<%BANNER%>

Guiding Loggerhead Hatchlings to the Sea

Permanent Link: http://ufdc.ufl.edu/UFE0021227/00001

Material Information

Title: Guiding Loggerhead Hatchlings to the Sea An Assessment of Ground-Level Barriers and Conflicting Cues
Physical Description: 1 online resource (82 p.)
Language: english
Creator: Scarpino, Russell A
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: caretta, disorientation, hatchling, loggerhead
Interdisciplinary Ecology -- Dissertations, Academic -- UF
Genre: Interdisciplinary Ecology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Artificial lighting disrupts the orientation ability of hatchling sea turtles as they crawl from their nest to the sea. Experiments were conducted at night on beaches exposed to artificial lighting to determine whether the placement of ground-level nest shields landward of hatchlings would act to restore proper orientation. Shielding failed to reduce hatchling disorientation during seven of nine testable sampling events. Of the two events with positive results, only one found the shield to restore orientation to the majority of hatchlings in the group. Therefore, nest shielding is a rather poor technique for managing the impacts of artificial lighting on hatchling turtles. Management plans should focus on applying methods that reduce light pollution levels, rather than manipulating hatchlings and the nest environment. Coastal light management practices have darkened many beaches throughout Florida to successfully reduce hatchling disorientation rates. As coastal lighting problems become more adequately resolved, attention must turn to larger-scale sources of light pollution that affect turtles, such as skyglow emanating from more urbanized areas.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Russell A Scarpino.
Thesis: Thesis (M.S.)--University of Florida, 2007.
Local: Adviser: Carthy, Raymond R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021227:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021227/00001

Material Information

Title: Guiding Loggerhead Hatchlings to the Sea An Assessment of Ground-Level Barriers and Conflicting Cues
Physical Description: 1 online resource (82 p.)
Language: english
Creator: Scarpino, Russell A
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: caretta, disorientation, hatchling, loggerhead
Interdisciplinary Ecology -- Dissertations, Academic -- UF
Genre: Interdisciplinary Ecology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Artificial lighting disrupts the orientation ability of hatchling sea turtles as they crawl from their nest to the sea. Experiments were conducted at night on beaches exposed to artificial lighting to determine whether the placement of ground-level nest shields landward of hatchlings would act to restore proper orientation. Shielding failed to reduce hatchling disorientation during seven of nine testable sampling events. Of the two events with positive results, only one found the shield to restore orientation to the majority of hatchlings in the group. Therefore, nest shielding is a rather poor technique for managing the impacts of artificial lighting on hatchling turtles. Management plans should focus on applying methods that reduce light pollution levels, rather than manipulating hatchlings and the nest environment. Coastal light management practices have darkened many beaches throughout Florida to successfully reduce hatchling disorientation rates. As coastal lighting problems become more adequately resolved, attention must turn to larger-scale sources of light pollution that affect turtles, such as skyglow emanating from more urbanized areas.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Russell A Scarpino.
Thesis: Thesis (M.S.)--University of Florida, 2007.
Local: Adviser: Carthy, Raymond R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021227:00001


This item has the following downloads:


Full Text

PAGE 1

1 GUIDING LOGGERHEAD HATCHLINGS TO THE SEA: AN ASSESSMENT OF GROUND-LEVEL BARRIERS AND CONFLICTING CUES By RUSSELL A. SCARPINO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Russell A. Scarpino

PAGE 3

3 To my parents Joseph and Sarah Scarpino

PAGE 4

4 ACKNOWLEDGMENTS I cannot thank my parents Joseph and Sarah S carpino enough for their continuous love and support. I dedicate my work to them. I am es pecially grateful for th e advice and support of my major advisor, Raymond Carthy, and my gra duate committee members Nat Frazer and Blair Witherington. This work would not be possible without the numerous vol unteers and technicians that sacrificed their summers to help out. They made it happen. I would like to thank the Natu ral Resources division of Tyndall Air Force Base for their logistical support in the field. I am also indebted to BAE Systems and Eglin Air Force Base for their companionship and logistical support in the field. Other assist ance was provided by the Florida Cooperative Fish & Wildlife Research Unit, Kirt Rusenko at the Gumbo Limbo Nature Center, Bob Miller at Eglin Air Force Base, John Calabro, and the staff from St. Joseph Peninsula State Park. Work was perfor med under FL DEP Marine Turtle Permit #94.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES................................................................................................................ .........8 ABSTRACT....................................................................................................................... ............10 CHAPTER 1 INTRODUCTION..................................................................................................................11 2 ORIENTATION AND NEST SHIELDING..........................................................................16 Introduction................................................................................................................... ..........16 Materials and Methods.......................................................................................................... .17 Study Area..................................................................................................................... ..17 Orientation Trials.............................................................................................................17 Nest Shields................................................................................................................... ..18 Sample Gathering ( Caretta caretta Hatchlings)..............................................................18 Data Analysis.................................................................................................................. .18 Results........................................................................................................................ .............19 Discussion..................................................................................................................... ..........20 3 ORIENTATION AND LIGHT CUES...................................................................................40 Introduction................................................................................................................... ..........40 Materials and Methods.......................................................................................................... .41 Study Area..................................................................................................................... ..41 Light Intensity Measurements.........................................................................................42 Orientation trials............................................................................................................. .42 Tyndall Air Force Base Ligh ting and Dune Evaluation..................................................42 Data Analysis.................................................................................................................. .43 Results........................................................................................................................ .............44 Discussion..................................................................................................................... ..........45 4 ORIENTATION AN D MULTIPLE CONFLICTING CUES................................................65 Introduction................................................................................................................... ..........65 Materials and Methods.......................................................................................................... .66 Results........................................................................................................................ .............66 Discussion..................................................................................................................... ..........67

PAGE 6

6 5 SUMMARY........................................................................................................................ ....69 Results........................................................................................................................ .............69 Management Implications......................................................................................................70 APPENDIX. LOGISTIC REGRESSION TABLE.......................................................................72 LIST OF REFERENCES............................................................................................................. ..79 BIOGRAPHICAL SKETCH.........................................................................................................82

PAGE 7

7 LIST OF TABLES Table page 2-1 Hatchling orientation tria l site description, 2004-2005.....................................................23 2-2 Summary of experi mental design, 2004-2005...................................................................24 3-1 Artificial light sources visi ble from experimental sites.....................................................47 3-2 Summary of sampling event light conditions....................................................................48 3-3 Artificial light sources observed from TAFB beaches......................................................49 3-4 Light intensity measurements at Site 1, Tyndall Air Force Base......................................50 3-5 Light intensity measurements at Site 2, Tyndall Air Force Base......................................51 4-1 Logistic regression output for the combined site/treatment/lunar variable. Site/treatment/lunar variable combinations in bold are significan tly correlated with the disorientation of a hatchling.........................................................................................68

PAGE 8

8 LIST OF FIGURES Figure page 2-1 Map of Florida displaying Tyndall Ai r Force Base (TAFB) and Boca Raton...................25 2-2 Aerial image of Tyndall Air Force Ba se (TAFB), located in Bay County on the northwestern coast of Florida.............................................................................................26 2-3 Aerial image of Boca Raton, located in Palm Beach County, FL on the southeastern coast of Florida............................................................................................................... ...27 2-4 Hatchling orientation arenas (2m and 8m radii)................................................................28 2-5 Orientation of hatchling loggerh ead turtles in sampling event A......................................29 2-6 Orientation of hatchling loggerh ead turtles in sampling event B1....................................30 2-7 Orientation of hatchling loggerh ead turtles in sampling event B2....................................31 2-8 Orientation of hatchling loggerh ead turtles in sampling event B3....................................32 2-9 Orientation of hatchling loggerh ead turtles in sampling event C......................................33 2-10 Orientation of hatchling loggerh ead turtles in sampling event D......................................34 2-11 Orientation of hatchling loggerh ead turtles in sampling event E1....................................35 2-12 Orientation of hatchling loggerh ead turtles in sampling event E2....................................36 2-13 Orientation of hatchling loggerh ead turtles in sampling event F1.....................................37 2-14 Orientation of hatchling loggerh ead turtles in sampling event F2.....................................38 2-15 Orientation of hatchling loggerhea d turtles in sampling event G. ...................................39 3-1 Aerial image of Tyndall Air Force Base located in Bay County on the northwestern coast of Florida............................................................................................................... ...52 3-2 Aerial image of Boca Raton, located in Palm Beach County, FL on the southeastern coast of Florida............................................................................................................... ...53 3-3 Orientation and light inte nsity at sampling event A..........................................................54 3-4 Orientation and light inte nsity at sampling event B1.........................................................55 3-5 Orientation and light inte nsity at sampling event B2.........................................................56 3-6 Orientation and light inte nsity at sampling event B3.........................................................57

PAGE 9

9 3-7 Orientation and light inte nsity at sampling event C...........................................................58 3-8 Orientation and light inte nsity at sampling event D..........................................................59 3-9 Orientation and light inte nsity at sampling event E1.........................................................60 3-10 Orientation and light inte nsity at sampling event E2.........................................................61 3-11 Orientation and light inte nsity at sampling event F1.........................................................62 3-12 Orientation and light inte nsity at sampling event F2.........................................................63 3-13 Orientation and light inte nsity at sampling event G..........................................................64

PAGE 10

10 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science GUIDING LOGGERHEAD HATCHLINGS TO THE SEA: AN ASSESSMENT OF GROUND-LEVEL BARRIERS AND CONFLICTING CUES By Russell A. Scarpino August 2007 Chair: Raymond R. Carthy Major: Interdisciplinary Ecology Artificial lighting disrupts the orientation ability of hatchling sea turtles as they crawl from their nest to the sea. Experiments were conduc ted at night on beache s exposed to artificial lighting to determine whether th e placement of ground-level nest shields landward of hatchlings would act to restore proper orie ntation. Shielding failed to redu ce hatchling disorientation during seven of nine testable sampling events. Of th e two events with positive results, only one found the shield to restore orientati on to the majority of hatchlings in the group. Therefore, nest shielding is a rather poor tec hnique for managing the impacts of artificial lighting on hatchling turtles. Management plans should focus on applying methods that reduce light pollution levels, rather than manipulating hatchlings and the nest environment. Coastal light management practices have darkened many beaches througho ut Florida to successfully reduce hatchling disorientation rates. As coastal lighting problem s become more adequate ly resolved, attention must turn to larger-scale sources of light pollution that affect turtles, such as skyglow emanating from more urbanized areas

PAGE 11

11 CHAPTER 1 INTRODUCTION For species and ecosystems that have evolved with a nightly allowance of darkness, our use of artificial night lighting can serve as a source of ecological disruption. Scientists are only now beginning to realize the impact of the ever-increasing brightening of the nighttime environment. This brightening if often referred to as ecological light pollution, arising from man-made sources such as lighted structures (e .g. buildings, bridges, towers), streetlights, security lighting, lights on vehicles and boats, and sky glow. Ecological light pollution is a global phenomenon, with consequences impacting a wi de range of species (Elvidge et al., 1997). It is shown to disrupt the daily movement of plankt on, disturb mammal disp ersal patterns, delay salmon migration and even disorient and cause th e death of migratory bi rds and sea turtles. Artificial night lighting produces distinct effects on the beha vior and population ecology of organisms in natural settings. These effects de rive from changes in orientation, attraction and repulsion caused by lights, which in turn, ma y affect foraging, re production, migration and communication (Longcore and Rich, 2006). Orient ation is caused by ambient light conditions, whereas, attraction and repulsion to luminance, or the brightness of a light source (Health Council of the Netherlands, 2000). Many diurnal birds and reptiles are known to forage under artificia l illumination, with seemingly beneficial results to all but their pr ey. In addition to fo raging, orientation under artificial light levels may induce other behaviors such as territoria l singing in birds or territorial displays in salamanders (Bergen and Abs, 1997). Artificial lighting may also disorient organisms accustomed to navigating in darkness The best known example of this is the disruption of hatchling sea turtle orientation as they crawl from their nests to the ocean. Under normal circumstances, hatchlings crawl quickly away from tall, dark objects characteristic of a

PAGE 12

12 dune landscape, and toward the lower, flatter, an d often brighter seaward horizon. With artificial lighting, hatchlings may no longer perceive dark silhouettes and crawl in circuitous paths (disorientation) or in direct paths towards light sources, away from the ocean (misorientation) (Witherington and Martin, 1996). The consequences of disrupted seafinding orientation can be fatal, and an estimated hundreds of thousands of hatchlings peri sh each year from dehydration, exhaustion, or capture by pr edators (Witherington, 1997). Marine turtles represent a group of species w hose life histories can best be described as highly vagile. Hatchling loggerhead sea turtles ( Caretta caretta ) from southeast Florida use three different sets of orientation cues to guide them safely offshore. After emerging from the nest, hatchlings utilize a variety of visual cues to quickly locate and maintain a direct course to the sea. They will generally orient toward th e flatter, and often brighter, seaward horizon, and away from higher, spatially va riable dune profile. Once they enter the surf zone, hatchling loggerheads are able to set their magnetic compa ss by orienting in response to wave surge (Wang et al., 1998) and the orbital motion of waves (Ma nning et al., 1997). As the turtle matures, so does its ability to incorporate more complex stim uli, as seen in pigeons and migratory birds (Wiltschko and Wiltschko, 1998). Non-neonate sea turt les are known to exhibit site fidelity and homing behavior (Dickerson et al., 1995; Lut cavage and Musick, 1985; Mendonca and Ehrhart, 1982), providing evidence that a more complex map sense develops with age. The environmental cues used by seafinding ha tchling sea turtles have been extensively reviewed. Earlier studies of these cues remain ed a matter of controversy (van Rhijn and van Gorkom, 1983; Mrosovsky and Kingsmill, 1985), but agre ed they are of a visual nature. This is evidenced by observations utilizi ng blindfolded turtles, in whic h none could successfully locate the sea (Ehrenfeld and Koch, 1967). Van Rhijn (1979) provided the only exception to this

PAGE 13

13 theory, demonstrating that blindf olded hatchlings preferred to de scend inclines which simulated the downward slope of a beach to the sea. While beach slope may assist hatchlings in finding the sea, no one has demonstrated it to independently aid seafinding (S almon et al., 1992). Previous experiments by Mrosovsky (1972) su ggested that hatchlings lo cate the sea by orienting and moving toward the brightest horizon, a mechanism known as positive phototropotaxis. The term brightness has been used in re lated literature to generally refer to the intensity and wavelengths of light visible to these turtle s. Although the positive phototr opotaxis hypothesis was the most widely held, studies later show ed hatchlings responding to cues associated with a horizons elevation (hatchlings oriented away from the higher silhouette produced by dunes and vegetation behind the nest), disregarding the direction of brightest light intensity (Limpus, 1971; Salmon et al., 1992). In conjunction with horizon elevatio n, Van Rhijn and van Gorkom (1983) revealed the importance of its shape as a potential cue. Tu rtles presented with a ve rtically striped horizon and an opposing, unmarked horizon consistently oriented away from the horizon displaying vertical stripes. Witherington ( 1992a) repeated this experiment and also found turtles to orient away from a silhouette containing ve rtical stripes, only to become attracted to it when made five times brighter than the open horizon. Some understanding of the visual systems of these turtles is required when discussing photic response behaviors such as seafinding. Tw o separate studies have sought to determine the spectral sensitivity of the sea turtle eye using a method known as electroretinography (ERG). This technique measures the electrical poten tial across the animals retina upon exposure to different wavelengths of light. Their data show that both loggerhead and green ( Chelonia mydas ) turtles are responsive to wavelengths from 440-700nm, and share similar peak sensitivity in the longer wavelength regi on, at approximately 580nm (Lev enson et al., 2004). Although

PAGE 14

14 spectral sensitivity is an important piece of inform ation in discerning the visual circuitry of these turtles, its use as an indicator of the animals behavioral response should be cautioned. Spectral sensitivity data show limited re tinal responses to shorter wavele ngths of light, even though this region of light will elicit the st rongest behavioral responses. Ha tchling loggerhead and green sea turtles will orient toward shorte r wavelength light of the near-ultraviolet to green region, whereas their response to longer wave length light is minimal (Witheri ngton, 1991). In addition to the characteristics of light such as intensity a nd wavelength, color has been implicated to elicit orientation responses in hatchli ngs. Exposure to higher intensities of yellow light was found to actually repel loggerhead hatchlings, and display a weak attraction to the colored source when the intensity is greatly reduced (Witherington, 1992a). The naturally occurring light from celestial bodies also aff ects hatchling seafinding ability. Th e moons phase, position in the sky, and temporal visibility can have a tremendous effect on hatchling orientation, especially on beaches where artificial landward lighting compromises proper sea-finding ability. Salmon and Witherington (1995) recorded natural and experi mental hatchling emergences at sites with directly visible artificial light so urces, and observed that more turt les fail to orient properly on dark nights around the new moon than under full moon illumination. Upon further analysis, it was revealed that background illumination from the moon, and not the moon itself restored normal sea-finding orientation as it decrease d the opposing anisotropi c light gradients. The purpose of this present study was to test methods of correcti ng seafinding behavior disrupted by artificial lighting. The principle objectives of this study were: 1) to determine whether the proximity and placement of groundlevel barriers (nest sh ields) landward of hatchlings affects their seafindi ng ability, 2) to evaluate the re sponse of hatchlings to multiple, photic stimuli present in a natural environment, and 3) to test an analytic al technique for relating

PAGE 15

15 the effects of multiple, visual cues on hatchling orientation response. To meet these objectives, hatchling orientation trials were conducte d on the beach at night, under a variety of environmental conditions. The first aspect of th e orientation trials invo lved a ground-level nest shielding experiment, which is discussed in Chapter 2. The various light measurements associated with these orientation trials are disc ussed separately in Chapter 3. Chapter 4 is a treatment synthesis of hatchling orientation meas urements and the combined effects of visual stimuli presented in the previous two chapters Chapter 5 summarizes the results of this experiment and discusses the management implications of nest shield use.

PAGE 16

16 CHAPTER 2 ORIENTATION AND NEST SHIELDING Introduction Most biologists agree with the hypothesis that hatchling sea turtles locate the ocean by relying on a combination of two sets of visual cues: light intensity gradients (hatchlings orient in the direction of highest light intensity) and hor izon elevation and shape (hatchlings crawl away from high, spatially variable s ilhouettes, characteristic of dun es and/or vegetation) (Limpus, 1971; Witherington, 1992b; Salmon et al., 1992, 1995). Past studies of dune profile and horizon shape were predominately conducted under labora tory conditions, and few have examined the effects of these conditions on hatchling or ientation in their natural environment. Adamany et al. (1997) studied the behavior of hatchling loggerhead turtles orienting from within open and shielded cages, in the pr esence of artificial landward light. Hatchlings emerging under an uncovered, or open, cage stayed within the confines of the cage for long periods of time until ambient light levels return ed to normal until the brightest direction was seaward approaching dawn. When the cages were shielded on the landward side, hatchlings were less likely to become tr apped. Although the shielded cage reduced cage trapping, hatchlings oriented toward th e landward artificial light upon exiti ng the confines of the cage. I would hypothesize that there would be sim ilar results with the use of ground-level barriers. To test this, I conduc ted orientation trials on the beach at night to determine if placement of ground-level artificial barriers (nest shields) landwa rd of orienting hatchlings affected their ability to locate th e sea. In addition, I attempted to assess whether a turtles proximity to the shield was a contributing fact or. Light intensity r eadings were taken in conjunction with these orientation trials, a nd are examined separately in Chapter 3.

PAGE 17

17 Materials and Methods Study Area The study was conducted on beaches owned by Tyndall Air Force Base (TAFB) in Bay County, FL and Boca Raton in Palm Beach County, FL (Figure 2-1). Artificial lighting has been known to disrupt hatchling orientation in both ar eas. TAFB spans approximately 25 kilometers of south-facing coastline along the Gulf Coast of Florida. These shores support turtles from the relatively small matrilineal stock of loggerheads known to nest throughout Northwestern Florida (Encalada et al., 1998). TAFB c onsists of (from east to west) Crooked Island East Beach (CIE), NCO Beach (NCO), and Shell Is land (Figure 2-2). Three additional sites in Boca Raton, FL were chosen for this study due to their high loggerhead nesting density, unique exposure to artificial lighting (Figure 2-3). These sites, adjacent to Red Reef Park, North Boca Inlet, and Spanish River Park, are within of a 5-mile stretch of beach that has been monitored for sea turtle activity by the City of Boca Raton since 1976. Orientation Trials Experimental trials were conducted in 2004 at four sites on TAFB beaches, and at three additional sites in Boca Raton, the following year. Trials were carried out during the night at pre-selected sites, which were chosen for th eir various dune profiles and exposure to light sources. At each site, two concen tric circles (arenas) were draw n in the sand, with radii of two and eight meters (Figure 2-4). The circumference of each arena was marked off at 10 intervals, with the 0 mark corresponding to the most direct route to the ocean. Prior to release, turtles were placed in a shallow wooden bowl, covered with a lid, and set flush with the sand surface in the middle of the arena center. Trials began by removing the lid via a pull cord operated by the observer. Trials began with the release of hatchlings, in groups of 3-10, in the center of the arenas, and concluded when all hatchlings exite d the outer, 8m arena. An observer monitored

PAGE 18

18 the release in a prone position, at least 1 meter from arena pe riphery, creating a silhouette no higher than 1 foot from the beach surface. Orient ation angles (two per turtle) were recorded for each hatchling, and individual turtles were used only once. Hatchlings that failed to orient seaward upon completion of a trial were retrieved a nd released at a nearby, darker area of beach. Nest Shields Nest shields consisted of black silt screen, also referred to as shade fencing, which was supported by wooden stakes, reaching a height of 1 meter. A double layer of fabric was used and coated in flat black paint to minimize reflecta nce. Shields were successful at blocking the majority of incident light, but were not comple tely opaque. During treatment releases, shields were positioned in a 180 arc on th e landward side of the arena, one meter from the release point, with the arc opening facing the ocean (Figure 24). Nest shields were erected immediately preceding hatchling trials and taken down later that night. Sample Gathering ( Caretta caretta hatchlings) Hatchlings were gathered in the early ev ening from loggerhead ne sts approaching their estimated date of emergence (approximately 55 da ys following deposition). Only turtles located just below the sand surface were gathered and ke pt light-naive in Styrofoam coolers before transport to an experimental aren a site. Hatchlings from multiple clutches were pooled together when possible to reduce sampling bias. Sample sizes for orientation trials were largely dependent upon the availability of nearly emer gent hatchlings on a given sampling night and location. Data Analysis Orientation data were analyzed (1) to determine if groups of turtles were significantly oriented, (2) to assess whether groups of turtle s were significantly oriented toward the most direct route to sea (0), (3) to determine if or ientation significantly di ffered between control and

PAGE 19

19 shielded treatment groups, and (4) between 2m arena and 8m arena da ta sets. Circular statistical methods were used to analyze these data (Fischer 1993). Exit angles for i ndividual turtles were plotted in circular histograms for visual comparison among groups Measurements of dispersion (group mean angle and r-vector) were calculated for each group. Rayleighs test was used to meet the first objective. A circular V-test wa s used to determine if groups of turtles were significantly oriented toward a pre-defined direction, in this case, directly seaward (0). Watsons U2 nonparametric test and Watson-Williams F-te sts (pair-wise comparisons) were used to determine whether significant differen ces exist among treatment and arena groups. Results During the summers of 2004 and 2005, 273 logge rhead hatchlings from 19 separate clutches were used in orientation trials at se ven sites on TAFB and Boca Raton beaches (Table 21). Of these 273 turtles, 128 were released in front of a nest shield (treatment) and 145 were released as control subjects (T able 2-2). Orientation trials conducted on a single night are given a sampling event ID (ex. B2=second sampling event at site B). Circular raw data plots of individual hatchling exit angles were constructed and sorted by sampling event, arena size, and treatment (Figures 2-5-15). Of the forty distributions tested, all but se ven were significantly oriented as groups (Rayleighs test, p<0.05). Thes e groups involved sampling events B2 (2m and 8m control and treatment groups), B3 (2m treatment group), and G (2m and 8m control groups) (see Figs. 2-7, 28, 2-15). A similar test, more suited for bimodal distributions, determined that all groups are significantly oriented (Raos spacing test, p<0.05 ). Only eight groups failed to significantly orient toward the most direct route to sea (V -test, p<0.05). These included the same sampling events B2 (2m and 8m control groups), B3 (2m and 8m control and treatment groups), and G (2m and 8m control groups). There were no sign ificant differences between distributions of exit

PAGE 20

20 angles measured from the 2m and 8m arenas (Watsons U2 test, p<0.05), w ith exception of sampling event B3 (Figure 2-8). It is unclear whether this resu lt is caused by the very small, group sample size in B3 (n=6). Si nce distributions of ex it angles taken at the 2m arenas matched those taken at the 8m arenas, I chose to use the 2m arena distributions in subsequent multisample comparisons. Distributions of hatchlings orie nting in the presence of a ground-level nest shield appear to be significantly different from their contro l groups during four of the nine paired control/treatment sampling events. This result is dependent upon the ty pe of statistical test used. These events are: B2 (Watsons U2 and Watson-Williams F-test, p<0.05), B3 (Watsons U2 and Watson-Williams F-test, p<0.05), C (Watson-Willi ams F-test only, p<0.05), and G (Watsons U2 test only, p<0.05) (Figures 2-7, 2-8, 2-9, and 2-15). Discussion Nest shielding trials were conducted at multip le sites under natural conditions. Conditions in the natural environment are complex, and differe nt sites varied with regard to dune profile, exposure to artificial light, a nd levels of lunar illumination. Natural and artificial lights, discussed in the following chapte r, will likely have some influen ce on hatchling orientation. For this reason, shielding effects were ascertained on a site-by-site and night-by-night basis, in an attempt to reduce the likelihood of light cues acting as overr iding, confounding factors when comparing distributions. With the exception of three sampling events (B2, B3, and G; Figures 2-7, 2-8, and 2-15), all groups of hatchlings were found to have directed orientations to ward the sea. The statistical test used, Rayleighs test, determined that even strongly bimodal di stributions (ex. groups directed seaward and landward) are not significantly orient ed. To account for possible bimodality, an additional test known as Watson's non-parametric two-sample U2 test was used.

PAGE 21

21 As with most analytical techniques, caveats such as these should be kept in mind when evaluating data. Individual hatchlings used in or ientation trials tended to crawl in rather directed paths, with very little wayward or circuitous movement. Ha tchling exit angles measured at the 2m and 8m arenas were not found to be si gnificantly different in pair-wis e or grouped comparisons. These data imply that hatchling orientation does not si gnificantly change with respect to a turtles distance from a visual barrier such as a nest shield, although caution s hould be taken when making this assumption. When exit angles are us ed as measures of hatchling orientation, one must keep in mind that orientation between arena perimeters is unaccounted for and could change drastically, even if the recorded 2m and 8m exit angles are alike. This was a factor when hatchlings crawling directly landw ard reoriented to evade the nest shield, only to continue its landward heading after clearing th e shield (event B3; Figure 2-8) Judging strictly by the exit angles, it appears the turtle crawled under th e shield, which was not possible under these conditions. Another assumption of this hypothesis regards the ini tial distance of hatchlings to ground-level barriers. Hatchlings were released at the same distance from a shield (1 meter) during all trials, and allowed to crawl before he adings were recorded from both arenas. Past research has suggested that a hatc hlings initial crawl heading may a ffect or set its direction of orientation. Therefore, this e xperiment does not quite discern be tween a hatchlings initial crawl and a hatchlings distance from the shield upon re lease as separate entiti es influencing further orientation. Sampling events with paired control and treatm ent trials (all except B1, E2) may provide some indication of a nest shields influence on orientation. Events A, C, D, E1, F1, and F2 consisted of groups of turtles whic h were directly oriented toward the sea, and, therefore did not

PAGE 22

22 display significant treatment effects (Figures 25, 9, 10, 11, 13, and 2-14). Events B2, B3, C, and G were found to have differences in mean v ector length (Watson-Williams F test) or other distributive properties (Watsons U2 test) between contro l and treatment groups. Of vital importance when dealing with circul ar data is the need to incorporate visual assessments of raw data plots into ones anal ytical framework to en sure that statistical assumptions are not violated. When relying on st atistical tests alone, one may assume that the shielding treatment in sampling event C had a co nsiderable effect on ha tchling orientation. Statistically, shielding did have a measurable effect on the mean vector length (r). Upon inspection of the raw data pl ots, it is obvious that shield ing did not have a significant biological effect on orientation at site C (Figure 2-9). The ability of a shield to restore seaward orientation in hatchlings would be considered a significant biologi cal effect. More research is needed to determine the suitability of testing for di fferences in mean vector length in such cases. Sample size is also a concern of statistical analyses and caution must be taken when interpreting results involving events B1 and B3. Both events have group sample sizes of 6 turtles. Event B1 was shown to have a statistically sign ificant orientation toward the sea (0o), although judging from the raw plot this may not be the case (Fig ure 2-6). When sample size, statistical method, and visual review of raw data are all taken in to account, one can make a more accurate judgment of a nest shields effect on orie ntating hatchlings. With all te sts considered, it is likely that events B2 and G displayed relevant nest shield ing effects on orientati on (Figures 2-7, 2-15). Additional factors affecting the orientation tr ial distributions are di scussed in Chapter 3.

PAGE 23

23 Table 2-1. Hatchling orientati on trial site de scription, 2004-2005 Site ID Location Beach Latitude Longitude Artificial Light Source A TAFB Crooked Island 29 58 18.75 85 30 32.68 PC glow, TAFB glow B TAFB NCO Beach 30 03 14.08 85 36 01.45 PC glow, TAFB lights C TAFB Shell Island 30 03.475 85 36.541 PC glow D TAFB Crooked Island 29 57 21.01 85 27 03.32 Mexico Beach, TAFB glow E Boca Raton Red Reef Park 26 21 59.28 80 04 05.27 Boca glow F Boca Raton N Boca Inlet 26 20 36.42 80 04 13.66 Boca spotlight G Boca Raton Spanish River 26 23 11.29 80 04 00.01 SR Road/A1A streetlight

PAGE 24

24 Table 2-2. Summary of e xperimental design, 2004-2005 # Hatchlings Date Event ID Control Shield Total 18-Aug-04 A 9 9 18 23-Aug-04 B1 6 0 6 25-Aug-04 C 23 23 46 29-Aug-04 B2 12 12 24 04-Sep-04 D 20 20 40 08-Sep-04 B3 6 6 12 20-Jul-05 E1 20 20 40 21-Jul-05 F1 10 10 20 02-Aug-05 F2 14 14 28 03-Aug-05 E2 15 0 15 04-Aug-05 G 10 14 24

PAGE 25

25 Figure 2-1. Map of Florida displaying Tyndall Air Force Base (TAFB) and Boca Raton

PAGE 26

26 Figure 2-2. Aerial image of Tyndall Air Force Base (TAFB), located in Bay County on the northwestern coast of Florida. TAFB consis ts of (from east to west) Crooked Island East Beach (CIE), NCO Beach (NCO), and Shell Island. Orientation trials were conducted at the four locations indicated by the bullets on the map, and designated by site name

PAGE 27

27 Figure 2-3. Aerial imag e of Boca Raton, located in Palm Beach County, FL on the southeastern coast of Florida. Orientation trials were conducted at the three locations indicated by the bullets on the map, and designated by site name

PAGE 28

28 Figure 2-4. Hatchling orientati on arenas (2m and 8m radii). Di splays hatchling release point, exit angles, and placement of nest shield (shown as bold arc).

PAGE 29

29 A B Figure 2-5. Orientation of hatchli ng loggerhead turtles in sampling ev ent A. Each triangle within a circular diagram represents the exit angle for a single turtle. Each turtle had 2 exit angles, corresponding to the 2m and 8m aren as. Control groups are represented in blue and treatment (shielded) groups are in red. Solid arrows represent the mean resultant vector (r) for each group. A) 2m arena exit angles. The 2m control group was significantly oriented with a mean angle of 20 (n= 9, r=0.897, p<0.05 Rayleighs test). The 2m treatment group was significan tly oriented with a mean angle of 30 (n=9, r = 0.774, p<0.05 Rayleighs test). B) 8m arena exit angles from the same groups of turtles. The distribution for th e 8m control group was significantly oriented with a mean angle of 22 (n=9, r=0.906, p< 0.05 Rayleighs test). The distribution for the 8m treatment group was significantly or iented with a mean angle of 20 (n=9, r=0.863, p<0.05 Rayleighs test).

PAGE 30

30 A B Figure 2-6. Orientation of hatch ling loggerhead turtles in samp ling event B1. Each triangle within a circular diagram represents the exit angle for a single turtle. Each turtle had 2 exit angles, corresponding to the 2m and 8m arenas. Control groups are represented in blue and treatment (shielde d) groups are in red. Solid arrows represent the mean resultant vector (r) for each group. A) 2m arena exit angles. The 2m control group was significantly oriented with a mean angle of 322 (n=6, r=0.914, p<0.005 Rayleighs test). B) 8m arena exit angles from the same groups of turtles. The distribution for the 8m control group was signi ficantly oriented with a mean angle of 319 (n=6, r=0.913, p<0.001 Rayleighs test). Treatment trials were not conducted due to lack of hatchlings or inclement weather.

PAGE 31

31 A B Figure 2-7. Orientation of hatch ling loggerhead turtles in samp ling event B2. Each triangle within a circular diagram represents the exit angle for a single turtle. Each turtle had 2 exit angles, corresponding to the 2m and 8m arenas. Control groups are represented in blue and treatment (shielde d) groups are in red. Solid arrows represent the mean resultant vector (r) for each group. A) 2m arena exit angles. The 2m control group failed to significantly orient with a mean angle of 149 (n=12, r=0.446, p<0.05 Rayleighs test). The 2m treatment group wa s not significantly or iented with a mean angle of 359 (n=12, r =0.459, p<0.05 Rayleighs te st). B) 8m arena exit angles from the same groups of turtles. The dist ribution for the 8m control group was not significantly oriented with a mean an gle of 168 (n=12, r=0.436, p<0.05 Rayleighs test). The distribution for the 8m treatment group was not significantly oriented with a mean angle of 14 (n=12, r= 0.385, p<0.05 Rayleighs test).

PAGE 32

32 A B Figure 2-8. Orientation of hatch ling loggerhead turtles in samp ling event B3. Each triangle within a circular diagram represents the exit angle for a single turtle. Each turtle had 2 exit angles, corresponding to the 2m and 8m arenas. Control groups are represented in blue and treatment (shielde d) groups are in red. Solid arrows represent the mean resultant vector (r) for each group. A) 2m arena exit angles. The 2m control group was significantly oriented with a mean angle of 176 (n=6, r=0.986, p<0.001 Rayleighs test). The 2m treatment group wa s not significantly or iented with a mean angle of 198 (n=6, r=0.228, p>0.05 Rayleighs test). B) 8m arena exit angles from the same groups of turtles. The di stribution for the 8m control group was significantly oriented with a mean an gle of 172 (n=6, r=0.974, p<0.001 Rayleighs test). The distribution for the 8m treatment group was significantly oriented with a mean angle of 181 (n=6, r=0.893, p<0.05 Rayleighs test).

PAGE 33

33 A B Figure 2-9. Orientation of hatchli ng loggerhead turtles in sampling ev ent C. Each triangle within a circular diagram represents the exit angle for a single turtle. Each turtle had 2 exit angles, corresponding to the 2m and 8m aren as. Control groups are represented in blue and treatment (shielded) groups are in red. Solid arrows represent the mean resultant vector (r) for each group. A) 2m arena exit angles. The 2m control group was significantly oriented with a mean angle of 10 (n=23, r=0.973, p<0.001 Rayleighs test). The 2m treatment group was significantly oriented with a mean angle of 1 (n=23, r=0.98, p<0.001 Rayleighs test). B) 8m arena exit angles from the same groups of turtles. Th e distribution for the 8m control group was significantly oriented with a mean angle of 9 (n =23, r=0.957, p<0.001 Raylei ghs test). The distribution for the 8m treatment group was significantly oriented with a mean angle of 1 (n=23, r=0.97, p<0.001 Rayleighs test).

PAGE 34

34 A B Figure 2-10. Orientation of hatc hling loggerhead turtles in samp ling event D. Each triangle within a circular diagram represents the exit angle for a single turtle. Each turtle had 2 exit angles, corresponding to the 2m and 8m arenas. Control groups are represented in blue and treatment (shielde d) groups are in red. Solid arrows represent the mean resultant vector (r) for each group. (A) 2m arena exit angles. The 2m control group was significantly oriented with a mean angle of 352 (n=20, r=0.952, p<0.001 Rayleighs test). The 2m treatment group was significantly oriented with a mean angle of 349 (n=20, r=0.909, p<0.001 Rayleigh s test). (B) 8m arena exit angles from the same groups of turtles. The distribution for the 8m control group was significantly oriented with a mean an gle of 353 (n=20, r=0.98, p<0.001 Rayleighs test). The distribution for the 8m treatment group was significantly oriented with a mean angle of 351 (n=20, r= 0.971, p<0.001 Rayleighs test).

PAGE 35

35 A B Figure 2-11. Orientation of hatc hling loggerhead turtles in samp ling event E1. Each triangle within a circular diagram represents the exit angle for a single turtle. Each turtle had 2 exit angles, corresponding to the 2m and 8m arenas. Control groups are represented in blue and treatment (shielde d) groups are in red. Solid arrows represent the mean resultant vector (r) for each group. A) 2m arena exit angles. The 2m control group was significantly oriented with a mean angle of 358 (n=20, r=0.954, p<0.001 Rayleighs test). The 2m treatment group was significantly oriented with a mean angle of 358 (n=20, r=0.974, p<0.001 Rayleighs te st). B) 8m arena exit angles from the same groups of turtles. The di stribution for the 8m control group was significantly oriented with a mean angl e of 0 (n=20, r=0.956, p<0.001 Rayleighs test). The distribution for the 8m treatment group was significantly oriented with a mean angle of 0 (n=20, r=0.962, p<0.001 Rayleighs test).

PAGE 36

36 A B Figure 2-12. Orientation of hatc hling loggerhead turtles in samp ling event E2. Each triangle within a circular diagram represents the exit angle for a single turtle. Each turtle had 2 exit angles, corresponding to the 2m and 8m arenas. Control groups are represented in blue and treatment (shielde d) groups are in red. Solid arrows represent the mean resultant vector (r) for each group. A) 2m arena exit angles. The 2m control group was significantly oriented with a mean angle of 11 (n=15, r=0.976, p<0.001 Rayleighs test). B) 8m arena exit angles from the same groups of turtles. The distribution for the 8m control group was signi ficantly oriented with a mean angle of 12 (n=15, r=0.986, p<0.001 Rayleighs test). Treatment trials were not conducted due to lack of hatchlings or inclement weather.

PAGE 37

37 A B Figure 2-13. Orientation of hatc hling loggerhead turtles in samp ling event F1. Each triangle within a circular diagram represents the exit angle for a single turtle. Each turtle had 2 exit angles, corresponding to the 2m and 8m arenas. Control groups are represented in blue and treatment (shielde d) groups are in red. Solid arrows represent the mean resultant vector (r) for each group. A) 2m arena exit angles. The 2m control group was significantly oriented with a mean angle of 4 (n=10, r=0.963, p<0.001 Rayleighs test). The 2m treatment group was significantly oriented with a mean angle of 357 (n=10, r = 0.949, p<0.001 Rayleighs test). B) 8m arena exit angles from the same groups of turtles. The distribution for the 8m control group was significantly oriented with a mean angl e of 3 (n=10, r=0.971, p<0.001 Rayleighs test). The distribution for the 8m treatment group was significantly oriented with a mean angle of 9 (n=10, r=0.964, p<0.001 Rayleighs test).

PAGE 38

38 A B Figure 2-14. Orientation of hatc hling loggerhead turtles in samp ling event F2. Each triangle within a circular diagram represents the exit angle for a single turtle. Each turtle had 2 exit angles, corresponding to the 2m and 8m arenas. Control groups are represented in blue and treatment (shielde d) groups are in red. Solid arrows represent the mean resultant vector (r) for each group. A) 2m arena exit angles. The 2m control group was significantly oriented with a mean angle of 337 (n=14, r=0.875, p<0.001 Rayleighs test). The 2m treatment group was significantly oriented with a mean angle of 354 (n=14, r = 0.977, p<0.001 Rayleighs test). B) 8m arena exit angles from the same groups of turtles. The distribution for the 8m control group was significantly oriented with a mean an gle of 342 (n=14, r=0.958, p<0.001 Rayleighs test). The distribution for the 8m treatment group was significantly oriented with a mean angle of 349 (n=14, r= 0.984, p<0.001 Rayleighs test).

PAGE 39

39 A B Figure 2-15. Orientation of hatc hling loggerhead turtles in samp ling event G. Each triangle within a circular diagram represents the exit angle for a single turtle. Each turtle had 2 exit angles, corresponding to the 2m and 8m arenas. Control groups are represented in blue and treatment (shielde d) groups are in red. Solid arrows represent the mean resultant vector (r) for each group. A) 2m arena exit angles. The 2m control group was not significantly oriented with a mean angle of 132 (n=10, r=0.516, p<0.05 Rayleighs test). The 2m treatment group was significantly oriented with a mean angle of 38 (n=14, r = 0.609, p<0.005 Rayleighs test). B) 8m arena exit angles from the same groups of turtles. The dist ribution for the 8m control group was not significantly oriented with a mean angl e of 124 (n=14, r=0.53, p<0.05 Rayleighs test). The distribution for the 8m treatment group was significantly oriented with a mean angle of 30 (n=14, r=0.609, p<0.005 Rayleighs test).

PAGE 40

40 CHAPTER 3 ORIENTATION AND LIGHT CUES Introduction Upon emerging from the nest, visu al cues appear to be an im portant orientation tool used by hatchling sea turtles. Loggerhead and green turtle hatchlings are particularly sensitive to short-wavelength light, such as that reflected from a blue ocean or sky; this sensitivity is thought to allow them to orient towards the ocean under most conditions. Conversely, long-wavelength light has been found to be less attractive to thes e hatchlings. In additi on to wavelength, hatchling turtles also respond to light intensity, orienting towards the brightest area on the horizon if placed in a highly directed light field (Witherington, 1992b). Brightness in this case is defined as light intensity and wavelength as a func tion of the spectral sensitivity of hatchling turtles (Lohmann et al., 1997). Often the brightest area on the horizon is the ocean due to the reflectance of moon and starlight. Artificial light sources are not as intense as celestial sources, but they are located closer to the hatchlings and create intense, directed light fields. Artificial light sources can attract the hatchlings and lead them back towards the land, even if they were previously oriented towards the ocean (Lohmann et al., 1997; Witham, 1982). Light pollution refers to man-made sources of light having conseque nces impacting a wide range of species (Elvidge et al. 1997). Deve loped areas with excessi ve amounts of artificial illumination can produce non-point sources of light and what is known as a skyglow, which may be visible on nesting beaches many miles awa y. Artificial light fields alter critical nocturnal behaviors involving adu lt nest-site selection, how they re turn to the sea after nesting, and, to a much greater extent, the ability of hatc hlings to locate the s ea quickly after exiting the nest (Witherington and Martin, 199 6). Artificial lighting can di srupt a hatchlings seafinding

PAGE 41

41 ability and greatly increase its risk of mortality by predati on, exhaustion, or dehydration (Mann, 1978). To investigate the effects of artificial and natural light on ha tchling orienta tion, directional light intensity measurements and lunar illuminati on were assessed in relation to trials conducted in the previous chapter. Materials and Methods Study Area The study was conducted at four sites on Tynda ll Air Force Base (TAFB) in Bay County, FL and three sites in Boca Raton, FL (Figures 3-1, 3-2). Both s ites were exposed to a variety of artificial light sources (Table 3-1). TAFB spans approximately 25 kilometers of south-facing coastline. TAFB consists of (from east to west) Crooked Island East Beach (CIE), NCO Beach (NCO), and Shell Island. Numerous areas along NCO are exposed to point sources of illumination originating from TAFB facilities. Visi ble from the eastern sections of CIE are point source luminaires of beachfront development with in the neighboring city of Mexico Beach, FL. Both of TAFBs beaches, including Shell Island, are exposed to a pronounced urban glow on the western horizon emanating from the urbanized Pana ma City, FL vicinity. Three additional sites in Boca Raton, FL, were chosen for this study du e to their high loggerhead nesting density and exposure to artificial li ghts. These sites, in Red Reef Par k, North Boca Inlet, and Spanish River Park, are part of a 5-mile stretch of beach that has been monitored for sea turtle activity by the City of Boca Raton since 1976. Visible from Re d Reef Park is a prominent glow from the urbanized areas of Boca Raton, and sites adjacent to the North Boca Inlet and Spanish River Park were within close proximity to direct lighti ng, emanating from a condominium spotlight and a streetlight, respectively.

PAGE 42

42 Light Intensity Measurements Light intensity measurements (irradiance) were taken using an Inte rnational Light IL 1700 research radiometer with a SHD033 high gain si licon detector and L30 high gain lens. The angle of acceptance of the lens is approximate ly 30. Light readings were recorded in watts/cm2. The receiving lens was positioned hori zontally in the arena center, approximately 2 inches from the sand surface, and readings were taken every 30 degrees beginning with the 0 bearing (most direct route to th e sea). Readings were taken us ing two filters to measure two spectral bandwidths: wavelengths of 300-500 nm, a spectral band of light more attractive to loggerhead hatchlings, and wavelengths of 545-700 nm, a spectral band less attractive to loggerhead hatchlings. Orientation trials A full description of orientati on trial procedure is referenced in Chapter 2. Orientation trials were conducted on nights ch aracterized as having either low or high levels of ambient light (Table 3-2). Nights on or around the new moon phase were consider ed to have low levels of ambient light. High ambient light conditions occurred on nights wh en over half of the moon was visible. Light intensity measurements were recorded dur ing hatchling orientation trials. Two sets of measurements were taken for each sampling ev ent (i.e. two sets per night). One set of measurements was recorded prior to releases of tu rtles in control trials (no shield), and another set was taken before shielded trials. TAFB Lighting and Dune Evaluation Artificial lighting and dune evaluations were conducted by request of TAFB Department of Natural Resources, and are discussed inde pendently from experiments in this study.

PAGE 43

43 Lighting surveys were conducted at night on TAFB s beaches to identify visible sources of artificial illumination that had the potential to interfere with sea turtle nesting and hatchling behavior. Information was gathered on the number, location, facility ID, and fixture type of each source of light visible from areas on the beach (Table 3-3). Bulb and fixture type were approximated based on color and shape when information was not available. As part of a dredging project to reopen East Pass on Tyndall Air Force Base, a large primary dune was constructed using the spoil, extending approximately one-half miles on either side of the dredge area. The construction of this manmade dune provided an opportunity to study the ability of such a structur e to block the visibility of arti ficial lights from the beach. Prior to dredging of the old pass, two sites we re identified for compar ing preand post-dune construction measurements of arti ficial light levels. Light measurements were taken May 2 and 5, 2002 from two sites on TAFB: Site 1 (N 30 04.003, W85 37.605), located approximately onehalf miles west of the project ed pass area, and Site 2 (N30 03 31.0, W85 36 34.8), one-half miles east of the projected pass on. Post-dune constr uction measurements were taken from locations within close distance of Sites 1 and 2, on August 17 and 18, 2004 (Tables 3-4, 3-5). Data Analysis Orientation and light intensity data were analyzed to (1) as sess whether groups of turtles were significantly oriented in directions of hi ghest light intensity, (2) determine if significant differences exist between groups of turtles or ienting under high and low levels of lunar illumination, and (3) investigate the role of li ght intensity gradients as orientation cues. Circular statistical methods were used to anal yze these data (Fischer, 1993). A circular V test was used to determine if gr oups of turtles were si gnificantly oriented toward the brightest directions. Watsons U2 tests were used to determine if significant differences exist between groups.

PAGE 44

44 Results Of the 273 turtles (collected from 19 clutches) used in orientation trials, 79 were released on nights with low ambient moonlight and 194 hatchl ings were tested under high intensities of lunar light. Each set of light in tensity measurements was scaled ac cording to highest intensity of that set and plotted along with individual, 2m ar ena hatchling exit angles (Figures 3-3 3-13). Light measurements taken during control trials we re used to calculate intensity gradients (mean landward/mean seaward intensity) to better represent site-specifi c lighting conditions (Table 32). Circular V tests showed that, of the eleven co ntrol treatment distributions tested, five were significantly oriented in the direc tion of highest light intensity. These groups involved events B2 (long wavelength light only), B3, C, D (short wa velength light only), an d G (see Figs. 3-5, 6, 7, 13). Events B2, B3 and G consisted of groups orienting towards bright landward directions, whereas, events C and D were oriented toward the brighter seaward horizon. Trial sites replicated under both lunar conditions were compared to test the effect of lunar phase on hatchling orientation (sit es B, E, and F). Exit angles for control groups in events B1 and B2 (both high lunar light c onditions) were combined and found to be significantly different from those in event B3 (Watsons U2 test, p<0.01). Hatchlings crawling under low lunar illumination at this site were significantly orient ed as a group in the landward direction (mean 174, n=6; Rayleighs test, p<0.01) whereas hatchlings under hi gh amounts of lunar light were not oriented as a group, but also tended to follo w landward paths (mean 247, n=18; Rayleighs test, p<0.05). Shielded groups at site B were also found to be significantly different (Watsons U2 test, p<0.01). When shielded trials were conduc ted at site B, the majority of hatchlings under high lunar light crawled seaward, compared to shielded trials under low lunar light, in which no hatchlings crawled in seaward directions. Contro l groups tested at site E directed seaward and

PAGE 45

45 were not significantly different with resp ect to levels of lunar light (Watsons U2, p<0.05). Lunar groups at site F were not found to be significantly different from each other during shielded trials, but did differ during control tests (Watsons U2 test, p<0.05). Discussion Before evaluating hatchling orientation in respon se to photic cues, it is important to assess the characteristics of light and the logistics of its measurement at th e variety of sites and conditions involved in this experi ment. In Figures 3-3 through 3-13, each bandwidth of light is graphed in proportion to the highe st measured intensity per site night and treatment, giving a more qualitative view of the light fields present at the time of each trial. With exception to one set of measurements, atmospheric conditions re mained relatively stable throughout the trial duration. During event A, cloud cover sporadic ally obscured the moons illumination as light intensity measurements were taken. Another li ghting anomaly involves site F, where a landward spotlight was illuminated during one sampling even t (F1; Figure 3-11) and was found off during the other (F2; Figure 3-12). Quantitatively, ener gy of long wavelength light was, on average, 8.7 times higher than shorter wavelengt h light readings. Average landwar d (90-240) levels of short and long bandwidths of light exceed ed seaward (270-60) levels at every sampling event, with exception to events B2, C, and D. Lunar illumination played a significant role in characterizing light fields during experimental trials. On nights with lower levels of lunar illumination, th e proportion of long to short-wavelength light was 6.4 times greater than that measured during nights with high lunar illumination, suggesting that moonlit nights increa se the amount of s horter-wavelength light which hatchlings are most attracte d to. Directional attributes of light intensity are also important to note. In general, the distri butions of shorter wavelength light (300-500nm) are rather similar to those of longer wavelength light (545-700nm). At replicated sites, the proportion of landward

PAGE 46

46 short-wave light to seaward s hort-wave light decreased by an average magnitude of 1.8 times when exposed to high levels of moonlight than w ith lower lunar levels. This implies that higher levels of moonlight can significantly shift the directionality of light fields. A similar effect is seen when nest shields are used. Nest shield s were effective at blocking the majority of landward light, altering the brightes t direction from landward to seaw ard, or in directions parallel to the sea. It was difficult to generalize hatchling orientatio n in relation to brightest direction or lunar phase, as nearly half of the dist ributions tested showed significan t responses to either condition. Five of the eleven distributions were found to be significantly oriented wi th respect to brightest direction, and an equally varied response was seen when distri butions were tested for lunar effect. This may be the result of low sample si ze relating to the conditions tested, and the fact that lighting, dune landscape, presence of a shie ld, and other factors are simultaneously used as cues by orienting hatchlings. Ch apter 4 will attempt to relate these cues, in combination, to hatchling orientation distributions.

PAGE 47

47 Table 3-1. Artificial light sources visible from experimental sites Site ID Location Artificial light source Light source bearing A TAFB Panama City (PC) and TAFB glow 85, 110 B TAFB PC glow, TAFB lights 90, 170 C TAFB PC glow 90 D TAFB Mexico Beach and TAFB glow 280, 110 E Boca Raton Boca glow 90 through 270 F Boca Raton Spotlight*, Boca glow 180, interspersed** G Boca Raton Streetlight 160 *spotlight was turned ON during event F1, and OFF during event F **dune silhouette alternated be tween darkened, high-rise co ndominiums and urban glow

PAGE 48

48 Table 3-2. Summary of samp ling event light conditions *Land/sea gradient = proportion of landward to s eaward light intensity pe r bandwidth (i.e. how much brighter the direction of land is than sea) EventID Lunar illumination Land/sea gradient* (300-500nm) Land/sea gradient* (545-700nm) Brightest direction (300-500nm) Brightest direction (545-700nm) A High 2.3 1.8 120o 90o B1 High 5.3 5.8 180o 120o B2 High 0.9 1.4 270o 150o B3 Low 6.3 8.9 180o 180o C High 0.5 1.0 0o 0o D High 0.7 0.8 60o 270o E1 High 1.2 1.1 150o 90o E2 Low 2.6 1.5 120o 120o F1 High 4.2 5.7 180o 180o F2 Low 3.2 4.2 150o 120o G Low 8.1 18.5 150o 150o

PAGE 49

49 Table 3-3. Artificial light source s observed from TAFB beaches *lights appeared clustered togeth er and were grouped accordingly Light ID Lighting type Bulb Fixture type Qty Location Drone launch wall-mounted area lighting MH open bottom 6 subscale drone launch facility TAFB lights* stadium lighting HPS multi-bulb 6 football field (Mississippi Road) TAFB lights* parking lot HPS Cobra head 5-10 parking lots of Burger King and Commissary TAFB lights* roadway lighting MH open bottom 9 Mississippi Road TAFB lights* roadway lighting MH open bottom 5 unknown roadway at NCO Beach entrance TAFB lights* wall-mounted area lighting LPS wall pak 4 Commissary: south facing wall TAFB glow glow n/a n/a n/a TAFB vicinity PC glow glow n/a n/a n/a west of TAFB (Panama City vicinity) Mexico Beach various variousvarious unk. east of TAFB (Mexico Beach vicinity) Marina parking lot, wall-mounted unk. unk. unk. Beacon Beach Road

PAGE 50

50 Table 3-4. Light intensity measuremen ts at Site 1, Tyndall Air Force Base Pre-dune construction Post-dune construction Bearing 300-500nm 545-700nm 300-500nm 545-700nm 0 0.47 1.74 0.87 3.20 45 0.81 3.90 1.18 6.82 90 0.68 2.95 0.85 4.70 135 0.37 1.29 0.16 0.75 180 0.34 1.15 0.26 1.13 225 0.41 1.45 0.32 0.25 270 0.85 3.47 0.96 4.76 315 0.75 3.42 1.25 6.20 Light intensity units = watts/cm2

PAGE 51

51 Table 3-5. Light intensity measuremen ts at Site 2, Tyndall Air Force Base Pre-dune construction Post-dune construction Bearing 300-500nm 545-700nm 300-500nm 545-700nm 0 11.45 22.40 9.67 16.30 45 5.08 56.50 8.32 28.50 90 0.75 3.90 0.48 3.10 135 0.55 3.03 0.32 3.65 180 0.54 3.25 0.52 1.50 225 0.55 3.41 0.27 2.74 270 1.10 5.34 1.52 10.40 315 2.70 9.67 7.58 13.10 Light intensity units = watts/cm2

PAGE 52

52 Figure 3-1. Aerial imag e of Tyndall Air Force Base, locate d in Bay County on the northwestern coast of Florida. TAFB consists of (f rom east to west) Crooked Island East Beach (CIE), NCO Beach (NCO), and Shell Island. Orientation trials were conducted at the four locations indicated by the bullets on the map, and designated by site name. Numerous areas along NCO are exposed to poi nt sources of illumination originating from TAFB facilities. Visible from the eastern sections of CIE are point source luminaires of beachfront development with in the neighboring city of Mexico Beach, FL. Both of TAFBs beaches, includi ng Shell Island, are exposed to a pronounced skyglow on the western horizon emanating from the urbanized Panama City, FL vicinity. Site A, located on CIE, was expos ed to artificial ligh ts originating from Panama City to the west (skyglow), and the landward skyglow from TAFB facilities. Site B, located on NCO, was exposed to westward Panama City skyglow, and a variety of directly visible TA FB luminaires. Site C, located on Shell Island, also received artificial light as skyglow from the west. Site D, on the eastern end of CIE, was exposed to direct lighting from the neighboring city of Mexico Beach, and skyglow originating from TAFB.

PAGE 53

53 Figure 3-2. Aerial imag e of Boca Raton, located in Palm Beach County, FL on the southeastern coast of Florida. Orientation trials were conducted at the three locations indicated by the bullets on the map, and designated by site name. Site E, adjacent to Red Reef Park, receives artifici al light exposure as skyglow from the city of Boca Raton. Site F, located just north of the Boca Inlet, wa s exposed to a seaward facing spotlight, and mounted at the base of a condominium. Site G, was exposed to a street light mounted at the intersection of Spanish River Road and A1A.

PAGE 54

54 A B Figure 3-3. Orientation a nd light intensity at sampling event A. Each triangle within a circular diagram represents the 2m arena exit angl e for a single turtle. Control groups are represented in blue and treatment (shielde d) groups are in red. Directional 300500nm (blue bars) and 545-700nm (blue arrows ) light intensity r eadings (watts/cm2) are shown radiating from the arena center. The solid line represents the orientation groups mean angle. A) Control group orie ntation angles and li ght intensities. B) Treatment group orientation an gles and light intensities.

PAGE 55

55 Figure 3-4. Orientation and light in tensity at sampling event B1. E ach triangle within a circular diagram represents the 2m arena exit angl e for a single turtle. Control groups are represented in blue and treatment (shielde d) groups are in red. Directional 300500nm (blue bars) and 545-700nm (blue arrows ) light intensity r eadings (watts/cm2) are shown radiating from the arena center. The solid line represents the orientation groups mean angle. A) Control group orie ntation angles and li ght intensities. B) Treatment group orientation an gles and light intensities.

PAGE 56

56 A B Figure 3-5. Orientation and light in tensity at sampling event B2. E ach triangle within a circular diagram represents the 2m arena exit angl e for a single turtle. Control groups are represented in blue and treatment (shielde d) groups are in red. Directional 300500nm (blue bars) and 545-700nm (blue arrows ) light intensity r eadings (watts/cm2) are shown radiating from the arena center. The solid line represents the orientation groups mean angle. A) Control group orie ntation angles and li ght intensities. B) Treatment group orientation an gles and light intensities.

PAGE 57

57 A B Figure 3-6. Orientation and light in tensity at sampling event B3. E ach triangle within a circular diagram represents the 2m arena exit angl e for a single turtle. Control groups are represented in blue and treatment (shielde d) groups are in red. Directional 300500nm (blue bars) and 545-700nm (blue arrows ) light intensity r eadings (watts/cm2) are shown radiating from the arena center. The solid line represents the orientation groups mean angle. A) Control group orie ntation angles and li ght intensities. B) Treatment group orientation an gles and light intensities.

PAGE 58

58 A B Figure 3-7. Orientation a nd light intensity at sampling event C. Each triangle within a circular diagram represents the 2m arena exit angl e for a single turtle. Control groups are represented in blue and treatment (shielde d) groups are in red. Directional 300500nm (blue bars) and 545-700nm (blue arrows ) light intensity r eadings (watts/cm2) are shown radiating from the arena center. The solid line represents the orientation groups mean angle. A) Control group orie ntation angles and li ght intensities. B) Treatment group orientation an gles and light intensities.

PAGE 59

59 A B Figure 3-8. Orientation a nd light intensity at sampling event D. Each triangle within a circular diagram represents the 2m arena exit angl e for a single turtle. Control groups are represented in blue and treatment (shielde d) groups are in red. Directional 300500nm (blue bars) and 545-700nm (blue arrows ) light intensity readings (watts/cm2) are shown radiating from the arena center. The solid line represents the orientation groups mean angle. A) Control group orie ntation angles and li ght intensities. B) Treatment group orientation an gles and light intensities.

PAGE 60

60 A B Figure 3-9. Orientation and light in tensity at sampling event E1. E ach triangle within a circular diagram represents the 2m arena exit angl e for a single turtle. Control groups are represented in blue and treatment (shielde d) groups are in red. Directional 300500nm (blue bars) and 545-700nm (blue arrows ) light intensity readings (watts/cm2) are shown radiating from the arena center. The solid line represents the orientation groups mean angle. A) Control group orie ntation angles and li ght intensities. B) Treatment group orientation an gles and light intensities.

PAGE 61

61 Figure 3-10. Orientation and light intensity at sampling event E2. E ach triangle within a circular diagram represents the 2m arena exit angl e for a single turtle. Control groups are represented in blue and treatment (shielde d) groups are in red. Directional 300500nm (blue bars) and 545-700nm (blue arrows ) light intensity r eadings (watts/cm2) are shown radiating from the arena center. The solid line represents the orientation groups mean angle.

PAGE 62

62 A B Figure 3-11. Orientation and light intensity at sampling event F1. E ach triangle within a circular diagram represents the 2m arena exit angl e for a single turtle. Control groups are represented in blue and treatment (shielde d) groups are in red. Directional 300500nm (blue bars) and 545-700nm (blue arrows ) light intensity r eadings (watts/cm2) are shown radiating from the arena center. The solid line represents the orientation groups mean angle. A) Control group orie ntation angles and li ght intensities. B) Treatment group orientation an gles and light intensities.

PAGE 63

63 A B Figure 3-12. Orientation and light intensity at sampling event F2. E ach triangle within a circular diagram represents the 2m arena exit angl e for a single turtle. Control groups are represented in blue and treatment (shielde d) groups are in red. Directional 300500nm (blue bars) and 545-700nm (blue arrows ) light intensity r eadings (watts/cm2) are shown radiating from the arena center. The solid line represents the orientation groups mean angle. A) Control group orie ntation angles and li ght intensities. B) Treatment group orientation an gles and light intensities.

PAGE 64

64 A B Figure 3-13. Orientation and light intensity at sampling event G. Each triangle within a circular diagram represents the 2m arena exit angl e for a single turtle. Control groups are represented in blue and treatment (shielde d) groups are in red. Directional 300500nm (blue bars) and 545-700nm (blue arrows ) light intensity r eadings (watts/cm2) are shown radiating from the arena center. The solid line represents the orientation groups mean angle. A) Control group orie ntation angles and li ght intensities. B) Treatment group orientation an gles and light intensities.

PAGE 65

65 CHAPTER 4 ORIENTATION AND MULTIPL E CONFLICTING CUES Introduction The response of hatchlings to light cues has been studied in much more depth than has their ability to orient with respec t to shape cues. The principal reason for this paucity of data is revealed in the inherent charac teristics of the coastal environment. A natural dune profile is typically higher, with a more va riable silhouette than the lowe r, flatter seaward horizon, and tends to absorb more light than would the surface of the sea. Herein lies the problem with experimentally assessing the relative impor tance of shape and li ght cues; under natural conditions these cues are rather complex. When a hatchling orients away fr om the darker, higher dune horizon, and crawls toward the often brighter and flatter seaward dir ection, it is difficult to determine whether the hatchlings response was dominated by its attraction to the brightest direction (light cue) or, a result of being repulsed by the higher, spatially variable dune (shape cue). Scientists have regarded shape and lig ht stimuli as conflicting cues, suggesting that a hatchlings overall reaction depe nds on the relative strength of the cues. This was evidenced when hatchlings were found to initially orient away from vertically st riped horizons until light intensity of the opposing, unmarked horizon wa s made five times brighter (Witherington, 1992a). Experiments involving naturally occurring cue c onflicts (light intensity vs. horizon profile) have provided significant results, as the coastal e nvironment consists of a multitude of coexisting environmental cues. This chapter will evalua te the use of a logist ic regression model to determine the probability of a hatchling failing to orient properly when exposed to multiple, visual cues in a natural environment.

PAGE 66

66 Materials and Methods A biased-reduced logistic regression appr oach was applied to data gathered during hatchling orientation trials refere nced in the previous chapters. Individual hatchling exit angles were transformed to create the binary response variable, disorien ted (y/n). Hatchlings were considered disoriented if thei r angle upon exiting the 2m aren a fell outside of a 90 range directed toward the sea (315 45 ). If the exit angle was greate r than 45 from the most direct route to sea, the hatchling wa s categorized as yes under the response vari able, disoriented. Independent variables used in th is test were both categorical and continuous, including site, shielding treatment (control or treatment), lunar illumination (l ow or high), 300-500nm intensity gradient, 545-700nm intensity gradient, and wave length gradient (Appendix A). Calculation of intensity gradients is discussed in Chapte r 3. Wavelength gradient was calculated by determining the ratio of mean 545-700nm intens ity readings to mean 300-500nm readings at each site. The slope and intercept of the best-f itting equation in the logistic regression were found using the penalized maximum-likelihood me thod. Analyses were conducted using the statistical program R. Results The intensity and wavelength gradients were dropped from the analysis, as they were considered replicates of the site, treatment, and lunar illumination variable s. Due to lack of replication and small sample sizes, the variables site, treatment, and lunar illumination had to be combined to run the reduced-bias logistic regression model. Probability estimates for the combined parameters site/treatment/lunar illumi nation, including estimates of regression error variance and confidence intervals are given in Table 4-1. The te st results suggest that it is probable a hatchling will disorien t at site G under low lunar il lumination, with or without a shield present, and at site B under most conditions (p-values in bold, Table 4-1).

PAGE 67

67 Discussion In this example, we categorized individual hatchlings as disoriented or not disoriented by whether or not they deviated from the seawar d route (0) by 45 or more. We chose our disorientation criteria arbitrarily, as our main goal was to test the applicability of using the logistic regression approach. The separation of interrelated environmental cues and their effects on hatchling orientation continues to be a difficult task. For this reas on, attempts were made to identify an analytical technique that would take into account the influence of multiple variables on orientation response. I decided to use a bi as-reduced logistic regression approach, which would allow for testing the relationship of disorientation probability to a variety of factors, such as site, treatment, lunar illumination, and lighting gradient. Applying th is approach to the data gathered in this study allowed us to determine the probability of a hatchling disorienting in relation to the single, combined variable, site/treatment/lunar illumina tion. This test fails to account for multiple variables at once due to the a posteriori analytical approach used in this study. A more appropriate research design woul d allow for greater sample sizes by replicating sampling events on both high and low moonlit nights, and incorporate fewer experimental sites.

PAGE 68

68 Table 4-1. Logistic regression output for the combined site/treatment/lunar variable. Site/treatment/lunar variable combinations in bold are significan tly correlated with the disorientation of a hatchling. For ex ample, there is a high likelihood that a hatchling will disorient at Site G under low ambient lunar light, with or without a nest shield present. Site/treatment/lunar variable coef se(coef) lower 0.95 upper 0.95 Chisq P-value (Intercept) -3.714 1.467 -8.559 -1.733 24.387 0.000001 E, Control, Low 0.280 2.086 -4.965 5.525 0.019 0.890518 E, Treatment, High 0.000 2.074 -5.241 5.241 0.000 1.000000 F, Control, High 0.669 2.111 -4.583 5.922 0.106 0.744444 F, Control, Low 2.524 1.597 0.069 7.459 4.092 0.043085 F, Treatment, High 0.669 2.111 -4.583 5.922 0.106 0.744444 F, Treatment, Low 0.346 2.090 -4.900 5.593 0.029 0.864808 G, Control, Low* 4.937 1.649 2.467 9.918 21.508 0.000004 G, Treatment, Low* 3.445 1.563 1.141 8.358 10.215 0.001393 A, Control, Low 1.979 1.739 -1.050 7.004 1.652 0.198740 A, Treatment, Low 3.095 1.625 0.589 8.047 6.167 0.013015 B, Control, High* 4.612 1.556 2.350 9.522 24.206 0.000001 B, Control, Low* 6.279 2.160 3.139 12.053 23.163 0.000001 B, Treatment, High* 3.714 1.576 1.378 8.635 11.827 0.000584 B, Treatment, Low* 6.279 2.160 3.139 12.053 23.163 0.000001 C, Control, High -0.137 2.069 -5.376 5.103 0.005 0.946186 C, Treatment, High -0.137 2.069 -5.376 5.103 0.005 0.946186 D, Control, High 1.149 1.704 -1.843 6.156 0.548 0.459033 D, Treatment, High 1.712 1.621 -0.879 6.664 1.558 0.211940

PAGE 69

69 CHAPTER 5 SUMMARY Results The effects of nest shielding were ascertained on a site-by-site and ni ght-by-night basis, in an attempt to reduce the likelihood of light cu es acting as overriding, confounding factors when comparing distributions of hatchli ng orientation angles. The nest shield appeared to significantly restore seaward orientation in groups of hatch lings during two of the nine testable sampling events. The most discernible effect was observe d at a site where hatchlings were previously observed misorienting toward a si ngle, landward NEMA-head street light, located directly behind the dune. Shielding created a shadow stretching to ward the sea, within wh ich hatchlings crawled until reaching the shoreline. In most cases, how ever, nest shielding failed to restore proper orientation. Results suggest that hatchling orientation does not drastically change with respect to a turtles distance from a nest shield, although more work is needed to determine the effects of a hatchlings initial distance from such a barrier. It appeared as though nest shields act more as visual barriers than physical obstructions to crawling hatchlin gs. Hatchlings were never observed coming in contact with th e shielding material and crawli ng along its length. Hatchlings attracted to landward light were able to aver t the shield by crawling around it without becoming physically blocked by it. It was difficult to generalize hatchling orientatio n in relation to brightest direction or lunar phase, as nearly half of the dist ributions tested showed significan t responses to either condition. Five of the eleven distributions were found to be significantly oriented wi th respect to brightest direction, and an equally varied response was seen when distri butions were tested for lunar effect. Testing the effect of combinations of environmental variables on orientation response

PAGE 70

70 was attempted, revealing the need for further st udy into similar statisti cal models, including the bias-reduced logistic regression approach. Management Implications Current management strategies fall into one of two groups based on their approach to protect hatchlings from artifi cial lighting. One strategy uses a hatchling-centered approach, seeking to prevent hatchlings from being a ffected by artificial lighting. This involves manipulation of the nest and/or hatchlings and includes management practices such as nest relocation, and nest caging. The us e of nest shields would also fall under this category. These strategies tend to be the leas t successful methods of management. In efforts to remove hatchlings from artificially lit areas or prevent them from responding to the lights, new problems are often created for the hatchlings and th e cause of the problem is not addressed. This study has shown that nest shielding is a poor technique for managing the impact of artificial lighting on hatc hling turtles. Shielding failed to reduce hatchling disorientation during seven of nine testable sampling events. Of th e two events with positiv e results, only one found the shield to restore orientation to the majority of hatchlings in the group. Conditions of this trial involved the blocking of light from a single, landward lumi naire that was close enough to the release site and in alignment with the shield such that a shadow was cast nearly all the way to the sea. Additional shielding tests under similar conditions are needed. In cases where optimal management techniques are not an option, altern ative methods may be justified. However, the very limited success of shielding nests to prev ent hatchling disorienta tion is evidence enough to exclude it from the list of possible, alte rnative methods of light management. The other approach to the lighting predicament seeks to directly manage the cause of the problem, often involving the atte nuation, redirection, or replacemen t of individual light sources. Light management techniques have been used throughout Florida with varying success, and

PAGE 71

71 serve to not only reduce the impact on hatchlings, but on nesting a dults as well. In Florida, artificial light pollution from on-going coastal development will continue to cause problems to wildlife, but cooperation from government agen cies, environmental groups, county regulators, and the concerned public has been quite evident. Pub lic awareness of the effects of lighting on sea turtles has grown, and most coastal countie s have adopted strict lighting ordinances. Although the enforcement of coastal lighting restrictions is a probl em in some areas more than others, positive strides continue to be made. In localized areas, comprehensive light mana gement plans have been applied within reasonable bounds of time, funding, and effort to significantly reduce disorientation. Additional levels of difficulty exist with managing light pollution produced by more urbanized areas, often resulting in skyglow, or with ar tificial light originating from ar eas farther inland. Light pollution from these areas can cause problems for nesting a nd hatching turtles from great distances away. Often, the sheer number of luminaires in these re gions precludes the use of traditional, light management methods. Larger scale management techniques and statewid e compliance would be required for effective mediation. Currently, much work is needed in Florida to resolve the lighting problems in coastal areas, but with co ntinued cooperation and public support, successful light management practices on a local level could se rve as a template for future statewide efforts.

PAGE 72

72 APPENDIX LOGISTIC REGRESSION TABLE Table A-1. Hatchling orientation data formatted for logistic regr ession analysis. Each hatchling used in orientation trials was given a sample number. Intensity gradient refers to the ratio of landward to seaward light intens ity for a given bandwidth at each site. Wavelength gradient is the ratio of mean light intensity at 300-500nm to mean intensity at 545-700nm at each site. Sample # Site Treatment Lunar 300-500nm intensity gradient 545-700nm intensity gradient Wavelength gradient Disoriented 1 A C L 2.27 1.80 3.77 N 2 A C L 2.27 1.80 3.77 N 3 A C L 2.27 1.80 3.77 N 4 A C L 2.27 1.80 3.77 N 5 A C L 2.27 1.80 3.77 N 6 A C L 2.27 1.80 3.77 N 7 A C L 2.27 1.80 3.77 N 8 A C L 2.27 1.80 3.77 N 9 A C L 2.27 1.80 3.77 Y 10 A T L 0.49 0.27 2.66 Y 11 A T L 0.49 0.27 2.66 N 12 A T L 0.49 0.27 2.66 N 13 A T L 0.49 0.27 2.66 Y 14 A T L 0.49 0.27 2.66 N 15 A T L 0.49 0.27 2.66 N 16 A T L 0.49 0.27 2.66 N 17 A T L 0.49 0.27 2.66 N 18 A T L 0.49 0.27 2.66 Y 19 B C H 5.32 5.83 4.64 N 20 B C H 5.32 5.83 4.64 N 21 B C H 5.32 5.83 4.64 Y 22 B C H 5.32 5.83 4.64 N 23 B C H 5.32 5.83 4.64 N 24 B C H 5.32 5.83 4.64 N 25 B C H 0.95 1.41 3.02 Y 26 B C H 0.95 1.41 3.02 Y 27 B C H 0.95 1.41 3.02 Y 28 B C H 0.95 1.41 3.02 Y 29 B C H 0.95 1.41 3.02 Y 30 B C H 0.95 1.41 3.02 Y 31 B C H 0.95 1.41 3.02 Y 32 B C H 0.95 1.41 3.02 Y 33 B C H 0.95 1.41 3.02 Y 34 B C H 0.95 1.41 3.02 Y 35 B C H 0.95 1.41 3.02 Y 36 B C H 0.95 1.41 3.02 Y 37 B T H 0.77 0.82 2.46 Y 38 B T H 0.77 0.82 2.46 Y 39 B T H 0.77 0.82 2.46 Y

PAGE 73

73 Table A-1. Continued Sample # Site Treatment Lunar 300-500nm intensity gradient 545-700nm intensity gradient Wavelength gradient Disoriented 40 B T H 0.77 0.82 2.46 N 41 B T H 0.77 0.82 2.46 N 42 B T H 0.77 0.82 2.46 N 43 B T H 0.77 0.82 2.46 Y 44 B T H 0.77 0.82 2.46 Y 45 B T H 0.77 0.82 2.46 Y 46 B T H 0.77 0.82 2.46 N 47 B T H 0.77 0.82 2.46 N 48 B T H 0.77 0.82 2.46 N 49 B C L 6.28 8.94 5.97 Y 50 B C L 6.28 8.94 5.97 Y 51 B C L 6.28 8.94 5.97 Y 52 B C L 6.28 8.94 5.97 Y 53 B C L 6.28 8.94 5.97 Y 54 B C L 6.28 8.94 5.97 Y 55 B T L 0.51 0.59 4.20 Y 56 B T L 0.51 0.59 4.20 Y 57 B T L 0.51 0.59 4.20 Y 58 B T L 0.51 0.59 4.20 Y 59 B T L 0.51 0.59 4.20 Y 60 B T L 0.51 0.59 4.20 Y 61 C C H 0.54 0.99 3.38 N 62 C C H 0.54 0.99 3.38 N 63 C C H 0.54 0.99 3.38 N 64 C C H 0.54 0.99 3.38 N 65 C C H 0.54 0.99 3.38 N 66 C C H 0.54 0.99 3.38 N 67 C C H 0.54 0.99 3.38 N 68 C C H 0.54 0.99 3.38 N 69 C C H 0.54 0.99 3.38 N 70 C C H 0.54 0.99 3.38 N 71 C C H 0.54 0.99 3.38 N 72 C C H 0.54 0.99 3.38 N 73 C C H 0.54 0.99 3.38 N 74 C C H 0.54 0.99 3.38 N 75 C C H 0.54 0.99 3.38 N 76 C C H 0.54 0.99 3.38 N 77 C C H 0.54 0.99 3.38 N 78 C C H 0.54 0.99 3.38 N 79 C C H 0.54 0.99 3.38 N 80 C C H 0.54 0.99 3.38 N 81 C C H 0.54 0.99 3.38 N 82 C C H 0.54 0.99 3.38 N 83 C C H 0.54 0.99 3.38 N 84 C T H 0.29 0.59 3.12 N 85 C T H 0.29 0.59 3.12 N

PAGE 74

74 Table A-1. Continued Sample # Site Treatment Lunar 300-500nm intensity gradient 545-700nm intensity gradient Wavelength gradient Disoriented 86 C T H 0.29 0.59 3.12 N 87 C T H 0.29 0.59 3.12 N 88 C T H 0.29 0.59 3.12 N 89 C T H 0.29 0.59 3.12 N 90 C T H 0.29 0.59 3.12 N 91 C T H 0.29 0.59 3.12 N 92 C T H 0.29 0.59 3.12 N 93 C T H 0.29 0.59 3.12 N 94 C T H 0.29 0.59 3.12 N 95 C T H 0.29 0.59 3.12 N 96 C T H 0.29 0.59 3.12 N 97 C T H 0.29 0.59 3.12 N 98 C T H 0.29 0.59 3.12 N 99 C T H 0.29 0.59 3.12 N 100 C T H 0.29 0.59 3.12 N 101 C T H 0.29 0.59 3.12 N 102 C T H 0.29 0.59 3.12 N 103 C T H 0.29 0.59 3.12 N 104 C T H 0.29 0.59 3.12 N 105 C T H 0.29 0.59 3.12 N 106 C T H 0.29 0.59 3.12 N 107 D C H 0.67 0.78 3.48 Y 108 D C H 0.67 0.78 3.48 N 109 D C H 0.67 0.78 3.48 N 110 D C H 0.67 0.78 3.48 N 111 D C H 0.67 0.78 3.48 N 112 D C H 0.67 0.78 3.48 N 113 D C H 0.67 0.78 3.48 N 114 D C H 0.67 0.78 3.48 N 115 D C H 0.67 0.78 3.48 N 116 D C H 0.67 0.78 3.48 N 117 D C H 0.67 0.78 3.48 N 118 D C H 0.67 0.78 3.48 N 119 D C H 0.67 0.78 3.48 N 120 D C H 0.67 0.78 3.48 N 121 D C H 0.67 0.78 3.48 N 122 D C H 0.67 0.78 3.48 N 123 D C H 0.67 0.78 3.48 N 124 D C H 0.67 0.78 3.48 N 125 D C H 0.67 0.78 3.48 N 126 D C H 0.67 0.78 3.48 N 127 D T H 0.46 0.52 2.74 Y 128 D T H 0.46 0.52 2.74 Y 129 D T H 0.46 0.52 2.74 N 130 D T H 0.46 0.52 2.74 N 131 D T H 0.46 0.52 2.74 N

PAGE 75

75 Table A-1. Continued Sample # Site Treatment Lunar 300-500nm intensity gradient 545-700nm intensity gradient Wavelength gradient Disoriented 132 D T H 0.46 0.52 2.74 N 133 D T H 0.46 0.52 2.74 N 134 D T H 0.46 0.52 2.74 N 135 D T H 0.46 0.52 2.74 N 136 D T H 0.46 0.52 2.74 N 137 D T H 0.46 0.52 2.74 N 138 D T H 0.46 0.52 2.74 N 139 D T H 0.46 0.52 2.74 N 140 D T H 0.46 0.52 2.74 N 141 D T H 0.46 0.52 2.74 N 142 D T H 0.46 0.52 2.74 N 143 D T H 0.46 0.52 2.74 N 144 D T H 0.46 0.52 2.74 N 145 D T H 0.46 0.52 2.74 N 146 D T H 0.46 0.52 2.74 N 147 E C H 1.23 1.13 3.67 N 148 E C H 1.23 1.13 3.67 N 149 E C H 1.23 1.13 3.67 N 150 E C H 1.23 1.13 3.67 N 151 E C H 1.23 1.13 3.67 N 152 E C H 1.23 1.13 3.67 N 153 E C H 1.23 1.13 3.67 N 154 E C H 1.23 1.13 3.67 N 155 E C H 1.23 1.13 3.67 N 156 E C H 1.23 1.13 3.67 N 157 E C H 1.23 1.13 3.67 N 158 E C H 1.23 1.13 3.67 N 159 E C H 1.23 1.13 3.67 N 160 E C H 1.23 1.13 3.67 N 161 E C H 1.23 1.13 3.67 N 162 E C H 1.23 1.13 3.67 N 163 E C H 1.23 1.13 3.67 N 164 E C H 1.23 1.13 3.67 N 165 E C H 1.23 1.13 3.67 N 166 E C H 1.23 1.13 3.67 N 167 E T H 0.45 0.25 2.59 N 168 E T H 0.45 0.25 2.59 N 169 E T H 0.45 0.25 2.59 N 170 E T H 0.45 0.25 2.59 N 171 E T H 0.45 0.25 2.59 N 172 E T H 0.45 0.25 2.59 N 173 E T H 0.45 0.25 2.59 N 174 E T H 0.45 0.25 2.59 N 175 E T H 0.45 0.25 2.59 N 176 E T H 0.45 0.25 2.59 N 177 E T H 0.45 0.25 2.59 N

PAGE 76

76 Table A-1. Continued Sample # Site Treatment Lunar 300-500nm intensity gradient 545-700nm intensity gradient Wavelength gradient Disoriented 178 E T H 0.45 0.25 2.59 N 179 E T H 0.45 0.25 2.59 N 180 E T H 0.45 0.25 2.59 N 181 E T H 0.45 0.25 2.59 N 182 E T H 0.45 0.25 2.59 N 183 E T H 0.45 0.25 2.59 N 184 E T H 0.45 0.25 2.59 N 185 E T H 0.45 0.25 2.59 N 186 E T H 0.45 0.25 2.59 N 187 E C L 2.61 1.47 6.41 N 188 E C L 2.61 1.47 6.41 N 189 E C L 2.61 1.47 6.41 N 190 E C L 2.61 1.47 6.41 N 191 E C L 2.61 1.47 6.41 N 192 E C L 2.61 1.47 6.41 N 193 E C L 2.61 1.47 6.41 N 194 E C L 2.61 1.47 6.41 N 195 E C L 2.61 1.47 6.41 N 196 E C L 2.61 1.47 6.41 N 197 E C L 2.61 1.47 6.41 N 198 E C L 2.61 1.47 6.41 N 199 E C L 2.61 1.47 6.41 N 200 E C L 2.61 1.47 6.41 N 201 E C L 2.61 1.47 6.41 N 202 F C H 4.23 5.66 7.17 N 203 F C H 4.23 5.66 7.17 N 204 F C H 4.23 5.66 7.17 N 205 F C H 4.23 5.66 7.17 N 206 F C H 4.23 5.66 7.17 N 207 F C H 4.23 5.66 7.17 N 208 F C H 4.23 5.66 7.17 N 209 F C H 4.23 5.66 7.17 N 210 F C H 4.23 5.66 7.17 N 211 F C H 4.23 5.66 7.17 N 212 F T H 0.16 0.28 8.19 N 213 F T H 0.16 0.28 8.19 N 214 F T H 0.16 0.28 8.19 N 215 F T H 0.16 0.28 8.19 N 216 F T H 0.16 0.28 8.19 N 217 F T H 0.16 0.28 8.19 N 218 F T H 0.16 0.28 8.19 N 219 F T H 0.16 0.28 8.19 N 220 F T H 0.16 0.28 8.19 N 221 F T H 0.16 0.28 8.19 N 222 F C L 3.18 4.18 11.20 Y 223 F C L 3.18 4.18 11.20 N

PAGE 77

77 Table A-1. Continued Sample # Site Treatment Lunar 300-500nm intensity gradient 545-700nm intensity gradient Wavelength gradient Disoriented 224 F C L 3.18 4.18 11.20 N 225 F C L 3.18 4.18 11.20 N 226 F C L 3.18 4.18 11.20 N 227 F C L 3.18 4.18 11.20 N 228 F C L 3.18 4.18 11.20 N 229 F C L 3.18 4.18 11.20 N 230 F C L 3.18 4.18 11.20 N 231 F C L 3.18 4.18 11.20 N 232 F C L 3.18 4.18 11.20 Y 233 F C L 3.18 4.18 11.20 Y 234 F C L 3.18 4.18 11.20 N 235 F C L 3.18 4.18 11.20 N 236 F T L 0.52 1.09 11.48 N 237 F T L 0.52 1.09 11.48 N 238 F T L 0.52 1.09 11.48 N 239 F T L 0.52 1.09 11.48 N 240 F T L 0.52 1.09 11.48 N 241 F T L 0.52 1.09 11.48 N 242 F T L 0.52 1.09 11.48 N 243 F T L 0.52 1.09 11.48 N 244 F T L 0.52 1.09 11.48 N 245 F T L 0.52 1.09 11.48 N 246 F T L 0.52 1.09 11.48 N 247 F T L 0.52 1.09 11.48 N 248 F T L 0.52 1.09 11.48 N 249 F T L 0.52 1.09 11.48 N 250 G C L 8.06 18.51 43.29 Y 251 G C L 8.06 18.51 43.29 Y 252 G C L 8.06 18.51 43.29 N 253 G C L 8.06 18.51 43.29 Y 254 G C L 8.06 18.51 43.29 Y 255 G C L 8.06 18.51 43.29 Y 256 G C L 8.06 18.51 43.29 Y 257 G C L 8.06 18.51 43.29 N 258 G C L 8.06 18.51 43.29 Y 259 G C L 8.06 18.51 43.29 Y 260 G T L 0.33 0.42 16.01 N 261 G T L 0.33 0.42 16.01 N 262 G T L 0.33 0.42 16.01 N 263 G T L 0.33 0.42 16.01 N 264 G T L 0.33 0.42 16.01 N 265 G T L 0.33 0.42 16.01 N 266 G T L 0.33 0.42 16.01 N 267 G T L 0.33 0.42 16.01 N 268 G T L 0.33 0.42 16.01 Y 269 G T L 0.33 0.42 16.01 Y

PAGE 78

78 Table A-1. Continued Sample # Site Treatment Lunar 300-500nm intensity gradient 545-700nm intensity gradient Wavelength gradient Disoriented 270 G T L 0.33 0.42 16.01 Y 271 G T L 0.33 0.42 16.01 Y 272 G T L 0.33 0.42 16.01 Y 273 G T L 0.33 0.42 16.01 Y

PAGE 79

79 LIST OF REFERENCES Able, K.P., 1977. The orientation of passerine noct urnal migrants following offshore drift. Auk 94, 320-330. Able, K.P., 1995. Orientation and navigation: a pe rspective on fifty years of research. The Condor 97, 592-604. Adamany, S.L., Salmon, M., Witherington, B.E., 1997. Behavior of sea turtles at an urban beach III. Costs and benefits of nest caging as a management strategy. Florida Scientist 60, 239253. Alerstam, T., 1985. Stategies of migratory flig ht, illustrated by arctic and common terns, Sterna paradisaea and Sterna hirundo In: Rankin, M.A. (Ed.), Migration: Mechanisms and Adaptive Significance. Contributions to Marine Science 27 (Supplement), pp. 580-603. Bergen, F., Abs, M., 1997. Etho-ecological study of the singing activity of the blue tit ( Parus caeruleus ), great tit ( Parus major ), and chaffinch ( Fringilla coelebs ). Journal of Ornithology 138, 451-467. Cain, S.D., Boles, L.C., Wang, J.H., Lohmann, K.J., 2005. Magnetic orientation and navigation in marine turtles, lobsters, and mollusks: c oncepts and conundrums. Integral Comparative Biology 45, 539-546. Dickerson, D.D., Reine, K.J., Nelson, D.A., Dick erson, C.E., Jr., 1995. Assessment of sea turtle abundance in six south Atlantic U.S. channe ls. Miscellaneous Paper EL-95-5, U.S. Army Engineer Waterways Experiment Station, Vicksburg, MS. Dingle, H., 1996. Migration: The Biology of Life on the Move. Oxford University Press, New York. Ehrenfeld, D.W., Koch, A.L., 1967. Visual accommodation in th e green turtle. Science 155, 827828. Elvidge, C., Baugh, K.E., Kihn, E.A., Davis, E. R., 1997. Nighttime lights of the world: 1994-95. ISPRS Journal of Photogrammetr y and Remote Sensing 56, 81-99. Fisher, N.I., 1993. Statistical Analys is of Circular Data. Cambridge University Press, Cambridge. Gwinner, D.R., 1952. Circadian and circannual pr ogrammes in avian migration. Journal of Experimental Biology 199, 39-48. Health Council of the Neth erlands, 2000. Impact of outdoor lighting on man and nature. Publication No. 2000/25E. The Hague: H ealth Council of the Netherlands. Kalmijn, A.J., 1978. Experimental evidence of geom agnetic orientation in elasmobranch fishes. In: Schmidt-Koenig, K., Keeton, W.T. (Eds.) Animal Migration, Navigation, and Homing. Springer-Verlag, Berlin, pp. 347.

PAGE 80

80 Levenson, D.H., Eckert, S.A., Crognale, M.A ., Deegan, J.F., Jacobs, G.H., 2004. Photopic spectral sensitivity of green and l oggerhead sea turtle s. Copeia 2004, 908. Limpus, C.J., 1971. Sea turtle ocean -finding behaviour. Search 2, 385-387. Lohmann, K.J., Lohmann, C.M.F., 1993. A light -independent magnetic compass in the leatherback sea turtle. Bi ological Bulletin 185, 149. Lohmann, K.J., Pentcheff, N.D., Nevitt, G.A., St etten, G.D., Zimmer-Faust, R.K., Jarrard, H.E., Boles, L.C., 1995. Magnetic orientation of spi ny lobsters in the ocean: experiments with undersea coil systems. Journal of Experimental Biology 198, 2041-2048. Lohmann, K.J., Witherington, B.E., Lohm ann, C.M.F., Salmon, M., 1997. Orientation, navigation, and natal beach homing in sea turtles. In: Lutz, P. L., Musick, J.A. (Eds.), The Biology of Sea Turtles Vol. I, CRC Press, Boca Raton, Florida, pp. 107-135. Longcore, T.M., Rich, C., 2004. Ecological lig ht pollution. Frontiers in Ecology and the Environment 2, 191-198. Longcore, T.M., Rich, C., 2006. Synthesis. In: Lo ngcore, T.M., Rich, C. (Eds.), Ecological Consequences of Artificial Night Lighti ng, Island Press, Washington, D.C., pp. 413-430. Lutcavage, M., Musick, J.A., 1985. Aspects of th e biology of sea turtles in Virginia. Copeia 1985, 449-456. Mauritzen, M., Derocher, A.E., Wiig, O., 2001. Spaceuse strategies of female polar bears in a dynamic sea ice habitat. Canadi an Journal of Zoology 79, 1704-1713. Mann, T.M., 1978. Impact of developed coastlin e on nesting and hatch ling sea turtles in Southeastern Florida. Masters Thesis, Florid a Atlantic University, Boca Raton, Florida. Manning, E.L., Cate, H.S., Lohmann, K.J., 1997. Di scrimination of ocean wave features by hatchling loggerhead sea turtles ( Caretta caretta ). Marine Biology 127, 539-544. Mendonca, M.T., Ehrhart, L.M., 1982. Activity, population size, and structure of immature Chelonia mydas and Caretta caretta in Mosquito Lagoon, Florida. Copeia 1982, 161-167. Mrosovsky, N., Kingsmill, S.F., 1985. How turtles fi nd the sea. Zeitschrift fur Tierpsychologie 67, 237-256. Mrosovsky, N., 1972. The water-finding ability of s ea turtles. Brain Behavior and Evolution 5, 202. Quinn, T.P., Merrill, R.T., Brannon, E.L., 1981. Magnetic field detection in sockeye salmon. Journal of Experimental Zoology 217, 137-142. Salmon, M., Wyneken, J., Fritz, E., Lucas, M., 1992. Seafinding by hatchling se a turtles: role of brightness, silhouette, and beach slope as orientation cues. Behaviour 122, 56-77.

PAGE 81

81 Salmon, M., Witherington, B.E., 1995. Artific ial lighting and seaf inding by loggerhead hatchlings: evidence for lunar modulation. Copeia 1995, 931-938. Ugolini, A., Pezzani, A., 1995. Magnetic compass and learning of the Y-axis (sealand) direction in the marine isopod Idotea baltica basteri Animal Behavior 52, 295-300. van Rhijn, F.A., 1979. Optic orientation in hatchlings of the sea turtle, Chelonia mydas I. Brightness: not the only op tic cue in sea-finding orient ation. Marine Behavior and Physiology 6, 105-21. van Rhijn, F.A., van Gorkom, J.C., 1983. Optic or ientation in hatchlings of the sea turtle, Chelonia mydas III. Sea-finding behaviour: the role of photic and visual orientation in animals walking on the spot under laboratory conditions. Marine Behavior and Physiology 9, 211-228. Wang, J.H., Jackson, J.K., Lohmann, K.J., 1998. Pe rception of wave surge motion by hatchling sea turtles. Journal of Experiment al Marine Biology and Ecology 229, 177-186. Wiltschko, W., Wiltschko, R., 1992. Migratory orient ation: magnetic compass orientation of Garden Warblers ( Sylvia borin ) after a simulated crossing of the magnetic equator. Ethology 91, 70-74. Wiltschko, R., Wiltschko, W., 1998. Pigeon homing and magnetic fields: a response to Luschi et al. 1996. Proceedings of the Royal Society of London B 265, 1203-1204. Witham, R., 1982. Disruption of sea turtle nesting habitat with em phasis on human influence. In: Bjorndal, K.A. (Ed.), Biology and Conservation of Sea Turtles. Smithsonian Institution Press, Washington, D.C., pp. 519-522 Witherington, B.E., 1991. Orientation of hatch ling loggerhead turtles at sea off artificially lighted and dark beaches. Journal of Expe rimental Marine Biology and Ecology 149, 1-11. Witherington, B.E., 1992a. Sea-finding behavior and the use of photic orientation cues by hatchling sea turtles. Doctor al Dissertation, University of Florida, Gainesville, FL. Witherington, B.E., 1992b. Behavior al responses of nesting sea tu rtles to artificial lighting. Herpetologica 48, 31-39. Witherington, B.E., 1997. The problem of photopo llution for sea turtles and other nocturnal animals. In: Clemmons, J.R., Buchholz, R. (E ds.), Behavioral Approaches to Conservation in the Wild. Cambridge University Press, Cambridge, England, pp. 303-328. Witherington, B.E., Martin, R.E., 1996. Understa nding, assessing, and resolving light-pollution problems on sea turtle nesting beaches. Florida Marine Research Inst itute Technical Report TR-2. Zar, J.H., 1984. Biostatistical Analysis Prentice Hall, Englewood Cliffs, N.J.

PAGE 82

82 BIOGRAPHICAL SKETCH Russell Asher Scarpino was born in Syracuse, NY on 7 February 1978 to parents Joseph A. and Sarah A. Scarpino. He received his B.S. in biotechnology from the Pennsylvania State University, State College, PA, in 2000. He enrolle d in the M.S. program at the University of Florida in 2002.