Citation
Ecology of Ornate Diamondback Terrapins (Malaclemys Terrapin Macrospilota) on a Small Gulf Coast Barrier Island and Their Behavior inside Crab Traps

Material Information

Title:
Ecology of Ornate Diamondback Terrapins (Malaclemys Terrapin Macrospilota) on a Small Gulf Coast Barrier Island and Their Behavior inside Crab Traps
Creator:
Suarez, Eric
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (75 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Wildlife Ecology and Conservation
Committee Chair:
JOHNSON,STEVEN ALBERT
Committee Co-Chair:
FREDERICK,PETER C
Committee Members:
KRYSKO,KENNETH L
TURNER,WILLIAM M,JR
Graduation Date:
8/8/2015

Subjects

Subjects / Keywords:
Crabs ( jstor )
Diamondback terrapins ( jstor )
Ecology ( jstor )
Environmental conservation ( jstor )
Experimentation ( jstor )
Female animals ( jstor )
Population size ( jstor )
Sex ratio ( jstor )
Turtles ( jstor )
Vegetation ( jstor )
Wildlife Ecology and Conservation -- Dissertations, Academic -- UF
terrapins
Gulf of Mexico ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Wildlife Ecology and Conservation thesis, M.S.

Notes

Abstract:
The diamondback terrapin (Malaclemys terrapin) has been experiencing population decline mostly through mortality in crab traps throughout its range. Population status is unclear, and the mechanisms by which they become entrapped are poorly known. To determine the ecological status of ornate diamondback terrapins (Malaclemys terrapin macrospilota) on a small Gulf Coast barrier island (Lanark Reef), a mark-recapture study was conducted to effectively develop a survey methodology and assess the population. I captured 381 individual M. t. macrospilota (32.8 % juveniles, 52.5 % adult males, and 14.7 % adult females). Sex ratio was male-biased (3.6:1) and there was an estimated 929.1 +/- 142.1 turtles inhabiting Lanark Reef. Analysis of environmental variables showed that the most efficient time to search for M. t. macrospilota is during a low tide on clear or partly cloudy days (CPUE = 0.27 or 1 terrapin every 3.6 minutes). While surveying the intertidal zone of Lanark Reef, turtles were only found in microhabitats associated with some form of vegetation (smooth cordgrass, tidal wrack, or both). An experiment was developed to determine which internal component (e.g., center baffle or bait cup) of a commercial crab trap limits their ability to escape. Thirty percent (16/53 turtles) successfully escaped from the control trap, 23% (12/53 turtles) escaped from experimental trap 1 (without center baffle), and 45% (24/53 turtles) escaped from experimental trap 2 (without bait cup). These results suggest that M. t. macrospilota smaller than the recommended Turtles Excluder Device (TED) size (45 mm x 120 mm) have a low probability of successfully escaping crab traps and that TEDs may not be the most successful tool for decreasing turtle mortality in crab traps. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, 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.
Thesis:
Thesis (M.S.)--University of Florida, 2015.
Local:
Adviser: JOHNSON,STEVEN ALBERT.
Local:
Co-adviser: FREDERICK,PETER C.
Statement of Responsibility:
by Eric Suarez.

Record Information

Source Institution:
UFRGP
Rights Management:
Copyright Suarez, Eric. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Classification:
LD1780 2015 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

ECOLOGY OF ORNATE DI AMONDBACK TERRAPINS (MALACLEMYS TERRAPIN MACROSPILOTA ) ON A SMALL GULF CO AST BARRIER ISLAND A ND THEIR BEHAVIOR INSIDE CRAB TRAPS By ERIC SUAREZ A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2015

PAGE 2

© 2015 Eric Suarez

PAGE 3

To my parents, Gilberto and Cynthia Suarez, for their constant encourag ement and unconditional support; without them, none of my success would be possible.

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my committee members: Steve Johnson, Peter Frederick, Kenneth Krysko, and Bill Tuner for their support. Wes Boone, Madeleine Cascarano, Bobbi Carpenter, Colleen Clossius , Branson Cone, Adam Casavant, Kevin Enge, Carolyn Enloe, Linda Harrison, Pearson Hil l, Linda Harrison, Amanda Heh, Natalie Lamneck , Tiffany Manteuffell, Elena Oehmig, Eric Phares, Amy Schwarzer, Veronica Stehouwer, and Travis Thomas assisted in the field. Ryan Gandy helped with study design. Jessica Burnett and Erin L eone pr ovided statistical consultation . Virginia Johnson and the Florida Fish and Wildlife field office in Carrabelle, FL provided housing. The Florida Fish and Wildlife Conservation Commission and the Whitney Lab provided equipment. I would also like to a cknowledge the Conservation Wildlife Tag Grant , Jennings Scholarship, Michael Moulton, and my parents for funding.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ........... 4 LIST OF TABLES ................................ ................................ ................................ ...................... 7 LIST OF FIGURES ................................ ................................ ................................ .................... 9 ABSTRACT ................................ ................................ ................................ ............................. 11 CHAPTER 1 BACKGROUND ON DIAMONDBACK TERRAPIN ( MALACLEMYS TERRAPIN ) ECOLOGY AND ECOLOGICAL THREATS ................................ ................................ ... 13 Diamondback Terrapin Ecology ................................ ................................ ...................... 13 Threats to Diamondback Terrapin Populations ................................ ................................ 15 Ecological Status of the Diamondback Terrapin ( Malaclemys terrapin ) in Florida .......... 18 Research Design and Thesis Structure ................................ ................................ ............. 19 2 DEVELOPMENT OF A STANDARDIZED AND ECOSYSTEM SPECIFIC POPULATION SURVEY METHODOLOGY FOR DETECTING ORNATE DIAMONDBACK TERRAPINS ( MALACLEMYS TERRAPIN MACROSPIL OTA ) ON BARRIER ISLANDS ................................ ................................ ................................ ......... 21 Introduction ................................ ................................ ................................ .................... 21 Methods ................................ ................................ ................................ .......................... 22 Study Site ................................ ................................ ................................ .................... 22 Sampling Design ................................ ................................ ................................ ......... 23 Statistical Analyses ................................ ................................ ................................ ..... 24 Results ................................ ................................ ................................ ............................ 25 Discussion ................................ ................................ ................................ ...................... 26 Environmental Va riables Influencing CPUE ................................ ................................ 26 Methods of Capture ................................ ................................ ................................ ..... 28 Habitat Use ................................ ................................ ................................ ................. 29 3 POPULATION ASSESMENT OF THE ORNATE DIAMON DBACK TERRAPIN ( MALACLEMYS TERRAPIN MACROSPILOTA ) ON A SMALL GULF CO AST BARRIER ISLAND ................................ ................................ ................................ ........... 35 Introduction ................................ ................................ ................................ .................... 35 Methods ................................ ................................ ................................ .......................... 36 Sampling Design ................................ ................................ ................................ ......... 36 Collection of Diamondback Terrapin Population Assessment Data for Pilot and Main Studies ................................ ................................ ................................ ............ 37 Statistical Analyses ................................ ................................ ................................ ..... 37

PAGE 6

6 Results ................................ ................................ ................................ ............................ 38 Pilot Study ................................ ................................ ................................ .................. 38 Main Study ................................ ................................ ................................ .................. 38 Discussion ................................ ................................ ................................ ...................... 39 Population Size and Structure ................................ ................................ ...................... 39 Sex Ratio ................................ ................................ ................................ ..................... 41 Body Size ................................ ................................ ................................ .................... 42 4 INVESTIGATION OF THE INTERACTIONS BETWEEN THE DIAMONDBACK TERRAPIN ( MALACLEMYS TERRAPIN MACROSPILOTA ) AND BLUE CRAB TRAPS ................................ ................................ ................................ ............................... 50 Introduction ................................ ................................ ................................ .................... 50 Methods ................................ ................................ ................................ .......................... 51 Pilot Study ................................ ................................ ................................ .................. 53 Main Study ................................ ................................ ................................ .................. 53 Statistical Analysis for Pilot Study ................................ ................................ .............. 54 Statistical Analysis for Main Study ................................ ................................ .............. 54 Results ................................ ................................ ................................ ............................ 55 Pilot Studies ................................ ................................ ................................ ................ 55 Main Study ................................ ................................ ................................ .................. 55 Discussion ................................ ................................ ................................ ...................... 56 5 CONSERVATION IMPLICATIONS AND FUTURE DIRECTIONS ............................... 64 LIST OF REFERENCES ................................ ................................ ................................ .......... 67 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ..... 75

PAGE 7

7 LIST OF TABLES Table page 2 1 Total captures, person minutes, and CPUE at different tide levels. ................................ . 31 2 2 Total captures, person minutes, and CPUE at different weather patterns. ....................... 31 2 3 Total captures of in different microhabitat types. ................................ ........................... 31 2 4 Results from the forward and backward stepwise multiple regression analyzing the affect of tide, weather pattern, and air temperature (daily high) on CPUE with AIC values for best model selection. Output from Program R. ................................ ............... 31 2 5 Comparison of CPUE among different sampling techniques throughout Malaclemys terrapin range. CPUE = terrapin catch/day (with specific habitat or trap type) unless noted otherwise. Most of the content from this table was taken from Selman and Baccigalopi (2012) . ................................ ................................ ................................ ....... 32 3 1 Total captures, person minutes, and CPUE at different sampling sections of Lanark Reef. ................................ ................................ ................................ .............................. 44 3 2 Morphometric measurements and mass from Malaclemys terrapin macrospilota captured in the main study. Data are presented as means (minimum maximum). ......... 44 3 3 Morphometric measurements and mass for each adult Malaclemys terrapin macrospilota in the main study. Data are presented as mean (minimum maximum). Raw data for males and females were compared using two sample t te st and sexual dimorphism index (SDI). ................................ ................................ ............................... 44 3 4 Mean PL lengths for adult male and female Malaclemys terrapin throughout their range. ................................ ................................ ................................ ............................ 45 3 5 Female PL length at sexual maturity for Malaclemys terrapin throughout their range. ... 45 3 6 Population estimates reported for Malaclemys terrapin throughout their range. Estimates are either reported alone, with standard error, or with confidence intervals. ... 45 3 7 Female maximum PL length for Malaclemys terrapin in Florida. ................................ .. 46 4 1 Number of Malaclemys terrapin macrospilota that escaped and did not escape from crab traps during the first pilot study. ................................ ................................ ............. 60 4 2 Number of Malaclemys terrapin macrospilota that escaped and did not escape from crab traps during the second pilot study. ................................ ................................ ........ 60 4 3 Number of Malaclemys terrapin macrospilota that escaped and did not escape from crab traps during the main study. ................................ ................................ ................... 60

PAGE 8

8 4 4 Morphometric measurements and mass from Malaclemys terrapin macrospilota that escaped and did not escape from crab traps during the main study. Data prese nted as mean (minimum maximum). ................................ ................................ ....................... 60

PAGE 9

9 LIST OF FIGURES Figure page 2 1 Map of Lanark Reef. ................................ ................................ ................................ ...... 33 2 2 Map of Lanark Reef showing the three sampling sections. ................................ ............. 33 2 3 Scatterplot showing the number of Malaclemys terrapin macrospilota captured in dry habitat within vegetation and dry habitat within vegetation under tidal wrack at different daily high air temperatures over the main study period. ................................ ... 34 3 1 Captured Malaclemys terrapin macrospilota by size class in Section 3 at Lanark Reef. ................................ ................................ ................................ .............................. 47 3 2 Captured Malaclemys terrapin macrospilota by size class in Section 2 at Lanark Reef. ................................ ................................ ................................ .............................. 47 3 3 Captured Malaclemys terrapin macrospilota by size class in Section 1 at Lanark Reef. ................................ ................................ ................................ .............................. 48 3 4 Histogram showing Malaclemys terrapin macrospilota captured each m onth by size class and sex at Lanark Reef. ................................ ................................ ......................... 48 3 5 Number of gravid Malaclemys terrapin macrospilota captured each month at Lanark Reef. ................................ ................................ ................................ .............................. 49 4 1 Standard commercial blue crab trap with labeled components. ................................ ....... 61 4 2 Control trap (standard commercial blue crab trap) used in experiment of Malaclemys terrapin macrospilota. Note the center baffle and bait cup are intact. ............................. 61 4 3 Experimental trap 1 used in experiment of Malaclemys terrapin macrospilota. Note the center baffle has been removed and th e bait cup is still intact. ................................ .. 62 4 4 Experimental trap 2 used in experiment of Malaclemys terrapin macrospilota . Note the center baffle is still intact and the bait cup has been removed. ................................ .. 62 4 5 Multiple pairwise comparison of mean escape time of Malaclemys t errapin macrospilota by trap type with standard error bars. Letters indicate significance. .......... 63 4 6 Modified crab trap with bait cup positioned flush with the plane of the center baffle. ..... 63

PAGE 10

10 LIST OF ABBREVIATIONS CL Mid line carapace length CPUE Catch per unit effort CW Carapace width Ex1 Experimental Trap 1 Ex2 Experimental Trap 2 FNAI F lorida Natural Areas Inventory Habitat Classification System FWC Florida Fish and Wildlife Conservation Commission HT Shell height PL Plastron length ppt Parts per thousand SDI Sexu al Dimorphism Index SSD Sexual size dimorphism TSD Temperature sex determination

PAGE 11

11 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science ECOLOGY OF ORNATE DI AMONDBACK TERRAPIN S ( MALACLEMYS TERRAPIN MACROSPILOTA ) ON A SMALL GULF CO AST BARRIER ISLAND A ND THEIR BEHAVIOR INSIDE CRAB TRAPS By Eric Suarez August 2015 Chair: Steven A. Johnson Co Chair: Peter C. Frederick Major: Wildlife Ecology and Conservation The diamondback terrapin ( Malaclemys terrapin ) has been experiencing population decline mostly through mortality in crab traps throughout its range . Population status is unclear, and the mechanisms by which they become entrapped are poorly known. To deter mine the ecological status of ornate diamondback terrapins ( Malaclemys terrapin macrospilota ) on a small Gulf Coast barrier island (Lanark Reef), a mark recapture study was conducted to effectively develop a survey methodology and assess the population. I captured 381 individual M. t. m acrospilota ( 32.8 % juveniles, 52.5 % a dult males, and 14.7 % adult females). Sex ratio was male biased (3.6:1) and there was an estimated 929.1 ± 142.1 turtles inhabiting Lanark Reef. Analysis of environmental variables show ed that the most efficient time to search for M. t. m acrospilota is during a low tide on clear or partly cloudy days (CPUE = 0.27 or 1 terrapin every 3.6 minutes). While surveying the intertidal zone of Lanark Reef, turtles were only found in microhabitats associated with some form of vegetation (smooth cordgrass, tidal wrack, or both). An experiment was developed to determine which inter nal component (e.g., center baffle or bait cup) of a commercial crab trap limits their ability to escape. Thirty percent (16/53 turtles) successfully escaped from the control trap, 23% (12/53 turtles) escaped from experimental trap 1

PAGE 12

12 (without center baffle), and 45% (24/53 turtles) escaped fro m experimental trap 2 (without bait cup). These results suggest that M. t . macrospi lota smaller than the recommended Turtles Excluder Device (TED) size ( 45 mm x 120 mm ) have a low probability of successfully escaping crab traps and that TEDs may not be the most successful tool for decreasing turtle mortality in crab traps.

PAGE 13

13 CHAPTER 1 BACKGROUND ON DIAMONDBACK TERRAPIN ( MALACLEMYS TERRAPIN ) ECOLOGY AND ECOLOGICAL THREATS Diamondback Terrapin Ecology There are only three genera ( Malaclemys , Callagur , and Orlitia ) of turtles specialized to brackish water habitats. Whereas Callagur and Orlitia occur in tropical systems in southeastern Asia, Malaclemys occurs in tropical, sub tropical, and temperate systems in eastern North America (Hart and Lee, 2006). Globally, brackish water systems such as estuaries have been suffering due to increased human impact on coastal areas (Dennison et al., 1993; Martinez et al., 2007). Costanza et al. (1997) ranked coastal waters as the number one provider of ecosystem services in the world. Estuaries offer an abu ndance of resources for a variety of organisms while serving as an important role in the functioning of coastal systems (Beck et al., 2001; Breine et al., 2007). With the exception of amphibians, reptiles are the least represented group of vertebrates that occur in estuarine habitats. The d iamondback terrapin ( Malaclemys terrapin ) ranges from Massachusetts southward to Texas (Ernst and Lovich, 2009 ). Within these brackish water systems, M. terrapin inhabit multiple habitat types, which include saltmarshes, barrier islands , creeks, rivers, and sandy beaches. Within estuarine habitats, M. terrapin can tolerate a wide range of salinities rang ing from 0 to 34 ppt (Robinson and Dunson, 1976). Its impermeable skin, the ability to regulate osmotic pressure, the pr esence of salt excreting glands, and other behavioral adaptations (e.g., drinking from the freshwater film after a rain event) have allowed this species to make a successful transition from freshwater to brackish water (Hart and Lee, 2006). By being the on ly semi aquatic turtle living in brackish habitats in North America, M. terrapin receive some major benefits by avoiding interspecific competition. Th us , M. terrapin hold a relatively high trophic position and exert a strong top down control within salt ma rsh and barrier island systems where

PAGE 14

14 the marsh periwinkle snail ( Littoraria irrorata ) c ould otherwise overgraze smooth cordgrass ( Spartina alterniflora ), which can lead to a decrease in overall biodiversity, increase in erosion , and an increase in habitat alteration (Hurd et al., 1979; Levesque, 2000; Silliman and Bertness, 2002). Malaclemys terrapin is a medium sized turtle that var ies by latitude and sex in body size, age at maturity, and behavior (Butler et al., 2006). Individuals are sexually dimorphic with adult females reaching sizes significantly larger than adult males. Mean male plastron length is 102 mm and they reach maturity between 4 to 7 years , whereas mean female plastron length is much larger at 148 mm and they reach maturity between 8 to 13 years of age (Lovich and Gibbons, 1990; Roosenburg, 1990). These differences in body size may lead to food partitioning between genders . However, both males and females have broad diet, which include snails, crab s , clams, mussels, annelid worms, fish, and plants; however, their diet is usually dependent on gender size dimorphism, size/age class, prey size and availability within a habitat (Tucker at al., 1995). Differences in body size along with climate also shap e M. terrapin nesting ecology. In more temperate climates , M. terrapin usually exhibit s a Type 1 nesting strategy, which entails larger clutches of small eggs; while M. terrapin in more tropical climates fall into the Type 2 nesting strategy, which entails smaller clutches of larger eggs (Moll, 1979 ). Laying multiple clutches has been observed in natural populations, but may be more common in populations at more tropical climates where their active season is longer (Hildebrand, 1932; Butler et al., 2006). C limate is also important in determining the length of dormancy during colder months . Many turtle species with broad latitudinal ranges in North America can remain dormant for more than half the year at the northern end of their range, while turtles at the southern end of their range will most likely have a shorter dormancy period (if any) with periods of activity during the

PAGE 15

15 colder months if conditions are tolerable (Ultsch, 2006). The relatively warm winter months in northern Florida could have important im plications for areas where M. terrapin are active most of the year and could potentially experience year round threats . Threats to Diamondback Terrapin Populations Turtles, along with other reptiles are experiencing global declines (Gibbons et al., 2000). T he ecological status of M alaclemys terrapin is unclear; however, researchers have reported declines in local populations in New York, New Jersey, Maryland, North Carolina, South Carolina, Florida (Atlantic Coast), Louisiana, and Mississippi ( Seigel and G ibbons, 1995; Dorcas et al., 2007). Currently, th is species has no federal protection in the United States . The IUCN Red List for Threatened Species categorizes M. terrapin as lower risk/near threatened (Tortoise & Freshwater Turtle Specialist Group, 1996) and th is species has recently been petitioned for listing by CITES under Appendix II . Nevertheless, population s continue to decline because of crab trap mortality, harvest for meat, habitat loss and degradation, predation (i.e. , via raccoons [ Procyon loto r ] , feral hogs [ Sus scrofa ] ) , and other anthropogenic processes such as boat and road mortality and collection for the pet trade ( Seigel and Gibbons, 1995; Butler et al., 2006; Ernst and Lovich, 2009). From the late 1800s through the mid 1900s, M . terrapin was harvested as a gourmet food item and populations were negatively impacted throughout much of their range (McCauley, 1945; Carr, 1952; Gibbons et al., 2001). In some areas, M. terrapin is still collected for personal consumption (Drabeck et al. , 2014), and similar to other North American turtles, it is exported to Asia for food and to Europe as pets (Green et al., 2010; Pfau and Roosenburg, 2010). Although harvest for consumption may still be a threat to local populations, M. terrapin is usually encountered as bycatch and are not targeted by fishermen (Roosenburg, 1990). Collection for the

PAGE 16

16 pet trade is a serious threat as these turtles can be rare to find and are aesthetically appealing because they vary remarkably in phenotype. According to the American Pet Products Manufacturers Association, in 2000 more than 3.9 million households had at least one reptile or amphibian as a pet in the United States. In 1997, turtles were the most exported reptile (96.6 %) and second in imports (15.0 %) , and it i s estimated that the trade in live herpetofauna and related products is worth at least two billion dollars annually (Franke and Telecky, 2001). From 1990 1992, 97 % (176 individuals) of all reported M. terrapin taken from the wild in Florida and sold in t he pet trade originated from the panhandle region of Florida , and when 1993 and 1994 are included, 64% (437 individuals) were removed from the Florida panhandle (Enge, 1993; Enge, 2005). Although exports of turtles has decrease d from 2002 2012 in Louisiana , California, Texas, and Florida, export numbers are still high (e.g., > 1 million wild turtles exported in 2009 from Louisiana) (Mali et al., 2014). Habitat loss and degradation along coastal ecosystems has been another threat to M . terrapin throughout i ts range. Over one third of the global human population resides within 100 km of a coastline (Cohen et al., 1997). With increasing human populations worldwide and the constant pressure to develop natural areas, the negative effects associated with human impacts (i.e. , habitat loss and degradation) on coastal ecosystems are likely to continue and possibly worsen (Stallings, 2009). Coastal regions comprise of only 17% of the contiguous land area of the United States, but 53% of the U.S. human population inh abits these areas, with population densities reaching their highest on the Atlantic Coast (Wilke et al., 2007). With the major increase in human population in the last 60 years, the large number of humans moving to the Atlantic Coast has accelerated habita t loss and degradation in areas where M. terrapin occur s

PAGE 17

17 (Hart and Lee, 2006). Loss of nesting beaches and alteration of foraging habitat could be detrimental to recruitment in local M. terrapin populations. L ike many other turtle species , M . terrapin is highly susceptible to predation by mammals and birds, especially while in the egg stage and after hatching. Predation on M. terrapin has been observed by a wide range of taxa , including but not limited to raccoons ( Procyon lotor ), ghost crabs ( Ocypode spp. ) , Bald E agles ( Haliaeetus leucocephalus ), herons and egrets (Family: Ardeidae ), Ruddy Turnstones ( Arenaria interpres ), feral hogs ( Sus scrofa ), foxes (Family: Canidae), river otters ( Lontra canadensis ), striped skunks ( Mephitis mephitis ), crows ( Covus spp . ), Laughing Gulls ( Leucophaeus atricilla ), rats ( Rattus spp. ); however, raccoons are likely the major predator of M. terrapin (Butler et al., 2006). In Florida, nearly 200 vertebrate species are known to be negatively impacted by P. lotor (MacLaren, 1992; Munscher et al., 2012). Although P. lotor and M. terrapin naturally share habitat, lack of harvest and the waste produced by humans (e.g., food) has perhaps increased P. lotor populations in such areas (Hoffman and Gottschang, 1977; Gehr t et al., 2002). Procyon lotor can prey on or injure every size/age class of M. terrapin . Many biologists agree that crab traps are the primary artificial threat to M. terrapin populations range wide (Butler et al., 2006). Throughout its range, M . terrapin share s habitat with the blue crab ( Callinectes sapidus ), and turtle mortality by means of drowning in man made crab traps has been known for over 50 years (Roosenburg, 2004, Davis, 1942 in Butler and Heinrich, 2007). A study in Maryland reported annual pop ulation mortality rates varying from 15% to 78% (Roosenburg et al., 1997). This variation is influenced by body size, season, population structure, and behavior (e.g., habitat use) (Roosenburg, 2004; Hart and Crowder, 2011) . Some states (New Jersey, Delawa re, Maryland) have adopted regulations for the

PAGE 18

18 implementation of turtle excluder devices (TEDs) on recreational and commercial (only New Jersey) blue crab traps to help reduce M. terrapin bycatch; however, it is important to note that the mandated implemen tation of TEDs may also affect the livelihood of crab fisherman. Ecological Status of t he Diamondback Terrapin ( Malaclemys terrapin ) i n Florida Malaclemys terrapin and accounts for 20% of the species total range (Butler et al., 2006). M. terrapin Wildlife Action Plan (Florida Fish and Wildlife Conservation Commission, 201 2). According to the Florida Fish and Conservation Commission (FWC), M. terrapin ha s no conservation status in Florida, but it does have a possession limit of two individuals . The ornate diamondback terrapin ( Malaclemys terrapin macrospilota ) inhabits bar rier islands, tidal creeks, coastal salt marshes, and mangrove islands from the Florida Keys, Monroe County , northward to Walton County in the Florida panhandle (Butler et al., 2006). According to Millsap et al., (1990), M. terrapin in Florida ha s one of t he highest biological scores, and is listed examined vertebrates (excluding marine fishes and marine vertebrates). According to this ranking system, M. terrapin ha s a median score between species listed as threatene d and endangered. The Florida Natural Areas Inventory (FNAI) recommends that statewide monitoring and population surveys are needed (Hipes et al., 2001). Still, of the three endemic subspecies that occur in Florida, the least is known about the biology and ecology of M. t. macrospilota ( Wood, 1992; Butler and Heinrich, 2013). In 2009, FWC began survey ing a population of M. t. macrospilota in Franklin County, Florida , and concluded that this population may be one of the largest and densest in the state.

PAGE 19

19 Res earch Design and Thesis S tructure There are many gaps in knowledge of Malaclemys terrapin throughout its range , but mostly in populations in the Gulf of Mexico. The overall aim of my study wa s to provide information on effective sampling techniques for monitoring populations of M. terrapin in the Florida panhandle and northwest ern peninsula, as well as providing information on its behavior inside a crab trap. The research I conducted for my th Chapter 5 explore s future research needed and provides recommendations for the conservation and management throughout the range of M. terrapin . Chapter 2 is a study developed to understand how the ornate diamondb ack terrapin ( Malaclemys terrapin macrospilota ) is spatially and temporally distributed on a small Gulf Coast barrier island (Lanark Reef) with regard to what microhabitats it is using. The objective of my study was to observe how certain environmental con ditions and habitat use affect capture rates of M. t. macrospilota during the spring and summer. I researched how ambient air temperature, tides, and weather patterns affect M. t. macrospilota catch per unit effort (CPUE). With these data, I proposed a sta ndardized and ecosystem specific population survey methodology for locating M. t. macrospilota on barrier island habitats. Chapter 3 is a study developed to assess the population of M . t. macrospilota inhabiting a small Gulf C oast barrier island . I designed a mark recapture study to better understand population dynamics with a primary goal of estimating population size. Other goals were to obtain demographic, sex ratio, and population structure data for comparison with other populations of M. ter rapin throughout its range. Chapter 4 is an experimental designed to observe M. terrapin behavior once inside a human constructed artificial crab trap. I specifically wanted to determine if M. t errapin could escape once inside a crab trap, while determinin g which internal component hindered escape the

PAGE 20

20 most. My results are then used to make recommendations on possible trap modifications to commercial trappers in order to facilitate M. terrapin escape.

PAGE 21

21 CHAPTER 2 DEVELOPMENT OF A STANDARDIZED AND ECOSYSTEM SPECIFIC POPULATION SURVEY METHODOLOGY FOR DETECTING ORNATE DIAMONDBACK TERRAPINS ( MALACLEMYS TERRAPIN MACROSPILOTA ) ON BARRIER ISLANDS Introduction A recurring problem in evaluating the need for management acti on due to the impact of crab trapping pressure, predation, and other anthropogenic effects on diamondback terrapin ( Malaclemys terrapin ) populations is the lack of population assessments. In order to effectively assess M. terrapin populations, the developm ent of a standardized survey methodology is crucial. Known s ampling techniques include seine nets (Lovich and Gibbons, 1990; Butler, 2000; Levesque, 2000; King and Ludlam, 2014), trammel nets (Dorcas et al., 2007), crab traps (Bishop, 1983; Butler and Hein rich, 2007; Hart and Crowder, 2011), modified crab traps (Mann, 1995; Roosenburg et al., 1997; Selman and Baccigalopi, 2012 ), active searching on boat ( Selman and Baccigalopi, 2012), head count surveys (Butler, 2002; Harden et al., 2009; Coleman, 2011), fyke nets (Selman and Baccigalopi, 2012; Selman et al., 2014), dip nets (Hart, 2005), otter trawls (Butler, 2000), cast netting (Butler, 2002), and eel pots (Radzio and Roosenburg, 2005). Active searching methods that involve walking on land or in shallow water to capture M. terrapin are limited in the literature (Boykin, 2004; Haskett, 2011). Most studies that report actively searching for M. terrapin focus only on locating nest sites rather than individuals within a population (Burger , 1976 ; Roosenburg, 1 992, 1994; Butler et al., 2004; Clowes, 2013; Mitchell and Walls, 2013) . In 2010 , FWC initiated population survey s of ornate diamondback terrapins ( Malaclemys terrapin macrospilota ) on a barrier island (Lanark Reef) in the Gulf of Mexico. By the summer of 2012, approximately 300 turtles had been marked, which allowed me to begin develop ing a standardized search methodology for detecting M. t. macrospilota that would be applicable to

PAGE 22

22 similar habitats in the Florida panhandle and northwest ern peninsula . Firs t, I determine d environmental conditions and habitat variables that were best for detecting and maximizing capture probability. Butler et al. (2006) suggested that tides could have a strong effect on habitat use . In South Carolina and Maryland, Tucker et a l. (1995) and Roosenburg et al. (1999) observed a difference in habitat use based on prey availability and sex of turtles. These studies suggest that environmental factors affect habitat use. Those environmental data gathered from this study can give us va luable insight on where to better search for and capture M. t. macrospilota while providing information on which habitats are crucial for their survival. T he focus of this Chapter is to highlight the importance of understanding how environmental factors affect behavior, numbers of individuals available to be counted , and spatial and temporal distribution of M . terrapin , which are extremely valuable in reducing fieldwork effort as well as providing sampling consistency . The standardized survey technique in this Chapter was developed to provide the sampling regime used to capture M. t. macrospilota for the population assessment and demographic description in Chapter 3. Methods Study Site The study is a small barrier island (Lanark Reef) located in the Gulf of Mexico, Figure 2 1). It stretches approximately 11 km east to west, parallel to the coast , and is periodically partially submerged , which results in a variation in the dry land area from 2 to 29 hectares (ha) . In 2012, the Audubon Society of Florida pur chased Lanark Reef to protect nesting shorebirds. For sampling purposes, I designated areas that remain emergent year round as Section 1, Section 2, and Section 3 (Figure 2 2). At a medium tide, Section 1 is approximately 7.0 h a , Section 2 approximately 0. 90 h a , and Section 3 approximately 0.67 h a . Lanark Reef is composed of salt marsh and salt flats as designated by the FNAI habitat classification system. Microh abitats

PAGE 23

23 sampled include d grasses, shrubs, shallow saline pools, and open sandy patches. The west ern side of the island is mostly sand and mud flats with patchy vegetation , whereas t he eastern side is almost completely covered by vegetation with a few bare sand or mud fl ats. The dominant vegetation is Smooth Cordgrass ( Spartina alterniflora ). Surrounding Thalassia spp . beds and other aquatic plants provide much of the vegetation that washes ashore in the form of tidal wrack (FNAI, 2010). The tidal wrack washes up onto the vegetation during high tide and is deposited at different elevations a s the tide recedes forming shaded areas . The soil types found on Lanark Reef are Mandarin Resota Leon and Duckston sand (Sasser et al., 1994). Sampling Design I captured M alaclemys t errapin macrospilota by hand while actively searching by walking in the in tertidal zone from 12 September 1 through 15 November 2012 (pilot study) and 9 April through 30 July 2013 (main study). No sampling occurred from December 2012 through February 2013 because M. t. macrospilota was not observed on Lanark Reef during these mon ths and preliminary telemetry data show that M. t. macrospilota overwinters at a different location (Eric Suarez and Travis Thomas, unpublished data). Water temperature, associated microhabitat, and behavioral observations were recorded for each turtle cap tured. Depending on weather, p atrolling for M. t. macrospilota occurred during daylight hours (i.e., especially midday; range 1000 1730) on high, low, and medium tides. I searched for M. t. macrospilota throughout the intertidal zone within patches of ve getation, under tidal wrack, in open sandy patches, and shallow pools created by the incoming tide. Wrack suspended on vegetation creates cover from predators and may potentially serve as a shaded thermal refuge. I generally avoided areas where wrack was c ompacted or was beyond the high tide zone. I searched the wrack by moving it or simply feeling it with my feet while walking . While both methods were effective, I preferred using my feet because it was less arduous , faster , and caused the least amou nt of disturbance to

PAGE 24

24 the habitat . Wrack and vegetation searches within the intertidal zone were the only method employed on Lanark Reef , because of vegetation, bird activity, and bird colony restrictions made it impossible to see crawls or other signs of M . t. macrospilota . Because gravid females and hatchling M. t. macrospilota have been found on Lanark Reef, it seems likely that nesting is occurring here; however, I did not perform nest searches in this study. Each of the three sections of Lanark Reef ar e different in size and shape, therefore from September through November 2012 and 4 March through 1 April 2013 the intertidal zones of Section 1 and Section 2 were sampled multiple times per week to allocate the appropriate sampling effort for each section . S ampling effort was based on the time it would take to sample the entire intertidal zone by walking a wandering patterned transect. During the development of standardizing sampling effort, on 20 March 2013, I discovered a new section connected by a sandb ar (Section 3) approximately 1,600 meters to the east of Section 2 (Figure 2 2). From 20 March through 8 April 2013 , the intertidal zone of Section 3 was searched to allocate the appropriate sampling effort. In order to effectively and consistently search the intertidal zone at each sampling occasion during the main study, I standardized time spent searching corrected for section size based on those pilot sampling events for each section (Section 1 [60 minutes], Section 2 [30 minutes], and Section 3 [30 min utes]). Statistical Analyses During each sampling period, tide level (low, medium [incoming or outgoing], high), weather pattern (cloudy, partly cloudy, clear), and associated microhabitat data (e.g., type of vegetation and water temperature [during high and medium tides]) were recorded. Daily high ambient air temperatures were taken from the nearest weather station (Apalachicola Regional Airport Weather Station) . I used CPUE (individuals captured/minute) to determine what tide levels , weather patterns , an d ambient air temperature (daily high) allowed for the most eff ective

PAGE 25

25 sampling. This metric was determined by dividing the number of individuals captured by the total number of person minutes (1 person searching for 1 minute = 1 person minute). All data we re analyzed for normality using the Shapiro Wilk normality test. A chi square goodness of fit analysis was used to determine if captures differed significantly between microhabitats (vegetation, vegetation with tidal wrack). A one tailed Wilcoxon signed ra nk test was used to determine if there was a significant difference in CPUE that was related to daily high air temperature. A Kendall Tau correlation was run to determine the relationship between daily high air temperatures and CPUE. Another Kendall Tau correlation was used to determine the relationship between water temperature upon capture and ambient air temperature (daily high ). A forward and backward stepwise multiple regression model was used to observe the combined effects of ti de levels , weather patterns , and ambient air temperature (daily high) on CPUE. I conducted all analyses using Program R statistical software (R Development Core Team, 2015 ) . Results The highest CPUE recorded during sampling at different tides was during low tide ( 0.27 or 1 terrapin every 3.6 minutes; Table 2 1). The highest CPUE recorded during sampling at different weather patterns was during partly cloudy days ( 0.30 or 1 terrapin ev ery 3.5 minutes; Table 2 2). There was a positive correlation between ambient air temperature (daily high) and CPUE (T b = 0.221, p = 0.029). Ambient air temperature (daily high) significantly affected CPUE (V = 1326, p < 0.0001). Water temperature recorded upon capture within the intertidal zone ranged from 23.6 °C 33.4 ° C (mean = 28.2 ° C). There was a positive correlation between ambient air and water temperature s ( T b = 0. 684, p <0.0001) . All M alaclemys t errapin macrospilota captured were found in microhabitats associated with some form of vegetation ( s mooth cordgrass patch with or without tidal wrack). Individuals were mostly found in dry habitat within

PAGE 26

26 vegetation under tidal wrack (Table 2 4 ). Total captures among microhabitat types varied signifi cantly ( 2 = 84.87, p < 0.001). Based on standard error and AIC values, the best model for predicting CPUE was the model that includes tide and weather effects, but excludes ambient air temperature (daily high) (Table 2 4). Discussion Environmental Variab les Influencing CPUE I determined the best conditions to observe M alaclemys t errapin macrospilota on Lanark Reef by using the highest CPUE for each environmental variable recorded and results from a microhabitat statistical analysis. Based on th e se results , I determined low tide, on hot, clear or partly cloudy days, within vegetation patches (preferably in Spartina alterniflora vegetation patches) and under tidal wrack to be the best conditions to search for M. t. macrospilota from April to November. Althou gh ambient air temperature (daily high) and CPUE differed significantly and were positively correlated, the correlation was weak. However, there was a strong correlation between ambient air temperature (daily high) and water temperature upon captures suggesting that M. t. macrospilota could be using the shaded microhabitats for thermoregulatory purposes. The strong correlation observed is due to the increase in both ambient air and water temperature as summer progres ses. Nevertheless, when searching during low tide, as temperature increase s , more turtles chose microhabitats that included tidal wrack (Figure 2 3 ). This observation could just be coincidental since more tidal wrack is present on Lanark Reef as air and wa ter temperatures get warmer . The stepwise multiple regression analysis shows that the variables most strongly influencing CPUE are tide level and weather pattern, suggesting that the effect of tide and weather are stronger predictors of capture rate than a ir with the lower AIC value because the standard error of air temperature is larger than the

PAGE 27

27 estimate, defining the estimate less accurate . The regression also shows how low tide has a significant positive effect on CP UE and cloudy weather has a significant negative effect on CPUE (Table 2 4). Overall CPUE (0.22 turtles /min ute, 13.2 turtles /hour or 1 turtle every 4.5 minutes) in my study on M . t. macrospilota was the highest reported for any study on M alaclemys terrapin . Other studies that used different sampling techniques reported much lower CPUE (Table 2 5). Boykin (2004) reported a lower CPUE (0.5 turtles/hour) using hand capture techniques, which could be due to difficulty in visual sampling i n mangrove habitats. In Texas, Haskett (2011) also hand captured turtles and reported a CPUE of 1.22 turtles/hour or 1 turtle every 49 minutes , which is also much lower than what was reported in my study. CPUE compared among other studies throughout M. terrapin range using different capture methods (e.g., otter trawls, seine nets, modified crab traps, trammel nets, dip nets, fyke nets) are presented in Table 2 5. Regardless of the different habitat types and environmental conditions in other studies, it is worthy to note what other studies have reported for the conditions under which they had the greatest amount of captures. Butler (2000) suggested that the best time to hand capture M alaclemys t erra pin centrata was just prior to through one hour after high tide. When comparing the total number of captures during high tide in my study (n = 115), it is still substantially greater than the total captures in Butler (2000) (n = 47), which suggest that hig h tide may be a good time to search for M. terrapin , but not the optimal time to search for M. terrapin. Butler (2000) did not report captures or effort at different tide levels. Estep (2005) reported that the greatest numbers of M. terrapin were observed at low tide, which is consistent with my findings. Hart and McIvor (2008) reported a large number of captures (n=300) while searching (by use of crab traps and dip nets) at low tide; however, they did not sample at other

PAGE 28

28 tide levels to compare captures at different tide levels. Harden et al. (2009) reported results on head count surveys under various environmental conditions. Their results were similar to this study in that they observed more individuals at low tide and under less cloud cover, suggesting t hat those two environmental variables are the most important in detecting M . terrapin using head counts surveys. Butler (2002) reported an increase in head counts with increasing water temperature and Avissar (2006) reported a decrease in activity with dec reasing water temperatures. In my study, I saw an increase in CPUE with increasing water and ambient air temperatures, but this could be due to the increase in tidal wrack abundance on Lanark Reef as temperatures warm, which makes it easier to detect M. t. macrospilota . Haskett (2011) and Hogan (2003) concluded that ambient air temperature was not a significant factor affecting CPUE in their sites, which is consistent with my findings. Methods of Capture Published c apture methods vary greatly making it difficult to compare capture rates among studies. Butler (2000) reported that otter trawling was most efficient, but capturing Carolina diamondback terrapins ( Malaclemys terrapin centrata ) by hand while searching on fo ot was the most successful technique as it was in my study (also see Seigel, 1984; Auger, 1989; Mann, 1995; Boykin, 2004 ; Mealey et al., 2014 ). Due to M . t . macrospilota habitat use on Lanark Reef, m ethods of capture such as crab trapping, trawling, seine netting, and fyke netting may be effective, but may not be the best method. Head count surveys within the intertidal zone was also not applicable in my study because of the dense vegetation at the site. Outside the intertidal zone, head count surveys will also be difficult because of heavy boat traffic and wave action. Each search method may be habitat specific; however, for population s that inhabit barrier islands in the Florida panhandle and the northwestern peninsula , searching by foot under environment al conditions presented here in may be most efficient .

PAGE 29

29 Habitat Use Habitat use by M. t. macrospilota in this study is similar to what has been observed elsewhere. Spivey (1998) and Harden et al. (2007) reported that M. terrapin generally avoided open water habitats, which is consistent with what has been observed in this study. Mealey et al. (2014) did not report if Malaclemys terrapin rhi zophorarum avoided open water habitats, but did have 97% of his captures inside (i.e., among the pneum atophores) or around black mangroves ( Avicennia germinans ), which is the dominant mangrove in the interior of the mangrove islands . Tucker et al. (1995) observed female turtle s entering marsh habitat to forage du ring the incoming tide in South Carolina, an d I observed multiple males and females eating marsh periwinkle ( Littoraria irrorata ) during high tide and fiddler crabs ( Uca spp. ) during low tide. Harden et al. (2007) suggested that turtles were using marsh habitats to avoid predators and for thermoreg u latory purposes in South Carolina, which could be occurring on Lanark Reef since there is lots of soft substrate and cover for thermoregulatory purposes, and plenty of S. alterniflora in which to forage. While all turtles were captured within the intertida l zone, it is still unknown how much they use habitat outside the intertidal zone (e.g., deeper/open water). A modified crab trap study reported only two captures outside the intertidal zone (water depth range 1.2 2.1 m) where traps were checked daily ( 2 925 trap days) ( Ryan Gandy, unpublished data ). In the same study, when traps were placed in the intertidal zone, 65 turtles were captured in 330 trap days. This does not necessarily mean that M. t. macrospilota are not using habitats outside the intertidal zone, but it suggest s that M. t. macrospilota may not be foraging or looking for structure (i.e. , cover) outside the intertidal zone and therefore not interested in entering those traps. They may be familiar w ith traps and therefore do not enter them, but unless they are interacting with traps elsewhere, it is unlikely that this is occurring around Lanark Reef (Ryan Gandy, unpublished data).

PAGE 30

30 My study shows an effective and efficient technique for surveying M. t. macrospilota at Lanark Reef and potentially other barrier island habitats in the panhandle of Florida. Searching on foot during low tide on clear or partly cloudy days were the most favorable conditions to capture M. t. macrospilota . Under these conditi ons in the warmer months of the year, searching in microhabitats that include tidal wrack and S. alterniflora would be most appropriate.

PAGE 31

31 Table 2 1. Total captures, person minutes, and CPUE at different tide levels. Tide Total Captures Person Minutes CPU E Low 272 990 0.27 (1 terrapin every 3.6 minutes) Medium 107 540 0.20 (1 terrapin every 5.0 minutes) High 115 720 0.19 (1 terrapin every 6.3 minutes) Table 2 2. Total captures, person minutes, and CPUE at different weather patterns. Weather Total Captures Person Minutes CPUE Clear 215 960 0.27 (1 terrapin every 4.5 minutes) Partly Cloudy 191 660 0.30 (1 terrapin every 3.5 minutes) Cloudy 88 630 0.14 (1 terrapin every 7.2 minutes) Table 2 3. Total captures of in different microhabitat types. Microhabitat Type Total Captures Dry habitat within vegetation* 43 Dry habitat within vegetation under wrack* 286 Wet habitat within vegetation** 93 Wet habitat within vegetation under wrack** 67 *Turtles captured in dry habitats were sampled during period of low and medium tides. **Turtles captured in wet habitats were sampled during medium and high tides. Table 2 4. Results from the forward and backward stepwise multiple regression analyzing the effect of tide, weather pattern, and air temperature (daily high) on CPUE with AIC values for best model selection. Output from Program R. Variables: 1. CPUE (Dependent Variable) 2. Weather Pattern (Independent Variable) 3. Tide Level (Independent Variable) 4. Air Temperature (Independent Variable) Model : CPUE ~ Weather Pattern + Tide Level + Air Temp Coefficients Estimate Std. Error t value p value (Intercept) 0.095633 0.172266 0.555 0.5815 air temp 0.001391 0.002089 0.666 0.5089 weather(cloudy) 0.091604 0.042046 2.179 0.0346 * weather (partly cloudy) 0.059736 0.040267 1.483 0.1449 tide (low) 0.084141 0.040847 2.060 0.0452 * tide (medium) 0.005300 0.043352 0.122 0.9032 Residual standard error: 0.113 on 45 degrees of freedom Multiple R squared: 0.4223, Adjusted R squared: 0.3581 F statistic: 6.58 on 5 and 45 DF, p value: 0.0001131 AIC= 218.32 (CPUE~Weather+Tide Level AIC= 218.82 (CPUE~Weather+Tide Level+Air Temp) * significant at p = 0.05

PAGE 32

3 2 Table 2 5. Comparison of CPUE among different sampling techniques throughout Malaclemys t errapin range . CPUE = terrapin catch/day (with specific habitat or trap type) unless noted otherwise. Most of the content from this table was taken from Selman and Baccigalopi ( 2012 ) . Capture Technique Study State CPUE Crab Trap Avissar 2006 Bishop 1983 Butler and Heinrich 2007 Cuevas et al. 2000 Guillory and Prejean 1998 Hart and Crowder 2011 Hoyle and Gibbons 2000 Rook et al. 2010 Morris et al. 2011 Roosenburg and Green 2000 Wood 1997 NJ SC FL MS LA NC SC VA VA MD NJ 0.15 0.16 0.007 0.147 0.0 0.0 0.003 0.008 0.027 0.20 0.10 0.044 0.23 0.15 0.48 Modified Crab Trap Butler 2000 Butler 2002 Hogan 2003 Mann 1995 Roosenburg 1992 Roosenburg et al. 1997 Smeenk 2010 Selman and Baccigalopi 2012 FL FL TX MS MD MD MD LA 0.047 0.013 0.025 0.0 0.87 0.163 0.48 1.2 0.17 0.38 0.0 Manual Searching Mangroves Tidal Ponds Land or Shallow Water Boykin 2004 Selman and Baccigalopi 2012 Haskett 2011 This Study FL LA TX FL 0.5/h 1.22/h 3.64/boat h 13.2/person h Trammel/Gill Net Butler 2000 Levesque 2000 FL SC 0.13/h 1.81/net set Surface Hoop Net/Crab Trap Mann 1995 Butler 2000 MS FL 0.0 0.0 Fyke Net Selman and Baccigalopi 2012 LA 0.5 0.8 Nesting Beach Pitfall Borden and Langford 2008 AL 0.05 Otter Trawl Butler 2000 FL 0.23 0.77/h Cast Netting Butler 2002 FL 1.9/h Eel Pot Radzio and Roosenburg 2005 MD 0.458

PAGE 33

33 Fig ure 2 1. Map of Lanark Reef. Figure 2 2. Map of Lanark Reef showing the three sampling sections .

PAGE 34

34 Figure 2 3. Scatterplot showing the number of Malaclemys terrapin macrospilota captured in dry habitat within vegetation and dry habitat within vegetation under tidal wrack at different daily high air temperature s over the main study period.

PAGE 35

35 CHAPTER 3 POPULATION ASSESMENT OF THE ORNATE DIAMON DBACK TERRAPIN ( MALACLEMYS TERRAPIN MACROSPILOTA ) ON A SMALL GULF CO AST BARRIER ISLAND Introduction Turtles, along with other reptiles are experiencing global declines (Gibbons et al., 2000). Diamondback terrapin ( Malaclemys terrapin ) populations have been reported as declining along the Atlantic Coast (Seigel, 1993; Seigel and Gibbons, 1995; Dorcas et al., 2007); however, it is uncertain whether M. terrapin populations are declining throughout Florida (Butler et al., 2006). Threats to M. terrapin populations include incidental drowning in human constructed artificial crab traps, predation, habitat loss, and illegal overexploitation across their entire range ( Seigel and Gibbons, 1995; Butler et al., 2006; Ernst and Lovich, 2009). As the ecological status of M. terrapin is unclear and human pressures on their habitats will probably worsen, the need for ecological and demographic data from poorly known M. terrapi n populations is essential to compare with other populations. Population level and behavioral information are necessary to inform management decisions about M. terrapin populations. Studies on M . terrapin in Florida have been generally limited to the penin sula . The n orthern Gulf Coast populations in the Florida panhandle have gone virtually unstudied , with the only information coming from museum specimens and anecdotal information (Butler et al., 2006). The only published studies on the ornate diamondback t errapin ( M alaclemys terrapin macrospilota ) in the Gulf Coast of Florida are distributional surveys of the Aquatic Preserve in Citrus County (Boykin, 2004), and nest surveys in the n orthwest ern peninsula (Butler and Heinrich, 2013). A p opulation and nest site selection study is currently being conducted in Lee County on the southwestern coast (C. Lechowicz, pers. comm.). Hart and McIvor (2008) implemented a population and demographic assessment on the mangrove

PAGE 36

36 diamondback terrapin ( M alac lemys t errapin rhizophorarum ) in Everglades National Park . On the mid Atlantic Coast of Florida, Seigel (1984, 1993) reported long term population declines from a marked population of the eastern Florida diamondback terrapins ( Malaclemys terrapin tequesta ) . On the Atlantic Coast in northeast ern Florida, Butler (2000, 2002) assessed the ecological status, distribution, home range, and seasonal behavior of the Carolina diamondback terrapin ( Malaclemys terrapin centrata ) . Of the five recognized subspecies of M. terrapin found in Florida, M. t. macrospilota has the largest distribution, but is also the least studied. Until my study, there had been no research completed on M. t. macrospilota west of the St. Marks River in Florida. The current ecological status of M. t. macrospilota is unknown in the Florida Panhandle and concern heightened after the Deepwater Horizon oil discharge beginning in April 2010. This event alarmed the Florida Fish and Wildlife Conservation Commission ( FWC ), wh ich then began to organize groups of volunteers to monitor M. t. macrospilota on a small Gulf Coast barrier island ( Lanark Reef ) in order to acquire baseline data to later assess impacts of the oil on that population . Fortunately, oil did not reached this area in the northeastern G ulf of Mexico , but as a result of multiple trips to mark and recapture turtles , more than 300 individuals were found by the end of summer in 2012. These preliminary data allowed me to develop the population asse ssment study described in this c hapter. The p urpose here is to report the results on population demographics, structure, and size of M. t. macrospilota on Lanark Reef using survey techniques described in Chapter 2. Methods Sampling Design See Chapter 2 under sampling design for methods used to surve y M alaclemys t errapin macrospilota .

PAGE 37

37 Collection of Diamondback Terrapin Population Assessment Data for Pilot and Main Studies Using 50 cm Haglof® aluminum calipers , I recorded (to the nearest 1 mm) the following measurements on each turtle captured: mid li ne carapace length (CL), carapace width (CW), shell height (HT), and plastron length (PL). Turtles were weighed using 1000 g and 2500 g Pesola® spring scales ( ± 0.3%) . Sex, date, time, GPS location (Garmin GPS 76Cx) were also recorded at each capture. Adult females were palpated to determine if they were gravid. Maturity was determined by the smallest gravid individual in females and by their enlarged tails and cloacal opening situated p ast the carapace in males (Seigel , 1984; Lovich and Gibbons, 1990). For Lanark Reef are here defined as having a M. t. macrospilota (PL > 49 mm) was individually marked by drilling hole s in the marginal scutes (Cagle, 1939) with a power dr ill and implanting a PIT (Passive Integrated Transponder, Biomark) tag underneath the bridge where the carapace and plastron connect. Statistical Analyses L ocations of each turtle were mapped using ArcGIS 10 (ESRI 2011) . All analyses except for the analysis on sex ratio were performed only on those data from the main study. All data were analyzed for normality using the Shapiro Wilk normality test. I used CPUE (individuals Rcapture used a Jolly Seber open population model assuming heterogeneous probability of mark recapture to estimate population size. I used a chi square goodness of fit test to determine if adult sex ratio s were significantly different (both pilot and main study). Population density was determined by dividing the number of individuals captured by Section area. To examine differences in mean CL, CW, HW, PL, and mass between sexes, a sexual dimorphism index (SDI) and two sample t -

PAGE 38

38 test was used for each parameter. The SDI is calculated by dividing the larger sex (females) by the smaller sex (males) and subtracting by 1 to get the degree of sexual size dimorphism (SSD) exhibited by a population (Lovich and Gibbons, 1992) . I conducted all analyses using Program R (R Devel opment Core Team, 2015). Results Pilot Study I captured 85 novel individuals (147 total captures) throughout the intertidal zone of Section 1 and Section 2 during the pilot study (September November 2012). All size classes (PL range = 46 mm 158 mm, mea n = 91.1 mm), exce pt hatchlings, were represented . Lanark Reef population structure consisted of 42.2% (n = 62) juveniles, 39.5% (n = 58) adult males, and 18.4% (n = 27) adult females. Sex ratio was male biased and differed significantly from 1:1 ( X 2 = 11.306, p = 0.0008 ). All individuals were captured in the intertidal zone under some form of vegetation (e.g., smooth cordgrass and/or tidal wrack) . Main Study I captured 381 novel individuals (494 total captures ; not including Pilot Study ) throughout the intertidal zone of the three sections of Lanark Reef (Table 3 1), which resulted in a CPUE of 0.22 or 1 turtle ev ery 4.5 minutes. The population e stimate for Lanark Reef was 929.1 ± 142.1 individuals . Population density in Section 1 was 22.2 individua ls/hectare (ha) , Section 2 had 145.6 individuals/h a , and Section 3 had 140.3 individuals/h a . Twelve turtles were recaptured two times, seven turtles three times, and two turtles four times. All size classes were represented on Lanark Reef (PL range = 27 m m 167 mm) (Table 3 2). Turtles were distributed evenly throughout the intertidal zone regardless of sex and size (Figures 3 1, 3 2, 3 3). The sample actually captured consisted of 125 (32.8 %) juveniles (includes sub adult s , and yearlings), 200 (52.5 %) adult males, and 56 (14.7 %) adult females.

PAGE 39

39 All morphometric measurements and mass differed significantly among males and females (Table 3 3). The overall adult male : female sex ratio was 3.6 :1 and differed significantly from 1:1 ( 2 = 81.000 , p < 0.0001 ). P opulation sex ratio remained male biased during the study period , and population structure shifted from a majority of females to a majority of juveniles during the la s t two months of sampling (Figure 3 4). I captured a total of 25 gravid females (45 palp ated females) from 9 April through 25 July 2013 (Figure 3 5). Discussion Population Size and Structure Based on total captures (n = 494) during four months of sampling and increasing number of new capture s (n = 381), the study population is very dense. L anark Reef growing or there is a lot of emigration and immigration occurring with many new individuals found at each subsequent sampling event. This population is one of the largest (929.1 ± 142.1 individual s ) reported throughout the range of M alaclemys terrapin . Capture techniques, sampling regime, study site characteristics and size, along with anthropogenic processes such as crab trapping, habitat loss, and harvest may explain the large differences in population estimates of M . terrapin in the literature (Table 3 6) . The area around Lanark Reef has the lowest crab trapping effort within its region, likely due to the low abundance of Callinectes sapidus (R. Gandy, pers. comm.) , and evidence of Procyon lotor have never been detect ed on this island (E. Suarez, pers. obs.; B. Turner, pers. comm.). These observations suggest mortality by predation and bycatch in crab traps is low or absent, which may explain the large size of this population. The model used in my study may not yield t he most appropriate population estimate because it is only considering three months of sampling in one year. This estimate is mostly providing the population size for this site during the peak months when M. t. macrospilota inhabits Lanark Reef . Turtles em igrate from Lanark

PAGE 40

40 Reef during the winter and immigrate back to the island during the spring. Also, based on preliminary telemetry data , some turtles m ove back and forth from nearby marshes during the fall. During this time away from Lanark Reef , this popu lation could be mixing with other populations, which may explain why new unmarked turtles continue to be found in such a small area. Whether my study population is one large population or consist s of several small populations aggregating at this one site i s uncertain . A robust population model that uses a combination of closed and open models and accounts for abundances during closed sampling periods and estimates survival rates between sampling periods should be used once more movement data can be collecte d. P opulation size structure shows a high percentage of juveniles and adult males, which is in consistent with other studies (S eigel, 1984; Butler, 2000; Gibbo n s et al., 2001; Hart, 1999; Boykin, 2004; Estep, 2005; Sheridan, 2010; Haskett, 2011; Mealey et al. 2014). Observed d ifferences could be due to the sampling technique, which allowed for all size classes to be captured. All size classes inhabiting Lanark Reef use d similar microhabitats (Figures 3 1, 3 2, 3 3 ) . Other studies using net based capture techniques could be using mesh sizes too large to capture smaller turtles, thus biasing results (Hart, 1999). Also, sampling may be occurring in habitats where juveniles may be absent (e.g., open water habitats) (Gib bons et al. 2001). Butler (2002) found juveniles, however, his search methods were biased because he was mostly searching in and around adult female nesting sites where juvenile s may overwinter in nest or vegetation (Burger, 1976; Roosenburg, 1991; Butler et al., 2004). Other possible reasons for the low number of juveniles captured in th ose studies could be due to low recruitment as threats such as crab trapping and predation can have a greater negative impact on smaller size classes. Lanark

PAGE 41

41 Reef offers su itable habitat to all size classes of M . t. macrospilota while other locations may have individuals of difference size classes stratifying habitat use based on the habitat available. Sex Ratio Sex ratios reported for M . terrapin populations are highly var iable . My observed sex ratio of 3. 6 :1 male:female indicates a male biased ratio, which has been observed in other areas in Florida (Butler, 2002), South Carolina (Bishop, 1983; Lovich and Gibbons, 1990; Estep, 2005), Louisiana (Cagle, 1952), and Delaware ( Hurd et al., 1979). Studies in other parts of Florida have reported female biased sex ratios (S eigel, 1984; Butler, 1998, 2000 ; Boykin, 1999, 2004). In New Jersey (Sheridan 2010), Delaware (Hurd et al., 1979), and Maryland (Roosenburg et al., 1997) sex rat ios are female biased as well . Hart and McIvor (2008) reported a sex ratio of 1:1 in south ern Florida, but included only adults in their analysis. These differences in reported sex ratios could be attributed to the type of data that were analy zed (e.g., al l males and females or only sexually mature males and females). Other reasons for this variation in sex ratios c o uld be due to differences in sampling techniques (e.g., trapping vs. hand capture), area/time sampled (e.g., nesting beach during nesting seaso n), imprecise results due to low sample sizes, or differences in survivorship among sexes and size classes (Gibbons et al., 2001). These reported differences in sex ratio can also be a result of crab trapping pressure or heightened predation leading to fem ale biased population, or increased road mortality or harvest pressure on large adults leading to male biased populations. Since M. terrapin exhibit temperature sex determination (TSD) , some of these populations may naturally be male or female biased based on where females may be laying their eggs. Although TSD likely varies by latitude throughout the range of M. terrapin ; generally, nest at lower incubation temperatures produce more males and nest at high incubation temperature produce more females (Ewert and Nelson, 1991; Roosenburg and Kelly, 1996).

PAGE 42

42 natural population than what has been reported elsewhere because of the lack of predators (i.e., raccoons), little or no crab trapping pressur e, and no bias sampling toward certain size classes. Body Size Throughout the range of M . terrapin , overall body size and size at maturity varies , even at similar lati tudes. A dult s inhabiting Lanark Reef have a smaller mean PL than in most other studied populations (Table 3 4). The smallest mature female (based on presence of eggs) I captured had a PL of 111 mm and the smallest mature male (based on enlarged tail) had a PL of 64 mm , suggesting that Lanark Reef ller size than what has been reporte d in other studies (Table 3 5). Hildebrand (1932) observed females as small as 120 mm PL oviposit in captivity under accelerated growth conditions. In my study, four gravid females (120 130 mm PL ) and six gravid females (< 132 mm PL ) were captured, which are the smallest size at maturity reported for M. terrapin (Table 3 5). Based on the observed high availability of prey on Lanark Reef , this population may be naturally undergoing accelerated growth compared to other popu lations. P ossible reason s for the low mean PL reported for M . t . macrospilota in my study, could be due to biased sampling techniques , local population genetics, or intraspecific competition . My imum PL for females (max female PL = 1 67 mm) is much smaller than those reported elsewhere in Florida (Table 3 7). Mean male PL is also substantially smaller on Lanark Reef compared to other studies throughout range of M. terrapin (Table 3 4). Other studies could be using sampling techniques t hat favor certain size classes such as looking for nesting sites or crab trapping (Butler, 2000; 2002). The high concentration of M. t. macrospilota in my study could be influencing the amount of food resources available to each individual, potentially pla cing a limit on the maximum size that could be reached by this population.

PAGE 43

43 Intraspecific competition may be a density dependent factor influencing size of adults. Population genetics and sampling bias can also be influencing these observed body sizes. Lana rk population could be reaching or at carrying capacity and the M. t. macrospilota that inhabit this island could be compensating for body size in order to support a large population. . Studies on other turtle species have reported a positive correl ation between body size and latitude (Iverson and Smith, 1993; Edmonds and Brooks, 1996; Iverson et al., 1997), however the reported mean PL for M. terrapin in the literature do not fit this pattern (Horn, 2012) . Differences in habitat type, habitat produc tivity (Johnston et al., 2012) or sampling techniques (Gibbon et al., 2001) used in these studies could be influencing the observed body size. When M . t. macrospilota inhabit Lanark Reef (April November), the population is large and dense, and appears to be healthy and thriving. The size at maturity observed for both sexes is the smallest size reported . The reported male biased sex ratio was not surprising, but the hi gh percentage of juveniles captured was unexpected. On Lanark Reef, the absence of P. lotor , little or no commercial crab trapping, and the high food density are all factors contributing to this large population.

PAGE 44

44 Table 3 1. Total captures, person minutes, and CPUE at different sampling sections of Lanark Reef. Section Total Captures Person Minutes CPUE Section 1 205 1140 0.18 (1 terrapin every 5.6 minutes) Section 2 181 600 0.30 (1 terrapin every 3.3 minutes) Section 3 108 510 0.21 (1 terrapin every 4.7 minutes) Table 3 2. Morphometric measurements and mass from Malaclemys terrapin macrospilota captured in the main study. Data are presented as mean s (minimum maximum). Size Class Carapace Length (mm) Plastron Length (mm) Carapace Width (mm) Shell Height (mm) Mass (g) Male (n=200 ) 103.1 (71 136) 88.4 (64 123) 80.6 (40 106) 41.6 (33 55) 200.5 (75 520) Female (n=56 ) 158.5 (120 186) 142.9 (111 167) 122.8 (95 142) 66.1 (51 79) 766.5 (320 1300) Juvenile (n=125 ) 88.1 (38 117 ) 78.62 (27 107 ) 69.88 (31 92 ) 39.0 (16 50 ) 150.2 (15 44 0) Total Sample (n=381 ) 106.3 (38 186) 93.2 (27 167) 83.3 (31 142) 44.3 (16 79) 267.2 (15 1300 ) Table 3 3. Morp hometric measurements and mass for each adult Malaclemys terrapin macrospilota in the main study . Data are presented as mean (minimum maximum). Raw data for males and females were compared using two sample t test and sexual dimorphism index (SDI) . Morphometric Measurement Male (n=197) Female (n=156) P value SDI Carapace Length (mm) 103.1 (71 136) 158.5 (120 186) < 0.001 0.531 Plastron Length (mm) 88.4 (64 123) 142.9 (111 167) < 0.001 0.617 Carapace Width (mm) 80.6 (40 106) 122.8 (95 142) < 0.001 0.524 Shell Height (mm) 41.6 (33 55) 66.1 (51 79) < 0.001 0.589 Mass (g) 200.5 (75 520) 766.5 (320 1300) < 0.001 2.823

PAGE 45

45 Table 3 4. Mean PL lengths for adult male and female M alaclemys terrapin throughout their range. * Calculated from appendix. Study Location Mean Male PL (mm) Mean Female PL (mm) This Study Florida (West Coast) 88.5 1 43.1 Seigel, 1984 Florida (East Coast) 104 154 Mealey et al. 2014 Florida (South Coast) 101.4 149.2 Boykin, 1999 Florida (Gulf Coast) 110 165 Boykin, 2004 Florida (Gulf Coast) 118 175 Butler, 2002 Florida (East Coast) 102 162 Hart and McIvor, 2008 Florida (South Coast) 105 160 Montevecchi and Burger, 1975 New Jersey n/a 154 Gibbons et al., 2001 South Carolina 103 144 Levesque, 2000 South Carolina 101 141 Spivey, 1998 North Carolina 96.5* 139.6* Haskett, 2011 Texas 111.1 164.2 Table 3 5. Female PL length at sexual maturity for M alaclemys terrapin throughout their range. Study Location Female PL at Sexual Maturity (mm) This Study Florida (West Coast) 111 Seigel, 1984 Florida (East Coast) 135 Butler, 2000 Florida (East Coast) 144 Montevecchi and Burger, 1975 New Jersey 132 Gibbons et al., 2001 South Carolina 138 Roosenburg , 1991 Maryland 175 Table 3 6. Population estimates reported for Malaclemys terrapin throughout their range. Estimates are either reported alone, with standard error, or with confidence intervals. Study Location Site Size Population Estimate This Study Florida (West Coast) 8.57 ha 929 ± 127 Seigel, 1984 Florida (East Coast) Site 1 Site 2 0.92 ha 0.66 ha 404 ± (183 792) 213 ± (59 627) Boykin, 2004 Florida (Gulf Coast) Site 1 Site2 n/a n/a 255 ± (143 367) 141 ± (81 201) Butler, 2002 Florida (East Coast) n/a 3,147 Hart, 2005 North Carolina 336 ha 1,545 Hurd et al., 1979 Delaware 0.9 km 1,655 Roosnburg et al., 1997 Maryland 10 km 2,778 3,730 Morreale, 1992 New York Site 1 Site 2 12 ha 200ha 341 344

PAGE 46

46 Table 3 7. Female maximum PL length for Malaclemys terrapin in Florida. Study Location Max Female PL (mm) This Study Florida (West Coast) 167 Seigel, 1984 Florida (East Coast) 177 Mealey et al. 2014 Florida (South Coast) 173 Butler, 2002 Florida (East Coast) 186 Hart and McIvor, 2008 Florida (South Coast) 185

PAGE 47

47 Figure 3 1. Captured Malaclemys terrapin macrospilota by size class in Section 3 at Lanark Reef . Figure 3 2. Captured Malaclemys terrapin macrospilota by size class in Section 2 at Lanark Reef .

PAGE 48

48 Figure 3 3. Captured Malaclemys terrapin macrospilota by size class in Section 1 at Lanark Reef . Figure 3 4 . Histogram showing Malaclemys terrapin macrospilota captured each month by size class and sex at Lanark Reef. 0 10 20 30 40 50 60 70 80 90 100 April May June July # of individuals Month Male Female Juvenile

PAGE 49

49 Figure 3 5. Number of gravid Malaclemys terrapin macrospilota captured each month at Lanark Reef.

PAGE 50

50 CHAPTER 4 INVESTIGATION OF THE INTERACTIONS BETWEEN THE DIAMONDBACK TERRAPIN ( MALACLEMYS TERRAPIN MACROSPILOTA ) AND BLUE CRAB TRAP S Introduction The ornate diamondback terrapin ( Malaclemys terrapin macrospilota ) inhabits barrier Monroe County, northward to Walton County in the Florida panhandle, where coastal development and crab fisheries are widespread (But ler et al., 2006). Throughout its range, Malaclemys terrapin also share habitat with the blue crab ( Callinectes sapidus ), and mortality from drowning in human constructed artificial crab traps has been documented for at least 72 years (Davis, 1942; Roosenb urg, 2004). Many researchers agree that crab traps are the primary threat to M. terrapin populations range wide (Butler et al., 2006). Based on the amount of turtles captured in crab traps, Roosenburg et al. (1997) estimated that from 15% to 78% of a local population might be captured in a single year. This large range is influenced by body size, season, and behavior (e.g., habitat use) (Roosenburg, 2004; Hart and Crowder, 2011) . Some s tates (New Jersey , Delaware, and Maryland) have adopted regulations requ iring implementation of turtle excluder devices (TEDs) on blue crab traps to help reduce M. terrapin bycatch; however, it is important to remember that if TEDs are reducing the number and size of trapped C. sapidus , the livelihood of crab fisherman could b e affected. It is also important to note that if TEDs were to be mandated throughout the crab trap fishery, it would not eliminate M. terrapin mortality since TEDS do not prevent smaller individuals from entering traps (Roosenburg and Green 2000). Roosenbu rg (2004) compiled a review, which included seven studies that used TEDs, showing how variable C. sapidus catch can be among studies. The push for the mandatory implementation of TEDs on commercial crab traps is high; however, these regulations may have

PAGE 51

51 no direct effect on reduction of M. terrapin mortality associated with crab traps used in Florida and could impose unnecessary economic burdens on the fishery there. Based on observations of individual turtle interactions with crab traps in the field (E. S uarez, unpublished data; C. Lechowicz, pers. comm ) , I design ed an experiment to test the duration and escape rates . The first objective of my study was to test whether M. terrapin (excluding adult females) can escape once inside a crab trap. I specifically wanted to test if individuals smaller than the recommended TED size for Florida (45 mm x 120 mm; Butler and Heinrich, 2007) can escape once inside a crab trap. My second objective was to determine which internal component (center baffle or b ait cup ; Figure 4 1 ) of a commercial crab trap limits the likelihood of M. terrapin escapement. Results from this experiment might lead to recommended changes to the existing trap design to facilitate M . terrapin escape ment without negatively effecting cra b trapping . Specifically I asked , do the internal component s of a commercial crab trap limit M. terrapin escapemen t ? My research hypothesis is that internal crab trap design is the cause for limiting M. t errapin escapement. Furthermore, I predicted that le ss crowded traps (e.g., baffle removed or bait cup removed) would allow M. t errapin to have a higher likelihood of escape ment. My statistical null hypothesis states that the l ikelihood of escape is equal among all trap types. Statistical hypothesis 1 state s that the l ikelihood of escape in the control crab trap is less than the likelihood of esc ape in experimental crab trap 2, which is less than the likelihood of escape in experimental crab trap 1 . Statistical hypothesis 2 states that the time it takes to e scape out of the control crab trap is greater than the time is takes to escape out of the experimental crab trap 2 , which is greater than the time it takes to escape out of experimental crab trap 1 . Methods Malaclemys terrapin macrospilota (n = 95; PL range = 56 106 mm; HT = 28 52 mm) were captured at Lanark Reef in the Florida panhandle and transported in 95 quart Rubbermaid

PAGE 52

52 containers to the Whitney L ab oratory, St. Augustine, F lorida . Two large (6.1 m diameter, 2.1 m deep) aquaculture tanks were used to house turtles . W ater used in these tanks came from a nearby estuary to insure that the chemistry is similar to what M. terrapin encounter in their natural habitat . Turtles housed for multiple days were kept in one large tank. They were fed dried and fre sh shrimp every other day. Structure and basking areas were included in the tank to help mimic natural habitat and provide cover, ba sking opportunities, and resting areas out of the water. No turtles were kept in tanks for more than three days , and turtles were released at their capture sites . In order to reduce undue stress on th e wild population and still reach a sample size large enough for robust statistical analysis , 60 turtles were used in the main study in addition to the 42 individuals used in my pi lot study . I modified commercial blue crab traps to fit three different designs. I used a standard unmodified commercial crab trap as my control (control trap; Figure 4 2 ). I test ed a commercial crab trap with the bait cup, but without the center baffle that separates the upper and lower parlors (experimental trap 1; Figure 4 3 ). I also test ed a commercial crab trap with the bait cup removed, but the center baffle still in place (ex perimental trap 2; Figure 4 4 ). Various measurements (CL, PL, CW, HT), sex, and mass were recorded before M. t. macrospilota were used in trials. All turtles followed the same protocol for each trial in the pilot and main study. The tank was set up with t he three crab traps inside , and barriers were set up to avoid any behavioral bias of subjects looking at each ot her while in the trap . Tanks were filled to 1.2 m deep and water temperature ranged from 31.1 32.2 ° C. Before each test, one person would get in side the tank and set up the barriers and crab traps. Then, a randomly selected individual was selected and the timer would begin once the turtle was put inside the crab trap. During the time the turtle was

PAGE 53

53 inside the crab trap, the volunteer inside the ta nk would hide behind the barriers to avoid disturbing the subject while I made behavioral observations from outside the tank. If an individual escaped, it was quickly removed with a dip net or by hand. After 10 minutes, if the turtle did not escape, the vo lunteer inside the tank removed it. Pilot Stud y Two pilot studies were conducted to determine if turtles could learn how to escape and to develop a more appropriate experimental design for my main study. For the first pilot study, turtles were randomly se lected to go through one of the two trap types (control, experimental trap 2) for the first trial, and then put through the different trap type for the second trial. For the second pilot study, there was no randomized block design implemented to determine which randomization scheme (i.e. , random block or no block) would be most appropriate for the main study. In the second pilot study, individuals were randomly selected and put through any trap type regardless of the trap type in which they were tested in t he previous trial. Some individuals went through the same trap type multiple times. I baited traps during the first pilot study, but decided against it for the rest of the experiment as individuals showed no interest in bait after 52 trial runs. Main Stud y Based on results of the pilot studies, I used a randomized block design in my main experiment . Subjects were assigned to blocks by trap type to ensure that each of the treatme nts (trap types) were given to each of the subjects . Although my pilot data sho wed no learning behavior or signs of injury after consecutive trials, I used a 24 hour rest period between trials to minimize stress on individual turtles. I started a timer once the individual was placed in a crab trap and stopped it when the turtle escaped . In cases when individuals did not escape, I manually

PAGE 54

54 remove d them after 10 mi nutes (or if the individual showed signs of lethargy ) based on what was reported in Baker et al. (2013) in order to avoid fatalities or injury. Statistical Analysis for Pilot Study All data were analyzed for normality using the Shapiro Wilk normality test. I ran multiple analyses after each of the two pilot studies to determine if trial s were independent . I used a chi square test of independence was used to determine if i ndividuals were learning how to escape from crab traps during multiple rounds . A Wilcoxon signed rank test was used to determine if individuals that got out in consecutive trials were getting out faster each subsequent trial. I also looked at a simple grap h plotting time (x axis) and trial (y axis) to see if time of escape decreased each subsequen t trial . I conducted all analyses using SPSS and R statistical software. Statistical Analysis for Main Study All data was analyzed for normality using the Shapiro Wilk normality test. A chi square test was used to determine if there was a trap effect on whether an individual escaped. If there was a significant difference in chi square values, I used a post hoc method described in Beasley and Shumacker (1995) that al lowed me to make pairwise comparisons among the different trap types. A Kruskal Wallis test was used to determine if time was significant ly different among trap types. If the Kruskal Wallis tests indicated a significant difference, multiple pairwise compar isons were used to detect differences among groups. I also compared mean mass and the va rious morphometric measurements to see if there was a difference in individual size and the likelihood of escape. I conducted all analyses using Program R statistical s oft ware (R Development Core Team, 2015).

PAGE 55

55 Results Pilot Studies The first pilot data set consisted of 26 individuals (n=52 trials). With both trials combined, only 50% of the escape attempts were successful (Table 4 1). Trap type had no significant effect on trial ( 2 = 0.38, p = 0.58). Time of escape was not significantly different at each consecutive successful escape attempt (V = 27, p = 0.65). The second pilot data set consisted of 16 individuals (n=48 trials ). With all trials combined, only 35% of the escape attempts were successful (Table 4 2). Trap type had no significant effect on trial ( 2 = 1.31, p = 0.52). Time of escape was not significantly different at each consecutive successful escape attempt (V = 19.5, p = 0.89). Main Study For the main study, my combined dataset consist ed of 53 individuals (n= 159 trials). Thirty percent escaped from the control trap, 23 % escaped from experimental trap 1, and 45 % escaped form experimental trap 2 (Table 4 3) . Trap type had a significant effect on whethe r an individual escaped ( 2 = 6.4 p = 0.04) and post hoc test show ed that experimental trap 2 was significantly affecting whether an individual escaped or not ( 2 = 5.71 p = 0.02) . Time of escape was significantly affected by trap type (H = 6.4, p = 0.04) and post hoc test show ed that experimental trap 2 had significantly faster escape time than the c ontrol and experimental trap 1 (Figure 4 5 ). Within my sample, t here was no significant difference in mean mass and all other morphometric measurements on whether an indi vidual escaped or not (Table 4 4 ). In traps that had the center baffle intact (control and experimental trap 2), 85.1% of the individuals went to the top level and only 2 individuals escaped through a cull ring.

PAGE 56

56 Discussion Overall, M alaclemys terr apin did not have great success escaping from crab traps. Modifications to the internal structure of commercial crab traps or the implementation of some type of escape panel may help turtles escape. My pilot study results suggest tha t no learning occurred during the experiment and t hat each trial was independent . T here was more of an individual effec t rather than a learning effect, suggesting that some individuals are naturally better at escaping than others. The turtles that did escape most of the time may have learned how to escape crab traps in their natural habitat; however, it that seems unlikely for the turtles inhabiting Lanark Reef, unless they are encountering traps outside of their time on the island since little or no commercial crab trapping occu rs around the island (Rya n Gandy, unpublished data). Based on the results from the main study, I reject ed all three hypotheses. I reject ed the null hypotheses because the likelihood of escape was not equal across all trap types. I reject ed hypothesis 1 and 2 because there was a difference in the likelihood of escape and faster escape time among trap types; there was no greater likelihood of escape and faster escape time in experimental trap 1 compared to the other trap types. However, I did see a difference in the likelihood of escape and faster escape time in experimental trap 2 versus experimental trap 1 and the control, which was not what I predicted. Experimental trap 1 had the fewest escapes largely due to the absence of the center baffle where only 22 .6% of the escape , which can be explained by turtle behavior s observed during this experiment. Once an individual was placed in a trap, its first reaction was to go up. Since experimental trap 1 had the baffle remove d, most of the individuals spen t their time trying to escape out of the top of the trap. When individuals were placed in the control and expe rimental trap 2, they also spent their time trying to go up to the second level of the trap , but

PAGE 57

57 the center baffle would block them forcing them s pend more time in the bottom level , which sometimes helped facilitate escape . This was the first known research testing the likelihood of M . terrapin escaping once inside a crab trap , thus comparisons to other studies cannot be made. These results should be taken with caution since this experiment was done in a lab setting where lots of variables are controlled for and may not truly explain what could happen in M. terrapin natural habitat . However, I assume that individuals will likely spend more time in a trap in the wild before they actually determine that they are trapped. In a recent modified crab trap study (Ryan Gandy, unpublished data) , I observed multiple turtles eating bait while inside the crab trap. If M. terrapin are attracted to bait and begin to feed once inside the trap, the time before the individual runs out of oxygen could substantially decrease. Many other factors that can be tested in the lab such as water turbidity, water temperature, number of organisms (i.e. , blue crab, catfish, or other aquatic organisms) and number of M . terrapin inside a trap may influence the probability of turtles to escape a crab trap . Furthermore , because of the differences in morphology throughout M. terrapin range, a TED size that wor ks in some areas may not be applicable to other areas. If TEDs are to be implemented, the TED size should be chosen so that it excludes the maximum number of turtles while allowing enough blue crabs to enter for crabbers to have sufficient profit. From thi s viewpoint, the mandated TED should be the smallest TED that the crabbers will allow. However, biologists may be wasting time, money, and resources focusing on research regarding crab traps with TEDs and M. terrapin interactions, especially if crab fisher men are not persuaded to put TEDs on their traps. Even in areas where TEDs have been mandated, fishermen are not always complying with these regulations (Radzio and Roosenburg, 2005) .

PAGE 58

58 If a standard TED of 45 mm x 120 mm were mandated by FWC, 56 out of 87 (64%) individuals smaller than the TED size would have drowned if they were caught in a crab trap. B iologist s should consider shifting their efforts on the development of an escape mechanism rather than the mandatory implementation of TEDs. Since crab fish erman and manufacturers are unlikely to remove any of the internal components (R. Gandy, pers. comm.); I suggest four modifications to crab traps that still keep the internal components intact. The first is a slight modification to the internal structure o f the crab trap. Instead of leaving the bait cup in its vertical position, making it smaller and laying it flush with the plane of the center baffle (Figure 4 6) will still keep the bait at the bottom level, but would leave an open space for M. terrapin to find the entrance while not continuously crashing into the bait cup in its original position. This modification would be similar to experimental trap 2, which had the most successful escape attempts. The second modification would be to change the size of the cull ring. I noticed that small ( 61 mm PL) turtles that find the cull ring are able to escape. The cull ring may be a component of the trap that can be altered as some crab fisherman in Florida have said they are capturing too many small Callinectes sapidus , which should be escaping out the cull rings (R. Gandy, pers. comm.). Nevertheless, these modifications would only potentially decrease the number of drowning incidents, not eliminate them completely. Based on turtle behaviors I observed, a different modification could be a trap that has an escape panel at the top that could allow M. terrapin to escape. The main problem with this modification is that the panel needs to prevent C . sapidus from escaping. Another modification could be to instal l another opening (i.e., similar entrance the crab traps already have at the bottom level) in the top level of the crab trap. Since M. terrapin spends most of its time trying to get to the top level or at the top level of the trap, including another entran ce at the top may help

PAGE 59

59 facilitate turtle escape, but not increase C . sapidus escapement rates. Regardless of the modification, it has to be something that crab trap manufacturers can install by making minimal modifications to the standard commercial crab t rap used. My experiment shows how crab traps can have a serious negative impact on M . terrapin . The most successful trap was experimental trap 2, which had the center baffle intact and free space in both the bottom and the top sections of the trap (Figure 4 4). Experimental trap 2 was the most successful design because it allowed for M. terrapin to swim more freely in the bottom level making it easier to find the exit. Since many states throughout M. terrapin range have decided not to adopt laws mandating t he implementation of TEDs, a shift to the development of an escape mechanism while minimizing modifications to crab trap design is crucial.

PAGE 60

60 Table 4 1. Number of Malaclemys terrapin macrospilota that escaped and did not escape from crab traps during the fi rst pilot study. Trap Type Trial 1 Escaped Trial 1 Did not Escape Trial 2 Escaped Trial 2 Did Not Escape Control 6 7 5 8 Ex2 6 7 9 4 Total 12 14 14 12 Table 4 2. Number of Malaclemys terrapin macrospilota that escaped and did not escape from crab traps during the second pilot study. Trap Type Trial 1 Escaped Trial 1 Did Not Escape Trial 2 Escaped Trial 2 Did Not Escape Trial 3 Escaped Trial 3 Did Not Escape Control 3 3 1 4 2 3 Ex1 1 4 4 2 1 5 Ex2 1 4 2 3 1 4 Total 5 11 7 9 4 12 Table 4 3. Number of Malaclemys terrapin macrospilota that escaped and did not escape from crab traps during the main study. Trap Type Trial 1 Escaped Trial 1 Did Not Escape Trial 2 Escaped Trial 2 Did Not Escape Trial 3 Escaped Trial 3 Did Not Escape Control 3 15 7 10 6 12 Ex1 5 13 6 12 1 16 Ex2 10 7 8 10 6 12 Total 18 35 21 32 13 40 Table 4 4. Morphometric measurements and mass from Malaclemys terrapin macrospilota that escaped and did not escape from crab traps during the main study. Data presented as mean (minimum maximum). Escaped? Carapace Length (mm) Plastron Length (mm) Carapace Width (mm) Shell Height (mm) Mass (g) Yes (n=52) 99.1 (67 116 ) 85.3 (58 99 ) 76.8 (50 90 ) 40.1 (29 45 ) 177.6 (50 240 ) No (n=107 ) 99.1 (67 123 ) 85.8 (58 103 ) 76.4 (50 95 ) 40.4 (29 4 9) 176.2 (50 300 )

PAGE 61

61 Figure 4 1 . Standard commercial blue crab trap with labeled components. Figure 4 2 . Control trap (standard commercial blue crab trap) used in experiment of Malaclemys terrapin macrospilota . Note the center baffle and bait cup are intact.

PAGE 62

62 Figure 4 3 . Experimental trap 1 used in experiment of Malaclemys terrapin macrospilota . Note the center baffle has been removed and the bait cup is still intact. Figure 4 4 . Experimental trap 2 used in experiment of Malaclemys terrapin macrospilota . Note the center baffle is still intact and the bait cup has been removed.

PAGE 63

63 Figure 4 5 . Multiple pairwise comparison of mean escape time of Malaclemys terrapin macrospilota by trap type with standard error bars. Letters indicate sig nificance. Figure 4 6 . Modified crab trap with bait cup position ed flush with the plane of the center baffle .

PAGE 64

64 CHAPTER 5 CONSERVATION IMPLICATIONS AND FUTURE DIRECTIONS My results show that there is a large population of ornate diamondback terrapins ( Malaclemys terrapin macrospilota ) at Lanark Reef in the Gulf Coast along the Florida panhandle, and long term population monitoring should continue to assure that anthropogenic threats (e.g., illegal collection for the pet trade and commercial crab trappin g) do not jeopardize this population. Long term monitoring would also be helpful in determining survival rates and long term population trends. Lanark Reef is not only important to the protected species of birds that nest there, but also for M. t. macrospi lota . P rotec ting Lanark Reef should be a high priority before threats can affect the species that inhabit it. The survey method described herein should be applied in other areas throughout the panhandle and northwest ern peninsula . Hotspots with a ggregation s like the one observed in this study have been observed elsewhere in Florida (Seigel, 1980, 1984). I recommend the implementation of species distribution models using habitat characteristics from Lanark Reef to potentially find other hotpots. Once these h otspots are found, the search methodology presented in Chapter 2 should be used to rapidly assess the population there. There are many gaps in research on M alaclemys terrapin throughout their geographic range with one of the largest gaps found in the Gulf of Mexico section of its range. Although it M. t. macrospilota population, other threats such as sea level rise and illegal collection for the pet trade are major thr eats this population will face in its future . Titus et al. (2009) reported that climate change is a major threat to M. terrapin . Sea level rise can drastically alter the habitat of Lanark Reef potentially making it unsuitable for M. t. macrospilota . The lo w elevation of Lanark Reef makes it prone to over wash that can destroy critical nesting and foraging habitat. The M. t.

PAGE 65

65 macrospilota from Lanark Reef have been collected for the pet trade in the past (B. Turner, pers. comm.) and collecting pressure will likely increase in the near future (J. Gray, pers. comm). Possession limits in Florida are good for keeping collectors from taking an excess amount of M. terrapin from the wild, but c hanging possession limits in Florida from two turtles per person to one turtle per person may deter collectors from capturing turtles for breeding proposes putting even less pressure on wild populations, but still allowing for collectors to keep one M. terrapin . Crab trapping is unlikely to affect the population of M. t. macrospilota inhabiting Lanark Reef, but is a large problem to populations elsewhere outside of Florida (Bishop, 1983; Wood and Herlands, 19 96; Roosenburg et al., 1997; Dorcas et al. 2007). According to the Florida Fish and Wildlife Fisheries Independent Monitoring program, turtles are not being captured at a high rate in Florida where crab trapping is occurring. From 2002 2011 in waters throu ghout Florida only 675 turtles were captured out of 95,871 attempts (CPUE = 0.0070) while using one of six capture techniques ( 21.3 m center bag seine , 183 m center bag seine , 183 m terminal bag seine , 61 m center bag seine , 366 m monofilament trammel net , 6.1 m otter trawl ). For the past 20 years, countless projects have been conducted to test the effectiveness of TEDs on crab traps, and only three states (New Jersey, Maryland, and Delaware) currently have regulations mandating the implantation of TEDs. Th e mandatory implementation of TEDs on crab traps is not getting through legislation fast enough and their needs to be a shift in research priorities. If the mandatory implementation of TEDs were to pass legislation and crab fisherman would abide by those r egulations, then a combination of TEDs and enlarged cull rings would be a good modification where larger individuals will still be prevented from entering the traps and smaller individuals that would get in through the TED can escape through the cull rings once inside. If no legislation gets passed, then I recommend a shift in research from testing the

PAGE 66

66 effectiveness of TEDs to the development of an escape mechanism or a modification to the existing crab trap that facilitates M. terrapin escape while not aff ecting crab capture rates and size. A modification such as an escape panel that opens outward needs to have enough resistance to only allow M. terrapin to push open and not C. sapidus . Since M. terrapin spend most of their time trying to do to the top leve l of the trap, the other modification, which entails the addition of an entrance on the top level of the trap would be the easiest modification to manufacturers; however placement of the new entrance that minimizes or keeps C. sapidus trespassing rates the same while increasing M. terrapin escapement need to be tested in the lab. The proposed modifications represented in Chapter 4 and any other modifications need to be tested thoroughly before anything gets presented to crab trap manufacturers. I hope to be gin experiments testing the modifications proposed in Chapter 4 in 2016.

PAGE 67

67 LIST OF REFERENCES Auger, P. J. 1989. Sex ratio and nesting behavior in a p opulation of Malaclemys terrapin displaying temperature d ependent sex determination. Ph.D. Dissertatio n, Tufts University, USA. Avissar , N . G . 2006. Changes in population structure of diamondback terrapins ( Malaclemys terrapin terrapin ) in a previously surveyed creek in southern New Jersey. Chelo nian Conservation and Biology 5: 154 159. Baker, P.J ., A. Thomson , I. Vatnick, and R.C. Wood. 2013. Estimating survival times for northern diamondback terrapins, Malaclemys terrapin terrapin , in submerged crab pots. Herpetologica l Conservation and Biology 8: 667 680. Beck, M.W., K.L. Heck , K.W. Able , D.L. Ch ilders , D.B. Eggleston , B.M. Gillanders , B. Halpern , C.G. Hays , K. Hoshino , T.J. Minello , R.J. Orth , P.F. Sheridan , and M.R. Weinstein. 2001. The identification, conservation, and management of estuarine and marine nurseries for fish an d invertebrates. Bio science 51: 633 641. Bishop, J.M. 1983. Incidental capture of diamondback t errapins by crab traps. Estuaries 6:426 430. Boykin, C. S. 1999. The Status of the Ornate Diamondback Terrapin at Tarpon Key, in the Pinellas National Wildlife Refuge. Senior Researc h Project, Eckerd College, USA . Boykin, C. 2004. The status and d emography of the ornate diamond back terrapin ( Malaclemys terrapin macrospilota ) within the Saint Marsh Aquatic Preserve. Unpubl. report to the Florida Dept. of Environmental Protecti on. 48 pp. Breine , J.J., J. Maes , P. Quataert , E. Van den Bergh , I. Simoens , G. Van Thuyne, and C. Belpaire. 2007. A fish based assessment tool for the ecological quality of the brackish Schelde estuary in Flander s (Belgium). Hydrobiologia 575: 141 159. B urber , J. 1976 . Behavior of hatchling diamondback terrapins ( Malaclemys terrapin ) in the field. Copeia 1976:742 748. Butler, J. A. 2000. Status and distribution of the Carolina diamondback t errapin, Malaclemys terrapin centrata , in Duval County, Florida. F i nal Report of Project NG94 103 to the Florida Fish Wildl. Conserv. Comm. 52 pp. Butler, J. A. 2002. Population e cology, home range, and seasonal movements of the Carolina diamondback t errapin, Malaclemys terrapin centrata , in n ortheastern Florida. Final Report of Project NG96 007 to the Florida Fish Wildl. Conserv. Comm. 65 pp. Butler, J.A., C. Broadhurst, M. Green, and Z. Mullin. 2004. Nesting, nest predation and hatchling emergence of the Carolina diamondback t errapin, Malaclemys terrapin centrata , in n ortheastern Florida. A merican Midland Naturalist 152:145 155 .

PAGE 68

68 Butler, J.A and G.L Heinrich. 2007. The effectiveness of bycatch reduc tion devices on crab pots at reducin g capture and mortality in dia mondback terrapins ( Malaclemys terrapin ) in Florida. Estu aries and Coasts 30:179 185. Butler, J.A. and G.L. Heinrich. 2013. Distribution of the ornate diamondback t errapin ( Malaclemys terrapin macrospilota ) in the Big Bend Region of Florid a. Southeastern Naturalist 12:552 567. Cagle, F. R. 1939. A s yste m of marking turtles for future i dentification. Copeia 1939:170 173. Cagle, F. R. 1952. A Louisiana terrapin p opulation ( Malaclemys ). Copeia 1952:74 76. Carr , A. 1952. Handb ook of Turtles. Cornell Univer sity Press, Ithaca, NY, USA. Clowes, E. L. 2013. Influences of vegetation on northern diamondback t errapin ( Malaclemys terrapin terrapin ) nest site selection. Ph . D . Dissertation, Ohio University, USA. Cohen, J. 1997. Population, economics, environment and culture: an introducti on to human carrying capaci ty. Journal of Applied Ecology 34:1 325 1333. Coleman, A., T. Wibbels , K. Marion , D. Nelson, and J. Dindo. 2011 . Population ecology of the diamondback terrapin ( Malaclemys terrapin pileata ) in Alabama. . Ph.D. Dissertation, University of Alabama, USA . Cost Grasso, B. Hannon, K. Limburg, S. Naeem, atural capital. Nature 387:253 260. Davis , C.C. 1 942. A study of the cra b pot as a fishing gear. Chesa peake Biologica l Laboratory Publication No. 53 . 20pp. Dennison, W. C., R. J. Orth, K. A. Moore, J. C. Stevenson, V. Carter, S. Kollar, P. W. Bergstrom, and R. A. Batiuk. 1993. Assessing water quality wit h submersed aquatic vegetation. BioScience 43:86 94. Dorcas, M. E., J. D. Willson, and J. W. G ibbons. 2007. Crab trapping causes population decline and demographic changes in diamond back terrapins over two decades. Biol. Conserv. 137:334 340. Drabeck , D . H . , M.W.H Chatfield , and C.L. Richards d iamondback terrapin ( Malaclemys terrapin ) pin the wake of the Deepwa ter Horizon Oil Spill: Insights from population genetic and contaminant analyses. J Herpetol 48:125 136 . Edmonds, J.H. and R.J. Brooks . 1996. Demography, sex ratio, and sexual size dimorphism in a northern population of common musk turtles ( Sternotherus odoratus ). Can. J. Zool. 74: 918 925.

PAGE 69

69 Enge, K. M. 1993. Herptile use and trade in Florida. Final Performanc e Report to the Florida Game and Fresh Water Fish Commi ssion, Nongame Wildlife Program. 102 pp. Enge, K. M. 2005. Commercial harvest of amphibians and reptiles in Florida for the pet trade. Pp. 198 211 in W.E M eshaka Jr. and K. J. Babbitt (E ds .), Amphibians and Reptiles: Status and Conser vation in Florida. Krieger, USA. Ernst, C.H. and J.E. Lovich. 2009. Turtles of the United States and Canada. 2nd ed. Johns Hopkins Uni versity Press, USA. ESRI 2001. ArcGIS Desktop: Release 10. Redlands, CA: Envir onmental Systems Research Institute. Estep, R.L. 2005. Seasonal movement and habitat use patterns of a diamondback t errapin ( Malaclemys terrapin ) p opulation. M.S. Thesis, College of Charleston, USA. Ewert, M.A. and C.E. Nelson. 1991. Sex deter mination in turtles : diverse patterns and some pos sible adaptive values. Copeia 1991:50 69. Franke, J. and T. M. Telecky. 2001. Reptiles as Pets: An E xamination of the Trade in Live Reptiles in the United States. Humane Society of the United States. Washington, DC. F plan. Available at http://myfwc.com/conservation/special initiatives/fwli/actionplan/. Accessed on 23 October 2014. Gehrt, S.D., G.F. Huber, and J.A. Ellis. 2002. Long term population trends of raccoons in Illinois. Wildlife Society Bulletin 30:457 463. Gibbons, J.W., J.E. Lovich, A.D. Tucker, N.N. Fitzsimmons, and J.L. Greene. 2001. Demographic and ecological factors affecti ng con servation and management of the d iamondback terrapin ( Mala clemys terrapin ) in South Carolina. Chel. Conserv. Biol. 4:66 74. Gibbons, J. W., D. E. Scott, T. J. Ryan, K. A. Buhlmann, T. D. Tuberville, B. S. Metts, J. L. Greene, T. Mills, Y. Leiden, S. Poppy, and C. T. Winne. 2000. The globa 666. Green, A. D., K.A. Buhlmann, C. Hagen , C. Romanek, and J.W. Gibbons. 2010. Mercury contamination in turtles an d implications for human health . Journal of Environmental Health 72(10):14 22. Harden , L . A ., N.A. Diluzio, J.W. Gibbons , M.E. Dorcas. 2007. Spatial and thermal ecology of the diamondback terrapin ( Malac l e mys terrapin ) in a South Carolina salt marsh. Journal of the North Carolina Academy of Sci ences 123:1 54 162.

PAGE 70

70 Harden, L.A., S.E. Pittman, J.W. Gibbons, and M.E. Dorcas. 2009. Development of a rapid assessment technique for diamondback terrapin ( Malaclemys terrapin ) populations using head count surveys. Applied Herpetology 6: 237 245. Hart, K.M. 1999. Declines in diamondbacks: terrapi n population modeling and implications for management. M.S. Thesis, Duke University, USA. Hart, K. M. 2005. Population biology of diamondback t errapins ( Malaclemys terrapin ): Defining and reducing threats across their geographic r ange. Ph.D. Dissertation, Duke Un iversity, USA . Hart, K.M. and L.B. Crowder. 2011. Mitigating by catch of diamondback terrapins in crab p ots. Jour nal of Wildlife Management 75:264 272. Hart, K . M . and D . S . Lee. 2007. The d iamondback terrapin: the biology, ecology, cultural history, and conservation status of an obligate estuarine turtle. Studies in Avian Biology 32:206 213. Hart, K.M. and C.C . McIvor. 2008. Demography and ecology of mangrove diamondback terrapins in a wilderness area of Everglades National P ark , Florida, USA. Copeia 2008:200 208. H askett, K. 2011. Abundance and movement of the Texas diamondback terrapin in the Deer Island c omplex, Galveston, TX. M.S. Thesis, Universi ty of Houston Clear Lake, USA Hilde brand, S.F. 1932. Growth of diamond back terrapins: size attained, sex ratios and longevity. Zoologica 9:551 563. Hipes, D., D.R. Jackson, K. NeSmith, D. Printiss, and K. Brandt. 2001. Field Guide to the Rare Animals of Florida. Florida Natural A reas Inve ntory, Tallahassee, FL, USA. Hoffman, C.O., and J.L. Gottschang. 1977. Numbers, di stribution, and movements of a r accoon population in a suburban residential community. Journal of Mammalogy 58:623 636. Hogan, J. L . 2003. Occurrence of the diamondback terra pin ( Malaclemys terrapin littoralis ) at South Deer Island in Galveston Bay, Texas, April 2001 May 20 02. Open File Report 02.022 to the U.S. Geological Survey. 24 pp. Horn, E. E. 2012. Life history v ariation in the diamondback t errapin ( Malaclemys terrapin ) . M.S. Thesis, Hofstra University, USA. Hurd, L.E., G.W. S medes, and T.A. Dean. 1979. An ecological study of natural population of diamondback t errapins ( Malaclemys terrapin ) in a Delaware salt marsh. Estuaries 2:228 233. Iverson, J. B., H. Higgins, A. Sir ulnik, and C. Griffiths. 1997. Local and geographic variation in the reproductive biology of the snapping turtle ( Chelydra serpentina ). Herpetologica 53:96 117.

PAGE 71

71 Iverson, J. B. and G.R. Smith . 1993. Reproductive ecology of the painted turtle ( Chrysemys pict a ) in the Nebraska sandh ills and across its range. Cop eia 1993:1 21. King, P. and J.P. Ludlam. 2014. Status of diamondback t errapins ( Malaclemys terrapin ) in North Inlet Winyah Bay, South Carolina. Che lonian Conservation and Biology 13(1): 119 124. Krysko, K.L., K.M. Enge, an d P.E. Moler. 2011. Atlas of amphibians and r ep tiles in Florida. Final Report of Project Agreement 08013 to the Florida Fish and Wildlife Conser vation Commission . 524 pp. Levesque , E. M. 2000. Distribution and ecology of the diamondback t errapin ( Malaclemys terrapin ) in South Carolina salt m arshes. M.S. T hesis, University of Charleston, USA. Lovich, J. E. and J. W. Gibbons. 1990. Age at maturity influences adult sex ratio in the turtle Malaclemys terrapin . Oikos 59:126 134. Lovich, J.E. and J.W. Gibbons. 1992. A review of techniques for quantifying sexual size dimorphism. Growth Development and Aging 56:269 281. Mann, T. M. 1995. Pop ulation surveys for diamondback t errapins (Malaclemys terrapin) and gulf salt marsh s nakes (Nerodia clarkii clarkii) in Mississip pi. Mississippi Mus. Nat. Sci. Mus. Tech. Rep. 37:1 75. McCauley, R.H. 1945. The r eptiles of Maryland and the District of Columbia. Privately printed. Hagerstown, Maryland, USA. MacLaren, P.A. 1992. Raccoon ( Procyon lotor ) depredation on lis ted and other species of native fauna. Florida Department of Natural Resources. Office of Resource Management, Gainesville , FL, USA . Mali I., M.W. Vandewege , S.K. Davis, and M.R.J. Forstner. 2014. Magnitude of the freshwater turtle exports from the long term trends and early effects of newly implemented harves t management regimes. PLoS ONE 9:e86478. Martinez, M.L., A. Intralawan, G. Vazquez, and O. Perez Maqueo. 2007 . The coast of our world: Ecolog ical, economic and social importance. Ecological Economics, 63: 254 272. Mealey, B., J.D. Baldwin, G.B. Parks Mealey, G.D. Bossart, M.R.J. Forstner. 2014. Characteristics of mangrove terrapins ( Malaclemys terrapin rhizophorarum ) inhabiting altered and natur al mangrove islands. Journal of North American Herpetology 1:77. Mitchell, J. C. and S.C. Walls. 2013. Nest site selection by diamond backed t errapins ( Malaclemys terrapin ) on a mid atlantic barrier i sland. Chel onian Conservation and Biology 12(2): 303 3 08. Moll, E.O. 1979. Reproductive cycles and adaptation. Pp. 305 332 in M. Harles and H. Morlock (Eds.), Turtles: Perspectives and Research. John Wiley and Sons, Inc., USA.

PAGE 72

72 Montevicchi, W. A. and J . Burger. 1975. Aspects of the reproductive biology of the nor thern diamondback t errapin, Malaclemys terrapin . American Midland Naturalist 94:166 175. Morreale, S. J. 1992. The status and population e colo gy of the diamondback t errapin, Malaclemys terrapin , in New York. Okeanos Ocean Research Foundation, Inc. Hampton Bays. 75pp. Munscher, E.C., E.H. Kuhns, C.A. Cox, and J.A. Butler. 2012. Decreased n est mortality for the Carolina diamondback t errapin ( Malaclemys terrapin centrata ) following removal of Raccoons ( Procyon lotor ) from a nesting beach in northeastern Florida. Herpetological Conservation Biology 7:167 184. Pfau, B. and W. M. Roosenburg. 2010. Diamondback terrapins in Maryland: researc h and conservation. Radiata 19: 2 34. R Development Core Team. 2015 . R: A language and environment for statistical computi ng. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3 900051 07 0, URL http://www.R project.org/. Radzio, T. A. and W. M. Roosenbur g. 2005. Diamondback terrapin mortality in the American eel pot fishery and evaluation of a by catch reducti on device. Estuaries 28:620 626. Robinson, G.D. and W.A. Dunson. 1976. Water and sodium balance in the estuarine diamondback terrapin ( Malaclemys ). J. Comp. Physiol. 105:129 152. Roose nburg, W.M. 1990. Chesapeake diamondback terrapin investigation 1987, 1 988 and 1989. Final Report to the Chesapeake Research Conso rtium Pub. No 133. 84 pp. Roosenburg, W. M. 1991. The Diamondback Terrapin: Habitat Requirements, Population Dynamics, and Opportunities for C onservation. New Perspectives in the Chesapeake System: A Research and Management and Partnership. Proceedings of a Conference. Chesapeake Research Consortium Pub. No 137. Solomons, Md. pp. 237 234. Roosenburg, W. M. 1992. Life history consequences of nest site choice by the diamondback terrapin, Malaclemys te rrapin . Ph.D. Dissertation, Univ. of Pennsylvania, USA. Roosenburg, W.M. and K.C. Kelley. 1996. The effect of egg size and incubation temperature on growth in the turtles, Malaclemys terrapin . Jou rnal of Herpetology 30:198 204. Roosenburg, W.M., W. Cresko, M. Modesitte, and M .B. Robbins. 1997. Diamondback t errapin ( Malaclemys terrapin ) mortality in crab p ots. Conservation Biology 11:1166 1172. Roosenburg, W. M., K. L. Haley, and S. McGuire. 1999. Habitat selection and movements of diamondback terrapins, Malaclemys terrapin , in a Maryland estuary. Chelonian Conservation Biology 3:425 429. Roosenburg, W.M., and J.P. Green. 2000. Impact of a by catch reduction device on diamondback terrapin and blue crab capture in crab pots. Ecol. Appl. 10:882 889.

PAGE 73

73 Roose nburg, W.M. 2004. The impact of crab pot fisheries on terrapin ( Malaclemys terrapin ) populations: where are we and where do we need to go? Pp. 23 30 i n C. Swarth, W. M. Roosenburg, and E. Kiviat (Eds.), Conservation and Ecology of Turtles of the Mid Atlant ic Region: A Symposium. Bi bliomania, USA. Sasser, L. D., K. L. Monroe, and J. N. Schuster. 1994. Soil s urvey of Franklin County, Florida. United States Department of Agric ulture. 192 pp. Seigel, R.A. 1980. Nesting habits of diamondback t errapins (Malaclemys terrapin) on the Atlantic c oast of Florida. Transactions of the Kansas Academy of Science 83:239 246. Seigel, R.A. 1984. Parameters of two populations of diamondback terrapins ( Malaclemys terrapin ) on the Atlantic coast of Florida. Pp. 77 87 in R.A. Siegel, L.E. Hunt, J.L. Knight , L. Malaret, and N.L. Zuschlag (E ds. ), Ver tebrate Ecology and Systematics: A Tribute to Henry S. Fitch. Special Publication 10 of the Museum of National History, University of Kansas, USA. Seigel, R.A. 1993. Apparent lo ng term decline in diamondback terrapin populations at the Kennedy Space Center, Florida. Herpetol. Rev 24:102 103. Seigel, R.A. and J.W. Gibbons. 1995. Workshop on the ecology, status, and management of the diamondback terrapin ( Malaclemys terrapin ) Savannah River Ecology Laboratory, 2 August 1994: final results and recommendations. Chelo nian Conservation and Biology 1: 241 243. Selman, W. and Baccigalopi. 2012. Effectively sampling Louisiana diamondback t errapin ( Malaclemys terrapin ) p opulations, with description of a New Capture Technique. Herpetological Review 43 (4):583 588. Sheridan, C.M. 2010. Mating system and dispersal patterns in the diamondback t errapin ( Malaclemys terrapin ). Ph.D . Dissertation. Drexel University, USA. Silliman , B . R . and M.D. B ertness. 2002. A trophic cascade regulates salt marsh primary production. Proceedings of the National Academy of Sciences of t he United States of America 99: 10500 10505. Smeenk, N.A . 2010. The population ecology of a headstart supple mented population of di amondback terrapins ( Malaclemys terrapin ) at the Poplar Island Environmental Restoration Project in the mid dle Chesapeake Bay. M.S. Thesis, Ohio Univ. 71 pp. Spivey, P.B. 1998. Home range, habitat selection, and diet of the diamondback t errapin ( Malaclemys terrapin ) in a North Carolina e stuary. M.S. Thesis, Univers ity of Georgia, USA.

PAGE 74

74 Titus, J. G., D. E. Hudgens, D. L. Trescott, M. Craghan, W. H. Nuckols, C. H. Hershner, J. M. Kassakian, C. J. Linn, P. G. Me rritt, T. M. McCue, J. F. Connell, J. Tanski, an d J. Wang. 2009. State and local governments plan for development of most land vulnerable to rising sea level along the US Atlantic Coast. Environmental Research Letters 4(4):044008. Tucker , A . D . , N.N. Fitzsimmons , and J.W. Gibbons. 1995. Resource partit ioning by the estuarine turtle Malaclemys terrapin : trophic, spatial and temporal foraging constraints. Herpetologica 51: 167 181. Ultsch, G. R. 2006. The ecology of overwintering among turtles: where turtles overwinter and its consequences. Biological Revi ews 81:339 367. Wilke, A. L., D. F. Brinker, B. D. Watts, A. H. Traut, R. Boettcher, J. M. McCann, B. R. Truitt, and P. P. Denmon. 2007. American oystercatchers in Maryland and Virginia, USA: status and distribution. Waterbirds 30:152 162. Wood, R.C. 199 2. Mangrove Terrapin. Pp. 204 209, In P.E. Moler (E d.) Rare and Endan gered Biota of Florida. Vol. III. Amphibians and Reptiles. University Press of Florida, Gainesville. FL. Wood , R. C. 1997. The impact of commercial crab traps on north ern diamondback terrapins, Malaclemys terrapin terrapin . In J. Van Abbema (ed.), Proceedings: Conservation, Restoration, and Management of Tortoises and Turtles An International Confer ence, pp. 21 27. New York Turtle and Tortoise Society, New York, USA. Wood, R. C. and R. Herlands. 1997. Turtles and tires: the impacts of roadkills on Northern Diamondback Terrapin, Malaclemys terrapin terrapin , populations on the Cape May Peninsula, southern New Jersey, USA. Pp. 46 53 i n J. V. Abbema, and P. C. H. Pritchard (E ds.), Conservation, Restoration, and Mana gement of Tortoises and Turtles: An Inte rnational Conference. New York Turtle and Tortoise Society, USA.

PAGE 75

75 BIOGRAPHICAL SKETCH Eric Suarez rece f rom the Department of Wildlife Ecology and C onservation at t he University of Florida . While pursuing his m in wildlife ecology and conversation , Eric was advised by Dr. Steve Johnson . His primary research is focused on the ecology and conserv ation of various species of chelonians in northern Florida . Currently his research focuses on population dynamics of the alligator snapping turtle in multiple rivers throughout northern and the panhandle of Florida. He is also researching the population dy He was hired by the Florida Museum of Natural History in 2010 and then by the Florida Fish and Wildlife Conservation Commission in 2011. During his time at the museum unde r the supervision of Dr. Kenneth Krysko, he catalogued and then transferred reptiles and amphibians to new holding tanks. While working for the Florida Fish and Wildlife Conservation Commission, he worked on numerous projects concerning chelonian ecology . Previous to his employment in 2010, he spent 4 year as a volunteer on numerous projects involving freshwater turtles in north Florida under the supervision of Dr. Gerald Johnston.