W&M ScholarWorks
Dissertations, Theses, and Masters Projects Theses, Dissertations, & Master Projects
2006
The Influence of Predation on the Nesting Ecology of Diamondback Terrapins (Malaclemys terrapin) in the Lower Chesapeake Bay
Victoria Ann Ruzicka College of William & Mary - Arts & Sciences
Follow this and additional works at: https://scholarworks.wm.edu/etd
Part of the Biodiversity Commons, Ecology and Evolutionary Biology Commons, and the Natural Resources and Conservation Commons
Recommended Citation Ruzicka, Victoria Ann, "The Influence of Predation on the Nesting Ecology of Diamondback Terrapins (Malaclemys terrapin) in the Lower Chesapeake Bay" (2006). Dissertations, Theses, and Masters Projects. Paper 1539626849. https://dx.doi.org/doi:10.21220/s2-6856-rr16
This Thesis is brought to you for free and open access by the Theses, Dissertations, & Master Projects at W&M ScholarWorks. It has been accepted for inclusion in Dissertations, Theses, and Masters Projects by an authorized administrator of W&M ScholarWorks. For more information, please contact [email protected]. THE INFLUENCE OF PREDATION ON THE NESTING ECOLOGY OF DIAMONDBACK TERRAPINS (MALACLEMYS TERRAPIN) IN THE LOWER CHESAPEAKE BAY
A Thesis
Presented to
The Faculty of the Department of Biology
The College of William and Mary in Virginia
In Partial Fulfillment
Of the Requirements for the Degree of
Master of Science
by
Victoria Ann Ruzicka
2006 APPROVAL SHEET
This thesis is submitted in partial fulfillment of
the requirements for the degree of
Master of Science
Victoria Ann Ruzicka
Approved by the Committee, October 2006
Randolph M. Chambers, Chair
/ M - <*Gre/ory M. Capdfli
John P. Swaddle DEDICATION
To my mom and dad, the two most important people in my life. You have always supported me through all of my many endeavors and challenged me to strive towards my dreams. I owe my past, present, and future success to both of you. I love you. TABLE OF CONTENTS
Page
Acknowledgements v
List of Table vi
List of Figures vii
Abstract ix
Chapter I. Introduction 2 Literature Review 4 Reproductive/Life History Traits 4 Threats to Turtle Reproduction 7 Diamondback Terrapins—A Case Study 13 Diamondback Terrapin Nesting Ecology-Natural History 16 Predation 19 Objectives 23
Chapter II. Methodology 25 Nesting Ecology-Natural History 30 Influence of Raccoon Predation 37
Chapter III. Results 41 Nesting Ecology-Natural History 41 Influence of Raccoon Predation 70
Chapter IV. Discussion 87 Nesting Ecology-Natural History 87 Influence of Raccoon Predation 101 Summary 105
Bibliography 110
Vita 124
iv ACKNOWLEDGEMENTS
There are many people that I must thank for their assistance on this project. First and foremost, I need to thank my committee. I owe the biggest debt of gratitude to my graduate advisor Dr. Randy Chambers. Thank you for all of your encouragement, guidance, and support during this project. Without your passion for turtles, I would have never found my love of diamondback terrapins. I would also like to thank my committee members Dr. John Swaddle and Dr. Greg Capelli for all of their input on my research and this manuscript.
There are many other people and organizations that I need to acknowledge for their assistance on the initiation, execution, and completion of this project. Willy Reay and Chris Clapp at CB-NERRVA, Virginia Department of Game and Inland Fisheries (VDGIF permit #026956), and the Institutional Animal and Use Committee at the College of William & Mary (IACUC permit #0420) authorized access to the Goodwin Islands and permits to handle turtles. Duane and Vera Heitkemper allowed me to use their property to store my canoe during the 2005 field season. Matthew Wolak was my dedicated terrapin field work partner, spending day after day in the hot summer sun canoeing, catching terrapins, and collecting nest data. Dave Wilcox and the Virginia Institute of Marine Science SAV monitoring program supplied the 2005 aerial photograph of the Goodwin Islands that I used in this manuscript. Tim Russell provided his knowledge on GPS/GIS related topics and kept me laughing through all of the stress. Lastly, George Gilchrist was always available to offer statistical advice when I had questions.
This acknowledgement would not be complete without thanking all of my William and Mary biology graduate student friends. These people made daily life inside and outside of Millington and Keck lab truly enjoyable. They were always there for me, through good and bad times, providing a helping hand when I needed assistance in the field and lending an ear when I needed to vent my frustrations. I will always cherish our time together and I will take my memories of all of you, The College of William and Mary, and graduate school with me for the rest of my life.
Finally, I would like to thank my family. Without your love and support these last two years I am not sure if any of this would have been possible. I am truly grateful to be surrounded by such wonderful people.
v LIST OF TABLES
Table Page
1. Habitat assessment of the Goodwin Islands 28
2. Total area/linear edge ratio of the Goodwin Islands 29
3. Nest totals by island 43
4. Egg totals by island 45
5. Nest totals by month 48
6. Incubation period 53
7. Environmental variable PC A extraction values 64
8. Loading factors for environmental variables PC A 65
9. Regression coefficients 70
10. Predation rate for nests and eggs 71
11. Mammalian predator movements 73
12. A comparison of nesting characteristics in various subspecies of Malaclemys 91
vi LIST OF FIGURES
Figure Page
1. York River watershed map 26
2. Map of the Goodwin Island complex 27
3. Diamondback terrapin nesting sites on the Goodwin Islands 42
4. Comparison of clutch size by island 46
5. Nesting activity 49
6. Nesting activity and partially depredated nests 50
7. Nesting activity and minimum air temperature 54
8. Nesting activity and maximum air temperature 55
9. Nesting activity and minimum water temperature 56
10. Nesting activity and maximum water temperature 57
11. Nesting activity and morning low tide 59
12. Nesting activity and morning high tide 60
13. Nesting activity and aquatic salinity 61
14. Nesting activity and relative humidity 62
15. Nesting activity and mammal tracks 63
16. Linear regression of nesting activity and PCI 66
17. Linear regression of nesting activity and PC2 67
18. Linear regression of nesting activity and PC3 68
19. Comparison of mammalian predator movements by island 74
20. Comparison of mammalian predator movements by month for each island 75
21. Nesting site habitat for all nests by nest status 77
vii 22. Nesting site habitat for all nests by island 78
23. Nesting site habitat for depredated nests by island 79
24. Nesting site habitat for hatched/emerged nests by island 80
25. Comparison of nearest neighbor distances for focal nests 82
26. Within island comparison of nearest neighbor distances for focal nests 83
27. Comparison of the mean number of nests occupying focal nest buffers 84
28. Within island comparison of the mean number of nests occupying focal nest buffers 85
viii ABSTRACT
The reproductive success of oviparous organisms, like turtles, is largely determined by reproductive life history traits and characteristics of the nesting site. Recent anthropogenic changes to turtle nesting sites have led to increases in populations of native, subsidized turtle nest predators. For many turtle species, a knowledge gap exists in the basic understanding of the reproductive life history characteristics and the potential effect of increasing nest predation. This information must be determined if management for population and/or species survival is to be effective.
For my research, I examined the nesting ecology of the diamondback terrapin (Malaclemys terrapin), the only North American turtle to reside exclusively within brackish-water environments, inhabiting estuaries from Massachusetts to Texas. From May 11-October 2, 2005,1 collected information on the nesting ecology and predation pressures on a terrapin population on the Goodwin Islands, an island complex within the lower Chesapeake Bay, VA. The complex was composed of three islands: Western, Middle, and Eastern Island. Differences in island habitat allowed for between island comparisons of the number of nests and eggs, predation rate, predator movements, and nest site density.
I found 1,362 nests on the Goodwin Islands during the 2005 nesting season, of which the majority were depredated or partially depredated. The number of hatched/emerged nests increased in a west-east gradient across the complex. Raccoons were the primary nest site predator at this site. Overall nest predation rate on the islands was 79.5% and egg predation rate was 84.7%. Both nest and egg predation rates decreased from west-east across the complex. Predator movements were compared across the three islands using 12 predator movement plots. Total predator movements were significantly different across islands (ANOVA, p<0.05), and decreased in a west-east gradient across the island complex. Predator movements differed within islands for the months of May, June, and July (ANOVA, p<0.05), increasing throughout the period for each island. The trend in predator activity for the season was also significantly different across islands (ANOVA, p <0.001). Both successful and depredated nests on the Eastern Island were located farther from their nearest neighbors and surrounded by fewer nests than nests on the other islands, indicating lowest nest density on this island where nest success was highest.
On the Goodwin Islands, I measured a west-east gradient in reproductive success, predation rate, and predator movements across the complex. Based on these results, raccoon predators appear to inhabit the Western Island, forage most intensely there, and move eastward through the complex in search of food. Eastern Island had the greatest reproductive success among the three islands, possibly due to its broadly distributed nesting area and distant location away from the predator source (Western Island). Continued predation on eggs by raccoons and low recruitment at this site may threaten the survival of the Goodwin Island terrapin population.
ix THE INFLUENCE OF PREDATION ON THE NESTING ECOLOGY OF DIAMONDBACK TERRAPINS (MALACLEMYS TERRAPIN) IN THE LOWER CHESAPEAKE BAY CHAPTER I
INTRODUCTION
Turtles are considered the oldest and most primitive reptiles (Ernst and Barbour
1989), part of the first group of vertebrates to adapt to life on dry land (Gibbons 1990).
Fossil evidence suggests that they evolved into their shelled form over 200 million years
ago during the Triassic Period. All extant turtles belong to the order Testudines, within the subclass Anapsida (Ernst and Barbour 1989). Modem turtle species have radiated throughout the world and within a variety of habitats. They can be found on all continents except Antarctica, occupying terrestrial, fresh water, brackish water, and marine environments. These aquatic environments include swamps, marshes, bogs, rivers, oceans, and streams (Ernst et al. 1994). In each environment, turtle species have evolved distinct life history traits that allow individuals to convert finite energy resources into the next generation (Pandian and Vemberg 1987). As for all species, the goal is to reproduce successfully, but the path to successful reproduction among turtle species is varied
(Gibbons 1990).
An organism’s basic life history is composed of distinct phenotypic traits: size at birth, growth patterns, age and size at maturity, offspring number, age- and size- specific reproductive investments and mortality schedules, and lifespan. Evolution has produced many successful combinations of these traits, resulting in a unique grouping and interaction of traits that determines an individual’s overall fitness (Steams 1992).
2 3
Among these traits, reproduction plays a critical role in the life history and overall fitness
of an organism (Gibbons 1990). Many life history traits directly influence an individual’s reproductive success (Steams 1992). The amount of energy an individual places into reproductive effort in part determines whether their genes are passed on to the next generation (Pandian and Vemberg 1987). The various aspects in the timing of reproduction (mating, nesting season duration, incubation period) provide further details and insight into life history characteristics and reproductive success (Gibbons 1990). The interaction of an organism with its physical environment during reproduction also determines reproductive success (Bell 1991). Nest site selection (Wilson 1998) and the macro- and micro- habitats (predation threats, moisture, and temperature) (Spencer and
Thompson 2003) of nest sites can affect the survival of both the females in the act of nesting as well as offspring produced within these habitats. A female has the ability to select the best nesting site that will ultimately improve her offspring’s chances of survival
(Bell 1991).
For this thesis, I have chosen to examine some of the reproductive life history characteristics of a population of diamondback terrapins (Malaclemys terrapin) occupying a small island complex within the York River sub-estuary of Chesapeake Bay in southern Virginia. Although terrapins are considered a keystone predator in saltmarsh habitats, life history information has never been collected for any terrapin population in
Virginia. Given ongoing threats to terrapins in the form of juvenile and adult mortality in crabpots, loss of nesting habitat, and nest destruction by mammalian predators, the information provided in this project will assist in the development of management plans for this unique estuarine species whose conservation status in Virginia is unknown. 4
Literature Review Reproductive/Life History Traits
Many reproductive life history traits of turtles are noteworthy. First, lifespan of a turtle is long. Turtles are one of the longest living vertebrates alive today (Gibbons 1990), many small and large sized turtle species are estimated to live over 20 and 30 years of age (Gibbons 1987). Biegler (1966) reported three of the oldest examples of turtles, the
Mediterranean tortoise ( Testudo graeca), Aldabra tortoise ( Geochelone gigantea), and
European pond turtle ( Emys orbicularis), that were confirmed to live more than 50, 60, and 70 years respectively. Also, many turtles have a long maturation period characterized by late sexual maturity (Ernst et al. 1994). Whereas small sized turtles, like eastern mud turtles ( Kinostemon subrubrum), reach sexual maturity at a few centimeters length around 4 to 5 years of age (Mahmoud 1967), larger sized turtles, like the loggerhead turtle ( Caretta caretta), reach sexual maturity at 65-104 cm in length (Dodd 1988c) around 10-30 years of age (Frazer and Ehrhart 1985). Extended longevity and late maturation allow for turtles to be iteroparous organisms, capable of producing multiple clutches of eggs both within and between nesting seasons (Gibbons 1987). Numerous environmental factors, however, can contribute to nest failure, leading to large variability in annual reproductive success (Plummer 1976; Burger 1977).
Since turtles are ectotherms, much of their reproductive activity is dependent on temperature and other environmental cues. Most turtles of the northern hemisphere are primarily active during the warmer months from April through October. Mating activities and the nesting season begin once water and air temperatures reach 15-20°C during the late spring (Ernst et al. 1994). 5
During nesting, female aquatic and semiaquatic turtles must leave the water to oviposit eggs on dry land above the high water line (Harless and Morlock 1979). Sea turtles like the Blanding’s turtle ( Emydoidea blandingi) will travel 2-1115 m to place their nests away from the high tide line (Congdon et al 1983). Obbard and Brooks (1980) found that nesting snapping turtles (C. serpentina) sometimes traveled more than 5km, away from the waters edge.
Nest site selection and the act of nesting are also thought to be triggered by environmental cues. Research on diamondback terrapins (Malaclemys terrapin) suggests that environmental factors such as tidal phase, weather, vegetation cover type, and slope of the land all influence turtle nesting activity (Burger and Montevecchi 1975). Multiple nest site characteristics such as slope, moisture, and salinity, may initiate turtle excavation of a nest (Wilson 1998; Wood and Bjomdal 2000). Nest site choice produces gregarious nesting behavior, in which nests are placed within the close vicinity of other turtle nests (Carr 1967; Doody et al. 2003). The beginning of the Blanding’s turtle (. E. blandingi) nesting season was found to correlate with spring air temperatures (Congdon et al 1983). Studies by Goode (1965) found that Australian freshwater turtles, including the eastern longneck ( Chelodina longicollis), broad-shelled ( Chelodina expansa), and
Macquarie tortoise (Emydura macquarri), tended to nest after heavy rains and during periods of high humidity. In addition to nesting and timing, turtles have been shown to exhibit nest site fidelity, returning to the same nest sites year after year (Congdon et al,
1983). Through genetic studies, this behavior has been shown in green turtles ( Chelonia mydas) to be a result of natal homing, in which sexually mature individuals return to their natal beach to nest (Meylan et al. 1990). A five year study by Valenzuela and Janzen 6
(2001) showed that female painted turtles (Chrysemys picta) exhibited significant nest site philopatry to particular geographic sites.
For most turtles, nesting substrate is composed primarily of loose soil, consisting of sand or loam with little to no vegetation. Loose substrate allows for aeration of the eggs within the nest cavity. Turtle nests may be located in dunes, road banks, railroad grade, dikes, and levees (Burger 1977; Obbard and Brooks 1980; Seigel 1980a; Snow
1982; Congdon et al. 1983). Since the sex of hatchlings in many turtle species is determined by the temperature of the nesting site environment, the location of nest sites can influence the sex ratio of a population (Ernst et al. 1994).
Nesting ecology characteristics vary from species to species and from population to population for North American aquatic turtles. Clutch size variation is apparent in most species, with the majority of turtles (60%) laying clutches composed of less than twenty eggs and only 10% regularly laying clutches greater than thirty eggs (Harless and
Morlock 1979). Some turtles, like the wood turtle (Clemmys insculpta), will only lay one clutch a year (Powell 1967). Shealey (1976) found that although Alabama map turtles
(Graptemys pulchra) averaged four clutches per season per female, individual turtles were capable of producing as many as six clutches. The number of clutches produced within a nesting season by the eastern mud turtle (K. subrubrum), was found to be greater in southern regions of the United States and fewer in northern regions (Gibbons 1983).
Further, clutch sizes of painted turtles (C. picta) tend to be larger in northern latitudes
(Moll 1973). Generally, nesting occurs earlier and lasts longer in southern climates than at northern latitudes, typically with smaller clutch sizes per nest (Harless and Morlock
1979). 7
Threats to Turtle Reproduction
The physical environment imparts environmental and evolutionary constraints on
the allocation of energy for reproduction (Pandian and Vemberg 1987). In essence,
characteristics of the nesting site environment determine in large part the reproductive
success of an individual. First, anthropogenic changes to the environment can reduce the
number of suitable places an organism can reproduce, thereby forcing nesting females to
compete for available habitat (Dunham et al. 1989). Predation of nesting females and her
eggs is another environmental factor that reduces the fitness of an individual (Bell 1991).
An increase in the number of predators surrounding nest sites will cause an increase in
adult and/or egg mortality, which ultimately will result in lower recruitment for a
population and thus threaten the overall survival of the population (Congdon et al. 1983).
On a global scale, the rapid rate of habitat change—often caused by humans— is
responsible for recent, dramatic declines in the world’s biodiversity (Wilson 1985).
Destruction of natural environments has led to habitat fragmentation, thereby reducing
suitable habitats for reproduction to isolated patches. Population extinctions of many
species unable to adjust to these anthropogenically altered landscapes has occurred
(Lande 1998). Species surviving drastic changes in their environment are forced to compete with other species and members of their own population for the remaining
suitable breeding, nesting, and feeding sites (Paton 1994; Houlahan and Findlay 2003).
Since approximately four billion people inhabit the world’s coastline, estuaries have experienced the devastating impacts of anthropogenic habitat alterations (Kennish
2002). Biologically productive coastal areas have been changed by pollution, nutrient enrichment, and destruction of natural shorelines (Kennish 1997; Fisher et al 2006). The disturbance of natural shorelines has impacted many species that rely on coastal estuarine habitats for reproductive purposes (Gibbons et al 2000; Klemmens 2000; Kennish 2002), including snapping turtles (Kinneary 1992), diamondback terrapins (Burger 1977), and sea turtles.
A secondary problem produced by anthropogenic habitat alteration is the appearance of large numbers of subsidized species (Klemens 2000). Subsidized species are native North American generalist and opportunistic organisms whose populations have dramatically increased during the past century. These species benefit from anthropogenic landscape changes and outcompete other native groups for resources.
Overabundant native species populations have been shown to reduce natural diversity in the environment by monopolizing resources, spreading diseases, and changing the relative abundance of local organisms (Garrott and White 1993). White-tailed deer
(Odocoileus virginianus) (Cote et al. 2004; Pellerin et al. 2006), cattail ( Typha spp.)
(Kostecke et al. 2004), stripped skunk ( Mephitis mephitis) (Johnson et al. 1989), red fox
(Vulpes vulpes) (Johnson et al. 1989; Renfrew and Ribic 2003), and gulls ( Larus spp.)
(Rome and Ellis 2004) are only a few species whose overabundance has had a negative impact on the sustainability of local plant, vertebrate, and invertebrate species. Coastal and inshore island populations are most at risk to increase in overabundant species
(Moore 2002). Eventually, these subsidized, overabundant species may overwhelm the more sensitive species within the habitat and threaten the ecological integrity of the natural environment (Garrott and White 1993).
As a local example of a subsidized species, the raccoon (. Procyon lotor) has thrived with human encroachment (Garrott and White 1993). Raccoon nest predation is 9 considered a major factor in reducing nest survival in almost all North American turtle species (as reviewed in Ernst and Barbour 1972; Ernst et al. 1994). Studies of fresh water
(Snow 1982; Robinson and Bider 1988; Herman et al. 1995; Marchand and Litvaitis
2004) and sea turtle populations (Stancyl et al. 1980; Ratnaswamy et al. 1997;
Garmestani and Percival 2005) indicate that raccoons dramatically decrease hatchling and egg survival. Nest predation on loggerhead turtle (Caretta caretta) nests was as high as
95% at one site in South Carolina and nearly 100% for nesting snapping turtles (Cheldra serpentine) in Massachusetts (Link et al. 1989). The number of turtle predators, such as raccoons, has been positively correlated with the extant of predation of turtle eggs and offspring (Burger and Garber 1995).
Raccoons are generalist and opportunistic predators (Garrott and White 1993) that forage throughout the night and sleep within arboreal dens during the day (Burt 1964).
Raccoon foraging is characterized by widespread movement interspersed with periods of intensive foraging on herbaceous or carnivorous food sources (Greenwood 1982). Male raccoons are larger than females and thus require a greater amount of food to sustain them energetically. As a consequence, male raccoons maintain larger home ranges than females throughout the year (Greenwood 1982; Gehrt and Fritzell 1997). The home range of male raccoons may be as large as 4,500 ha (Fritzell 1978) and minimum nightly foraging distances have been recorded as far as seven to nine miles in length (Greenwood
1982).
The resources required to satisfy the energetic requirements of the raccoon diet are commonly associated with habitat edges along aquatic environments (Greenwood
1982; Dijak and Thompson 2000; Kolbe and Janzen 2002a). Raccoons use ecotones, such 10 as forest/shrubland edges along riparian, marine, and estuarine environments, as foraging habitats (Greenwood 1982; Dijak and Thompson 2000; Erwin et al. 2001; Moore 2002;
Marchand and Litvaitis 2004). The transition zone from aquatic to terrestrial environments at ecological edges is also important to the reproductive success of turtles.
Habitat edges along riparian, wetland, and estuarine habitats provide the loose sandy soil and open canopy necessary for turtles to access nesting sites (Linck et al. 1989). Edges along aquatic environments also serve as ‘ecological traps’ for ovipositing organisms like turtles, by forcing nests to accumulate along the perimeter of a habitat and into the path of raccoons (Vickery et al. 1992; Gates and Gysel 1978; Kolbe and Janzen 2002b). The aquatic/terrestrial edge is usually the site of high nest predation (Paton 1994; Kolbe and
Janzen 2002a) and turtle eggs make up a large portion of the raccoon diet (Hamilton et al.
2002), supplying raccoons with a high energy food source (Moore 2002). Intense nest predation leads to low recruitment into adult turtle populations (Congdon et al. 1983;
Robinson and Bider 1988).
Turtle nests not predated by raccoons may be successful owing to location relative to habitat edge and immediate nest site surroundings. For example, box turtle (Terrapene omata), Blanding’s turtle (E. blandingi), and common snapping turtle (C. serpentine) nests have lower predation rates when located away from the edge of a stream or forest
(Temple 1987). Recent studies of painted turtles (C. picta) suggested that nests had less chance of raccoon predation when located away from the water’s edge (Kolbe and Janzen
2002a; Marchand and Litvaitis 2004). Also, survival of nests increased with an increase in vegetation cover for snapping turtle (C. serpentine) (Robinson and Bider 1988). Burke et al. (1998), however, found no indication that nest predation rate by raccoons on three 11
turtle species (mud turtle (K. sub rub rum), Florida river cooter ( Pseudemys concinna floridana), and red-eared slider ( Trachemys scripta)) varied with distance from the marsh edge or among habitat types.
A reduction in suitable nesting habitat may lead to increased nest density in the remaining sites, influencing nest success (Roosenburg 1994; Kolbe and Janzen 2002b).
Marchand and Litvaitis (2004) found that painted turtle nests clumped together were more likely to be predated by raccoons. Similarly, snapping turtle (C. serpentina) nests that were clumped with other nests had a higher likelihood of being predated by mammalian-predators like raccoons (Robinson and Bider 1988). Spencer (2002) found that mammalian-predated Australian fresh water turtle ( Emydura macquarii) nests were closer to other nests than were intact nests. Studies by Burke et al. (1998) (mud turtle (K. subrubrum), Florida river cooter (P. c. floridana), and red-eared slider ( T. scripta)) and
Doody et al (2003) (pig-nose turtle ( Carettochelys insculpta)), however, found that nest predation was independent of nest density.
The confluence of habitat destruction and increased nest predation by subsidized species like raccoons has many populations of turtle species at the tipping point of maintenance or local extinction (Ernst and Barbour 1989; Gibbons 1990). Under these environmental conditions, the two characteristics that were once associated with successful turtle reproduction-late sexual maturity and nest site fidelity-may now be detrimental to the survival of turtle species (Gibbons 1990; Burger and Garber 1995).
Anthropogenic changes to the environment have destroyed, altered, and decreased the number of nesting site habitats, forcing turtles to either adapt to the changes and compete for the remaining sites or find new sites for nesting (Gibbons et al. 2000; Klemens 2000). 12
Late sexual maturity means that turtles take years to mature, resulting in a slow response to rapid environmental change. Once they can reproduce nest site fidelity may force females to nest on small fragmented sites that are not suitable (Temple 1987; Roosenburg
1994). The few remaining adequate nesting sites within a population’s home range may cause competition among females for the remaining nest spots, in turn clumping nests together and possibly increasing the risk of mammalian predation (Robinson and Bider
1988; Spencer 2002; Doody et al. 2003; Marchand and Litvaitis 2004).
Recent environmental changes have benefited generalist predators, such as raccoons (Garrott and White 1993). As raccoon population numbers increase, so do the number of predation events on turtle nests (Burger and Garber 1995; Marchand and
Litvaitis 2004). Since predation threats are a major factor driving female nest site selection, turtles may choose to nest closer to the shore, increasing adult survival but decreasing offspring success by placing nests into the path of generalist mammalian predators (Spencer and Thompson 2003). Recruitment into turtle populations, although historically low due to environmental variability, may now be even lower do to an increase in nest predation (Plummer 1976; Burger 1977; Congdon et al. 1981). An increase in nest predation may ultimately alter the structure of turtle populations
(Marchand and Litvaitis 2004).
To sum, habitat degradation and nest predation are two environmental factors that may threaten the integrity of all turtle species within North America to the point of extinction within the twenty-first century (Gibbons 1990; Ernst et al. 1994). In the United
States there are currently thirty three endangered and four threatened turtle species
(Gibbons et al. 2000). Understanding the nesting patterns of turtle species is crucial to 13 determining how populations will respond to anthropogenic environmental change
(Marchand and Litvaitis 2004). For many turtle species, a knowledge gap exists in the basic understanding of the reproductive life history characteristics and new environmental constraints. This information must be determined if management for population and/or species survival is to be effective (Ernst et al. 1994).
Diamondback Terrapins-A Case Study
The diamondback terrapin (Malaclemys terrapin) is a species of turtle whose complete reproductive life history remains unknown. While many diamondback terrapin studies have been completed in the species northern range, few have focused on southern populations. For example, the nesting ecology of the diamondback terrapin has never been described for any population in the Commonwealth of Virginia (Seigel and Gibbons
1995). Previous nesting ecology studies of terrapins have determined that anthropogenic changes to the environment threaten the survival of terrapins, forcing U.S. Fish and
Wildlife Service to designate the diamondback terrapin a “status review” species.
Research on the life history and environmental constraints that influence diamondback terrapin reproductive success is needed in order to gain a better understanding of terrapin population status in the Commonwealth (Golder et al. 2000).
The diamondback terrapin is the only North American turtle to reside exclusively within brackish-water environments (Ernst et al. 1994; Moll and Moll 1994). Terrapins inhabit the saline channels of coastal salt marshes, tidal creeks, and estuaries (Orenstein
2001; Ernst et al. 1994; Moll and Moll 2004) and are suggested keystone species within the food chains of these brackish water environments, feeding primarily on marsh periwinkle (Littorina irrorata) (Silliman and Bertness 2002). Populations of 14 diamondback terrapins can be found along the eastern and Gulf coasts of the United
States from Cape Cod, Massachusetts to Corpus Christi Bay, Texas (Ernst and Bury
1982; Ernst et al. 1994; Orenstein 2001).
Diamondback terrapins are classified as members of the family Emydidae within the species-complex Chrysemys. This complex contains the freshwater genera
Chrysemys, Trachemys, Pseudemys, Deirochelys, and Graptemys plus Malaclemys.
Malaclemys, which contains the diamondback terrapin, is the only brackish-water genus found within the Chrysemys species complex (Ernst and Barbour 1972; Ernst et al. 1994).
Seven recognized subspecies of diamondback terrapins are found within the
United States (Ernst and Barbour 1972; Ernst and Bury 1982; Ernst et al. 1994). The three known Atlantic coast subspecies of diamondback terrapin are the Florida east coast diamondback ( M. t. tequesta), the Carolina diamondback (M. t. centrata), and the northern diamondback (M. t. terrapin). The northern diamondback terrapin ranges from
Cape Cod, Massachusetts to Cape Hatteras, North Carolina, and is found in coastal waters of the Commonwealth of Virginia.
Populations of the Atlantic Coast subspecies of terrapin have fluctuated dramatically over the past 100 years (Hildebrand 1932; Burger 1989). The diamondback terrapin was a staple food item in the diet of coastal Native Americans, colonists, and
African-American slaves (Garber 1990a; Moll and Moll 2004). In the late 19th and early
20th century, diamondback terrapins became a very popular food among the wealthiest
Americans, with the meat being incorporated into terrapin soup. To appease the demand for meat, the fishing industry increased the harvest of terrapins, and soon native populations of terrapins began to suffer from overexploitation. In 1891 over 89,000 15 pounds of diamondbacks were sold in Maryland fish markets and by 1920 the price had increased 500%. Overharvesting drove many diamondback terrapin populations close to extinction throughout the coastal United States (Garber 1990b).
Steps to increase the number of diamondback terrapins in the wild began as early as 1902. The Federal Bureau of Fisheries and the North Carolina Fisheries Commission developed a management plan to restore wild populations and produce greater numbers of terrapins for market sale. When wild terrapin numbers decreased substantially in later decades, North Carolina took more stringent measures, prohibiting the taking of terrapins between 1923 and 1938. The captive breeding program in North Carolina continued after the ban on terrapin fishing was installed and accounted for 250,000 hatchlings being released into the wild throughout the east coast of the United States. This practice saved many natural populations of diamondback terrapins from going extinct and probably mixed northern and Carolina subspecies (Hildebrand 1929; Coker 1951; Garber 1990b).
While many decimated diamondback terrapin populations have recovered from overharvesting in the early twentieth century, additional threats to their survival have emerged (Hurd et al. 1979). Current populations of terrapins suffer from numerous detrimental effects arising directly or indirectly from human activity: a reduction in nesting site habitats (Roosenburg 1991; Seigel 1993); drowning in crabpots (Roosenburg
1991; Roosenburg 1996; Roosenburg et al. 1997; Wood 1997; Roosenburg and Green
2000; Feinberg and Burke 2003; Hoyle and Gibbons 2000); overharvesting (Garber
1990a; Garber 1990b); motorboat mortality (Roosenburg 1991); automobile mortality
(Roosenburg 1991; Wood and Herlands 1997, Szerlag and McRobert 2006); and predation by subsidized species (Burger 1977; Auger and Giovannone 1979; Seigel 16
1980b; Lazell and Auger 1981; Roosenburg 1991; Seigel 1993; Roosenburg and Place
1995; Butler et al. 2004 ). Diamondback terrapins, like many other turtle species, have
suffered tremendously from these threats due to their life history characteristics
(Orenstein 2001).
Terrapins are estimated to have a long life span of possibly over 40 years
(Hildebrand 1932). A long life span is typically characterized by the late arrival of sexual
maturity as well as low replacement rates. Roosenburg (1991) found that wild females became sexually mature at 8-13 years of age and males at 4-7 years. In the same study it
was estimated that, due to high nest predation rates, a female terrapin needed to reproduce for three years at maximal reproductive output in order to replace herself as a
hatchling. Given these life history characteristics and the many new anthropogenic threats
to the survival of diamondback terrapins, major concern for the future sustainability of
wild populations is warranted.
The overall status of diamondback terrapin populations remains in decline or uncertain in many locations throughout the United States (Seigel and Gibbons 1995).
With data on basic reproductive life history characteristics and the new anthropogenic constraints on populations of diamondback terrapins, effective management for species survival may be possible (Congdon et al. 1983). Failure to conduct these studies may result in continued population declines and a need for federal protection status (Seigel and Gibbons 1995). To date, no in-depth study on the nesting ecology of diamondback terrapins and the influence of mammalian predation has been conducted within the commonwealth of Virginia (Seigel and Gibbons 1995).
Diamondback Terrapin Nesting Ecology-Natural History 17
Diamondback terrapins use coastal uplands along marshes, tidal creeks, and estuaries for nesting (Auger and Giovannone 1979; Roosenburg 1994; Ernst et al. 1999).
Terrapins typically construct their nests in open, sandy habitats (Goodwin 1994;
Roosenburg 1994) located above the mean high tide line (Roosenburg and Place 1995;
Roosenburg 1996). Female terrapins are iteroparous, nesting annually as well as producing more than one clutch within a nesting season (Roosenburg 1991).
Environmental sex determination (ESD) is a characteristic found in all diamondback terrapin populations: the temperature of the nesting substrate surrounding the egg determines the sex of the developing embryo (Burke 1993; Roosenburg and Place 1995;
Roosenburg 1996).
The physical characteristics of nesting female terrapins are similar throughout their range. Most mature adult females have an average carapace length between 6.5 in and 7 in (Burger 1989). The weight of adult nesting females ranges from a minimum nesting size of 1100 g to a maximum size of 1800 g (Roosenburg 1996).
Daily nesting excursions are thought to be colonial, with one or more females accessing the nesting beach at the same time (Auger and Giovannone 1979). Nesting females dig and conceal nest holes with their back legs (Burger 1977). The process of nesting may take as little as 15 minutes or as long as two hours, from the time the female leaves the water until she returns (Roosenburg 1994). Individuals that have laid 4 or more eggs and are disturbed during the process of nesting will complete their clutch and cover the nest. Females that that have started digging a nest hole and have laid fewer than 4 eggs prior to disturbance typically will abandon the act of nesting altogether (Burger
1977). 18
Diamondback terrapins choose many types of habitats for their nests. At a Florida site, diamondback terrapins chose to nest in man-made dikes (Seigel 1980a). Burger
(1977) found the majority of diamondback terrapin nests on sand dunes at one site in
New Jersey. Terrapins in New Jersey seem to prefer nesting in areas with little vegetation
(Burger and Montevecchi 1975). In New York, however, a large percentage of nests were found in dune/mixed grassland habitats and shrubland habitats (Feinburg and Burke
2003).
The timing and duration of the diamondback terrapin nesting season varies geographically (Zimmerman 1992). The nesting season for southern terrapin populations begins earlier and lasts longer than for northern populations. Florida terrapins begin nesting in April and continue through early July, a period of over 60 days (Seigel 1980a).
In more northern populations, terrapins begin nesting later in the spring and have a shorter nesting season measured at 51 days in a New York population (Feinberg and
Burke 2003) and 43 days in a New Jersey population (Burger 1977).
Geographic clinal variation also is observed in nest incubation period. Butler et al.
(2004) found an average incubation period of 69 days in a Florida population. In a New
York population the average incubation period was approximately 81 days (Cook 1989);
Burger (1977) found an average incubation period of 75 days in New Jersey; average incubation period for a population in Massachusetts (the northern limit of its range) was
108 days (Auger and Giovannone 1979). Due to long incubation periods and short nesting periods, northern terrapin hatchlings may overwinter within the nest cavity, resulting in the emergence of young early the following spring (Lazell and Auger 1981;
Auger and Giavannone 1979; Roosenburg 1994; Baker et al. 2006). 19
Clutch sizes for diamondback terrapin nests also vary over their geographic range.
The average clutch size in a Florida population was 6.7 eggs (Seigel 1980a), whereas terrapins in a New York population averaged 11.8 eggs per clutch (Giambanco 2002).
Average clutch sizes were similar in New Jersey (Burger 1977) and Connecticut populations (Aresco 1996), with 9.2 and 9.6 eggs, respectively.
The nesting activities of diamondback terrapins are thought to be influenced by physical environmental variables. For example, past studies have suggested that high tides are positively associated with nesting activity (Burger and Montevecchi 1975;
Auger and Giovannone 1979; Zimmerman 1992; Goodwin 1994; Aresco 1996; Feinburg and Burke 2003). From visual observations, female terrapins seem to prefer nesting on the higher tides and nesting activity increases during high tide events, presumably because high water allows more rapid access to nesting areas, decreasing the time a turtle is exposed on land to possible predation (Burger and Montevecchi 1975).
Temperature is another environmental factor thought to influence nesting activity.
Feinburg and Burke (2003) found nesting activity was influenced by daily temperatures, as terrapins preferred to nest during medium cloud cover and avoided more extreme temperatures associated with either full cloud cover or full sun. Further, terrapins at their site in New York did not nest when air temperatures exceeded 35°C, and in Florida, terrapins did not nest once the temperatures exceeded 36°C (Seigel 1980a).
Predation
Predation of diamondback terrapins by plants, birds, and invertebrates has had a significant negative impact on the reproductive success of diamondback terrapins.
Coastal plant species, such as American beachgrass (Ammophila breviligulata), have 20 destroyed entire clutches of eggs within a nest (Auger and Giovannone 1979;
Zimmerman 1992; Ners 2003). Rootlets of American beachgrass destroyed eggs and hatchlings in 25% of the nests monitored at one site in New Jersey (Lazell and Auger
1981) and were involved in 65% of all root-predated egg occurrences in New York
(Giambanco 2002). Saltwort ( Batis maritima) and seashore saltgrass (Distichilis spicata) are also possible terrapin nest predators (Butler et al. 2004).
Avian predation on diamondback terrapins has been recorded in many locations throughout its range. Watkins-Colwell and Black (1997) produced the first recorded account of a gull ( Larus sp.) attack on an adult male terrapin. Adult diamondback terrapin carcasses have been found in the nests of bald eagles ( Haliaeetus leucocephalus) in
Maryland and Virginia (Clark 1982). Burger (1977) found that 18% of all terrapin egg predation was due to avian predators. In that study laughing gulls ( Larus atricilla) consumed eggs while terrapins were nesting. American crows ( Corvus brachyrhynchos), fish crows (Corvus ossifragus), and boat-tailed grackles ( Quiscalus major) have been recorded unearthing concealed terrapin nests and eating the eggs (Burger 1977; Butler et al. 2004).
Invertebrates, such as crabs and ants, are documented diamondback terrapin nest predators. Soon after hatching, terrapins are very susceptible to predation by ghost crabs
(O. quadrata) (Arndt 1994; Butler et al. 2004; Barton 2005). Ghost crabs are attracted to hatching nests and consume or injure the hatchlings. Ant predation was first recorded by
Burger (1977) and documented in later years by Roosenburg (1992). The most recent report of possible ant predation was by Butler et al. (2004) who found two species-red 21 imported fire ant ( Solenopsis invicta) and pyramid ants ( Conomyrma spp)-that raided terrapin nests.
Mammalian predation has had the greatest negative impact on diamondback terrapin populations along the east coast of the United States (Burger 1977; Seigel 1980b;
Seigel 1993; Aresco 1996; Feinberg and Burke 2003; Butler et al. 2004; Draud et al.
2004). Many of these predators include native and non-native mammals whose populations have thrived with anthropogenic landscape changes (Garrott and White
1993). Mammalian predators of the diamondback terrapin include the red fox (V. fulva)
(Zimmerman 1992), striped skunk (M. mephitis) (Burger, 1977), otter ( Lutra canadensis)
(Roosenburg 1992), bobcat (Lynx rufus) (Zimmerman 1992), and Norway rat ( Rattus norvegicus) (Ners 2003; Draud et al. 2004). The most detrimental native mammalian predator to the diamondback terrapin is the raccoon (P. lotor) (Burger 1977; Seigel 1980;
Roosenburg 1992; Zimmerman 1992; Goodwin 1994; Aresco 1996; Feinberg and Burke
2003; Ners 2003; Butler et al. 2004).
Changes by humans to the environment have increased the contact between diamondback terrapins and mammalian predators such as raccoons (Seigel 1980b; Burger
1989; Seigel 1993; Feinburg and Burke 2003). Raccoon populations are large and especially active within the Chesapeake Bay coastal estuary system (Roosenburg and
Place 1995). Extirpation of natural predators like the red wolf ( Canis rufus) and gray wolf (Canis lupus), has allowed raccoon populations to flourish along the eastern coast of the United States (Webster et al. 1985; Garrettson et al. 1996). An increase in raccoon populations has placed greater demand on natural prey items such as the diamondback terrapin adults and eggs (Roosenburg 1991). Seigel (1980b) reported that 10% of all adult 22 female terrapins were depredated by raccoons at a Florida study site, while Feinberg and
Burke (2003) and Ners (2003) also recorded raccoon predation on adult terrapins in New
York. Diamondback terrapin nests, however, suffer the most from raccoon predation.
Raccoons had a significant impact on terrapin nests in New York, with only 5.2% of all unprotected nests surviving raccoon predation (Feinberg and Burke 2003). At the same site Giambanco (2003) found that 46% of all nests covered by predator excluders were still predated by raccoons.
Raccoon predation intensifies with an increase in vegetation and density of nests.
Burger (1977) found that terrapin nests located in an open area with little vegetation had less mammalian predation than nests in an area dominated by shrubs and trees. Feinberg and Burke (2003) found the most depredated nests located in dune/mixed grassland and shrubland habitats. How close a nest is to other nests may have some influence on raccoon predation. Roosenburg and Place (1994) and Burger (1977) suggested that nest predation rate may be correlated with nesting density and that nest success may be higher in areas that have fewer nests.
The effect of mammalian predation on populations of diamondback terrapins and their nests can be devastating. Raccoons are thought to have eliminated all nesting females at three study sites within Florida (Seigel 1993). Raccoons have had a severely negative effect on terrapin populations on Long Island, New York and Little Beach, New
Jersey (Burger 1989). The rate of mammalian predation at other sites has left many to wonder whether diamondback terrapin populations will decrease to dangerously low levels (Roosenburg 1991; Feinburg and Burke 2003; Butler et al. 2004; Draud et al.
2004). Indeed, virtually every study of terrapin nesting has documented at least 23
anecdotally the negative impact of mammalian predation on diamondback terrapins and their nests.
Objectives
In this project I chose to study the reproductive life history and nesting ecology of a population of diamondback terrapins in Virginia. My goal was to determine the impacts of raccoon predation on terrapin reproductive success on a complex of three small islands within the lower Chesapeake Bay estuary. The island closest to the mainland was inhabited year-round by raccoons, while the two remaining islands farther from the mainland were visited by foraging raccoons but were not inhabited by raccoons year- round. This complex allowed for between-island comparisons of terrapin nesting ecology, raccoon activities, and predation pressures.
The objectives of this study were to:
1. Determine the nesting ecology and reproductive life history characteristics (the number and location of nests and eggs, clutch size, duration of nesting season and nesting activity, information on nesting females, and incubation period) of a population of diamondback terrapins in Virginia.
2. Determine the relationship between terrapin nesting activity and abiotic environmental variables (air temperature, relative humidity, water temperature, aquatic salinity, and tidal cycle) and a biotic environmental variable (predator movements).
3. Determine the nest and egg predation rate.
4. Determine the spatial and temporal variation in mammalian predator movements throughout the terrapin nesting season. 24
5. Determine the relationship between nest success and the immediate habitat surrounding a nest.
6. Determine the relationship between nest success and nest site density.
This project will provide further details about the nesting ecology of diamondback terrapins, allowing scientists to gain a better understanding about the reproductive life history and predation threats to this species. Information gathered on the nesting ecology of terrapins from this site will supply site managers of the Goodwin Islands the means to make better conservation policies in regards to predator removal programs. CHAPTER II
METHODOLOGY
I conducted field research for this project on the Goodwin Islands, Virginia
(37°13’00”N; 76°23’37”W). The Goodwin Islands are a 127 hectare salt-marsh archipelago located on the southern tip of the mouth of the York River (Figure 1). The islands are geologically designated as Pleistocene-age Poquoson members of the Tabb
Formation and modem marsh (Johnson 1972; Mixon et al. 1989). The island complex consists of sandy beaches, scrub/shrub ridges, inter-tidal flats and marsh, and forested upland. Submerged aquatic vegetation (SAV) beds (314 hectares) surround much of the islands. The Goodwin Islands complex is part of the lower Chesapeake Bay tidal estuary system and the Chesapeake Bay National Estuarine Research Reserve of Virginia (CB-
NERRVA) (http://www.dnr.state.md.us/bay/cbnerr/).
The Goodwin Islands are separated into eastern and Western portions by a tidal creek. The creek spans roughly 25 ft across at its greatest width. At high tide the creek is approximately 4-5 ft deep. The eastern portion of the island complex is divided further by a small inlet at high tide. The eastern island was designated into a larger island and a smaller island. Therefore, in this study the islands were categorized and referred to as
“Western Island,” “Middle Island,” and “Eastern Island” (Figure 2).
Habitat and island area varies among the three islands (Tables 1, 2). The western portion of the island complex is dominated by forested uplands consisting of pine and oak
25 FIGURE 1 YORK RIVER WATERSHED MAP II
Map of the York River watershed and location of Goodwin Island complex, courtesy of Virginia Institute of Marine
Science (http://ccrm.vims.edu/cci/york/index.html). 26 FIGURE 2 MAP OF THE GOODWIN ISLAND COMPLEX ) N Z) < Z - 1 9 s ) O 00 CO T— T— CO z o c The Goodwin Islands complex, part of the Chesapeake Bay Virginia National Estuarine Research Reserve (CB-NERRVA) (2005 27 aerial photograph provided by the Virginia Institute of Marine Science Annual SAV Monitoring Program; Orth et al. 2005). 28 TABLE 1 HABITAT ASSESSMENT OF THE GOODWIN ISLANDS WI MI El Total Upland Forest 27.00 0.00 0.00 27.00 Intertidal Marsh 55.19 37.91 5.49 98.59 Potential Nesting Habitat 0.46 0.45 0.50 1.41 Total 82.65 38.36 5.99 127.00 Habitat assessment (ha) for the Goodwin Islands, 2005. WI=Westem Island, MI=Middle Island, EI=Eastern Island 29 TABLE 2 TOTAL AREA/LINEAR EDGE RATIO OF THE GOODWIN ISLANDS WI MI El Total Total Area 826,449 383,552 59,911 1,269,912 Linear Edge 6,747 11,069 2,103 19,919 Ratio (m) 122.49 34.65 28.49 186 Total area (m2) to linear edge (m) ratio for the Goodwin Islands, 2005. WI=Westem Island, MI=Middle Island, EI=Eastem Island 30 communities. Mammalian predators including raccoons reside in these upland habitats (Willy Reay, pers. comm.). The eastern portion of the island complex is dominated by tidal marsh habitat, consisting of salt marsh cordgrass ( Spartina alterniflora), salt grass (Distichlis spicata), and salt meadow hay (, Spartina patens). The upland borders of these wetland habitats are invaded by common reed (. Phragmites australis). Mammalian predators forage but do not reside within these tidal marshes. Terrapin nesting ecology and mammalian predator movement were studied within the Western, Eastern, and Little Islands. I conducted research during the diamondback terrapin nesting season from May 11-October 2, 2005 and performed a follow-up nest site survey on May 12, 2006. Transportation to the islands was by canoe. Nesting Ecology-Natural History Nests and Eggs I determined the total number of terrapin nests at this site through daily site visits to the Goodwin Islands. Nests were discovered through surveys of beach and upland areas of the Western, Middle, and Eastern Islands. The status of nests was classified as intact, predated, partially predated, or hatched/emerged. Egg shells were carefully reconstructed to obtain an egg count for each nest. The locations of intact nests were determined through observations of nesting females. A nesting female was left undisturbed during the nesting period and observed from a distance. Once the female had finished nesting the observer collected the female and marked the location of the nest with flagging. 31 Intact nest locations were also discovered through visual identification of terrapin crawls. Observers followed terrapin crawls from the water and identified cover-up patterns, triangular shaped areas of disturbance in the loose sand that signified the location of a nest (Burger 1977; Roosenburg 1992). Nests were dug up by hand, then eggs were counted and the nest cavity was covered with sand. Intact nests were flagged to follow future nest progress. Predated and partially predated terrapin nests were recognized by the appearance of broken white egg shells on the sand near the nest cavity. Observers found the exact locations of the nest cavity by digging within the vicinity of the broken shells. Once the location was determined the eggs were counted and placed back into the nest cavity and the hole was refilled with sand. With this procedure we avoided the possibility of recounting any of the predated nests. Partially predated nests were marked with flagging to follow the future progress of the nest. Hatched/emerged nests are nests that have escaped destruction by predators and have shown evidence of hatchling emergence from the nest cavity. Although the final outcome of the nest (whether hatchlings survived their emergence or were predated upon emergence) is unknown, these nests were considered as successful nests for this study. Hatched/emerged nests were located by the appearance of broken eggs shells on the sand. These nests were distinguished from predated or partially predated nests by the appearance of the egg shells and the nest cavity. The egg shells of hatched nests differed in color from the predated and partially predated eggs. Hatched/emerged egg shells are characterized by a chalky white exterior and a shiny silver interior (Butler et al. 2004). 32 Nest cavities often appeared as a mound of sand with broken egg shells embedded inside. The eggs were counted, placed back into the nest cavity, and recovered with sand. I recorded all nest locations in the field using Trimble GeoExplorer 3 Global Positioning System with accuracy within 1 m. The date each nest was discovered, time of day of discovery, status of nest (active, predated, partially predated, hatched/emerged), and egg count (if available) were recorded within the GPS file. Each nest was assigned an individual nest identification number. Once out of the field, I downloaded nest points from the GPS unit and placed them onto a 2005 orthorectified aerial photo of the Goodwin Islands and formatted them using ESRI ArcGIS software (version 9). Clutch Size I determined the mean clutch size from predated, partially predated, hatched and intact nests found on Western, Middle, and Eastern Island. All eggs were counted, returned to the nest cavity, and covered with sand. Covering the nest with sand after the egg count eliminated the possibility of recounting eggs or nests. Diamondback terrapin predated eggshells do not persist between years (Feinberg and Burke 2003), assuring that the eggs counted during this study were laid during the 2005 field season. Mean clutch sizes between predated and successful nests, as well from the Western, Middle, and Eastern Islands were compared using a one way ANOVA, setting the level of significance at p<0.05. Statistical analyses were performed using SPSS v. 13.0 (SPSS Inc. Chicago, Illinois). Duration of the Nesting Season and Nesting Activity I conducted daily surveys for the first signs of nesting activity (i.e. terrapin crawls, predated nests, and females on the beach) on May 11, 2005. Terrapin crawls and 33 females on the beach during the season indicated the location of possible nests and also helped identify active nesting areas. I continued daily surveys throughout the summer to determine the last signs of nesting activity (i.e. disappearance of terrapin crawls, predated nests, and females on the beach). Since diamondback terrapin hatchlings for some populations have been found to hibernate over the winter in nest cavities, there is the possibility that successful nests were underestimated for the season. In their northern range, diamondback terrapin hatchlings will enter hibemacula within the nest cavity and emerge in the spring (Baker et al. 2006). Whether diamondback terrapin hatchlings over winter in the nest cavity in Virginia is unknown. I conducted a one-day survey for overwintered, hatched/emerged successful nests in the spring of 2006, but no nests were detected. Since most terrapin nests are predated within 24-48 hours of laying (Roosenburg 1992; Goodwin 1994; Feinburg and Burke 2003; Butler et al. 2004), predated and partially predated nests were back-tracked 24 hrs from the date they were found to provide the best estimate of their lay date. These predated and partially predated nests were combined with intact nests to represent diamondback terrapin nesting activity. Predated, partially predated, and intact nests were plotted over time to view the seasonal pattern in nesting activity. Also, the number of partially predated nests was plotted separately from the nesting activity data to see whether partially predated nests correlated with terrapin activity. Female Terrapins I collected information on nesting females through daily site visits to the Western, Middle, and Eastern Islands. Females were captured when they were found crawling on 34 the beach or within the marsh grass. A female found in the act of nesting was left undisturbed until she had abandoned the nest area. The date, time of capture, and location of capture were recorded for each female. Females were weighed (±25g) using a spring scale. Straight line measurements of maximum carapacial width and length (±lmm) were taken using calipers (Roosenburg and Dunham 1997). Abnormalities in appearance, such as propeller scars on shells, barnacles, or missing limbs were noted. All females were palpated for eggs. I estimated the age of the terrapin by counting annular rings on the carapacial scutes (Lovich and Gibbins 1990; Gibbons et al. 2001). Terrapins were marked for future identification using a Log 2 carapace notching technique (Cagle 1939). Each individual female was given a unique number based on this system. The number was filed as notches into the marginal scutes of the carapace, allowing for identification of recaptured females (Auger and Giovannone 1979). Terrapins were released as soon as possible after capture to minimize stress on the animal. Incubation Period The incubation period of nests on the Goodwin Islands was determined by following the progress of nests that were found intact or partially predated. These nests were marked with flagging and checked on daily site visits throughout the season. Many intact and partially predated nests were completely predated shortly after oviposition, and could not be used in this analysis. For the remaining nests, the emergence of hatchlings or the appearance of entire clutches of hatched eggs was a sign that the incubation period had ended. The hatch date of the nests was recorded for each nest to determine the mean incubation time for all nests. 35 Influence of the Environment on Nesting Activity I completed principle components analysis (PCA) to address the research question whether observed differences in nesting activity (timing and intensity) were related to environmental variation. Diamondback terrapins are hypothesized to alter their nesting activity in response to environmental factors (Burger 1977; Cook 1989; Roosenburg 1991; Feinburg and Burke 2003). The time period investigated for this analysis was May 27-August 1, 2005, which includes time before and after the diamondback terrapin nesting season. I chose eight independent abiotic environmental variables and one biotic variable for this analysis. Since diamondback terrapins are reptiles and their activity is thought to be influenced by temperature, I included daily maximum and minimum air temperature and water temperature. Spring tides and increases in daily humidity have been suggested to spur terrapin nesting activity, so the mean height of daily morning high and low tides (heights measured above mean lower low water) and mean daily relative humidity were also included in this analysis. Although salinity has never been shown to influence nesting activity in any turtle species, I decided to include the mean daily aquatic salinity in this analysis. All abiotic environmental variable data were obtained from CBNERRVA Goodwin Islands weather archives, where data collected from meteorological/hydrological stations positioned just offshore the Goodwin Island complex were posted. Finally, I used daily predator movements, obtained from predator movement plots on the islands (see below), as the biotic independent variable for the PCA. The total number of predated, partially predated, and intact nests found on a daily basis represented female diamondback terrapin nesting activity and was the dependent 36 variable used in the correlation with each of the principal components. Since most terrapin nests are predated within 24 hours of oviposition (Feinburg and Burke 2003; Butler et al. 2004) and most raccoon movement occurs in the evening (Burt and Grossenheider 1964), the predated nests, partially predated nests, and predator movements were back-tracked 24 hours from the date they were found to provide a more accurate assessment of turtle and raccoon activity. Predator movement and nesting activity from all three islands were combined for each respective variable in the analysis. Both daily predator movements and nesting activity data contained some nonconsecutive days of observation. From these data, the total number of tracks and nests found on days after a period of no observation was averaged over the number of days of missed observation. This was done to avoid misleading observations of data, since the tracks and nests found on a day after a period of no observation would include tracks and nests from all previous days. Prior to running statistical tests, all nine environmental variables were plotted individually with the dependent variable of nesting activity to examine visually the relationships between variables. A QQ plot was completed to check for normality of all variables used in the principle components analysis. A correlation was run using PCI, PC2, and PC3 values obtained from the PCA of the independent variables and the dependent variable. PCI, PC2, and PC3 were then regressed against the dependent variable to determine whether a significant relationship existed between environmental variables and variation in nesting activity. All statistical analyses were performed using SPSS v. 13.0 (SPSS Inc. Chicago, Illinois). Influence of Raccoon Predation Predation Rate To determine the predation rate for the Western, Middle, and Eastern Islands the total number of predated, partially predated, and unsuccessful intact nests/eggs were divided by the overall number of nests/eggs (including intact successful nests/eggs and hatched/emerged successful nests/eggs) found on each island. These totals were multiplied by 100 to receive total percent predated. Identification of nest predators was determined by tracks within the nesting sites, bite patterns on egg shells, visual sightings of organisms within areas, and nest cavity disturbance patterns (Burger 1977; Aresco 1996; Feinburg and Burke 2003). Mammalian Predator Movements To determine patterns of mammalian predator movements within the Goodwin Island complex I constructed twelve predator movement plots. I compared mammalian predator movements for May, June, and July between and within the three islands. Six plots were randomly distributed on the Western Island and three plots were distributed on Middle and Eastern Islands. All plots had an area of 11.6 m and were constructed on unvegetated, loose, sandy soil. Four stakes were used to designate the location of a plot and its area. Plots were raked after a visit to eliminate all evidence of predator movement for that day. Mammalian predator movement plots were checked on a daily basis from May 27-August 1, 2005. The identities of the tracks were determined using a tracking guide (Burt and Grossenheider 1964). Each predator track that entered and left the plot was considered one predator movement. Predator movements were counted and recorded for 38 each plot on daily visits. Tracks counted on days following nonconsecutive sampling periods represented an accumulation of movements from the previous day(s) of no observation. Tracks counted on these days were averaged over the missing days of observation, to avoid misleading track totals on days following periods of no sampling. Movements from all plots were totaled for Western, Middle, and Eastern Islands and differences in predator movement were compared within islands for the months of May, June, and July (May 27-July 31, 2005) using a univariate ANOVA with post-hoc comparisons. Total predator movements and trends in predator movements over the nesting season were compared between islands using a repeated measures ANOVA with post-hoc comparisons. Statistical analyses were performed using SPSS v. 13.0 (SPSS Inc. Chicago, Illinois). Nesting Habitat and Location To determine the relationship between nest location and nest status, I analyzed the immediate nesting site habitat for every nest upon discovery. The location of every nest was determined to within 1 m with Trimble GPS equipment. Using a classification scheme similar to that described by Palmer and Cordes (1988), the immediate habitat surrounding a nest was classified as sandy beach (0-25% vegetation), mixed grassland/shrubland (25-75% vegetation), and wrack (nest location composed of dried/fresh marine vegetation). The number of predated, partially predated, intact, and hatched/emerged successful nests in each of these locations was determined. To determine whether nest success was related to the distance of its nearest neighbor, whether successful nests were located in low density nesting areas or off the beaten path of nesting activity, the spatial relationships among predated nests and 39 hatched/emerged successful nests was analyzed. Nests that were found partially predated were classified as predated nests for this analysis. Nests that were found as intact nests were eliminated from the analysis because the future outcome of some was uncertain. The nearest neighbor nest, the nest found at the smallest distance to the focal nest, was identified for every nest in the analysis. The nearest neighbor distance was calculated in meters using ESRI ArcGIS software (version 9). Statistical analyses using SPSS v. 13.0 (SPSS Inc. Chicago, Illinois) compared the mean distance between a nest and its nearest neighbor using a oneway ANOVA, p<0.05. A QQ plot was completed on all the variables to check for normality. Outliers found through initial analysis were removed from the data to increase normality. The ANOVA compared the average distances of focal nests to nearest nest neighbors. To determine whether a nest success was related to the number of nests surrounding a focal nest during the season, whether successful nests were more likely to be surrounded by fewer nests and found in low nest density areas, buffers were placed around nests using ESRI ArcGIS software (version 9) and the number of neighboring nests within a focal buffer was counted. A buffer with a radius of 2.59 m was placed around each nest. The small buffer size was chosen in order to analyze the immediate areas surrounding focal nests. The buffer radius used in this analysis was the mean minimum distance of a sample of focal nests to their nearest neighbor nests, determined during the nearest neighbor analysis (see above). Only nest sites that were visited on a daily basis throughout the nesting season were included in this analysis. As in the nearest neighbor analysis, partially predated nests were considered predated nests and intact nests were eliminated from this analysis. 40 Statistical analyses using SPSS v. 13.0 (SPSS Inc. Chicago, Illinois) compared the mean number of nests found within the buffers of predated and successful focal nests using one-way ANOVAs, p<0.05. A QQ plot was completed on all the variables to check for normality. Outliers found through initial analysis were removed from the data to increase normality. CHAPTER III RESULTS Nesting Ecology-Natural History Nests and Eggs A total of 69 sampling days were completed in the field during the 2005 nesting season. Diamondback terrapin nesting sites were found primarily along open beaches and mixed vegetation on the northern side of the Goodwin Islands (Figure 3). Terrapins nested at northern, western, and southern location on the Western Island. Nesting sites on the Middle Island were found at northern and southern locations only. The Eastern Island had nesting sites located throughout the island. All conspicuously open beach habitats above the high tide line on the islands were used by terrapins as nesting sites. A total of 1,362 nests were found on the Goodwin Islands over the course of the study period. Of the 1,362 nests, 1,091 nests were found to be depredated during the season. An additional 21 nests were partially depredated, i.e., the predator did not consume the entire clutch, and another 41 nests were found intact at the time of oviposition. Of these 62 nests, only 5 partially depredated and 5 intact nests successfully hatched during the season. Finally, 209 successful nests were also found on the islands after hatchling emergence. The number and types of nests varied across the island complex; however there was a west-east gradient in nest success, with Eastern Island having the greatest number of hatched/emerged nests (Table 3). 41 FIGURE 3 DIAMONDBACK TERRAPIN NESTING SITES ON THE GOODWIN ISLANDS Diamondback terrapin nest sites on the Goodwin Islands, 2005: Nesting sites (•). (2005 aerial photograph provided by the Virginia Institute of Marine Science Annual SAV Monitoring Program; 42 Orth et al. 2005). 43 TABLE 3 NEST TOTALS BY ISLAND WI MI El Total Depredated 302 570 219 1,091 Partially Depredated 0 11 10 21 Intact 9 8 24 41 Hatched/Emerged 4 59 146 209 Total 315 648 399 1,362 Total number of diamondback terrapin nests, categorized by status and island, on the Goodwin Islands during the 2005 nesting season. WI=Westem Island, MI=Middle Island, EI=Eastem Island 44 A total of 12,030 eggs were found in the 1,349 nests (the egg count for 13 intact nests was not determined). Of these, 9,975 eggs were from depredated nests, with 182 eggs from partially depredated nests, 234 eggs from intact nests, and 1,639 eggs from successful nests, i.e., nests that hatched during the season. The number of hatched/emerged eggs increased from west to east across the complex, with Eastern Island having the greatest number of successful eggs (Table 4). Clutch Size The overall mean ± S.E. clutch size for diamondback terrapin nests on the Goodwin Islands was 8.90 ± 0.07 eggs/nest (range = 1-17). Depredated nests on the Goodwin Islands had a mean clutch size of 9.16 ± 0.08 eggs/nest. The mean clutch size for partially depredated nests on the islands was 8.67 ± 0.72eggs/nest. The mean clutch size for intact nests and hatched/emerged successful nests was 8.52 ± 0.59 eggs/nest and 7.57 ± 0.17 eggs/nest, respectively. A comparison between depredated nest mean clutch size and successful nest mean clutch size was significantly different (ANOVA, p=0.001). Clutches from all nests were totaled and the mean ± S.E. clutch size varied significantly among islands (ANOVA, p<0.05) (Figure 4). A post-hoc comparison determined that average clutch size from the Eastern island was significantly smaller than both the Western and Middle Islands which were not significantly different (EI-WI: p=0.001; EI-MI: p=0.001), implying a possible selection pressure on reduced clutch size. Duration of the Nesting Season and Nesting Activity No nests or terrapin crawls were discovered during the month of May. The first nest was found on June 4th. Nests and female nest crawls continued throughout the months of June and July. The last sign of terrapin nesting activity was two nests and a 45 TABLE 4 EGG TOTALS BY ISLAND WI M I E l Total Depredated 2,777 5,270 1,928 9,975 Partially Depredated — 104 78 182 Intact 53 60 121 234 Successful 18 485 1,136 1,639 Total 2,848 5,919 3,263 12,030 Total number of diamondback terrapin eggs, categorized by status and island, on the Goodwin Islands during the 2005 nesting season. Exact numbers of eggs from 13 intact nests could not be determined and were not included in this table. WI=Westem Island, MI=Middle Island, EI=Eastem Island Among islands comparison of overall mean ± S.E. clutch size. Islands that are are that Islands size. clutch S.E. ± mean overall of comparison islands Among Overall Mean Clutch Size 10 2 - 4 6 8 - - - significantly different from one another have different letters. different have another one from different significantly COMPARISON OF CLUTCH SIZE BY ISLAND BY SIZE CLUTCH OF COMPARISON etr Middle Western FIGURE 4 FIGURE Island Eastern 46 47 nest crawl on the Eastern Island on July 28th. Based on this information the nesting season lasted approximately 55 days. The greatest numbers of nests were found during June and July (Table 5). The majority of depredated nests were found during these two months. Partially depredated and intact nests were only found during the June and July period. Hatched/emerged successful nests were not found in June or July. The first hatched/emerged successful nest was found on August 5th. The number of hatched/emerged successful nests decreased from August to October, with the most hatched/emerged nests observed during the month August. No emergence from potential overwintering nests was detected during a beach survey of Western and Middle Islands on May 12, 2006. Depredated (back-tracked 24 hrs), partially depredated(back-tracked 24 hrs), and intact nests found during daily site surveys represented diamondback terrapin nesting activity during 2005 (Figure 5). Six distinct peaks in nesting activity—two large peaks and four small peaks—were observed. The first significant peak in nesting activity occurred around June 8th. Three small peaks in activity were observed around June 14th, June 24th, and June 30th. The second significant peak in activity occurred around July 6th, followed by another smaller peak on July 11th. No nesting activity was observed prior to June 4th or after July 28th. Further, increases in nesting activity correlated with the number of partially depredated nests found (Figure 6). Partially depredated nests only occurred during peaks of nesting activity. Adult Female Terrapins A total of 59 female diamondback terrapins were captured during the research period. All but one of the females was found on land during nesting site surveys. The first 48 TABLE 5 NEST TOTALS BY MONTH May June July Aug Sep Oct May(2006) Total Depredated 0 537 550 3 1 0 0 1,091 Partially Depredated 0 12 9 0 0 0 0 21 Intact 0 33 8 0 0 0 0 41 Hatched/Emerged 0 0 0 159 47 3 0 209 Total 0 582 567 162 48 3 0 1,362 Total number of diamondback terrapin nests found on the Goodwin Islands during each month of the research period in 2005 and the follow-up survey in May 2006. Diamondback terrapin nesting activity (depredated, partially depredated, and intact nests) nests) intact and depredated, partially (depredated, activity nesting terrapin Diamondback Number of Nests 100 120 //056/20/2005 6/6/2005 on the Goodwin Islands from June 4-July 28, 2005. 28, 4-July June from Islands Goodwin the on NESTING ACTIVITY NESTING FIGURE 5 FIGURE Date 7/4/2005 7/18/2005 8/1/2005 49 depredated nests found daily during the 2005 nesting season on the Goodwin Islands. Goodwin the on season nesting 2005 the during daily found nests depredated Nesting Activity Diamondback terrapin nesting activity (depredated and intact nests) and partially partially and nests) intact and (depredated activity nesting terrapin Diamondback 100 120 20 - 40 0 - 60 0 - 80 - - NESTING ACTIVITY AND PARTIALLY DEPREDATED NESTS DEPREDATED PARTIALLY AND ACTIVITY NESTING //0562/057420 /820 8/1/2005 7/18/2005 7/4/2005 6/20/2005 6/6/2005 Nesting Activity Nesting Partially Depredated Nests Depredated Partially FIGURE 6 FIGURE Date Number of Partially Depredated Nests 50 51 nesting female was captured on June 5th and the last female was captured on July 18th. One female, female #59, was captured in a crab pot on July 30th. Only 2 females were recaptured during the research period. Female #24 was found 3 days after her initial capture. She was discovered within 50 m of her initial capture and was found to be gravid on both occasions. Female #43 was recaptured 15 days after her initial capture, also within 50 m of her initial capture. This female was not gravid at her initial capture but was gravid at recapture. Since females only emerge from the water to nest, the behavior of female #43 is indicative of multiple clutching. From annular rings on scutes, the mean ± S.E. age of captured females was found to be 7.70 ± 0.15 years. Since the shells of older turtles typically become worn as they age, annular rings and the age of the turtle become difficult to interpret. The shells of turtle #42 and #58 were so worn that their age could not be determined and they were eliminated from the analysis. The mean ± S.E. weight of a nesting female diamondback terrapin on the Goodwin Islands was 1,433.6 ± 31.4 g. The mean ± S.E. carapace length was 20.8±0.15 cm and carapace width was 15.7 ±0.12 cm. Scars on the carapace and plastron were found on 13 females. Three of the captured females were missing limbs. From the 60 total nesting female captures during the season, 43 were captured during the morning and 17 were captured in the afternoon/evening hours. There were 24 captures on the Western Island, 26 captures on the Middle Island, and 10 captures on the Eastern Island. At the time of capture, 19 nesting females were found to be gravid and 41 had already nested. 52 Incubation Period Incubation period for diamondback terrapin nests on the Goodwin Islands was determined using 10 nests for which lay dates were known (Table 6). Of the 10 nests, 5 were found initially as partially depredated nests and 5 were found initially as intact nests. The earliest lay date was June 6th and the latest lay date was July 15th. The earliest emergence date was August 5th and the latest emergence date was September 8th. Based on the emergence of the hatchlings from the nest cavity, the mean ± S.E. incubation period for all 10 nests was 59.0 ±1.9 days and ranged from 44 to 65 days. Influence of the Environment on Nesting Activity The principal component analysis (PCA) on environmental variables consisted of 8 abiotic variables (daily maximum and minimum air temperature, maximum and minimum water temperature, mean height of morning high and low tide, mean aquatic salinity and mean relative humidity) and 1 biotic variable (daily mammalian predator movements back-tracked 24 hrs). The dependent variable representing female terrapin nesting activity was composed of depredated (back-tracked 24 hrs), partially depredated (back-tracked 24 hrs), and intact nests found each day, measured independently from mammalian predator movements. The time period used for this analysis was from June 4- July 28, 2005. Before running the PCA, diamondback terrapin nesting activity was plotted over the time period. All nine independent variables were graphed individually with the dependent variable to examine their relationships within the designated time period. Although initial nesting activity during the season corresponded with increases in air and water temperatures (Figures 7-10), these variables did not track variation in nesting 53 TABLE 6 INCUBATION PERIOD Status When Emergence Nest # Island Found Lay Date Date Days Incubation 20 El Intact 6/5/2005 8/5/2005 61 24 El Intact 6/5/2005 8/5/2005 61 49 El Intact 6/6/2005 8/5/2005 60 59 El Intact 6/6/2005 8/10/2005 65 154 El PP 6/9/2005 8/8/2005 60 168 El PP 6/9/2005 8/10/2005 62 424 MIPP 6/22/2005 8/24/2005 63 847 MIPP 7/9/2005 9/8/2005 59 882 MI PP 7/10/2005 8/24/2005 44 1048 El Intact 7/15/2005 9/8/2005 55 Incubation period for 10 diamondback terrapin nests found on the Goodwin Islands found during the 2005 nesting season. Nesting activity (depredated, partially depredated, and intact nests) and minimum daily minimum and nests) intact and depredated, partially (depredated, activity Nesting Number of Nests 100 120 0 - 40 20 0 - 60 0 - 80 - - NESTING ACTIVITY AND MINIMUM AIR TEMPERATURE AIR MINIMUM AND ACTIVITY NESTING 6/6/2005 Minimum Air Temperature (°C) Temperature Air Minimum Nesting Activity Nesting /020 //0571/058/1/2005 7/18/2005 7/4/2005 6/20/2005 air temperature (°C). temperature air FIGURE 7 FIGURE Date Minimum Daily Air Temperature (°C) 54 Nesting activity (depredated, partially depredated, and intact nests) and maximum daily maximum and nests) intact and depredated, partially (depredated, activity Nesting Number of Nests 100 120 20 - 40 0 - 60 - - NESTING ACTIVITY AND MAXIMUM AIR TEMPERATURE AIR MAXIMUM AND ACTIVITY NESTING 6/6/2005 Maximum Air Temperature (°C) Temperature Air Maximum Activity Nesting /020 //0571/058/1/2005 7/18/2005 7/4/2005 6/20/2005 air temperature (°C). temperature air FIGURE 8 FIGURE Date 26 - 28 - 30 - 32 - 34 - 3680 - 38 - 40 - - Maximum Daily Air Temperature (°C) 55 Nesting activity (depredated, partially depredated, and intact nests) and minimum daily minimum and nests) intact and depredated, partially (depredated, activity Nesting Number of Nests 100 40 20 60 80 . . . « NESTING ACTIVITY AND MINIMUM WATER TEMPERATURE WATER MINIMUM AND ACTIVITY NESTING //05 /020 7420 71/05 8/1/2005 7/18/2005 7/4/2005 6/20/2005 6/6/2005 ...... Min. Water Temperature (°C) Temperature Water Min. Activity Nesting water temperature (°C). temperature water FIGURE 9 FIGURE Date —I- 18 32 r - - 24 - 26 - 28 - 30 - 20 22 Minimum Daily Water Temperature (°C) 56 57 FIGURE 10 NESTING ACTIVITY AND MAXIMUM WATER TEMPERATURE Nesting Activity Max. Water Temperature (°C) ...... ' ...... i ...... i ...... 6/6/2005 6/20/2005 7/4/2005 7/18/2005 8/1/2005 <5 Date Nesting activity (depredated, partially depredated, and intact nests) and maximum daily water temperature (°C). 58 throughout the season. Morning high and low tide levels varied according to lunar periodicity, but did not correspond to nesting activity (Figures 11 and 12). Aquatic salinity generally increased throughout the season (Figure 13) and relative humidity generally varied between 60-80% (Figure 14). Nesting activity was not clearly associated with variation in salinity or relative humidity. Finally, predator movements (measured as the number of mammal tracks in test plots) showed strong correlation with terrapin nesting activity (Figure 15). The PCA was run on all abiotic and biotic environmental variables. The PCA produced 9 significant components. Of the 9 components, the first three were extracted and accounted for 75.04% of the variance in the data. PCI explained 46.61% of the variation. PC2 explained 16.99% of the variation. PC3 explained 11.44% of the variation. The remaining 6 components had low loading factors (eigenvalues lower than 1) and will not be discussed further (Table 7). PCI loaded positively and most strongly for maximum and minimum air temperature, maximum and minimum water temperature and salinity (Table 8). The strongest loading factors for PC2 were high tide, which loaded negatively, and low tide, which loaded positively. Finally, PC3 loaded positively and most strongly for relative humidity. The one biotic environmental variable-raccoon tracks-did not load strongly into any of the three PCs. All three regression factors were graphed against the dependent variable of daily nesting activity to see the distribution of data points (Figures 16, 17, 18). A Pearson’s bivariate correlation was run between the two variables to test for significance. The Nesting activity (depredated, partially depredated, and intact nests) and daily morning daily and nests) intact and depredated, partially (depredated, activity Nesting Number of Nests 100 120 20 - 40 0 - 60 0 - 80 0 - 0 - - -| -| 6/6/2005 • ------NESTING ACTIVITY AND MORNING LOW TIDE LOW MORNING AND ACTIVITY NESTING — Nesting Activity Nesting — Morning Low Tide (ft) Tide Low Morning /020 //0571/058/1/2005 7/18/2005 7/4/2005 6/20/2005 low tide (ft). tide low FIGURE 11 FIGURE 0.4 - 0.6 Daily Morning Low Tide (ft) 59 Nesting activity (depredated, partially depredated, and intact nests) and daily morning daily and nests) intact and depredated, partially (depredated, activity Nesting Number of Nests 100 2 i i 120 0 - 20 - 40 0 - 60 0 - 80 0 - 0 - //05 /020 7420 71/05 8/1/2005 7/18/2005 7/4/2005 6/20/2005 6/6/2005 NESTING ACTIVITY AND MORNING HIGH TIDE HIGH MORNING AND ACTIVITY NESTING Morning High Tide (ft) Tide High Morning Activity Nesting high tide (ft). tide high IUE 12 FIGURE Date 3.4 r - - 2.4 - - - 3.0 - 3.2 - 2.0 2.2 2.6 2.8 Daily Morning High Tide (ft) 60 Nesting activity (depredated, partially depredated, and intact nests) and mean daily mean and nests) intact and depredated, partially (depredated, activity Nesting Number of Nests 120 100 20 0 - 40 0 - 60 0 - 80 0 - 0 - | - - -A — Nesting Activity • Nesting - — -A ------NESTING ACTIVITY AND AQUATIC SALINITY AQUATIC AND ACTIVITY NESTING Aquatic Salinity (psu) Salinity Aquatic 6/20/2005 aquatic salinity (psu). salinity aquatic IUE 13 FIGURE 7/4/2005 7/18/20056/6/2005 8/1/2005 Mean Daily Aquatic Salinity (psu) 61 Nesting activity (depredated, partially depredated, and intact nests) and mean daily mean and nests) intact and depredated, partially (depredated, activity Nesting Number of Nests 100 120 20 40 0 - 60 0 - 80 - 6/6/2005 -A — Nesting Activity Nesting — -A • ------NESTING ACTIVITY AND RELATIVE HUMIDITY RELATIVE AND ACTIVITY NESTING al H%) Daily RH(% /020 7/4/2005 6/20/2005 relative humidity (RH) (%). (RH) humidity relative l | 1 —i i r i 1——i 1 1 1 | 1 _l IUE 14 FIGURE Date 7/18/2005 8/1/2005 60 - 65 - 70 - 55 90 75 95 80 85 Mean Daily RH (%) 62 Nesting activity (depredated, partially depredated, and intact nests) and mammal tracks. mammal and nests) intact and depredated, partially (depredated, activity Nesting Number of Nests 100 20 - 40 0 - 60 0 - 80 - - //05 /020 7420 71/05 8/1/2005 7/18/2005 7/4/2005 6/20/2005 6/6/2005 NESTING ACTIVITY AND MAMMAL TRACKS MAMMAL AND ACTIVITY NESTING Mammal Tracks Mammal Nesting Activity Nesting IUE 15 FIGURE Date LA a A!! - 0 - - 40 - 60 - 80 - 20 100 Number of Mammal Tracks 63 64 TABLE 7 ENVIRONMENTAL VARIABLE PCA EXTRACTION VALUES Extraction Sums of Squared Loadings Total % of Cumulative ponent Eigenvalues Variance % 1 4.20 46.61 46.61 2 1.53 16.99 63.60 3 1.03 11.44 75.04 4 0.90 9.98 85.02 5 0.55 6.14 91.16 6 0.49 5.45 96.61 7 0.16 1.75 98.36 8 0.11 1.23 99.59 9 0.04 0.41 100.00 65 TABLE 8 LOADING FACTORS FOR ENVIRONMENTAL VARIABLES PCA Component 1 2 3 Mammal Tracks 0.387 0.234 0.142 Air Temp Min 0.802 0.095 0.364 Air Temp Max 0.858 0.113 0.014 Relative Humidity -0.272 -0.328 0.881 Aquatic Salinity 0.845 -0.111 -0.142 Water Temp Min 0.940 -0.020 0.069 Water Temp Max 0.955 -0.104 0.011 High Tide 0.274 -0.750 -0.276 Low Tide 0.073 0.872 -0.001 PCI ■2 3 0 2 3 1 1 LINEAR REGRESSION OF NESTING ACTIVITY AND PCI AND ACTIVITY NESTING OF REGRESSION LINEAR 0 20 Nesting Activity/Day Nesting 40 IUE 16 FIGURE 080 60 100 p = 0.27 = p r = -0.15 =r 120 140 66 PC2 ■2 3 0 2 3 1 1 LINEAR REGRESSION OF NESTING ACTIVITY AND PC2 AND ACTIVITY NESTING OF REGRESSION LINEAR 0 04 08 0 2 140 120 100 80 60 40 20 Nesting Activity/Day Nesting IUE 17 FIGURE p = 0.58 = p r = 0.08 =r 67 PC3 •4 ■2 ■3 0 2 1 1 LINEAR REGRESSION OF NESTING ACTIVITY AND PC3 AND ACTIVITY NESTING OF REGRESSION LINEAR 0 T • 20 Nesting Activity/Day Nesting 40 • • IUE 18 FIGURE 0120 60 80 100 p = 0.02 = p r = 0.33 =r 140 68 69 correlations on PCI and PC2 were found not to be significant. The correlation on PC3 was found to have a significant influence on nesting activity. The environmental data were analyzed further using a forward stepwise multiple linear regression, to test for significant correlations between the independent variables and nesting activity (Regression ANOVA, r2=0.30 F=10.99, p<0.001). Of the nine independent variables, only raccoon tracks (B=0.47, p<0.001) and aquatic salinity (6=- 0.41, p<0.001) were significant factors in the regression model. The remaining environmental factors were not significant and were not included in the model (Table 9). The model used for this analysis was: nesting activity = C + x(Mammal Tracks) + y(Salinity). Influence of Raccoon Predation Predation Rate Raccoons were the primary predators of terrapin nests at the Goodwin Island complex. This conclusion was based on tracks within the nesting areas, 3 visual sightings of adults, and characteristics of nest cavity disturbances. On two occasions nest sight disturbance indicated avian predation, possibly conducted by American crows {Corvus brachyrhynchos) within the area. Nest predation by invertebrates, vegetation, or other avian species was not detected at this site. Also, there was no indication of predation on adult terrapins at this site. Overall, nest predation within the Goodwin Island complex was 79.5% but varied in intensity across the islands from west to east. The highest nest predation was 95.8% on the Western Island. The lowest nest predation rate was 54.8% on the Eastern Island (Table 10). Similarly, total egg predation rate for the Goodwin Island complex was 70 TABLE 9 REGRESSION COEFFICIENTS Standardized Variable Beta Significance Mammal Tracks 0.466 0.000 Aquatic Salinity -0.407 0.001 Relative Humidity 0.221 0.066 Water Temp Max 0.267 0.187 Low Tide -0.107 0.366 High Tide -0.064 0.597 Water Temp Min 0.092 0.704 Air Temp Min 0.034 0.804 Air Temp Max 0.005 0.974 71 TABLE 10 PREDATION RATE FOR NESTS AND EGGS Island Nest Egg Western 95.8% 99.4% Middle 87.9% 91.6% Eastern 54.8% 63.0% Predation rate for nests and eggs found on the Goodwin Islands, 2005. 72 84.7% and also varied across the island in a gradient from west to east. Egg predation was highest on the Western Island (99.4%) and lowest on the Eastern Island at 63.0%. Mammalian Predator Movements Based on the presence of tracks in test plots, mammalian predator movements were recorded during 46 days of the 69 day sampling period. The first recording of movements occurred on May 27th. The last recording of movements occurred on August 1st. Raccoons were the primary mammalian predators on the Goodwin Islands and produced all of the tracks used in this analysis. A west-east gradient in total mammalian predator movement was observed (Table 11). More tracks were observed from each of the six plots from Western Island than from any of the three plots from the Middle Island. In turn, track numbers from the three plots on the Eastern Island were lower than all other plots. Over the 46 days of observation, the greatest amount of raccoon traffic occurred in a Western Island plot (419 tracks) and the least traffic occurred in an Eastern Island plot (24 tracks). Averaged across all plots and all months, the highest mean number of predator movements occurred on the Western Island and were lowest on the Eastern Island (Figure 19). The mean ± S.E. number of daily predator movements among islands was significantly different (ANOVA, p<0.05). The Western Island had the greatest average number of tracks, decreasing to least average number of tracks on the Eastern Island (post-hoc comparisons, p<0.05). The mean ± S.E. number of daily predator movements by month for each island also was significantly different (Figure 20). The fewest average number of tracks occurred in 73 TABLE 11 MAMMALIAN PREDATOR MOVEMENTS Island Plotl Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Total Western 419 316 244 263 203 385 1830 Middle 186 157 239 ——— 582 Eastern 53 24 41 —— — 118 Total mammalian predator movements recorded for 12 movement plots on the Goodwin Islands, 2005. Between islands comparison of the mean ± S.E. number of mammal predator movements. movements. predator mammal of number S.E. ± mean the of comparison islands Between Total number of observations is shown for each island. Islands that are significantly significantly are that Islands island. each for shown is observations of number Total Mean Number of Mammal Movements COMPARISON OF MAMMALIAN PREDATOR MOVEMENTS BY ISLAND BY MOVEMENTS PREDATOR MAMMALIAN OF COMPARISON 0 2 4 3 5 6 1 different from one another have different letters. different have another one from different Western IUE 19 FIGURE Island Middle Eastern 74 each island. Total number of observations is shown by month. For each island, months months island, each For month. by shown is observations of number Total island. each Comparison of the mean ± S.E. number of mammal predator movements by month for for month by movements predator mammal of number S.E. ± mean the of Comparison Mean Number of Mammal Movements FOR MONTH BY MOVEMENTS PREDATOR MAMMALIAN OF COMPARISON 4 2 3 5 6 1 that are significantly different from one another have different letters. different have another one from different significantly are that etr Mdl Eastern Middle Western b EACH ISLAND EACH FIGURE 20 FIGURE Island 75 76 May, increasing to greatest number of tracks in July on each island (ANOVA, p<0.05). The trend in predator activity for May, June, and July was also significantly different across islands (ANOVA, p <0.001). Nesting Habitat The immediate habitat surrounding a nest site was analyzed for all 1,362 nests (Figure 21). The majority of depredated nests were found in mixed grassland habitat. Hatched/emerged successful nests and partially depredated nests were found in similar numbers in open beach and mixed vegetation habitat. The majority of intact nests were originally found in open beach habitat. For both the Western and Middle Islands, most nests were found in mixed grassland, followed by open beach habitat (Figure 22). On the Eastern Island, more nests were found in open beach habitat than in mixed vegetation. The fewest nests were found in wrack habitat. Depredated nest locations were compared between Western, Middle, and Eastern Islands (Figure 23). For all three islands the pattern of nest location was the same, with most depredated nests located within mixed vegetation, followed by open beach, and wrack habitat. Finally, hatched/emerged successful nest locations were compared for all three islands (Figure 24). A gradient of increasing numbers of successful nests occurred from west to east within the island complex and within all types of habitat types. The majority of 59 successful nests were located within mixed vegetation on the Middle Island. On the Eastern Island the majority of 146 hatched/emerged successful nests were located within open sandy beach habitat. In contrast, only four successful nests were found on the Western Island. Nesting site habitat locations for all nests by status when found on the Goodwin Islands, Goodwin the on found when status by nests all for locations habitat site Nesting Total Number of Nests 200 400 300 500 600 100 0 NESTING SITE HABITAT FOR ALL NESTS BY NEST STATUS NEST BY NESTS ALL FOR HABITAT SITE NESTING erdtd ucsfl Intact Successful Depredated 1 FIGURE 21 FIGURE Status 2005 . rack W Mixed Vegetation Mixed Open Beach Open Depredated Partially 77 Nesting site habitat locations for all nests by islands found on the Goodwin Islands, 2005. Islands, Goodwin the on found islands by nests all for locations habitat site Nesting Total Number of Nests 0 - 400 200 0 - 300 100 - NESTING SITE HABITAT FOR ALL NESTS BY ISLAND BY NESTS ALL FOR HABITAT SITE NESTING 1 Western FIGURE 22 FIGURE Island Middle i Eastern xdVegetation V ixed M O pen Beach pen O rack W T 78 Nesting site habitat locations for depredated nests by island on the Goodwin Islands, Goodwin the on island by nests depredated for locations habitat site Nesting Number of Predated Nests 200 400 0 - 300 100 NESTING SITE HABITAT FOR DEPREDATED NESTS BY ISLAND BY NESTS DEPREDATED FOR HABITAT SITE NESTING - - 1 etr Middle Western n FIGURE 23 FIGURE Islands 2005. Mixed Vegetation Vegetation Mixed Wrack Open Beach Open Eastern 79 Nesting site habitat locations for hatched/emerged nests by island on the Goodwin the on island by nests hatched/emerged for locations habitat site Nesting Number of Hatched/Emerged Nests NESTING SITE HABITAT FOR HATCHED/EMERGED NESTS BY ISLAND BY NESTS HATCHED/EMERGED FOR HABITAT SITE NESTING 20 - 40 0 - 60 0 - 80 - Mixed Vegetation Vegetation Mixed Wrack Open Beach Beach Open Western -P- Islands, 2005. Islands, FIGURE 24 FIGURE 1 1 1 Islands Middle Eastern 80 81 Nearest Neighbor To determine whether nest success is related to the focal nest distance to its nearest neighbor, the mean minimum distance between a focal nest and its nearest nest neighbor was compared for each island within the complex. The mean minimum distance between all focal nests and their nearest neighbors was 2.59 m. A comparison of the mean ± S.E. minimum distance of depredated and successful focal nests to their nearest neighbors was not significantly different (p=0.991) (Figure 25). A comparison within islands of depredated and successful focal nests was significantly different for Eastern Island (p=0.002) but not for Western or Middle Island (p=0.940 and p=0.134, respectively) (Figure 26). Less crowding on the Eastern Island increased focal nest distances to their nearest neighbors. Buffers To determine whether nest success was related to the number of nests surrounding a focal nest, a buffer with a radius of 2.59 m (determined from the nearest neighbor analysis) was placed around each nest. For this analysis the number of nests within depredated and successful focal nest buffers was compared for each island within the complex. A comparison of the mean ± S.E. number of nests within buffers between depredated and successful focal nests was not significantly different (p=0.085) (Figure 27). A comparison within islands of depredated and successful focal nests was significantly different for the Middle Island (p=0.015) but not the Western or Eastern Island (p=0. I l l and p=0.112, respectively) (Figure 28). Nests on the Eastern Island are successful focal nest to its nearest neighbor. Focal nests that are significantly different significantly are that nests Focal neighbor. nearest its to nest focal successful Comparison of the overall mean ± S.E. minimum distance (m) from a depredated and and depredated a from (m) distance minimum S.E. ± mean overall the of Comparison Mean Distance to Nearest Neighbor (m) NESTS FOCAL FOR DISTANCES NEIGHBOR NEAREST OF COMPARISON 0.0 0.2 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 from one another have different letters. different have another one from erdtd Successful Depredated Focal Nest StatusNestFocal FIGURE 25 FIGURE 82 Within island comparison of the mean ± S.E. minimum distance (m) from a depredated depredated a from (m) distance minimum S.E. ± mean the of comparison island Within WITHIN ISLAND COMPARISON OF NEAREST NEIGHBOR DISTANCES FOR DISTANCES NEIGHBOR NEAREST OF COMPARISON ISLAND WITHIN and successful focal nest to its nearest neighbor. For each island, focal nests that are are that nests focal island, each For neighbor. nearest its to nest focal successful and Mean Distance to Nearest Neighbor (m) 0.5 2.0 2.5 1.0 1.5 significantly different from one another have different letters. different have another one from different significantly etr Mdl Eastern Middle Western FOCAL NESTS FOCAL FIGURE 26 FIGURE ] Successful Nests Successful ] Nests Depredated I Islands 83 successful focal nest buffers. Focal nests that are significantly different from one another one from different significantly are that nests Focal buffers. nest focal successful COMPARISON OF THE MEAN NUMBER OF NESTS OCCUPYING FOCAL NEST FOCAL OCCUPYING NESTS OF NUMBER MEAN THE OF COMPARISON Comparison of the overall mean ± S.E. number of nests occupying depredated and depredated occupying nests of number S.E. ± mean overall the of Comparison Mean Number of Nests/Buffer 0 2 4 3 5 1 Depredated 1018 have different letters. different have Focal Nest Status Nest Focal FIGURE 27 FIGURE BUFFERS Successful 214 84 Within island comparison of the mean ± S.E. number of nests occupying depredated and and depredated occupying nests of number S.E. ± mean the of comparison island Within successful focal nest buffers. For each island, focal nests that are significantly different significantly are that nests focal island, each For buffers. nest focal successful Mean Number of Nests/Buffer 10 12 WITHIN ISLAND COMPARISON OF THE MEAN NUMBER OF NESTS NESTS OF NUMBER MEAN THE OF COMPARISON ISLAND WITHIN - 4 - OCCUPYING FOCAL NEST BUFFERS NEST FOCAL OCCUPYING Western from one another have different letters. different have another one from FIGURE 28 FIGURE 478 Island Middle I 60 I I Successful Nests Successful I I 1 234 Eastern D epredated Nests epredated D 150 85 86 more spread out over the island area compared to nests on the Middle or Eastern Islands, yielding fewer nests within a focal nest buffer on this island. CHAPTER IV DISCUSSION This study documented the location, timing, and extent of nesting by diamondback terrapins on the Goodwin Islands complex, and the extent of nest predation by raccoons. Observed cross island gradients in predator activity and terrapin nest success suggested that mammalian predators inhabited and foraged most intensely on the Western Island, and moved eastward through the complex in search of food. The specific aspects of terrapin nesting ecology observed at Goodwin Islands are compared with results from prior studies conducted from other locations throughout the species range, however due to no prior research on diamondback terrapins in Virginia a comparison among local populations could not be made. Nesting Ecology-Natural History Nests and Eggs Unlike previous terrapin island nesting studies (Seigel 1980a; Feinburg and Burke 2003; Ners 2003), all sites determined feasible for nesting before the start of the nesting season were used by terrapins in the Goodwin Islands complex. Terrapins nested within all conspicuous open beaches and marginal mixed grasslands located above the high tide zone. Nesting did not occur in uplands inhabited by dense pine/oak forest or by stands of common reed (Phragmites australis). On the Western and Middle Island, females primarily nested along the northern shore, within narrow, marginal mixed grassland areas and small open beaches were located (Table 1; Figure 3). Although Western and Middle 87 Island were the two largest islands in the complex, their potential nesting areas were smaller than the Eastern Island. Nesting on the Eastern Island was extensive, with nests found throughout the island area. The distribution of nests on this island was due to large areas of suitable upland nesting habitat (beaches, small dunes, and grasslands) (Table 1). The total number of nests found at the Goodwin Islands (1,362) was among the highest recorded totals for any terrapin nesting location (Burger 1976; Burger 1977; Auger and Giovannone 1979; Roosenburg 1991; Roosenburg and Place 1995; Aresco 1996; Butler et al 2004). For example, this 127 ha island complex produced equivalent nest totals to those found within the Jamaica Bay Wildlife Refuge in New York (Feinburg and Burke 2003), a site composed of 3,700 ha of aquatic and upland habitat. More nests were produced within 1.41 ha total nesting area on the Goodwin Islands than at similar sized terrapin nesting areas at other terrapin study sites (Burger 1976; Burger 1977; Burger and Montevecchi 1975; Butler et al 2004). Nest totals per total island area and per nesting area make the Goodwin Islands one of the most densely used sites ever reported. The number and different types of nests (depredated, partially depredated, intact, and successful) varied across the island complex (Table 3). The Middle Island had the highest nesting activity on the Goodwin Island complex during 2005. Middle Island also had the highest number of depredated nests, whereas Eastern Island had the greatest nest success. Nest success increased in a west-east gradient. High nest predation and fewer successes on Middle Island and Western Island are consistent with the observation that mammalian predators resided on the Western Island, an island suitable for raccoon habitation, and moved off the island during foraging excursions to exploit food resources on the neighboring islands (Schneider et al 1971; Greenwood 1982). Eastern Island, as 89 the island farthest away from Western Island, had the greatest number of successful nests and fewest depredated nests because it was farthest from the predator source (the Western Island) and lacked a resident raccoon population. Hatched/emerged successful nests were not found in June or July, and first appeared in early August (Table 5). Considering the mean incubation period for nests at this site (59 days) most successful nests were laid in earlier June, during and just after the first peak in nesting activity (Figure 5). Fewer hatched/emerged nests were found in September and October, suggesting that nests deposited in July or early August had less of a chance of survival. A previous turtle nesting study has suggested that nests that are deposited earlier in the season have a greater chance of nest survival because of a lag time in the predator response to the new prey items (Robinson and Bider 1988). Once predators recognize the new prey items and their locations, predation intensity increases and nests have a greater chance of being consumed. The large numbers of hatched/emerged nests on the Goodwin Islands in early August were deposited in the beginning of the nesting season, so that nest success may depend on when a nest is created during the season. Partially depredated nests have been reported at only one other terrapin nesting site in New Jersey (Burger 1977). At the Goodwin Islands complex, partially depredated nests only occurred on the Middle and Eastern Islands, islands with higher nesting activity than the Western Island (Tables 3, 4). Partially depredated nests were found during periods of peak terrapin nesting activity (Figure 6), suggesting that raccoons may have been overwhelmed by the number of available nests and eggs, causing them not spend all of their energy searching for every egg within a nest cavity (Bell 1991). Further, 90 the distribution of partially depredated nests is consistent with the movement of raccoons from west to east within the island complex in search of food. Raccoons foraged most intensively on the Western Island and may have become overwhelmed by the number of nests on Middle and Eastern Island, leaving some nests partially depredated. Partially depredated nests may represent a decrease in the foraging efficiency of raccoons, suggesting that nesting during peaks in terrapin nesting activity may increase an individual eggs chance of survival (Robinson and Bider 1988). Unlike sea turtles, diamondback terrapins do not exhibit nesting “arribada,” in which massive groups of individuals emerge to nest over one or more short term periods within a nesting season (Pritchard 1969). Diamondback terrapins at this site (Figure 5) and at other locations (Burger and Montevecchi 1975; Burger 1977; Roosenburg 1994; Feinburg and Burke 2003; Butler et al 2004) are found to be colonial nesters, emerging singularly or coincidentally with a few other females during nesting throughout the nesting season (Caldwell et al 1959). The practice of arribada and group nesting may reduce nest predation, by overwhelming predators with more prey items and decreasing their searching efficiency (Caldwell et al 1959; Pritchard 1969; Robinson and Bider 1988; Tucker 1997). Clutch Size The mean clutch size at the Goodwin Islands (8.9±0.1 eggs/nest) fell within the expected geographic latitudinal variation seen in other terrapin nesting ecology studies (Table 12). Clutch size at this site was larger than clutches produced in more southern terrapin populations (Hildebrand 1932; Seigel 1980a; Zimmerman 1992) and smaller than average clutches produced in more northern populations (Burger 1977; Goodwin 1994; TABLE 12 A COMPARISON OF THE NESTING CHRACTERISTICS IN VARIOUS SUBSPECIES OF MALACLEMYS 0 £ 3 y *3 C/3 o> S3 53 u X « CJ3 w 3 £3 o> a WD 03CS !/5 o s C/3 X 1/3 -2 * 2 N s o e < < 03 O —H OS ' O OS 3 43 fc 3 3 3 3 (50 43 s- 3 3 O 3 3 O > 3 3 43 co of Pi 'a o 03 —H O OS f o 343 3 43 3 afc fa O O £ 3 n U ’(EL H < O O so o O n n o os n i CO § 43 SO r~- bo m in m t"- X 03 '<5 X CN O O CO o XI pp CN O O co O CO £ CN o c 3 43 3 3 S-, 3 B 3 3 o O < C/3 00 O 'O 3 to n 3 3 '3 £ £ 3 ^ H -2 e X * u 03 X 03 O CO CN 8 43 3 s-. 3 n SO o GO U O OS CN N NO O B 43 •— s 3 3 n n - r r-~ PP CN o o f o SO oo O 3 n GO O NO r~ NO in m CN in c-~ 43 CT Captive bred diamondback terrapins 91 92 Aresco 1996; Roosenburg 1992; Giambanco 2002; Feinburg and Burke 2003; Ners 2003). These geographic clutch size differences may be caused by the differences in abiotic mortalities of eggs and hatchlings found in these locations (Tinkle 1961; Tinkle et al. 1970; Pianka 1970). Due to variable and/or unpredictable climatic conditions, abiotic mortality is thought to be high in northern climates, resulting in selection for increased fecundity (larger clutch sizes) as well as smaller eggs. Due to constant and/or predictable climatic conditions, abiotic mortality in southern climates is thought to be less important. Selection in southern locations might favor improved offspring survivorship by increasing the size of eggs, while reducing fecundity (smaller clutch sizes). The Goodwin Islands are located midway between the northern and southern most range of the species along the Atlantic coast, and terrapins at this site produced an average clutch size comparable to this gradient. When compared to Western and Middle Island, Eastern Island produced a significantly smaller overall clutch size (Figure 4). The difference in clutch size among islands may be due to the relative number of hatched/emerged nests (Table 3), as the overall clutch size for hatched/emerged nests was significantly smaller than depredated nests. Because far more hatched/emerged nests were found on Eastern Island, the overall clutch size was smaller compared to the other two islands. The smaller clutch size observed for nests on the Eastern Island may be the result of smaller nesting females. Clutch size in terrapins (Seigel 1980a; Goodwin 1994; Roosenburg and Dunham 1997) and other turtle species (Congdon and Gibbons 1983; DePari, 1987; Nieuwolt-Dacanay 1997; Souza and Abe 2001; Iverson 2002) has been found to correlate positively with female body size. Small turtles almost never have large clutches, because clutch size is limited by a female’s size (Gibbons 1990). In addition, the growth rate of terrapins slows tremendously at maturity, with little change in body size occurring once an adult becomes sexually mature (Hildebrand 1932). Smaller body size in reproductively mature females suggests that individuals are reaching sexual maturity quickly at a younger age (Gibbons 1990). Since the Eastern Island is the only island in the complex to experience low predation and substantial nest success, these terrapins may be producing more individuals into the population that are smaller and younger at sexual maturity. Since female terrapins exhibit nest site fidelity (Auger and Giovannone 1979; Roosenburg 1991) and hatchling turtles have been found to return to natal beaches to nest (Meylan et al 1990), terrapins will consistently nest in the same location year after, no matter what the predation intensity or nest success rate is for that location. Therefore, this may result in short term selection pressure on the islands, with nesting female terrapins on the Eastern Island having greater nest success, yielding a population of females reaching sexual maturity at a younger age than nesting females on the other two islands. Alternatively, the smaller clutch size on the Eastern Island may simply be the result of a population with a large number of younger, smaller mature females. Since recruitment into the Western and Middle Island populations is low, the only mature females nesting on these islands might be older, larger terrapins. No matter what the explanation, there was no difference in female age among islands recorded at this site. More data is needed to support the suspected age structure differences among the Goodwin Islands. Clutch size may also vary with the number of sires associated with each clutch. A study on painted turtles (C. picta) found that multiple paternity positively correlated with 94 female clutch size, i.e., larger clutches indicated multiple paternity and smaller clutches indicated single paternity (Pearse et al 2002). If the number of males a female will mate with determines clutch size in terrapins, then the small clutches on Eastern Island may indicate that many females from this particular population may be mating with only one male. Since males are routinely caught in crab pots due to their small size (Roosenburg 1991; Roosenburg 1996; Roosenburg et al 1997; Wood 1997; Roosenburg and Green 2000; Feinberg and Burke 2003; Hoyle and Gibbons 2000), the lack of multiple male partners among Eastern Island females may be due to an increase in male deaths from local crab pot use around the island (pers. obsev.). On the other hand, the lack of male sires may be the result of fewer males being produced on this Eastern Island. Temperature of the nest cavity determines the sex of diamondback terrapin hatchlings. Males are produced in cooler conditions (26°C-27°C) than females (30°C-32°C) (Herlands et al. 2004). Most successful nests on the Eastern Island were produced in open beach habitat (Figure 24). These open, sunny habitats are usually warmer and have been found to produce more female terrapin hatchlings (Roosenburg and Place 1995). The Eastern Island, due to the abundance of open beach nesting habitat may be producing more females than males into the population, resulting in fewer mature male sires in the population. Regardless of the cause, average clutch size on the Eastern Island is significantly smaller than the average clutch size on Western and Middle Island. Since Eastern Island is the only island in the complex with substantial recruitment, the smaller overall clutch size may indicate smaller numbers of individuals entering into the population. 95 Duration of the Nesting Season and Nesting Activity The nesting season for the diamondback terrapin population at the Goodwin Islands began on June 4th and lasted until July 28th, a period of 55 days. The timing and duration of this nesting season fell within the expected latitudinal cline seen in other terrapin nesting ecology studies (Table 12). The nesting season at the Goodwin Islands began later and had a shorter duration than southern terrapin populations (Seigel 1980a; Zimmerman 1992; Butler et al 2004). The nesting season at the Goodwin Islands commenced at a similar time to northern populations but had a longer duration period (Burger 1977; Goodwin 1994; Feinberg and Burke 2003). The start of the 2005 nesting season for the Goodwin Island terrapins does not support previous studies, which suggest that based on geography and warm climate Virginia terrapins should have started nesting earlier than was witnessed during this season (Seigel 1980a). Nesting during 2005 may have been delayed due to water and air temperatures that remained low during late May. Temperature may stimulate the onset of nesting at this site, as has been shown at other terrapin nesting sites (Seigel 1980; Feinburg and Burke 2003). Air and water temperature did rise dramatically by 10°C just prior to terrapin nesting activity in 2005, suggesting a threshold temperature may exist to initiate nesting. In 2006, nesting at the Goodwin Islands commenced on May 25th (M. Wolak, pers. comm.), 10 days earlier than the start of the 2005 nesting season. Temperature data are not yet available for 2006 in order to make a comparative analysis between years; as a result it is not yet clear if air and water temperatures were warmer during May in 2006, reaching the temperature threshold sooner and instigating an earlier arrival of the nesting season. 96 While overwintering of hatchlings within the nest cavity has been shown to occur in northern terrapin populations (Lazell and Auger 1981; Auger and Giovannone 1979; Roosenburg 1994; Ners 2003; Draud et al. 2004; Baker et al 2006) and to possibly occur in one southern population (Butler et al. 2004), overwintering was not detected in this Virginian population. Nesting site surveys during early May 2006 revealed no signs of hatched/emerged nests on the Goodwin Islands, although the emergence of hatchlings had been observed during March in the Guinea marshes along the northern shore of the York River (C. Van Dover, pers. obs.). The lack of evidence of hatchling emergence during the May 2006 survey may be due to timing. A previous diamondback terrapin nesting study from New York found that overwintering hatchlings emerged in the early spring during the month of April (Draud et al. 2004). Since my survey was completed in late spring, I may have missed the emergence of overwintering hatchlings from their nests. Either way, if terrapin hatchlings due overwinter within the nest cavity in Virginia, the state’s southern location and climate may severely reduce its occurrence, a result found in other southern terrapin nesting studies (Roosenburg 1994; Butler et al 2004). Further research on hatchling terrapins at the Goodwin Islands is needed to determine the occurrence and frequency of nest cavity overwintering at this site. Female Terrapins The Goodwin Island female terrapins are younger in age but similar in size to terrapins studied from natural populations. The average age of nesting females on the Goodwin Islands was 7.7±0.15 years. Nesting female terrapins at this site were younger than the females in Maryland (Roosenburg 1991; Roosenburg and Dunham 1997) but similar in age to captive nesting females (Hildebrand 1932). Average weight 97 (1433.6±31.4 g) and size (CL: 20.8±0.15 cm and CW: 15.7±0.12 cm) for Goodwin Island females was also similar to terrapins from previous studies (Roosenburg 1991; Aresco 1996; Roosenburg 1996; Szerlag and McRobert 2006), and extremely similar in size to a nesting female terrapin found by Reid (1955) along the York River in Virginia (CL: 19 cm and CW: 15.3 cm). Accurate comparisons of terrapin size were difficult to interpret because most studies measured plastron dimensions, not carapace dimensions as in this study. A larger sample size of nesting female terrapins from the Goodwin Islands may provide greater accuracy on the age and size at maturity for females from this population, and whether these metrics vary across the island complex. Female terrapins may be affected by the motor boat traffic surrounding the Goodwin Islands. Of 59 nesting females captured during 2005, 27% had injuries to their limbs or carapace. Also, during the May 2006 nest site survey, a deceased female with a propeller scar through the carapace was found within the wrack line on the northern shore of Western Island. Roosenburg (1991) and Gibbons et al (2001) have both found injured female terrapins with propeller injuries to their shells. Roosenburg suggests that females are especially vulnerable to the propellers of fast moving boats because turtles spend a large portion of their time in channels during the spring and summer near the surface of water column where it is warm. A large channel with significant boat traffic during the warmer months separates the mainland from the Western Island, creating an opportunity for injuries and mortality to female terrapins at the site. Incubation Period The incubation period for terrapin nests on the Goodwin Islands was 59±2 days (Table 6). Length of incubation period tends to vary on a geographical cline, with 98 northern populations exhibiting longer periods of incubation than southern populations of terrapins (Table 12). The incubation period of terrapin nests at the Goodwin Islands was shorter than the incubation periods studied in northern populations (Burger 1977; Auger and Giovannone 1979; Cook 1989; Goodwin 1994). However, the average incubation period for nests on the Goodwin Islands was not longer than the emergence period found in a previous study on a southern population (Seigel 1980a; Butler et al 2004), where a wide range in days until emergence was observed with some nests exhibiting shorter incubation periods than the average period found on the Goodwin Islands. Although the methods used to obtain the incubation periods on the Goodwin Islands were fairly precise, a larger sample size would improve the overall estimate. Many intact and partially depredated nests (12 and 7, respectively) were difficult to locate as time passed (due the removal of nest site markers by environmental or human disturbance), and thus the final outcomes of these nests could not be determined. Future nesting studies at this site should use different nest marking techniques to identify and track individual nests throughout the season. Diamondback terrapins have been shown to produce two clutches per season (Auger and Giavannone 1979; Goodwin 1994; Feinberg and Burke 2003) and may produce as many as three a season in natural (Seigel, 1980a; Roosenburg 1991) or more in captive settings (Hildebrand 1932). At the Goodwin Islands, female #43 was recaptured on the beach 15 days after her initial capture. Female #43 was not gravid at her initial capture but was gravid at recapture, suggesting that this female completed at least two beach excursions and two clutches during the 2005 season. The 15 day period corresponded with the intemesting intervals found by Roosenburg (1992), Goodwin 99 (1994), and Feinburg and Burke (2003) (15, 16, and 17.5 days, respectively). Although no information exists on terrapin nesting in Virginia, terrapin studies from northern portions of the Chesapeake Bay estuary in Maryland have verified double clutching there (Roosenburg 1991). Based on this information, Goodwin Island female terrapins are producing at least two clutches per season, as has been suggested as typical for most populations (Russell Burke, pers. comm.). A total of 1,362 depredated, intact, partially depredated, and hatched/emereged successful nests were found during the 2005 nesting season. Assuming that on average females double clutch, then an estimate of 681 female terrapins reproduced on the Goodwin Islands in 2005. If the sex ratio of sexually mature individuals for this population is 1:1, then approximately 1,200 terrapins live, feed, and nest in and around the complex. Compared to previous terrapin population studies (Hurd et al 1979; Roosenburg 1991; Gibbons et al 2001; Feinburg and Burke 2003), the adult terrapin population size on the Goodwin Islands marsh complex is fairly small. Influence of the Environment on Nesting Activity The PCA only found daily relative humidity to have a significant positive correlation with diamondback terrapin nesting activity (Figures 16, 17, 18), providing no support for previous studies that suggested the influence of other environmental variables (Seigel 1980a; Burger and Montevecchi 1975; Auger and Giovannone 1979; Goodwin 1994; Aresco 1996; Feinburg and Burke 2003). Other turtle nesting studies have suggested that relative humidity may play an important role in predicting female turtle nesting activity (Goode, 1965; Lopez-Castro et al, 2004). Increases in relative humidity may influence daily nesting activity because relative humidity is thought to decrease 100 turtle water loss when they are making overland movements, like those made during nest attempts (Finkler, 2001). It has also been suggested that relative humidity plays an important role in influencing the seasonal changes in the gonadal cycle of turtles (Singh, 1974; Singh, 1977). Daily relative humidity may play an important role in terrapin nesting activity at the Goodwin Islands. Unlike the PCA, the step-wise regression found significant relationships between other environmental variables and nesting activity (Table 9). First, a significant positive relationship was found between predator tracks and nesting activity (Figure 15). Tracking plots at the Goodwin Islands captured raccoon nocturnal foraging excursions along the water’s edge prior to the start of the terrapin nesting season. Since terrapin nests on the Goodwin Islands lie within the home range of foraging raccoons, they are susceptible to predation. As generalist and opportunistic predators, raccoons will alter their nocturnal foraging movements to take advantage of new resources and resource patches that provide assured prey items (Schneider 1971; Greenwood 1982; Kolbe and Janzen 2002), such as terrapin eggs. Raccoon activity at the Goodwin Islands responded to the onset of terrapin nesting with a short lag, followed by an increase in activity in close correlation with nesting throughout the season. Roosenburg (1992) also reported a lag time in raccoon response to the start of the terrapin nesting season. As more nests were being laid by terrapins on the Goodwin Islands there was more activity by raccoons to dig up those nests. The nesting activity on the Goodwin Islands was promoting raccoon nest predation. Mean daily aquatic salinity negatively correlated with nesting activity, although this relationship was not visually obvious (Figure 13). Aquatic salinity has been found to influence both the physiology and behavior of marine and brackish water turtle species. 101 Sea turtles and diamondback terrapins reside exclusively within highly saline environments (Schmidt-Nielsen and Fange 1958). To survive such high levels of salinity and maintain homeostasis, both groups of turtles have a low integumentary permeability to salts and a lachrymal gland located in the orbit of the eye to excrete a concentrated salt solution. When placed in a highly saline environment terrapins alter their behavior by ceasing food consumption (Davenport and Ward 1993), dinking low salinity water from rainfall (Davenport and Macedo 1990), and basking at the waters surface (Davenport and Magill 1996). These physical and behavioral responses in terrapins indicate that osmoregulation and changes in aquatic saline play a key role in regulating terrapin activity. However, no previous turtle or terrapin nesting studies have found aquatic salinity to correlate with nesting activity. B. Influence of Raccoon Predation Predation Rate Overall terrapin nest predation was 79.5% and overall egg predation was 84.7% at the Goodwin Islands. Overall raccoon predation at the Goodwin Islands was not as high as 92.2%, reported by Feinburg and Burke (2003), but was higher than rates from prior terrapin nesting ecology studies (Burger 1976; Burger 1977; Roosenburg 1992; Roosenburg 1994; Giambanco 2002; Ners 2003; Butler et al 2004). Raccoons were the primary predators of terrapin nests on the Goodwin Islands, and this was true at all other terrapin (Burger 1976; Burger 1977; Seigel 1980b; Roosenburg 1991; Roosenburg 1992; Seigel 1993; Zimmerman 1992; Goodwin 1994; Aresco 1996; Feinberg and Burke 2003; Butler et al 2004) and many other turtle (Stancyl et al 1980; Snow 1982; Robinson and Bider 1988; Herman et al 1995; Ratnaswamy et al 1997; Marchand and Litvaitis 2004; 102 Garmestani and Percival 2005) nesting sites. Unlike other terrapin studies (Seigel 1980b; Seigel 1993; Feinberg and Burke 2003; Ners 2003), no adult terrapins were killed by raccoons at the Goodwin Islands. American crows (C. brachyrhynchos) were possible nest predators in two single predation events but unlike other terrapin nesting sites (Burger 1977; Butler et al 2004), were not a major factor in predation of terrapin nests at the Goodwin Islands. As shown in previous studies, raccoons at this site seemed to follow the edge habitat along the aquatic environments in search of food (Dijak and Thompson 2000; Erwin et al 2001; Moore 2002; Marchand and Litvaitis 2004) and, although it is unclear, may have either used the fresh odors left by females during nesting to detect the location of nests (Congdon et al 1983) or discovered nests because of nesting site sand disturbance (Burke et al 2005; D.M. Nally, pers. comm.). Although raccoon predation has been recorded as a threat to marine turtle nests in Virginia (Musick 1979), this is the first study to document the impacts of raccoon predation on diamondback terrapin nests within the Commonwealth. A gradient in predation rate occurred from west to east across the complex, with the most predation and lowest nest/egg success occurring on the Western Island (Table 10). This result further supports the idea that raccoons move eastward within the island complex in search of food. Mammalian predators were able to exploit almost all terrapin nests located on the Middle Island and Western Island, thereby opening windows for successful laying on Eastern Island. Based on these results, the Eastern Island is the only island in the complex characterized by high recruitment into the population. Since female terrapins are thought to exhibit nest site fidelity (Auger and Giovannone 1979; Roosenburg 1991; Goodwin 1994), creek site fidelity (Gibbons et al 2001), and hatchling 103 turtles have been found to return to natal beaches to nest (Meylan et al 1990), the Eastern Island may be the only area within the complex to successfully recruit hatchling terrapins into the population. Mammalian Predator Movements The overall number of tracks and average tracks per plot decreased from west to east across the Goodwin Islands complex, with the lowest predator activity found on the Eastern Island, further supporting the notion that raccoons inhabit the Western Island and forage terrapin nesting areas for food from west to east (Tables 11; Figures 19, 20). The number of raccoon tracks increased throughout the nesting season. Fewer tracks occurred during the month of May, and track numbers increased from June to July for all three islands. Raccoons responded to the onset of the terrapin nesting season by increasing their movements. Predator movement was not significantly different between May and June on the Middle and Eastern Island, suggesting a possible lag time in raccoon response to terrapin nesting (Roosenburg 1992). In response to the new food source - terrapin eggs- over time, raccoons expanded their searching distance later in the nesting season, venturing farther eastward towards the Middle and Eastern Islands, areas containing more terrapin nests. The increase in movements and expansion of searching distance in response to terrapin nesting is indicative of raccoon foraging behavior within wetlands, in which raccoons, as generalist predators (Garrott and White 1993), will forage along designated paths, altering their foraging movements in response to new food sources as they become available (Schneider et al 1971; Greenwood 1982). 104 Nesting Habitat Overall and within island comparisons found that the most nests were located in mixed grassland (Figure 21), a similar finding to that of previous terrapin nesting ecology studies (Burger 1977; Roosenburg 1992; Zimmerman 1992; Feinburg and Burke 2003). However, this result may represent the type of nesting site habitat that is most available to nesting females at the Goodwin Islands. The overall complex was found to be composed of fewer open, sandy beaches with a greater number of mixed grassland areas. This may also explain why more nests were found in mixed grassland areas on the Western and Middle Islands but not on Eastern Island (Figure 22). Eastern Island had more areas of open sandy beach and therefore contained more nests in this type of habitat. Wrack always contained the fewest nests and this is the first terrapin study to identify nesting within this type of habitat (Seigel 1980a; Burger 1977; Burger and Montevecchi 1975; Feinburg and Burke 2003; Butler et al 2004). No matter what the habitat type, the greatest number of depredated nests was found on the Middle Island and the number of hatched/emerged successful nests increased from west to east. Based on these data, nest success on the Goodwin Islands does not seem to be determined by habitat type but rather by raccoon predation intensity. Buffers and Nearest Neighbor Although some turtle nesting studies have suggested that an increase in nest density will increase nest predation (Burger 1977; Robinson and Bider 1988; Roosenburg and Place 1995; Kolbe and Janzen 2002b; Spencer 2002; Doody et al 2003; Marchand and Litvaitis 2004), neither depredated nor successful nests were located more frequently away from areas of high nesting density during the 2005 season on the Goodwin Islands 105 (Figures 25, 26, 27, 28). These results coincide with other turtle nesting studies that suggest that nest predation may be independent of nest density (Burke et al 1998; Doody et al 2003). Conversely, high nest densities among all nests were common throughout the island complex, due to the small total area of nesting site habitats. Previous turtle nesting studies have found nest clumping to occur in areas devoid of large, open sandy beach and comprised primarily of small, sandy patches interspersed by mixed grassland (Roosenburg 1994; Robinson and Bider 1988; Kolbe and Janzen 2002b), a common landscape characteristic of the Goodwin Islands. Clustering of nests on Eastern Island was less dense (fewest number of nests within buffers and largest nearest neighbor distance) than on any other island in the complex. Open, broad, nesting areas were found on the Eastern Island, whereas nesting habitat was narrower on the Western and Middle Island. Small patches or narrow bands of suitable nesting sites along the water’s edge will cause nests to clump and create an ecological trap, placing nests into the path of foraging raccoons and increasing their chances of being depredated (Vickery et al 1992; Gates and Gysel 1978; Kolbe and Janzen 2002b). Once a predator finds one nest within a small patch of clumped nests, it will likely find all the nests within a patch because they are so close together (Fretwell and Lucas 1970; Sutherland 1996). This explains the high nest predation on the Western and Middle Island, islands dominated by small, narrow patches of suitable nesting habitat along the water’s edge. Summary The Goodwin Islands complex is an important habitat for nesting diamondback terrapins within the lower Chesapeake Bay estuary. The density of nests within suitable nesting habitat area was among the highest for any previously studied terrapin nesting 106 location. As was found in other terrapin nesting studies, most nests were located in mixed grassland areas. Clutch size, incubation period, and length of nesting season were all found to fall within the expected latitudinal gradient of the species range for these traits. This study found evidence that female terrapins at this site exhibit nest site fidelity but did not find evidence of hatchling overwintering within the nest cavity. The PCA and the multiple regression analysis found a significant correlation between nesting activity and three environmental variables (relative humidity, mammalian predator movements, and aquatic salinity). Nesting activity was constant once the nesting season began, providing no evidence of an arribada and supporting the notion that diamondback terrapin are colonial nesters. Raccoons were overwhelmingly the primary terrapin nest predators. Mammalian predator movements correlated with the increase in terrapin nesting and the deposition of eggs, increasing significantly between months and islands as the nesting season progressed. High nest densities among nests was common throughout the island complex, but clustering of nests on Eastern Island was less dense than on any other island due to the presence of large, open nesting areas. In contrast, the small, narrow nest patches may have created ecological traps where predation rates were very high. From west to east across the Goodwin Islands complex, significant gradients in hatching success (total successful nests and hatched eggs), predation intensity (nests and eggs), and predator movements were observed. Raccoons inhabit the Western Island closest to the mainland and move west to east in search of food. The impact of raccoon predation on terrapin nests decreased from west to east. Since the majority of nests were found on the Middle and Eastern Islands, raccoons may move in a pattern described by the ideal free distribution theory, foraging and moving from patch to patch within their 107 environment while looking for prey items (Fretwell and Lucas 1970; Greenwood 1982; Sutherland 1996)). An individual will ultimately settle on the most ideal patches, the ones that will yield the highest rewards (the most prey items), for the least amount of energy expenditure (Bell 1991). Raccoons inhabited Western Island and consumed the most prey items, while expended the least amount of energy during nightly foraging. Seeking more rewards, raccoons moved eastward toward the Middle and Eastern Islands, islands with more prey items. Based on the ideal free distribution, it is understandable why raccoons would venture eastward, off of the island they inhabit, to obtain the large patches of available terrapin eggs. The gradient in predation intensity and nest success across the Goodwin Island complex may be deleterious to terrapin population survival. Predation intensity is lowest and nest success is highest on the Eastern Island, where females produce smaller clutch sizes and perhaps a population of females that reach sexual maturity faster than populations from other parts of the island complex. If female terrapins at the Goodwin Islands exhibit nest site fidelity, then females nesting on Western and Middle Island will consistently recruit fewer individuals into the population due to the extreme predation intensity, resulting in an aging group of reproductive individuals from those locations. The effects of predation, nest site fidelity, and nest success may be producing short term selection pressure on the islands, with nesting female terrapins on the Eastern Island having greater nest success, higher recruitment, and producing a younger population than nesting females on the other two islands. If this is the case, then the number of females nesting on the Western and Middle Islands should begin to decline over the years, decreasing the number of nests at these sites and forcing more raccoons to venture 108 eastward in search of food. In time, the success of nests on the Eastern Island may improve as more females return to nest, but eventually the east will be the only nesting site and raccoons will compensate by moving more extensively to the Eastern Island for the only abundant food source. Raccoon populations must be considered in the conservation management of the Goodwin Island diamondback terrapin population. Removal of raccoons from nesting sites has been shown to increase turtle nest success and recruitment (Christiansen and Gallaway 1984; Garmestani and Percival 2005). However, predator removal does not always yield the desired management result. For example, if a predator removal program only removes part of the raccoon population, predation pressures will not decrease (Ratnaswamy et al 1997; Prange et al 2003) and raccoons will easily re-populate from elsewhere. Further, the removal of mammalian predators may have drastic impacts on the natural environment (Stancyk et al 1980; Ratnaswamy and Warren 1998), and Barton (2005) found that the removal of raccoons within turtle nesting habitats eliminated raccoon nest predation but increased predation by ghost crabs. Since ghost crabs comprised a large percentage of the raccoon diet, crab populations flourished with the removal of raccoons and quickly replaced raccoons as the primary terrapin nest predators. A frequently used nonlethal alternative to predator removal is the construction of predator excluder screens over nests to decrease predation (Ratnaswamy et al 1997). Heppell’s (1998) life-history model on multiple species suggests that spending time and money on ways to decrease predation and increase juvenile recruitment into turtle populations may not be the answer at all, and that in order to stabilize declining populations, conservation efforts should focus on ways to reduce adult survival. More studies are needed in this 109 area to determine whether raccoons are overabundant on the Goodwin Islands (Garrott and White 1993) and what their true impact is on terrapin populations before predator removal programs are initiated within this island complex. Finally, this project has shown that terrapin eggs are extremely vulnerable to predation by subsidized predators. But does an increase in nest predation and a decrease in recruitment really put terrapin populations at risk? Low chances of survival of turtle eggs and juveniles are often associated with predation and other environmental factors (Plummer 1976; Burger 1977). Survivorship of individuals increases with age, with lowest survival occurring from 0-1 year of age (Gibbons, 1990; Iverson, 1991), and size (Haskell et al. 1996). With this in mind, increasing the number of predators at a location will certainly decrease the survivorship of eggs and young turtles for that site, and this has led researchers to conduct management techniques, such as “capping” nests (Giambanco 2003), predator removal (Garmestani and Percival 2005), and head starting hatchlings (Heppell et al. 1996), to increase juvenile survival. But does juvenile survival matter to the viability of a turtle population? Population models of sea turtles have found that population persistence is dependent on the survival of juveniles and egg/hatchlings (Mazaris et al. 2005). Also, a population model performed for the terrapin population surrounding the Goodwin Islands found that the youngest age classes were contributing the most to the overall growth/decline of the population (M. Wolak, pers. comm.). Considering the low recruitment at the Goodwin Islands due to raccoon predation and the importance of young terrapins to the population, the future survival of this population may be in jeopardy. 110 BIBLIOGRAPHY Aresco MJ. 1996. Malaclemys terrapin terrapin (northern diamondback terrapin): reproduction and nest predation. Herpetological Review 27: 77. Arndt RG. 1994. Predation on hatchling diamondback terrapin, Malaclemys terrapin, by ghost crab, Ocypode quadrata. Florida Scientist 45(3-4): 215-217. Auger PJ, Giovannone P. 1979. On the fringe of existence: diamondback terrapins at Sandy Neck. Cape Naturalist 8: 44-58. Baker PJ, Costanzo JP, Herlands R, Wood RC, Lee RE. 2006. Inoculative freezing promotes winter survival in hatchling diamondback terrapin, Malaclemys terrapin. Canadian Journal of Zoology 84: 116-124. Barton BT. 2005. Cascading effects of predator removal on the ecology of sea turtle nesting beaches [dissertation]. Orlando (FL): University of Central Florida. 48 p. Bell WJ. 1991. Searching Behaviour: The behavioural ecology of finding resources. London (UK): Chapman and Hall. 358 p. Biegler R. 1966. A survey of recent longevity records for reptiles and amphibians in zoos. International Zoo Yearbook 6: 487-493. Brown L, Macdonald DW. 1995. Predation on green turtle Chelonia my das nests by wild canids at Akyatan Beach, Turkey. Biological Conservation 71: 55-60. Burger J. 1976. Behavior of hatchling diamondback terrapins (. Malaclemys terrapin) in the field. Copeia 4: 742-748. Burger J. 1977. Determinants of hatchling success in diamondback terrapin, Malaclemys terrapin. American Midland Naturalist 97(2): 444-464. Burger J. 1989. Diamondback terrapin project. Plastron Papers 19(1): 35-40. Burger J, Garber SD. 1995. Risk assessment, life history strategies, and turtles: could declines be prevented or predicted? Journal of Toxicology and Environmental Health 46: 483-500. Burger J, Montevecchi WA. 1975. Nest site selection in the terrapin Malaclemys terrapin. Copeia: 113-119. Burke RL. 1993. Adaptive value of sex determination mode and hatchling sex ratio bias in reptiles. Copeia 3: 854-859. I l l Burke RL, Schneider CM, Dolinger MT. 2005. Cues used by raccoons to find turtle nests: Effects of flags, human scent, and diamond-backed terrapin sign. Journal of Herpetology 39 (2): 312-315. Burke VJ, Rathbun SL, Bodie JR, Gibbons JW. 1998. Effect of density on predation rate for turtle nests in a complex landscape. Oikos 83 (1): 3-11. Burt WH. 1964. A field guide to the mammals. Cambridge (MA): The Riverside Press. 284 p. Butler JA, Broadhurst C, Green M, Mullin Z. 2004. Nesting, nest predation, and hatchling emergence of the Carolina diamondback terrapin, Malaclemys terrapin centrata, in Northeastern Florida. American Midland Naturalist 152: 145-155. Cagle FR. 1939. A system of marking turtles for future identification. Copeia: 170-173. Cagle FR. 1950. The life history of the slider turtle, Pseudemys scripta troostii. Ecological Monographs 20: 31-54. Caldwell DK. et al. 1959. Multiple and group nesting by the Atlantic loggerhead turtle. Bulletin of the Florida State Museum. Biological Sciences 4: 308. Carr AF. et al. 1966. The ecology and migrations of sea turtles, 6. The hawksbill turtle in the Caribbean Sea. American Museum Novitates. 2248: 1-29. Carr AF. 1967. So excellent a fishe; a natural history of sea turtles. Garden City (NY): Natural History Press. 248 p. Chesapeake Bay National Estuarine Reserve in Virginia (CB-NERRVA). 2001. Goodwin Islands home page, Christiansen JL, Gallaway BJ. 1984. Raccoon removal, nesting success, and hatchling emergence in Iowa turtles with special reference to Kinosternon flavescens (Kinostemidae). Southwestern Naturalist 29: 343-348. Clark WS. 1982. Turtles as a food source of nesting bald eagles in the Chesapeake Bay Region. Journal of Field Ornithology 53: 49-51. Coker RE. 1951. The diamond-back terrapin in North Carolina. In (ed) H. F. Taylor, Survey of Marine Fisheries of North Carolina. Chapel Hill (NC): University of North Carolina Press. 219-230 pp. Congdon JD, Gibbons JW. 1983. Relationships of reproductive characteristics to body size in Pseudemys scripta. Herpetologica 39(2): 147-151. 112 Congdon JD, Tinkle DW, Breitenbach GL, van Loben Sels RC. 1983. Nesting ecology and hatching success in the turtle Emydoidea blandingi. Herpetologica 39(4): 417-429. Congdon JD, Tinkle DW, Rosen PC. 1981. Nesting frequency and success: implications for the demography of painted turtles. Ecology 62(6): 1426-1432. Cook B. 1989. A natural history of the diamondback terrapin. Underwater Naturalist 18(1): 25-31. Cote SD, Rooney TP, Tremblay JP, Dussault C, Waller DM. 2004. Ecological impacts of deer overabundance. Annual Review of Ecology, Evolution and Systematics 35: 113-47. Davenport J, Macedo E-A. 1990. Behavioural osmotic control in the euryhaline diamondback terrapin Malaclemys terrapin: responses to low salinity and rainfall. Journal of Zoology 220: 487-496. Davenport J, Magill SH. 1996. Thermoregulation or osmotic control? Some preliminary observations on the function of emersion in the diamondback terrapin Malaclemys terrapin (Latreille). Herpetological Journal 6: 26-29. Davenport J, Ward JF. 1993. The effects of salinity and temperature on appetite in the diamondback terrapin Malaclemys terrapin (Latreille). Herpetological Journal 3: 95-98. DePari JA, Linck MH, Graham TE. 1987. Clutch size of the Blanding's turtle, Emydoidea blandingi, in Massachusetts. Canadian field-naturalist 101(3): 440-442. Dijak WD, Thompson III FR. 2000. Landscape and edge effects on the distribution of mammalian predators in Missouri. Journal of Wildlife Management 64(1): 209- 216. Dodd CK. 1988. Synopsis of the biological data on the loggerhead sea turtle Caretta caretta. U.S. Fish & Wildlife Service Biological Report 88(14): 110. Doody JS, Sims RA, Georges A. 2003. Gregarious behavior of nesting turtles (■Carettochelys insculpta) does not reduce nest predation risk. Copeia 4: 894-898. Draud M. 2004. Predation on hatchling and juvenile diamondback terrapins (.Malaclemys terrapin) by the Norway rat (Rattus norvegicus). Journal of Herpetology 38(3): 467-470. 113 Dunham AE, Grant BW, Overall KL. 1989. Interfaces between biophysical ecology and the population ecology of terrestrial vertebrate ectotherms. Physiological Zoology 62(2): 335-355. Ernst CH, Barbour RW. 1972. Turtles of the United States. Lexington (KY): The University Press of Kentucky. 347 p. Ernst CH, Barbour RW. 1989. Turtles of the World. Washington (D.C.): Smithsonian Institute Press. 313 p. Ernst CH, Bury, RB. 1982. Malaclemys, M. terrapin. Catalogue of American Amphibians and Reptiles 299: 1-4. Ernst CH, Lovich JE, Barbour RW. 1994. Turtles of the United States and Canada. Lexington (KY): The University Press of Kentucky. 578 p. Erwin RM, Truitt BR, Jimenez JE. 2001. Ground-nesting waterbirds and mammalian carnivores in the Virginia barrier island region: Running out of options. Journal of Coastal Research 17 (2): 292-296. Finkler MS. 2001. Rates of water loss and estimates of survival time under varying humidity in juvenile snapping turtles ( Chelydra serpentina). Copeia 2: 521-525. Fisher TR, Hagy JD III, Boynton WR, Williams MR. 2006. Cultural eutrophication in the Choptank and Patuxent estuaries of Chesapeake Bay. Limnology and Oceanography 51(1): 435-447. Feinberg JA, Burke RL. 2003. Nesting ecology and predation of diamondback terrapins, Malaclemys terrapin, at Gateway National Recreation Area, New York. Journal of Herpetology 37(3): 517-526. Frazer NB, Ehrhart LM. 1985. Preliminary growth models for green, Chelonia mydas, and loggerhead, Caretta caretta, turtles in the wild. Copeia 1: 73-79. Fretwell SD, Lucas HL. 1970. On territorial behavior and other factors influencing habitat distribution in birds. Acta Biotheoretica 19: 16-37. Fritzell EK. 1978. Habitat use of prairie raccoons during the waterfowl breeding season. Journal of Wildlife Management 42: 118-127. Garber SD. 1990a. Diamondback terrapin. The Urban Naturalist 40: 33-36. Garber SD. 1990b. The ups and downs of the diamondback terrapin. The Conservationist (NYSDEC): May-June: 44-47. 114 Garber SD, Burger J. 1995. A 20-yr study documenting the relationship between turtle decline and human recreation. Ecological Applications 5(4): 1151-1162. Garmestani AS, Percival HF. 2005. Raccoon removal reduces sea turtle nest depredation in the Ten Thousand Islands of Florida. Southeastern Naturalist 4(3): 469-472. Garrott RA, White PJ. 1993. Overabundance: An issue for conservation biologists? Conservation Biology 7(4): 946-949. Garrettson PR, Rohwer FC, Zimmer JM, Mense BJ, Dion N. 1996. Effects of mammalian predator removal on waterfowl and non-game birds in North Dakota. Transactions of the North American Wildlife and Natural Resources Conference: 94-101. Gates JE, Gysel LW. 1978. Avian nest dispersion and fledging success in field-forest ecotones. Ecology 59: 871-883. Gehrt SD, Fritzell EK. 1997. Sexual differences in home ranges of raccoons. Journal of Mammalogy 78: 921-931. Gehrt SD, Fritzell EK. 1998. Resource distribution, female home range dispersion and male spatial interactions: group structure in a solitary carnivore. Animal Behaviour 55(5): 1211-1227. Giambanco MR. 2002. Comparison of viability rates, hatchling survivorship, and sex ratio of laboratory- and field-incubated nests of the estuarine, Emydid turtle Malaclemys terrapin [dissertation]. Hempstead (NY): Hofstra University. 83 p. Gibbons JW. 1983. Reproductive characteristics and ecology of the mud turtle Kinostemon subrubrum (Lacepede). Herpetologica 39: 254-271. Gibbons JW. 1987. Why do turtles live so long? Bioscience 37(4): 262-269. Gibbons JW. 1990. Life history and ecology of the slider turtle. Washington (D.C.): Smithsonian Institutional Press. 368 p. Gibbons JW, Lovich JE, Tucker AD, FitzSimmons NN, Greene JL. 2001. Demographic and ecological factors affecting conservation and management of the diamondback terrapin (.Malaclemys terrapin) in South Carolina. Chelonian Conservation and Biology 4(1): 66-74. Gibbons JW, Scott DE, Ryan TJ, Buhlmann KA, Tuberville TD, Metts BS, Greene JL, Mills T, Leiden Y, Poppy S, Winne CT. 2000. The global decline of reptiles, deja vu amphibians. BioScience 50(8): 653-666. 115 Golder WD, Lee D, Whilden M. 2000. Terrapin conservation efforts. Turtle and tortoise newsletter; the newsletter of conservationists and biologists 2: 7. Goode J. 1965. Nesting behavior of freshwater tortoises in Victoria. Victorian Naturalist 82: 218-222. Goodwin CC. 1994. Aspects of nesting ecology of the diamondback terrapin (.Malaclemys terrapin) in Rhode Island [dissertation]. Kingston (RI): University of Rhode Island. 84 p. Greenwood RJ. 1982. Nocturnal activity and foraging of prairie raccoons ( Procyon lotor) in North Dakota. American Midland Naturalist 107(2): 238-243. Hamilton AM, Freedman AH, Franz R. 2002. Effects of deer feeders, habitat and sensory cues on predation rates on artificial turtle nests. American Midland Naturalist 147: 123-134. Harless M, Morlock H. 1979. Turtles: perspectives and research. New York (NY): A Wiley-Interscience Publication. 695 p. Haskell A, Graham TE, Griffin CR, Hestbeck JB. 1996. Size related survival of headstarted redbelly turtles {Pseudemys rubriventris) in Massachusetts. Journal of Herpetology 30 (4): 524-527. Heppell SS, Crowder LB, Crouse DT. 1996. Models to evaluate headstarting as a management tool for long-lived turtles. Ecological Applications 6 (2): 556-565. Heppell SS. 1998. Application of life-history theory and population model analysis turtle conservation. Copeia 2: 367-375. Herlands R, Wood R, Pritchard J, Clapp H, Le Furge N. 2004. Diamondback terrapin, {Malaclemys terrapin) head-starting project in southern New Jersey. In: Swarth C, Roosenburg WM, Kiviat E, editors. Conservation and Ecology of Turtles of the Mid-Atlantic Region: A Symposium. Salt Lake City (UT): Biblomania. p 13-18. Herman TB, Power TD, Eaton BR. 1995. Status of Blanding’s turtles, Emydoidea- Blandingii, in Nova-Scotia, Canada. Canadian Field-Naturalist 109 (2): 182-191. Hildebrand SF. 1929. Review of the experiments on artificial culture of the diamond- back terrapin. Bulletin of the United States Bureau of Fisheries 45: 25-70. Hildebrand SF. 1932. Growth of diamondback terrapins: size attained, sex ratio, and longevity. Zoologica 9(15): 551-563. 116 Houlahan JE, Findlay CS. 2003. The effects of adjacent land use on wetland amphibian species richness and community composition. Canadian Journal of Aquatic Science 60: 1078-1094. Hoyle ME, Gibbons JW. 2000. Use of a marked population of diamondback terrapins (.Malaclemys terrapin) to determine impacts of recreational crab pots. Chelonian Conservation and Biology 3(4): 735-737. Hurd LE, Smedes GW, Dean TA. 1979. An ecological study of a natural population of diamondback terrapins in a Delaware salt marsh. Estuaries 2(1): 28-33. Iverson JB. 1991. Patterns of survivorship in turtles (order Testudines). Canadian Journal of Zoology 69: 385-391. Iverson JB. 2002. Reproduction in female razorback musk turtles ( Stemotherus carinatus: Kinostemidae). Southwestern Naturalist 47(2): 215-224. Johnson DH, Sargeant AB, Greenwood RJ. 1989. Importance of individual species of predators on nesting success of ducks in the Canadian Prairie Pothole Region. Canadian Journal of Zoology 67(2): 291-297. Johnson GH. 1972. Geology of the Yorktown, Poquoson West, and Poquoson East Quadrangles, Virginia. Report of Investigations 30, Virginia Division of Mineral Resources. 57 p. Kamler JF, Gipson PS. 2003. Space and habitat use by male and female raccoons, Procyon lotor, in Kansas. Canadian Field-Naturalist 117(2): 218-223. Kennish MJ. 1997. Practical Handbook of Estuarine and Marine Pollution. Boca Raton (FL): CRC Press. 524 p. Kennish MJ. 2002. Environmental threats and environmental future of estuaries. Environmental Conservation 29(1): 78-107. Kinneary JJ. 1992. The effect of water salinity on growth and oxygen consumption of snapping turtle ( Chelydra serpentina) hatchlings from an estuarine habitat. Journal of Herpetology 26(4): 461-467. Klemens MW. 2000. Turtle Conservation. Washington (D.C.): Smithsonian Institution Press. 334 p. Kolbe JJ, Janzen FJ. 2002a. Spatial and temporal dynamics of turtle nest predation: edge effects. Oikos 99: 538-544. Kolbe JJ, Janzen FJ. 2002b. Impact of nest-site selection on nest success and nest temperature in natural and disturbed habitats. Ecology 83(1): 269-281. 117 Kostecke RM, Smith LM, Hands HM. 2004. Vegetation response to cattail management at Cheyenne Bottoms, Kansas. Journal of Aquatic Plant Management 42(1): 39- 45. Lande R. 1998. Anthropogenic, ecological and genetic factors in extinction and conservation. Researches on Population Ecology 40(3): 259-269. Lawler AR, Musick JA. 1972. Sand beach hibernation by a northern diamondback terrapin, Malaclemys terrapin terrapin (Schoepff). Copeia 2: 389-390. Lazell JD, Auger PJ. 1981. Predation on diamondback terrapin (Malaclemys terrapin) eggs by dunegrass (Ammophila breviligulata). Copeia 3: 723-724. Linck MH, DePari JA, Butler BO, Graham TE. 1989. Nesting behavior of the turtle, Emydoidea blandingi, in Massachusetts. Journal of Herpetology 23: 442- 444. Lopez-Castro MC, Carmona R, Nichols WJ. 2004. Nesting characteristics of the olive ridley turtle (Lepidochelys olivacea) in Cabo Pulmo, southern Baja California. Marine Biology 145(4): 811-820. Lovich JE, Gibbons JW. 1990. Age at maturity influences adult sex ratio in the turtle Malaclemys terrapin. Oikos 59(1): 126-134. Mahmoud IY. 1967. Courtship behavior and sexual maturity in four species of Kinostemid turtles. Copeia: 314-319. Marchand MN, Litvaitis JA. 2004. Effects of habitat features and landscape composition on the population structure of a common aquatic turtle in a region undergoing rapid development. Conservation Biology 18(3): 758-767. Mazaris AD, Fiksen O, Matsinos YG. 2005. Using an individual-based model for assessment of sea turtle population viability. Population Ecology 47: 179-191. Meylan AB, Bowen BW, Avise JC. 1990. A genetic test of the natal homing versus social facilitation models for green turtle migration. Science 248(4956): 724-727. Mixon RB, Berquist CR, Jr., Newell, WL, Johnson GH, Powars DS, Schindler JS, Rader EK. 1989. Geologic map and generalized cross sections of the coastal plain and adjacent parts of the piedmont, Virginia. Miscellaneous Investigation Series, Map 1-2033, U.S. Geological Survey. Moll EO. 1973. Latitudinal and intersubspecific variation in reproduction of the painted turtle, Chrysemys picta. Herpetolica 29: 307-318. 118 Moll D, Moll EO. 2004. The ecology, exploration, and conservation of river turtles. New York (NY): Oxford University Press. 393 p. Moore PG. 2002. Mammals in intertidal and maritime ecosystems: interactions, impacts and implications. Oceanography and Marine Biology: An Annual Review 40: 491-608. Musick JA. 1979. The marine turtles of Virginia, families Chelonidae and Dermochelyidae, with notes on identification and natural history. Gloucester Point (VA): Educational Series, Virginia Institute of Marine Science. 23 p. Ners SE. 2003. Distribution and predation of diamondback terrapin nests at six upland islands of Jamaica Bay and Sandy Hook Unit, Gateway National Recreation Area [dissertation]. Hempstead (NY): Hofstra University. 104 p. Nieuwolt-Dacanay PM. 1997. Reproduction in the western box turtle, Terrapene ornata luteola. Copeia 4: 819-826. Obbard ME, Brooks RJ. 1980. Nesting migrations of the snapping turtle (Chelydra serpentina). Herpetologica 36(2): 158-162. Orenstein R. 2001. Turtles, tortoises, and terrapins: survivors in armor. Buffalo (NY): Firefly Books. 308 p. Orth RJ, Wilcox DJ, Nagey LS, Owens AL, Whiting JR, Kenne AK. 2005. 2004 Distribution of Submerged Aquatic Vegetation in the Chesapeake Bay and Coastal Bays. VIMS Special Scientific Report Number 146. Final report to U.S. EPA, Chesapeake Bay Program, Annapolis (MD): Grant No.CB973013-01-0. Available from: http://www.vims.edu/bio/sav/sav04 Palmer WM, Cordes CL. 1988. Habitat suitability index models: diamondback terrapin (nesting)—Atlantic Coast. U.S. Fish Wildlife Service Biological Report 82(10.151): 1-18. Pandian TJ, Vemberg FJ. 1987. Animal Energetics. San Diego (CA): Academic Press. 631 p. Paton PW. 1994. The effects of edge on avian nest success: how strong is the evidence? Conservation Biology 8(1): 17-26. Pearse DI, Janzen FI, Avise, JI. 2002. Multiple paternity, sperm storage, and reproductive success of female and male painted turtles (Chrysemys picta) in nature. Behavioral Ecology and Sociobiology 51(2): 164-171. Pellerin S, Huot J, Cote SD. 2006. Long-term effects of deer browsing and trampling on the vegetation of peatlands. Biological Conservation 128(3): 316-326. 119 Pianka ER. 1970. On r- and K-Selection. The American Naturalist 104(940): 592-597. Plummer MV. 1976. Some aspects of nesting success in the turtle, Trionyx muticus. Herpetologica 32(4): 353-359. Powell C. 1967. Female sexual cycles of Chrysemys picta and Clemmys insculpta in Nova Scotia. Canadian Field-Naturalist 81: 134-1340. Prange S, Gehrt SD, Wiggers EP. 2003. Demographic factors contributing to high raccoon densities in urban landscapes. Journal of Wildlife Management 67(2): 324-333. Pritchard PCH. 1969. Studies of the systematics and reproductive cycles of the genus Lepidochelys [dissertation]. Gainsville (FF): University of Florida, p. Ratnaswamy MJ, Warren RJ. 1998. Removing raccoons to protect sea turtle nests: are there implications for ecosystem management? Wildlife Society Bulletin 26: 846- 850. Ratnaswamy MJ, Warren RJ, Kramer MT, Adam MD. 1997. Comparisons of lethal and nonlethal techniques to reduce raccoon depredation of sea turtle nests. Journal of Wildlife Management 61(2): 368-376. Reid GK. 1955. Reproduction and development in the northern diamondback terrapin, Malaclemys terrapin terrapin. Copeia 1955(4): 310-311. Renfrew RB, Ribic CA. 2003. Grassland passerine nest predators near pasture edges identified on videotape. Auk 120(2): 371-383. Robinson C, Bider JR. 1988. Nesting synchrony-A strategy to decrease predation of snapping turtle ( Chelydra serpentine) nests. Journal of Herpetology 22(4): 470- 473. Rome MS, Ellis JC. 2004. Foraging ecology and interactions between herring gulls and great black-backed gulls in New England. Waterbirds 27(2): 200-210. Roosenburg W. 1991. The diamondback terrapin: population dynamics, habitat requirements, and opportunities for conservation. In: Chaney A, Mihursky JA, editors. New Perspectives in the Chesapeake System: A Research and Management Partnership. Proceedings of a Conference. Solomons (MD): CRC Publication 137. p 227-234. Roosenburg W. 1992. Life History Consequences of nest site selection by the diamondback terrapin, Malaclemys terrapin [dissertation]. Philadelphia (PA): University of Pennsylvania. 218 p. 120 Roosenburg W. 1994. Nesting habitat requirements of the diamondback terrapin: a geographic comparison. Wetlands Issues 6(2): 8-11. Roosenburg W. 1996. Maternal condition and nest site choice: An alternative for the maintenance of environmental sex determination? American Zoologist 36: 157- 168. Roosenburg W, Cresko W, Modesitte M, Robbins, MB. 1997. Diamondback terrapin (.Malaclemys terrapin) mortality in crab pots. Conservation Biology 11: 1166- 1172. Roosenburg W, Cresko W, Modesitte M, Robbins MB. 1997. Diamondback terrapin {Malaclemys terrapin) mortality in crab pots. Conservation Biology 11(5): 1166- 1172. Roosenburg W, Dunham AE. 1997. Allocation of reproductive output: egg- and clutch- size variation in the diamondback terrapin. Copeia 2: 290-297. Roosenburg W, Green JP. 2000. Impact of a bycatch reduction device on diamondback terrapin and blue crab capture in crab pots. Ecological Applications 10(3): 882- 889. Roosenburg W, Place AR. 1995. Nest predation and hatchling sex ratio in the diamondback terrapin: implications for management and conservation. Toward a Sustainable Coastal Watershed: The Chesapeake Experiment. Proceedings of a Conference 149: 65-70. Schmidt KA. 2003. Nest predation and population declines in Illinois songbirds: a case for mesopredator effects. Conservation Biology 17(4): 1141-1150. Schmidt-Nielsen K, Fange R. 1958. Salt glands in marine reptiles. Nature 182(4638): 783-785. Schneider DG, Mech LD, Tester JR. 1971. Movements of female raccoons and their young as determined by radio-tracking. Animal Behaviour Monographs 4: 1-43. Seigel RA. 1980a. Nesting habits of diamondback terrapins {Malaclemys terrapin ) on the Atlantic Coast of Florida. Transactions of the Kansas Academy of Sciences 83(4): 239-246. Seigel RA. 1980b. Predation by raccoons on diamondback terrapins, Malaclemys terrapin tequesta. Journal of Herpetology 14(1): 87-89. 121 Seigel RA. 1993. Recent population changes: apparent long-term decline in diamondback terrapin populations at the Kennedy Space Center, Florida. Herpetological Review 24(3): 102-103. Seigel RA, Gibbons JW. 1995. Workshop on the ecology, status, and management of the diamondback terrapin (. Malaclemys terrapin), Savannah River Ecology Laboratory, 2 August 1994: Final Recommendations. Chelonian Conservation and Biology 1(3): 240-243. Shealy RM. 1976. The natural history of the Alabama map turtle, Graptemys pulchra, in Alabama. Bulletin of the Florida State Museum, Biological Science 21(2): 47- 111. Silliman BR, Bertness MD. 2002. A trophic cascade regulates salt marsh primary production. Proceedings of the National Academy of Sciences, 6 August 2002. 99(16): 10500-10506. Singh DP. 1974. Analysis of environmental factors regulating the gonadal cycle in a tropical pond turtle, Lissemys p. granosa (Schoepff). Experimentia 30(8): 967- 968. Singh DP. 1977. Annual sexual rhythm in relation to environmental factors in a topical pond turtle, Lissemys punctata granosa. Herpetologica 33(2): 190-194. Snow JE. 1982. Predation on painted turtle nests: nest survival as a function of nest age. Canadian Journal of Zoology 60: 3290-3292. Souza FL, Abe AS. 2001. Population structure and reproductive aspects of the freshwater turtle, Phrynops geojfroanus, inhabiting an Urban River in Southeastern Brazil. Studies on Neotropical Fauna & Environment 36(1): 57-62. Spencer RJ. 2002. Experimentally testing nest site selection: fitness trade-offs and predation risk in turtles. Ecology 83(8): 2136-2144. Spencer RJ, Thompson MB. 2003. The significance of predation in nest site selection of turtles: an experimental consideration of macro- and microhabitat preferences. Oikos 102: 592-600. Stancyk SE, Talbert OR, Dean JM. 1980. Nesting activity of the loggerhead turtle Caretta caretta in South Carolina, II. Protection of nests from raccoon predation by transplantation. Biological Conservation 18: 289-298. Steams SC. 1992. The evolution of life histories. New York (NY): Oxford University Press. 249 p. 122 Sutherland WJ. 1996. From individual behaviour to population ecology. New York (NY): Oxford University Press. 213 p. Szerlag S, McRobert SP. 2006. Road occurrence and mortality of the northern diamondback terrapin. Applied Herpetology 3: 27-37. Temple SA. 1987. Predation on turtle nests increases near ecological edges. Copeia 1: 250-252. Tinkle DW. 1961. Geographic variation in reproduction, size, sex ratio, and maturity of Sternothaerus odoratus (Testudinata: Chelydridae). Ecology 42(1): 68-76. Tinkle DW, Wilbur HM, Tilley SG. 1970. Evolutionary strategies in lizard reproduction. Evolution 24(1): 55-74. Tucker JK. 1997. Natural history notes on nesting, nests, and hatchling emergence in the red-eared slider turtle, Trachemys scripta elegans, in west-central Illinois. Illinois Natural History Survey Biological Notes 140: 13. Valenzuela N, Janzen FJ. 2001. Nest-site philopatry and the evolution of temperature- dependent sex determination. Evolutionary Ecology Research 3: 779-794. Vickery PD, Hunter ML, Wells JV, 1992. Evidence of incidental nest predation and its effects on nests of threatened grassland birds. Oikos 63: 281-288. Virginia Institute of Marine Science-Center for Coastal Resources Management. 2004. York River Watershed Map. Watkins-Colwell GJ, Black M. 1997. Malaclemys terrapin terrapin (northern diamondback terrapin): predation. Herpetological Review 28(2): 87-88. Webster WD, Parnell JF, Biggs WC Jr. 1985. Mammals of the Carolinas, Virginia, and Maryland. Chapel Hill (NC): The University of North Carolina Press. 255 p. Whitelaw DM, Zajac RN. 2002. Assessment of prey availability for diamondback terrapins in a Connecticut salt marsh. Northeastern Naturalist 9(4): 407-418. Wilson DS. 1998. Nest-site selection: microhabitat variation and its effects on the survival of turtle embryos. Ecology 79(6): 1884-1892. Wilson EO. 1985. The biological diversity crisis. BioScience 35(11): 700-706. Wood DW, Bjomdal KA. 2000. Relation of temperature, moisture, salinity, and slope to nest site selection in loggerhead sea turtles. Copeia (1): 119-128. 123 Wood RC. 1997. The impact of commercial crab traps on northern diamondback terrapins, Malaclemys terrapin terrapin. Proceedings: Conservation, Restoration, and Management of Tortoises and Turtles-An International Conference. New York (NY): p 21-27. Wood RC, Herlands R. 1997. Turtles and tires: the impact of roadkills on northern diamondback, Malaclemys terrapin terrapin, populations on the Cape May peninsula, southern New Jersey, USA. Proceedings: Conservation, Restoration, and Management of Tortoises and Turtles-An International Conference; New York (NY): p 46-53. Zimmerman TD. 1992. Latitudinal reproductive variation of the salt marsh turtle, the diamondback terrapin {Malaclemys terrapin) [dissertation]. Charleston (SC): University of Charleston. 54 p. 124 VITA Victoria Ann Ruzicka Victoria Ann Ruzicka was bom in Smithtown, NY on December 17, 1981. As a child she was encouraged by her parents to be confident, inquisitive, and love learning. Victoria thrived in school, becoming an active member in varsity soccer, orchestra, clubs, and honors societies. She graduated within the top 3% of her class from Sachem High School in June 2000. Victoria began her college career at the State University of New York at Geneseo in August 2000. At S.U.N.Y. Geneseo, she became a member of Phi Eta Sigma, Golden Key, Beta Beta Beta and Phi Beta Kappa honors societies. During her spring 2003 semester, the author studied abroad and attended the University College of London in London, England. Victoria graduated Magna cum Laude from S.U.N.Y. Geneseo in May 2004 with a B.S. in Biology. In August 2004 the author entered The College of William & Mary as a graduate student in the Department of Biology. During her time in graduate school she presented her data at the Atlantic Estuaries Research Society conference in Philadelphia, PA and the William & Mary Graduate Symposium. Victoria Ann Ruzicka defended her thesis in October of 2006.