Title Page
 Table of Contents
 Chapter 1: Introduction
 Chapter 2: Evaluation of physical...
 Chapter 3: Evaluation of aerial...
 Chapter 4: Movement patterns of...
 Chapter 5: Recommendations

Title: Sea turtle nesting in the Ten Thousand Islands of Florida
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00067404/00001
 Material Information
Title: Sea turtle nesting in the Ten Thousand Islands of Florida
Physical Description: ix, 96 leaves : ill. ; 29 cm.
Language: English
Creator: Garmestani, Ahjond S, 1971-
Publication Date: 1997
Subject: Sea turtles -- Nests -- Florida -- Ten Thousand Islands   ( lcsh )
Sea turtles -- Florida -- Ten Thousand Islands   ( lcsh )
Loggerhead turtle -- Nests -- Florida -- Ten Thousand Islands   ( lcsh )
Loggerhead turtle -- Florida -- Ten Thousand Islands   ( lcsh )
Ten Thousand Islands (Fla.)   ( lcsh )
Dissertations, Academic -- Wildlife Ecology and Conservation -- UF   ( lcsh )
Wildlife Ecology and Conservation thesis, M.S   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Thesis: Thesis (M.S.)--University of Florida, 1997.
Bibliography: Includes bibliographical references (leaves 86-95).
Statement of Responsibility: by Ahjond S Garmestani.
General Note: Typescript.
General Note: Vita.
Funding: Florida Historical Agriculture and Rural Life
 Record Information
Bibliographic ID: UF00067404
Volume ID: VID00001
Source Institution: Marston Science Library, George A. Smathers Libraries, University of Florida
Holding Location: Florida Agricultural Experiment Station, Florida Cooperative Extension Service, Florida Department of Agriculture and Consumer Services, and the Engineering and Industrial Experiment Station; Institute for Food and Agricultural Services (IFAS), University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: aleph - 002327603
oclc - 39107309
notis - ALT1219

Table of Contents
    Title Page
        Title Page
    Table of Contents
    Chapter 1: Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
    Chapter 2: Evaluation of physical paramters as indicators of nesting beach selection for the loggerhead sea turtle (caretta caretta)
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
    Chapter 3: Evaluation of aerial survey for estimation of sea turtle nesting for the ten thousand islands: Application of a survey technique
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
    Chapter 4: Movement patterns of the Ten Thousand islands raccoon (procyon lotor marinus) and the effect of its removal on sea turtle hatch success
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
    Chapter 5: Recommendations
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
Full Text

Sea turtle nesting in the Ten Thousand Islands of Florida

August 1997

Technical Report #56

Completed for
U.S. Fish and Wildlife Service &
National Park Service

Ahjond S. Garmestani, H. Franklin Percival, Kenneth G. Rice and Kenneth M. Portier

Citation: Garmestani, A.S., H.F. Percival, K.G. Rice and K.M. Portier. 1997. Sea turtle nesting
in the Ten Thousand Islands of Florida. Fla. Coop. Fish and Wildl. Res. Unit, USGS-Biological
Resources Division Tech. Rep. 56. 96 pp.


First and foremost I would like to thank Dr. H. Franklin Percival for providing me

with advice, financial support, folk euphemisms, and the opportunity to work with an

individual as excellent as him. I am also very grateful to Dr. Ken Portier, Dr. Mel

Sunquist and Dr. Alan Bolten for serving as committee members and for offering

valuable advice throughout the course of this research. Special thanks are in order for

SCA/Americorps assistants Kimberley Augenfeld, Kevin LoGiudice, Victor Noerdlinger,

Wayne Taylor, Fred Williams, Jill Witt, and particularly Chuck Lane and Ellen Cheng.

None of the research I conducted would have been possible without their constant

assistance in the field and the laboratory.

Thanks are also in order for Maura Kraus and her staff with the Collier County

Department of Natural Resources and Dr. Todd Hopkins and his staff at Rookery Bay

National Estuarine Research Reserve. Dr. Ray Carthy and Dr. Ken Rice provided

invaluable assistance with research design and analysis. I also would like to extend my

gratitude to Dr. John Davis, Tess Korhnak, Dr. Ken Clark, Dr. Henry Gholz and the

Archie Carr Center for Sea Turtle Research, particularly Dr. Karen Bjomdal, for the use

of lab space and supplies. For their secretarial and administrative support, I am indebted

to Debra Hughes and Barbara Fesler. Others who offered invaluable advice are: Dr.

Clarence Abercrombie, Dave Addison, Craig Allen, Jill Blue, Dr. Mary Collins, Jason

Croop, Dr. Stanley Gehrt, Brendan Godley, Dr. Tim Gross, Dr. Ed Hanlon, Sally

Hopkins-Murphy, Seihun Kong, Dr. JoAnn Mossa, Tom Murphy, Larry Ogren, Dr. Mike

Pelton, Dr. Scott Smith, Phil Wilkinson and John Wooding.

This work would not have been possible without the funding and support of the

Florida Panther/Ten Thousand Islands National Wildlife Refuge (TTINWR), Everglades

National Park (ENP) and the Florida Cooperative Fish and Wildlife Research Unit. In


particular, Ben Nottingham, Jr. (TTINWR), Larry Richardson (TTINWR), Jim

Krakowski (TTINWR), Cindy Garcia (TTINWR), Lennie Jones (TTINWR), Jim

Durrwachter (TTINWR), Mike Mayer (ENP), Daryl Rhodes (ENP), Kathy Clossin

(ENP), Larry Anderson (ENP), Gene Wesloh (ENP), Sid Capo (ENP), Jim Brown (ENP)

and Judy Hayes (ENP) as well as the rest of the staff at ENP and TTINWR, and Sandy

McPherson of the Jacksonville, Florida U.S. Fish and Wildlife Service office. The U.S.

Coast Guard and the National Marine Fisheries Service provided air time in 1996. This

work was funded by the U.S. Fish and Wildlife Service (Cooperative Agreement # 14-16-

0009-1544, Research Work Order No. 119) through the Florida Cooperative Fish and

Wildlife Research Unit (U.S Geological Survey- Biological Resources Division, the

University of Florida, the Florida Game and Freshwater Fish Commission and the

Wildlife Management Institute cooperating).



Loggerhead sea turtles (Caretta caretta) nest in numerous substrate and beach

types within the Ten Thousand Islands (TTI) of southwest Florida. Nesting beach

selection was analyzed on 12 islands within this archipelago. Numerous physical

characteristics were recorded to identify the relatedness of these variables and determine

their importance for nesting beach selection in C. caretta. These variables were chosen

after evaluating the islands, conducting literature searches and soliciting personal

communications. Along transects, data were collected, on the following: height of

canopy, beach width, overall slope (beach slope and slope of offshore approach) and sand

samples analyzed for pH, percentage of water, percentage of organic content, percentage

of carbonate and particle size (8 size classes). Data on ordinal aspect of beaches and

beach length were also recorded and included in the analysis. All of the variables were

analyzed by tree regression, incorporating the nesting data into the analysis. In the TTI,

loggerheads appear to prefer wider beaches (p< 0.001; R2= 0.56) that inherently have less

slope, and secondarily, wider beaches that have low amounts of carbonate (p< 0.001). In

addition, C. caretta favors nest sites within or in close proximity to the supra-littoral

vegetation zone of beaches in the TTI (p< 0.001).

Aerial surveys were flown over the TTI region during June and July of 1996.

Twenty-nine of the possible 34 islands with potential nesting beaches were chosen for the

study. Ground surveys were conducted on 8 islands to provide a correction factor

between ground and aerial counts. All crawls counted on the ground were counted from

the air. This technique resulted in estimates of 906 nests and 997 false crawls in 1996 for

the 29 islands. It appears that sea turtles (Caretta caretta and Chelonia mydas) nest in

relatively high numbers in an area (TTI) not previously known to support significant sea

turtle nesting activity.


Predation by raccoons (Procyon lotor) is the primary factor in nest success for

sea turtles in the TTI archipelago. Eight islands within or adjacent to the Ten Thousand

Islands National Wildlife Refuge (TTINWR) have been surveyed for sea turtle nesting

activity from 1991-96. Raccoon trapping was conducted during the 1992 nesting season

and in 1995-96 from January through April, removing 21, 15 and 2 raccoons,

respectively. Removals in 1992 (21) resulted in continued high predation, indicative of a

high raccoon population and insufficient removals (nests= 42, 95% predation). Although

fewer raccoons were removed prior to the 1995 and 1996 nesting seasons (15, 2), no nest

predation occurred during either year (1995, nests= 41; 1996, nests= 62). These figures

are in contrast to 76-100% nest predation for the previous four years. Eight raccoons

were also trapped, tagged and released on Gullivan Key in 1995 (276 trap-nights, 3.2%

trap success). The following year only 1 raccoon was trapped and removed (114 trap-

nights, 0.9% trap success) resulting in no significant decrease in nest predation (nests=

33, 97% predation). The results from the TTI indicate that raccoon removal may be an

effective management tool for increasing sea turtle nest success on relatively remote

barrier islands, although some raccoons may become sensitized to traps if exposed to

traps prior to removal work. More work is needed to provide conclusive evidence that

sufficient and cost-effective trapping and removal can be conducted on these barrier

island beaches in order to increase sea turtle hatching success.



ACKNOWLEDMENTS ............................................................................i

ABSTRACT........................................................................................ iii


1 INTRODUCTION ...............................................................................1

2 EVALUATION OF PHYSICAL PARAMETERS AS..................................... 6

Introduction and Background.............................................................. 6
Homing behavior................................................................... 7
Nesting beach selection............................................................ 9
Nest-site selection................................................................. 11
Objectives...................................... ............................. ... 12
Materials and Methods................................................................... 13
Study Area.......................................................................13
Field Sampling Techniques.................................................... 13
Laboratory analysis............................................................. 17
Statistical analysis............................................................... 18
R esults............................................. ......................................... 18
Nesting beach selection......................................................... 18
Nest-site selection...............................................................24
Discussion .................................................................................24
Nesting beach selection......................................................... 27
Nest-site selection...............................................................30
Conclusion .................................................................... 32


Introduction and Background..........................................................34
Materials and Methods....................................................................35


Study Areas............................................ ........................ 35
Sampling Techniques...........................................................36
Formula ............................................................. ...... .... ..37
R esults...................................................... ..... ......... .................. 38
D discussion ....................................... .......... ................ ... ... ... ........ 4 1
Conclusion ................................................................... ..47

4 MOVEMENT PATTERNS OF THE TEN THOUSAND................................48

Introduction and Background..........................................................48
Materials and Methods................................................................... 50
Study Areas....................................................................... 50
Sampling Techniques...........................................................51
1995 Field Season...............................................................52
1996 Field Season...............................................................53
R results .................................................................................... ... 55
1995 Field Season...............................................................55
1996 Field Season................................................................60
Discussion........................................ ....................................... 64
Conclusion................................................. ............... ..... 65

5 RECOMMENDATIONS....................................................................68





Sea turtles have been the subject of much scientific and public attention due to

their imperiled status and prominent role on beaches of the world. The direct contact

with the public has heightened awareness and concern for sea turtle conservation.

Understanding distribution, nesting and habitat preferences, and sources of mortality is

critical to management of sea turtles and their habitats.

The Ten Thousand Islands (TTI) extend roughly 40 kilometers from Marco Island

to Pavilion Key along Florida's southwest coast (Edwards, 1991). Located

approximately 25 kilometers southeast of Naples, these largely undeveloped islands are

utilized by the public year-round, primarily for fishing (Cheng, 1996). The climate is

subtropical with 110 to 160 centimeters of average annual precipitation, two-thirds of

which occurs between May and October (Cheng, 1996). Average annual temperature is

23 degrees Celsius, with occasional freezing temperatures in the winter and regular mid-

thirties in the summer (Cheng, 1996).

The mangrove ecosystem of the TTI supports a wide range of invertebrates,

fishes, amphibians, reptiles, birds, and mammals. Roughly eighty-six species of fish and

over eighty species of birds have been documented in the area (Cheng, 1996). The

waters serve as a rich nursery ground and an important foraging area for many marine

creatures. Red (Rhizophora mangle) and black (Avicennia germinans) mangroves

dominate most of the islands. Where they exist, beaches are usually narrow and

interspersed with mangroves. Some of the most seaward islands have one or two long

stretches (several hundred meters) of uninterrupted beach. These islands are


characterized by a pioneer zone where vegetation meets the beach. A variety of grasses,

herbaceous plants, and vines are found in this zone, including seashore saltgrass

(Distichlis spicata), sea oats (Uniola paniculata), sea purslane (Sesuvium

portulacastrum), and railroad vine (Ipomoea pescaprae) (Cheng, 1996).

The loggerhead sea turtle (Caretta caretta), is a member of the family

Cheloniidae. It is known to nest in Florida, particularly on the east coast, in the largest

nesting aggregation in the world. Various techniques have been developed to estimate

sea turtle population characteristics. These include a description of techniques for

studying sea turtle population dynamics (Bustard, 1979), a study of population dynamics

of green turtles based on nesting females and hatching (Thompson, 1980) and estimation

of population dynamics for the southeastern United States based on netting of sea turtles

(Henwood, 1987). Chaloupka and Musick (1997) describe the latest techniques for

assessing population dynamics of sea turtles, discussing in detail the numerous variables

that have to be considered when attempting to create a population estimate. Several

researchers have based population estimates on nesting loggerhead females that were

tagged over the course of a 20-year period in Georgia (Crouse et al., 1987; Frazer, 1983;

Frazer, 1984; Richardson et al., 1978; Richardson, 1982).

Numerous studies have been conducted pertaining to sea turtles in Florida, with

some work conducted in southwest Florida. LeBuff & Beatty (1971) and LeBuff (1990)

reported on the nesting status of C. caretta on the Gulf coast of Florida, and LeBuff

(1969) discussed the status of marine turtles at Sanibel and Captiva islands along the

southwest coast of Florida. The studies mentioned above refer to basic descriptions of

the nesting status in southwest Florida in predominantly urban areas. Addison (1994)

commented on the nesting behavior of C. caretta on beaches in the Naples, Florida area,

conducting some work on uninhabited beaches similar to the TTI. LeBuff and Hagan

(1978) assessed the effectiveness of aerial survey as a means of documenting sea turtle

nesting activity on the beaches of the Ft. Myers-Naples, Florida area, with incidental


flights over the TTI included in the study. However, little is known of C. caretta in the

TTI of the west coast of Florida. The TTI have been neglected, with annual ground

surveys conducted by the U.S. Fish and Wildlife Service the only indicator of sea turtle

activity in the TTI. Further down the coast of southwest Florida, Klukas (1967) reported

on the status of marine turtles in Everglades National Park (ENP), while Davis and

Whiting (1977) created a nesting estimate for ENP lands and reported on sources of

mortality for sea turtle nests. Kushlan (1986) repeated the aerial methodology utilized by

Davis and Whiting (1977) in order to create an updated nesting estimate for ENP lands.

The flights over ENP did not cover the TTI in either of the previously mentioned aerial

studies in the TTI. Due to the lack of information concerning sea turtles, it became

important to develop techniques for assessing the nesting status of sea turtles in the TTI.

Included in these techniques are methods for assessing nesting preferences and

distribution of nesting by sea turtles, as well as conservation (i.e. predator control)


Once a sea turtle has navigated to its natal area, it may exhibit nesting

preferences. Only a few researchers directly attempted to clarify reasons for nesting

beach selection in sea turtles. Among these possible factors are an open offshore

approach, sediment type, vegetation, lighting, salinity and area of nesting beach

(Mortimer, 1990, 1995; Johannes and Rimmer, 1984). However, these factors may not

have inter-specific applications, and may differ from study area to study area around the

world. Furthermore, these particular characteristics have been treated individually by

these researchers; they were not subjected to multiple variable analysis.

Miller (1995) reported that sea turtles tend to renest in relatively close proximity

(0-5 km) during subsequent nesting attempts, while a small percentage may utilize more

distant nesting sites. Carr et al. (1978) found that green sea turtles select the same

beaches year after year for nesting, while avoiding other seemingly almost identical

beaches. Melucci et al. (1991) found evidence that hawksbill turtles also exhibit a


significant degree of site-fixity on Antigua. Bjorndal et al. (1985) also found evidence

that the hawksbill turtle exhibits a significant degree of site-specificity at Tortuguero,

although a greater degree of site-fixity existed between seasons rather than within

seasons. Mortimer and Portier (1989) found that green turtles exhibit a high degree of

nest site fidelity at Ascension Island. Tagging data showed that 70 percent of observed

renesting emergences occurred in the same beach cluster as the initial observed

emergence (Mortimer and Portier, 1989). In Chapter 2, I use tree regression to evaluate

numerous physical characteristics of the beaches in the TTI in order to establish

loggerhead sea turtle nesting beach preferences for a particular characteristic or

combination of characteristics. I also discuss nest placement and/or selection by C.

caretta in the TTI.

Aerial surveys have been employed to assess wildlife throughout the United

States and the world. The value of this type of assessment is dependent upon the species

or activity of the species being evaluated. Aerial surveys have been utilized to census sea

turtles and dugongs by several researchers (Marsh and Sinclair, 1989; Bayliss, 1986;

Irvine et al., 1981; Fritts et al., 1983). Marsh and Sinclair (1989) and Bayliss (1986)

reported success in counting dugongs from the air, but reported that due to the diving

behavior of sea turtles, it was difficult to assess their abundance. Fritts et al. (1983)

reported numerous sightings of a variety of species in the TTI region, including large

numbers of loggerheads, and lesser numbers of leatherback, Kemp's Ridley, and green

sea turtles. However, their counts may have coincided with weather and water conditions

that were favorable to spotting sea turtles, as was indicated by Mourao et al. (1994).

Mourao et al. (1994) clearly state that population estimates derived from aerial surveys

are subject to a host of sampling errors. In Chapter 3, I assess the problem of sea turtle

nesting distribution in the TTI. I discuss the possibility of creating a seasonal nesting

estimate for the TTI archipelago based on an aerial survey technique adapted from

Hopkins-Murphy and Murphy (1983).


Raccoons are known to prey on the eggs of numerous species, including:

American crows (Kilham, 1986), waterfowl (Urban, 1970), painted turtles (Christens &

Bider, 1987), wading birds (Frederick & Collopy, 1989), yellow mud turtles

(Christiansen & Gallaway, 1984), coastal burrow-nesting birds (Hartman, 1993) and

diamondback terrapins (Seigel, 1979). The impacts of raccoons on sea turtles are well

known throughout the United States, particularly in areas where continuous studies have

been conducted. The Georgia Department of Natural Resources assessed the impacts of

raccoons, as well as feral hogs on sea turtle nests on several barrier islands for six

seasons. Raccoons have been shown to have definitive impacts on the success of sea

turtle nesting over the six-year period (Harris and Maley, 1990; Maley and Harris, 1991,

1992; Maley and Murphy, 1993, 1994; Maley, 1995). Raccoons have been determined to

be a major factor in the success of sea turtle nesting in the TTI of southwest Florida (L.

Richardson, pers. comm.). The TTI are inhabited by the Ten Thousand Islands raccoon

(Procyon lotor marinus), a subspecies of P. lotor which was first identified by Nelson

(1930). P. 1. marinus is characterized by a smaller stature than the mainland subspecies,

as well as cranial differences (Nelson, 1930; Bigler et. al., 1977). Due to the high

incidence of depredation of sea turtle nests by raccoons (ranging from 59-85% for 1991-

1994) (Garmestani, 1995), a better understanding of raccoon re-population trends,

movement patterns and effects of removal on sea turtle hatch success in the Ten

Thousand Islands National Wildlife Refuge was necessary to create a more

comprehensive sea turtle management strategy (Shabica et al., 1978). I discuss this

research in Chapter 4 and offer management suggestions for raccoons in regard to sea

turtle conservation.

Finally, in Chapter 5, I suggest management options for the TTI with respect to

sea turtle conservation.




Introduction and Background

The Ten Thousand Islands (TTI) have been neglected when assessing sea turtle

populations in southwest Florida. No accurate estimates have been made for the region

because surveys conducted in the archipelago have been incomplete. Only the area

encompassed by the Ten Thousand Islands National Wildlife Refuge (TTINWR) has

been effectively monitored. TTINWR accounts for approximately 20% of the possible

turtle nesting beaches in the archipelago; the remainder is already under the ownership of

Everglades National Park (ENP) and the Florida Department of Environmental


The TTI are characterized by undeveloped mangrove islands with low human

activity, predominantly fishermen and campers (fall, winter and spring). The islands are

not subject to the kind of nesting disruption due to human usage that affects distribution

of loggerhead nests in other research areas (e.g., North Carolina; Fangman and

Rittmaster, 1994). Fangman and Rittmaster (1994) found that loggerhead sea turtles

tended to abort nesting attempts with the increased presence of humans (e.g., camping,

night-walking, parties). With the continued protection of these islands by the federal

government (National Park Service, U.S. Fish and Wildlife Service) and the Department

of Environmental Protection (Rookery Bay National Estuarine Research Reserve), the

relatively unspoiled turtle nesting beaches offered a unique opportunity for study.

Because of the potential significance of the TTI for sea turtle nesting, further exploration


of these islands and accurate predictions on sea turtle nesting was necessary to determine

the status of turtle populations in this area.

Information gathered for eight monitored islands in the refuge indicated that the

total nesting effort for the Ten Thousand Islands could be significant (Table 2-1).

Table 2-1. Total TTINWR turtle survey data (1991-1995)

Year # of crawls # of known # of depredated % of
nests nests depredated
1991 194 157 128 82
1992 255 139 118 85
1993 189 93 61 66
1994 232 157 92 59
1995 364 206 85* 41*

*- includes raccoon removal island (Panther Key)

Sea turtles exhibit three basic behaviors when returning to natal areas to nest:

homing behavior, nesting beach selection and nest-site selection. Homing behavior is

discussed as a means of inter-connecting these behaviors, although it was not examined

in this study. Nesting preferences are discussed in detail, with nesting beach selection the

focus of this research.

Homing behavior

Sea turtles are known to exhibit varying degrees of site-specificity to their natal

areas. Grassman et al. (1984) found evidence that sea turtles detect differences in water

samples when exposed to different samples in a laboratory setting. In laboratory

experiments, the researchers found that sea turtles preferred to spend time in water

solutions in which they had been artificially imprinted. This research suggests that sea

turtles may have the ability to identify their natal areas by imprinting on olfactory cues


specific to that particular area. Lohmann (1992) commented that tagging studies at

Ascension Island have shown that green turtle females remain faithful to their nesting

site; no turtle seen at Ascension has ever been found nesting elsewhere. Loggerhead

turtles also exhibit some nest site-fixity (Lund, 1986), as evidenced by the 49% return of

tagged female turtles over ten years at Cumberland and Little Cumberland Island,

Georgia (Richardson et al., 1978). The methods by which sea turtles orient and return to

natal areas has been the subject of much debate and research. Lohmann (1992) has

suggested that cues utilized by other migratory animals such as the position of the sun or

stars, polarized light, odors, wind direction, infrasound and/or the earth's geomagnetic

field may play a role in navigation for sea turtles. Lohmann (1992) does not rule out the

possible reliance upon chemosensory factors, as suggested by Grassman et al. (1984), or

other as of yet unidentified cues, but hypothesizes that they are not a primary factor in


Lohmann (1992) suggested that adult sea turtles appear to fix their position with

respect to their destination. Furthermore, he discounted the use of stars as an orientation

cue based on anatomical studies of sea turtle eyes that revealed them to be extremely

myopic when their heads are above water (Ehrenfeld and Koch, 1967; Koch et al., 1969).

He further suggested that adults may have a "map sense", that allows them to orient

according to the magnetic field line inclination, or by detecting pathways along the ocean

floor created by magnetic maxima and minima. As with hatchlings, wave direction may

be a possible cue, due to the independence of swell direction from local weather patterns,

for adult turtles migrating back to nesting areas.


The idea of "social facilitation", suggested by Owens et al. (1982), was refuted by

conducting mtDNA analyses on three different sea turtle species (Chelonia mydas,

Caretta caretta and Eretmochelys imbricata) (Allard et al. 1994; Bowen et al., 1992;

Bowen et al., 1993; Broderick et al., 1994; Meylan et al., 1990). Bowen (1995) stated

that the mtDNA evidence suggests that nesting populations are distinct demographic

units, and depletion of nesting aggregates will not be compensated by recruitment from

other populations. What this evidence suggested is that sea turtles are fixed upon a

particular natal area and return to that area utilizing a navigation mechanism that is not

yet understood. These aforementioned studies dealt with some of the possible

mechanisms sea turtles may use to return to their natal area, in this case the TTI and/or

southwest Florida, but further exploration was necessary to indicate preferences for

particular beaches.

Nesting beach selection

Shoop et al. (1985) made the observation that a high-density loggerhead sea turtle

nesting beach was abandoned over the course of a five-year period due to, in their

estimation, a mud flat that developed on the ocean side of the island. This sort of drastic

change in a suitable nesting habitat does not necessarily affect the homing behavior of sea

turtles to a particular area, but it would drastically affect nesting beach selection.

Therefore, it appears that certain chemical and/or physical factors influence loggerhead

sea turtle nesting beach selection. Mortimer (1995) has conducted the most

comprehensive research dealing with nesting beach preferences in sea turtles. She

conducted research on green turtles at Ascension Island, analyzing numerous components

of the island in an attempt to determine nesting beach preferences for C. mydas.


Mortimer (1995) commented that there are some basic requirements indicative of a good

nesting beach. Among these characteristics are: easy accessibility from the sea, a beach

platform that is high enough to not be inundated by high tides or the water table, and

beach sand that facilitates gas diffusion, but moist enough and fine enough to prevent

excess slippage during nest construction (Mortimer, 1995). Taking these factors into

account, Mortimer (1995) evaluated beach length, rock cluttering the foreshore, rock

obstructing the offshore approach and artificial light visible on the shore as possible

indicators of nesting beach selection.

Sea turtle nesting preferences for C. caretta in the TTI were assessed by

evaluating each nesting beach based on a subset of physical characteristics of the islands.

These characteristics included: height of canopy (Bustard, 1972); beach width, overall

slope (beach slope and slope of offshore approach) (Mortimer, 1995; Shoop et al., 1985)

and sand samples analyzed for pH, percentage of water, percentage of organic content,

percentage of carbonate and particle size (8 size classes) (Mortimer, 1995; Mortimer,

1990; Stancyk and Ross, 1978; Hughes, 1974). Data on ordinal aspect of beaches and

beach length (Mortimer, 1995) were also recorded and included in the analysis. These

factors were based upon characteristics of the islands, past research and personal

communications (R. Carthy, A. Bolten, K. Rice and H. Percival). Although green sea

turtles were known to nest in the TTI, they were not included in this study.

Nest-site selection

Numerous studies have been conducted on nest-site selection in sea turtles,

addressing various factors as potential indicators of nest-site selection. Johannes and

Rimmer (1984) found that nest sites for the green turtle were on beaches that had lower


salinity at nesting depth, lower salt content at the surface of the sand and were sheltered

from prevailing winds. Stoneburner and Richardson (1981) found that sea turtles cross a

temperature gradient after they have landed on a nesting beach. A nesting female will lay

her eggs once she has passed over a zone of the beach where a rapid rise in temperature is

encountered. However, Hays et al. (1995) were critical of the latter study, indicating that

the sand temperatures Stoneburner and Richardson (1981) recorded were from disturbed

sand at C. caretta nest sites. Hays et al. (1995) suggested that the abrupt temperature

increase at nest sites may be a result of deeper, warmer sand being brought to the surface.

They were also critical of the lack of explanation for this abrupt temperature increase (i.e.

no correlation with any of the features of the beach topography). Hays et al. (1995)

compared nest-sites for loggerhead turtles in southwest Florida, and green turtles on

Ascension island, and found that loggerheads clump their nesting attempts close to

vegetation backing nesting beaches, while green turtles appear to cue on uneven beach

topography when choosing nesting sites. Stancyk and Ross (1978) found no correlation

between sand characteristics of the beaches at Ascension Island and nesting frequency of

green turtles. Mortimer (1990) also conducted a study at Ascension Island and found no

preference for a particular sand type based on clutch survival, and concluded that factors

other than the physiognomy of the sand on nesting beaches may be more important in

nest-site selection for the green turtle. Bjorndal and Bolten (1992) concluded that annual

factors may have a greater effect on nest placement than individual nesting patterns for

green turtles at Tortuguero. This lack of consistent nest distribution may result from

environmental uncertainty and low predictability of nest success (Bjorndal and Bolten,

1992). Mrosovsky (1983) was unable to explain the poor nest-site selection of


leatherback turtles in Guianas and Malaysia. In addition, Camhi (1993) conducting

research at Cumberland Island, Georgia found that loggerhead turtles did not show strong

site preferences, even though certain areas of the beach were more likely to produce

hatchlings. Whitmore and Dutton (1985) found that leatherbacks nested more in open,

sandy areas, while green turtles nested in predominantly vegetated areas. Daud and

deSilva (1987), found that leatherbacks preferred to nest in fine sand on steep sloped

beaches, while green turtles preferred to nest in coarser sand on broad, flat beaches.

Nest-site selection was also evaluated for C. caretta utilizing nesting data from

the 1996 field season. No attempt was made to evaluate the possible cues indicating nest-

site selection.


The primary objective of this research was to establish if the loggerhead sea turtle

exhibits preferences for a particular physical characteristic, or combination of

characteristics when selecting nesting beaches. The secondary objective of this research

was to document the overall distribution of loggerhead sea turtle nests and determine if

there was a significant trend of nest placement for the TTI.

Materials and Methods

Study Area

The TTI extend roughly 40 kilometers from Marco Island to Pavilion Key along

Florida's southwest coast (Edwards, 1991). Located approximately 25 kilometers

southeast of Naples, Florida, these largely undeveloped islands are utilized by the public

year-round, primarily for fishing (Cheng, 1996). The climate is subtropical with 110 to


160 centimeters of average annual precipitation, two-thirds of which occurs between May

and October (Cheng, 1996). Average annual temperature is 23 degrees Celsius, with

occasional freezing temperatures in the winter and regular mid-thirties in the summer

(Cheng, 1996).

The study area included 12, low-relief mangrove islands, further broken down

into 35 beaches based on the ordinal aspect of the individual beaches. Islands in the

study area directly face the Gulf of Mexico and were known to support sea turtle nesting.

These islands included: Brush, B, Turtle, Gullivan, White Horse, Hog, Panther and

Round Keys, located in the Cape Romano Aquatic Preserve and Ten Thousand Islands

National Wildlife Refuge, and Kingston, Indian, Picnic, and Tiger Keys in Everglades

National Park. These islands are located to the southeast of Marco Island, and are part of

the western delta of Florida's Everglades. Beaches are made up of differing substrate

types that are usually backed by a variety of supralittoral vegetation (Table 2-2). The 35

beaches examined totaled 7.404 kilometers in total length.

Field Sampling Techniques

Twelve (12) islands were selected for study based on logistical constraints and

supporting agencies' (TTINWR and ENP) preferences. This included the 8 islands

covered in the five-year monitoring period for TTINWR (Edwards, 1991) and 4

additional islands within ENP. Although these islands do not represent a random sample

of possible islands in the TTI, they are representative (Garmestani, pers. observe ) of the

range for each of the variables selected for analysis. The 12 islands are a sample of the

19 northernmost keys, with 10 other keys in the TTI also included in the generalization of

the results of this research.

Sea turtle surveys were conducted from May 13, 1996 until August 15, 1996.

Surveys were conducted twice a week for each island. This strategy was developed by

the U.S. Fish and Wildlife Service, due to the remoteness of the TTI and lack of

personnel. A total of 252 loggerhead nests was used in the analysis for nesting beach


Table 2-2. List of plant species sampled within quadrats at
transect points

Common Name

Scientific Name

,Black mangrov
Saffron plum
Gray nickerbea
Sea rocket
Bay bean
Australian pine
Beach spurge
Sea grape
Latherleaf *
Seashore saltgi
Water grass
Devil's potato
Black torch
Finger grass
Spider lily
Wild potato vine
Railroad vine
Morning glory
White mangrov
Trailing ludwigih
Poorman's patc
Prickly pear
Bahia grass
Painted leaf
Red mangrove
Cabbage palm
Sea purslane
Cord grass
Bay cedar
Sea oats
Spanish bayoni
Dildoe cactus
Exotic specie

Arenaria lateriflora
e Avicennia germinans
Bumelia celastrina
n Caesalpinia bonduc
Cakile edentula
Canavalia rosea
* Casuarina equisetifolia
Cenchrus incertus
Chamaesyce mesembrianthemifoluim
i Chrysobalanus icaco
Coccoloba uvifera
Colubrina asiatica
Conocarpus erectus
Cyperus retrorsus
rass Distichlis spicata
Echinochloa walteri
Echites umbellata
Erithalis fruticosa
Eustachys petraea
Hymenocallis latifolia
S Ipomoea pandurata
lpomoea pes-caprae
Ipomoea trichocarpa
e Languncularia racemosa
Ludwigia alterniflora
a U Ludwigia palustris
Melanthera nivea
zh Mentzelia floridana
Opuntia compressa
Paspalum notatum
Passiflora suberosa
Poinsettia heterophylla
Rhizophora mangle
Sabal palmetto
Scaveola plumieri
Sesuvium portulacastrum
Spartina patens
Suriana maritima
Uniola paniculata
et Yucca aloifolia
Cereus pentagonus




selection, while a subset of 236 nests was used in the analysis for nest-site selection.

Crawls were described and data were collected from each nest site. Included in the data

set were species of turtle, vegetation (if any), distance of nest from mean high tide line

(MHTL), distance of nest from the vegetation (calculated as a percentage of the beach

width for each nest), distance of nest from the nearest transect and status of nest (e.g.,

depredated, inundated).

Island characteristics were measured in April and May of 1996, prior to sea turtle

nesting season. Each of the 12 islands was assessed for a subset of physical

characteristics that might influence loggerhead nesting beach selection along 82 transects

that were placed at 100m intervals along nesting beaches. Each transect was run

perpendicular from the beach/vegetation transition (i.e. starting point) to a point 100m

offshore using a compass to maintain the perpendicular vector of the transect, and a

survey tape to record distance. The beach/vegetation transition was defined by the first

occurrence of vegetation on the open beach and marked with a stake along the transects.

Data were collected on the following characteristics:

1) Beach width Beach width was defined as the distance from the MHTL to the

beach/vegetation transition, and measured in meters along the transects using a distance

measuring wheel.

2) Slope Slope was assessed along the transects from each of the individual

stakes. Sample point distances were taken in meters with a survey tape. Depth

measurements were made in meters using a laser transit and survey pole, or in the case of

deeper areas, a Lowrance sonar (X-16 paper graph unit; Lowrance Electronics, Inc.,

Tulsa, OK, USA). Measurements were taken in increments based upon the rate of

change of the slope. Least squares technique was used to estimate slope based on linear,

quadratic or cubic models.

3) Height of canopy Canopy height was defined as the height of the vegetation

taken while standing at MHTL, and observing the canopy with a clinometer. This


characteristic was assessed along each transect using the clinometer and a survey tape,

and measured in meters.

4) Nesting substrate Nesting substrate samples were taken at MHTL along each

transect (i.e. shell content, particle size, organic content, water content and pH levels).

Samples were taken at MHTL, because after landing a sea turtle may exhibit a preference

for a particular nesting beach based on factors in the sand with which she comes in

contact as she initially leaves the surf for the open beach. The nesting substrate samples

were collected by digging vertically and taking a 300 gm sample, placing the sample in a

ziplock bag and storing the sample on ice, to preclude chemical changes in the samples.

The samples were maintained on ice as they were transported to the field headquarters in

Everglades City and refrigerated until they were taken to Gainesville on ice. After

arriving in Gainesville, the samples were refrigerated until they were analyzed in the


5) Ordinal aspect Ordinal aspect is defined by the angle of the vector running

parallel to a nesting area. This feature was assessed with US Geological Survey, Collier

County and NOAA maps, and compass measurements. The 12 islands were divided into

35 "beaches" based on the ordinal aspect of their nesting areas.

6) Beach length Beach length was defined as the horizontal cross-section

distance of each beach, and was measured in meters using a distance measuring wheel

along MHTL.

7) Vegetation Nesting area vegetation was classified according to common

name and scientific name at each transect starting point in a 1m2 quadrat.

Laboratory analysis

Sand samples were assessed for pH, particle size, water content, organic content

and carbonate content. For pH analysis, two 5g sub-samples were taken. Since sea

turtles would come into contact with wet sand, one was assessed directly, while the other

was processed in a drying oven at 105 degrees Celsius for 12-16 hours as per standard


soil science technique. Each sample was mixed with de-ionized and distilled water in a

1:2 mixture, and allowed to sit for one hour. Then the mixture was stirred and a pH

reading was taken and recorded.

For particle size, a 50cc. sub-sample was air-dried for 72 hours and was shaken

through 8 sieves (< 0.246mm, 0.246, 0.351, 0.495, 0.701, 0.991, 1.397, 1.981mm). Each

size class was weighed and the results recorded. For water, organic and carbonate

content, one 25cc. sub-sample was taken and subjected to analysis outlined in Dean

(1974) and Schulte (1988). The 25 cc. sub-sample was dried in a drying oven at 105

degrees Celsius for 12-16 hours, allowed to cool to room temperature in a desiccator and

then weighed to determine water content. The dried sub-sample was tested for organic

content by heating in a muffle furnace for 24 hours at 360 degrees Celsius and weighing

after cooling to room temperature in a desiccator. The resulting sample was processed by

heating in a muffle furnace at 850 degrees Celsius for 4 hours, and weighing after it had

cooled to room temperature in a desiccator, to determine carbonate content.

Sand samples for organic and carbonate content were assessed in triplicate in

order to validate consistent laboratory technique, as well as account for variability among

sub-samples. The average values obtained from the laboratory analysis were utilized for

statistical analysis.

Statistical analysis

The multi-dimensional data which resulted from the field measurements and

laboratory analysis were plotted in their first two principal components to detect obvious

clustering of transects. The relationship between beach utilization for nesting and beach

characteristics was examined using a robust tree regression technique (Portier and

Anderson, 1995). The final tree regression model was used to indicate which

characteristics should be considered important in predicting beach use by nesting turtles.

A binomial test was used to compare nest counts on the beach to nest counts in the


vegetation. A Chi-square goodness-of-fit test was used to determine if nesting occurred

uniformly across the width of the beach.

Computations associated with the formal statistical tests and computations of

beach slope were performed in SAS (SAS Institute, Inc, 1989) on a personal computer.

The tree regression and principal components analysis computations were performed

using S-PLUS (Statistical Sciences, 1993; Venables and Ripley, 1994) on a UNIX



Nesting beach selection

The principal components plots did not demonstrate significant clustering within

the data. A more formal cluster analysis using the average linkage method also failed to

demonstrate clustering.

The final regression tree (R2= 0.56; Figure 2-1) identified beach width (p< 0.001)

as the primary factor affecting nesting beach selection for C. caretta. Beaches that are

wider (>8.5m; n= 15) on average have much higher nesting than narrower beaches

(<8.5m; n= 20). Beach width was negatively correlated (r= -0.6) with beach slope, wider

beaches having inherently less slope. For wider beaches (> 8.5m) the regression tree

indicated a further split using percentage of beach carbonate (p< 0.001). On wide

beaches loggerheads preferred less than 9% carbonate (n= 10) (i.e. function of the

shelliness of a beach), over beaches that had over 9% carbonate (n= 5). Wider beaches

having low carbonate levels will be expected to have the highest nesting. The percentage

of carbonate (i.e. shelliness) was also positively correlated (r= 0.88) with the three largest

particle size classes (0.991mm, 1.397mm, 1.981mm). Narrow beaches (n= 20) (< 8.5m

beach width) had 153 crawls, of which 60 (39%) resulted in nests. Wide beaches (n= 5)

(> 8.5m beach width) with high levels of carbonate (> 9% carbonate), received 50 crawls,


A .


0 C3

l5 a5 o
c v
1 --.

J 0 v

C E-o

a o-g i g

CO? co
^~ v- ^


of which 21 (42%) resulted in nests. Finally, beaches that were wide (> 8.5m) and had

less than 9% carbonate (n= 10), had a total of 292 crawls, of which 171 (59%) resulted in

nests. The number of nests on beaches increased dramatically in conjunction with

increased beach width (Figure 2-2). The nesting to non-nesting emergence ratio was

decidedly higher on beaches possessing the favored characteristics (Figure 2-3). All of

the beach data are summarized in Table 2-3.

The other variables included in the analysis were not as highly correlated to

loggerhead sea turtle nesting beach selection as were the primary and secondary

components. Although these variables may be of tertiary importance in nesting beach

selection, there influence will be much less than the two characteristics identified in the

regression tree. With a larger data set it might be possible to determine which of these

variables further discriminate nesting beach selection.

After running the duplicate analyses on the sand sub-samples, it was determined

that there was greater than 2% variability between several of the samples. Therefore, it

was necessary to run a third sample set in order to get an average of the three samples and

alleviate concerns over variability between the samples. Overall, the triplicate samples

had a 9.2% coefficient of variation.

Nest-site selection

There were 236 loggerhead nests used in the analysis, 111 of which were located

in the vegetation and 125 on the open beach (Figure 2-4). No significant preference

between open beach and vegetation was detected for C. caretta nest placement (p= 0.82,

binomial test). However, location of nests was not uniformly distributed in relation to the

vegetation and vegetation lines on beaches (p< 0.001, Chi-square test; Figure 2-5). Of

these nests (n= 236), 127 were depredated, 106 hatched and 3 were inundated. For the

depredated nests (n= 127), there was no significant difference in depredation rates

detected for nests in the vegetation (n= 59) or nests on the open beach (n= 58) (p= 0.14).


12 15

-*-- Beaches


Beach width (m)

Figure 2-2. Number of nests on TTI beaches in relation to beach width


120 -
115 -
110 -
105 -
100 -
95 -
90 -
85 -
80 -
75 -
70 -
65 -
60 -
55 -
45 -
40 -
30 -
20 -
15 -
10 -









z z

6 0o 5

A M A*-
(seoue~jweu -0 equlnu)
^0 rt~~ 03 -
s: ili
:~ C. Il

o:H* ~^

e~euoqje~ o 10e~elueojed pue Lfp|M ipeeg


Table 2-3. Index of physical components for Ten Thousand Islands


Beach slope 82
Offshore slope 82
Beach width 82
Height of canopy 82
pH (wet) 82
pH (dry) 82
% H20 82
% Organic material 82
% Carbonate 82
% Sand 1 (1.981mm) 82
% Sand 2 (1.397mm) 82
% Sand 3 (0.991mm) 82
% Sand 4 (0.701 mm) 82
% Sand 5 (0.495mm) 82
% Sand 6 (0.351 mm) 82
% Sand 7 (0.246mm) 82
%/ Sand 8 (<0.246mm) 82
Beach length 35
Ordinal aspect 35

N Mean S.E.









4 80

C 70-


20 -

10 -

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Distance from vegetation line (m)

Figure 2-4. Distance (m) of loggerhead nests to supra-littoral vegetation line




0 0

Je -4l

) Io w I N
*0 C


a- 0
'l^-^1 '-&

sisa JO T-Wfl
'. ; / .' **.0 ?)
.__ ^. ._ -rf,,a.joe- (
^ *-a ..*' ^ -- .^ ^V I. (


There was no significant difference in the hatch rate for the hatched nests (n=

106) whether they were in the vegetation (n= 51) or on the open beach (n= 55) (p= 0.39).


It is now widely accepted that sea turtles have the ability to home to particular

areas of the world's seas for feeding, mating and nesting. Lohmann (1992) has described

mechanisms of navigation utilized by hatchling sea turtles, and has hypothesized about

the possible mechanisms that adults use to home to their natal areas for nesting. The

distinctiveness of sea turtle populations as demographic units indicates that sea turtles are

programmed to home to natal areas (Bowen, 1995). Meylan et al. (1990) concluded that

social facilitation to non-natal sites was rare due to evidence of mtDNA sequences for

turtles sampled. Allard et al. (1994) supported the results found by Meylan et al. (1990)

by conducting a study on green turtles at Tortuguero, Costa Rica and the southeast coast

of Florida. They found evidence, in the form of mtDNA sequences, that the Costa Rican

and Florida populations were significantly different (Allard et al., 1994).

On a finer scale, Murphy and Hopkins-Murphy (1990) conducted research on the

homing behavior of gravid loggerhead turtles in South Carolina. Of the 27 radio-tagged

turtles they collected data from, 23 (85%) showed evidence of returning to their home or

primary beach (Murphy and Hopkins-Murphy, 1990). They found that there is a

plasticity to site selection which allowed for the acceptance of suitable alternate nesting

beaches for one or more nesting events for several of the turtles (Murphy and Hopkins-

Murphy, 1990). They hypothesized that loggerhead sea turtles may select beaches

according to offshore topography and/or the suitability of the beaches for nesting at the

translocation site (Murphy and Hopkins-Murphy, 1990). They suggest that loggerhead

sea turtles may be exhibiting preferences when selecting nesting beaches. Carr and Carr

(1972) reported that green turtles showed strong site fixity, although some re-nesting


attempts stray up to 7 km from previous nesting attempts. They suggest that colony

expansion might occur in this manner (Carr and Carr, 1972). Another possible

explanation for this phenomenon is that green turtles have a preference for particular

beach characteristics. After homing to natal areas, sea turtles may have several beach

types from which to choose, and select nesting beaches according to criteria specific to

their natal area.

Nesting beach selection

Mortimer (1995) reported that the heaviest nesting by green turtles on Ascension

Island occurred on beaches that were unlighted, with open offshore approaches and

foreshores relatively free of rock clutter. In addition, most nesting occurred on stretches

of beach where the offshore approach was deepest (Mortimer, 1995). For the beaches

that met these criteria, the correlation between beach length and the estimated number of

clutches laid per season was highly significant (Mortimer, 1995). Mortimer (1995) found

no relationship between beach vegetation or beach sand characteristics and density of

nesting. Johannes and Rimmer (1984) reported that no relationship has been found

between nesting beach selection and pH, calcium carbonate content, water content,

organic content and particle size. However, they found beaches favored by the green

turtle in Australia had lower salt content and less exposure to prevailing winds (Johannes

and Rimmer, 1984). Although Johannes and Rimmer (1984) found a positive

relationship between these characteristics and green turtle nest density, there is no

evolutionary explanation for the observed phenomenon. While sea turtles could possibly

detect elevated levels of salinity upon immediate landing on a nesting beach and choose

another nesting beach, there is no clear relationship between elevated levels of salinity

and sea turtle nesting beach selection. There is no literature referring to the potential

impacts of increased salinity on egg development in sea turtles. One of the only

references in the literature to salinity and its effects on development, deals with its

impacts on egg development in Baltic cod (Westin and Nissling, 1991). Therefore, the


connection between salinity and nesting beach selection is questionable. Furthermore,

the connection between prevailing winds and nesting beach selection is not clear. This

point is reinforced by the fact that prevailing winds in the study area frequently ceased at

night, making them a non-factor in nesting beach selection (Johannes and Rimmer,


The tree regression approach was used in the analysis because it is very flexible,

allowing the data to specify complex interactions where justified, and very simple to

interpret. This approach is also less concerned with underlying distributional

assumptions than are traditional linear regression approaches. Tree regression models

expected nest counts as a binary decision process, which is a result of comprehensive

examinations of all levels of each characteristic in a hierarchical model-building process.

The process involves constructing a large binary tree, then pruning the tree back to a size

which is consistent among sub-samples of the complete data set.

For the loggerhead sea turtle, a positive relationship was found between beach

width, which is highly correlated with slope (i.e. wider beach=decreased slope), and the

number of nests (n= 192; 76%) laid on beaches. On these wider beaches, loggerheads

preferred those with a lower shell content, which is highly correlated with the three

largest particle size classes analyzed. The fact that the number of non-nesting

emergences increased on beaches that do not possess the favored criteria only lends

strength to the tree model.

Loggerheads may prefer wider beaches because the loss of nests from tidal

inundation or erosion decreases with distance from the surf (Fowler, 1979). Choosing

beaches that are relatively wide may be the only alternative for a sea turtle attempting to

produce a successful clutch in a region like the TTI, where beaches are narrow and

subject to wide variation in tidal fluctuation from season to season and year to year.

Slope may simply be a construct of beach width in the area, or C. caretta may be cueing

on the slope of the beach. This makes sense, because the gradient of the beach


determines the distance a turtle must crawl overland in order to reach a nest site

(Pritchard, 1971; Schulz, 1975). Loggerheads may prefer more gently sloping beaches

because they afford the quickest and easiest approach and means of escape after laying a

nest. Bjorndal and Bolten (1992) have suggested that in long-lived species that cannot

accurately assess the nesting environment, it may be better to select to nest on the basis of

the survivorship of the adult female rather than on the survivorship of the clutch.

C. caretta may choose to nest on wide beaches that are relatively free of large

shell fragments because the shells may make it more difficult for the turtle to create an

adequate nest cavity. Digging in shells is difficult, even for a creature as large and

powerful as a loggerhead sea turtle. Furthermore, nesting cavities made up of shells will

be more prone to collapse as the turtle excavates the chamber. Considering that the

quality of beach sand can so strongly influence reproductive success, one would expect

turtles to use sand textures as a criteria in selection (Mortimer, 1990). Data gathered by

Mortimer (1995) at Ascension Island suggest otherwise. There was no correlation

between the average percent hatchling emergence at Ascension Island beaches and

nesting density (Mortimer, 1995). It appears that the choice of nesting beach is primarily

determined by the ease with which a nest site can be reached and excavated, with little

choice exhibited for more favorable reproductive substrates.

No significant relationship was found between pH, percentage of water,

percentage of organic content, or the 5 smallest classes of the 8 particle size classes and

the number of nests laid on beaches in the TTI. From these data, it appears that C.

caretta does not select nesting beaches according to the height of the canopy immediately

adjacent to nesting beaches, the beach length, the slope of the offshore approach, or the

ordinal aspect of the beaches. Although some of these factors could potentially affect the

reproductive success of sea turtles, none affected the nesting beach selectivity of C.



Nest-site selection

Nest-site selection has been the subject of much sea turtle research (Bjomdal and

Bolten, 1992; Camhi, 1993; Carr and Carr, 1972; Daud and deSilva, 1987; Miller, 1995;

Mrosovsky, 1983; Whitmore and Dutton, 1985). Numerous theories have been suggested

as possible explanations for this critical process in sea turtle ecology, with multiple

variables assessed as potential indicators of nest-site selection. Among these are salinity

(Johannes and Rimmer, 1984), temperature (Stoneburner and Richardson, 1981), particle

size, pH, water content, organic content, shell content, substrate conductivity (Mortimer,

1990; Stancyk and Ross, 1978) and vegetation (Hays and Speakman, 1993). A

connection has been made between temperature, salinity and nest-site selection (Johannes

and Rimmer, 1984; Stoneburner and Richardson, 1981), and some researchers have

suggested the vegetation zone backing beaches as a potential cue for sea turtles to select

nest sites (Hays and Speakman, 1993). The importance of this process on the

reproductive success of the turtle is critical, however, sea turtles often choose nest sites

that do not maximize their reproductive output, and therefore their fitness. When it is

apparent that certain beach types drain better, provide better gas and water exchange and

are more favorable for maintaining a nest cavity (Pollock and Hummon, 1971), it is hard

to imagine that sea turtles are not programmed to detect these obvious advantages in

selecting certain nest sites over others (Mortimer, 1995). Bolten and Bjorndal (1992)

have suggested that a female may have a limited ability to assess the current nesting

environment when selecting a nest site, and unpredictable changes in the nesting

environment over a 60-day incubation period. They believe this environmental

uncertainty and the inability of female turtles to assess the nesting habitat in terms of

ultimate hatching success may be responsible for the lack of any consistent patterns in

nest distribution for green turtles at Tortuguero (Bjorndal and Bolten, 1992). Hays and

Speakman (1993) found that loggerheads tended to lay nests away from the sea, and close

to, but not beyond the vegetation line. The increase in the distance turtles laid from the


sea when the vegetation line was further from the water suggests that the vegetation may

have constrained the length of the turtle's inland crawl (Hays and Speakman, 1993).

Some researchers have suggested that loggerhead sea turtles simply crawl a random

distance above the most recent MHTL prior to digging a nest site (Hays et al., 1995).

They constructed a random-crawl-distance model that closely reproduced the observed

spatial pattern of nests (Hays et al., 1995). Hays et al. (1995) stated that the distribution

of nests on Sanibel and Captiva islands in southwest Florida can be explained by the

random-crawl-distance model without implicating micro-habitat cues.

Although it has been suggested that supra-littoral vegetation backing nesting

beaches may act as a cue for nesting sea turtles (Hays and Speakman, 1993),

documentation of nest placement by C. caretta in the TTI indicates that no such

relationship exists. Of the nests analyzed for 1996 (n= 236), 111 were laid within the

supra-littoral vegetation backing nesting beaches, while 125 were laid on the open beach

at varying lengths from the vegetation line. The majority of the 125 nests laid on the

open beach were clumped in close proximity to the vegetation line. Furthermore, nests

laid in the vegetation suffered nearly equal depredation rates when compared to nests laid

on the open beach, and therefore no appreciable difference in hatch rates. Since no

micro-habitat cues were analyzed for this research, it is difficult to explain why C. caretta

nests in this manner in the TTI. The most obvious explanation for this phenomenon is

that loggerheads may select nest sites that will lessen the chance of tidal inundation or

inundation of nests by the water table (Fowler, 1979). The possibility exists that they

simply crawl a random distance beyond the last MHTL and lay their clutch, as has been

suggested by Hays et al. (1995). No attempt was made to verify the theory proposed by

Hays et al. (1995); the theory only exists as one possible explanation for nest placement

by the loggerhead sea turtle in the TTI.



The TTI is a unique ecosystem, and results from this study are unlikely to be

directly applicable to other loggerhead nesting areas. The same is true of the green turtle

work conducted by Mortimer (1995) at Ascension Island. Therefore, the conclusions

reached about green turtles and loggerhead turtles may be specific to the environments

evaluated. Without testing these methodologies on turtles in other parts of the world,

there is no way of validating the results. Therefore, the only way of commenting on

loggerhead nesting beach selection in other parts of the world is to replicate this

methodology in another part of the world where C. caretta nests.

Loggerhead sea turtles in the TTI appear to have chosen nesting beaches

according to the width of the beach, preferring relatively wide beaches. Secondarily, C.

caretta preferred wider beaches according to the shelliness (i.e. carbonate content) of the

beach. C. caretta also selected nest-sites in the TTI that are in or close to supra-littoral


From a management perspective, this means that biologists/conservationists in the

TTI could concentrate their efforts for loggerhead sea turtle conservation on beaches that

meet the criteria for favorable nesting beach selection. Beach width measurements could

be taken, and sand from the beaches could be shaken through a series of sieves in order to

assess the quality of a particular beach. Beach width could be assessed quickly and

easily; beaches that are on average greater than 8.5 meters in width should be selected.

After these wide beaches were selected, the resource manager could judge the amount of

shells (i.e. carbonate content) that remains in the three largest sand classes (0.991mm,

1.397mm, 1.981mm) after the sample has been shaken through the sieves. If there is a

large proportion of shells (> 9%) in the sample, then the manager has the ability to make

decisions about allocation of time, money and personnel to particular islands and/or

beaches, based on the quality of the particular nesting area for C. caretta.


This management option is presented as a possibility for application in the TTI.

However, at least one more year of data would need to be collected in order to determine

if the nesting beach selection observed in 1996 represented a trend in the behavior of C.

caretta. If the replicate study re-affirms the findings of this study, then the suggested

management scheme could be implemented on a trial basis.



Introduction and Background

The remoteness of the Ten Thousand Islands (TTI) provided a unique opportunity

to conduct biological research in a relatively undisturbed setting. However, this desirable

characteristic (i.e. remoteness) of the TTI limits access to the area, requiring large

expenditures of personnel, time and expenses to effectively monitor and protect sea turtle

nests. Numerous researchers have used aerial surveys to assess large areas of sea turtle

nesting beaches. These include an assessment of sea turtle nesting in the southeastern

United States (Shoop et al., 1985), North Carolina (Crouse, 1984), South Carolina

(Stancyk et al., 1979), Baja California (Fritts et al., 1982), Panama (Meylan and Meylan,

1985) and a survey of Pacific Mexico coupled with a new estimate of the world's

leatherback (Dermochelys coriacea) turtle population (Pritchard, 1982). Pritchard (1982)

used aerial surveys to assess an area of Mexico that resulted in discovery of a large,

previously unknown leatherback rookery. With this new information, Pritchard (1982)

was able to create a crude estimate of the world's population of mature, female

leatherback turtles. Sarti et al. (1996) followed up the work of Pritchard (1982) and other

researchers, and found this large Mexican conglomeration of nesting leatherbacks to be in

serious decline from previously documented numbers. Crouse (1984) conducted aerial

surveys in North Carolina in conjunction with ground truthing, allowing for a conversion


factor to create a seasonal estimate for the state of North Carolina. Unfortunately,

differential crawl retention on individual beaches precluded any season-long estimate.

Aerial surveys have been flown over the TTI area, but documentation of sea turtle

nesting was not the primary objective of the flights. Irvine et al. (1981) flew fixed-wing

manatee and dolphin surveys over the TTI, documenting incidental observations of sea

turtle crawls in the area. LeBuff and Hagan (1978) flew fixed-wing aerial surveys over

the beaches of Cape Romano and Marco Island, and reported missing approximately 50%

of the nests documented by surveyors on the ground. Fixed-wing aerial surveys by Davis

and Whiting (1977) over the Cape Sable region extended into the southern portion of the

Ten Thousand Islands archipelago, encompassing Pavilion and Rabbit Keys, included in

the current study. Kushlan (1986) repeated Davis and Whiting's surveys 8 years later,

and found results similar to the previous work.

The objective of this study was to develop a seasonal nesting estimate for the TTI

using an aerial survey coupled with ground truth counts. The estimate developed

depended upon two basic assumptions: nesting activity on the 8 ground-truth islands is

similar to activity on the other study islands and the nest-to-non-nesting emergence ratio

on other islands in the TTI is similar to that experienced on the 8 ground-truth islands.

Materials and Methods

Study Areas

The Ten Thousand Islands (TTI) extend roughly 40 kilometers from Marco Island

to Pavilion Key along Florida's southwest coast (Edwards, 1991). Located

approximately 25 kilometers southeast of Naples, these largely undeveloped islands are

utilized by the public year-round, primarily for fishing (Cheng, 1996). The climate is

subtropical with 110 to 160 centimeters of average annual precipitation, two-thirds of

which occurs between May and October (Cheng, 1996). Average annual temperature is


23 degrees Celsius, with occasional freezing temperatures in the winter and regular mid-

thirties in the summer (Cheng, 1996).

These islands are dominated by mangrove trees, with some invasion by Australian

pine (Casuarina equisetifolia). Beaches are characterized by sand, false coral, or oyster

shells, either individually or in some combination on each island. Characteristically,

beaches of these islands face the Gulf of Mexico from many different aspects and are

much shorter and narrower than the long, ribbon-like beaches on the east coast of Florida

or in Naples, Florida.

Sampling Techniques

From aerial photographs and fly-overs prior to initiation of the study, 29 keys in

the TTI were identified that could potentially support sea turtle nesting. From this

information, a flight plan was developed that allowed observation of all 29 keys for

sufficient numbers of days to create an overall seasonal nesting estimate for the TTI in

1996. To assess the accuracy of the aerial counts for the 29 islands, 8 of these islands

were chosen as ground truth islands for this study. Sea turtle nesting surveys were

conducted from May 13, 1996 until August 15, 1996 for the 8 ground-truth islands.

Ground surveys were conducted twice a week for each island, providing a seasonal

nesting total. This strategy was developed by the U.S. Fish and Wildlife Service due to

the remoteness of the TTI and lack of personnel.

Aerial surveys were conducted in a U.S. Coast Guard Sikorsky Jayhawk

helicopter at an altitude between 46.9 to 62.5 meters and a speed of 30 to 40 knots over

the surf zone. This placement of the helicopter afforded the best view of the narrow TTI

beaches. Flights were made on two successive mornings after new moons in June (June

19 and 20) and July (July 9 and 10) and on two successive mornings after full moons in

June (June 26 and 27) and July (July 25 and 26). The flight on June 19 was a training


flight, and the flight on June 27 was canceled due to mechanical problems with the


The full moon/new moon flight plan provided the greatest proportion of "fresh"

crawls (i.e. crawls made the night prior to aerial survey), since the tides following these

lunar cycles erase most old crawls from the beaches. In addition, all old crawls on the

ground-truth beaches were erased by raking prior to aerial survey. Crawls were ground-

truthed by assistants conducting daily surveys to provide the basis for a visibility index

and an accurate nest/false crawl ratio. Aerial surveys were begun between 0800 and

0900 h, usually concluding within two hours, beginning at Coon Key at the northwest end

of the study area, and ending at Pavilion Key at the extreme southeastern end. The

survey was conducted from the port side of the helicopter. This methodology was

adapted from research conducted by Hopkins-Murphy and Murphy (1983) in South



The following formula was used to calculate a nesting estimate for the entire

study area:

If the total ground and aerial counts for ground-truth islands are not equal, then a

correction factor has to be calculated:

G = ground count of crawls for ground-truth islands

Ag = aerial count of crawls for ground-truth islands

C = correction factor

= (Ag/G)

The correction factor can then be applied to the observed number of aerial counts

to create a corrected estimate (a) of the total number of aerial counts at all sites:

Ta = total aerial count of crawls at all sites

= total corrected # of aerial counts at all sites



If a correction factor is necessary then the total aerial estimate (a) would then be

substituted for the total aerial count (Ta) in subsequent calculations. The percentage of

total nesting activity on the ground-truth islands observed during the aerial survey is

calculated in order to obtain a seasonal nesting estimate:

Tg = total seasonal ground count of crawls for ground-truth islands

P = % of total seasonal crawl activity for ground-truth islands

= (G/Tg)

Once the percentage of the total seasonal nesting activity (P) has been calculated,

a total seasonal nesting estimate can be calculated:

S= total seasonal crawl estimate for TTI

= (Ta/P)

After the total seasonal nesting estimate has been calculated, the number of total

nests can be calculated based on the nest-to-total-crawl ratio for the 8 ground-truth


Ng = # of nests observed on ground-truth islands during aerial survey

f = total # of nests for all sites

= e(Ng/Tg)


A total of 26 crawls was observed in both ground and aerial surveys during the

study period on the ground-truth islands (Table 3-1). Although the total counts were

identical, the overall aerial survey failed to detect one crawl on one occasion and

recorded one when it did not exist on another occasion. In addition, the ground and aerial

counts for individual islands exhibited a much higher degree of variability when

compared to the total count for the entire survey period (Table 3-2).


Table 3-1. Ground and aerial counts of sea turtle crawls on 8 islands in TTI for June-July

Date Ground count Aerial count

6/20/96 5 5

6/26/96 8 8

7/9/96 5 4

7/10/96 4 4

7/25/96 3 4

7/26/96 1 1

TOTAL 26 26


Table 3-2. Aerial and ground counts on ground-truth islands for
each survey by island

Date Key Ground Air

6/20/96 Brush

6/26/96 Brush
Turtle I

7/9/96 Brush 7



0 -








7/10/96 Brush

7/25/96 Brush

7/26/96 Brush


Date Key Ground



| 0
- 1




The total number of crawls observed based on ground surveys conducted twice a

week throughout the 1996 nesting season (Figure 3-1), does not accurately represent the

levels of activity on the 8 ground truth islands on any given night. Figure 3-2 depicts the

estimated numbers of nests that were laid on the 8 ground truth islands, based on the

twice-a-week ground surveys. These nightly estimates were calculated by totaling the

number of nests laid during lapses in ground surveys and dividing that total by the

number of nights in the interval between ground surveys. The average obtained was then

applied to each of the nights that fell between the interval between ground surveys. This

graph (Figure 3-2) represents a trend in the amount of nightly nesting activity that

occurred on these islands during the 1996 season.

An estimate was derived utilizing the formulas described previously in this paper.

There were 26 crawls (G) counted on the 8 ground-truth islands during the aerial survey.

For these 8 ground-truth islands, there were a total of 422 crawls (Tg) in 1996. Thus, the

aerial observations accounted for 6.2% (P) of the seasonal crawl activity for these 8

islands. Since there were 118 aerial observations (Ta) at all sites, a total seasonal crawl

estimate (8) of 1903 crawls was derived. By calculating the % of nests for the 8 ground-

truth islands (Ng/Tg= 47.6%) for the entire nesting season, the total number of nests at all

sites (u = 906) was calculated. It is estimated that 906 nests and 997 non-nesting

emergences occurred on the 29 study islands during the 1996 nesting season.


Aerial surveys have been used in many sea turtle studies (Shoop et al., 1985;

Crouse, 1984; Stancyk et al., 1979; Fritts et al., 1982; Meylan and Meylan, 1985;

Pritchard, 1982) as a technique for covering large areas of coastline, reaching remote

areas and reducing costs of conducting season-long daily or nightly surveys. Aerial

surveys have been employed on several occasions in the area immediately adjacent to the



C 4

I---- 5

L I I *h' U) =


1 01 0 0 10 0N
I Q-

0 .

peiuesqo SIMR-- -o .Ioq fnN
42 -





In 0 I n

in 0

peAjesqo sIM Jo j joeqwunN


. ?
* a g
:! ' *
s i~
*S -a
' a >
a T
' B
a T-
' K s
' p: a
' R s
' R R
^ R Fi

^ R S
' K a
' ^ R
'. r:: H
' K a
' -= a
' R "
R ^
ft ?
' ft s
' 1= ?
' K S

R g
is o>
' R a

ft T

a a
ft S
3 K
a fc
a K
a S
a S
a a
a a
a S
a a
a 5
a s
a e
a S
a S
a ;
a ?
a s
a ^
a g
a *
a r-
a a>
a in
a *
a n
a r-
a f
* 8
*B S
B ft
5) H
.;> a
a R
a R
a *

a &
a S
a s
a ;
a p





































TTI (Davis and Whiting, 1977; LeBuff, Jr. and Hagan, 1978; Irvine, 1981; Kushlan,

1986), reporting varying degrees of success at documenting sea turtle activity.

The most efficient means of gathering data on marine turtle nesting activity over a

large area appears to be aerial survey (Hopkins-Murphy and Murphy, 1983). But, aerial

survey is only useful if it is effective, that is, producing accurate and reproducible

estimates (Hopkins-Murphy and Murphy, 1983). There are a number of factors that

influence the effectiveness of aerial survey nesting estimates. LeBuff and Hagan (1978)

reported that strong westerly or northwesterly winds in conjunction with high tides often

caused beach erosion and shifting sands, and that severe weather left marine vegetation

upon beaches that concealed crawls. Hopkins-Murphy and Murphy (1983) discuss the

importance of aerial sampling parameters associated with aerial surveys such as: speed of

the aircraft, height of the aircraft, position of the aircraft, angle of view (beach), skill of

the pilot and observer experience on estimator accuracy. Other factors that could

potentially bias aerial estimates, include: amount of turtle activity, the type of beach, the

amount of sunlight, human activity on the beach, location of crawls on the beach and in

the case of the TTI, the ordinal aspect of the islands. LeBuff and Hagan (1978) noted

that crawls in close proximity to each other or crawls partially hidden caused problems

when attempting to assess nesting activity from the air. Not all factors can be controlled

or accounted for in creating nesting estimates.

Crouse (1984) identified one of the inherent difficulties with creating estimates

from aerial survey, is that a limited number of aerial surveys are used to extrapolate to a

total seasonal nesting estimate. Crouse (1984) is critical of the use of a conversion factor

as a means of creating nesting estimates via aerial survey. She noted that variability of

crawl retention and visibility from beach to beach adversely affects conversion indices.

These factors may vary from beach to beach depending upon characteristics of the

nesting substrate and prevailing tides (Crouse, 1984). She also stated that weather

patterns could affect crawl retention, both geographically and temporally, and that


changes in beach profile due to erosion or accretion may affect crawl retention over time

(Crouse, 1984).

The TTI are used minimally in the summer months by humans, therefore, that

impact on crawl retention is minimal. In addition weather patterns and the type of

nesting substrate can cause differential crawl retention on TTI beaches temporally; these

factors can be controlled by only documenting "fresh" crawls. Recognizing these

limitations of aerial survey, efforts to reduce potential sources of bias and make

corrections prior to the beginning of the flight schedule, were taken. For example, a test

flight was conducted where fine-tuning of the aerial survey technique to the TTI took

place (i.e. adjusting speed, altitude and angle of observation of the helicopter). Another

problem associated with aerial survey is detecting all crawl activity after nights of high

nesting activity or storm activity. One of the methods of dealing with this problem is the

creation of a conversion factor. Although the overall accuracy of a seasonal estimate will

suffer if aerial and ground counts are not in fairly close agreement, the conversion factor

still allows for an estimate to be made. The accuracy of this sort of estimate would be

debatable, although it would provide managers with a good idea of the trends in sea turtle

nesting activity for their study area. The usefulness of such data cannot be discounted.

Since the survey technique suggested here has numerous limitations, it is

necessary to describe methods that could improve upon the protocol used for the 1996

season. One of the key limitations of this particular study is the fact that we conducted 6

flights. After assessing the data, it was determined that in order to accurately create a

nesting estimate that is worthy of application for threatened and endangered species,

several years of flights would need to be flown, with the flights occurring no less than a

daily basis for at least the first year. After collecting this data, it would be possible to

determine the minimum number of flights needed to create estimates that can limit the

variability in the data. By adding many more flights to the protocol, a more accurate

picture of sea turtle nesting trends could potentially be developed for the TTI.


Although the technique applied in the TTI for the 1996 nesting season is not the

answer to developing an accurate nesting estimate for the TTI, it did provide answers

about the difficulties associated with aerial survey in the TTI. Therefore, the technique

served as the first step in potential development of a method for accurately assessing sea

turtle nesting activity in the TTI. Furthermore, the nesting estimate suggested in this

paper should not be viewed as a hard estimate of the total activity in the TTI. In fact, due

to the variability of crawls observed, the estimate should be cautiously viewed as an

indicator of the amount of sea turtle nesting that is occurring in the TTI. This preliminary

research should be expanded upon by implementing the changes to the methodology

suggested in this paper.

Davis and Whiting (1977), in work that is now over 20 years old, found an

estimated 1553 loggerhead nests in 1972 and 919 loggerhead nests in 1973 for nesting

beaches stretching from Cape Sable northwest to Morman Key, just southeast of the

furthest extent (Pavilion Key) of my study area. Kushlan (1986) conducting aerial

surveys in the same area and employing the same methodology as Davis and Whiting

(1977) found 1362 nests in 1980. Taking into account the estimate I made for the TTI,

this portion of coastline, long thought to be fairly insignificant in terms of overall nesting

effort by sea turtles, could potentially host between two to three thousand sea turtle nests

in a given year. These nesting numbers identify the significance of this coastline as a

high use area for sea turtles, particularly the loggerhead sea turtle. Furthermore, I was

unable to differentiate crawls by species of sea turtle from the aerial survey, therefore, it

is not possible to say how many of the nests in the TTI are from green turtles (Chelonia

mydas), which are known to nest in the islands.

This technique may be used as an indicator of nesting activity in the TTI

although, there are several additional changes in the methodology used in this study that

could potentially increase the accuracy of aerial surveys. The use of two observers is a

strategy used in the work conducted by Murphy and Hopkins-Murphy (1983) in South


Carolina. They found no significant difference in the number of crawl identifications

between two separate observers. By using two observers, like Murphy and Hopkins-

Murphy (1983), a basic mark-recapture study could be created, and the results could be

subjected to statistical analyses to lend strength to any aerial estimates. Since aerial

observations are conducted by helicopter in the TTI, there is greater flexibility with

regard to how flights could be conducted. It may be possible to safely reduce the speed at

which the helicopter flies, based on the discretion of the pilot. It may be beneficial to

hover over beaches at the discretion of the observerss. Hovering might allow for more

accurate documentation of crawl activity. The observers) may request additional fly-bys

of beaches in order to get more accurate counts of crawl activity.

The methodology suggested here is imperfect. Although the estimates for 1996

appear to be sound, these results cannot predict the accuracy of aerial surveys for future

years in the TTI. Therefore, when implementing this strategy, the conclusions drawn

from the results presented should be considered cautiously.


Aerial survey, in conjunction with seasonal ground surveys, appears to be a useful

tool for evaluation of sea turtle activity in the Ten Thousand Islands. In order to develop

low-cost methods for assessing trends in sea turtle nesting in the TTI, natural resource

managers may consider annual aerial surveys with partial ground-truth counts punctuated

with aerial survey corrected by season-long ground surveys. Detailed information on the

accuracy of aerial survey, as well as the trends in sea turtle nesting in the TTI may allow

for the development of a lower-cost method for evaluating nesting activity. However,

before any protocol can be established, at least several more years of aerial surveys need

to be conducted, before nesting trends can be more clearly understood.



Introduction and Background

Raccoons (Procyon lotor) have been the subject of considerable study pertaining

to various aspects of behavior, feeding ecology, and population dynamics. In addition,

several studies have dealt directly with the Ten Thousand Islands raccoon (Procyon lotor

marinus). This subspecies was identified by Goldman (1950), and later confirmed by

Hall and Kelson (1959). This spurred further study, including descriptions of

morphometric characteristics (Bigler et al, 1977), and population characteristics of P. 1.

marinus (Bigler et al, 1981).

Raccoons are effective predators of sea turtle nests in Florida (Ratnaswamy, 1995;

McMurtray, 1986; Lewis et al., 1994; Kushlan, 1986; Klukas, 1967; Davis & Whiting,

1977), Georgia (Anderson, 1981), and South Carolina (Stancyk et al., 1980; Andre &

West, 1979; Hopkins & Murphy, 1980; Hopkins et al., 1977). Raccoon depredation rates

of sea turtle nests ranged from 49-87% annually at Cape Sable, FL (Davis & Whiting,

1977) and from 59-85% annually from 1991-1994 for Ten Thousand Islands National

Wildlife Refuge (TTINWR), FL (Garmestani, 1995). Raccoons have been identified as

the primary predator of sea turtle nests with humans, fire ants, ghost crabs and bobcats as

possible secondary predators in TTINWR (L. Richardson, pers. comm.).

Due to the imperiled status of sea turtles, various means of decreasing raccoon

depredation of sea turtle nests have been explored. Numerous studies have been


conducted to evaluate the effectiveness of conditioned taste aversion (CTA) and its

impacts on consumption of eggs by raccoons. Linhart et al. (1991) formulated baits that

were readily taken by raccoons, tested the baits in the field, and found that 31-53% of the

baits were taken by non-target species (Linhart et al., 1994). Conover (1990) reported

that emetine dihydrochloride was determined to have some effectiveness in reducing egg

depredation. Nicolaus (1987) and Semel & Nicolaus (1992) found CTA using estrogen-

injected eggs can be an effective strategy to reduce raccoon depredation of eggs.

However, Nicolaus et al (1989), found mixed results when treating eggs with estrogen,

and researchers in South Carolina found CTA using lithium chloride to be completely

ineffective at reducing depredation of sea turtle nests (Hopkins & Murphy, 1982). One

study conducted in South Dakota concluded that raccoons could be baited into ingesting

chemosterilants, thereby inhibiting reproduction (Nelson, 1970). CTA techniques were

rejected by TTINWR because of their mixed results, combined with the likelihood of

ingestion by non-target species.

Other methods of sea turtle nest protection have been attempted. McMurtray

(1986) concluded that the most effective means of turtle nest protection was flat screens

left on nests until hatchlings emerged. Ratnaswamy (1995) supported the claim that

screens are the most effective form of nest protection from predators, and claimed that

CTA and predator removal were ineffective. Several researchers argue the effectiveness

of nest transplantation as an effective means of decreasing turtle egg mortality (Andre &

West, 1979; Stancyk et al., 1980). Hopkins and Murphy (1982) concluded that the most

effective form of nest protection includes nest transplantation and predator removal,

thereby decreasing chances of inundation and depredation. This conclusion was

supported by the findings of researchers at St. Vincent's National Wildlife Refuge (Lewis

et al., 1994). However, the U.S. Fish and Wildlife Service has recently determined that

nest transplantation should only be considered when a nest is threatened by potential

inundation (MacPherson, personal comm.). Klukas (1967) reported a 25% reduction in


nest destruction using predator elimination as a management tool on Cape Sable in

Everglades National Park (ENP).

Raccoons depredated nests on the eight continuously surveyed islands in

TTINWR at varying rates (Garmestani, 1995). Although the literature indicated that

other forms of nesting management may be more effective, particularly screening in

conjunction with predator removal (Hopkins & Murphy, 1982; Lewis et al., 1994), lack

of personnel and the remoteness of TTINWR precluded any attempt at labor-intensive or

season-long nesting management. Due to these constraints, the effectiveness of short-

term raccoon control was assessed as a possible management strategy for the Ten

Thousand Islands.

Materials and Methods

Study Areas

Study areas were 4 low-relief mangrove islands located within TTINWR and the

Cape Romano-Ten Thousand Islands Aquatic Preserve in southwest Florida. This rich

estuary, which supports breeding and feeding grounds for a multitude of species, consists

of a labyrinth of tidal creeks, bays, and passes. The 4 study islands were dominated by

red (Rhizophora mangle), black (Avicennia germinans), and white (Laguncularia

racemosa) mangrove trees, interspersed with sandy and/or shelly beaches. These beaches

are of varying lengths (50m-815m) and widths (1.4m-18.3m), and support beach

vegetation that varies with the substrate and elevation of the particular island and/or

beach. Study islands were of varying sizes, in close proximity (< 2 km) and

interconnected by the observed motility of raccoons: Panther (54.8 ha), White Horse

(39.3 ha), Hog (26.7 ha), and Gullivan (19.6 ha). These 4 islands (Panther, Hog, White


Horse, and Gullivan Keys) were known to support sea turtle nesting and to suffer varying

rates of raccoon depredation of sea turtle nests (Garmestani, 1995). White Horse was the

site of 75 documented sea turtle nests from 1991 through 1994, and had an average

depredation rate of 46% for those four years; Gullivan was the site of 124 documented

sea turtle nests from 1991-1994, with a 91% depredation rate for that time period; Hog

Key had a total of 21 documented sea turtle nests for the years 1991-1994, with an

average depredation rate of 72%.

Sampling Techniques
Live traps (91.44cm X 27.94cm X 27.94cm [2-door]; 81.28cm X 30.48cm X

25.4cm [1-door]; Havahart, Inc., Lititz, Pennsylvania, USA; 121.92cm X 30.48cm X

30.48cm [1-door]; custom traps, South Carolina Wildlife and Marine Resources

Department, Georgetown, South Carolina, USA) were placed in a randomly spaced array

in the vicinity of nesting beaches, utilizing vegetation as camouflage. Raccoons were

attracted into live traps using a variety of baits, including sardines, a sardine/cat food

mixture, mullet, marshmallows, and chicken eggs coated with fish oil. After a raccoon

was captured, it was compressed within the live trap using either a homemade plunger or

stakes. The animal was then injected intramuscularly with 8-10 mg/kg of ketamine (100

mg/ml; Fort Dodge Laboratories, Inc., Fort Dodge, Iowa, USA). Once the animal was

anesthetized, one of two things then occurred: 1) if the animal was part of the mark-

recapture study it received a Monel (National Band and Tag Co., Newport, KY, USA) ear

tag in each ear and a Passive Integrated Transponder (PIT) (Avid Co., Norco, CA, USA)

which was placed in the scruff of the neck, and a 4 cc sample of blood (1995) was taken

from the heart and placed on ice; 2) if the animal was part of the removal study, blood

was drawn (1995), and the animal euthanized with an intercardiac injection of euthasol


(Delmarvia Laboratories, Inc., Midlothian, VA, USA). Next, numerous aspects of the

anatomy of the raccoon were measured and evaluated, including: sex, testes lengths,

baculum length, evidence of lactation, weight, age (Grau et al., 1970), canine lengths,

total length (head-body and tail), right hind-foot, right ear, length of vibrissae, neck, head

and chest circumference, pelage, and physical condition. Euthanized raccoons were then

buried in three-foot deep holes on the beach. Tagged raccoons were released by placing

them in the shade, covering their eyes with horseshoe crab shells to alleviate the potential

damage from sunlight. Raccoons were captured and removed on Panther Key in January,

February, May, June, and July for 1995, and in January for 1996; sampled by mark-

recapture on White Horse Key in March of 1995 and 1996, and on Hog Key in April of

1995 and 1996; Gullivan Key was sampled by mark-recapture in March and April of

1995, and removal in March and April of 1996.

1995 Field Season
On January 11th, 1995, raccoon trapping commenced on Panther Key in the

TTINWR. A 100 mile round trip drive, and a 12 mile round trip boat ride were required

each day in order to arrive at the field site in the TTINWR. Initially, 12 Havahart one-

door live traps were used to capture raccoons on the west beach of Panther Key.

However, after compiling 8 nights using only the one-door traps, 12 Havahart two-door

live traps were deployed on the east beach of Panther Key for an additional 15 nights, for

a total of 23 nights on Panther Key. Only one door was opened on the two-door traps

each night. Beginning May 1, 12 one-door and 10 custom live traps on loan from South

Carolina Wildlife and Marine Resources Department (SCWMRD) were re-set on Panther

Key. After 5 nights of trapping, the 10 custom traps were removed, and the remaining 12

live traps were run until July 14, 1995. Raccoons were enticed into traps with a variety


of baits including sardines, a sardine/cat food mixture, marshmallows and mullet.

Beginning March 3, 1995, raccoon trapping commenced on White Horse Key. A

total of 33 traps was initially deployed on White Horse, including 12 Havahart one-door

traps, 11 Havahart two-door traps, and 12 custom traps on loan from Phillip Wilkinson of

SCWMRD. Six of the two-door traps were closed after two nights of trapping, because

of time constraints. All raccoons captured were marked and released. Mullet was used

exclusively to bait traps on White Horse Key.

Beginning March 28, 1995, raccoon trapping commenced on Gullivan Key.

Twenty-two traps were set on Gullivan, including 12 Havahart one-door traps, and 10 of

the custom traps on loan from SCWMRD. Mullet was used exclusively as bait, and all

raccoons were tagged and released.

Beginning April 17, 1995, 7 one-door Havahart traps were set on Hog Key. This

number was upgraded the following day, by adding 5 more Havahart one-door traps, and

10 of the custom traps on loan from SC Wildlife. A total of 8 nights were run on Hog

Key from April 17 through April 28. Mullet and marshmallows were used as bait, and all

captured raccoons were tagged and released.

1996 Field Season

Raccoon trapping for the 1996 field season was based upon the rate of first-time

captures for the 1995 field season (Table 4-1).


Table 4-1. Initial raccoon capture rates for TTINWR
(1995) initial trapping regime

Island new captures # of nights %new captures

Panther 13 8 100

White Horse 18 5 100

Gullivan 8 6 100

Hog 10 5 100

In addition, the study methodology was further altered from the 1995 field season

by adding Gullivan Key as a removal island, along with Panther Key. Gullivan Key was

selected due to its distance from Panther Key and high depredation rate. Also, during

1996, mark-recapture efforts were conducted on White Horse and Hog Keys, in order to

accumulate data on changes in raccoon population characteristics of the two islands.

Raccoon trapping commenced on Panther Key beginning on January 18, 1996,

running traps for 8 nights. Twenty-one traps were deployed on the island, including 11

Havahart one-door and 10 Havahart two-door live traps. Each of these traps was baited

with chicken eggs scented with fish oil. This bait was devised as a method of targeting

egg-depredating raccoons. All raccoons captured were euthanized and buried on site.

On March 5, 1996, trapping began on White Horse Key, running 5 nights.

Twenty traps were originally deployed on White Horse, including 20 Havahart live traps

(e.g. ten one-door, ten two-door). But, after two nights, one of the two-door traps was

stolen, so the remaining three nights were run with 19 traps. All traps were baited with

the chicken egg/fish oil bait, and all raccoons were marked and released.

On March 30, 1996, trapping commenced on Gullivan Key, running 6 nights.

Nineteen Havahart live traps were deployed on Gullivan, including 10 one-door and 9


two-door traps. All traps were baited with egg/oil, and all raccoons were eliminated and

buried on site.

On April 18, 1996, trapping was initiated on Hog Key, running 5 nights. Eighteen

Havahart live traps were deployed on Hog Key, including 10 one-door and 8 two-door

traps. All traps were baited with egg/oil, and all raccoons were marked and released.


1995 Field Season

Panther Key (PK), an island continuously surveyed for the past 5 years at

TTINWR, was chosen as a test island for raccoon removal because of its high incidence

of sea turtle nests and raccoon depredation. In addition to the limited success with live

trapping, baiting the beach with piles of mullet was completely unsuccessful for spot-

lighting and shooting of raccoons. There were a total of 15 captures on Panther Key,

including one escape. So, a total of 14 raccoons was removed from Panther Key in 750

trap nights for the 1995 season. Traps were reset on Panther Key in May to determine

whether raccoons will move from adjacent islands to exploit an open niche and a plentiful

seasonal food resource (i.e. sea turtle nests). However, only 1 unmarked raccoon was

captured after the start of sea turtle nesting. Raccoon tracks were observed on numerous

occasions on Panther Key during turtle nesting season, but no depredation occurred in

1995 (Table 4-2).


Table 4-2. Sea turtle nesting and depredation on Panther Key
for the 1991-1995 nesting seasons

Year total nests # depredated % depredated

1991 63 63 100

1992 42 40 95

1993 36 29 81

1994 38 29 76

1995* 41 0 0


Even with the elimination of raccoon depredation as a factor in sea turtle nest

mortality for Panther Key in 1995, overall hatch success did not significantly improve

due to the high rate of inundation of nests (Table 4-3). This high rate of inundation was

due to an extreme aberration in the occurrence of tropical storms and hurricanes for the

1995 storm season.

Table 4-3. Sea turtle hatch success on Panther Key for the 1995 nesting season

In addition to the animals trapped on Panther Key, raccoons were marked and

released on Gullivan (n=8), White Horse (n=18), and Hog (n=10) Keys. All 3 of these

islands are in close proximity to Panther Key, and all have turtle nesting beaches that

have been documented for the past 5 years.

Ten nights of trapping were run on White Horse Key with 18 initial captures, and

a total of 30 captures for White Horse Key. Raccoons were tagged and released,


accounting for the 12 recaptures (Table 4-4). A total of 12 nights of trapping was run on

Gullivan Key, with 8 initial captures and 1 recapture (Table 4-4). Ten initial captures

were recorded on Hog Key in 8 nights of trapping (Table 4-4).

Table 4-4. Results of raccoon trapping activity on the four study islands for the
1995 trapping season

# of trap Initial Total # of captures % of
Island nights captures recaptures per night successful
Panther 750 14 0 0.33 2.0
White 286 18 12 3.00 10.5
Gullivan 264 8 1 0.75 3.4
Hog 161 10 1 1.38 6.8
TOTAL 1461 50 14 0.67 3.4

Of these 50 captures, 36 were males, ranging in size from 1.4 kg to 4.8 kg, with mean

weight and standard error of 3.55 +/- 0.136 kg. Fourteen were females ranging in size

from 1.4 kg to 3.6 kg, with mean weight and standard error of 2.83 +/- 0.151 kg. Age

class distribution for males (n=36) captured in 1995 was fairly even (Figure 4-1), but age

class distribution was slightly skewed towards young (Class 1) and middle-aged (Class 3)

females (n=14) (Figure 4-1). Although a population estimate could not be created,

raccoon densities were calculated for the period of trapping on a particular island (Table






( O
(0 in



e o

(0 C

sOo0e NO #

SU0033Rej Jo #











Table 4-5. Raccoon densities during trapping periods
for 1995

Island # of raccoons captured Density during capture
period (per ha.)
Panther 14 0.26
White Horse 18 0.46
Gullivan 8 0.37
Hog 10 0.41
TOTAL 50 0.36


1996 Field Season

Raccoon removal was conducted again in 1996 on Panther Key. There were a

total of 2 captures in 168 trap nights. No depredation was observed in 1996, and I

assume that this is attributable to the removal of raccoons on Panther Key (Table 4-6).

Table 4-6. Number of depredated nests for Panther Key

Year # of nests # depredated % depredated

1996 61 0 0

The hatch success for sea turtles on Panther Key drastically improved for the

1996 field season when compared to hatch rates for the 1995 field season (Table 4-7). I

attribute this drastic increase to the effectiveness of raccoon removal and the lower

incidence of devastating storms (i.e. hurricanes and tropical storms) in 1996.

Table 4-7. Hatch success for Panther Key (1996)

# of nests # hatched # of eggs # of emerged # unhatched
evaluated hatchlings
61 58 (95%) 5809 4875 (84%) 934 (16%)

In addition to the animals trapped on Panther Key, raccoons were removed on

Gullivan Key (n=l), and mark-recapture was conducted on White Horse (n=15) and Hog

(n=5) Keys.

There were 15 initial captures and 1 recapture for a total of 16 captures on White

Horse for 1996. Of these 15 captures, 6 were captured on White Horse in 1995 and 1 was

tagged on Gullivan in 1995. This male raccoon is the only record of movement between


islands based on my mark-recapture data. Therefore, 8 "new" raccoons were trapped and

tagged on White Horse Key in 1996. All raccoons were tagged and released.

There was 1 capture on Gullivan for 1996. This individual was not one of the 8

raccoons tagged on Gullivan in 1995. The individual was eliminated and buried on-site.

The raccoon removal method was completely ineffective for Gullivan in 1996. Raccoons

depredated nests at a similar rate in 1995, when no removal was attempted (Table 4-8).

Table 4-8. Number of depredated nests for Gullivan Key

Year # of nests # depredated % depredated

1995 36 34 94

1996* 33 32 97


Five nights of trapping were run on Hog Key, from April 18-22, 1996. There

were a total of 6 captures on Hog, including 1 recapture. Of these 5 initial captures, 2

were captured in 1995, and 3 were "new" raccoons. All raccoons were marked and


Of these 23 captures, 13 were males, ranging in size from 2.1 kg to 4.7 kg, with

mean weight and standard error of 3.45 +/- 0.182 kg, and 10 were females ranging in size

from 1.6 kg to 3.6 kg, with mean weight and standard error of 2.64 +/- 0.208 kg. Age

class distribution for females (n= 10) captured in 1996 was fairly even (Figure 4-2), but

age class distribution was slightly skewed towards young (Class 1) and middle-aged

(Class 3) males (n=13) (Figure 4-2). Trapping data are summarized in Table 4-9.



2 u-I-.

m 0

S #


Table 4-9. Raccoon activity in TTINWR (1996)

# of trap Initial Total # of captures % of
Island nights captures recaptures per night successful
Panther 168 2 0 0.25 1.2
White 97 15 1 3.00 16.5
Gullivan 114 1 0 0.17 0.8
Hog 108 5 1 1.2 5.6
TOTAL 487 23 2 1.04 5.1


Although a population estimate could not be created for 1996, raccoon densities during

trapping periods are summed in Table 4-10.

Table 4-10. Raccoon densities during trapping periods
for 1996

Island # of raccoons captured Density during capture
period (per ha.)
Panther 2 0.04
White Horse 15 0.38
Gullivan 1 0.05
Hog 5 0.04
TOTAL 23 0.16


The data I collected provide a closer look at the Ten Thousand Islands raccoon

(Procyon lotor marinus), which has been studied by several other researchers (Nelson,

1930; Bigler et al., 1977; Bigler et al., 1977). Although I was able to capture quite a few

raccoons (1995= 50; 1996= 23), I was unable to get enough recaptures to run any sort of

population model, such as the examples suggested in Hallett et al. (1991). Therefore, I

was unable to create a population estimate for the 4 islands I surveyed.

However, I was able to collect information pertaining to the natural history of P. 1.

marinus. For instance, in 1995, I caught more males (n= 36) than females (n= 14). It is

possible that this male to female ratio is due to a sex-biased response to live traps as has

been suggested by Gehrt and Fritzell (1996), who found that males were captured more

frequently during all of their trapping periods. Yet, in 1996, although I have a small

sample size, I caught raccoons at nearly a 1:1 ratio (males n= 13, females n= 10). This


ratio (1:1) is what is to be expected in nature, although the exploration of this sex-bias

theory merits a closer look.

Depredation of sea turtle nests on Panther Key was reduced to zero for the two-

year study period. I assume this drastic decrease in depredation on Panther Key was due

to raccoon control. This finding is inconclusive, but unconnected research conducted on

St. Catherine's Island in Georgia found that there was no shift in raccoon home ranges or

centers of activity with regard to sea turtle nesting (Anderson, 1981). Wilkinson (pers.

comm.) observed in South Carolina that raccoons not familiar with sea turtle nests may

forage on beaches without taking eggs obviously available for exploitation.

Finally, when considering the poor trap success for the 1996 field season, the

intangible factor of trap-smart raccoons must be taken into account. Short of conducting a

radio-telemetry study, it would be difficult to quantify what percentage of the TTI

raccoons are trap-smart, although there is evidence to prove their persistence in the

population. For example, low-capture rates on Hog Key and to a lesser degree White

Horse Key were experienced in 1996 after those islands were subjected to trapping in

1995. But the greatest evidence is provided by the abysmal trap success on Gullivan Key

coupled with extremely high nest depredation for the island (see Table 4-8 and Figure 4-

3). From the limited mark-recapture data, only 1 raccoon moved from 1 island to

another. This evidence and the fact that low numbers of raccoons re-populated Panther

Key, suggest that raccoon removal could be an effective sea turtle management policy for

the Ten Thousand Islands (Figure 4-4).


The Ten Thousand Islands raccoon has proven to be the single greatest threat to

sea turtle hatch success in the mangrove forests of southwest Florida. Therefore, any sea

turtle management strategy should include live trapping on islands that have not

previously been exposed to trapping, and explore the possibility of utilizing leg-hold




40 -
35 -
30 -
25 --Nests Laid
20 -.- Depredated
15 -
10 -
1991 1992 1993 1994 1995 1996


Figure 4-3. Raccoon depredation rates on Gullivan Key, 1991-1996.
Raccoon control program in effect this year

. 50
Z 40
z 230
a 20


1991 1992 1993 1994 1995 1996
1991 1992 1993 1994 1995 1996


Figure 4-4. Raccoon depredation rates on Panther Key, 1991-1996.
Raccoon control program in effect these years


traps and shooting as methods of removal for islands that have been exposed to live

trapping. In addition, removal efforts should be concentrated on islands with high

numbers (>10) of sea turtle nests and/or with the highest probability of hatch success, due

to slope, sediment type, etc. These islands include: B, Gullivan, White Horse and Hog

Keys. Finally, hatch success will be dramatically affected by weather patterns for the

area. High hatch success cannot be expected if numerous storms have direct and indirect

effects on the islands as in 1995, a year of heavy storm activity, and 1996, a year of

minimal storm activity (see Table 4-3 and 4-7).

It would be beneficial to know about the movement patterns and re-population

trends for the TTI raccoons. Conducting radio-telemetry on P.I. marinus may provide

answers that allow for a long-term raccoon management strategy to be employed. By

conducting this study, information could be gathered that may indicate how often raccoon

control measures would need to be implemented. It would also be necessary to evaluate

the population genetics of each islands' raccoon population. Understanding the

relatedness of raccoons on each of the islands, coupled with radio-telemetry, would

provide a more complete understanding of the raccoon populations and their movement

patterns in the TTI, and allow for a more comprehensive raccoon management strategy.




It appears that loggerhead sea turtles select nesting beaches according to certain

criteria of beaches in the TTI. The conclusions reached about loggerheads in the TTI

may not be true of C. caretta in other parts of the world. Other factors may play a larger

part in the decision-making process in sea turtles. For instance, beach lighting, the

presence of humans, sea walls and re-nourished beaches could potentially have a greater

effect on nesting beach and nest-site selection, particularly in inhabited areas, than the

factors I analyzed for the TTI. However, the most important element of my study is that

the nesting habitat in the TTI is relatively unspoiled. This means that the nesting

preferences exhibited by loggerheads can be assumed to be based on cues inherent to the


1) Aerial surveys can be effective in assessing sea turtle activity in the TTI,

reducing expenditures of time, personnel and expenses. Obviously aerial surveys

need to be conducted in a helicopter due to the configuration of the islands, but

with continued cooperation of the U.S. Coast Guard, an effective, low-cost survey

plan could be implemented for the region. The use of aerial survey also provided

the first comprehensive evaluation of sea turtle nesting activity for the TTI. This

season-long estimate may fluctuate with sea turtle nesting cycles, and changes to

the survey technique, but this indicator of sea turtle nesting activity in the TTI

should be considered a minimum nesting estimate for the region when you take


into account the five islands that were not included in this study, including one

island (Cape Romano) that is known to support from 50 to 100 nests per season

(Kraus, pers. comm.).

2) The removal of raccoons is a potential management strategy for improvement

of sea turtle hatch success. Even if depredation is reduced significantly, hatch

success will not necessarily improve in years of heavy storm activity in the area.

Therefore, raccoon removal is simply a method for decreasing depredation of sea

turtle nests and indirectly increasing hatch success. In addition, the raccoon

research provided some basic data on the status of Procyon lotor marinus in the


3) Other research addressing the effects of exotic invasive species (i.e. Australian

pine and the red imported fire ant) into the TTI needs to be undertaken to

understand their impacts on sea turtles, as well as other native flora and fauna in

the archipelago. In addition, personal observations of Kemp's Ridley

(Lepidochelys kempii) sea turtles and juvenile green turtles in the area warrants

investigation of this area as a potential feeding ground for these two species.



Addison, D.A. 1994. Caretta caretta (loggerhead sea turtle) nesting. Herpetological
Review 25: 63.

Allard, M.W., M.M. Miyamoto, K.A. Bjorndal, A.B. Bolten and B.W. Bowen. 1994.
Support for natal homing in green turtles from mitochondrial DNA sequences.
Copeia 1994: 34-41.

Anderson, S. 1981. The raccoon (Procyon lotor) on St. Catherines Island, Georgia: 7.
Nesting sea turtles and foraging raccoons. American Museum Novitates 2713: 1-

Andre, J. B., and L. West. 1981. Nesting and management of the Atlantic loggerhead,
Caretta caretta caretta (Linnaeus) (Testudines: Cheloniidae) on Cape Island,
South Carolina, in 1979. Brimleyana No. 6: 73-82.

Bayliss, P. 1986. Factors affecting aerial surveys of marine fauna, and their relationship
to a census of dugongs in the coastal waters of the Northern Territory. Australian
Wildlife Resources 13: 27-37.

Bigler, W.J., A.S. Johnson, and G.L. Hoff. 1977. Morphometric characteristics of the
Ten Thousand Islands raccoon. Florida Scientist 40: 261-262.

Bigler, W.J., G.L. Hoff, and A.S. Johnson. 1981. Population characteristics of Procyon
lotor marinus in estuarine mangrove swamps of southern Florida. Florida
Scientist 44: 151-157.

Bjorndal, K.A., A. Carr, A.B. Meylan and J.A. Mortimer. 1985. Reproductive biology
of the hawksbill (Eretmochelys imbricata) at Tortuguero, Costa Rica, with notes
on the ecology of the species in the Caribbean. Biological Conservation 34: 353-

Bjorndal, K.A. and A.B. Bolten. 1992. Spatial distribution of green turtle (Chelonia
mydas) nests at Tortuguero, Costa Rica. Copeia 1992: 45-53.

Bowen, B.W., A.B. Meylan, J.P. Ross, C.J. Limpus, G.H. Balazs and J.C. Avise. 1992.
Global population structure and natural history of the green turtle (Chelonia
mydas) in terms of matriarchal phylogeny. Evolution 46: 865-881.


Bowen, B.W., J.C. Avise, J.I. Richardson, A.B. Meylan, D. Margaritoulis and S.
Hopkins-Murphy. 1993. Population structure of the loggerhead turtle (Caretta
caretta) in the northwest Atlantic Ocean and Mediterranean Sea. Conservation
Biology 7: 834-844.

Bowen, B.W. 1995. Molecular genetic studies of marine turtles. In Biology and
Conservation of Sea Turtles, K.A. Bjorndal, Ed.-Rev. Ed., pp. 585-587.

Broderick, D., C. Moritz, J.D. Miller, M. Guinea, R.J. Prince and C.J. Limpus. 1994.
Genetic studies of the hawksbill turtle: evidence for multiple stocks and mixed
feeding grounds in Australian waters. Pacific Conservation Biology 1: 121-131.

Bustard, H.R. 1972. Australian Sea Turtles. London: Collins and Sons.

Bustard, H.R. 1979. Population dynamics of sea turtles. In Turtles: Perspectives and
research: Marion Harless & Henry Morlock, editors; 523-540, 1979.

Camhi, M.D. 1993. The role of nest site selection in loggerhead sea turtle (Caretta
caretta) nest success and sex ratio control. Ph.D. Dissertation, Rutgers
University, New Brunswick, NJ. 255pp.

Carr, A. and M.H. Carr. 1972. Site fixity in the Caribbean green turtle. Ecology 53:

Carr, A., M.H. Carr and A.B. Meylan. 1978. The ecology and migration of sea turtles. 7.
The west Caribbean turtle colony. Bulletin of the American Museum of Natural
History 162: 4-46.

Chaloupka, M.Y. and J.A. Musick. 1997. Age, growth, and population dynamics. In
The Biology of Sea Turtles, P.L. Lutz and J.A. Musick, Eds., pp. 233-276.

Cheng, E. 1996. Marine turtle report. Ten Thousand Islands National Wildlife Refuge,
Naples, FL. Unpublished. Mimeo. 27pp.

Christens, E. and J.R. Bider. 1987. Nesting activity and hatching success of the painted
turtle (Chrysemys picta marginata) in southwestern Quebec. Herpetologica 43:

Christiansen, J.L. and B.J. Gallaway. 1984. Raccoon removal, nesting success, and
hatchling emergence in Iowa turtles special reference to Kinosternonflavescens
(Kinosternidae). The Southwestern Naturalist 29: 343-348.

Conover, M.R. 1990. Reducing mammalian predation on eggs by using a conditioned
taste aversion to deceive predators. Journal of Wildlife Management 54: 360-


Crouse, D.T. 1984. Loggerhead sea turtle nesting in North Carolina: applications of an
aerial survey. Biological Conservation 29: 143-155.

Crouse, D.T., L.B. Crowder and H. Caswell. 1987. A stage-based population model for
loggerhead sea turtles and implications for conservation. Ecology 68: 1412-

Daud, S.K. and M.W.R.N. De Silva. 1987. Nesting site selection of leatherback and
green turtles on Terengganu Beach. Proceedings of the 10th Annual Seminar of
the Malaysian Society of Marine Sciences, pp.78-83.

Davis, G.E. and M.C. Whiting. 1977. Loggerhead sea turtle nesting in Everglades
National Park, USA. Herpetologica 33: 18-28.

Dean, Jr., W.E. 1974. Determination of carbonate and organic matter in calcareous
sediments and sedimentary rocks by loss of ignition: comparison with other
methods. Journal of Sedimentary Petrology 44: 242-248.

Edwards, K. 1991. Marine turtle nesting report. Ten Thousand Islands National
Wildlife Refuge, Naples, FL. Unpublished. Mimeo. 23pp.

Ehrenfeld, D.W. and A.L. Koch. 1967. Visual accommodation in the green turtle. Science
155: 827-828.

Fangman, M.S. and K.A. Rittmaster. 1994. Effects of human beach usage on the
temporal distribution of loggerhead nesting activities. Proceedings of the 13th
Annual Symposium on Sea Turtle Biology and Conservation, pp. 222-224.

Fowler, L.E. 1979. Hatching success and nest predation in the green turtle, Chelonia
mydas, at Tortuguero, Costa Rica. Ecology 60: 946-955

Frazer, N.B. 1983. Demography and life history evolution of the Atlantic loggerhead sea
turtle, Caretta caretta (Georgia). Ph.D. Dissertation, University of Georgia,
Athens. 247pp.

Frazer, N.B. 1983. Effect of tidal cycles on loggerhead sea turtles (Caretta caretta)
emerging from the sea. Copeia 1983: 516-519.

Frazer, N.B. 1984. A model for assessing mean age-specific fecundity in sea turtle
populations. Herpetologica 40: 281-291.

Frederick, P.C. and M.W. Collopy. 1989. The role of predation in determining
reproductive success of colonially nesting wading birds in the Florida Everglades.
The Condor 91: 860-867.


Fritts, T.H., M.L. Stinson and R. Marquez M. 1982. Status of sea turtle nesting in
southern Baja California, Mexico. Bulletin of the Southern California Academy
of Sciences 81: 51-60.

Fritts, T.H., W. Hoffman and M.A. McGehee. 1983. The distribution and abundance of
marine turtles in the Gulf of Mexico and nearby Atlantic waters. Journal of
Herpetology 17: 327-344.

Garmestani, A.S. 1995. Island biogeography: Raccoon movement patterns with respect
to a seasonal food resource. U.S. Fish and Wildlife Service, Naples, FL. Mimeo.

Gehrt, S.D. and E.K. Fritzell. 1996. Sex-biased response of raccoons (Procyon lotor) to
live traps. American Midland Naturalist 135: 23-32.

Goldman, E.A. 1950. Raccoons of north and middle America. North American Fauna
60. U.S. Fish and Wildlife Service. Washington, D.C.

Grassman, M.A., D.W. Owens, J.P. McVey and R. Marquez M. 1984. Olfactory-based
orientation in artificially imprinted sea turtles. Science 224: 83-84.

Grau, G.A., G.C. Sanderson and J.P. Rogers. 1970. Age determination of raccoons.
Journal of Wildlife Management 34: 364-372.

Hall, E.R. and K.R. Kelson. 1959. The mammals of North America. Vol. 2: 884-891.
Ronald Press Co. New York.

Hallett, J.G., M.A. O'Connell, G.D. Sanders and J. Seidensticker. 1991. Comparison of
population estimators for medium-sized mammals. Journal of Wildlife
Management 55: 81-93.

Harris, M.J. and C.G. Maley. 1990. Loggerhead sea turtle nesting on Ossabaw, St.
Catherines, Sapelo, and Little St. Simons Island, Georgia in 1989. U.S. Fish and
Wildlife Service Study Completion Report Project E-4, Study I: 28pp.

Hartman, L.H. 1993. Ecology of coastal raccoons (Procyon lotor) on the Queen
Charlotte Islands, British Columbia, and evaluation of their potential impact on
native seabirds. M.S. Thesis, University of Victoria, Vancouver, 155pp.

Hays, G.C. and J.R. Speakman. 1993. Nest placement by loggerhead turtles, Caretta
caretta. Animal Behaviour 45: 47-53.

Hays, G.C., A. MacKay, C.R. Adams, J.A. Mortimer, J.R. Speakman and M. Boerema.
1995. Nest site selection by sea turtles. Journal of Marine Biological
Assessments (U.K.). 75: 667-674.


Henwood, T.A. 1987. Sea turtles of the southeastern United States, with emphasis on
the life history and population dynamics of the loggerhead turtle, (Caretta
caretta). Ph.D. Dissertation, Auburn University, Auburn, AL. 170pp.

Hopkins, S.R., T.M. Murphy, Jr., K.B. Stansell, and P.M. Wilkinson. 1978. Biotic and
abiotic factors affecting nest mortality in the Atlantic loggerhead turtle.
Proceedings of the Annual Conference of the Southeastern Association of Fish
and Wildlife Agencies 32: 213-223.

Hopkins, S.R., and T.M. Murphy, Jr. 1982. Management of loggerhead turtle nesting
beaches in South Carolina. South Carolina Wildlife and Marine Resources
Department Report E-1, Study No. VI-A-2, 19pp.

Hopkins, S.R., and T.M. Murphy, Jr. 1982. Testing of lithium chloride aversion to
mitigate raccoon depredation of loggerhead turtle nests. Proceedings of the
Annual Conference of the Southeastern Association of Fish and Wildlife
Agencies 36: 484-491.

Hopkins-Murphy, S.R. and T.M. Murphy, Jr. 1983. Distribution of loggerhead turtle
nesting activity in South Carolina by aerial beach survey. South Carolina
Wildlife and Marine Resources Department Report E-1, Study No. VI-A-2, 60pp.

Hopkins-Murphy, S.R., and T.M. Murphy, Jr. 1985. Factors affecting nest survival of
Caretta caretta in South Carolina: implications for management. Unpublished,

Hughes, G.R. 1974. The sea turtles of southeast Africa. 1. Status, morphology and
distributions. Oceanographic Research Institute Investigational Report, Durban
35: 1-144.

Irvine, A.B., J.E. Caffin and H.I. Kochman. 1981. Aerial surveys for manatees and
dolphins in western peninsular Florida (with notes on sightings of sea turtle and
crocodiles). U.S. Fish and Wildlife Service, Washington, D.C. FWS/OBS-80/50,

Johannes, R.E. and D.W. Rimmer. 1984. Some distinguishing characteristics of the
green turtle (Chelonia mydas) on North West Cape Peninsula, Western Australia.
Marine Biology 83: 149-154.

Kilham, L. 1986. Renestings of American crows in Florida and predation by raccoons.
Florida Field Naturalist 14: 21- 23.

Klukas, R.W. 1967. Factors affecting nesting success of loggerhead turtles at Cape
Sable, Everglades National Park. File Number N1415 National Park Service, PO
Box 279, Homestead, Florida, 33033. 58pp.


Koch, A.L., A. Carr and D.W. Ehrenfeld. 1969. The problem of open-sea navigation:
the migration of the green turtle to Ascension Island. Journal of Theoretical
Biology 22: 163-179.

Kushlan, J.A. 1986. Atlantic loggerhead nesting status in southwest Florida.
Herpetological Review 17: 51- 52.

LeBuff, Jr., C.R. 1969. The marine turtles of Sanibel and Captiva islands. Sanibel-
Captiva Conservation Foundation, Inc., Report No.l: 1-13.

LeBuff, Jr., C.R. and R.W. Beatty. 1971. Some aspects of nesting of the loggerhead
turtle, Caretta caretta caretta (Linne), on the gulf coast of Florida. Herpetolgica
27: 153-156.

LeBuff, Jr., C.R. and P.D. Hagan. 1978. The role of aerial surveys in estimating nesting
populations of the loggerhead turtle. Florida Marine Research Publications 33:

LeBuff, Jr., C.R. 1990. The loggerhead turtle in the eastern Gulf of Mexico. Caretta
Research, Inc., Sanibel, FL, 236pp.

Lewis, T.E., G.O. Bailey, H.L. Edmiston. 1994. Effects of predator control on sea turtle
nest success on the barrier islands of Apalachicola Bay. Proceedings of the 14th
Annual Symposium on Sea Turtle Biology and Conservation. NOAA Technical
Memorandum NMFS-SFESC-351: 242-243.

Linhart, S.B., F.S. Blom, G.J. Dasch, J.D. Roberts, R.M. Engeman, J.J. Esposito, J.H.
Shaddock, and G.M. Baer. 1991. Formulation and evaluation of baits for oral
rabies vaccination of raccoons (Procyon lotor). Journal of Wildlife Diseases 27:

Linhart, S.B., F.S. Blom, R.M. Engeman, H.L. Hill, T. Hon, D.I. Hall, and J.H.
Shaddock. 1994. A field evaluation of baits for delivering oral rabies vaccines to
raccoons (Procyon lotor). Journal of Wildlife Diseases 30: 185-194.

Lohmann, K.J. 1992. How sea turtles navigate. Scientific American January: 100-106.

Lund, P.F. 1986. Nest production and nesting-site tenacity of the loggerhead turtle,
Caretta caretta, on Jupiter Island, Florida. Unpublished M.S. Thesis, University
of Florida, Gainesville. 32pp.

Maley, C.G. and M.J. Harris. 1991. Loggerhead sea turtle nesting on Ossabaw, St.
Catherines, Sapelo, and Little St. Simons Island, Georgia in 1990. U.S. Fish and
Wildlife Service Study Completion Report Project E-4, Study I: 36pp.


Maley, C.G. and M.J. Harris. 1992. Loggerhead sea turtle nesting on Ossabaw, St.
Catherines, Sapelo, and Little St. Simons Island, Georgia in 1991. U.S. Fish and
Wildlife Service Study Completion Report Project E-4, Study I: 37pp.

Maley, C.G. and M.S. Murphy. 1993. Loggerhead sea turtle nesting on Ossabaw, St.
Catherines, Sapelo, and Little St. Simons Island, Georgia in 1992. U.S. Fish and
Wildlife Service Study Completion Report Project E-4, Study I: 23pp.

Maley, C.G. and M.S. Murphy. 1994. Loggerhead sea turtle nesting on Ossabaw, St.
Catherines, Sapelo, and Little St. Simons Island, Georgia in 1993. U.S. Fish and
Wildlife Service Study Completion Report Project E-4, Study I: 21pp.

Maley, C.G. 1995. Loggerhead sea turtle nesting on Ossabaw, St. Catherines, Sapelo,
and Little St. Simons Island, Georgia in 1994. U.S. Fish and Wildlife Service
Study Completion Report Project E-4, Study I: 16pp.

Marsh, H. and D.F. Sinclair. 1989. An experimental evaluation of dugong and sea turtle
aerial survey techniques. Australian Wildlife Research 16: 639-650.

McMurtray, J.D. 1986. Reduction of raccoon predation on sea turtle nests at Canaveral
National Seashore, Florida. M.S. Thesis, University of Georgia, Athens. 48pp.

Melucci, C.L., J.I. Richardson, R. Bell and L.A. Corliss. 1991. Nest site preference and
site fixity of hawksbills on Long Island, Antigua. Proceedings of the 11th Annual
Workshop on Sea Turtle Biology and Conservation, pp. 171-174.

Meylan, A. and P. Meylan. 1985. Nesting of Dermochelys coriacea in Caribbean
Panama. Journal of Herpetology 19: 293-297.

Meylan, A.B., B.W. Bowen and J.C. Avise. 1990. A genetic test of the natal homing
versus social facilitation models for green turtle migration. Science 248: 724-

Miller, J.D. 1995. Nesting biology of sea turtles. In Biology and Conservation of Sea
Turtles, K.A. Bjorndal, Ed.-Rev. Ed., pp.573-575.

Mortimer, J.A. and K.M. Portier. 1989. Reproductive homing and interesting behavior
of the green turtle (Chelonia mydas) at Ascension Island, South Atlantic Ocean.
Copeia 1989: 962-977.

Mortimer, J.A. 1990. The influence of beach sand characteristics on the nesting
behavior and clutch survival of green turtles (Chelonia mydas). Copeia 1990:

Mortimer, J.A. 1995. Factors influencing beach selection by nesting sea turtles. In
Biology and Conservation of Sea Turtles, K.A. Bjorndal, Ed.-Rev. Ed., pp. 45-52.


Mourao, G.M., P. Bayliss, M.E. Coutinho, C.L. Abercrombie and A. Arruda. 1994. Test
of an aerial survey for caiman and other wildlife in the Pantanal, Brazil. Wildlife
Society Bulletin 22: 50-56.

Mrosovsky, N. 1983. Ecology and nest-site selection of leatherback turtles,
Dermochelys coriacea. Biological Conservation 26: 47-56.

Murphy, T.M. and S.R. Hopkins-Murphy. 1990. Homing of translocated gravid
loggerhead turtles. Proceedings of the 10th Annual Workshop on Sea Turtle
Biology and Conservation, pp. 123-124.

Nelson, E.W. 1930. Four new raccoons from the keys of southern Florida. Smithsonian
Miscellaneous Collections 82: 1-17.

Nelson, R.L. 1970. Evaluation of chemosterilants as inhibitors of reproduction in
raccoons. South Dakota Department of Game, Fish and Parks progress report W-
75-R- 12: 10pp.

Nicolaus, L.K. 1987. Conditioned aversions in a guild of egg predators: implications
for aposematism and prey defense mimicry. Department of Biological Sciences,
Northern Illinois University, DeKalb 60115.

Nicolaus, L.K., J. Herrera, J.C. Nicolaus, and C.R. Gustavson. 1989. Ethinyl estradiol
and generalized aversions to eggs among free-ranging predators. Applied Animal
Behaviour Science 24: 313-324.

Pollock, L.W. and W.D. Hummon. 1971. Cyclic changes in interstitial water content,
atmospheric exposure, and temperature in a marine beach. Limnology and
Oceanography 16: 522-535.

Portier, K.M. and D.L. Anderson. 1995. Using tree regression to identify nutritional and
environmental factors affecting sugarcane production. Applied Statistics in
Agriculture 7: 182-189.

Pritchard, P. 1971. The leatherback or leathery turtle, Dermochelys coriacea. IUCN
Monograph, Marine Turtle Series 1: 1-39.

Pritchard, P.C.H. 1982. Nesting of the leatherback turtle, Dermochelys coriacea, in
Pacific Mexico, with a new estimate of the world population status. Copeia 1982:

Ratnaswamy, M.J. 1995. Raccoon depredation of sea turtle nests at Canaveral National
Seashore, Florida: implications for species management and conservation. Ph.D.
Dissertation, University of Georgia, Athens. 79pp.


Richardson, J.I. 1982. A population model for adult female loggerhead sea turtles
(Caretta caretta) nesting in Georgia. Ph.D. Dissertation, University of Georgia,
Athens. 204pp.

Richardson, J.I., T.H. Richardson and M.W. Dix. 1978. Population estimates for nesting
female Loggerhead sea turtles (Caretta caretta) in the St. Andrew Sound area of
southeastern Georgia. Florida Marine Research Publications 33: 34-38.

Richardson, T.H., J.I. Richardson and M.W. Dix. 1978. Remigration patterns of
loggerhead sea turtles (Caretta caretta) nesting on Little Cumberland and
Cumberland Islands, Georgia. Florida Marine Research Publications 33: 39-44.

Sarti, M.L., S.A. Eckert, T.N. Garcia and A.R. Barragan. 1996. Decline of the world's
largest nestingassemblage of leatherback turtles. Marine Turtle Newsletter 74:

SAS Institute, Inc. 1989. SAS/STAT User's guide. SAS Institute, Cary, NC. 846pp.

Schulte, E.E. 1988. Recommended soil organic matter tests. In Recommended chemical
soil test procedures for the north-central region, W.C. Dhanke, Ed. North Dakota
Agricultural Experiment Station, North Dakota State University Bulletin #499:
Schulz, J.P. 1975. Sea turtles nesting in Surinam. Stichting Natuurbehoud Suriname
(STINASU) 3: 1-143.

Seigel, R.A. 1979. Predation by raccoons on diamondback terrapins, Malaclemys
terrapin tequesta. Journal of Herpetology 14: 87-89.

Semel, B. and Nicolaus, L.K. 1992. Estrogen-based aversion to eggs among free-
ranging raccoons. Ecological Applications 2: 439-449.

Shabica, S., D. Stoneburner, D. Gilmore, J. Hedden, J. Hinesley and J. Swigert. 1978.
Raccoon predation on sea turtle nests. Bulletin of the Ecological Society of
America 59: 102.

Shoop, C.R., C.A. Ruckdeschel and N.B. Thompson. 1985. Sea turtles in the southeast
United States: nesting activity as derived from aerial and ground surveys, 1982.
Herpetologica 41: 252-259.

Stancyk, S.E. and J.P. Ross. 1978. An analysis of sand from green turtle nesting beaches
on Ascension Island. Copeia 1978: 93- 99.

Stancyk, S.E., O.R. Talbert and A.B. Miller. 1979. Estimation of loggerhead turtle
nesting activity in South Carolina by aerial survey. American Zoologist 19: 954.


Stancyk, S.E., O.R. Talbert and J.M. Dean. 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.

Statistical Sciences. 1993. S-PLUS User's manual. StatSci, Seattle.

Stoneburner, D.L. and J.I. Richardson. 1981. Observations on the role of temperature in
loggerhead turtle nest site selection. Copeia 1981: 238-241.

Thompson, N.B. 1980. Population dynamics of the Atlantic green sea turtle, Chelonia
mydas, (Linnaeus) 1758. Ph.D. Dissertation, University of Rhode Island,
Kingston. 95pp.

Urban, D. 1970. Raccoon populations, movement patterns, and predation on a managed
waterfowl marsh. Journal of Wildlife Management 34: 35-39.

Venables, W.N. and B.D. Ripley. 1994. Modern Applied Statistics with S-PLUS.
Springer-Verlag, New York. 462pp.

Westin, L. and A. Nissling. 1991. Effects of salinity on spermatozoa motility,
percentage of fertilized eggs and egg development of Baltic cod (Gadus morhua),
and implications for cod stock fluctuations in the Baltic. Marine Biology 108: 5-

Whitmore C.P. and P.H. Dutton. 1985. Infertility, embryonic mortality and nest-site
selection in leatherback (Dermochelys coriacea) and green sea turtles (Chelonia
mydas) in Surinam. Biological Conservation 34: 251-272.

Witherington, B.E. 1995. Hatchling orientation. In Biology and Conservation of Sea
Turtles, K.A. Bjorndal, Ed.-Rev. Ed., pp. 577-578.


University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs