<%BANNER%>

Characterization of a Marine Turtle Aggregation in the Big Bend of Florida

HIDE
 Title Page
 Dedication
 Acknowledgement
 Table of Contents
 List of Tables
 List of Figures
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 Biographical sketch
 

PAGE 1

CHARACTERIZATION OF A MARINE TURTLE AGGREGATIO N IN THE BIG BEND OF FLORIDA By WILLIAM J. BARICHIVICH 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 2006

PAGE 2

Copyright 2006 by William J Barichivich

PAGE 3

To the late Captain Edgar Yellowlegs Campbell.

PAGE 4

iv ACKNOWLEDGMENTS I would like to thank my committee, Drs. Raymond R. Carthy and C. Kenneth Dodd, Jr., for their patience with me during my many distractions. My unofficial committee of friends and coworkers, Dr. Jeffrey R. Schmid, Dr. Lora L. Smith, Dr. Steve A. Johnson, Howard L. Jelks, and Dr. Kevin G. Smith, contributed more than they may ever know. I thank the Fates for making my path cro ss with Jeff Schmid. Through his tutelage I gained an appreciation of th e nuances of turtle fishing he learned from Larry Ogren and the last of the commercial fishermen in Ce dar Key. One fisherman, Edgar Campbell, and his family, were especially generous and enlightening to all th ings Old Florida, although I never did learn my ni ckname if I ever earned one. It should not be forgotten that Dr. Cath i Campbell started this project and I am indebted to her for that. Without the selfless dedicati on of Mike Randall, Jennifer Staiger, Duane Houkom, Les Parker, Gary H ill and innumerable University of Florida and Long Island University students, the fieldwork would have ground to a halt. This research would not have been possibl e without the support of Drs. Churchill B. Grimes and Kenneth J. Sulak. Funding for this project was provided by the National Marine Fisheries Service Panama City Offi ce and equipment and logistics by the United States Geological Survey Flor ida-Caribbean Science Center. I am grateful to my parents, Pat and Da ve, and my maternal grandparents, Pat and Bill, for helping me become who I am, and al ways letting me do what I thought was right

PAGE 5

v regardless of how they felt about my decisions Finally, I am thankful for my fiance, Jennifer.

PAGE 6

vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................ix ABSTRACT....................................................................................................................... ..x CHAPTER 1 INTRODUCTION........................................................................................................1 2 MATERIALS AND METHODS.................................................................................3 Study Area....................................................................................................................3 Data Collection.............................................................................................................3 Biometric and Nonbiometric Data..............................................................................5 Data Analysis................................................................................................................6 3 RESULTS...................................................................................................................11 Captures and Effort.....................................................................................................11 Recaptures and Local Movements..............................................................................12 Seasonal and annual size distributions.......................................................................14 Carapace Regression Equations..................................................................................14 Growth Analysis.........................................................................................................15 4 DISCUSSION.............................................................................................................29 Ontogenetic Habitat....................................................................................................30 Migratory Behavior....................................................................................................31 Site Fidelity.................................................................................................................3 2 Temporal and Geographic Shift in Size......................................................................34 Growth........................................................................................................................3 5 Headstarts...................................................................................................................36 Future Research..........................................................................................................39

PAGE 7

vii LIST OF REFERENCES...................................................................................................41 BIOGRAPHICAL SKETCH.............................................................................................46

PAGE 8

viii LIST OF TABLES Table page 3-1. Annual set-net effort (km-hr) and CPUE (turtles/km-hr) by species for Apalachee and Deadman Bay from 1995 to 1997....................................................17 3-2. Proportion of marine turtle captures made in Deadman Bay from 1996 to 1999 by gear type and species...........................................................................................18 3-3. Identification, release, and subseque nt recapture data of NMFS headstart Kemps ridley turtles................................................................................................19 3-4. Seasonal water temperatures and Kemp's ridley turtle, Lepidochelys kempii carapace lengths in Deadman Bay from 1996-1999................................................20 3-5. Formulae for converting between straight-line and curved carapace measurements of Kemp's ridley turtles....................................................................21 3-6. Mean annual growth rates of Kemp's ridley turtles..................................................22 3-7. Estimated values of asymptotic length ( a ) and intrinsic growth rate ( k ) from nonlinear regression of von Bertanlanffy growth interval equation for Kemp's ridley turtles..............................................................................................................23

PAGE 9

ix LIST OF FIGURES Figure page 2-1. Map of Florida Big Bend (black box ) showing Apalachee (green box) and Deadman Bays (blue box)..........................................................................................9 2-2. USGS R/V Plastron prior to outfitting with a tower................................................10 3-1. Length-frequency distributions................................................................................24 3-2. Example of fibropapillomatos is exhibited by a green turtle, Chelonia mydas from Deadman Bay..................................................................................................25 3-3. Photograph of biofouling by the sea turtle barnacle ...............................................26 3-4. Annual relative size compositi on of Kemps ridley turtles, Lepidochelys kempii captured in Deadman Bay from 1996 to 1999.........................................................27 3-5. Kyphotic green turtle, Chelonia mydas captured in Deadman Bay........................28

PAGE 10

x 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 CHARACTERIZATION OF A MARINE TURTLE AGGREGATIO N IN THE BIG BEND OF FLORIDA By William J. Barichivich December 2006 Chair: Raymond R. Carthy Major Department: Wildlife Ecology and Conservation The Kemps ridley sea turtle, Lepidochelys kempii is critically endangered. Away from the nesting beach, little is known about the species. Data from all stages of their life cycle are critical to understa nding their recovery. Theref ore, this in-water study was initiated to fill in gaps in our knowledge about the juvenile and subadult ridleys inhabiting the nearshore waters of the northeastern Gulf of Mexico. Between 1995 and 1999, 159 individuals of th ree species of marine turtle ( L. kempii Caretta caretta and Chelonia mydas ) were captured in Apalachee and Deadman bays, Florida. Captures were dominated by 20 to 40 cm standard straight carapace length Kemps ridleys. Ninety-six percent of all captures were made in Deadman Bay and no C. caretta were captured in Apalachee Bay. Caretta caretta and C. mydas were captured during the warmer months and L. kempii was captured from March to December, although fewer captures were made in cooler months. No turtles were captured or observed when water temperatures were less

PAGE 11

xi than 20 C. Reasonable estimates of annual gr owth for ridleys ranged from 1.25 to 8.92 cm. The high degree of short-term recap tures of Kemps ridleys in Deadman Bay indicates a high level of site fidelity with in a season and recaptures between seasons may suggest more long term affinity to the area. This is the only study of marine turtle s to include Deadman Bay and the results show the area has been overlooked as an im portant developmental habitat for the three species captured. Heavy industry and coasta l developmental poten tially threaten the health of the area and the role it plays in th e recovery and maintenance of these federally listed species.

PAGE 12

1 CHAPTER 1 INTRODUCTION The Kemps ridley ( Lepidochelys kempii Garman, 1880) is the most endangered of the seven extant marine turtle species (R oss et al. 1989). The US Fish and Wildlife Service (USFWS) and the National Marine Fisheries Service (NMFS) estimate the breeding population at 1,500 to 3,000 individual s. The population has been reduced from approximately 40,000 observed nesting on one da y to no more than 700 nesting annually (Magnuson et al. 1990, USFWS and NMFS 1992) Conservation measures for the species have focused on the protection of the nesting beach, captive rearing (head starting), and the implementation of turtle excluder devices (TEDs) on shrimp nets. Despite the mandatory use of TEDs, annual incidental take of Kemps ridleys by shrimp trawls has been estimated between 500 a nd 5,000 (Magnuson et al. 1990). Lack of knowledge about early life stages of the Kemp 's ridley sea turtle currently hinders recovery efforts for this federally listed species. The recovery plan for the Kemps ridl ey (U.S. Fish and Wildlife Service and National Marine Fisheries Service (NMFS), 1992) identif ied in-water, live capture studies as a Priority I Task for determining seasonal use of nearshor e habitat by juveniles and subadults. The U.S. Geological Surv ey, Biological Resources Division (USGSBRD) targeted marine turtles on their Bi ological Resource and Management Issues agenda. In addition, an independent scie ntific review team (Eckert et al. 1994) recommended that research efforts for Kemps ridley be focused on a large-scale mark and recapture program that should, in part, provide information on growth and survival

PAGE 13

2 rates, size-frequency distributi ons, sex ratios, habitat use, a nd movement patterns for wild and headstarted juvenile turtles. Juvenile and subadult Kemp's ridleys use th e shallow nearshore waters of the north and central West Coast of Florida (True 1887, Ogren 1989, Schmid and Ogren 1990, Rudloe et al. 1991, Schmid and Ogren 1992, Schmid 1998, Schmid and Barichivich 2005, Schmid and Barichivich 2006). In the nearshore waters of Cedar Key, Florida, Schmid and Ogren (1990, 1992, Schmid 1998) conducted a long-term (1986-1995) study of wild subadult Kemp's ridleys. This is one of few studies to ch aracterize the population of Kemps ridleys using devel opmental habitat in the region. The current study was undertaken as part of a collaborative effort between the NMFS, Southeast Fisheries Science Center Panama City, Florida, and USGS-BRD, Florida Caribbean Science Center Gainesville, Florida, to establish sampling methods for development of population indices for mon itoring Kemps ridleys in the Florida Panhandle. The goals of the NMFS/USGS ridl ey research in the Florida Big Bend area were to define patterns of occurrence, re lative abundance (v is-a-vis other sea turtle species), growth rate, sex ratio, size freque ncy distribution, habita t use, and movement.

PAGE 14

3 CHAPTER 2 MATERIALS AND METHODS Study Area Sampling was conducted from May 1995 to August 1999 from Apalachee Bay to Suwannee Sound within the Florida Big Bend. The specific areas targeted for sampling included Apalachee and adjacent bays, and two areas in D eadman Bay. The majority of netting in 1995 was conducted in and around Dickerson and Levy bays, adjacent to western Apalachee Bay. Netting in subsequent years was conducted at the Pepperfish Keys and Fishermans Rest in Deadman Bay (Figure 1). Much of the Apalachee Bay is characterized as estuarine habitat with oyster beds, seagrass, sand, and mud patches th roughout. Seagrass beds with sand substrate characterize the more southern capture sites in Deadman Bay. Several significant paleo-river channels bisect the broad seagrass shelf underlyi ng the bay. Kemps ridleys are known to utilize these bathymetric features to move in and o ffshore while exploiting th e abundant prey along the edge of the shallower grass flats (Carr and Caldwell 1956, Rudloe et al. 1991, Schmid 1998). Sampling was concentrated in Deadman Bay due to the abundance of Kemps ridleys, clear water, and the high capture rate. The bulk of effort in 1998 and 1999 was con centrated north of the Steinhatchee River channel and south of Fishermans Rest (Figure 1). Data Collection A 7 m Tremblay flat-bottomed boat (birddawg) wa s used for all research activities. This shallow draft tunnel-hulled skiff was outfitted with a 2.25 m tall gunwale mounted tower to aid in the observation of marine tu rtles (Figure 2). The large aft netwell and forward mounted

PAGE 15

4 outboard motor allowed the rapid deployment of nets at any speed as well as the easy retrieval of turtles and swimmers without fear of entanglement in the propeller. Several capture methods, including set-netting, st rike-netting, and hand ca pture, were used. All proved successful but varied in efficacy depe nding on conditions. The set-netting technique and gear employed was similar to that of the fo rmer commercial Cedar Key-Crystal River fishery (Caldwell and Carr 1956). A 50 m x 6 m multi-filament nylon ne t (25 cm bar) was stretched across a natural channel and monitored. Each end was held in place by a large (approximately 25 kg) kedge (fisherman's) anchor attached by a 15 m polypropylene bridle Closed-cell foam buoys were attached along the head line to supplement floatation a nd serve as navigational aides. The net was checked for turtle s hourly. Any turtles observed in the net were immediately removed. These methods were consistent with those of other contemporary fishery independent in-water marine turtle studies in the n earby area (Schmid and Ogren 1992, Schmid 1998). Strong tides and a high frequency of boat activity hindered set-netti ng. It was therefore best to conduct this type of sampling durin g neap tides, on weekdays, and early in the summer (< July 1) before the recreational sc allop season commenced. Active capture techniques also were utilized to capture turtles, esp ecially when conditions for set-netting were poor. Two observers, one po rt and one starboard, st ationed in the tower looked into the water with polari zed glasses while a third person piloted the boat. When a turtle was spotted, a floating marker was dropped at that point. After marking the spot where the turtle was seen, the observer continued to watch the often stationary turtle while directing the boat pilot. The pilot maneuvered th e boat alongside the turtle and th e second observer released the net on the pilots command. The p ilot circled the turtle as the net ran off the stern. Turtles generally became entangled in the net and were easily removed by observers in the boat. A

PAGE 16

5 snorkeler removed individuals that were encirc led but continued to avoid becoming entangled. Initially a 150 m x 6 m nylon net (25 cm bar, very similar to th e set-net except for length) was used for strike-netting but was replaced by a 150 m x 2.5 m monofilament ne t (10 cm bar) for its superior ability to prevent the escape of sma ll turtles. Hand capture (rodeo) was generally reserved as a last resort if a turtle escaped enta nglement in the strike-net or if the water was too shallow (< 0.25 m) to run the boat. After pursu ing an individual for a short distance a diver jumped off the boat onto the turtle while the rema ining crew returned in the boat to pick up both turtle and diver. Biometric and Non-biometric Data Turtles were checked for scars from previ ous tagging and for living, flipper, and PIT (Passive Integrated Transponder) tags Living tags appeared as a wh ite patch near the center of a carapacial scute. Living tags are formed by tran splanting a piece of lighter colored plastral tissue into a scute on the darker carapace at different scut e locations to distinguish between year classes (Fontaine et al. 1993). The NMFS Head-start Program in Galveston, Texas, has performed this procedure on all head-started Kemps ridley turtle s since 1984. If flipper tags were not present, #681 inconel flipper tags (Na tional Band and Tag Co., supplied by NMFS, Miami, FL) were placed on the proximal trailing edge of both anteri or flippers of all ma rine turtle species captured. If a PIT tag was not detected by s canning the anterior fli ppers and shoulder region, one was placed subcutaneously in the dorsal surf ace of the left anterior flipper of all Kemps ridleys. Any biofouling was removed from the ta gs of recaptured animals to aid tag retention. Measurements including carapace and plastr on lengths and widths, and overall body mass for each individual were recorded. The carap ace measurements included both curved and straight-line measurements for the following: 1) standard carapace leng th (from the precentral scute at carapace midline to posterior margin of postcentrals), 2) minimum carapace length, 3)

PAGE 17

6 notched carapace length, and 4) total carapac e length (see Pritchard et al. 1983 for full descriptions and diagrams of carapace measuremen ts). Curved and straight-line carapace widths were measured at the widest point of the dorsal side. Tree calipers (95 cm or 40 cm length) were used for all straight-line meas urements and a 150 cm flexible tape measure was used for all curved measurements, all to the nearest mm. Using hanging Pesola spring scales, turtles weighing less than 2 kg were weighed to the nearest 0.02 kg, between 2 and 20 kg were weighed to the nearest 0.2 kg, and those greater than 20 kg to the nearest 0.5 kg. To minimize error in extrapolating growth rates, all measurements were made by a single observer throughout any single season (Bjorndal and Bolten 1988). P hotographs were taken of the full body of each individual (carapace and plastron) and of a ny abnormalities. When possible, blood was drawn from Kemps ridleys for radioimmunoassay (R IA) sex determination (Geiss et al. 2005). Salinity and water temperatures were obtained at the time of capture using a YSI model 30 meter. Depth, tidal vector and velocity, and substrate and vegetative composition were also recorded. Position was recorded at each captu re location, as Universal Transverse Mercator (UTM) coordinates, using a real-time differe ntially corrected Magellan NAV DLX-10 Global Positioning System with accuracy better than 10 m. Data Analysis Length-frequency distribution and species composition of Kemps ridley, green, and loggerhead turtles were assessed for each of the two major embayments surveyed. Further analyses of Kemps ridley capture data were performed to address potential gear biases, morphometrics, growth rates, and seasonal occu rrence and habitat charact eristics. Standard straight-line carapace length (SSC L) was used to evaluate size distribution, gear, and growth. Unless otherwise noted, means ( x) are followed by one standard deviation.

PAGE 18

7 Set-netting effort (CPUE) was standardized (Shaver 1994) with the formula: Hrs Length Nets E 1000 where E = the netting effort in hours fished by km tangle net Nets = the number of tangle nets fished Length = the length (m) of a net; and Hrs = the number of hours fished Morphometric relationships of Kemps ridl eys were examined by regressing carapace length on carapace length and log-transformed body mass. Kemps ridley carapace conversion formulae were calculated by regressing paired straight-line and curv ed carapace lengths. Annual growth rates were calculat ed with the following formula: 365 Days Length G where G = the growth rate in cm/yr; Length = difference between th e length at recapture and th e length at initial capture (cm); and Days = the number of days at large. Kemps ridley growth rates were compared by duration of intercapture period, season, and size class at initial capture. The von Bertalanffy growth interval equation (Fabens 1965): kte CL a a CL 1 2 where CL2 = the carapace length at recapture a = the asymptotic length CL1 = the carapace length at first capture k = the intrinsic growth rate

PAGE 19

8 t = the time in years between captures was fitted to the recapture data using a nonlin ear least squares regression procedure (SAS Institute Inc. 1998).

PAGE 20

9 Figure 2-1. Map of Florida Big Bend (black box) showing Apalachee (g reen box) and Deadman Bays (blue box). Note that multiple individuals and species were captured at set-net locations and individual turt les were captured at species specific hand and strike-net locations.

PAGE 21

10 Figure 2-2. USGS R/V Pl astron prior to outfitting with a tower. This 7 m Tremblay birddawg was the primary capture vessel o ver the course of the study. Note nylon strike net in the aft netw ell and release float (aka let -go) next to the observer.

PAGE 22

11 CHAPTER 3 RESULTS Captures and Effort One (16%) green and five (84%) Kemps ridley turtles were captured in Apalachee Bay over 18.45 km net hours. All turtle s in the Apalachee Bay area were captured in setnets with the maximum Kemps ridley CPUE of 2.1 turtles/km net hr in 1996 (Table 3-1). The green turtle (Chelonia mydas Linnaeus, 1758), measured 37.3 cm SSCL and the Kemps ridleys ranged from 28.6 to 38.4 cm SSCL ( x = 34.0 3.4 cm, Fig. 3-1). Nettin g in this area was conducted June through October 1995, August 1996, and July 1997. The green turtle captur e occurred in July 1997 and Kemps ridleys were captured from July through October 1995, and August 1996. Eleven (6.9%) loggerhead (Caretta caretta Linneaus, 1758), 27 (17.0%) green, and 121 (76.1%) Kemps ridley turtles we re tagged during netting activities in Deadman Bay. In addition to 8.3 km net hr of set-net effort, 100 days of active searching were conducted, including opportunistic captures while traveling to and returning from set-net sampling. Though setnetting was successful for capturi ng all species, 54.5, 66.0 and 88. 3% of loggerhead, green and Kemps ridley captures were made by rodeo and st rike-netting (Tables 3-1 & 3-2). Four capture techniques, large-mesh nylon set-net, large-mesh nylon strike-net, monofilament strike-net, and hand capture, were compared. Carapace lengths of Kemps ridley captured with the nylon nets, regardless of technique, were si gnificantly greater than those cap tured with the other methods (F = 5.575, P = 0.001). Loggerhead turtles ranged from 23.7 cm to > 1 m SSCL ( x = 63.5 24.7 cm, excluding two individuals too large to measure straight-line), green turtles from 27.9 to 70.7

PAGE 23

12 cm SSCL ( x = 42.2 9.8 cm), and Kemps ridley turtles from 20.7 to 64.2 cm SSCL ( x = 35.0 8.8 cm, Figure 3-1). RIA of 3 loggerhead bl ood plasma samples coll ected in 1996 yielded a sex ratio of 1M: 0.5F. Four of the 11 loggerhead s captured were potentia lly adults, yet only one individual could be positively iden tified as a male by its proportionally larger tail. Gender could not be determined for any of the green turtles ba sed on external characters because they were all immature. However, RIA of 8 individuals reveal ed a 1M: 7F sex ratio. RIA of 12 blood plasma samples from immature Kemps ridley collected in 1996 resulted in a 1M: 2F ratio, and a later analysis of 48 Kemps ridleys collected from 199 7 to 1999 suggested a more highly skewed ratio of 1M: 3.7F (Geis et al. 2005). Only one Ke mps ridley captured was large enough (64.2 cm SSCL) to be considered adult. This Galves ton NMFS lab headstart appeared to be female. Netting in Deadman Bay was conducted March to December: loggerheads were captured from March to August, green turtles from April to December, and Kemps ridleys from March to December. The incidence of fibropapillomatosis in green turtles from this area varied from year to year; 55% in 1996, 18% in 1997, and no observati ons in the final two y ears. Additionally, the severity of fibropapillomas ranged from small le sions of the nictitating membrane to large (approximately 15 cm diameter by 6 cm thick) tu mors on the plastron, head and limbs (Figure 32). Recaptures and Local Movements One green turtle and 14 Kemps ridleys, excluding NMFS headstarted turtles, were recognized as individuals previously marked in this study. All recaptures were made in Deadman Bay north of the Steinhatchee River ch annel with the exception of a single ridley recapture in the southern extreme of the bay n ear the Pepperfish Keys. Twenty-one percent of the Kemps ridleys recaptured exhi bited flipper tag loss. All indi viduals at large for more than

PAGE 24

13 158 days lost at least one of th eir two flipper tags, resulting in a 17% loss of individual tags applied to ridleys. Heavy biof ouling by the sea turtle barnacle Chelonobia testudinaria (Linnaeus, 1758) was documented after 55 days (F igure 3-3) and significa nt (> 1cm diameter) growth was observed in as few as 20 days. Thre e PIT tags were lost from two small (<25cm SSCL) Kemps ridleys over multiple recaptures. After the apparent loss of these PIT tags, Nexaband SC was used to seal inje ction site wounds after tagging. The sole green turtle recaptured was at la rge for 357 days. Kemps ridley recapture intervals ranged from 2 to 370 days. The green turtle was recaptured 720 meters (m) from the original point of capture and Kemps ridley in tercapture distances spanned 168 to 2466 m. There was no relationship between the number of days at large and the distance between capture locations (r2 = 0.0566, n = 21). In 1998 three Kemps ri dleys from the area between the Steinhatchee River channel and Fishermans Rest were recaptured multiple times. One individual tagged in June was recaptured in July, August, and September (88 day duration). Another turtle tagged in June was recaptured in August and De cember (158 day duration). The third turtle was tagged in August, recaptured later in the m onth, and again in December (112 day duration). In addition to the 14 Kemps ridleys marked and recaptured in this study, four NMFS Galveston Lab headstart turtles were recaptured. In 1995, three coastal-benthic juveniles (28.6, 36.1, and 38.4 cm SSCL) were recaptured in Ap alachee Bay (see Campbell 1996). Six days after the initial recapture, the largest of the three turtles was recaptu red a second time at the same location. The two smallest turtles had readable PIT tags and had been released near Panama City, Florida. In May of the following year an adult female (64.2 cm SSCL) was recaptured at the Pepperfish Keys. Based on the locations of li ving tags, the turtles had been repatriated for

PAGE 25

14 323, 885, 1206 and at least 2388 days respectively (T able 3-3; Fontaine et al. 1993, Caillouet 1995). Seasonal and annual size distributions Mean seasonal water temperatures for Deadma n Bay capture locations were calculated by date: spring (Mar-May), summer (Jun-Aug), fall (S ep-Nov), and winter (D ec-Feb, Table 3-4). Loggerheads were captured in wate r temperatures between 20.7 and 32.7 C, green turtles from 22.2 to 32.7 C, and Kemps ridleys exceeding 19.7 and less than 34 C. Similarly, loggerheads were captured in salinities ranging from 13.3 to 31.3 parts per thousand (ppt), green turtles greater than 15.0 but less than 31.7 ppt, and Kemps ridleys exceeding 9.0 but no more than 35.0 ppt. Carapace lengths of Kemps ridleys captu red in Deadman Bay were pooled by season and compared. There was no significant effect of season on the size of the turtles captured (ANOVA, F = 2.084, P = 0.105). No significant differences in SSCL between seasons were detected with a Bonferroni multiple comparison procedure. The annual relative size distribut ion of Kemp's ridley turtle s captured in Deadman Bay was dominated by the 30-40 cm size class in all years (Figure 3-4). The mean annual SSCL of ridley turtles captured in Deadman Bay di d not vary between years (ANOVA, F = 1.861, P = 0.139), and the annual carapace length distributions di d not differ when compared using the KolmogrovSmirnov two-sample test. Carapace Regression Equations Straight-line carapace width and le ngth were strongly correlated (r = 0.993, n = 137) for Kemps ridley turtles. The following equation describes the relationship: SCW = -3.10 + 1.03 (SSCL). Again, a strong relationship (r = 0.995, n = 137) exists for natural log transformed mass-tolength data resulting in the equation:

PAGE 26

15 ln WT = -8.50 + 2.91 (ln SSCL). Straight-line and curved carapace length measurement conversion equations were calculated to allow for comparison with othe r studies that may use different measuring techniques (Table 3-5). Growth Analysis One green turtle was recaptured during this study. This kyphotic indi vidual was initially 36.7 cm SSCL and grew 1.3 cm over 357 days resu lting in an annual estim ated growth rate of 1.33 cm/yr (Figure 3-5). Between 1997 and 1999, 14 Kemps ridley turtles were recaptured a total of 21 times in Deadman Bay, allowing for th e calculation of 21 annua l growth rates. The small proportion (14%) of long-term recaptures an d short duration of within-season recaptures ( x = 61.2 43.9 days) potentially confounds analys is of growth by season and size due to the amplification of measurement errors in the extrapolation of annual growth rates. The effect of this can be observed in the precipitous decrease in standard errors of annual growth rate with an increase in recapture interval. Although mean annual growth rate with in and between seasons did not vary stat istically (ANOVA, F = 0.76, P = 0.396), there appears to be a slight tendency toward greater growth within a season. Mean gr owth rate did not vary with size class of the turtle (ANOVA, F=1.30, P=0.299) but showed a numeric tr end for 30-40 cm SSCL turtles to grow faster than those in the 2030 cm SSCL size class (Table 3-6). Each recapture interval was fitted with the von Bertalanffy growth interval equation. Asymptotic length estimates ranged -784.2 to 230.8 cm and intrinsic growth rates estimates from 0.00417 to 0.0198 (Table 3-7). The best fit growth model with the lowest residual error was that of all captures, although it did no t produce the most realistic asymptotic length. Marquez (1994) reported the mean carapace length of 65 cm for nesting females. Estimates of asymptotic length

PAGE 27

16 approaching this figure could be considered the mo st biologically realisti c. Therefore the model for recaptures exceeding 90 days would fit best.

PAGE 28

17Table 3-1. Annual set-net effort (km-hr) and CPUE (turtles/km-hr) by species for Ap alachee and Deadman bays from 1995 to 1997. LK=Lepidochelys kempii, CC=Caretta caretta, and CM=Chelonia mydas. Apalachee Bay Deadman Bay Year (months) Effort (km-hr) LK/km-hr CC/km-hr CM/km-hr Effort (km-hr) LK/km-hr CC/km-hr CM/km-hr 1995 (Jun-Sep) 17.57 0.28 0 0 1996 (May-Sep) 0.48 2.08 0 0 5.72 1.05 0.52 0.52 1997 (May-Jul) 0.4 0 0 2.5 2.58 1.94 0.39 2.33

PAGE 29

18 Table 3-2. Proportion of marine turtle cap tures made in Deadman Bay from 1996 to 1999 by gear type and species. Caretta caretta n = 11 Chelonia mydas n = 28 Lepidochelys kempii n = 145 Rodeo 0.00 3.57 15.17 Monofilament strike-net 36.36 35.71 59.31 Nylon strike-net 18.18 28.57 13.79 Set-net 45.45 32.14 11.72

PAGE 30

19Table 3-3. Identification, release, and subsequent recapture data of NM FS headstart Kemps ridley turtles, Lepidochelys kempii, captured and identified in this st udy. Original NMFS tags in bold. Flipper tag numbers PIT Living tag location Year class NMFS release date Capture date NMFS release location Capture location SSCL N-T (cm) at recapture Estimated growth rate (cm/yr) SSN801 SSN802 2242190B60 2nd left costal scute 1991 19May92 7Sep95 13Sep95 32 km off Galveston, TX Fiddlers Point 38.4 Unknown size at release SSH019 SSN803 1F09221422 3rd left costal scute 1993 25Oct94 13Sep95 off Panama City, FL Fiddlers Point 28.6 4.6 SSN806 SSN807 7F7D31127A 3rd right costal scute 1992 18May93 20Oct95 off Panama City, FL Fiddlers Point 36.1 18.7 SSN808 SSN809 2242253909 4th vertebral scute 1986 NA 15May96 NA Pepperfish Keys 64.2 NA

PAGE 31

20 Table 3-4. Seasonal water temperat ures and Kemp's ridley turtle, Lepidochelys kempii, carapace lengths in Deadman Bay from 1996-1999 (stand ard deviation given in parentheses). Season Mean water temperature Mean carapace length n Size range Spring (Mar-May) 26.1 (3.4) 36.1 (8.9) 19 23.3-64.2 Summer (Jun-Aug) 30.1 (1.1) 34.2 (8.3) 96 20.7-55.0 Fall (Sep-Nov) 26.7 (2.3) 35.7 (10.3) 16 22.8-58.0 Winter (Dec-Feb) 22.4 (0.6) 27.7 (5.1) 8 22.6-35.1

PAGE 32

21 Table 3-5. Formulae for converting between stra ight-line and curved ca rapace measurements of Kemp's ridley turtles, Lepidochelys kempii. TSCL=total straight -line carapace length, SSCL=standard straight-line carapace lengt h, MSCL=minimum straight-line carapace length, and MCCL=minimum curved carapace length. Converted length Conversion formula n Adjusted r2 TSCL 1.01 SSCL 0.0199 121 0.999 TSCL 1.02 MSCL + 0.157 121 0.999 TSCL 0.958 MCCL + 0.676 121 0.995 SSCL 0.989 TSCL + 0.0469 121 0.999 SSCL 1.01 MSCL + 0.176 121 1.00 SSCL 0.948 MCCL + 0.687 121 0.995 MSCL 0.977 TSCL 0.113 121 0.999 MSCL 0.929 SCCL + 0.408 121 0.989 MSCL 0.937 MCCL + 0.511 121 0.996 MCCL 1.04 TSCL0.511 121 0.995 MCCL 1.05 SSCL0.562 121 0.995 MCCL 1.06 MSCL0.386 121 0.996

PAGE 33

22 Table 3-6. Mean annual growth ra tes of Kemp's ridley turtles, Lepidochelys kempii, by capture interval, netting season, and size class (standard deviations given in parentheses). Turtles were assigned to size classes by mean of initial and recapture SSCL. Data treatments n Mean SSCL growth rate (cm/yr) Range of growth rates (cm/yr) Recapture interval All recaptures 21 4.21 (2.83) 0.0 8.92 Recaptures > 90 d 8 3.34 (1.91) 1.25 6.47 Recaptures > 180 d 3 3.38 (0.66) 2.78 4.09 Netting season Within season 18 4.34 (3.04) 0.0 8.92 Between season 3 3.38 (0.66) 2.78 4.09 Size class 20 30 cm 12 3.42 (2.64) 0.0 8.26 30 40 cm 8 5.50 (2.98) 1.25 8.92 40 50 cm 1 3.25 3.3

PAGE 34

23 Table 3-7. Estimated values of asymptotic length (a ) and intrinsic growth rate (k) from nonlinear regression of von Bertanlanffy growth interv al equation for Kemp's ridley turtles, Lepidochelys kempii, captured in Deadman Bay between 1997 and 1999. One asymptotic standard error in parentheses. Data treatment n a k All recaptures 21 230.8 cm (463.0) 0.0180 cm (0.0418) Residual mean square error = 0.3588 All recaptures > 90 days 8 206.5 cm (499.9) 0.0198 cm (0.0567) Residual mean square error = 0.6428 All recaptures > 180 days 3 -784.2 cm (13383.2) -0.00417 cm (0.0682) Residual mean square error = 0.7817

PAGE 35

24 Lepidochelys kempiiSSCL N-T (cm) 20253035404550556065 0 5 10 15 20 25 30 35 Deadman Bay Apalachee Bay Chelonia mydas 202530354045505560657075 Number of turtles 0 2 4 6 8 10 Caretta caretta 20406080100 0 1 2 3 4 Figure 3-1. Length-frequency distributions for loggerhead, Caretta caretta, green, Chelonia mydas and Kemps ridley turtles, Lepidochelys kempii, captured in Apalachee and Deadman bays from 1995 to 1999. Note difference in scales of both axes.

PAGE 36

25 Figure 3-2. Example of fibropapillomatosis exhibited by a green turtle, Chelonia mydas, from Deadman Bay.

PAGE 37

26 Figure 3-3. Photograph of biofou ling by the sea turtle barnacle Chelonobia testudinaria on a #681 inconel flipper tag of a recaptured Kemps ridley turtle, Lepidochelys kempii, after 55 days at large.

PAGE 38

27 1996 203040506070 0 10 20 30 40 1997 203040506070 Relative size composition (%) 0 10 20 30 40 1998 203040506070 0 10 20 30 40 1999Carapace length (cm) 203040506070 0 10 20 30 40 x = 40.1 cm SCL sd = 11.7 n = 17 x = 44.4 cm SCL sd = 8.5 n = 12 x = 32.7 cm SCL sd = 7.3 n = 63 x = 33.1 cm SCL sd = 6.2 n = 29 Figure 3-4. Annual relative size com position of Kemps ridley turtles, Lepidochelys kempii, captured in Deadman Bay from 1996 to 1999.

PAGE 39

28 Figure 3-5. Kyphotic green turtle, Chelonia mydas, captured in Deadman Bay.

PAGE 40

29 CHAPTER 4 DISCUSSION Based on CPUE, it may appear that setnetting was an effective method for capturing Kemps ridley turtles. Difficulties utilizing CPUE data to make comparisons are often left unaddressed and have led to the suggestion by the Turtle Expert Working Group (TEWG, 1998) to standardize CPUE to aid in making statistically defensible comparisons through space and time. Intuitivel y this makes sense. For example, CPUE data generated from unequal effort between ye ars and netting locations in this study show a decline in CPUE with an in crease in effort (Table 3-1). However, standardization would also require all in-wat er studies to use the same gear, and the same effort regardless of local conditions. This highly im practical approach would thereby preclude the use of most, if not all, of the data from this study, as well as other highly productive fishery independent in-water studies, based purely on methodology (e.g. Witzell and Schmid 2004). TEWG has subsequently sugg ested that trawl surveys may be the only reliable source of data for analysis of l ong term trends (TEWG 2000). The results of a mixed method, including trawls, fishery dependent study of marine turtles of Apalachicola-Apalachee Bay showed that all Kemps ridley turtles < 25 cm SSCL were captured in less than 1 m depth (Rudloe et al 1991). If a trawl-only survey was implemented in the area, not only would a si gnificant component of the size frequency distribution be lost, thereby sk ewing the apparent size and ag e structure, but much of the bay would be inaccessible to the sampling gear. Insights into marine turtle habitats and areas of abundance, age and size distributions temporal and spatia l migratory patterns,

PAGE 41

30 sex ratios, and growth rate can be achieved without the use of CPUE as the primary measure. Ontogenetic Habitat Though data collected in Apal achee Bay were limited, data generated by marine turtle captures in Deadman Bay demonstrated the importance of the area as a critical developmental habitat. While the presence of marine turtles was known in the nearshore waters of the Big Bend region of the eas tern Gulf of Mexico, Deadman Bay was previously unrecognized as an area of marine turtle con centration (Caldwell and Carr 1957, Carr and Caldwell 1956, Rudloe et al 1991, Schmid 1998, True 1887). Prior to this study, research in the region was focused we st of the St. Marks River and south of Suwannee Sound. Similar to these areas, Deadman Bay is locat ed in the largest remaining seagrass bed in North America, alt hough at the local scale each embayment is quit different (CSA 1985). Schmid (1998) sugge sted habitat and prey partitioning within his study site resulted in differences in mari ne turtle species composition between the two capture locations within Waccasassa Bay. No such obvious habitat choices are possible in the rather homogenous Deadman Bay. Th e larger Kemps ridleys of Waccasassa Bay preferred hard bottom habitat and all observations of the sm aller ridleys of Deadman Bay were made over seagrass beds (Schmid and Barichivich 2005, Schmid et al 2003). The size distribution of Kemps ridl eys from Deadman Bay is very similar to that of the upper Texas and Louisiana coast, but once again the habitats are quite diffe rent (Landry et al 2005). The beachfront habitats from their west ern Gulf of Mexico study are most like the preferred ridley habitat of Waccasassa Bay. The large proportion (11.5, 35.7, and 27.2% ) of post pelagic Kemps ridley (20-25 cm SSCL), green (< 40 cm SSCL) and loggerhead (< 50 cm SSCL) turtles support the

PAGE 42

31 hypothesis that this area may be an ejection po int for marine turtles recruiting from the pelagia to the benthos (Collard and Ogren 1990). Additionally, the observation of small loggerhead turtles (23.7, 33.5 and 49.6 cm SSC L) suggests these indi viduals did not enter the North Atlantic Gyre before recruiting to a demersal habitat. A possible alternative ontogenetic migration route for loggerhead turtles may lie wi thin the Gulf of Mexico. Migratory Behavior Historically, Kemps ridleys in this regi on were not known to migrate to other areas to overwinter. Few ridleys marked on th e Florida west coast have been recovered elsewhere, but one notable record was a movement from Homosassa Florida to the Chesapeake Bay (Lutcavage and Musick 1985). A pattern of sout herly migration has been described for Kemps ridleys inhabiti ng the nearshore waters of the Northwest Atlantic (Henwood and Ogren 1987, Morre ale and Standora 2005, Schmid 1995). Despite the evidence for seasonal movements in the Atlantic, no such pattern can be confirmed for the eastern Gulf of Mexico. Interestingly, the former Florida west coast commercial fishery operated from April until Oct ober (the first cold weather of the year). This and other fishery independent studies in the region have faile d to capture marine turtles in water less than 20C. Commercial marine turtle fishermen suggested to Carr and Caldwell (1956) that at least a portion of the green and Kemps ridley turtles in the Crystal River-Withlacoochee area brumated rather than migrated to warmer water, citing observations of some individuals captured in the early part of the season coming up out of the mud (Carr 1995). Although overwinteri ng behavior has been demonstrated for loggerhead turtles in the Cape Canaveral ship ch annel, it has yet to be verified in Kemps ridleys (Carr et al 1980). R ecent satellite telemetry of si x Kemps ridleys captured and released in Waccasassa Bay demonstrated an offshore and southerly migratory pattern,

PAGE 43

32 thereby confirming at least some segment of the local population of ridleys leaves the area during the coldest months of the year. Telemetered turtles left the area in late November and returned in March. Their de parture, and perhaps return, seems to have coincided with a change in sea surface temp erature (Schmid and Witzell, in press). While these dates are in agreement with months not represented by hist oric fishing effort in the area, Kemps ridleys were captured in Deadman Bay late into December. The possibility that these late season observa tions were turtles passing through while migrating to other locales seems unlikely, si nce several individuals were local residents that had been captured earlier in the year. Additionally this populat ion of smaller turtles would have to begin their migr ation earlier in the year than individuals from the Cedar Keys, since they likely travel slower a nd are initially further north (Schmid and Barichivich 2005). It cannot be ruled out that those indi viduals observed in Deadman Bay did not migrate, and may represent a non-migratory component of the local population. A similar year-round occurrence of Kemps ridley turtles has been observed in the Apalachicola and Apalachee es tuarine area (Rudloe et al. 1991). Site Fidelity In foraging grounds south of Deadman Bay, some marked marine turtles were subsequently recaptured at the point of initia l capture. Both Kemps ridley and green turtles returned to the Crystal River-Withl acoochee fishing grounds after being marked and released in the Cedar Keys. Two of 25 (8 %) Kemps ridleys marked and released in the Cedar Keys returned to their feeding grounds some 40 km away before being recaptured after 43 and no less than 91 days (Carr and Caldwell 1956). Likewise, 9.4% of the ridleys marked at Corrigan Reef were subsequently recaptured at the same location 14 to 839 days later (Schmid 1998). The recapture rate for Deadman Bay was similar

PAGE 44

33 (11.5%), although the turtles were not recaptur ed at the same location but in the same vicinity. Additionally, half the ridleys r ecaptured, including all three individuals recaptured multiple times, had been held in captivity for fecal sample collection and returned to the bay within 48 hours. Even th is added disruption to a species known to be unstable and irascible did not deter these individuals from remaining in the area (Carr 1942). Though the proportion of multi-annual to within-season recaptures from the area is low, the seasonal fidelity of immature Kemps ridleys cannot be overstated. An issue which impacts both growth and recap ture analyses is tag retention. There was a slightly lower rate of tag loss observed in Deadman than the 35% seen in Waccasassa Bay, but the inter capture durations were very different. Although the total number of recaptures was similar between stud ies, there were near ly equal numbers of recaptures within and between seasons in Waccasassa, whereas 85.6% of recaptures in Deadman Bay occurred within a given season (Schmid 1998). If the recapture intervals in Deadman Bay were of greater duration, li ke Wacassassa Bay, the tag loss rate would likely increase and may be similar to those reported by Schmid (1998). In both studies, tag loss seems directly attribut able to biofouling, mostly by barnacles. In response to their observations, Schmid and Ogren (1992) tested plastic Jumbo-Roto tags (Dalton Supplies Ltd.) as an alternative to #681 incone l. Each turtle recei ved one inconel and one plastic tag. It had been s uggested the smooth radius of the plastic post would be less likely to sever the trailing edge of the turtles flipper, especially with the added resistance and sharp cutting edge created by barnacle gr owth, than the bare edge of the punched sheet steel used for the #681 (Bjorndal and Bolten, unpubl.). In addition to barnacles, coral growth has been reported on flipper tags of turtles captured in Cape Canaveral. A

PAGE 45

34 stainless steel alloy containi ng copper, like monel, though less corrosion resistant than inconel which lacks copper, may be a better choice for flipper tags given its inherent resistance to biofouling (Schmid and Ogren 1992). Temporal and Geographic Shift in Size Based on observations of the former Cr ystal River-Withlacoochee commercial fishery, Carr and Caldwell (1956) suggested larg er Kemps ridleys were captured early in the April to November commercial season. Neither the results of their analysis of commercial landings nor those of this study c ould statistically suppor t seasonal shifts in size class (Carr and Caldwell 1956). Howeve r, the mean SSCL of ridleys varied seasonally in Waccasassa Bay; turtles captured in summer were significantly larger than those captured in fall (Schmid 1998). In c ontrast, a fishery-dependent survey of Apalachicola-Apalachee Bays found the mean si ze of Kemps ridleys also varied with season, but larger turtles were capture d in the winter months, December through February. These results were likely biased by the increase in trawl e ffort coincident with the shrimp season and an increase in survey depth (Rudloe et al. 1991). This suggests an offshore migration can not be ruled out (Ogren 1989). No difference in annual size distribution was shown for Kemps ridley turtles captured in Deadman Bay over the four year s of survey. A similar comparison of 10 years of data from Waccasassa Bay resulted in at least one anomalous year, 1991. Two possible explanations for the shift in size class were cited; an influx of pelagic Sargassum, and the introduction of new gear, a smaller mesh net (Schmid 1998). While a change in Gulf currents with an associated influx of Sargassum, and presumably accompanying commensals (e.g. small turtles), ma y be a plausible explanation, the effect of gear could not be ruled out. During surveys in Deadman and Gullivan bays, small

PAGE 46

35 turtles (<30cm SSCL) were observed escap ing through 25.0 and 20.5 cm mesh tangle nets, respectively (Witzell and Schmid 2004). Consequently the nets used in Deadman Bay were replaced by a 10 cm mesh size. An analysis of the relationship between capture methods utilized in Deadman Bay and the mean SSCL of the Kemps ridleys captured there demonstrated a bias toward larg er size classes with la rger mesh sizes. The extent of the effect of gear bias cannot be determined, but to date no other published study of Kemps ridley turtles has used nets wi th this small a mesh size. It would appear simple to recommend switching to smaller mesh nets, but the ability to tangle turtles while preventing the smallest size classes from escaping is a difficult balance to achieve. If mesh sizes become too small not only will la rger turtles bounce off the net rather than tangle, but bycatch and subsequent net repair may increase. In the case of Deadman Bay, the benign nature of the area allowed the use of the encirclement net as a blockade as well as a tangle net. The clear shallow waters allowed for the recovery of turtles, tangled or not, by a snorkeler. In many locations, such as Gullivan Bay, biotic and abiotic conditions (e.g. strong currents, dark wate r, sharp limestone and oysters, marine mammals and large sharks) precluded the us e of anything other than an entanglement strike net. Ultimately the solution may requi re testing gear and accounting for associated bias or using multiple sampling methods to be st suit the objectives of a long term study. Growth A comparison of published growth rates of Kemps ridleys captur ed along the west coast of Florida shows a north to south trend of increasing growth rate in nearly every category of interval, season, and size class with greater than one observation. This trend is complemented by a decreasing difference between within-season and between-season growth (Schmid 1998, Witzell and Schmid 2004) This latitudinal gradient may be

PAGE 47

36 associated with a shorter, or complete lack of, migration or torpor in the southwestern corner of state. Alternatively, this may be an artifact of seasonal resource availability in the temperate waters of the northern penins ula versus those of subtropical southwest Florida. The greatest rate of growth by size class was different for each study and showed no trend, suggesting each location may best suit the nutri tional requirements of various classes differentially. Although di fferences in diet may be responsible for differences in growth rate, growth may also be affected by food availability, environmental quality, and temperature (Morreale and Standora 2005). Length-based estimates of growth for all re captures from Long Island Sound, Cape Canaveral, and Sabine/Calcasieu passes show no clear geographic pattern when compared to those from the Florida gulf co ast. The reported growth rate of Kemps ridleys recaptured in Long Is land Sound (4.1 cm/yr) is most similar to that observed in Deadman Bay (4.21 cm/yr, Table 3-6) as is the size range (27 cm SSCL) of the turtles upon which the estimates are based. Despite an activity period of le ss than 4 months per year, Kemps ridleys in northern developmenta l habitats appear to exhibit comparable growth rates to ridleys from the northeastern Gulf of Mexico, where they are active at least three-quarters of the year (Morreale and Standora 2005). Headstarts There appears to be a pattern of decreasing occurrence of headstarted Kemps ridley turtles from Apalachee Bay clockwise around the Florida Gulf coast. During the 1995 netting season of this study, 60% of th e ridleys captured in Apalachee Bay were positively identified as headstarts, whereas only one or two individuals were observed in each of the following sites, (0.83%, 1 of 121) Deadman Bay, (0.78%, 2 of 254) Waccasassa Bay, and possibly (0.52%, ma ybe 1 of 192) Gullivan Bay (Schmid 1998,

PAGE 48

37 Witzell and Schmid 2004). The pattern is possibly a function of distance between the release and capture locations. Two of the three turtles captured in Apalachee Bay were released off Panama City, the nearest head start release point to the study area. In addition, a temporal shift may be occurring. Of 106 Kemps ridley turtles captured in surveys from 1984 to 1988 in the Apalachicol a-Apalachee bays, no h eadstarted turtles were identified (Rudloe et al 1991). Seven year s later, most of the turtles captured in the Apalachee Bay were headstarts. One headstarted Kemps ridley was capt ured in Deadman Bay. This 1986 year class turtle was the only adult Kemps ridley observed in the area during this study. A female, she could not be further identified due to loss of all in dividual marks. No readable PIT tag could be found in either fore-flipper, although a living tag in her carapace was obvious, as was a tag scar in th e right fore-flipper. Records of adult Kemps ridleys along the Florid a west coast are rare, especi ally since the end of the commercial fishery. Beginning in May 1989 Kemps ridleys began nesting on both coasts of Florida (Meylan et al. 1990). Since then, 21 incidences of nesting, or attempted nesting, scattered through ei ght counties, including two in the panhandle, have been documented through the 2005 nesting season (F WRI 2006). While none of these nesting females could be positively identified as head starts, those origins could not be ruled out (Johnson et al. 1999). Bowen et al. (1994) saw this marked increase in aberrant nesting 11 to 14 years (the presumed time to matura tion) post initiation of the headstart program as strikingly coincidental (Schmid and Witzell 1997, Zug et al. 1997). While an argument can be made that nesting by Kemps ridleys has always occurred in Florida, it seems very unlikely this species conspicuous habit of nesting duri ng the day would have

PAGE 49

38 gone unnoticed (Johnson et al. 1999). Additionally, the natal origin of this species was so uncertain, largely due to the ab sence of any nesting or reprod uctive observations prior to 1960, it had earned the name bastard turtle because it was thought to be a hybrid between a loggerhead, hawksbill, and perhaps a green turtle (Carr 1942, 1952). Early investigations would have surely turned up re ports of nesting had it ever been observed in the state. A similar observation of an adu lt, 65.2cm SSCL, female Kemps ridley was made in Gullivan Bay. Though not specifica lly cited as such in publication, Schmid (pers. com.) is confident she was a headstar t turtle. A size frequency outlier, she was marked only by a flipper tag scar, but no read able PIT or living tag (Witzell and Schmid 2004). The inability to confirm the origin of this suspected headstart turtle captured in Gullivan Bay is one example of the many vagaries of this often criticized NMFS program (Huff 1989, Woody 1991, Frazer 1992, Hewavise nthi 1993). Only 9.6% of the 8,026 individuals released in the fi rst six years of the program ha d living tags, and they were not applied to all turtles un til 1990. Until 1988, no more than 5.6% of any year class was PIT tagged, with an overall application rate of 27.5% of the first 22,576 headstarts released. The greatest tagging rate (98.3%) of any of the methods was flipper tags. As previously discussed, flipper tags are qui ckly lost and not suitable for long-term reidentification, assuming retention rates similar to those of tags applied to wild turtles. Further, gross errors in describing h eadstart marking techniques confound their identification. For example, publications on th e identification of head start living tags use scute and bone nomenclature in terchangeably, and frequencies of PIT tags applied to headstarts are incorrectly st ated and could lead to the use of the wrong equipment to attempt to read them (Fontaine and Shaver 2005, Caillouet et al. 1986). The use of coded

PAGE 50

39 wire tags should also be noted. Beginning in 1984, they were applied to the right or left flipper of all headstarted turtles. We were fortunate to have the rather esoteric and expensive equipment required to read wire tags but could not detect these tags in any of the four headstarts captured in this study. Had the tag been readable, no information other than the presence of the tag could be gained without surgical removal. Future Research Currently the states only industrial river, the Fe nholloway, is located approximately 40 km north of Deadman Bay. It has been estimated that effluent from the Buckeye Florida L.P. Kraft mill in Pe rry has resulted in the loss of 25 km2 of seagrass beds near the river mouth. Wh ile the direct impact on marine turtles is unknown, the real danger may lie in the sublethal effects of the consumption, and subsequent biomagnification, of contaminated crabs and other prey. Plans to mitigate discharge into the river have focused on a pipeline inte nded to bypass the river and pump effluent directly into the Gulf of Mexico. Another potential source of pollution, a coal fired powerplant meant to supply distant municipalitie s, is intended to be built on the same river. Though the atmospheric deposition of mercury into local terrestrial, freshwater, and marine foodwebs is certain, the impacts of powerplant intakes and coal delivery barges and associated structures in the shal low waters of the area are not. While coastal development of the Forgotten Coast is very low, it is increasing and will lead to more boats on the water. Presently the area is enormously popul ar during the summer months for the recreational scallop harvest. Though repor ts of marine turtle strandings in the area are low, this may not be a true reflection of the number of d eaths, but the result of little effort and the low-energy vegetated shoreline. Current statistics from the Florida Sea Turtle Stranding and Salvage Network show th at approximate 50% of ridleys and 33% of

PAGE 51

40 the loggerheads stranded in Dixie and Taylor counties are related to boat impacts (Alan Foley, pers. com.). Tagging should continue in the area, but fo cused research should be directed at behavioral studies addressi ng the effects of increased boating activity and seasonal migrations. Further characteriz ation of specific habitats a nd how turtles are using them seasonally and annually should be emphasized Toxicological research should be considered to address the potential risk s from pulp mill and coal powerplant in conjunction with a trophic study to elucidate contaminant pathways

PAGE 52

41 LIST OF REFERENCES Bjorndal, K. A., and A. B. Bolten. 1988. Gr owth rates of immature green turtles, Chelonia mydas, on feeding grounds in the southern Bahamas. Copeia 1988:555 564. Bowen, B. W., T. A. Conant, and S. R. Hopkins-Murphy. 1994. Where are they now? The Kemps ridley headstart pr oject. Conserv. Biol. 3:853. Caillouet, C. W., C. T. Fontaine, S. A. Man zella, T. D. Williams, and D. B. Revera. 1986. Scutes reserved for living tags. Mar. Turtle Newsletter 36:5. Caillouet, C. W., C. T. Fontai ne, S. A. Manzella-Tirpak, an d D. J. Shaver. 1995. Survival of head-started Kemps ridley sea turtles (Lepidochelys kempii) released in the Gulf of Mexico or adjacent bays. Chelonian Conserv. Biol. 1:285. Caldwell, D. K., and A. Carr. 1957. Status of th e sea turtle fishery in Florida. Trans. 22nd N. Am. Wildl. Conf., p. 457. Campbell, C. L. 1996. Capture of juvenile Ke mps ridleys in the nearshore waters of Apalachee Bay Florida. In R. Byles and Y. Fernandez (compilers), Proceedings of the 16th annual workshop on sea turtle biology and conservation, p. 2829. U.S. Dep. of Commer., NOAA Tech Memo. NMFS-SEFSC-436. Carr, A. 1942. Notes on sea turtles. Proc. New Eng. Zool. Club 21:1. Carr, A. 1952. Handbook of turtles: The turtle s of the United States, Canada, and Baja California. Comstock Publ. Assoc., Co rnell Univ. Press, Ithaca, NY, 542 p. Carr, A. 1995. Notes on the behavioural ecology of sea turtles. In K.A. Bjorndal (ed.) Biology and conservation of sea turtles, p. 19. Smithsonian Inst. Press, Washington, D.C. Carr, A., and D. K Caldwell. 1956. The ecology and migrations of sea turtles: 1. Results of field work in Florida, 1955. Am. Mus. Nov. 1793:1. Carr, A., L. Ogren, and C. McVea. 1980. Apparent hibernation by the Atlantic loggerhead turtle Caretta caretta off Cape Canaveral, Florida. Biol. Conserv. 19:7 14. Collard, S. B., and L. H. Ogren. 1990. Disper sal scenarios for pela gic post-hatchling sea turtles. Bull. Mar. Sci. 47:233.

PAGE 53

42 Continental Shelf Associates, Inc., and Martel Laboratories, Inc. 1985. Florida Big Bend seagrass habitat study narra tive report. A final repo rt by Continental Shelf Associates, Inc to the Minerals Manageme nt Service, Metairie, LA. Contract No. 14-12-0001-30188. Eckert, S. A., D. Crouse, L. B. Crowder, M. Maceina, and A. Shah. 1994. Review of the Kemps ridley sea turtle headstart pr ogram, 22-23 September 1992, Galveston, Texas. U.S. Department of Commer ce. NOAA Tech. Memo. NMFS-OPR-3. 9 p. Fabens, A. J. 1965. Properties and fitting of the von Bertalanffy growth curve. Growth 29:265. Florida Wildlife Research Institute (FWRI). 2006 Reported Nesting Activity of the Kemps Ridley, Lepidochelys kempii in Florida, 1979-2005. Fish and Wildlife Research Institute, Florida Fish and Wildlife Conservation Commission Data Summary Date: 13 Feb. 2006 Available: http://www.floridamarine.org/engine/ download_redirection_process.asp?file=lk_79 -05_0218.pdf&objid=2377&dltype=article Last accessed Aug. 2006. Fontaine, C. T., D. B. Revera, T. D. Willia ms, and C. W. Caillouet. 1993. Detection, verification and decoding of tags and ma rks in head started Kemps ridley sea turtles, Lepidochelys kempii. U.S. Department of Commerce. NOAA Tech. Memo NMFS-SEFC-334. 40 p. Fontaine, C. T., and D. Shaver. 2005. Head -starting the Kemps ridley sea turtle, Lepidochelys kempii, at the NMFS Galveston Laboratory, 1978-1992: a review. Chelonian Conserv. and Biol. 4:838. Frazer, N. B. 1992. Sea turtle conservati on and halfway technology. Conserv. Biol. 6:179. Geis, A. A., W. J. Barichivich, T. Wibbels, M. Coyne, A. M. Landry, and D. Owens. 2005. Predicted sex ratio of juvenile Ke mps ridley sea tu rtles captured near Steinhatchee, Florida. Copeia 2005:293. Henwood, T. A., and L. H. Ogren. 1987. Di stributions and migrations of immature Kemps ridley turtles (Lepidochelys kempi) and green turtles (Chelonia mydas) off Florida, Georgia, and South Caro lina. Northeast Gulf Sci. 9:153. Hewavisenthi, S. 1993. Turtle hatcheries in Sri Lanka: Boon or bane. Mar. Turtle Newsletter 60:19. Huff, J. A. 1989. Florida terminates headst art program. Mar. Turtle Newsletter 46:1. Johnson, S. A., A. L. Bass, B. Libert, M. Marshall, and D. Fulk. 1999. Kemp's ridley (Lepidochelys kempi) nesting in Florida. Fla. Sci. 3/4:194.

PAGE 54

43 Landry, A. M., D. T. Costa, F. L. Kenyon II, and M. S. Coyne. 2005. Population characteristics of Kemps ridley se turt les in nearshore wate rs of the upper Texas and Louisianan coasts. Chelonian Conserv. Biol. 4:801. Lutcavage, M., and J. A. Musick. 1985. Aspect s of the biology of sea turtles of Virginia. Copeia 1985:449. Magnuson, J. J., K. A. Bjorndal, W. D. DuPa ul, G. L. Graham, D. W. Owens, C. H. Peterson, P. C. H. Pritchard, J. I. Rich ardson, G. E. Saul, and C. W. West. 1990. Decline of the sea turtles: causes an d prevention. National Academy Press. Washington, D. C. 274 p. Mrquez, M. R. 1994. Synopsis of the biolog ical data on the Kemp's ridley turtle, Lepidochelys kempi (Garman, 1880). U.S. Departme nt of Commerce. Tech. Memo NMFS-SEFSC-343. 91 p. Meylan, A., P. Castandea, C. Coogan, T. Lo zon, and J. Fletemeyer. 1990. First recorded nesting of Kemps ridley in Florida, USA. Mar. Turtle Newsletter 48:8. Morreale, S. J., and E. A. Standora. 2005. Western North Atlantic water: crucial developmental habitat for Kemps ridley and loggerhead tu rtles. Chelonian Conserv. Biol. 4:873. Ogren, L. H. 1989. Distribution of juven ile and subadult Kemp's ridley turtles: Preliminary results from the 1984-1987 surveys. In: C.W. Caillouet, Jr. and A.M. Landry, Jr. (eds.), Proc. Of the 1st intern ational symposium on Kemp's ridley sea turtle biology, conservation, and ma nagement, p. 116-123. Texas A & M Univ. (TAMU)-SG-89-105:116. Pritchard, P., P. Bacon, F. Berry, A. Carr, J. Fletemeyer, R. Gallagher, S. Hopkins, R. Lankford, R. Marquez M., L. Ogren, W. Pr engle, Jr., H. Reichart, and R. Witham. 1983. Manual of sea turtle research and conservation techniques, second ed. K.A. Bjorndal and G.H. Balazs (eds.), Center for Environmental Education, Washington, D.C. 126 p. Ross, J. P., S. Beavers, D. Mundell, and M. Airth-Kindree. 1989. The status of Kemp's ridley. Washington, D. C.: Center for Marine Conservation. 51 p. Rudloe, A., J. Rudloe, and L. Ogren. 1991. Occurrence of immature Kemp's ridley turtles, Lepidochelys kempi, in coastal waters of northwest Florida. Northeast Gulf Sci. 12:49. SAS Institute Inc. 1999. SAS/STAT User's Gu ide, Version 8. SAS Institute Inc., Cary, NC. Schmid, J. R. 1995. Marine turtle populations on the east-central coast of Florida: Results of tagging studies at Cape Canaveral, Florida, 1986-1991. Fish. Bull. 93:139.

PAGE 55

44 Schmid, J. R. 1998. Marine turtle populati ons on the west-central coast of Florida: Results of tagging studies at the Cedar Keys, Florida, 1986-1995. Fish. Bull. 96:589. Schmid, J. R., and W. J. Barichivich. 2005. Developmental biology and ecology of the Kemps ridley turtle, Lepidochelys kempii, in the eastern Gulf of Mexico. Chelonian Conserv. and Biol. 4:828. Schmid, J. R., and W. J. Barichivich. 2006. Lepidochelys kempiiKemps ridley. In P. A. Meylan (ed.), Biology and conser vation of Florida turtles, p. in press. Chelonian Research Monographs. Schmid, J. R., A. B. Bolten, K. A. Bjorndal, W. J. Lindberg, H. F. Percival, and P. D. Zwick. 2003. Home range and habitat use by Kemps ridley turtle s in west-central Florida. J. Wildl. Manage. 67:196. Schmid, J. R., and L. H. Ogren. 1990. Results of a tagging study at Cedar Key, Florida, with comments on Kemp's ridley dist ribution in the southeastern U.S. In T. I. Richardson, J. I. Richardson, and M. Donne lly (compilers), Proceedings of the 10th annual workshop on sea turtle biology and conservation, p. 129. U.S. Dep. of Commer., NOAA Tech. Memo. NMFS-SEFSC-278. Schmid, J. R., and L. H. Ogren. 1992. Subadult Kemp's ridley sea turtles in the southeastern U.S.: Results of long-term tagging studies. In M. Salmon and J. Wyneken. (compilers), Proceedings of the 11th annual workshop on sea turtle biology and conservation, p. 102. U. S. Dep. of Commer., NOAA Tech. Memo. NMFS-SEFSC-32. Schmid, J. R., and W. N. Witzell. 1997. Age and growth of wild Kemps ridley turtles (Lepidochelys kempi): cumulative results of tagging studies in Florida. Chelonian Conserv. and Biol. 2:532. Schmid, J. R., and W. N. Witzell. In press. Seasonal migrations of immature Kemps ridley turtles along the west coas t of Florida. Gulf Mex. Sci. Shaver, D. J. 1994. Relative abundance, temporal patterns, and growth of seat turtles at the Mansfield Channel, Texa s. J. Herpetol. 25:491-497. True, F. W. 1887. The turtle and terrapin fisheries. In G.B. Goode, The fisheries and fishery industries of the United States, s ec. 5, vol. 2, part 19, p. 493-454. U.S. Comm. Fish Fish. Turtle Expert Working Group. 1998. An assessment of the Kemps ridley (Lepidochelys kempii) and loggerhead (Caretta caretta) sea turtle populations in the western North Atlantic. U.S. Department of Commerce. NOAA Tech. Memo NMFSSEFSC-409. 96 p.

PAGE 56

45 Turtle Expert Working Group. 2000. Assessment update for the Kemps ridley and loggerhead sea turtle populations in the we stern North Atlantic. U.S. Department of Commerce. NOAA Tech. Memo NMFS-SEFSC-444. 115 p. U. S. Fish and Wildlife Service (USFWS ) and National Marine Fisheries Service (NMFS). 1992. Recovery Plan for the Kemp's Ridley Sea Turtle (Lepidochelys kempii). St. Petersburg, FL: National Ma rine Fisheries Service. 40 p. Witzell, W. N., and J. R. Schmid. 2004. Im mature sea turtles in Gullivan Bay, Ten Thousand Islands, Southwest Florida. Gulf of Mex. Sci. 1:54. Woody, J. B. 1991. Guest Edito rial: Its time to stop headst arting Kemps ridleys. Mar Turtle Newsletter 54:7. Zug, G. R., H. J. Kalb, and S. J Luzar. 1997. Age and growth on wild Kemps ridley sea turtles Lepidochelys kempii from skeletochronological data. Biol. Conserv. 80:261 268.

PAGE 57

46 BIOGRAPHICAL SKETCH William James Jamie Barichivich was born in Fort Monmouth, New Jersey, on May 21, 1970. He grew up in Sarasota, Florid a, spending much of his time fishing in the bays and Gulf of Mexico between Tampa Bay and Charlotte Harbor. He graduated from Sarasota High in 1988 and received an A ssociate of Arts degree from Manatee Community College. He began pursuing a degr ee in architecture at the University of Florida in 1991 but changed majors to wild life ecology and conser vation in his second year. His background as a waterman secured him a job with a gene rous PhD student, Jeff Schmid, aiding in the capture and telemetry of Kemps ridley sea turtles near the Cedar Keys, Florida. He worked three field s easons on that project and graduated with a Bachelor of Science in forest resources and conservation in the spring of 1996. After graduation, he began working for the United States Geological Survey (USGS) on the development of a capture/recapt ure study of marine turtles in the Florida Big Bend. This work soon evolved into Jamie s thesis research when he began graduate study in the Department of W ildlife Ecology and Conservation. Jamie has continued to work for the USGS as his research interests have expanded. Since 2000 he has studied amphibians as part of the Department of Interior Amphibian Research and Monitoring Initiative, working across the Southeastern United States. He has also worked on other USGS projects, including mitigation of vertebrate road mortality, Gulf Sturgeon population censu s, and trophodynamics of deep sea benthic communities.


Permanent Link: http://ufdc.ufl.edu/UFE0016242/00001

Material Information

Title: Characterization of a Marine Turtle Aggregation in the Big Bend of Florida
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0016242:00001

Permanent Link: http://ufdc.ufl.edu/UFE0016242/00001

Material Information

Title: Characterization of a Marine Turtle Aggregation in the Big Bend of Florida
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0016242:00001


This item has the following downloads:


Table of Contents
    Title Page
        Page i
        Page ii
    Dedication
        Page iii
    Acknowledgement
        Page iv
        Page v
    Table of Contents
        Page vi
        Page vii
    List of Tables
        Page viii
    List of Figures
        Page ix
    Abstract
        Page x
        Page xi
    Introduction
        Page 1
        Page 2
    Materials and methods
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
    Results
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
    Discussion
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
    References
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
    Biographical sketch
        Page 46
Full Text












CHARACTERIZATION OF A MARINE TURTLE AGGREGATION IN THE BIG
BEND OF FLORIDA















By

WILLIAM J. BARICHIVICH


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2006

































Copyright 2006

by

William J Barichivich



























To the late Captain Edgar "Yellowlegs" Campbell.















ACKNOWLEDGMENTS

I would like to thank my committee, Drs. Raymond R. Carthy and C. Kenneth

Dodd, Jr., for their patience with me during my many distractions. My "unofficial"

committee of friends and coworkers, Dr. Jeffrey R. Schmid, Dr. Lora L. Smith, Dr. Steve

A. Johnson, Howard L. Jelks, and Dr. Kevin G. Smith, contributed more than they may

ever know.

I thank the Fates for making my path cross with Jeff Schmid. Through his tutelage

I gained an appreciation of the nuances of turtle fishing he learned from Larry Ogren and

the last of the commercial fishermen in Cedar Key. One fisherman, Edgar Campbell, and

his family, were especially generous and enlightening to all things "Old Florida,"

although I never did learn my nickname if I ever earned one.

It should not be forgotten that Dr. Cathi Campbell started this project and I am

indebted to her for that. Without the selfless dedication of Mike Randall, Jennifer

Staiger, Duane Houkom, Les Parker, Gary Hill and innumerable University of Florida

and Long Island University students, the fieldwork would have ground to a halt.

This research would not have been possible without the support of Drs. Churchill

B. Grimes and Kenneth J. Sulak. Funding for this project was provided by the National

Marine Fisheries Service Panama City Office and equipment and logistics by the United

States Geological Survey Florida-Caribbean Science Center.

I am grateful to my parents, Pat and Dave, and my maternal grandparents, Pat and

Bill, for helping me become who I am, and always letting me do what I thought was right









regardless of how they felt about my decisions. Finally, I am thankful for my fiancee,

Jennifer.
















TABLE OF CONTENTS



A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ......... .. .. ....................................... .... ... .............. viii

LIST OF FIGURES ......... ......................... ...... ........ ............ ix

A B ST R A C T ................. .......................................................................................... x

CHAPTER

1 IN TRODU CTION ................................................. ...... .................

2 M ATERIALS AND M ETHOD S ........................................... ........... ............... 3

Stu dy A rea ......................................................................... . 3
Data Collection ........................................ ........3
Biom etric and N on-biom etric D ata ........................................ ........................ 5
D ata A nalysis.................................................... 6

3 R E S U L T S ............................................................................................. 1 1

Captures and Effort .................. ........................ .... ... ...... .............. .. 11
R ecaptures and Local M ovem ents..................................................................... .... 12
Seasonal and annual size distributions ............................................ ............... 14
C arapace R egression E quations..................................................................................14
Growth Analysis ...................................... ........................... .... ......... 15

4 DISCUSSION ....................................................... ........... .... .......... 29

O ntogenetic H habitat .................. ................................. .. .. .. .. .............. 30
M igratory B behavior ........................ .............. ............. ...... .......... 1
S ite F id e lity ........................................................................................................... 3 2
Temporal and Geographic Shift in Size.................................... ...................... 34
G ro w th ............................................................................................. 3 5
H e a d sta rts .......................................................................... 3 6
F future R research ....................................................................... 39









L IST O F R E F E R E N C E S ...................................... .................................... ....................4 1

B IO G R A PH IC A L SK E TCH ...................................................................... ..................46















LIST OF TABLES


Table p

3-1. Annual set-net effort (km-hr) and CPUE (turtles/km-hr) by species for
Apalachee and Deadman Bay from 1995 to 1997..................................................17

3-2. Proportion of marine turtle captures made in Deadman Bay from 1996 to 1999
by gear type and species .................. ............. ........ ... ....................... 18

3-3. Identification, release, and subsequent recapture data ofNMFS headstart
K em p 's ridley turtles................................................ ... .... .... .......... 19

3-4. Seasonal water temperatures and Kemp's ridley turtle, Lepidochelys kempii,
carapace lengths in Deadman Bay from 1996-1999 ........................ .............. 20

3-5. Formulae for converting between straight-line and curved carapace
measurements of Kemp's ridley turtles. .......................... ............... 21

3-6. M ean annual growth rates of Kemp's ridley turtles ..............................................22

3-7. Estimated values of asymptotic length (a) and intrinsic growth rate (k) from
nonlinear regression of von Bertanlanffy growth interval equation for Kemp's
ridley turtles .............. ........................ ......... ......... ... .. ......... 23
















LIST OF FIGURES


Figure page

2-1. Map of Florida Big Bend (black box) showing Apalachee (green box) and
D eadm an Bays (blue box).......................................................................9

2-2. USGS R/V Plastron prior to outfitting with a tower.. ..............................................10

3-1. Length-frequency distributions ........................................... ......................... 24

3-2. Example of fibropapillomatosis exhibited by a green turtle, Chelonia mydas,
from D eadm an B ay. ........................... ................ ......................... 25

3-3. Photograph of biofouling by the sea turtle barnacle. .............................................26

3-4. Annual relative size composition of Kemp's ridley turtles, Lepidochelys kempii,
captured in Deadman Bay from 1996 to 1999. .............................. ......... ...... .27

3-5. Kyphotic green turtle, Chelonia mydas, captured in Deadman Bay ......................28















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

CHARACTERIZATION OF A MARINE TURTLE AGGREGATION IN THE BIG
BEND OF FLORIDA

By

William J. Barichivich

December 2006

Chair: Raymond R. Carthy
Major Department: Wildlife Ecology and Conservation

The Kemp's ridley sea turtle, Lepidochelys kempii, is critically endangered. Away

from the nesting beach, little is known about the species. Data from all stages of their life

cycle are critical to understanding their recovery. Therefore, this in-water study was

initiated to fill in gaps in our knowledge about the juvenile and subadult ridleys

inhabiting the nearshore waters of the northeastern Gulf of Mexico.

Between 1995 and 1999, 159 individuals of three species of marine turtle (L.

kempii, Caretta caretta, and Chelonia mydas) were captured in Apalachee and Deadman

bays, Florida. Captures were dominated by 20 to 40 cm standard straight carapace length

Kemp's ridleys. Ninety-six percent of all captures were made in Deadman Bay and no C.

caretta were captured in Apalachee Bay.

Caretta caretta and C. mydas were captured during the warmer months and L.

kempii was captured from March to December, although fewer captures were made in

cooler months. No turtles were captured or observed when water temperatures were less









than 200 C. Reasonable estimates of annual growth for ridleys ranged from 1.25 to 8.92

cm. The high degree of short-term recaptures of Kemp's ridleys in Deadman Bay

indicates a high level of site fidelity within a season and recaptures between seasons may

suggest more long term affinity to the area.

This is the only study of marine turtles to include Deadman Bay and the results

show the area has been overlooked as an important developmental habitat for the three

species captured. Heavy industry and coastal developmental potentially threaten the

health of the area and the role it plays in the recovery and maintenance of these federally

listed species.














CHAPTER 1
INTRODUCTION

The Kemp's ridley (Lepidochelys kempii Garman, 1880) is the most endangered of

the seven extant marine turtle species (Ross et al. 1989). The US Fish and Wildlife

Service (USFWS) and the National Marine Fisheries Service (NMFS) estimate the

breeding population at 1,500 to 3,000 individuals. The population has been reduced from

approximately 40,000 observed nesting on one day to no more than 700 nesting annually

(Magnuson et al. 1990, USFWS and NMFS 1992). Conservation measures for the

species have focused on the protection of the nesting beach, captive rearing (head

starting), and the implementation of turtle excluder devices (TEDs) on shrimp nets.

Despite the mandatory use of TEDs, annual incidental take of Kemp's ridleys by shrimp

trawls has been estimated between 500 and 5,000 (Magnuson et al. 1990). Lack of

knowledge about early life stages of the Kemp's ridley sea turtle currently hinders

recovery efforts for this federally listed species.

The recovery plan for the Kemp's ridley (U.S. Fish and Wildlife Service and

National Marine Fisheries Service (NMFS), 1992) identified in-water, live capture

studies as a Priority I Task for determining seasonal use of nearshore habitat by juveniles

and subadults. The U.S. Geological Survey, Biological Resources Division (USGS-

BRD) targeted marine turtles on their Biological Resource and Management Issues

agenda. In addition, an independent scientific review team (Eckert et al. 1994)

recommended that research efforts for Kemp's ridley be focused on a large-scale mark

and recapture program that should, in part, provide information on growth and survival









rates, size-frequency distributions, sex ratios, habitat use, and movement patterns for wild

and headstarted juvenile turtles.

Juvenile and subadult Kemp's ridleys use the shallow nearshore waters of the north

and central West Coast of Florida (True 1887, Ogren 1989, Schmid and Ogren 1990,

Rudloe et al. 1991, Schmid and Ogren 1992, Schmid 1998, Schmid and Barichivich

2005, Schmid and Barichivich 2006). In the nearshore waters of Cedar Key, Florida,

Schmid and Ogren (1990, 1992, Schmid 1998) conducted a long-term (1986-1995) study

of wild subadult Kemp's ridleys. This is one of few studies to characterize the population

of Kemp's ridleys using developmental habitat in the region.

The current study was undertaken as part of a collaborative effort between the

NMFS, Southeast Fisheries Science Center, Panama City, Florida, and USGS-BRD,

Florida Caribbean Science Center, Gainesville, Florida, to establish sampling methods for

development of population indices for monitoring Kemp's ridleys in the Florida

Panhandle. The goals of the NMFS/USGS ridley research in the Florida Big Bend area

were to define patterns of occurrence, relative abundance (vis-a-vis other sea turtle

species), growth rate, sex ratio, size frequency distribution, habitat use, and movement.














CHAPTER 2
MATERIALS AND METHODS

Study Area

Sampling was conducted from May 1995 to August 1999 from Apalachee Bay to

Suwannee Sound within the Florida Big Bend. The specific areas targeted for sampling included

Apalachee and adjacent bays, and two areas in Deadman Bay. The majority of netting in 1995

was conducted in and around Dickerson and Levy bays, adjacent to western Apalachee Bay.

Netting in subsequent years was conducted at the Pepperfish Keys and Fishermans Rest in

Deadman Bay (Figure 1). Much of the Apalachee Bay is characterized as estuarine habitat with

oyster beds, seagrass, sand, and mud patches throughout. Seagrass beds with sand substrate

characterize the more southern capture sites in Deadman Bay. Several significant paleo-river

channels bisect the broad seagrass shelf underlying the bay. Kemp's ridleys are known to utilize

these bathymetric features to move in and offshore while exploiting the abundant prey along the

edge of the shallower grass flats (Carr and Caldwell 1956, Rudloe et al. 1991, Schmid 1998).

Sampling was concentrated in Deadman Bay due to the abundance of Kemps ridleys, clear water,

and the high capture rate. The bulk of effort in 1998 and 1999 was concentrated north of the

Steinhatchee River channel and south of Fishermans Rest (Figure 1).

Data Collection

A 7 m Tremblay flat-bottomed boat (birddawg) was used for all research activities. This

shallow draft tunnel-hulled skiff was outfitted with a 2.25 m tall gunwale mounted tower to aid

in the observation of marine turtles (Figure 2). The large aft netwell and forward mounted









outboard motor allowed the rapid deployment of nets at any speed as well as the easy retrieval of

turtles and swimmers without fear of entanglement in the propeller.

Several capture methods, including set-netting, strike-netting, and hand capture, were used.

All proved successful but varied in efficacy depending on conditions. The set-netting technique

and gear employed was similar to that of the former commercial Cedar Key-Crystal River fishery

(Caldwell and Carr 1956). A 50 m x 6 m multi-filament nylon net (25 cm bar) was stretched

across a natural channel and monitored. Each end was held in place by a large (approximately

25 kg) kedge (fisherman's) anchor attached by a 15 m polypropylene bridle. Closed-cell foam

buoys were attached along the headline to supplement floatation and serve as navigational aides.

The net was checked for turtles hourly. Any turtles observed in the net were immediately

removed. These methods were consistent with those of other contemporary fishery independent

in-water marine turtle studies in the nearby area (Schmid and Ogren 1992, Schmid 1998).

Strong tides and a high frequency of boat activity hindered set-netting. It was therefore best to

conduct this type of sampling during neap tides, on weekdays, and early in the summer (< July 1)

before the recreational scallop season commenced.

Active capture techniques also were utilized to capture turtles, especially when conditions

for set-netting were poor. Two observers, one port and one starboard, stationed in the tower

looked into the water with polarized glasses while a third person piloted the boat. When a turtle

was spotted, a floating marker was dropped at that point. After marking the spot where the turtle

was seen, the observer continued to watch the often stationary turtle while directing the boat

pilot. The pilot maneuvered the boat alongside the turtle and the second observer released the

net on the pilot's command. The pilot circled the turtle as the net ran off the stem. Turtles

generally became entangled in the net and were easily removed by observers in the boat. A









snorkeler removed individuals that were encircled but continued to avoid becoming entangled.

Initially a 150 m x 6 m nylon net (25 cm bar, very similar to the set-net except for length) was

used for strike-netting but was replaced by a 150 m x 2.5 m monofilament net (10 cm bar) for its

superior ability to prevent the escape of small turtles. Hand capture (rodeo) was generally

reserved as a last resort if a turtle escaped entanglement in the strike-net or if the water was too

shallow (< 0.25 m) to run the boat. After pursuing an individual for a short distance a diver

jumped off the boat onto the turtle while the remaining crew returned in the boat to pick up both

turtle and diver.

Biometric and Non-biometric Data

Turtles were checked for scars from previous tagging and for living, flipper, and PIT

(Passive Integrated Transponder) tags. Living tags appeared as a white patch near the center of a

carapacial scute. Living tags are formed by transplanting a piece of lighter colored plastral tissue

into a scute on the darker carapace at different scute locations to distinguish between year classes

(Fontaine et al. 1993). The NMFS Head-start Program in Galveston, Texas, has performed this

procedure on all head-started Kemp's ridley turtles since 1984. If flipper tags were not present,

#681 inconel flipper tags (National Band and Tag Co., supplied by NMFS, Miami, FL) were

placed on the proximal trailing edge of both anterior flippers of all marine turtle species

captured. If a PIT tag was not detected by scanning the anterior flippers and shoulder region,

one was placed subcutaneously in the dorsal surface of the left anterior flipper of all Kemp's

ridleys. Any biofouling was removed from the tags of recaptured animals to aid tag retention.

Measurements including carapace and plastron lengths and widths, and overall body mass

for each individual were recorded. The carapace measurements included both curved and

straight-line measurements for the following: 1) standard carapace length (from the precentral

scute at carapace midline to posterior margin of postcentrals), 2) minimum carapace length, 3)









notched carapace length, and 4) total carapace length (see Pritchard et al. 1983 for full

descriptions and diagrams of carapace measurements). Curved and straight-line carapace widths

were measured at the widest point of the dorsal side. Tree calipers (95 cm or 40 cm length) were

used for all straight-line measurements and a 150 cm flexible tape measure was used for all

curved measurements, all to the nearest mm. Using hanging Pesola spring scales, turtles

weighing less than 2 kg were weighed to the nearest 0.02 kg, between 2 and 20 kg were weighed

to the nearest 0.2 kg, and those greater than 20 kg to the nearest 0.5 kg. To minimize error in

extrapolating growth rates, all measurements were made by a single observer throughout any

single season (Bjomdal and Bolten 1988). Photographs were taken of the full body of each

individual (carapace and plastron) and of any abnormalities. When possible, blood was drawn

from Kemp's ridleys for radioimmunoassay (RIA) sex determination (Geiss et al. 2005).

Salinity and water temperatures were obtained at the time of capture using a YSI model 30

meter. Depth, tidal vector and velocity, and substrate and vegetative composition were also

recorded. Position was recorded at each capture location, as Universal Transverse Mercator

(UTM) coordinates, using a real-time differentially corrected Magellan NAV DLX-10 Global

Positioning System with accuracy better than 10 m.

Data Analysis

Length-frequency distribution and species composition of Kemp's ridley, green, and

loggerhead turtles were assessed for each of the two major embayments surveyed. Further

analyses of Kemp's ridley capture data were performed to address potential gear biases,

morphometrics, growth rates, and seasonal occurrence and habitat characteristics. Standard

straight-line carapace length (SSCL) was used to evaluate size distribution, gear, and growth.

Unless otherwise noted, means ( x) are followed by + one standard deviation.









Set-netting effort (CPUE) was standardized (Shaver 1994) with the formula:

E Nets x Length Hrs
E = ------ lOOOs
S 1000

where E = the netting effort in hours fished by km tangle net

Nets = the number of tangle nets fished

Length = the length (m) of a net; and

Hrs = the number of hours fished

Morphometric relationships of Kemp's ridleys were examined by regressing carapace

length on carapace length and log-transformed body mass. Kemp's ridley carapace conversion

formulae were calculated by regressing paired straight-line and curved carapace lengths.

Annual growth rates were calculated with the following formula:


G = ALength 365
G Days )

where G = the growth rate in cm/yr;

A Length = difference between the length at recapture and the length at initial capture

(cm); and

Days = the number of days at large.

Kemp's ridley growth rates were compared by duration of intercapture period, season, and

size class at initial capture. The von Bertalanffy growth interval equation (Fabens 1965):

CL2 =a (a CL )e-kt

where CL2 = the carapace length at recapture

a = the asymptotic length

CL1 = the carapace length at first capture

k = the intrinsic growth rate






8


t = the time in years between captures

was fitted to the recapture data using a nonlinear least squares regression procedure (SAS

Institute Inc. 1998).















843O-W
I


f1S"P~

C ru


0 2 4 8 12 16
,i m= Kilometers


84-2I0W


Figure 2-1. Map of Florida Big Bend (black box) showing Apalachee (green box) and Deadman

Bays (blue box). Note that multiple individuals and species were captured at set-net

locations and individual turtles were captured at species specific hand and strike-net

locations.


z9sev~i~ I
B'3aewV


I


3011Nm


a;",yuW


s8,;ir'W"


I ,..N


0I .t-"r















I | '. .f"


Figure 2-2. USGS R/V Plastron prior to outfitting with a tower. This 7 m Tremblay "birddawg" was the primary capture vessel over
the course of the study. Note nylon strike net in the aft netwell and release float (aka "let-go") next to the observer.














CHAPTER 3
RESULTS

Captures and Effort

One (16%) green and five (84%) Kemp's ridley turtles were captured in Apalachee Bay

over 18.45 km net hours. All turtles in the Apalachee Bay area were captured in set-nets with the

maximum Kemp's ridley CPUE of 2.1 turtles/km net hr in 1996 (Table 3-1). The green turtle

(Chelonia mydas Linnaeus, 1758), measured 37.3 cm SSCL and the Kemp's ridleys ranged from

28.6 to 38.4 cm SSCL ( x = 34.0 + 3.4 cm, Fig. 3-1). Netting in this area was conducted June

through October 1995, August 1996, and July 1997. The green turtle capture occurred in July

1997 and Kemp's ridleys were captured from July through October 1995, and August 1996.

Eleven (6.9%) loggerhead (Caretta caretta Linneaus, 1758), 27 (17.0%) green, and 121

(76.1%) Kemp's ridley turtles were tagged during netting activities in Deadman Bay. In addition

to 8.3 km net hr of set-net effort, 100 days of active searching were conducted, including

opportunistic captures while traveling to and returning from set-net sampling. Though set-

netting was successful for capturing all species, 54.5, 66.0 and 88.3% of loggerhead, green and

Kemp's ridley captures were made by rodeo and strike-netting (Tables 3-1 & 3-2). Four capture

techniques, large-mesh nylon set-net, large-mesh nylon strike-net, monofilament strike-net, and

hand capture, were compared. Carapace lengths of Kemp's ridley captured with the nylon nets,

regardless of technique, were significantly greater than those captured with the other methods (F

= 5.575, P = 0.001). Loggerhead turtles ranged from 23.7 cm to > 1 m SSCL ( x = 63.5 24.7

cm, excluding two individuals too large to measure straight-line), green turtles from 27.9 to 70.7









cm SSCL ( x = 42.2 9.8 cm), and Kemps ridley turtles from 20.7 to 64.2 cm SSCL ( x = 35.0

+ 8.8 cm, Figure 3-1). RIA of 3 loggerhead blood plasma samples collected in 1996 yielded a

sex ratio of 1M: 0.5F. Four of the 11 loggerheads captured were potentially adults, yet only one

individual could be positively identified as a male by its proportionally larger tail. Gender could

not be determined for any of the green turtles based on external characters because they were all

immature. However, RIA of 8 individuals revealed a 1M: 7F sex ratio. RIA of 12 blood plasma

samples from immature Kemp's ridley collected in 1996 resulted in a 1M: 2F ratio, and a later

analysis of 48 Kemp's ridleys collected from 1997 to 1999 suggested a more highly skewed ratio

of 1M: 3.7F (Geis et al. 2005). Only one Kemp's ridley captured was large enough (64.2 cm

SSCL) to be considered adult. This Galveston NMFS lab headstart appeared to be female.

Netting in Deadman Bay was conducted March to December: loggerheads were captured

from March to August, green turtles from April to December, and Kemp's ridleys from March to

December. The incidence of fibropapillomatosis in green turtles from this area varied from year

to year; 55% in 1996, 18% in 1997, and no observations in the final two years. Additionally, the

severity of fibropapillomas ranged from small lesions of the nictitating membrane to large

(approximately 15 cm diameter by 6 cm thick) tumors on the plastron, head and limbs (Figure 3-

2).

Recaptures and Local Movements

One green turtle and 14 Kemp's ridleys, excluding NMFS headstarted turtles, were

recognized as individuals previously marked in this study. All recaptures were made in

Deadman Bay north of the Steinhatchee River channel with the exception of a single ridley

recapture in the southern extreme of the bay near the Pepperfish Keys. Twenty-one percent of

the Kemp's ridleys recaptured exhibited flipper tag loss. All individuals at large for more than









158 days lost at least one of their two flipper tags, resulting in a 17% loss of individual tags

applied to ridleys. Heavy biofouling by the sea turtle barnacle Chelonobia testudinaria

(Linnaeus, 1758) was documented after 55 days (Figure 3-3) and significant (> 1cm diameter)

growth was observed in as few as 20 days. Three PIT tags were "lost" from two small (<25cm

SSCL) Kemp's ridleys over multiple recaptures. After the apparent loss of these PIT tags,

Nexaband SC was used to seal injection site wounds after tagging.

The sole green turtle recaptured was at large for 357 days. Kemp's ridley recapture

intervals ranged from 2 to 370 days. The green turtle was recaptured 720 meters (m) from the

original point of capture and Kemp's ridley intercapture distances spanned 168 to 2466 m. There

was no relationship between the number of days at large and the distance between capture

locations (r2 = 0.0566, n = 21). In 1998 three Kemp's ridleys from the area between the

Steinhatchee River channel and Fishermans Rest were recaptured multiple times. One individual

tagged in June was recaptured in July, August, and September (88 day duration). Another turtle

tagged in June was recaptured in August and December (158 day duration). The third turtle was

tagged in August, recaptured later in the month, and again in December (112 day duration).

In addition to the 14 Kemp's ridleys marked and recaptured in this study, four NMFS

Galveston Lab headstart turtles were recaptured. In 1995, three coastal-benthic juveniles (28.6,

36.1, and 38.4 cm SSCL) were recaptured in Apalachee Bay (see Campbell 1996). Six days after

the initial recapture, the largest of the three turtles was recaptured a second time at the same

location. The two smallest turtles had readable PIT tags and had been released near Panama

City, Florida. In May of the following year an adult female (64.2 cm SSCL) was recaptured at

the Pepperfish Keys. Based on the locations of living tags, the turtles had been repatriated for









323, 885, 1206 and at least 2388 days respectively (Table 3-3; Fontaine et al. 1993, Caillouet

1995).

Seasonal and annual size distributions

Mean seasonal water temperatures for Deadman Bay capture locations were calculated by

date: spring (Mar-May), summer (Jun-Aug), fall (Sep-Nov), and winter (Dec-Feb, Table 3-4).

Loggerheads were captured in water temperatures between 20.7 and 32.7 C, green turtles from

22.2 to 32.7 C, and Kemp's ridleys exceeding 19.7 and less than 34 C. Similarly, loggerheads

were captured in salinities ranging from 13.3 to 31.3 parts per thousand (ppt), green turtles

greater than 15.0 but less than 31.7 ppt, and Kemp's ridleys exceeding 9.0 but no more than 35.0

ppt. Carapace lengths of Kemp's ridleys captured in Deadman Bay were pooled by season and

compared. There was no significant effect of season on the size of the turtles captured

(ANOVA, F = 2.084, P = 0.105). No significant differences in SSCL between seasons were

detected with a Bonferroni multiple comparison procedure.

The annual relative size distribution of Kemp's ridley turtles captured in Deadman Bay was

dominated by the 30-40 cm size class in all years (Figure 3-4). The mean annual SSCL of ridley

turtles captured in Deadman Bay did not vary between years (ANOVA, F = 1.861, P = 0.139),

and the annual carapace length distributions did not differ when compared using the Kolmogrov-

Smirnov two-sample test.

Carapace Regression Equations

Straight-line carapace width and length were strongly correlated (r = 0.993, n = 137) for

Kemp's ridley turtles. The following equation describes the relationship:

SCW= -3.10 + 1.03 (SSCL).

Again, a strong relationship (r = 0.995, n = 137) exists for natural log transformed mass-to-

length data resulting in the equation:









In WT= -8.50 + 2.91 (In SSCL).

Straight-line and curved carapace length measurement conversion equations were

calculated to allow for comparison with other studies that may use different measuring

techniques (Table 3-5).

Growth Analysis

One green turtle was recaptured during this study. This kyphotic individual was initially

36.7 cm SSCL and grew 1.3 cm over 357 days resulting in an annual estimated growth rate of

1.33 cm/yr (Figure 3-5). Between 1997 and 1999, 14 Kemp's ridley turtles were recaptured a

total of 21 times in Deadman Bay, allowing for the calculation of 21 annual growth rates. The

small proportion (14%) of long-term recaptures and short duration of within-season recaptures

( x = 61.2 43.9 days) potentially confounds analysis of growth by season and size due to the

amplification of measurement errors in the extrapolation of annual growth rates. The effect of

this can be observed in the precipitous decrease in standard errors of annual growth rate with an

increase in recapture interval. Although mean annual growth rate within and between seasons

did not vary statistically (ANOVA, F = 0.76, P = 0.396), there appears to be a slight tendency

toward greater growth within a season. Mean growth rate did not vary with size class of the

turtle (ANOVA, F=1.30, P=0.299) but showed a numeric trend for 30-40 cm SSCL turtles to

grow faster than those in the 20-30 cm SSCL size class (Table 3-6).

Each recapture interval was fitted with the von Bertalanffy growth interval equation.

Asymptotic length estimates ranged -784.2 to 230.8 cm and intrinsic growth rates estimates from

0.00417 to 0.0198 (Table 3-7). The best fit growth model with the lowest residual error was that

of all captures, although it did not produce the most realistic asymptotic length. Marquez (1994)

reported the mean carapace length of 65 cm for nesting females. Estimates of asymptotic length






16


approaching this figure could be considered the most biologically realistic. Therefore the model

for recaptures exceeding 90 days would fit best.












Table 3-1. Annual set-net effort (km-hr) and CPUE (turtles/km-hr) by species for Apalachee and Deadman bays from 1995 to 1997.
LK=Lepidochelys kempii, CC=Caretta caretta, and CM=Chelonia mydas.

Apalachee Bay Deadman Bay
Year (months) Effort LK/km-hr CC/km-hr CM/km-hr Effort LK/km-hr CC/km-hr CM/km-hr
(km-hr) (km-hr)
1995 (Jun-Sep) 17.57 0.28 0 0
1996 (May-Sep) 0.48 2.08 0 0 5.72 1.05 0.52 0.52
1997 (May-Jul) 0.4 0 0 2.5 2.58 1.94 0.39 2.33






18


Table 3-2. Proportion of marine turtle captures made in Deadman Bay from 1996 to 1999 by
gear type and species.


Caretta Chelonia Lepidochelys
caretta mydas kempii
n= 11 n = 28 n= 145
Rodeo 0.00 3.57 15.17
Monofilament
strike-net 36.36 35.71 59.31
Nylon
strike-net 18.18 28.57 13.79
Set-net 45.45 32.14 11.72












Table 3-3. Identification, release, and subsequent recapture data ofNMFS headstart Kemp's ridley turtles, Lepidochelys kempii,
captured and identified in this study. Original NMFS tags in bold.

SSCL Estimated
NMFS NMFS N-T growth
Flipper tag Living tag Year release Capture release Capture (cm) at rate
numbers PIT location class date date location location recapture (cm/yr)
SSN801 2242190B60 2nd left 1991 19May92 7Sep95 32 km off Fiddler's 38.4 Unknown
SSN802 costal 13Sep95 Galveston, Point size at
scute TX release
SSH019 1F09221422 3rd left 1993 250ct94 13Sep95 off Panama Fiddler's 28.6 4.6
SSN803 costal City, FL Point
scute
SSN806 7F7D31127A 3rd right 1992 18May93 200ct95 off Panama Fiddler's 36.1 18.7
SSN807 costal City, FL Point
scute
SSN808 2242253909 4th vertebral 1986 NA 15May96 NA Pepperfish 64.2 NA
SSN809 scute Keys









Table 3-4. Seasonal water temperatures and Kemp's ridley turtle, Lepidochelys kempii, carapace
lengths in Deadman Bay from 1996-1999 (standard deviation given in parentheses).

Mean carapace
Season Mean water temperature length n Size range
Spring 26.1 36.1
19 23.3-64.2
(Mar-May) (3.4) (8.9)
Summer 30.1 34.2
96 20.7-55.0
(Jun-Aug) (1.1) (8.3)
Fall 26.7 35.7
16 22.8-58.0
(Sep-Nov) (2.3) (10.3)
Winter 22.4 27.7
(Dec-Feb) (0.6) (5.22.6-35.1
(Dec-Feb) (0.6) (5.1)









Table 3-5. Formulae for converting between straight-line and curved carapace measurements of
Kemp's ridley turtles, Lepidochelys kempii. TSCL=total straight-line carapace length,
SSCL=standard straight-line carapace length, MSCL=minimum straight-line carapace
length, and MCCL=minimum curved carapace length.


Conversion formula
1.01 SSCL- 0.0199
1.02 MSCL + 0.157
0.958 MCCL + 0.676
0.989 TSCL + 0.0469
1.01 MSCL + 0.176
0.948 MCCL + 0.687
0.977 TSCL 0.113
0.929 SCCL + 0.408
0.937 MCCL + 0.511
1.04 TSCL- 0.511
1.05 SSCL- 0.562
1.06 MSCL- 0.386


Adjusted r2
0.999
0.999
0.995
0.999
1.00
0.995
0.999
0.989
0.996
0.995
0.995
0.996


Converted
length
TSCL
TSCL
TSCL
SSCL
SSCL
SSCL
MSCL
MSCL
MSCL
MCCL
MCCL
MCCL









Table 3-6. Mean annual growth rates of Kemp's ridley turtles, Lepidochelys kempii, by capture
interval, netting season, and size class (standard deviations given in parentheses).
Turtles were assigned to size classes by mean of initial and recapture SSCL.


Mean SSCL Range of
growth rate growth rates
Data treatments n (cm/yr) (cm/yr)
Recapture interval
All recaptures 21 4.21 0.0 8.92
(2.83)
Recaptures > 90 d 8 3.34 1.25 6.47
(1.91)
Recaptures > 180 d 3 3.38 2.78 4.09
(0.66)
Netting season
Within season 18 4.34 0.0 8.92
(3.04)
Between season 3 3.38 2.78 4.09
(0.66)
Size class
20 30 cm 12 3.42 0.0 8.26
(2.64)
30- 40 cm 8 5.50 1.25 8.92
(2.98)
40 50 cm 1 3.25 3.3









Table 3-7. Estimated values of asymptotic length (a) and intrinsic growth rate (k) from nonlinear
regression of von Bertanlanffy growth interval equation for Kemp's ridley turtles,
Lepidochelys kempii, captured in Deadman Bay between 1997 and 1999. One
asymptotic standard error in parentheses.


Data treatment n a k
All recaptures 21 230.8 cm 0.0180 cm
(463.0) (0.0418)
Residual mean square error = 0.3588

All recaptures > 8 206.5 cm 0.0198 cm
90 days (499.9) (0.0567)
Residual mean square error = 0.6428

All recaptures > 3 -784.2 cm -0.00417 cm
180 days (13383.2) (0.0682)
Residual mean square error = 0.7817











Caretta caretta


20 40 60 80 100


Chelonia mydas


20 25
20 25


30 35 40 45 50 55 60


65 70 75


Lepidochelys kempii


mm


20 25 30 35 40


1 Deadman Bay
I I Apalachee Bay









45 50 55 60 65


SSCL N-T (cm)


Figure 3-1. Length-frequency distributions for loggerhead, Caretta caretta, green, Chelonia
mydas, and Kemp's ridley turtles, Lepidochelys kempii, captured in Apalachee and
Deadman bays from 1995 to 1999. Note difference in scales of both axes.























-4<'


- 'I ,
-r; -


s, "y ,-i


E 'I" r KEY
Pc -1a,,
^^'-*^

.. i .
r ..- .


Figure 3-2. Example of fibropapillomatosis exhibited by a green turtle, Chelonia mydas, from Deadman Bay.


m-gE

U.











































Figure 3-3. Photograph of biofouling by the sea turtle barnacle Chelonobia testudinaria on a #681 inconel flipper tag of a recaptured
Kemp's ridley turtle, Lepidochelys kempii, after 55 days at large.













x = 40.1 cm SCL
sd = 11.7
n= 17








20 30 40 50 60 71

x = 44.4 cm SCL
sd = 8.5
n= 12








20 30 40 50 60 71

x = 32.7 cm SCL
sd = 7.3
n = 63








20 30 40 50 60 71

x= 33.1 cm SCL
sd = 6.2
n= 29


1996


1997













1998


1999


20 30 40 50 60 70

Carapace length (cm)


Figure 3-4. Annual relative size composition of Kemp's ridley turtles, Lepidochelys kempii,
captured in Deadman Bay from 1996 to 1999.













































Figure 3-5. Kyphotic green turtle, Chelonia mydas, captured in Deadman Bay.














CHAPTER 4
DISCUSSION

Based on CPUE, it may appear that set-netting was an effective method for

capturing Kemp's ridley turtles. Difficulties utilizing CPUE data to make comparisons

are often left unaddressed and have led to the suggestion by the Turtle Expert Working

Group (TEWG, 1998) to standardize CPUE to aid in making statistically defensible

comparisons through space and time. Intuitively this makes sense. For example, CPUE

data generated from unequal effort between years and netting locations in this study show

a decline in CPUE with an increase in effort (Table 3-1). However, standardization

would also require all in-water studies to use the same gear, and the same effort

regardless of local conditions. This highly impractical approach would thereby preclude

the use of most, if not all, of the data from this study, as well as other highly productive

fishery independent in-water studies, based purely on methodology (e.g. Witzell and

Schmid 2004). TEWG has subsequently suggested that trawl surveys may be the only

reliable source of data for analysis of long term trends (TEWG 2000). The results of a

mixed method, including trawls, fishery dependent study of marine turtles of

Apalachicola-Apalachee Bay showed that all Kemp's ridley turtles < 25 cm SSCL were

captured in less than 1 m depth (Rudloe et al 1991). If a trawl-only survey was

implemented in the area, not only would a significant component of the size frequency

distribution be lost, thereby skewing the apparent size and age structure, but much of the

bay would be inaccessible to the sampling gear. Insights into marine turtle habitats and

areas of abundance, age and size distributions, temporal and spatial migratory patterns,









sex ratios, and growth rate can be achieved without the use of CPUE as the primary

measure.

Ontogenetic Habitat

Though data collected in Apalachee Bay were limited, data generated by marine

turtle captures in Deadman Bay demonstrated the importance of the area as a critical

developmental habitat. While the presence of marine turtles was known in the nearshore

waters of the Big Bend region of the eastern Gulf of Mexico, Deadman Bay was

previously unrecognized as an area of marine turtle concentration (Caldwell and Carr

1957, Carr and Caldwell 1956, Rudloe et al 1991, Schmid 1998, True 1887). Prior to this

study, research in the region was focused west of the St. Marks River and south of

Suwannee Sound. Similar to these areas, Deadman Bay is located in the largest

remaining seagrass bed in North America, although at the local scale each embayment is

quit different (CSA 1985). Schmid (1998) suggested habitat and prey partitioning within

his study site resulted in differences in marine turtle species composition between the two

capture locations within Waccasassa Bay. No such obvious habitat choices are possible

in the rather homogenous Deadman Bay. The larger Kemp's ridleys of Waccasassa Bay

preferred hard bottom habitat and all observations of the smaller ridleys of Deadman Bay

were made over seagrass beds (Schmid and Barichivich 2005, Schmid et al 2003). The

size distribution of Kemp's ridleys from Deadman Bay is very similar to that of the upper

Texas and Louisiana coast, but once again the habitats are quite different (Landry et al

2005). The beachfront habitats from their western Gulf of Mexico study are most like the

preferred ridley habitat of Waccasassa Bay.

The large proportion (11.5, 35.7, and 27.2%) of post pelagic Kemp's ridley (20-25

cm SSCL), green (< 40 cm SSCL) and loggerhead (< 50 cm SSCL) turtles support the









hypothesis that this area may be an ejection point for marine turtles recruiting from the

pelagia to the benthos (Collard and Ogren 1990). Additionally, the observation of small

loggerhead turtles (23.7, 33.5 and 49.6 cm SSCL) suggests these individuals did not enter

the North Atlantic Gyre before recruiting to a demersal habitat. A possible alternative

ontogenetic migration route for loggerhead turtles may lie within the Gulf of Mexico.

Migratory Behavior

Historically, Kemp's ridleys in this region were not known to migrate to other areas

to overwinter. Few ridleys marked on the Florida west coast have been recovered

elsewhere, but one notable record was a movement from Homosassa Florida to the

Chesapeake Bay (Lutcavage and Musick 1985). A pattern of southerly migration has

been described for Kemp's ridleys inhabiting the nearshore waters of the Northwest

Atlantic (Henwood and Ogren 1987, Morreale and Standora 2005, Schmid 1995).

Despite the evidence for seasonal movements in the Atlantic, no such pattern can be

confirmed for the eastern Gulf of Mexico. Interestingly, the former Florida west coast

commercial fishery operated from April until October (the first cold weather of the year).

This and other fishery independent studies in the region have failed to capture marine

turtles in water less than 200C. Commercial marine turtle fishermen suggested to Carr

and Caldwell (1956) that at least a portion of the green and Kemp's ridley turtles in the

Crystal River-Withlacoochee area brumated rather than migrated to warmer water, citing

observations of some individuals captured in the early part of the season coming "up out

of the mud" (Carr 1995). Although overwintering behavior has been demonstrated for

loggerhead turtles in the Cape Canaveral ship channel, it has yet to be verified in Kemp's

ridleys (Carr et al 1980). Recent satellite telemetry of six Kemp's ridleys captured and

released in Waccasassa Bay demonstrated an offshore and southerly migratory pattern,









thereby confirming at least some segment of the local population of ridleys leaves the

area during the coldest months of the year. Telemetered turtles left the area in late

November and returned in March. Their departure, and perhaps return, seems to have

coincided with a change in sea surface temperature (Schmid and Witzell, in press).

While these dates are in agreement with months not represented by historic fishing effort

in the area, Kemp's ridleys were captured in Deadman Bay late into December. The

possibility that these late season observations were turtles passing through while

migrating to other locales seems unlikely, since several individuals were local residents

that had been captured earlier in the year. Additionally this population of smaller turtles

would have to begin their migration earlier in the year than individuals from the Cedar

Keys, since they likely travel slower and are initially further north (Schmid and

Barichivich 2005). It cannot be ruled out that those individuals observed in Deadman

Bay did not migrate, and may represent a non-migratory component of the local

population. A similar year-round occurrence of Kemp's ridley turtles has been observed

in the Apalachicola and Apalachee estuarine area (Rudloe et al. 1991).

Site Fidelity

In foraging grounds south of Deadman Bay, some marked marine turtles were

subsequently recaptured at the point of initial capture. Both Kemp's ridley and green

turtles returned to the Crystal River-Withlacoochee fishing grounds after being marked

and released in the Cedar Keys. Two of 25 (8%) Kemp's ridleys marked and released in

the Cedar Keys returned to their feeding grounds some 40 km away before being

recaptured after 43 and no less than 91 days (Carr and Caldwell 1956). Likewise, 9.4%

of the ridleys marked at Corrigan Reef were subsequently recaptured at the same location

14 to 839 days later (Schmid 1998). The recapture rate for Deadman Bay was similar









(11.5%), although the turtles were not recaptured at the same location but in the same

vicinity. Additionally, half the ridleys recaptured, including all three individuals

recaptured multiple times, had been held in captivity for fecal sample collection and

returned to the bay within 48 hours. Even this added disruption to a species known to be

"unstable and irascible" did not deter these individuals from remaining in the area (Carr

1942). Though the proportion of multi-annual to within-season recaptures from the area

is low, the seasonal fidelity of immature Kemp's ridleys cannot be overstated.

An issue which impacts both growth and recapture analyses is tag retention. There

was a slightly lower rate of tag loss observed in Deadman than the 35% seen in

Waccasassa Bay, but the intercapture durations were very different. Although the total

number of recaptures was similar between studies, there were nearly equal numbers of

recaptures within and between seasons in Waccasassa, whereas 85.6% of recaptures in

Deadman Bay occurred within a given season (Schmid 1998). If the recapture intervals

in Deadman Bay were of greater duration, like Wacassassa Bay, the tag loss rate would

likely increase and may be similar to those reported by Schmid (1998). In both studies,

tag loss seems directly attributable to biofouling, mostly by barnacles. In response to

their observations, Schmid and Ogren (1992) tested plastic Jumbo-Roto tags (Dalton

Supplies Ltd.) as an alternative to #681 inconel. Each turtle received one inconel and one

plastic tag. It had been suggested the smooth radius of the plastic post would be less

likely to sever the trailing edge of the turtle's flipper, especially with the added resistance

and sharp cutting edge created by barnacle growth, than the bare edge of the punched

sheet steel used for the #681 (Bjomdal and Bolten, unpubl.). In addition to barnacles,

coral growth has been reported on flipper tags of turtles captured in Cape Canaveral. A









stainless steel alloy containing copper, like monel, though less corrosion resistant than

inconel which lacks copper, may be a better choice for flipper tags given its inherent

resistance to biofouling (Schmid and Ogren 1992).

Temporal and Geographic Shift in Size

Based on observations of the former Crystal River-Withlacoochee commercial

fishery, Carr and Caldwell (1956) suggested larger Kemp's ridleys were captured early in

the April to November commercial season. Neither the results of their analysis of

commercial landings nor those of this study could statistically support seasonal shifts in

size class (Carr and Caldwell 1956). However, the mean SSCL of ridleys varied

seasonally in Waccasassa Bay; turtles captured in summer were significantly larger than

those captured in fall (Schmid 1998). In contrast, a fishery-dependent survey of

Apalachicola-Apalachee Bays found the mean size of Kemp's ridleys also varied with

season, but larger turtles were captured in the winter months, December through

February. These results were likely biased by the increase in trawl effort coincident with

the shrimp season and an increase in survey depth (Rudloe et al. 1991). This suggests an

offshore migration can not be ruled out (Ogren 1989).

No difference in annual size distribution was shown for Kemp's ridley turtles

captured in Deadman Bay over the four years of survey. A similar comparison of 10

years of data from Waccasassa Bay resulted in at least one anomalous year, 1991. Two

possible explanations for the shift in size class were cited; an influx of pelagic

Sargassum, and the introduction of new gear, a smaller mesh net (Schmid 1998). While a

change in Gulf currents with an associated influx of Sargassum, and presumably

accompanying commensals (e.g. small turtles), may be a plausible explanation, the effect

of gear could not be ruled out. During surveys in Deadman and Gullivan bays, small









turtles (<30cm SSCL) were observed escaping through 25.0 and 20.5 cm mesh tangle

nets, respectively (Witzell and Schmid 2004). Consequently the nets used in Deadman

Bay were replaced by a 10 cm mesh size. An analysis of the relationship between

capture methods utilized in Deadman Bay and the mean SSCL of the Kemp's ridleys

captured there demonstrated a bias toward larger size classes with larger mesh sizes. The

extent of the effect of gear bias cannot be determined, but to date no other published

study of Kemp's ridley turtles has used nets with this small a mesh size. It would appear

simple to recommend switching to smaller mesh nets, but the ability to tangle turtles

while preventing the smallest size classes from escaping is a difficult balance to achieve.

If mesh sizes become too small not only will larger turtles bounce off the net rather than

tangle, but bycatch and subsequent net repair may increase. In the case of Deadman Bay,

the benign nature of the area allowed the use of the encirclement net as a blockade as

well as a tangle net. The clear shallow waters allowed for the recovery of turtles, tangled

or not, by a snorkeler. In many locations, such as Gullivan Bay, biotic and abiotic

conditions (e.g. strong currents, dark water, sharp limestone and oysters, marine

mammals and large sharks) precluded the use of anything other than an entanglement

strike net. Ultimately the solution may require testing gear and accounting for associated

bias or using multiple sampling methods to best suit the objectives of a long term study.

Growth

A comparison of published growth rates of Kemps' ridleys captured along the west

coast of Florida shows a north to south trend of increasing growth rate in nearly every

category of interval, season, and size class with greater than one observation. This trend

is complemented by a decreasing difference between within-season and between-season

growth (Schmid 1998, Witzell and Schmid 2004). This latitudinal gradient may be









associated with a shorter, or complete lack of, migration or torpor in the southwestern

corner of state. Alternatively, this may be an artifact of seasonal resource availability in

the temperate waters of the northern peninsula versus those of subtropical southwest

Florida. The greatest rate of growth by size class was different for each study and

showed no trend, suggesting each location may best suit the nutritional requirements of

various classes differentially. Although differences in diet may be responsible for

differences in growth rate, growth may also be affected by food availability,

environmental quality, and temperature (Morreale and Standora 2005).

Length-based estimates of growth for all recaptures from Long Island Sound, Cape

Canaveral, and Sabine/Calcasieu passes show no clear geographic pattern when

compared to those from the Florida gulf coast. The reported growth rate of Kemp's

ridleys recaptured in Long Island Sound (4.1 cm/yr) is most similar to that observed in

Deadman Bay (4.21 cm/yr, Table 3-6) as is the size range (27-39 cm SSCL) of the turtles

upon which the estimates are based. Despite an activity period of less than 4 months per

year, Kemp's ridleys in northern developmental habitats appear to exhibit comparable

growth rates to ridleys from the northeastern Gulf of Mexico, where they are active at

least three-quarters of the year (Morreale and Standora 2005).

Headstarts

There appears to be a pattern of decreasing occurrence of headstarted Kemp's

ridley turtles from Apalachee Bay clockwise around the Florida Gulf coast. During the

1995 netting season of this study, 60% of the ridleys captured in Apalachee Bay were

positively identified as headstarts, whereas only one or two individuals were observed in

each of the following sites, (0.83%, 1 of 121) Deadman Bay, (0.78%, 2 of 254)

Waccasassa Bay, and possibly (0.52%, maybe 1 of 192) Gullivan Bay (Schmid 1998,









Witzell and Schmid 2004). The pattern is possibly a function of distance between the

release and capture locations. Two of the three turtles captured in Apalachee Bay were

released off Panama City, the nearest headstart release point to the study area. In

addition, a temporal shift may be occurring. Of 106 Kemp's ridley turtles captured in

surveys from 1984 to 1988 in the Apalachicola-Apalachee bays, no headstarted turtles

were identified (Rudloe et al 1991). Seven years later, most of the turtles captured in the

Apalachee Bay were headstarts.

One headstarted Kemp's ridley was captured in Deadman Bay. This 1986 year

class turtle was the only adult Kemp's ridley observed in the area during this study. A

female, she could not be further identified due to loss of all individual marks. No

readable PIT tag could be found in either fore-flipper, although a living tag in her

carapace was obvious, as was a tag scar in the right fore-flipper. Records of adult

Kemp's ridleys along the Florida west coast are rare, especially since the end of the

commercial fishery. Beginning in May 1989 Kemp's ridleys began nesting on both

coasts of Florida (Meylan et al. 1990). Since then, 21 incidences of nesting, or attempted

nesting, scattered through eight counties, including two in the panhandle, have been

documented through the 2005 nesting season (FWRI 2006). While none of these nesting

females could be positively identified as headstarts, those origins could not be ruled out

(Johnson et al. 1999). Bowen et al. (1994) saw this marked increase in aberrant nesting

11 to 14 years (the presumed time to maturation) post initiation of the headstart program

as "strikingly coincidental" (Schmid and Witzell 1997, Zug et al. 1997). While an

argument can be made that nesting by Kemp's ridleys has always occurred in Florida, it

seems very unlikely this species conspicuous habit of nesting during the day would have









gone unnoticed (Johnson et al. 1999). Additionally, the natal origin of this species was so

uncertain, largely due to the absence of any nesting or reproductive observations prior to

1960, it had earned the name "bastard" turtle because it was thought to be a hybrid

between a loggerhead, hawksbill, and perhaps a green turtle (Carr 1942, 1952). Early

investigations would have surely turned up reports of nesting had it ever been observed in

the state. A similar observation of an adult, 65.2cm SSCL, female Kemp's ridley was

made in Gullivan Bay. Though not specifically cited as such in publication, Schmid

(pers. com.) is confident she was a headstart turtle. A size frequency outlier, she was

marked only by a flipper tag scar, but no readable PIT or living tag (Witzell and Schmid

2004). The inability to confirm the origin of this suspected headstart turtle captured in

Gullivan Bay is one example of the many vagaries of this often criticized NMFS program

(Huff 1989, Woody 1991, Frazer 1992, Hewavisenthi 1993). Only 9.6% of the 8,026

individuals released in the first six years of the program had living tags, and they were

not applied to all turtles until 1990. Until 1988, no more than 5.6% of any year class was

PIT tagged, with an overall application rate of 27.5% of the first 22,576 headstarts

released. The greatest tagging rate (98.3%) of any of the methods was flipper tags. As

previously discussed, flipper tags are quickly lost and not suitable for long-term

reidentification, assuming retention rates similar to those of tags applied to wild turtles.

Further, gross errors in describing headstart marking techniques confound their

identification. For example, publications on the identification of headstart living tags use

scute and bone nomenclature interchangeably, and frequencies of PIT tags applied to

headstarts are incorrectly stated and could lead to the use of the wrong equipment to

attempt to read them (Fontaine and Shaver 2005, Caillouet et al. 1986). The use of coded









wire tags should also be noted. Beginning in 1984, they were applied to the right or left

flipper of all headstarted turtles. We were fortunate to have the rather esoteric and

expensive equipment required to read wire tags, but could not detect these tags in any of

the four headstarts captured in this study. Had the tag been readable, no information

other than the presence of the tag could be gained without surgical removal.

Future Research

Currently the state's only industrial river, the Fenholloway, is located

approximately 40 km north of Deadman Bay. It has been estimated that effluent from the

Buckeye Florida L.P. Kraft mill in Perry has resulted in the loss of 25 km2 of seagrass

beds near the river mouth. While the direct impact on marine turtles is unknown, the real

danger may lie in the sublethal effects of the consumption, and subsequent

biomagnification, of contaminated crabs and other prey. Plans to mitigate discharge into

the river have focused on a pipeline intended to bypass the river and pump effluent

directly into the Gulf of Mexico. Another potential source of pollution, a coal fired

powerplant meant to supply distant municipalities, is intended to be built on the same

river. Though the atmospheric deposition of mercury into local terrestrial, freshwater,

and marine foodwebs is certain, the impacts of powerplant intakes and coal delivery

barges and associated structures in the shallow waters of the area are not. While coastal

development of the "Forgotten Coast" is very low, it is increasing and will lead to more

boats on the water. Presently the area is enormously popular during the summer months

for the recreational scallop harvest. Though reports of marine turtle standings in the area

are low, this may not be a true reflection of the number of deaths, but the result of little

effort and the low-energy vegetated shoreline. Current statistics from the Florida Sea

Turtle Stranding and Salvage Network show that approximate 50% of ridleys and 33% of









the loggerheads stranded in Dixie and Taylor counties are related to boat impacts (Alan

Foley, pers. com.).

Tagging should continue in the area, but focused research should be directed at

behavioral studies addressing the effects of increased boating activity and seasonal

migrations. Further characterization of specific habitats and how turtles are using them

seasonally and annually should be emphasized. Toxicological research should be

considered to address the potential risks from pulp mill and coal powerplant in

conjunction with a trophic study to elucidate contaminant pathways
















LIST OF REFERENCES


Bjorndal, K. A., and A. B. Bolten. 1988. Growth rates of immature green turtles,
Chelonia mydas, on feeding grounds in the southern Bahamas. Copeia 1988:555-
564.

Bowen, B. W., T. A. Conant, and S. R. Hopkins-Murphy. 1994. Where are they now?
The Kemp's ridley headstart project. Conserv. Biol. 3:853-856.

Caillouet, C. W., C. T. Fontaine, S. A. Manzella, T. D. Williams, and D. B. Revera. 1986.
Scutes reserved for living tags. Mar. Turtle Newsletter 36:5-6.

Caillouet, C. W., C. T. Fontaine, S. A. Manzella-Tirpak, and D. J. Shaver. 1995. Survival
of head-started Kemp's ridley sea turtles (Lepidochelys kempii) released in the Gulf
of Mexico or adjacent bays. Chelonian Conserv. Biol. 1:285-292.

Caldwell, D. K., and A. Carr. 1957. Status of the sea turtle fishery in Florida. Trans. 22nd
N. Am. Wildl. Conf., p. 457-463.

Campbell, C. L. 1996. Capture of juvenile Kemp's ridleys in the nearshore waters of
Apalachee Bay Florida. In R. Byles and Y. Fernandez (compilers), Proceedings of
the 16th annual workshop on sea turtle biology and conservation, p. 28-29. U.S.
Dep. of Commer., NOAA Tech. Memo. NMFS-SEFSC-436.

Carr, A. 1942. Notes on sea turtles. Proc. New Eng. Zool. Club 21:1-16.

Carr, A. 1952. Handbook of turtles: The turtles of the United States, Canada, and Baja
California. Comstock Publ. Assoc., Cornell Univ. Press, Ithaca, NY, 542 p.

Carr, A. 1995. Notes on the behavioral ecology of sea turtles. In K.A. Bjorndal (ed.)
Biology and conservation of sea turtles, p. 19-29. Smithsonian Inst. Press,
Washington, D.C.

Carr, A., and D. K Caldwell. 1956. The ecology and migrations of sea turtles: 1. Results
of field work in Florida, 1955. Am. Mus. Nov. 1793:1-23.

Carr, A., L. Ogren, and C. McVea. 1980. Apparent hibernation by the Atlantic
loggerhead turtle Caretta caretta off Cape Canaveral, Florida. Biol. Conserv. 19:7-
14.

Collard, S. B., and L. H. Ogren. 1990. Dispersal scenarios for pelagic post-hatchling sea
turtles. Bull. Mar. Sci. 47:233-243.









Continental Shelf Associates, Inc., and Martel Laboratories, Inc. 1985. Florida Big Bend
seagrass habitat study narrative report. A final report by Continental Shelf
Associates, Inc to the Minerals Management Service, Metairie, LA. Contract No.
14-12-0001-30188.

Eckert, S. A., D. Crouse, L. B. Crowder, M. Maceina, and A. Shah. 1994. Review of the
Kemp's ridley sea turtle headstart program, 22-23 September 1992, Galveston,
Texas. U.S. Department of Commerce. NOAA Tech. Memo. NMFS-OPR-3. 9 p.

Fabens, A. J. 1965. Properties and fitting of the von Bertalanffy growth curve. Growth
29:265-289.

Florida Wildlife Research Institute (FWRI). 2006 Reported Nesting Activity of the
Kemps Ridley, Lepidochelys kempii, in Florida, 1979-2005. Fish and Wildlife
Research Institute, Florida Fish and Wildlife Conservation Commission Data
Summary Date: 13 Feb. 2006 Available:
http://www.floridamarine.org/engine/downloadredirection_process. asp?file=lk_79
-05_0218.pdf&objid=2377&dltype=article, Last accessed Aug. 2006.

Fontaine, C. T., D. B. Revera, T. D. Williams, and C. W. Caillouet. 1993. Detection,
verification and decoding of tags and marks in head started Kemp's ridley sea
turtles, Lepidochelys kempii. U.S. Department of Commerce. NOAA Tech. Memo
NMFS-SEFC-334. 40 p.

Fontaine, C. T., and D. Shaver. 2005. Head-starting the Kemp's ridley sea turtle,
Lepidochelys kempii, at the NMFS Galveston Laboratory, 1978-1992: a review.
Chelonian Conserv. and Biol. 4:838-845.

Frazer, N. B. 1992. Sea turtle conservation and halfway technology. Conserv. Biol.
6:179-184.

Geis, A. A., W. J. Barichivich, T. Wibbels, M. Coyne, A. M. Landry, and D. Owens.
2005. Predicted sex ratio of juvenile Kemp's ridley sea turtles captured near
Steinhatchee, Florida. Copeia 2005:293-298.

Henwood, T. A., and L. H. Ogren. 1987. Distributions and migrations of immature
Kemp's ridley turtles (Lepidochelys kempi) and green turtles (Chelonia mydas) off
Florida, Georgia, and South Carolina. Northeast Gulf Sci. 9:153-159.

Hewavisenthi, S. 1993. Turtle hatcheries in Sri Lanka: Boon or bane. Mar. Turtle
Newsletter 60:19-22.

Huff, J. A. 1989. Florida terminates "headstart" program. Mar. Turtle Newsletter 46:1-2.

Johnson, S. A., A. L. Bass, B. Libert, M. Marshall, and D. Fulk. 1999. Kemp's ridley
(Lepidochelys kempi) nesting in Florida. Fla. Sci. 3/4:194-204.









Landry, A. M., D. T. Costa, F. L. Kenyon II, and M. S. Coyne. 2005. Population
characteristics of Kemp's ridley se turtles in nearshore waters of the upper Texas
and Louisianan coasts. Chelonian Conserv. Biol. 4:801-807.

Lutcavage, M., and J. A. Musick. 1985. Aspects of the biology of sea turtles of Virginia.
Copeia 1985:449-456.

Magnuson, J. J., K. A. Bjorndal, W. D. DuPaul, G. L. Graham, D. W. Owens, C. H.
Peterson, P. C. H. Pritchard, J. I. Richardson, G. E. Saul, and C. W. West. 1990.
Decline of the sea turtles: causes and prevention. National Academy Press.
Washington, D. C. 274 p.

Marquez, M. R. 1994. Synopsis of the biological data on the Kemp's ridley turtle,
Lepidochelys kempi (Garman, 1880). U.S. Department of Commerce. Tech. Memo
NMFS-SEFSC-343. 91 p.

Meylan, A., P. Castandea, C. Coogan, T. Lozon, and J. Fletemeyer. 1990. First recorded
nesting of Kemp's ridley in Florida, USA. Mar. Turtle Newsletter 48:8-9.

Morreale, S. J., and E. A. Standora. 2005. Western North Atlantic water: crucial
developmental habitat for Kemp's ridley and loggerhead turtles. Chelonian
Conserv. Biol. 4:873-881.

Ogren, L. H. 1989. Distribution of juvenile and subadult Kemp's ridley turtles:
Preliminary results from the 1984-1987 surveys. In: C.W. Caillouet, Jr. and A.M.
Landry, Jr. (eds.), Proc. Of the 1st international symposium on Kemp's ridley sea
turtle biology, conservation, and management, p. 116-123. Texas A & M Univ.
(TAMU)-SG-89-105:116-123.

Pritchard, P., P. Bacon, F. Berry, A. Carr, J. Fletemeyer, R. Gallagher, S. Hopkins, R.
Lankford, R. Marquez M., L. Ogren, W. Prengle, Jr., H. Reichart, and R. Witham.
1983. Manual of sea turtle research and conservation techniques, second ed. K.A.
Bjorndal and G.H. Balazs (eds.), Center for Environmental Education, Washington,
D.C. 126 p.

Ross, J. P., S. Beavers, D. Mundell, and M. Airth-Kindree. 1989. The status of Kemp's
ridley. Washington, D. C.: Center for Marine Conservation. 51 p.

Rudloe, A., J. Rudloe, and L. Ogren. 1991. Occurrence of immature Kemp's ridley
turtles, Lepidochelys kempi, in coastal waters of northwest Florida. Northeast Gulf
Sci. 12:49-53.

SAS Institute Inc. 1999. SAS/STAT User's Guide, Version 8. SAS Institute Inc., Cary,
NC.

Schmid, J. R. 1995. Marine turtle populations on the east-central coast of Florida: Results
of tagging studies at Cape Canaveral, Florida, 1986-1991. Fish. Bull. 93:139-151.









Schmid, J. R. 1998. Marine turtle populations on the west-central coast of Florida:
Results of tagging studies at the Cedar Keys, Florida, 1986-1995. Fish. Bull.
96:589-602.

Schmid, J. R., and W. J. Barichivich. 2005. Developmental biology and ecology of the
Kemp's ridley turtle, Lepidochelys kempii, in the eastern Gulf of Mexico.
Chelonian Conserv. and Biol. 4:828-834.

Schmid, J. R., and W. J. Barichivich. 2006. Lepidochelys kempii-Kemp's ridley. In P. A.
Meylan (ed.), Biology and conservation of Florida turtles, p. inpress. Chelonian
Research Monographs.

Schmid, J. R., A. B. Bolten, K. A. Bjorndal, W. J. Lindberg, H. F. Percival, and P. D.
Zwick. 2003. Home range and habitat use by Kemp's ridley turtles in west-central
Florida. J. Wildl. Manage. 67:196-206.

Schmid, J. R., and L. H. Ogren. 1990. Results of a tagging study at Cedar Key, Florida,
with comments on Kemp's ridley distribution in the southeastern U.S. In T. I.
Richardson, J. I. Richardson, and M. Donnelly (compilers), Proceedings of the 10th
annual workshop on sea turtle biology and conservation, p. 129-130. U.S. Dep. of
Commer., NOAA Tech. Memo. NMFS-SEFSC-278.

Schmid, J. R., and L. H. Ogren. 1992. Subadult Kemp's ridley sea turtles in the
southeastern U.S.: Results of long-term tagging studies. In M. Salmon and J.
Wyneken. (compilers), Proceedings of the 11th annual workshop on sea turtle
biology and conservation, p. 102-103. U.S. Dep. of Commer., NOAA Tech.
Memo. NMFS-SEFSC-32.

Schmid, J. R., and W. N. Witzell. 1997. Age and growth of wild Kemp's ridley turtles
(Lepidochelys kempi): cumulative results of tagging studies in Florida. Chelonian
Conserv. and Biol. 2:532-537.

Schmid, J. R., and W. N. Witzell. Inpress. Seasonal migrations of immature Kemp's
ridley turtles along the west coast of Florida. Gulf Mex. Sci.

Shaver, D. J. 1994. Relative abundance, temporal patterns, and growth of seat turtles at
the Mansfield Channel, Texas. J. Herpetol. 25:491-497.

True, F. W. 1887. The turtle and terrapin fisheries. In G.B. Goode, The fisheries and
fishery industries of the United States, sec. 5, vol. 2, part 19, p. 493-454. U.S.
Comm. Fish Fish.

Turtle Expert Working Group. 1998. An assessment of the Kemp's ridley (Lepidochelys
kempii) and loggerhead (Caretta caretta) sea turtle populations in the western
North Atlantic. U.S. Department of Commerce. NOAA Tech. Memo NMFS-
SEFSC-409. 96 p.









Turtle Expert Working Group. 2000. Assessment update for the Kemp's ridley and
loggerhead sea turtle populations in the western North Atlantic. U.S. Department of
Commerce. NOAA Tech. Memo. NMFS-SEFSC-444. 115 p.

U. S. Fish and Wildlife Service (USFWS) and National Marine Fisheries Service
(NMFS). 1992. Recovery Plan for the Kemp's Ridley Sea Turtle (Lepidochelys
kempii). St. Petersburg, FL: National Marine Fisheries Service. 40 p.

Witzell, W. N., and J. R. Schmid. 2004. Immature sea turtles in Gullivan Bay, Ten
Thousand Islands, Southwest Florida. Gulf of Mex. Sci. 1:54-61.

Woody, J. B. 1991. Guest Editorial: It's time to stop headstarting Kemp's ridleys. Mar
Turtle Newsletter 54:7-8.

Zug, G. R., H. J. Kalb, and S. J Luzar. 1997. Age and growth on wild Kemp's ridley sea
turtles Lepidochelys kempii from skeletochronological data. Biol. Conserv. 80:261-
268.















BIOGRAPHICAL SKETCH

William James "Jamie" Barichivich was born in Fort Monmouth, New Jersey, on

May 21, 1970. He grew up in Sarasota, Florida, spending much of his time fishing in the

bays and Gulf of Mexico between Tampa Bay and Charlotte Harbor. He graduated from

Sarasota High in 1988 and received an Associate of Arts degree from Manatee

Community College. He began pursuing a degree in architecture at the University of

Florida in 1991 but changed majors to wildlife ecology and conservation in his second

year. His background as a waterman secured him a job with a generous PhD student, Jeff

Schmid, aiding in the capture and telemetry of Kemp's ridley sea turtles near the Cedar

Keys, Florida. He worked three field seasons on that project and graduated with a

Bachelor of Science in forest resources and conservation in the spring of 1996.

After graduation, he began working for the United States Geological Survey

(USGS) on the development of a capture/recapture study of marine turtles in the Florida

Big Bend. This work soon evolved into Jamie's thesis research when he began graduate

study in the Department of Wildlife Ecology and Conservation.

Jamie has continued to work for the USGS as his research interests have expanded.

Since 2000 he has studied amphibians as part of the Department of Interior Amphibian

Research and Monitoring Initiative, working across the Southeastern United States. He

has also worked on other USGS projects, including mitigation of vertebrate road

mortality, Gulf Sturgeon population census, and trophodynamics of deep sea benthic

communities.