The majority of gopher tortoise activity is due to thermoregulation and occurs at or near the mouth of the burrow (Eubanks et al. 2003, Jodice e t al. 2006). Foraging typically takes place within a 30 meter radius of the burrow entrance (McRae et al. 1981, Alexy et al. 2003). Gopher tortoises typically use more than one burrow, although the number is highly variable, and males tend to use more b urrows than females. There are still many areas in which little is known about gopher tortoises and their activity. For example, we know that gopher tortoises sometimes lay their eggs in the apron of burrows, however other times they will lay them in op en understories (Butler and Hull 1996). Moreover, we are unsure what factors play into this decision and which is the preferred habitat, since the ones laid away from aprons are much harder to locate. Although we have gradually been able to build upon ou r knowledge of gopher tortoise activity, there are few comprehensive methods for surveys of their behavior. Because of the difficulty of studying gopher tortoise behavior and activity patterns numerous innovative approaches have been developed and impleme nted. Gopher tortoise activity has been monitored through intensive manual observation; however, this is labor intensive and difficult due to the infrequency of activity. Several methods have been used to detect gopher tortoise activity at burrows

PAGE 33

26 Goph er tortoises play an important role in sandhill and xeric habitats and are relied upon by hundreds of species for either shelter or habitat (Eubanks et al. 2003). As an ecosystem engineer, gopher tortoises also have a large impact on the surrounding habit at through the creation of microclimates and open areas (Kaczor and Hartnett 1990). Although the importance of this keystone species is widely recognized, we still lack knowledge of many of their behaviors and activities. The objective of this study is t o use infrared triggered digital cameras in order to comprehensively monitor gopher tortoise activity at the entrance of burrows and to provide a better understanding of temporal patterns of activity. I hypothesize that gopher tortoise activity will be pr imarily diurnal and that many species of burrow commensals will be observed. Materials and Methods The following study was conducted on the Ordway Swisher Biological Station (OSBS 2941' N, 82 W http://www.ordway.ufl.edu/index.htm). All protocols and r esearch procedures for working with animals were approved by the University of Florida Institute of Food and Agricultural Sciences Animal Research Committee under protocol 008 09WEC The station is located in Putnam County, Florida and encompasses approxi mately 3,900 ha. The station encompasses land originally purchased by the is used as a private hunting reserve. In 1979, a total of 1,400 ha were donated to The Nature Conservancy as a wildlife sanctuary. In 1980, the Go odhill foundation, founded by Katherine Ordway, provided the University of Florida with a grant to purchase an adjacent 2,500 ha from the Swisher Foundation. In 2006, the University of Florida became responsible for managing both properties. The collecti ve properties were then named the Ordway Swisher Biological Station.

PAGE 34

27 Prior to the acquisition of the land by the Nature Conservancy and the University of Florida, land encompassed by the station was used for cattle ranching as well as for game hunting. The Ordway Swisher Biological Station has since initiated ongoing restoration efforts in an attempt to return the habitat to its original state. In addition to fire restoration, there is an ongoing effort to clean up the remains of internal fences as we the station has generated over 200 publications, many of which focus on dynamics within the sandhill ecosystem. Evidence of human impact at the station persists in the form of fences and gates, although the majority of the remaining internal fences are so degraded they do not meaningfully fragment the habitats. The station is crisscrossed with roads and fire breaks which also help to divide the station into units. Rur al residential properties are found along the border of the station. The units used during the course of this study (B6, C10, G2, I5 Appendix I ) are scattered across the station and chosen based on the habitat type and time of last burn. The station is home to over 500 species of plants and 284 species of invertebrates. squirrel ( Sciurus niger L. ), gopher tortoise, gopher frog and black bear ( Ursus americanus Pall as). The high number of species found within the station is due in part to the many habitat types within the station. The Ordway Swisher Biological Station encompasses a wide variety of habitats including both wetlands and uplands comprised of sandhills, xeric hammock, mixed

PAGE 35

28 forest, swamps, marshes and lakes. Xeric hammocks are created through the exclusion of fire for greater than 30 years and are the climax community on sandy uplands. The community of plants found in xeric hammocks is largely dictated by the community from which it originated and is often characterized by multiple species of oak and saw palmetto ( Serenoa repens J.K. Small) among others. Upland mixed forests generally occur on hilly terrain with sandy clay soil. A closed canopy leads t o reduced air movement and increased humidity. The multiple types of swamps and marshes found on the station require frequent fires to prevent encroachment of shrubs and trees. Sandhill habitats are dominated by longleaf pines, which are spaced widely, w ith a sparse midstory of turkey oak and an understory dominated by wiregrass and other herbaceous plants. Sandhill habits are dependent on wildfire to prevent the succession to xeric hammocks and fires naturally occur in these habitats every 2 5 years. M uch of the historic sandhill habitat has been lost to development, agriculture and fire exclusion; thus the study and proper maintenance of these habitats has greater importance. It is also the habitat in which this study was conducted. Nine gopher tortoi se burrows were periodically monitored from June 16, 2009 to August 13, 2009 for a total of 58 days at the Ordway Swisher Biological Station. For the purpose of this study only gopher tortoise burrows that appeared to be active were selected to be monitor ed. Burrows that were blocked by leaves and other detritus material were considered to be inactive (Figure 2.3 A). Inactive burrows often displayed substantial amounts of erosion. Burrows were identified as active (Figure 2.3 B) where there was recent e vidence that the burrow had been used by gopher tortoises, such as tracks or displaced sand in front of the burrow. Active burrows may also be easily

PAGE 36

29 identified by fresh gopher tortoise tracks, however frequent rain prevents use of this as the sole method of identifying active burrows. Burrows selected for study were all located at least 30 m from the road and 100 m away from any other burrows that were being used in the study to reduce the probability of recording the activity of the same tortoise at two different burrows. Figure 2.3. A. An inactive gopher tortoise burrow at the OSBS. B. An active gopher burrow at the OSBS. Active infrared trail monitors (TM1550v and TM1500v active infrared trail m onitors, Goodson & Associates, I nc.) were placed a cross the entrances of the gopher tortoise burrows. The triggers were placed so that the infrared transmitter and receiver were just above ground level and would be triggered when anything one inch in height passed between them. To ensure that each camer a and trigger set would be capable of picking up similar disturbances, each was calibrated using a one inch PVC pipe (Figure 2.4). If the pipe could be passed along the ground at any point between transmitters and registered then the system was considered to be satisfactorily calibrated. The triggers

PAGE 37

30 were secured by stakes to prevent accidental movement by animals. Stakes were determined to be a more efficient and secure method of securing the sensors than strapping them to embedded PVC pipe while causing fewer disturbances to the burrow apron. A small trowel was used to dig a small hole for each unit in the apron of the burrow allowing the sensors to be placed flush with the ground with minimal disturbance. Figure 2.4. Calibration of the infrared trai l monitors is checked using a one inch piece of PVC pipe. Several modifications were ma de to the outside of the sensor casing s to increase reliability and reduce the risk of recently excavated sand from blocking the infrared beam clear sheet of Plexiglass cut to the size of each unit. The PVC was placed such that it encompassed the transmitter and receivers. A piece of electrical tape with a hole punched in it was p laced over the receiver. The Plexiglass sheet was in turn secured to the unit using two zip ties. The effect of these modifications was twofold. First, the PVC ends and electrical tape helped to increase the sensitivity of the triggers. The triggers ar e intended to be used for larger animals at much greater distances and thus needed to be more sensitive to fully capture the movement of smaller animals. Secondly, the PVC ends helped to keep excavated sand from blocking the sensor. The units were then p laced in gallon Zip lock bags, with a hole cut for the PVC to protrude from, to

PAGE 38

31 further protect them from the sand and elements. Although the triggers are weatherproof, they are not designed to be exposed to constant harsh sunlight or to be buried in sand The cable connecting the sensors and cameras was suspended above the ground using low vegetation or sticks to prevent insects or small mammals from chewing through the cord. The excess cord was coiled loosely on the tripod and allowed to spool out easi ly should a large animal run into the cord to reduce the risk of the tripod being knocked. Each infrared trigger was connected to a digital camera (EOS 350D EOS digital camera or EOS 400D EOS digital camera, Canon U.S.A., Inc.), which captures a picture each time the infrared beam is broken. Each camera had a 2 GB compact flash drive for picture storage. The cameras were mounted in a weatherproof box, on a tripod and situated so that they would look into the entrance of the burrow with as few obstructi ons as possible (Figure 2.6). Each time the cameras were set up they were set into programmable mode (P) and autofocus was used to focus on the entrance of the burrow. Then the flash on the camera manually opened and the focus on the lens was set to manu al focus to prevent the camera from focusing each time a picture was taken. The camera was then set to the fully automatic mode to ensure proper exposure even with varied lighting during a day. Throughout the duration of this study, cameras and infrared t riggers were checked bi daily. The battery and compact flash card were replaced with blank cards and fully charged batteries. Debris such as leaves and sticks that may have accumulated between the infrared triggers was removed to reduce any chance of int erference. Between camera checks the spare camera batteries were recharged and compact flash cards were downloaded to a computer, then the pictures were entered into an excel spreadsheet. The

PAGE 39

32 cameras were rotated in sets of three between the nine burrows used in this study in a bi weekly basis. Figure 2.5. Active infrared trail monitors are enclosed in zip lock bags and staked to the ground to prevent movement; meanwhile the triggers are checked using a one inch piece of PVC at OSBS. Figure 2.6. A d igital camera in a weatherproof box is set up to capture pictures of activity at the entrance of a gopher tortoise burrow.

PAGE 40

33 Results During the course of the study, nine gopher tortoise burrows were periodically monitored us ing six infrared triggered camer as. A total of 2,628 successful events were photographed during the study, 82 percent of which were of gopher tortoises (2,152 of 2,627) (Table 2.1). A total of eight species of vertebrates were also photographed at the entrance of the gopher tortoise bu rrow. Following gopher tortoises, the most common species observed were gopher frogs and Florida mice in this order. Gopher tortoises were found to be al most exclusively diurnal, with greater than 80% of gopher tortoise observations falling between 0700h and 1800h. Upon exiting the burrow, gopher tortoises remained out of the burrow for a mean time of 24.67 minutes before returning. The mean duration spent inside the burrow between exits during the day was 64.64 minutes. Peak gopher tortoise activity o ccurred during 13 : 00h and 14 : 00h (Figure 2.7). The 95% confidence interval for gopher tortoise activity fell between 0 9: 30h and 20:40h. In this study greater than 95% of activity fell between 10:00h and 20:00h. Table 2.1. Frequency of vertebrate species observed during the study over the summer of 2009 at the Ordway Swisher Biological Station. Name Scientific Name Number Obs. Gopher Tortoise Gopherus polyphemus 2152 Gopher Frog Rana areolata 200 Florida Mouse Podomys floridanus 80 Florida Pine Snake Pituophis melanoleucus mugitus 13 Snakes Serpentes 4 Virginia Opossum Didelphis virginiana 3 Whiptail Lizard Teiidae 2 Raccoon Procyon lotor 1 Total Observations 2455 Gopher tortoises were found to have a unimodal pattern of activity and were almos t exclusively diurnal. The mean time of first gopher tortoise activity per day was

PAGE 41

34 approximately 13:25h, while the mean time of the last recorded activity per day was approximately 16:46h. The mean time of all activity was 15:06 h:m. Gopher tortoises we re reliably recorded entering then leaving the burrow 421 times. A little fewer than 30 percent of these records were less than 15 minutes in duration (Figure 2.8). The number of records decreased gradually, with a substantial increase in records of stay s longer than ten hours. The longest recorded duration spent in the burrow was over 46 hours. Gopher tortoises were recorded exiting the burrow and later returning a total of 428 times (Figure 2.9). Close to 65 percent of these events were less than 15 minutes in duration. There was a sharp decrease in the number of events as the duration increased, with 95% of records lasting less than 60 minutes. There were only nine records of activities lasting more than two hours. The second most commonly observe d species next to gopher tortoises were gopher frogs. During the duration of the project, gopher fro gs were recorded at the e ntrance of the burrows 188 times (Figure 2.10). Gopher frogs were observed most hours of the day and appeared to be active during both the day and night. Observation peaks during different hours were frequently due to the same frog triggering an individual camera many times by blocking the infrared beam. The next most observed category consists of all observations of class Insecta grouped together. Of 165 observations, insects were found to be active at gopher tortoise burrow entrances almost exclusively during the day with only one observation at night

PAGE 42

35 Figure 2.7. Number of gopher tortoise observations sandhill pine habitat s relative to the time of day observations occurred. Figure 2.8. Number of observations of different lengths of time (minutes) gopher tortoises spent in the burrow between exits. Figure 2.9. Number of observations of different lengths of time (minut es) gopher tortoises spent out of the burrow before returning.time (Figure 2.11). 0 50 100 150 200 250 300 350 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Number of Observations Time of Day (Hours) 0 50 100 150 200 250 300 <15 15-30 30-60 60-120 120-240 240-600 >600 Number of Events Duration (Minutes) n=421 0 50 100 150 200 250 300 <15 15-30 30-60 60-120 120-240 240-600 >600 Number of Events Duration (Minutes) n=428

PAGE 43

36 Similar to the gopher frogs, large numbers of observations were due to a single individual triggering the camera multiple times. The third most observed species was the Florida mouse, which was observed a total of 74 times at gopher tortoise burrow entrances (Figure 2.12). The majority of observations were after dark or during the early morning. Figure 2.10. Number of observations of gopher frogs in relation to time of day. Figure 2.11. Number of observations of class Insecta in relation to time of day. 0 10 20 30 40 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Number of Observations Time of Day (Hours) 0 10 20 30 40 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Number of Observations Time of Day (Hours)

PAGE 44

37 Figure 2.12 Number of observations of Florida mice in relation to time of day. Discussion Observed gopher tortoise activity was primarily diurnal and followe d a unimodal pattern of activity. These support published works which have observed the same patterns of activity (McRae et al. 1981, Alexy et al. 2003). Gopher tortoises were found to spend much longer periods of time inside their burrows between exits than spent outside before returning to the burrows. The longest duration gopher tortoises were observed to stay inside their burrows before exiting was greater than 46 hours. This is supported by other studies, which monitored gopher tortoises using radi o transmitters, which have observed tortoises spending multiple days inside a burrow ( Diemer 1992, Yager et al. 2007). Duration spent inside the burrow appears much more variable than the amount of time spent out of burrows. Time out of burrows was domin ated by activities lasting fewer than 15 minutes. Gopher frogs were observed to be active throughout the day and did not exhibit any temporal patterns. This is with observed behavior of gopher frogs using screen funnels at the entrance of tortoise burro ws (Franz 1984). The advantage of using camera 0 10 20 30 40 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Number of Observations Time of Day (Hours)

PAGE 45

38 traps in this situation is that screen funnels may only be left in place over night and must be removed in the morning to prevent the tortoise from inadvertently destroying the traps. Unfortunately, distingu ishing individuals is difficult using camera traps and is reliant on capture recapture methods. Insect activity was observed to be almost exclusively diurnal. Although other studies have recorded insect presence at gopher tortoise burrows, insect activity in these habitats remains poorly documented and understood (Knight and Holt 2005). Different methodologies should be employed in these habitats to better describe the presence and activity of insects. Although camera traps provide a convenient method of remote monitoring, sensitivity of the infrared triggers would have to be increased substantially in order to comprehensively record insect presence and activity. Florida mice were observed to be primarily nocturnal, with only two observations during the d ay, which is consistent with previous behavioral studies (Alexy et al. 2003). Florida mice tend to be almost exclusively reliant on gopher tortoise burrows (Layne and Jackson 1994). Observation of Florida mice outside the burrow was complicated by their relatively small size and quick movements. Even with the quick on time of modern DSLR camera, in numerous pictures the mouse was just visible on the edge of the frame, suggesting that other pictures were entirely missed. In addition, Florida mice constru ct chimneys, providing additional small entrances to the burrow, and thus do not always pass through the camera sensors (Layne and Jackson 1994). Although we were able to detect activity patterns consistent with published literature, more comprehensive mo nitoring of chimneys and other entrances would be necessary to capture all Florida mice activity.

PAGE 46

39 Eight species of vertebrates were captured using digital cameras actively using the gopher tortoise burrow or surrounding apron (Table 2.1) The use by a mu ltitude of species, some of which are threatened, further reinforces the importance gopher tortoise burrows play in their habitats. The infrared triggered camera traps were able to capture gopher tortoise activity very well, however, in order to learn mor e about commensals, higher sensitivity settings or different infrared triggers should be used to monitor burrows. Infrared trigged cameras present a unique approach of monitoring gopher tortoise burrows and allow for better monitoring of species which mak e only short term visits to burrows when compared to burrow cameras or manual observation. In some pictures, it is possible to discern the sex of the tortoise by examining the carapace. However the camera angle used in this study is not ideal for capturi ng distinguishing characteristics. In order to determine differences in activity patterns between sexes, the camera position should be reevaluated and more in depth examination of individual photographs would be required. The level of inference that can be drawn about gopher tortoise activity patterns from this study is limited by the small sample size and the duration of the project. Ideally, larger randomized samples would be used to give increased confidence about activity patterns of this population of gopher tortoises. However, the burrows selected for this study were fairly representative of the majority located within the Ordway Swisher Biological Station and thus should provide a reasonable depiction of activity patterns for this population. Gop her tortois e activity is known to vary seasonally (Diemer 1992a, Diemer 1992b, McRae et al 1981), so in order to more comprehensively record

PAGE 47

40 gopher tortoise activity patterns, the study should be continued for the duration of at least one year.

PAGE 48

41 Litera ture cited Alexy, K.J., K.J. Brunjes, J.W. Gasset and K.V. Miller. 2003. Continuous remote monitoring of gopher tortoise burrows use. Wildlife Society Bulletin 31(4): 1240 1243. Aresco, M.J. and C. Guyer. 1999. Burrow abandonment by gopher tortoises in sla sh pine plantations of the Conecuh National Forest. The Journal of Wildlife Management 63(1): 26 35. Breininger, D.R., P.A. Schmalzer, and C.R. Hinkle. 1991. Estimating occupancy of gopher tortoise ( Gopherus polyphemus ) burrows in costal scrub and slash pi ne forests. Journal of Herpetology 25(3): 317 321. Butler, J.A. and T.W. Hull. 1996. Reproduction of the tortoise, Gopherus polyphemus in Northeastern Florida. Journal of Herpetology 30(1): 14 18. Diemer, J.E. 1992a. Demography of the tortoise Gopherus po lyphemus in Northern Florida. Journal of Herpetology 26(3): 281 289 Diemer, J.E. 1992b. Home range and movements of the tortoise Gopherus polyphemus in Northern Florida. Journal of Herpetology 26(2): 158 165. Eubanks, J.O., W.K. Michener, and C. Guyer. 200 3. Patterns of movement and burrow use in a population of gopher tortoises ( Gopherus polyphemus ). Herpetologica 59(3): 311 321. Franz, R. 1986. The Florida gopher frog and the Florida pine snake as burrow associates of the gopher tortoise in Northern Flor ida. Pp. 16 20 in D.R. Jackson and R.J. Bryant (eds.). The gopher tortoise and its community. Proceedings from the 5 th annual meeting of gopher tortoise council, Florida State Museum, Gainesville. Glitzenstein, J.S., W.T. Platt, and D.R. Streng. 1995. E ffects of fire regime and habitat on tree dynamics in North Florida longleaf pine savannas. Ecological Society of America 65(4): 441 476. Izhaki, I., D.J. Levey, and W.R. Silva. 2003. Effects of prescribed fire on an ant community in Florida pine savanna. Ecological Entomology 28: 439 448. Jodice, P.G., D.M. Epperson, and G.H. Visser. 2006. Daily energy expenditure in free ranging gopher tortoises. American Society of Ichthyologists and Herpetologists 2006(1): 129 136.

PAGE 49

42 Kaczor, S.A, and D.C. Hartnett. 1990. Gopher tortoise ( Gopherus polyphemus ) effects on soils and vegetation in a Florida sandhill community. American Midland Naturalist 123(1): 110 111. Knight, T.M., and R.D. Holt. Fire generates spatial gradients in herbivory: an example from a Florida sandh ill ecosystem. Ecology 86(3) 587 593. Layne, J.N. and R.J. Jackson. 1994. Burrow use by the Florida mouse ( Podomys floridanus ) in South Central Florida. American Midland Naturalist 131(1): 17 23. McRae, W.A., J.L. Landers, and J.A. Garner. 1981. Movement p atterns and home range of the gopher tortoise. American Midland Naturalist. 106(1): 165 179. Varner J.M., D.R. Gordon, F.P. Putz, and J.K. Hiers. 2005. Restoring fire to long unburned Pinus palustris ecosystems: novel fire effects and consequences for lo ng unburned ecosystems. Restoration Ecology 13(3): 536 544. Walters, T.W., D.S. Decker Walters, and D.R. Gordon. 1994. Restoration considerations for wiregrass ( Aristida stricta ): allozymic diversity of populations. Conservation Biology 8(2): 581 585. Wi tz, B.W., D.S. Wilson, and M.D. Palmer. 1991. Distribution of Gopherus polyphemus and its vertebrate symbionts in three burrow categories. American Midland Naturalist 126(1): 152 158. Yager, Y.L., M.G. Hinderliter, C.D. Heise, and D.M. Epperson. 2007. Gop her tortoise response to habitat management by prescribed burning. Journal of Wildlife Management 71(2): 428 434.

PAGE 50

43 Chapter 3 Effects of Prescribed Fire on Temporal Activity of Gopher Tortoises Abstract O ur understanding of the habitat ecology of the go pher tortoise ( Gopherus polyphemus Daudin) has increased substantially in the past decade, although effects of habitat manipulation, fire, and habitat quality on the behavioral ecology of tortoises remains poorly understood. Specifically, little is known about the impact of prescribed fire on gopher tortoise activity. Use of infrared triggered cameras combined with digital photographic technology has proved a useful tool for comprehensive monitoring of gopher tortoise burrows, allowing for examin ation and assessment of gopher tortoise activity during all hours. Fifty three burrows were monitored for one week each from June to August for a total of 60 days Infrared camera traps were placed in front of burrows that were visually determined to have been rece ntly active and triggered by a tortoise entering or exiting the burrow. The six cameras used in the study were split between habi tats recently treated with fire and areas which were unburned for over a year The time stamps embedded in the images were re corded in order to examine the temporal patterns and effects of prescribed fi re on gopher tortoise activity. Eight species were photographed actively using gopher tortoise burrows; the most commonly observed species were gopher tortoises, Florida mice ( Pod omys floridanus Chapman), and gopher frogs ( Rana capito LeConte) in this order. The vast majority of photos captured were of gopher tortoises (887 of 1171). Time of activity in burned and

PAGE 51

44 unburned habitats differed significantly. The mean time of observ ation for unburned habitats was approximately 14:30 h while the mean time of observations for burned habitats was approximately 15:54 h The mean time of activity was found to vary significantly between burned and unburned habitats. Further studies are ne cessary in order to better describe the patterns of activity observed in this study.

PAGE 52

45 Introduction The Sandhill ecosystem found throughout the southeast United States evolved to become highly dependent on naturally occurring fires every 1 10 years (Frost 2006, Glitzenstein et al. 1995, Drew et al. 1998). Today only a small fragment of sandhill ecosystem s remain L oss of this habitat has been caused by numerous factors including agriculture practices, development, and fire suppression (Frost 2006). Natu ral fires play an important role in the maintenance of these habitats. As our knowledge of naturally occurring fires has increased, there has also been a substantial increase in efforts to maintain s andhill habitats through reintroduction of burning (Varn er et al. 2005, Walters et al. 1994). Longleaf pine ecosystems are characterized by widely spaced longleaf pine ( Pinus palustris Mill.), a sparse midstory of turkey oak ( Quercus laevis Walter), and a dense understory of wiregrass ( Astrida stricta Michx. ) and other herbaceous plants (Frost 2006, Drew et al. 1998). Historically, primeval sandhill habitats had a complete lack of midstory because of regular fires When fire is suppressed in these ecosystems the majority eventually succeed into Xeric Hammock characterized by an abundance of hard woods with a dense midstory and a closed canopy (Frost 2006, Drew et al. 1998, Glitzenstein et al. 1995). A much denser layer of detritus and humus is also formed on the floor (Varner et al. 2005). Regular burning of this habitat has three highly noticeable impacts: opening of the canopy, elimination of the midstory, and the growth of wire grass and other understory plants (Figure 3.1) Each of these helps to prevent succession of Sandhill habitats.

PAGE 53

46 Figure 3.1. A gopher tortoise burrow in traditional longleaf pine habitat undergoing restoration at the Ordway Swisher Biological Station. The succession of longleaf pine habitat to Xeric Forest habitat has many implications ; greatly alter ing the ecosystem and the spec ies within it. Species with low tolerance for fire, including many hardwood species, experience rapid growth with fire suppression (Frost 2006, Glitzenstein et al. 1995). Suppression of fire for more than 30 years can lead to the complete transition s to Xeric Forest, in which hardwood species like turkey oak, outcompete the longleaf pine (Varner et al. 2005). The loss of wiregrass and the closing of the canopy alter the habitat, greatly reducing the level of biodiversity

PAGE 54

47 (Drew et al. 1998). Much higher levels of biodiversity are found in sandhill ecosystems when compared to x eric h ammocks (Walters et al. 1994). Additionally, habitats with wiregrass have greater species diversity (Varner et al. 2005). Closing of the forest canopy also results in less fav orable habitat for several vertebrate species, including the gopher tortoise ( Gopherus polyphemus Daudin) Sandhill habitats with denser canopies have a reduc ed density of gopher tortoise burrows ( Aresco and Guyer 1999 ). In addition to reduced biodiver sity, the succession of longleaf pine habitats to x eric hammock can pose a threat to humans because of the dangerous fires that may arise (Varner et al. 2005). When suppressed habitats eventually burn the fire is usually catastrophic, more difficult to co ntrol, and may threaten nearby residential properties (Varner et al. 2005). In a healthy longleaf pine ecosystem the fire is generally cool, spreads quickly, and, due to the lack of mid story, unable to reach tree canopies (Glitzenstein et al. 1995, Drew e t al. 1998). Regions with suppressed fire show a larger midstory contribute to the increased accumulation of detritus and hummus also known as duff, which acts as a slow burning fuel that encourages canopy fires (Frost 2006). Duff fires may smolder for l ong periods of time and are almost impossible to put out (Varner et al. 2005). Smoke from duff fires can present a health risk to local residents, since the smoke is heavier than that from active flames (Varner et al. 2005). Duff fires also present a ris k to the larger trees, because tree roots are unable to survive the long period of increased heat, which eventually cause s the death of the tree (Glitzenstein et al. 1995). A solid understanding of fire dynamics is very important for multiple reasons. N a tural fire cycles have been disrupted by development and fire suppression P rescribed fire plays a crucial role in maintaining and preserving these ecosystems (Varner et al.

PAGE 55

48 2005, Glitzenstein et al. 1995, Frost 2006). A better understanding of the relat ionship between dependent species and fire enable s better application of prescribed fire management to delicate ecosystems and progress towards the goal of restoring and maintaining longleaf pine ecosystems (Varner et al. 2005, Glitzenstein et al. 1995). Gopher tortoise abandonment of burrows significantly increases under a closed canopy when compared to the open canop ies of sandhill pine habitat (Aresco and Guyer 1999, Yager et al. 2007). Over time, the density of gopher tortoise burrows in xeric hammo cks becomes lower than that of longleaf pine habitats. Naturally occurring fires, as well as prescribed fires, open up the forest canopy, eliminate the midstory, and encourage the growth and proliferation of wiregrass each of which produce a more desirab le habitat for gopher tortoises (Eubanks et al 2003, Hermann et al. 2002, Yager et al. 2007). Over 350 species of animals are dependent on gopher tortoise burrows for habitat or refuge (Russell et al. 1999). The importance of gopher tortoises as ecosystem engineers and the extent to which their burrows are used by other species necessitates the preservation of their preferred habitat. Our understanding of gopher tortoise activity patterns has increased substantially as new applications of technology have a llowed more comprehensive monitoring of activity. Traditionally, wildfires have played an important role in maintaining sandhill ecosystems. Due to the effects of modern wildfire suppression techniques prescribed burning has been implemented in order to maintain forest habitats. Today, prescribed fire plays a crucial role in the maintenance of these ecosystems and has a large impact on the local flora and fauna. Gopher tortoises, a keystone species in these ecosystems, play an important role in habit at creation for many other species. As we continue to expand our

PAGE 56

49 knowledge and understanding of natural ecosystems, it is important to determine the impacts of prescribed fire on key species such as gopher tortoises. The objective of this study is to det ermine the impact of recently burned habitats on gopher tortoise activity patterns through the use of infrared triggered cameras. Methods The following study was conducted on the Ordway Swisher Biological Station ( OSBS 2941' N, 82 W http://www.ordway.u fl.edu/index.htm). All protocols and research procedures for working with animals were approved by the University of Florida Institute of Food and Agricultural Sciences Animal Research Committee under protocol 008 09WEC The station is located in Putnam C ounty, Florida and encompasses approximately 3,900 ha The land the station was originally purchased by the Swisher ha were donated to The Nature Conservancy as a wildl ife sanctuary. In 1980, the Goodhill foundation, founded by Katherine Ordway, provided the University of Florida with a grant to purchase an adjacent 2,500 ha from the Swisher Foundation. In 2006, the University of Florida became responsible for managing both properties. The collective properties were then named the Ordway Swisher Biological Station. Prior to the acquisition of the land by the Nature Conservancy and the University of Florida, by the station was used for cattle ranching as well as game hunting. The Ordway Swisher Biological Station has since initiated restoration efforts in an attempt to restore the habitat to its original state. In addition to prescribed fire (Figure 3. 2) there is an ongoing effort to clean up the remains of intern al fences as well as other areas research at the station has

PAGE 57

50 generated over 200 research publications, many of which focus on dynamics within the sandhill ecosystem. by humans persists in the form of fences and gates, although the majority of the remaining internal fences are degraded such that they do not fragment the habitats. The station is crisscrossed with roads and fire breaks which also help to divide the stati on into units. Rural residential properties are found along the border of the station. The units used during the course of this study (B6, C10, G2, I5) are scattered across the station and chosen based on the habitat type and time of last burn. The Ord way Swisher Biological Station encompasses a wide variety of habitats including both w etlands and uplands comprising sandhills, xeric hammock, upland mixed forest, swamps, marshes and lakes. Xeric hammocks habitats are created through the exclusion of fir e for more than 30 years and are the climax community on sandy uplands. The community of plants found in xeric hammocks is largely dictated by the com munity from which it originated and is often characterized by multiple species of oak and saw palmetto ( S erenoa repens J.K. Small). When these habitats do burn, the result is almost always a catastrophic fire and substantial alteration of the community. Upland mixed forests generally occur on hilly terrain with sandy clay soil. A closed canopy leads to red uced air movement and increased humidity, resulting in infrequent burns. The multiple types of swamps and marshes found on the station require frequent fires in order to prevent encroachment of shrubs and trees. Sandhill or longleaf pine habitats compris e a substantial portion of the station and are dependent on frequent burns. These habitats are dominated by longleaf pines, which are widely spaced The mid story is sparse and largely comprised of turkey oak, while the understory is dominated by wiregras s and

PAGE 58

51 Figure 3.2. A drip torch is used to spread fire in a longleaf pine habitat at the Ordway Swisher Biological Station. Photograph courtesy of John P. Hayes.

PAGE 59

52 other herbaceous plants. Sandhill habits are dependent on wildfire to prevent the succes sion to xeric hammocks and fires naturally occur in these habitats every 2 5 years. Much of the historic sandhill habitat has been lost to development, agriculture and fire exclusion; thus the importance of studying and properly maintaining these habitats has been assigned additional importance. It is also the habitat in which this study was conducted. Over the course of this study, six camera and infrared trigger units were used to monitor 53 gopher tortoise burrows one week each over the period of June 1 2010 to August 5, 2010 for a total of 60 days at the Ordway Swisher Biological Station (OSBS) covering four units. In this study, burrows were classified into two group s: active and inactive (Figure 2.3 ). Other studies have used more expansive classi fication systems ( Alexy et al. 2003, Aresco and Guyer 1999 Diemer 1992); however, because studying gopher tortoise activity is the primary objective of this study, only fully active burrows were monitored. Since gopher tortoises have been known to remain inside their burrows for multiple days, sites that appeared inactive may have contained gopher tortoises but were not included in this study. Burrows were identified as active when there was recent evidence that the burrow had been used by gopher tortois es The same criteria used in the previous study for determining whether or not a burrow was active was used again in this study. The burrows selected for study were located at least 30 m from the road and 40 m away from any other burrows included in the study to reduce the probability of recording the activity of the same tortoise at two different burrows (Alexy et al. 2003). Active infrared trail monitors (TM1550v and TM1500v active infrared trail monitors, Goodson & Associates, inc., Lenexa, Kans. ) were placed adjacent to the

PAGE 60

53 entrances of the gopher tortoise burrows. To ensure that each camera and trigger set pipe (Figure 3.4). If the pipe could be passed alon g the ground at any point between transmitters and registered then the system was considered to be satisfactorily calibrated. The triggers were secured using metal stakes (Figure 3.4 ) to prevent accidental movement by animals. In a previous study gophe r tortoises accidently pushed one of the sensors into the burrow and the use of stakes eliminated this problem (Hayes, unpublished data). A small trowel was used to dig a small hole for each unit in the apron of the burrow allowing the sensors to be place d flush while minimizing disturbance to the burrow apron. The units used in this study used the PVC and Plexiglass modifications designed to increase infrared trigger sensitivity and reliability (Figure 3.4 ) (for more detail see previous chapter). The s ensors were also sealed in plastic in order to prevent degradation of the sensor casing and accumulation of sand inside the units. In order to further reduce the risk of the camera units being accidently knocked over by wildlife or high winds, a stake tied to the center column weight hook was embedded in the ground directly beneath the tripod. Each infrared trigger was plugged into a digital camera (EOS 350D EOS digital camera or EOS 400D EOS digital camera, Canon U.S.A., Inc.), which captured a picture each time the infrared beam was broken. Each camera had a rechargeable battery and a 2 GB compact flash drive for picture storage. The cameras were mounted in a weatherproof box on a tripod and focused at the entrance of the burrow with as few obstructio ns as possible ( Figure 2.6). Each time the cameras were set up they were set

PAGE 61

54 into programmable mode (P) and autofocus was used to focus on the entrance of the burrow. The flash on the camera was manually opened and the focus on the lens was set to manual to prevent the camera from focusing each time an image was captured. The camera was then set to the fully automatic mode to automate image exposure and flash use. Figure 3.4 Active infrared trail monitors are enclosed in zip lock bags and staked to the ground to prevent movement; meanwhile the triggers are checked using a one inch piece of PVC at OSBS. For the purpose of this study, the gopher tortoise burrows were divided into two categories; burrows in areas that had been recen tly burned (less th an one year) and those without recent ly prescribed fire treatment (greater than one year). In order to reduce potential external factors such as temperature or rainfall, three burrows in each habitat were monitored simultaneously. Cameras were rotated to three new burrows in each habitat on a weekly basis.

PAGE 62

55 A survey of tree diameter at breast height ( DBH ) as well as ground cover and vegetation was conducted around each burrow. Measuring from the apron of the burrow, a circle (r=20 m) was marked using flag s. Trees and snags within this circle with a circumference of 10 cm or greater were measured and recorded. Trees were categorized as longleaf p ine, t urkey o ak, l ive o ak ( Quercus virginiana Mill.), or other which included snags. The categories selected account for the majority of trees found in the sandhill ecosystem. Further information was gathered by conducting v egetative and ground cover samples two and five meters from the center of the apron in each of the cardinal directions. Groundcover and veg etation were divided into the following categories: wire grass, green herbaceous, uncharred woody, bare ground, charred wire grass, charred woody, charred other, and detritus. The categories were based on the most abundant plant species, wire grass, and other types of ground cover that appeared prominent. A one square meter grid was constructed from PVC pipe and divided into 16 equal squares using twine (Figure 3.5 and 3.6). The grid was placed at each sample point parallel to the measuring tape, star ting at the point of measurement, and extending one meter away from the burrow apron. Ground cover and vegetation were approximated by recording the rough number of squares of each of the previous categories. The percentage of each category was calculate d from the recorded values. Throughout the duration of this study, cameras and infrared triggers were checked each week day. The units remained active during weekends in order to maximize data. C amera battery and card replacement was performed each day cameras were checked. D ebris such as leaves and sticks that may have accumulated between the infrared triggers

PAGE 63

56 was removed to reduce any chance of interference. Between camera checks, the spare camera batteries were recharged and compact flash cards were downloaded to a computer, then the pictures activity events were entered into an excel spreadsheet. Figure 3.5. A 1 m 2 grid is used to survey ground cover at a recently burned unit.

PAGE 64

57 Figure 3.6. Ground cover percentages are recorded using a 1 m2 gri d at the Ordway Swisher Biological Station. Photograph courtesy of Danielle Fasig.

PAGE 65

58 Results During the course of the study 53 burrows were successfully monitored for the course of one week 27 burrows in burned habitats and 26 in unburned habitats. Twenty burrows in each habitat recorded gopher tortoise activity. A total of 1171 successful observations were recorded during the duration of the project (Table 3. 1). Gopher tortoise observation accounted for the majority of observations with 887. Observatio ns in areas categorized as unburned accounted for 469 observations, while burned areas accounted for 418. In addition to gopher tortoises, eight other species were observed actively using the gopher tortoise burrows or aprons As in the previous study, t he two next most common species o bserved after the gopher tortoise were the Florida mouse and the gopher frog Observations of the Florida mouse were almost exclusively nocturnal. Ninety five percent of all gopher tortoise activity in both burned an unbu rned units occurred between 11:46 h :m and 18:37 h :m In unburned units, 95 % of activity occurred between 10:58 h :m and 18:10 h :m while in burned units, 95 % of activity occurred between 11:46 h :m and 18:36 h. The median time of all gopher tortoise obse rvations was 15:12 h:m, while the medians of activity in unburned and burned areas were 14:30 h:m and 15:54 h:m, respectively. All probabilities were calculated using the Kruskal Wallis H Test. The average first time of activity, usually in mornings, for days in which gopher tortoise activity was recorded was 14:02 h:m in unburned units 14.21 h:m in burned units and was determined not to be statistically significant ( p =0. 9084 ). The average last time of activity, usually in evenings, for days in which goph er tortoise activity was recorded in unburned and burned units was 16:30 h:m and 16:51 h:m respectively and was determined not to be statistically significant as well ( p = 0 893 ). The average number of

PAGE 66

59 events, per burrow per day that gopher activity was rec orded, was 9.5 in unburned units and 6.3 in burned units and was not statistically significant ( p =0. 0698 ) (Table 3.2) Table 3.1. Frequency of vertebrate species observed during the study in both burned and unburned habitats during the summer of 2010 at t he Ordway Swisher Biological Station. Name Scientific Name Number Obs. Gopher Tortoise Gopherus polyphemus 887 Florida Mouse Podomys floridanus 182 Gopher Frog Rana areolata 80 Eastern Cottontail Sylvilagus floridanus 7 Whiptail Lizard Teiidae 6 Arm adillo Dasypodidae 4 Snakes Serpentes 4 Racoon Procyon lotor 1 Total Observations 1171 Table 3.2. Statistical values for gopher tortoise activity. Variable Chi square DF p Time of Activity 29.3656 1 <.0001 Duration in Burrow 0.7905 1 0.374 Durat ion Out of Burrow 2.4857 1 0.1149 Average First Time of Activity 0.0132 1 0.9084 Average Last Time of Activity 0.0181 1 0.893 Events per burrow per day 3.2877 1 0.0698 Groundcover and vegetation samples were categorized as samples from either burned o r unburned areas. Eight samples from each burrow were combined and averaged for each burrow. Percentages of g round cover and vegetation for each burrow were averaged and the 95 % confidence interval calculated (Figure 3.7). Wire grass cover accounted for under 12 % of ground cover in burned habitats and 35.5 % percent of unburned areas. Similarly, other green herbaceous material accounted for 6 % in burned areas and under 20 % in unburned habitats. Bare ground or sand accounted for 24 % of

PAGE 67

60 cover in b urned habitats and 14 % in unburned habitats. Charred wire grass accounted for approximately 4.25 % in burned areas and 0.59 % in unburned areas. Charred detritus composed 27 % of cover in burned areas and 4 % in unburned areas. The preceding categories were all determined to be statistically significant with a p value of <0.0001 for each category. Charred woody cover and detritus were not statistically significant. Charred woody material covered approximately 5 % in burned areas and slightly more than 2.5 % in unburned areas ( p =0.8070). Percentage of ground covered by detritus was similar in burned and unburned habitats, accounting for 24.17 and 24.47 %, respectively ( p =0. 6888 ) (Table 3.3) Figure 3.7. Percent ground and vegetation cover averaged for burned (blue) and unburned (red) habitats. Error bars represent 95% confidence intervals. 0 5 10 15 20 25 30 35 40 Wire Grass Green Herbaceous Uncharred Woody Bare Ground Charred Wire Grass Charred Woody Charred Other Detritus Percent Cover

PAGE 68

61 Table 3.3. Statistical values for ground cover variables Type of ground cover Chi square DF p Wire Grass 32.3297 1 <.0001 Green Herbaceous 21.7437 1 <.0001 U ncharred Woody 1.2203 1 0.2693 Bare Ground 16.9773 1 <.0001 Charred Wire Grass 40.1749 1 <.0001 Charred Woody 13.9987 1 0.0002 Charred Other 32.4152 1 <.0001 Detritus 0.1604 1 0.6888 Temporal gopher tortoise activity was divided between burned and unburned areas. Gopher tortoise activity from burned and unburned habitats was grouped by the number of events which occurred each hour and graphed relative to the time of day the activity occurred (Figures 3.8 and 3.9). The Kruskal Wallis H Test was con ducted to compare the activity time between burned and unburned habitats. The two categories were highly significa ntly different ( p < 0 .0001). The mean time of activity observation for unburned habitats was 14:30 h:m, while the mean time of observation for burned habitats was approximately 15:54 h:m.

PAGE 69

62 Figure 3.8 Number of gopher tortoise observations in burned habitats relative to the time of day observations occurred. Figure 3.9 Number of gopher tortoise observations in unburned habitats relative to the time of day observations occurred. 0 10 20 30 40 50 60 70 80 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Number of Observations Time of Day (Hours) 0 10 20 30 40 50 60 70 80 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Number of observations Time of Day (Hours)

PAGE 70

63 Discussion The vegetation and ground cover surveys provide a serviceable comparison between the burned and unburned habitats. Burned habitats had significantly less wire grass and green herbaceous material with si gnificantly more bare ground, charred wire grass, charred woody material, and other charred material. Levels of uncharred woody material and detritus did not vary significantly between the two habitat types. Ground cover was divided into these eight cat egories in order to enable efficient and feasible data collection although ideally more in depth recording of species variation between these habitats should have been recorded. The data collected gives an accurate representation of the difference betwee n the two habitats. Eight sample replications at each burrow aided in providing a representative sample. Distinctions drawn between the different habitat types will be used to assist in explaining the behavioral differences observed between habitats. Go pher tortoise activity in both burned an d unburned units was found to be almost exclusively diurnal, as is in accordance with published behavioral studies (Axely et al. 2003). The mean time of activity within the studied population in burned areas proved to be just under 1:30 h:m later than those in unburned units. The difference in ground cover between the two different habitats suggests that the differences in activity time can be explained through closer examination of these variables. Closer examina tion of activity patterns can further elucidate the difference between activity in burned and unburned units, particularly why the shift in activity occurs. A broader use of other survey methods such as visual monitoring and mark and recapture could also assist in determining the differences in behavior.

PAGE 71

64 No significant differences were found between the mean time of the first exit from the burrow and the mean time of the last entrance. This indicates that the overall duration of gopher tortoise activity was not significantly different between burned and unburned units (Table 3.2) These data suggest that the difference in mean time of activity is caused by increased activity during a different time, rather than a change in the overall time of day they r emain active. About 90% of gopher tortoise activity is related to basking and thermoregulation and is spent close to the burrow (Figure 3.9). Similarly, gopher tortoises will also enter the burrow for thermoregulation. Thus, a large portion of gopher t ortoise activity at the burrow entrance is accounted for by thermoregulation. No significant differences between the numbers of events per burrow per day were observed between the different habitat types. This indicates that individuals in one habitat we re not significantly more active than those in the other and also suggests that the differences in habitat do not lead to increased thermoregulatory responses. Although only burrows that appeared to have had recent gopher tortoise activity were included in the study, 14 of 53 burrows had no recorded activity. Many of the burrows with recorded gopher tortoise activity also had days in which no gopher tortoise activity was recorded. The lack of or pause in activity may partly explain the tendency of gopher tortoises to occasionally enter and remain in their burrow for multiple days, or the use of alternative burrows. However, this does not fully explain the lack of any activity for an entire week at 14 burrows. Likely, these burrows were visited by a nonre sident gopher tortoise, leaving fresh tracks in the sand, but not returning in the following week.

PAGE 72

65 Figure 3. 10 A gopher tortoise in a burned unit basks in the evening sun. Image captured at 5:21 h:m, at the Ordway Swisher Biological Station. It is imp ortant to consider the scope and limitations of the project to determine the level of inference that can be drawn about gopher tortoise activity. I was unable to use a fully random samp ling method due to timing and habitat availability. This limitation re duced the ability to draw conclusions about the general behavior of the gopher tortoise population s However, gopher tortoise burrows were chosen with out intentional sample bias I n my opinion, the conditions observed for the different habitats are fairl y representative of the habitat characteristics at the Ordway Swisher Biological Station I suggest that the activity patterns observed in these individuals would also be observed in other gopher tortoises in N orth C entral Florida sandhill habitats. Whil e activity patterns in other regions may not align exactly with my observations, I would also expect to find a

PAGE 73

66 relationship between the vegetation around gopher tortoise burrows and their activity throughout a larger portion of their region in the Southeas tern United States. The results of this study support general conclusions made by other s depicting gopher tortoise activities and it expands on areas that are poorly studied in current literature ( Alexy et al. 2003, McRae et al. 1981 ) Although several st udies have examined the density and populations between areas that are regularly burned and those where fire is suppressed, none to date have compared the differences in gopher tortoise activity between these two habitats in the same region A better unde rstanding of activity and response of gopher tortoises to prescribed fire will grant increased understanding of fire dominated ecosystems and the approaches and management strategies best suited to maintaining them. A fully randomized study should be condu cted in order to draw larger inferences about the behavior of the gopher tortoise population. Similarly, a larger sample size over longer period s of time will allow for higher confidences as well as an even more accurate depiction of gopher tortoise activ ity relative to the time of day. Finally, the study should be conducted for at least the period of one year in order to examine whether or not the time of year factors into the activity response s to habitat.

PAGE 74

67 Literature Cited Alexy, K.J., K.J. Brunjes, J.W. Gasset and K.V. Miller. 2003. Continuous remote monitoring of gopher tortoise burrows use. Wildlife Society Bulletin 31(4): 1240 1243. Aresco, M.J. and C. Guyer. 1999. Burrow abandonment by gopher tortoises in slash pine plantations of the Conecuh Nat ional Forest. The Journal of Wildlife Management 63(1): 26 35. Drew, M.B., L.K. Kirkman, and A.K. Gholson Jr. 1998. The vascular flora of Ichauway, Baker County, Georgia: a remnant longleaf pine/wiregrass ecosystem. Southern Appalachian Botanical Society 63(1): 1 24. Eubanks, J.O., W.K. Michener, and C. Guyer. 2003. Patterns of movement and burrow use in a population of gopher tortoises ( Gopherus polyphemus ). Herpetologica 59(3): 311 321. Frost, C. 2006. History and future of the longleaf pine ecosystem. P ages 9 48 in S. Jose, E. J. Jokela, and D. L. Miller, editors. The longleaf pine ecosystem. Springer, New York, New York, USA. Glitzenstein, J.S., W.T. Platt, and D.R. Streng. 1995. Effects of fire regime and habitat on tree dynamics in North Florida lon gleaf pine savannas. Ecological Society of America 65(4): 441 476. 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 pol yphemus Biolog ical Conservation 108: 289 298. Russell, K.R., D.H. Van Lear, and D.C. Guynn Jr. 1999. Prescribed fire effects on herpetofauna: review and management implications. Wildlife Society Bulletin 27(2): 374 384. Varner J.M., D.R. Gordon, F.P. Pu tz, and J.K. Hiers. 2005. Restoring fire to long unburned Pinus palustris ecosystems: novel fire effects and consequences for long unburned ecosystems. Restoration Ecology 13(3): 536 544. Walters, T.W., D.S. Decker Walters, and D.R. Gordon. 1994. Restorat ion considerations for wiregrass ( Aristida stricta ): allozymic diversity of populations. Conservation Biology 8(2): 581 585.

PAGE 75

68 Yager, Y.L., 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(2): 428 434.

PAGE 76

69 Chapter 4 Conclusion Substantial declines in gopher tortoise populations over the past several decades have heightened the importance of studying this species in order to implement better conservat ion and management practices. In this work I used infrared triggered digital cameras to better understand temporal activity patterns and the impact of prescribed fire on gopher tortoise activity. The results obtained from these studies present a clearer picture of go pher tortoise activity patterns and begin to describe the impact of fire ecology on gopher tortoise behavior. Few direct comparisons can be drawn between t he work described in the earlier chapters due to a break in monitoring over two years, a nnual variation in activity patterns, and lacking vegetative information from the first study. Although both studies were conducted at the Ordway Swisher B iological S tation, they were conducted in different areas. However comparing both studies will beg in to describe variability of gopher tortoise activity throughout multiple years and provide groundwork for future studies. Almost all gopher tortoise activity in both studies was diurnal and the pattern of g opher tortoise activity was primarily unimodal. The mean time of activity during the summer of 2010 was 14:30 h:m for unburned habitats and 15:54 h:m for burned areas. The grand mean of all activity was 15:12 h:m. During 2009, the mean of all gopher tortoise activity was 15:06 h:m. Comparing these results suggests that the gopher tortoises in both studies originated from the same normally distributed population and time of activity does not vary substantially between years. Finally, this suggests that if the relationship observed between gopher tor toises and habitat type (burned or unburned) holds true over multiple

PAGE 77

70 years, then the habitat chosen for the first study shared vegetative characteristics with both the burned and unburned habitats. Unfortunately, insufficient data were gathered during the first study to allow in depth comparisons to be made. Although it may not be possible to draw conclusions directly from these observations, they pave the way for future studies and allow for the formulation of better hypotheses. Temporal patterns of goph er tortoise activity observed during this project may be used as a reference for different locations. Specific time of activity is likely to vary due to differences in temperature, habitat and geographic location. I anticipate the similarity in temporal activity from year to year to reflect what has been observed in these two studies even in different geographic locations. By building off of these observations, multiple year studies may focus to a lesser extent on the variab ility of behavior between year s and more on specific behavioral activities. Future studies should control for habitat variation over multiple years to ascertain variability of gopher tortoise activity among years. Based on the results of these two studies, I hypothesize that there wou ld be little variability. The study should also be conducted during the full course of a year in order to describe the variability of activity between seasons. A comprehensive understanding of basic gopher tortoise activity patterns is paramount in furt hering the study of this reclusive species. Future studies of gopher tortoise activity benefit greatly from a basic understanding of activity times, frequency of events, and the duration of time gopher tortoises spend in and outside of their burrows. Th ese studies strive to describe and understand the activity patterns of gopher tortoises, setting the groundwork for further studies and better conservation practices.

PAGE 78

71 Gopher tortoises play a key role in th e ecosystems which they inhabit and should remain a priority for future study and conservation efforts. Still declining populations, low visibility, long lifespans and low reproductive success necessitate special care in order to preserve this species. Gopher tortoise populations should also be consider ed a priority because of the role the play in the habitat maintenance of longleaf pine forests. The benefits of conserving gopher tortoise populations extend to the many species dependent on gopher tortoise burrows and their habitat and are ever more impor tant given the increasing pressures placed on remaining sandhill habitats by development and agriculture. T his study contribut es meaningful information about gopher tortoise behavior, which may be applied to further the field of conservation of some of Fl most diverse and important ecosystems.

PAGE 79

72 Appendix 1. Map of Ordway Swisher Biological Station Research Units

PAGE 80

73 Appendix 2. Research Protocol All protocols and research procedures for working with animals were approved by the University of Flor ida Institute of Food and Agricultural Sciences Animal Research Committee under protocol 008 09WEC


ERROR LOADING HTML FROM SOURCE (http://ncf.sobek.ufl.edu//design/skins/UFDC/html/footer_item.html)