Pristis Pectinata) Using Acoustic Monitoring and Tracking in a Nursery Habitat in Southwest Florida Lisa D

Home , Florida pompano, Houndfish, Orthopristis chrysoptera, Pristis

Florida State University Libraries

Electronic Theses, Treatises and Dissertations The Graduate School

2012 Monitoring Movement Patterns of Juvenile Smalltooth Sawfish (Pristis Pectinata) Using Acoustic Monitoring and Tracking in a Nursery Habitat in Southwest Florida Lisa D. Hollensead

Follow this and additional works at the FSU Digital Library. For more information, please contact [email protected] THE FLORIDA STATE UNIVERSITY

COLLEGE OF ARTS AND SCIENCES

MONITORING MOVEMENT PATTERNS OF JUVENILE SMALLTOOTH SAWFISH

(PRISTIS PECTINATA) USING ACOUSTIC MONITORING AND TRACKING IN A

NURSERY HABITAT IN SOUTHWEST FLORIDA

LISA D. HOLLENSEAD

A Thesis submitted to the Department of Biological Science in partial fulfillment of the requirements for the degree of Master of Science

Degree Awarded: Summer Semester, 2012 Lisa D. Hollenesad defended this thesis on June 26, 2012.

The members of the supervisory committee were:

R. Dean Grubbs Professor Co-Directing Thesis

Don R. Levitan Professor Co-Directing Thesis

Emily H. DuVal Committee Member

Jeanette L. Wulff Committee Member

John K. Carlson Committee Member

The Graduate School has verified and approved the above-named committee members, and certifies that the thesis has been approved in accordance with university requirements.

ACKNOWLEDGEMENTS

Where to start? Funding for this project was made possible with a grant from the Cooperative Endangered Species Conservation Fund (Section 6 ESA). I am grateful to Shelley Norton from the NOAA Southeastern Regional Office in St. Pete, Florida for her tireless work on helping acquire funds and organize the sawfish recovery team meetings. I am grateful to my committee members for all their help and insightfulness with regards to the design of this project and the writing of this thesis. I appreciate all of their time and their enthusiasm. Special thanks goes out to my co advisors Don and Dean. Don was always very willing to meet with me whenever I had questions and has a unique way of approaching data analysis that I admire. I could write a book on what I have learned from Dean the past four years. His knowledge of everything from natural ecology, to proper writing, to all the Florida locations of a particular eating establishment is impressive. He has got the great balance of being able to sit and discuss animal movement and habitat theory all afternoon and share a beer with you afterwards. It makes him a fantastic advisor. He certainly has made me a better speaker, writer and scientist. I also acknowledge John for not only serving on my committee but starting my career. He took me in as a college intern and from there has given me multiple opportunities to learn new things and supported me both professionally and personally. He has cooked me fabulous meals when I lived in Panama City, taught me how to fly fish, and let me share his beloved dogs. It is a rare occurrence to have such a great mentor and friend. I am lucky to have him. Also, he will not ever let me forget requesting projects involving field work while I was employed in Panama City which brings me to my next acknowledgements. Thank you so much to everyone who helped me with the field aspects of this project. They are A. Canning, E. Tewes, C. Custer, J. Bender, J. Boston, A. Pacicco and C. Peterson. Special recognition goes out to Drew Foley, Christine Singer, and Lisa Ailloud for enduring hours of misery through damp and murderously bugging conditions in the middle of the everglades during my night tracks. They never complained once and were always there to help me. I also thank Mollie Taylor for her effort during the 2010 season. I almost drowned her and we were nearly struck by lightening but we made it through.

iii

The field work for this project would have never gotten done without Kelcee Smith and Dana Bethea. Thank you, Kelcee for crawling through the mud to count mangrove prop roots, having your boyfriend bake for us, and purchasing our project mascot Greg Greg. Thanks to Dana for running the boat, taking all the photos, cooking all the meals, and helping me tag animals. As for the staging and logistics of this project, I am indebted to all the hardworking folks down at the FSU marine lab. They are fantastic people to work with. Special thanks goes to Bobby, Mark, and Dennis for all they did to keep me supplied with a truck and boat. Thanks as well to Maranda, Kathy, and Sharon for making sure I did not go broke during the field season. I acknowledge Barney Hale and his group from Terracon Industries Inc. They did an excellent job processing some rather stinky sediment samples. I am grateful for all their great work. Lastly, I want to thank my family. To my sister, Celeste: Thank you for putting up with me as a big sister. I am sure it was not easy. Thanks for always helping me “clean” my room and laughing at my dumb jokes when no else was. To my mother, Sandra: You say that society is always so quick to blame the world’s moms for all its problems. Well, you should know this thesis will be archived at FSU forever and available world wide and I am going to put on this on the record for everyone to see. Thank you for all you have ever done for me. You inspire me to be more than I ever thought I could be. You are simply the best. To my dad, Paul: Thank you for all your love and support. You are a fantastic dad and have always put me and the rest of the family first and I appreciate that so much. Again for the rest of the world to see: Go Cards!

TABLE OF CONTENTS

List of Tables ...... vi List of Figures ...... vii Abstract ...... ix CHAPTER ONE: ANALYSIS OF SMALL SCALE AND DAILY MOVEMENT PATTERNS OF JUVENILE PRISITIS PECTINATA IN A NURSERY HABITAT ...... 1 1.1 Introduction ...... 1 1.2 Methods...... 3 1.2.1 Study Areas ...... 3 1.2.2 Sampling ...... 4 1.2.3 Active Tracking ...... 5 1.2.4 Data Analysis ...... 5 1.3 Results ...... 7 1.3.1 Active Tracking ...... 7 1.3.2 Activity space as a function of track duration and home range testing ...... 8 1.4 Discussion ...... 9 1.5 Conclusions ...... 10 CHAPTER TWO: ASSESSING RESIDENCY TIME AND HABITAT USE OF PRISTIS PECTINATA USING ACOUSTIC MONITORING IN A NURSERY HABITAT ....38 2.1 Introduction ...... 38 2.2 Methods...... 41 2.2.1 Study Areas ...... 41 2.2.2 Sampling ...... 41 2.2.3 Acoustic Monitoring ...... 42 2.2.4 Habitat Characterization ...... 43 2.2.5 Data Analysis ...... 43 2.3 Results ...... 44 2.3.1 Residency Time and Corridor Movements ...... 44 2.3.2 Spearman Rank Test ...... 45 2.3.3 Logistic Regression ...... 45 2.4 Discussion ...... 46 2.5 Conclusions ...... 47 Appendix ...... 62 References ...... 103 Biographical Sketch ...... 109

LIST OF TABLES

1.1 Summary of P. pectinata active tracking in 2011 ...... 19

1.2 Comparison of activity space and rate of movement between study sites ...... 21

2.1 List of P. pectinata acoustically tagged in 2011 ...... 51

2.2 Residency time by sub region ...... 55

2.3 A matrix of Rho values resulting from the Spearman’s Rank correlation tests ...... 59

2.4 Results of three sets of logistic regressions (GLM)...... 60

2.5 Result summary of a stepwise regression (GLM) ...... 61

LIST OF FIGURES

1.1 Map of the federally assigned critical habitat ...... 13

1.2 Map of the study areas in Everglades National Park ...... 14

1.3 A map of Faka Union Bay located in the Ten Thousand Island National Wildlife Refuge ..15

1.4 Photo of methodology for attaching acoustic tags to the first dorsal of P. pectinata using a cattle ear tag ...... 16

1.5 Examples of animal fixes visualized in ArcGIS and the calculation of MCPs ...... 17

1.6 Examples of animal fixes visualized in ArcGIS and the calculation of kernel density estimates ...... 18

1.7 Low tide at Mud Bay ...... 20

1.8 Activity spaces (Minimum Convex Polygons) for all tracks across bays ...... 22

1.9 Activity spaces (kernel density estimates) for all tracks across all bays ...... 23

1.10 Average rate of movement (ROM) for all tracked fish across area...... 24

1.11 Activity spaces (Minimum Convex Polygons) for ebb versus flood tidal flow direction .....25

1.12 Activity spaces (kernel density estimates) for ebb versus flood tidal flow direction ...... 26

1.13 Average rates of movement (ROM) for ebb versus flood tidal flowdirection...... 27

1.14 Activity spaces (Minimum Convex Polygons) for high versus low tidal amplitude ...... 28

1.15 Activity spaces (kernel density estimates) for high versus low tidal amplitude ...... 29

1.16 Average rates of movement (ROM) for high versus low tidal amplitude ...... 30

1.17 Diel comparisons of activity spaces (Minimum Convex Polygons) ...... 31

1.18 Diel comparisons of activity spaces (kernel density estimates) ...... 32

1.19 Diel comparisons for average rates of movement (ROM) ...... 33

1.20 Activity space (100% Minimum Convex Polygons) over track duration for four Mud Bay fish...... 34

vii

1.21 Activity space (100% Minimum Convex Polygon) over track duration for two Chokoloskee Island fish ...... 35

1.22 Scatter of slopes (rate of habitat expansion) calculated for each tracked animal over total minutes tracked ...... 36

1.23 Measure of activity space (100% Minimum Convex Polygon) over time for age one ID 3306...... 37

2.1 Locations of anchored VR2w receivers within ENP ...... 49

2.2 Map illustrating detection radius of VR2w receivers ...... 50

2.3 Abacus plot showing residency times for all P. pectinata tagged in 2011 in ENP ...... 52

2.4 Abacus plots illustrating an emigration corridor though the Lopez River system ...... 53

2.5 Abacus plot illustrating ID 346 moving south from Mud Bay (MB) and exiting out the Lopez River system...... 54

2.6 Residency time within a subregion ...... 56

2.7 A stacked graph illustrating each tagged sawfish days present and movement between subregions ...... 57

2.8 Abacus plot illustrating greater presence in tidal bays among the back water region habitats (tidal bay, river, and creek) ...... 58

viii

ABSTRACT

Chapter 1 Habitat use studies can be used to both investigate ecological and behavioral patterns of animals as well as provide a useful management tool for conservation planners. However, essential habitat can be difficult to determine for highly mobile marine animals, especially when these species are rare or endangered. While critical habitat has been very broadly delineated for the endangered smalltooth sawfish (Pristis pectinata), essential fish habitat (EFH) within the nursery has not been fully described. I used telemetry methods to determine daily activity spaces and rates of movement (ROM) of juvenile P. pectinata in a nursery in southwest Florida. These results were tested for differences in diel and tidal patterns of activity. Seven juvenile animals ranging in size from 85 - 175 cm fork length were tagged in April - September 2011. Overall, activity spaces ranged from 0.07 - 0.17 km2 using 95% Minimum Convex Polygons (MCP), 0.01 - 0.16 km2 based on 50% kernel density estimates (KDE), and 0.08 - 0.68 km2 based on 95% (KDE). Average ROMs ranged from 2.4 - 6.1 meters/min. Activity space and ROMs reflected the morphology of the bay in which the animal was tracked such that fish in small bays had small activity spaces and ROMs. There were no detectable differences in activity space or ROM between ebb and flood tide or high or low tide. Activity space decreased and ROM increased at night indicating possible foraging behavior at night. A home range (1.7 km2) was calculated for one animal. Daily asymptotes in space used were reached for all other tracks suggesting daily activity spaces were determined despite relatively short tracking durations.

Chapter 2 Bays, estuaries, and other discrete coastal habitats are highly productive and serve as nurseries for a variety of marine fishes. Nurseries are particularly crucial for batoids whose life histories are dependent on rapid growth in the first year, and they may be especially important for rare or endangered species within the group. The smalltooth sawfish, Pristis pectinata, are an endangered marine elasmobranch that makes use of specific nurseries in southwest Florida. While habitat and environmental parameters have been described within the nursery, specific characteristics of the habitats, such as mangrove morphology and sediment types associated with habitat use have not been identified. Two mangrove characteristics (prop root density and limb overhang) and two sediment characteristics (percent organic and percent silt) were used as

ix independent variables to construct a habitat model. Acoustic monitoring was used to examine long term (weeks or months) patterns in habitat use in nursery areas during the critical first year of life. Twenty young-of-the-year sawfish were acoustically tagged between April and October of 2011, and detected by an array of 32 VEMCO VR2w receivers in a documented nursery within Everglades National Park. Presence in the array for individual smalltooth sawfish ranged from one day to 197 days, and overall P. pectinata were present within the acoustic array for 334 days. There was also evidence of overwintering specifically in Chokoloskee Bay. In the back water region (Turner River, Mud Bay, Cross Bays, Wilderness Waterway, and Lopez River), residency times were longer in tidal bays rather than creeks or rivers. A potential emigration corridor from the back water region was observed through the Lopez River. Using receiver data for animals moving between neighboring areas, a step wise logistic regression model in a generalized linear model framework for receiver hits per hour was significant for mangrove prop root density (Stepwise GLM: Partial R square = 0.22, C(p) = 6.02 p = 0.023). This model indicated a higher probability of seeing a juvenile smalltooth sawfish when mangrove prop root density was high.

CHAPTER ONE ANALYSIS OF SMALL SCALE AND DAILY MOVEMENT PATTERNS OF JUVENILE PRISTIS PECTINATA WITHIN A NURSERY HABITAT

1.1 Introduction

Sawfishes (Family: Pristidae) are a small group (7 species) of elasmobranches that inhabit tropical and subtropical waters worldwide. They are dorsolventrally flattened batoids and can be identified by a large toothed rostrum and two dorsal fins. Mature animals of some species can grow to lengths of five meters (Simpfendorfer et al. 2008). Globally, all species of sawfish have exhibited large declines in abundance and range contractions which led the International Union for Conservation of Nature to list them all as “critically endangered” (IUCN 2012). Therefore, life history information (habitat use, age at maturity etc.) is needed to implement effective conservation and management plans. In US waters, smalltooth sawfish, Pristis pectinata, historically ranged from east Texas to North Carolina along the Atlantic coast (NMFS 2009). However, the population is now primarily concentrated in southwest Florida (Wiley and Simpfendorfer 2010, Seitz and Poulakis 2002, Poulakis and Seitz 2004, Simpfendorfer and Wiley 2005). Extirpation from areas of historical concentration along the Atlantic coast such as the Indian River Lagoon (Snelson and Williams 1981) has been hypothesized. Because of this large population decline and range reduction, the United States National Marine Fisheries Service (NMFS) received a petition from the Center for Marine Conservation (now the Ocean Conservancy) requesting listing of the North American population of P. pectinata as endangered under the U.S. Endangered Species Act (ESA). Subsequently, NMFS conducted a formal status review (NMFS 2000) and on 1 April 2003, P. pectinata became the first domestic marine fish added to the U.S Endangered Species Act (2003, ESA Listing Rule for the U.S. DPS, Federal Register 68

FR 15674). By U.S. federal law, critical habitat must be established for an endangered species within five years of its listing. Critical habitat protection is implemented in an effort to protect areas that are or maybe crucial to the recovery of an endangered species. Little data existed concerning specific habitat requirements for P. pectinata, therefore NMFS designated most of southwest Florida and the Florida Keys as critical habitat in 2009 (Federal Register 74 FR 45353, Fig. 1.1). This habitat identification is important for initial conservation efforts (Primack 2006). However, more efficient conservation plans can be implemented if species-specific habitats (or areas of essential fish habitat) can be determined. For example, identification of nursery habitats allows for study of juvenile animals soon after their birth. Small concentrations of juvenile P. pectinata have been found over repeated years in the Ten Thousand Island National Wildlife Refuge and Everglades National Park (Simpfendorfer and Wiley 2005, Wiley and Simpfendorfer 2007, Simpfendorfer et al. 2010). However, an area containing elevated densities of juvenile individuals may not necessarily indicate an appropriate nursery habitat as other factors such as availability of limited resources, protection from predators and other factors affecting growth or mortality must be considered (Heupel et al. 2007). Delineation of nursery habitats is important for management and conservation and has been documented for many elasmobranch species (e.g. Morrissey and Gruber 1993a, Morrissey and Gruber 1993b, Heupel and Simpfendorfer 2002, Grubbs et al. 2007). Given the endangered status, identification of nurseries may be critical to recovery of the U.S. population of P. pectinata. Like most batoids growth is very rapid in the first few years of life (Simpfendorfer 2008), presumably an adaptation to limit the risk of predation. Juvenile smalltooth sawfish double in size during the first year of life (Simpfendorfer 2008). A P. pectinata nursery must have the prey biomass to facilitate the animal’s rapid growth the first year. Knowing the fine-scale spatial and temporal use of P. pectinata in its nursery habitat is the first step to understanding its juvenile ecology, and can be conducted using methods of acoustic telemetry tracking. Animal movement studies may not directly answer some of the questions involving an area’s nursery value, but it does provide a foundation upon which other components of a viable nursery habitat can be tested (e.g. possible seasonal variation in prey availability). Specifically, the goals

2 of this study were to measure daily activity space, test for differences in diel and tidal movement patterns, and measure home range in order to determine fine-scale habitat use of juvenile P. pectinata within a nursery. I used acoustic telemetry techniques and tested for differences in movement patterns using spatial analysis software. With these results, managers can more efficiently and accurately conserve nursery habitat that could increase survival through the critical first year for P. pectinata.

1.2 Methods

1.2.1 Study Areas

Active tracks were conducted in two bays in Everglades National Park (ENP), Mud Bay and Chokoloskee Island Bay (Fig. 1.2) and one bay, Faka Union, in Ten Thousand Island National Wildlife Refuge (Fig. 1.3). Mud Bay is the smallest of the study areas (0.21 km2) and is located in the back waters of ENP. A small mangrove island in the northern half of the bay is known to be used by juvenile smalltooth P. pectinata in abundance over multiple years (Simpfendorfer et al. 2010). While much of this island is exposed during low tides, the southern portion of the bay remains submerged for the duration of the tidal cycle. A relatively deep cut runs through the most northern edge and connects the two openings of the bay. This cut experiences a high rate of flow during ebb and flood tides. The west bay of Chokoloskee Island is located at the very northeastern border of ENP, and is the largest of the study areas (3.06 km2). This island is anthropogenically altered making it unique among other habitats in the surrounding national park. A man- made causeway links the island to the mainland, and it is lined on either side with mangroves. Active tracking took place on the western side of the causeway. In this bay, the tide propagates as a large wave moving against the causeway flooding mangroves during high tide and retreating out during ebb tide leaving the surrounding flat exposed. The third site is located within the Ten Thousand Island National Wildlife Refuge in an area called Faka Union Bay (1.24 km2). This bay is located at the terminus of a man-made channel, and the resulting spoil island has historically been a location of several juvenile P. pectinata encounters across years (Simpfendorfer et al. 2010). Much

like the island in Mud Bay, this island is accessible to juvenile smalltooth sawfish during the high tide, but the flat surrounding the island is exposed during low tide. While these three areas differ in morphology and anthropogenic disturbance, they have similar overall features and experience the same seasonal abiotic trends. They are all highly turbid, shallow, and lined with mangrove shorelines. During the dry season (November - May) salinities can be up to 37 (practical salinity scale), while in the rainy season (June - October) salinities can be as low as 7 (practical salinity scale). Water temperatures in the summer months (June - August) are > 35 °C for the majority of the summer across all areas.

1.2.2 Sampling

Juvenile P. pectinata were collected under guidelines approved by Protected Species Permit # SEFSC-NMFS-13330 with permission to collect in Everglades National Park approved by permit # EVER-2011-SCI-0010. Collection for juvenile P. pectinata occurred during March - October 2011. Gear used for animal capture included two 30.5 and two 61 m monofilament gillnets each with 76 and 102mm stretched mesh. Nets were soaked (placed and oriented in the water column so has to maximize the area fished) and checked very 0.5 hour and fished in a particular area for half of a tidal cycle. Abiotic factors including salinity (practical salinity scale), dissolved oxygen (mg/L), and temperature (°C) were taken at water surface at every fished location using a YSI-85 environmental meter (YSI Inc. Yellow Springs, Ohio). Captured animals were carefully removed from the gillnet and brought alongside the boat for processing. During this process the animal’s mouth and spiracles remained underwater to allow the animal to breathe normally and reduce stress. Rostral length, precaudal length, fork length (tip of rostrum to fork of caudal fin), and stretch total length were measured. Sex was also recorded with males being identified by the presence of claspers. Animals less than 1.5 m fork length (the approximate size at one year of age) were fitted with a VEMCO (VEMCO Ltd. Halifax, Nova Scotia) V9 or V13 (depending on animal size) continuous tag by affixing the tag to a Primer 1 Swivel TM Ear Tag (Premier1Supplies© Washington, Iowa) using epoxy. The tag was attached through the most anterior portion of the first dorsal fin close to the body (Fig. 1.4.) as thicker

4 connective tissue in this region allowed for better tag retention. Tags emitted an acoustic signal at a particular frequency (between 60 and 81 kHz) every 2-3 seconds.

1.2.3 Active Tracking

Acoustic “pings” emitted by the continuous tags were received by a VEMCO VR100 directional hydrophone with receiver. The receiver was able to monitor up to 8 frequencies simultaneously, so multiple individuals could be tracked at the same time. Animals were tracked either by an anchored motor vessel with its engine off or monitored using a kayak. A space of approximately 60 m was maintained between the tracker and the animal. These steps were taken to minimize any influence tracking may have on the animal’s natural movement and behaviors. “Fixes” were recorded every 10 minutes during the track. A fix includes latitude and longitude position of the tracking vessel, a compass bearing based on the direction of the hydrophone, and an intensity of the ping which is displayed on the VR 100 interface. Tracking was conducted with a goal of following individuals for at least 24 hours.

1.2.4 Data Analysis

I analyzed fix data using Microsoft Excel 2003. The information recorded from these fixes was used to triangulate the position of the tracked P. pectinata. First, the latitude and longitude positions of the vessel were converted from decimal degrees to Universal Transverse Mercator (UTM) values using an online conversion website (http://www.rcn.montana.edu/resources/tools/coordinates.aspx). Distance in the UTM projection is measured in meters, and the Cartesian coordinate system makes for simple triangulation calculations which are ideal in scientific reporting (White and Garrott 1990). In Excel, the compass bearings were converted into radians, and the intensity recorded by the VR100 was converted to meters using information obtained during range testing of the acoustic equipment at the beginning of the study. The following equations were then calculated in Excel to identify the position of the animal during the track:

UTM Fish X = (UTM Boat X + (Distance * (sin(radians)))

UTM Fish Y = (UTM Boat Y + (Distance * (cos(radians)))

The calculated positions were then visualized in ArcGIS 9.3 (ERSI, Redlands, California, USA) for analysis. Minimum Convex Polygons (MCPs) and Kernel Density Estimates (KDE) were calculated as estimates of activity space. Non parametric methods were used in analysis of active tracking because spatial data is inherently autocorrelated (Worton 1989, White and Garrott 1990). These are the most common analysis tools used among active tracking studies, and therefore allow comparisons with other studies. MCPs and density kernels were calculated using the Home Range Tools extension (Rodgers et al. 2007) for ArcGIS 9.3. MCPs are calculated by drawing a polygon around the most outer points of a track and measuring the area. Most often, 5% of points farthest from the center of the polygon are removed as outliers, and the resulting 95% MCPs are reported. Though MCPs are easily calculated, core activity space within these polygons is often more informative about an animal’s habitat use (Fig. 1.5). Kernel density estimates were therefore used to calculate core areas of movement. In this case, a normal bivariate density distribution was applied to data points within an assigned area or kernel. A grid can then be placed over these distributions, and the amount of distribution overlap between the grids was measured. Areas with high calculated overlap are areas with a large number of points very close together (Worton 1989). However, placing a distribution function to over dispersed data resulted in over fitting of the data and required the use of a smoothing parameter h (Gitzen 2006). Selecting h is important as over-smoothing or under-smoothing data can cause a misinterpretation of the final estimated activity space results. While different smoothing parameters exist (reference bandwidth, least-square cross-validation, and biased cross- validation), the smoothing parameter that yielded the best results for this study was the

reference bandwidth, href. It is calculated by the following equation (Worton 1995):

-(1/6) href = n √(varx+ vary)/2

where, n equals the number of data points, and varx and vary are the variances around the x and y coordinates. The smoothing parameter href works best for unimodial or “nested”

telemetry data (Worton 1995 and Gitzen 2006) and is most appropriate for data points in high concentrations within a single core area. This was the observed pattern for P. pectinata tracked locations. I reported 50% and 95% isopleths from the KDEs (Fig. 1.6). The Home Range Tool extension also calculated rate of movement (ROM) between each triangulated location. ROM is described as:

ROM = (distancen+1 – distancen) / (time n+1 – timen)

Differences in activity space and ROM between tidal flow, tidal amplitude and day versus night were compared using Wilcoxon Signed Rank Tests. All comparison testing was performed in the statistical package R (R Development Core Team 2011). To test for bias in home range estimation as a function of track duration, 100% MCPs were calculated using sequential fixes for each individual over the time of the individual’s track. I calculated an individual’s home range if these individual points (or fixes) resulted in an asymptote where ten consecutive observations resulted in a <1% increase in area.

1.3 Results

1.3.1 Active Tracking Seven animals were manually tracked (Table 1.1) between April - October 2011. Tracking periods for individual fish ranged from 3 to 74 hours for a total of 224 hours tracked and a total of 952 fixes. The longest consecutive individual track lasted 36 hours. Four animals were tracked in Mud Bay, two were tracked on the western side of Chokoloskee Island and one age 1 animal (this animal had been captured in August 2010) was tracked at Faka Union. Two animals in Mud Bay were tracked across multiple months. In April, four animals (ID 338, 399, 340, 341) were monitored immediately after being tagged. A paired t-test was conducted to determine if a post-tagging recovery period should be eliminated from subsequent analysis of movement. A comparison between the average ROM for the first hour and the average ROM for the rest of the track was compared. There was no difference in the average ROM between the first hour, and the rest of the track (t-test t = 1.2, df = 3, p value = 0.3) for the four fish tracked

7 immediately after tagging so all fixes were used for analysis. All other tracked animals had at least a 24 hour recovery period before tracking, so all fixes were used for analysis. The acoustic equipment did not perform well in shallow (approximately less than 0.5 m) highly turbid environments and animals were often lost at low tide hampering the continuous tracking. This problem was exacerbated by individual animals using small tidal pools during low tides (Fig. 1.7). Continuing to monitor the animal in small tidal pools would have violated the 60 m distance buffer that ensured normal animal behavior. When this happened, I abandoned tracking efforts targeting that individual and searched for other animals that had moved to other portions of the bay that were still submerged. Overall, the average activity spaces and average ROMs among all fish reflected the size and morphology of the bay in which the fish were tracked (Table 1.2). Mud Bay animals trended towards smaller activity spaces, and Chokoloskee Island fishes trended toward higher activity spaces (Fig. 1.8 and Fig. 1.9). Overall, average ROMs trended low at Mud Bay and high at Chokoloskee Island (Fig. 1.10). For tidal flow, there was no statistical difference between ROM (Wilcoxon Signed Rank Test: V = 3, df = 5, p = 0.17), 95% MCP (V = 9, df = 5, p = 0.84), 50% KDE (V = 8, df = 5, p = 0.69), and 95% KDE (V = 9, df = 5, p = 0.84) (Fig. 1.11 - 1.13) between ebb and flood tides. There was also no difference between high and low tide for ROM (V = 12, df = 6, p = 0.81), 95% MCP (V = 17, df = 6, p = 0.69), 50% KDE (V = 16, df = 6, p = 0.81) and 95% KDE (V = 14, df = 6, p = 1.0) (Fig. 1.14 - 1.16). Only individuals that had observations for both night and day were used in diel analysis. Since only four animals were tracked both day and night, mean comparison testing was not conducted. There was a slight trend in MCP and kernel activity space between day and night (Fig. 1.17 and Fig. 1.18). Night generally had lower activity space than day. There was a large trend towards higher ROMs at night than during the day (Fig. 1.19). 1.3.2 Activity space as a function of track duration and home range testing Only ID 3306 fulfilled the requirements of a defined home range (Fig. 1.23). The curve reached an asymptote with a less than 1% increase area at 0.17 km2 after 36 hours of observation. Regression curves for raw 100% MCP over time tracked yielded asymptotes for daily activity space, but not for the overall track for all other tracks (Fig.

1.20 and 1.21). To test for bias related to tracking duration, activity space and time were log transformed and fitted to a linear function to calculate a slope value for each track. These linear slopes were compared as a function of total minutes observed for all tracks. There was no relationship between the track slopes and over all track duration (Fig. 1.22) suggesting that rate of activity space expansion was independent of time tracked. 1.4 Discussion Diel trends suggest that individual smalltooth sawfish have increased ROMs, but a decrease in activity space at night. Rapid movement in smaller areas may indicate feeding behavior at night. During night tracks in Mud Bay, a large bait fish school (most likely mullet, Mugil sp., and ladyfish, Elops saurus) was observed on the flat around the northern mangrove island every night. Several P. pectinata were tracked in the middle of the bait fish school, and could have been feeding (L. Hollensead, personal observation). There are two possible explanations for the inability to determine home range for all but one sawfish tracked in this study. First, the duration of tracks may have been inadequate to determine a home range. Logistical constraints due to tides and the remote field site limited continuous tracking over 36 hours and tracks in months during the wet season were often suspended due to afternoon thunderstorms. Second, young-of-the-year P. pectinata may not establish home ranges. While daily asymptotes in activity space were reached, habitat use continued to expand over multiple days. Home ranges are typically established to optimize resource allocation due to density-dependent effects such as competition. If resources are not limited, home ranges may be unnecessary. Young-of-the-year juveniles may exhibit more random movements due to inexperience in foraging and predator avoidance from juvenile bull sharks, Carcharhinus leucas, which inhabit the same estuaries (Wiley and Simpfendorfer 2007). Only one P. pectinata, a one-year old animal who was much larger than others tracked in this study, exhibited a defined home range. A one-year old sawfish is twice the size of a neonate and therefore likely faces a much lower predation risk. Average ROMs in the current study were similar to those observed by Simpfendorfer et al. (2010) in both Mud Bay (2.87 m/min) and Faka Union (3.25 m/min). Simpfendorfer et al. (2010) suggested three P. pectinata in their study (two were young- of-the-year and one was a one-year old) exhibited home ranges, however, they used a

different method to calculate home range. The previous study’s longest individual track was 24 hours, and the analysis run was a Monte Carlo random walk comparison to test for site fidelity. The result of this analysis was said to be the animal’s home range (ranging between 0.004 – 0.1 km2). Since no tracks were longer than 24 hours, these estimates would more appropriately describe daily activity space. For all but one track in the current study, only daily activity spaces were determined with an increasing rate of expansion even with tracking durations longer than 24 hours. It is plausible that juvenile P. pectinata (especially young-of-the-year) do not exhibit a home range. If resources are not limited and competition is low, there is no ecological reason to limit movements to a defined home range. In this study, activity space of P. pectinata instead reflected the size of the bay where tracking occurred. 1.5 Conclusions Accurate delineation of essential fish habitat (EFH) especially critical nursery areas is vital to the successful recovery of an endangered marine species. It is convenient for managers to protect discrete areas of well established home ranges for a species. However on a broad scale, young-of-the-year sawfish may not always adhere to traditional bounds of a home range in certain areas. It is possible that P. pectinata behave differently depending on area and that management strategies may have to be tailored to meet these dissimilarities. While I described fine-scale and daily patterns of movement for juvenile smalltooth sawfish in a nursery, the broader implications may lie in the results from the active tracks over time. Simply measuring daily activity spaces may underestimate the amount of area required for a smalltooth sawfish in its first year of growth, and this could have implications for proper management. This study found only one animal that exhibited a home range, and this one result could also be argued as constraint of the morphology of Faka Union Bay. Instead, I would encourage managers to examine and extrapolate the linear transformations in this study when thinking about managing nursery areas for juvenile P. pectinata. Rate of activity space expansion and bay size must be accounted for when estimating the area required for a young-of-the-year to grow into an age 1 juvenile.

Discrete areas of juvenile abundance may require even more conservation consideration given the potential female philopatry seen in sawfish. Genetic studies of sawfishes in Australia have suggested high female philopatry to certain pupping areas (Phillips et al. 2011). A similar genetic study of P. pectinata was inconclusive as to female philopatry (Chapman et al. 2011). The ability of young of the year P. pectinata to adapt and quickly move as well as the presence of suitable neighboring habitat may alleviate dangers of such birthing strategies where recruitment failure can occur due to damages in particular critical habitats. If the P. pectinata pupping strategy involves female philopatry, daily habitat expansion rather than a constrained home range behavior would be advantageous for young-of-the-year smalltooth sawfish. Juveniles could move into other areas in the event of habitat disturbance of a site of high fidelity for pregnant females. Also, in areas with high concentrations of young-of-the-year during peak pupping seasons, rapid expansion into other areas could limit intraspecies competition for resources. Because daily activity spaces for juvenile smalltooth sawfish are relatively small within the first few months of life, individuals found in areas of high environmental disturbance maybe particularly vulnerable. Juvenile smalltooth sawfish found in and around the Caloosahatchee River, an area with extremely high anthropogenic influences, can be found in high concentrations over multiple years in specific areas or “hot spots” along the river (Poulakis et al. 2011). Since the river is controlled by a dam system, freshwater influxes can cause aggregations of animals of various cohorts to be constrained at the river mouth, and it is uncertain as to how these types of events affect survival or population recovery (Simpfendorfer et al. 2011). Direct comparison of long term and small scale movement patterns between the two nurseries should be conducted to discern the differences in movement strategies between the two areas. This would include direct comparisons between the rate of habitat expansion between nurseries. In this study, I examined the short term and fine-scale movement behaviors of young-of-the-year P. pectinata using acoustic telemetry tracking in order to examine core habitat use within the nursery area. Daily activity spaces are very small and overall movements suggest no defined home range or tidal patterns. The next step in understanding the spatial ecology of this species is to examine what happens at larger

11 temporal scales. How long does an individual stay within the nursery area during its first year? Are there discrete habitats within the nursery area that have higher residency time than others, and what are the biotic characteristics of these areas? These questions are addressed in Chapter 2 using acoustic monitoring.

Figure 1.1: Map of the federally assigned critical habitat. Shaded areas indicate federal designation of P. pectinata critical habitat as determined by NOAA NMFS in 2009. Map credit to Amanda Frick NOAA South East Regional Office.

A. B.

Figure 1.2: Map of the study areas in Everglades National Park (A) Map of Mud Bay: Located in the backwaters of ENP. The mangrove island in the northern section of the bay is an area high in P. pectinata encounters. (B) Map of Chokoloskee Island Bay: This area experiences high anthropogenic influence. A mangrove lined man made causeway links the island to the mainland.

Figure 1.3: A map of Faka Union Bay located in the Ten Thousand Island National Wildlife Refuge. A spoil island (highlighted in red) created by a dredged canal at the northern side of the bay, is an area high in juvenile P. pectinata encounters.

Figure 1.4: Photo of methodology for attaching acoustic tags to the first dorsal of P. pectinata using a cattle ear tag. Photo courtesy of Dana M. Bethea, NOAA NMFS Panama City Laboratory.

A. B. Figure 1.5: Examples of animal fixes visualized in ArcGIS and the calculation of MCPs. (A) Animal locations recorded during 28 hour track visualized in ArcGIS 9.3. (B) A 95% MCP (drawn in red) calculates the area in which 95% of the fix points are encircle

A. B. Figure 1.6: Examples of animal fixes visualized in ArcGIS and the calculation of kernel density estimates. (A) Animal locations during a 28 hour track visualized in ArcGIS 9.3. (B) Resulting data layer from kernel density analysis. Colors distinguish the core areas within the track with higher utilization represented by the warmer colors. Solid lines indicate the 50% (innermost), 75% (middle) and 95%(outermost) kernel.

Table 1.1: Summary of P. pectinata active tracking in 2011. Tag ID, fork length, sex, area caught, and month(s) tracked for seven fish. Hours represent how long an individual was tracked and fixes indicate how many animal locations were recorded during the track.

Fork Hours Length Fish ID Tracked Fixes (cm) Sex Area Month

338 28 116 85 F Mud Bay April

339 20 78 87 F Mud Bay April/May

340 28 107 86.5 F Mud Bay April

341 74 154 89 M Mud Bay April/May

345 12 96 95.5 M Chok Isl June

347 25 176 98.5 F Chok Isl July

3306 37 225 175 M Faka Union October

Figure 1.7: Low tide at Mud Bay. During these low tides, tracked P. pectinata would use these small tides pools and the track was abandoned, because attempting to track in such close proximity would have violated the set 60 m buffering space between animal and tracker. Photo courtesy of Dana M. Bethea, NOAA NMFS Panama City Laboratory.

Table 1.2: Comparison of activity space and rate of movement between study sites. Averages and standard deviations (for all but Faka Union) were calculated for activity spaces and rates of movement (ROM). Activity spaces were calculated as 95% Minimum Convex Polygons (95% MCP), 50% kernel density estimates (50% KDE) and 95% kernel density estimates (95% KDE). Activity spaces and ROM averages reflect bay morphology Mud Bay was the smallest bay, Faka Union was middle sized, and Chokoloskee Island Bay was the largest.

ROM 95% MCP 50% KDE 95% KDE Area (meters/min) (km2) (km2) (km2) n Mud Bay 2.41 (±0.24) 0.07 (± 0.02) 0.01 (± 0.01) 0.08 (± 0.05) 4 Faka Union Bay 3.84 0.1 0.06 0.24 1 Chokoloskee Island Bay 6.06 (± 0.26) 0.17 (± 0.05) 0.16 (± 0.03) 0.68 (± 0.03) 2

Mud Bay Fish Chok. Island Fish Faka Union Fish

Figure 1.8: Activity spaces (Minimum Convex Polygons) for all tracks across bays. Activity space reflected bay morphology with the smallest bay, Mud Bay, exhibiting the lowest activity spaces, and the largest bay around Chokoloskee Island showing the largest activity spaces. Colored bars represent the bay in which the animal was tagged and tracked. Grey is Mud Bay, black is Faka Union, and red is Chokolosskee Bay.

Mud Bay Fish Chok. Island Fish Faka Union Fish

A. B. Figure 1.9: Activity spaces (kernel density estimates) for all tracks across all bays. (A) Fifty percent kernel density estimates of activity space of all seven tracked fish across areas. (B) Ninety five percent kernel density activity space of all seven tracked fish across areas. Activity space reflected bay morphology with the smallest bay, Mud Bay, exhibiting the lowest activity spaces, and the largest bay around Chokoloskee Island showing the largest activity spaces. Colored bars represent the bay in which the animal was tagged and tracked. Grey is Mud Bay, black is Faka Union, and red is Chokolosskee Bay.

Mud Bay Fish Chok. Island Fish Faka Union Fish

Figure 1.10: Average rates of movement for all tracked fish across area. Average ROMs reflected bay size morphology with the smallest bay, Mud Bay, exhibiting the lowest ROMs, and the largest bay around Chokoloskee Island showing the highest ROMs. Colored bars represent the bay in which the animal was tagged and tracked. Grey is Mud Bay, black is Faka Union, and red is Chokolosskee Bay.

Figure 1.11: Activity spaces (Minimum Convex Polygons) for ebb versus flood tidal flow direction. Box plots were created around the median (dark line) and box “edges” represent the first and third quantile. Plot “whiskers” indicated highest and lowest values while dots represent outliers. For the comparisons, ebb and flood tidal flow had a sample size of six individuals.

A. B. Figure 1.12: Activity spaces (kernel density estimates) for ebb versus flood tidal flow direction. (A) 50% kernel density estimates and (B) 95% kernel density estimates of activity spaces were calculated for flood and ebb tides. Box plots were created around the median (dark line) and box “edges” represent the first and third quantile. Plot “whiskers” indicated highest and lowest values while dots represent outliers. For the comparisons, ebb and flood tidal flow had a sample size of six individuals.

Figure 1.13: Average rates of movement (ROM) for ebb versus flood tidal flow direction. Box plots were created around the median (dark line) and box “edges” represent the first and third quantile. Plot “whiskers” indicated highest and lowest values while dots represent outliers. For the comparisons, ebb and flood tidal flow had a sample size of six individuals.

Figure 1.14: Activity spaces (Minimum Convex Polygons) for high versus low tidal amplitude. Box plots were created around the median (dark line) and box “edges” represent the first and third quantile. Plot “whiskers” indicated highest and lowest values while dots represent outliers. For the comparisons, high and low tidal amplitude had a sample size of seven individuals.

A. B. Figure 1.15: Activity spaces (kernel density estimates) for high versus low tidal amplitude. (A) 50% kernel and (B) 95% kernel activity spaces compared as high versus low tidal amplitude. MCP activity spaces compared as high versus low tidal amplitude. Box plots were created around the median (dark line) and box “edges” represent the first and third quantile. Plot “whiskers” indicated highest and lowest values while dots represent outliers. For the comparisons, high and low tidal amplitude had a sample size of seven individuals.

Figure 1.16: Average rates of movement (ROM) for high versus low tidal amplitude. MCP activity spaces compared as high versus low tidal amplitude. Box plots were created around the median (dark line) and box “edges” represent the first and third quantile. Plot “whiskers” indicated highest and lowest values while dots represent outliers. For the comparisons, high and low tidal amplitude had a sample size of seven individuals.

Figure 1.17: Diel comparisons of activity space (Minimum Convex Polygons). Overall, MCP activity space area tended to be larger during the day than at night.

A. B.

Figure 1.18: Diel comparisons of activity spaces (kernel density estimates). (A) 50% kernel density estimates and (B) 95% kernel density estimates were calculated for day and night. Overall, activity space area tended to be larger during the day than at night.

Figure 1.19: Diel comparisons for average rates of movement (ROM). ROM tending to be higher during the day than at night.

Figure 1.20: Activity space (100% Minimum Convex Polygons) over track duration for four Mud Bay fish. ID 339 and 340 were both tracked in two different months (April and May). Daily activity space measurements were estimated, but total movement area did not asymptote and therefore home range could not be calculated.

Figure 1.21: Activity space (100% Minimum Convex Polygon) over track duration for two Chokoloskee Island fish. Daily activity space measurements were estimated, but total movement area did not asymptote and therefore home range could not be calculated.

35 Slope v Mins Tracked All y = -1E-04x + 1.0682 R2 = 0.1671 1.6 1.4 1.2 1 0.8

Slope 0.6 0.4 0.2 0 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 Minutes Tracked

Figure 1.22: Scatter of slopes (rate of habitat expansion) calculated for each tracked animal over total minutes tracked. ID 339 and ID 341 had April and May tracks which were calculated separately for the two months. Therefore, there are nine data points though only seven fish were tracked.

Figure 1.23: Measure of activity space (100% Minimum Convex Polygon) over time for age one ID 3306. The plot reached an asymptote and there fore a define home range at 0.171 km2.

37 CHAPTER TWO ASSESSING RESIDENCY TIME AND HABITAT USE OF JUVENILE PRISTIS PECTINATA USING ACOUSTIC MONITORING IN A NURSERY

2.1 Introduction

Discrete coastal habitats that are highly productive, such as bays and estuaries, have been shown to serve as nurseries for several marine fishes (e.g. Carlson and Straty 1981, Laedsgaard and Johnson 1995, Layman 2000, Meyer et al. 2000 Ross and Lancaster 2002, Pape et al. 2003) including elasmobranchs (e.g. DiGirolamo et al. 2012, Curtis et al. 2011, Carlson et al. 2008, Grubbs et al. 2007, Holland et al. 1993, Castro 1993). These nursery habitats often support juvenile fishes by providing benefits such as food resources for growth and protection from predation as well as other benefits (Spinger 1967). Once these habitats are identified, however, few studies quantitatively measure the environmental and physical characteristics of these areas that often drive their function as nurseries (Morrissey and Gruber 1993a, Morrissey and Gruber 1993b, Grubbs et al. 2007, Knip et al. 2011). Habitat uses and what drives animal movements are often overlooked. The definition of nursery habitat has been refined and is often taxon-dependent (Heck et al. 2003, Dahlgren et al. 2006, Heupel et al. 2007, Knip et al. 2010). Regardless of species, it is imperative to understand how juveniles use the habitat and to assess the area’s functions as a nursery. Identifying habitat characteristics that are correlated with species presence can aid conservation managers when developing management plans for exploited species and recovery plans for rare or endangered species (Primack 2006, Simpfendorfer et al. 2010). Telemetry is an effective method to ascertain habitat use within an area, and can indirectly provide information regarding other ecological questions including population abundance (Godin 1997), diet and resource selection (Manly et al 1993), behavior (Gosling and Sutherland 2000), taxonomy (Cody 1985, Godin 1997), and migration patterns (Cody 1985). In the event that basic life history questions cannot be directly

38 determined (e.g. lethal sampling) due to conservation laws, as is the case with endangered species, investigations as to the motivation behind habitat use can be used to infer about other ecological processes (Godin 1997). In spatial ecology the terms habitat use, habitat selection, and habitat preference are often used interchangeably in the literature but have distinct definitions. Johnson (1980) defined habitat use as resource consumption by an animal over a fixed temporal scale. Habitat selection refers to disproportionate use of a resource relative to its availability whereas habitat preference would indicate that an observer would have some indication of an animal’s “wants”. Likewise, measuring resource/consumer interactions is difficult in marine settings especially when examining a large mobile predator (Heithaus et al. 2002). For marine environments, habitat variables associated with animal presence or absence are often used to describe habitat use and selection (Godin 1997). Identifying an animal’s habitat is important for conservation (Primack 2006). Identification of habitat, especially nursery habitat, is so important to the life history of animals, the United States federal government requires critical habitat be delineated no more than five years after a species is listed on the Endangered Species Act (ESA). Pristis pectinata was the first domestic marine fish listed by the ESA (2003, ESA Listing Rule for the U.S. DPS, Federal Register 68 FR 15674) and critical nursery habitat was determined for southwest Florida in 2009 (Federal Register 74 FR 45353). The area defined is well documented for having discrete concentrations of juvenile P. pectinata (Seitz and Poulakis 2002, Poulakis and Seitz 2004, Simpfendorfer and Wiley 2005). Nursery habitats for P. pectinata are of particular interest because of the rapid growth rate they experience within the first year of life (Simpfendorfer et al. 2008). It is believed that females have protracted parturition between November and July with a peak in April/May (Poulakis et al. 2011). Pups are born at about 85 cm fork length (tip of rostrum to fork of caudal fin), but double in size within their first year (Simpfendorfer et al. 2008). This life history strategy is common to batoids and may facilitate rapid decrease in predation risk. For young-of-the-year P. pectinata, potential predators such as bull sharks, Carcharhinus leucas and lemon sharks, Negaprion. brevirostris co-occur in the nursery habitats (Wiley and Simpfendorfer 2007).

39 Telemetry studies within nursery habitats in Everglades National Park (ENP) (Simpfendorfer et al. 2010, Chapter 1 of this thesis) and the Caloosahatchee River (Poulakis et al. 2011) have been conducted on juvenile P. pectinata in southwest Florida. These studies indicated that juvenile P. pectinata have small daily activity spaces (Simpfendorfer et al. 2010) and that discrete areas of concentration exist which are stable over several years (“hot spots”) (Poulakis et al. 2011, Simpfendorfer et al. 2010) within the larger nursery area. Habitat models using environmental data suggest juvenile P. pectinata abundances are highest in salinities ranging from 18 to 30 (practical salinity scale), water temperature of greater than 30°C, dissolved oxygen of greater than 6 mg/L, water depths less than 1 meter, and close proximity to mangrove lined habitat (Poulakis et al. 2011, Wiley and Simpfendorfer 2010, Simpfendorfer 2006). Affinity for shallow water has also been demonstrated using telemetry methods for juvenile Pristis microdon found in river systems in Northern Australia (Whitty et al. 2009, Phillips et al. 2009) as well a larger juveniles (Stevens et al. 2008). However, the physical characterization of benthos and mangrove morphology of sawfish habitat has not been described as a function of animal presence. The results from active telemetry indicate activity spaces are very small, therefore identifying physical characters of these microhabitats are important. If P. pectinata are selecting specific areas within the greater nursery habitat based on specific environmental and physical variables this may help illuminate uncertain aspects of juvenile P. pectinata ecology (e.g. foraging behavior). I expand on previous works of P. pectinata nursery habitats by creating variable selection models that will identify the most informative physical parameters as they relate to P. pectinata presence in a relatively pristine environment. Physical parameters considered were sediment characteristics and mangrove morphology. Theses parameters were selected based on P. pectinata’s morphology as a demersal batoid and increased observational encounters with increased proximity to mangrove-lined shores (Simpfendorfer 2006, Simpfendorfer et al. 2010) In order to accomplish these goals, I used acoustic telemetry monitoring (anchored listening stations) methods to measure spatial and temporal movements of P. pectinata within their nursery habitat. Acoustic monitoring is a useful tool in habitat studies because it allows for constant 24-hour 7-day- per-week monitoring of subjects (Heupel et al. 2005). These methods can identify areas

40 of greater use by P. pectinata as well as migration corridors and allow calculations of residency time within the nursery. Residency time within the relative protection of the nursery habitat is especially critical information necessary for successful conservation and recovery of P. pectinata. It is also important to determine if residency is related to predation risk or changes in resource availability. Once P. pectinata presence within the nursery is documented, quantitative measurements of mangrove morphology and benthic sediment composition were measured and incorporated into the habitat models. Models were examined in order to illuminate possible correlations between physical habitat characteristics and juvenile P. pectinata presence. 2.2 Methods

2.2.1 Study Areas

Acoustic monitoring was conducted at two study sites: Mud Bay and Chokoloskee Island in southern Florida. Mud Bay is located in the back water region of ENP (Fig. 1.2a). A mangrove island in the north of the bay has previously been identified as an area where juvenile P. pectinata have concentrated over several years (or a “hot spot”) (Simpfendorfer et al. 2010). A series of narrow creeks and neighboring bays connect Mud Bay to the coast on either side. This topography is ideal for monitoring possible migration corridors using acoustic monitoring. Since most of the back water area is located within a national park, it is relatively pristine. However, Chokoloskee Island in the northeastern region of the study site is anthropogenically influenced. A man-made causeway links the island to the mainland and is lined with mangroves on either side. 2.2.2 Sampling Juvenile P. pectinata were collected under guidelines approved by Protected Species Permit # SEFSC-NMFS-13330. Sampling for juvenile P. pectinata occurred during March - October 2011. Gear used for animal capture included two 30.5 m and two 61 m monofilament gillnets each with 76 and 102 mm stretched meshes. Nets were soaked (placed and oriented in the water column so has to maximize the area fished) and checked every 0.5 hour and fished in a particular area for half a tidal cycle. Abiotic factors including salinity (practical salinity scale), dissolved oxygen (mg/L), and

41 temperature (°C) were taken at the water surface at every fished location using a YSI-85 environmental meter (YSI Inc. Yellow Springs, Ohio). Captured animals were carefully removed from the gillnet and brought alongside the boat for processing. During this process the animal’s mouth and spiracles remained underwater to allow the animal to breathe normally and reduce stress. Rostral length, precaudal length, fork length, and stretch total length were measured. Sex was also recorded with males being identified by the presence of claspers. Animals less than 1.5 m fork length (the approximate sixe at one year of age) were fitted with a VEMCO (VEMCO Ltd. Halifax, Nova Scotia) V9 or V13 (depending on animal size) 1-L Global Coding tag by affixing the tag to a Primer 1 Swivel TM Ear Tag (Premier1Supplies© Washington, Iowa) using epoxy. The tag was attached through the most anterior portion of the first dorsal fin close to the body (Fig. 1.4) as thicker connective tissue in this region allowed for better tag retention. Tags emitted an acoustic signal at 69 kHz on a nominal delay of 60 seconds. 2.2.3 Acoustic Monitoring In ENP, 32 VEMCO VR2w acoustic hydrophones were anchored in sites that radiated from Mud Bay. The array was designed to both assess residency time within the study area and identify possible corridors for exiting the system (Fig. 2.1). When a tagged animal was within detection range of a receiver, the tag’s unique ID number and the time were recorded and archived. Gates or areas of overlapping VR2w (Heupel et al. 2002) detection coverage were placed at river mouths to maximize the probability that animals would be detected when leaving the system. In order to ensure the monitoring radius was maximized, the anchoring method was designed to keep the receiver up right and close to the bottom. Receivers were anchored upright to a PVC pole (using cable zip ties and duct tape) which was cemented to a concrete block. Tying the VR2w to the PVC ensured the unit remained in an ideal orientation and the concrete block kept the receiver near the bottom. All receivers were downloaded and cleared of fouling every month. The array was range tested to evaluate individual receiver performance and to validate that gates were functioning as designed. The range test was performed by towing an acoustic tag behind a kayak and recording a time-stamped location every 15 meters using a handheld GPS. Figure 2.2 shows the resulting map of the detection

42 radiuses and demonstrates that receivers acting as gates were properly placed in the back water region (Turner River, Mud Bay, Cross Bays, Wilderness Waterway, and Lopez River). 2.2.4 Habitat characterization Benthic samples were collected at every sample location throughout the study duration and in areas around receivers. These samples were sent for analysis to Terracon Industries Inc., North Carolina and analyzed for grain size analysis and percent organic content in accordance with American Society for Testing Materials (tests ASTM D2974 and ASTM D422). A P200 sieve was used to distinguish between sand and the silt/clay complex while a P270 sieve removed the silt from clay particles. Mangrove prop root density and limb overhang were measured in areas within a receiver’s detection radius. A 1.5 m PVC pole with 1.5 m of line attached to each end was used as a quadrat. The PVC pole was placed on the ground at the water line and then moved inshore until any part of the pole abutted a prop root. Then both lines were fully extended through the prop roots. Using the rope and PVC as a guide, a counter estimated the area as if there was a complete 1.5 m by 1.5 m quadrat and counted the prop roots within the area. A measuring tape was used to measure any tree limb overhang (m) by placing one end of the tape at the location of the PVC pole and then extending the tape out to the furthest outreaching tree limb. These measurements were taken twice more by moving 6 m to either side of the first measured area along the coastline. The values for all three measurements were averaged for each receiver station and used as mangrove variables in the habitat model. 2.2.5 Data Analysis If an animal ID was detected at least twice on a receiver within one 24 hour period, the individual was recorded as present that day. An animal ID that appeared only once during a day was omitted from analysis in order to eliminate the possibility of counting a false detection. Residency time was calculated as the sum of days present within any part of the array. “Days monitored” were calculated as the total number of days between the initial tagging of the animal until the last time it was detected. “Days present” indicate the number of days when at least two hits by an individual tag were detected on a receiver within a 24 hour day, and “longest consecutive days present” are

43 the maximum number of days in a row an animal was said to be present. “Days undetected” and “longest consecutive days undetected” illustrates days when the presence of the animal could not be confirmed but the animal may have been in the region. I constructed habitat models using VR2w records for each station as a dependent variable. Three different dependent variables were calculated and tested. The first calculation was to determine the probability of seeing a specific sawfish within an area (hits/fish/hour), the second calculation was to determine the probability of seeing any sawfish within an area (hits/hour), and the third was calculated to test for possible variable selection over a larger temporal scale (fish/week). The dependent variables were tested as functions of percent sediment organic, percent sediment silt, prop root count (per 2.25 m2), and mangrove limb overhang (m). Spearman’s rank correlation tests were run to assess correlations between variables before they were processed in the model. The three dependent variables were tested against the independent variables and correlation testing was also performed among the independent variables. Analyses were performed using R statistical software (R Development Core Team 2011). The habitat model was constructed using a stepwise logistic regression as a means of variable selection (Allison 2001). All possible combinations of variables were tested starting with a null model with no variables to a full model using all variables for each dependent variable. Models were run using generalized linear model (GLM) framework. Variable selection models with a p-value less than 0.05 were said to be significant. Analyses were performed using SAS® software (9.1, PROC GLM). 2.3 Results 2.3.1 Residency Time and Corridor Movements

Twenty juvenile P. pectinata were tagged and recorded by acoustic monitoring stations (Table 2.1). Presence in the array for individual smalltooth sawfish ranged from one day to 197 days and overall P. pectinata were present within the acoustic array for 334 days (Fig. 2.3). Five P. pectinata were captured in Mud Bay and the rest were captured around Chokoloskee Island. Of the five Mud Bay fish, two (ID 340 and 341) exited the backwaters area through the Lopez River system (Fig. 2.4). ID 340 left the system and never returned to the array whereas ID 341 left the back water area through

44 Lopez River and then spent several weeks in Chokoloskee Bay before its tag expired. A third tag (ID 346) followed the same exit route; however, the very rapid movement of ID 346 (Fig. 2.5) suggests this P. pectinata may have been consumed by a larger animal (possibly C. leucas). The corridor movements of ID 346 will not be reported because of these suspicions, but it is interesting to note that this corridor maybe used by potential predators of P. pectinata. The other two P. pectinata remained in Cross Bay 1, a neighboring bay to Mud Bay, until their tags expired. The subregions within the array having the longest residency times were Mud Bay, Cross Bay 1, and the bay around Chokoloskee Island (Table 2.2, Fig. 2.6). Overall, animals tagged in Chokoloskee Bay did not enter the back water region and vice versa (Fig. 2.7). Animals originally tagged in Mud Bay did emigrate from the bay in the summer months (June and July) into neighboring areas in the back water region. However, while smalltooth sawfish were detected throughout the back water region, tidal bays (as opposed to creeks or rivers) were discrete areas of greater residency (Fig 2.8). Given the design of the array, fish tagged in Mud Bay had an increased probability of being detected moving through the back water region: therefore, possible selection of certain habitat characteristics could be best determined examining the back water region. For this reason, only receivers located in the back water region (Turner River, Mud Bay, Cross Bays, Wilderness Waterway and Lopez River) were used in the subsequent habitat logistic regression model. 2.3.2 Spearman’s Rank Test Results of the Spearman’s Rank correlation tests among the independent variables and the three dependent variables showed there was low to moderate correlation between the dependent variables and any of the independent variables (Table 2.3). The strongest correlation was between the mangrove measurements prop root and overhang (rho = 0.54) and the sediment measurements silt and organic (rho = 0.57). 2.3.3 Logistic Regression Three sets of logistic regression were calculated each having a different dependent variable. For the models testing dependent variables hits per fish per hour and fish per week, no variables were significant. A non-significant trend indicated that detections of monitored fish were somewhat greater in areas with high prop root density (GLM: Standard Estimate = 0.61, Error = 0.32, t-value = 1.97, p = 0.06) (Table 2.4). A

45 subsequent step wise regression analysis of hits per hour incorporating all habitat variables indicated that sawfish had an increased probability of being encountered in areas with high prop root density (Stepwise GLM: Partial R square = 0.22, C(p) = 6.02 p = 0.023) (Table 2.5). 2.4 Discussion Acoustic monitoring was used to answer questions about long (e.g. weeks or months) temporal and spatial scales of habitat use. Residency time within the nursery was longer than previously reported. Simpfendorfer et al. (2010) had three receiver stations anchored in Mud Bay (2003 – 2007) and found that smalltooth sawfish left the area in late summer (average days present was 16.8 days). Our findings also show exiting of Mud Bay in the summer, but data from the larger array suggest P. pectinata reside for months afterwards in the surrounding system. Near Chokoloskee Island, an area previously unstudied, there is evidence of overwintering when water temperatures remain habitable (28° C). Interestingly, P. pectinata that were tagged around Chokoloskee Island never entered the back waters into Mud Bay, and only one animal tagged in Mud Bay entered Chokoloskee Bay. Animals originally tagged in Mud Bay were never recorded on the west side of Chokoloskee Island. Additional monitoring is necessary to see if movement patterns are repeated across years, but if consistencies in habitat use (e.g. overwintering) are observed then Chokoloskee Bay may have to be managed as nursery year-round. The acoustic array also demonstrated that while P. pectinata will use most of the available habitat, there are areas where P. pectinata will remain for weeks and other areas that act more like corridors. Specifically, animals tagged in Mud Bay emigrated during summer months (June and July) into the neighboring back water area. There was greater use of tidal habitats rather than creeks or rivers. Longer detection times were recorded in tidal bays despite having a decrease in detection coverage as opposed to receivers along narrow creeks or rivers where coverage was greater. The design of the array to capture animals leaving the back water areas from Mud Bay allowed for construction of a habitat model using multiple logistic regression (GLM). The three dependent variables measured were fish per week, hits per fish per hour, and hits per hour. These served to test separate possible correlation between habitat

46 use and physical habitat properties. The results indicated that fish were more likely to be detected in areas with high prop root densities (number of prop roots/2.5m2). Previous studies of smalltooth sawfish presence have reported a correlation between habitat use and proximity to mangroves (Simpfendorfer 2006). Perhaps, areas with high mangrove prop root density may be particularly important to preserve as potential juvenile P. pectinata habitat. 2.5 Conclusions In this study, I calculated residency times for juvenile P. pectinata in a nursery, identified migration corridors, and determined if mangrove or sediment characteristics were important variables in sawfish habitat use. Residency time was greater than previously reported with overwintering occurring in Chokoloskee Bay. Also, while animals tagged in Mud Bay did leave the area in summer, three out of five animals originally tagged stayed within the system until their tag batteries expired. These individuals may have remained in the system for longer than calculated. Chokoloskee Bay and the neighboring back water region have potential to serve as year-round nursery habitats for juvenile smalltooth sawfish, and conservation managers should consider this finding when implementing conservation strategy plans. A possible migration corridor through the Lopez River system was documented; however, this pattern would have to be repeated over several years to make a definite statement about corridor use. For fish in the back water region, movement between areas was informative. While detections were recorded throughout the system, there were discrete areas (specifically tidal bays) where individual sawfish remained for a longer time than others. Tidal bays probably allow juvenile P. pectinata to move within certain depth strata regardless of tidal flow direction and tidal amplitude. Tidal bays also may provide a refuge from high currents that are associated with narrow and deep channels like rivers or creeks. The habitat model suggests that within the back water region time spent by sawfish in a particular area correlated with mangrove prop root density (number of prop roots/2.5m2). The higher the mangrove prop root density the higher the probability of encountering a juvenile P. pectinata. These results are valuable because they are the first indication of possible selection of habitat by juvenile P. pectinata. Since young-of-the-year sawfish have small

47 daily activity spaces (Simpfendorfer et al. 2010) they are potentially restricted in the first months of their lives to the area in which they were born. However, after a few months of experiencing rapid growth young-of-the-year sawfish may require different habitat characteristics than those of the area in which they were pupped, and the variables that cause them to emigrate from the bays where they were born remains unknown. Future studies of juvenile P. pectinata habitat use in a nursery should be conducted with the goal of identifying and testing potential variables in an effort to better delineate areas of essential fish habitat.

Figure 2.1: Locations of anchored VR2w receivers within ENP. Red balloons indicated receiver stations where Onset® HOBO Temperature/Light data loggers were attached

Figure 2.2: Map illustrating detection radius of VR2w receivers. Array set up was designed to maximize chance of recording an animal leaving the backwaters area as well as monitoring residency time.

50 Table 2.1: List of P. pectinata acoustically tagged in 2011.

Fish ID Date Captured Subregion STL (cm) Sex

338 4/25 Mud Bay 85.0 F

339 4/25 Mud Bay 87.0 F

340 4/25 Mud Bay 86.5 F

341 4/27 Mud Bay 89.0 M

342 5/23 Chokoloskee Bay 95.0 F

343 5/23 Chokoloskee Bay 90.0 F

346 5/26 Mud Bay 93.0 F

344 6/18 Chokoloskee Island 88.0 F

345 6/18 Chokoloskee Island 95.5 M

347 7/26 Chokoloskee Island 98.5 F

1748 8/25 Chokoloskee Island 120.0 M

1749 8/25 Chokoloskee Island 120.0 F

1750 8/25 Chokoloskee Island 127.0 F

1751 8/25 Chokoloskee Island 113.0 F

1752 8/25 Chokoloskee Island 107.0 M

1753 8/25 Chokoloskee Island 118.0 F

3308 9/21 Chokoloskee Bay 128.0 F

3307 10/22 Chokoloskee Island 83.0 M

3309 10/22 Chokoloskee Island 85.0 F

3310 10/22 Chokoloskee Island 85.0 M

Figure 2.3: Abacus plot showing residency times for all P. pectinata tagged in 2011 in ENP. Black colored plots indicate that an animal was present within the back water region (between the Turner and Lopez Rivers). Red colored plots indicate that an animal was present on the west side of Chokoloskee Island or Chokoloskee Bay. Stars indicate the date the animal was acoustically tagged. Arrows ilustrate probable tag battery expatriation and these animals were no longer detectable in the array.

A. B. Figure 2.4: Abacus plots illustrating an emigration corridor though the Lopez River system. (A) Abacus plot for P. pectinata ID 340. (B) Abacus plot for P. pectinata ID 341. Stations names on the Y axis are arranged geographically. Stations to the south of Mud Bay (Station name = MB) are lower on the axis and stations to the north of Mud Bay are higher on the axis. Both plots illustrate movement out of the back water region via the Lopez River (LR). ID 341 was then detected for several weeks in Chokoloskee Bay (CHOK) after leaving the back water region.

Figure 2.5: Abacus plot illustrating ID 346 moving south from Mud Bay (MB) and exiting out the Lopez River system. Rapid movement suggests the animal being monitored was not a P. pectinata and possible a C. leucas.

54 Table 2.2: Residency time by sub region. Sub regions with the highest days present are Mud Bay, Cross Bay 1 (CB1), and the bay around Chokoloskee Island (CISL and CHOK Bay).

Days Days Longest Conseq. Long Conseq. Days Sub Region ID Monitored Present Present Undetected Undetected MB 338 23 23 23 0 0 MB 339 33 33 33 0 0 MB 340 20 20 20 0 0 MB 341 67 66 33 1 1 MB 346 2 2 2 0 0 CB1 338 128 73 28 33 65 CB1 339 123 52 21 46 71 CB1 340 3 3 3 0 0 CB1 341 2 2 2 0 0 CB2 338 2 2 2 0 0 CB2 339 2 2 2 0 0 CB2 340 2 2 2 0 0 CB2 341 3 2 2 1 1 CB3 339 2 2 2 0 0 CB3 340 1 1 1 0 0 CB3 341 2 2 2 0 0 LR 340 1 1 1 0 0 LR 341 3 3 3 0 0 WW 338 1 1 1 0 0 WW 341 3 2 2 1 1 TR 338 1 1 1 0 0 TR 339 2 2 2 0 0 TR 340 2 2 2 0 0 TR 342 6 6 6 0 0 CHOK BAY 340 1 1 1 0 0 CHOK BAY 341 72 25 14 27 47 CHOK BAY 342 110 8 5 88 102 CHOK BAY 343 1 1 1 0 0 CHOK BAY 344 197 10 4 135 187 CHOK BAY 3308 1 1 1 0 0 CHOK BAY 1753 58 28 7 6 30 CISL 344 1 1 1 0 0 CISL 345 14 14 14 0 0 CISL 347 23 23 23 0 0 CISL 1748 115 20 19 95 95 CISL 1749 62 57 44 3 5 CISL 1750 26 24 17 2 2 CISL 1751 81 42 26 31 39 CISL 1752 74 57 18 8 17 CISL 3307 89 36 11 23 53 CISL 3309 102 84 29 6 18 CISL 3310 133 118 56 5 18

55 A.

Figure 2.6: Residency time within a subregion. (A) A map of the array broken down into 7 sub regions. (B) A plot of days present for each tagged P. pectinata showing residency time within each of the subregions.

Figure 2.7: A stacked graph illustrating each tagged sawfish days present and movement between subregions. Days present within in a subregion with individual fish on the X axis (Mud Bay fish are the first five Fish IDs). Given the layout of the array, Mud Bay fish were more likely to be detected moving through the back water region (Turner River, Cross Bays, Wilderness Waterways, Lopez River). Overall, animals tagged near Chokoloskee Island did not move into the back water region, and vise versa.

Figure 2.8: Abacus plot illustrating greater presence in tidal bays among the back water region habitats (tidal bay, river, and creek). Fish tagged in the open bay system (Chokoloskee Island and Bay) could not be detected in other areas, but fish emigrating out of Mud Bay could. For this reason, back water area receivers (located in Turner River, Mud Bay, Cross Bays, Wilderness Waterway, and Lopez River) were used to construct the habitat model.

58 Table 2.3: A matrix of Rho values resulting from the Spearman’s Rank correlation tests. Tests were conducted between all independent values and the three dependent values: fish per week, hits per hour, hits per fish per hour. Prop root density and mangrove limb overhang were moderately correlated (rho = 0.57, bold red) as well as percent organic sediment and percent silt sediment (rho = 0.54, bold red).

59 Table 2.4: Results of three sets of logistic regressions (GLM). Dependent variables fish per week, hits per fish per week, and hits per hour we all tested as a function of four habitat variables (density of mangrove prop roots, mangrove limb overhang, percent silt sediment and percent organic sediment). No variables resulted as significant for the dependent variables fish per week or hits per fish per hour. The model testing the dependent variable hit per hour resulted in a nearly significant p value for mangrove prop root density (bold red).

60 Table 2.5: Result summary of a stepwise regression (GLM). Hits per hour (dependent variable) was tested as a function of four habitat variables (mangrove prop root density, mangrove limb overhang, percent organic sediment, and percent silt sediment). The only variable to result as significant (p = 0.023) was mangrove prop root density. This variable accounted for 22% of the explained variance.

61 APPENDIX

FLORIDA STATE UNIVERSITY ANIMAL CARE AND USE COMMITTEE PROTOCOL REVIEW FORM

20 Submissio January Principal n Date: 2011 Investigator: Christopher C. Koenig [email protected] or E-Mail Address: [email protected] Telephone: Campus FSU Coastal and Marine Address: Laboratory, Bldg 407 Mail Code: 1259 ACUC Protocol 110 Department: FSUCML, Office of Research #: 6 Habitat and demographics of fishes of the Proposal Title: Gulf of Mexico and the South Atlantic Bight.

______THIS IS A NEW PROTOCOL.

___X___ THIS PROTOCOL IS SUBMITTED AS A TRI-ANNUAL RENEWAL, formerly #9902

The FSU ACUC acknowledges that replication of previous research may be an essential element in scientific inquiry. In planning this experiment, I have reviewed the relevant literature (e.g., database search, consultation with colleagues, other). Based upon the available resources, I certify that the work described in this protocol does not unnecessarily duplicate previous work.

P.I. Signature: Date: 20 January 2011

62 Your signature acknowledges your responsibility for the contents of this questionnaire and the conduct of any animal use that may be approved by the Committee, by yourself or your staff. Any significant change in procedure, animal use or personnel must be approved by the ACUC prior to implementation. All animal procedures proposed in grant submissions must be included in an ACUC Animal Care and Use Protocol and be reviewed and approved by the ACUC. Failure to do so may result in suspension of approval to perform animal research at FSU as well as suspension of funding.

This Questionnaire must contain all animal use information necessary for ACUC approval. Please exclude any confidential information from this questionnaire (e.g. proprietary information, potential trade secrets, and patentable material) as the document is considered a public record and is available to outside parties by request under the Freedom of Information Act. If confidential material must be included, please mark it "Confidential" and contact the secretary, ACUC, for further instructions.

ACUC DMR March 18, 2011 Next Review Date: February

Approved: 1, 2014

Protocol March 18, 2014

Expiration:

1. List the names of all individuals that will contact animals under this animal use description (include DIS and Honors Students):

Name Phone E-Mail Address Status (Grad, Lab Emergency Number Undergrad, Emergency contact Post-Doc) Contact* priority (1, (Y/N) 2, 3, 4 … ) CC Koenig [email protected] FACULTY Y 2 FC [email protected] FACULTY Y 1 Coleman Robert GRAD N Ellis David [email protected] FACULTY Y 5 Kimbro Kelly [email protected] GRAD N

63 Kingon Kevin [email protected] FACULTY Y 4 Craig Dean [email protected] FACULTY Y 3 Grubbs Chris [email protected] POSTDOC Y 8 Stallings Alejandra TECHNICIAN N Mickle Lisa [email protected] GRAD N Hollensead Matthew [email protected] GRAD Y 9 Kolmann Cheston [email protected] GRAD N Peterson Anne [email protected] FACULTY Y 7 Randall Hughes Hanna TECHNICIAN N Garland Tanya TECHNICIAN N Rogers Evan TECHNICIAN N Pettis Emily Field [email protected] TECHNICIAN N Robyn TECHNICIAN N Zerebecki Althea GRAD N Moore *If Yes, Emergency Lab Contact, also indicate priority in which persons should be contacted (1,2,3…etc.)

Basic Species Hazardous Medical Name Anesthesia Allergens training Specific Agents Monitoring CC Koenig 10/14/1994 Grandfathered Grandfathered 03/14/2011 09/10/2008 02/02/2009 FC 11/16/1994 Grandfathered Grandfathered 03/14/2011 02/22/2010 01/09/2009 Coleman

64 Robert Ellis 08/13/2010 08/13/2010 08/13/2010 08/13/2010 08/13/2010 08/27/2010

David 09/13/2010 09/13/2010 09/13/2010 09/13/2010 09/13/2010 09/20/2010 Kimbro Kelly Kingon 03/20/2007 03/03/2011 03/03/2011 03/20/2007 09/22/2010 03/20/2009

Kevin Craig 02/18/2008 02/18/2008 02/18/2008 02/18/2008 02/18/2008 02/23/2011

Dean 02/18/2008 02/18/2008 02/18/2008 02/18/2008 02/10/2010 02/23/2011 Grubbs Chris 02/18/2008 02/18/2008 02/18/2008 02/18/2008 08/12/2008 03/11/2011 Stallings Alejandra 03/01/2011 03/01/2011 03/01/2011 03/01/2011 03/01/2011 03/04/2011 Mickle Lisa 03/16/2010 03/16/2010 03/16/2010 03/16/2010 02/28/2011 03/12/2010 Hollensead Matthew 08/13/2010 08/13/2010 08/13/2010 08/13/2010 08/23/2010 08/05/2010 Kolmann Cheston 02/18/2011 02/18/2011 02/18/2011 02/18/2011 08/1/2010 02/18/2011 Peterson Anne Randall 02/01/2011 02/01/2011 02/01/2011 02/01/2011 01/25/2011 02/02/2011 Hughes Hanna 2/04/2011 2/04/2011 2/04/2011 2/04/2011 02/01/2011 02/04/2011 Garland Tanya 2/04/2011 2/04/2011 2/04/2011 2/04/2011 09/22/2010 02/04/2011 Rogers

Evan Pettis 01/24/2011 01/24/2011 01/24/2011 01/24/2011 01/24/2011 01/24/2011

Emily Field 02/02/2011 02/02/2011 02/02/2011 02/02/2011 10/06/2010 03/02/2011 Robyn 02/16/2011 02/16/2011 02/16/2011 02/16/2011 09/22/2010 02/24/2011 Zerebecki Althea 02/18/2011 02/18/2011 02/18/2011 02/18/2011 08/18/2010 02/21/2011 Moore

NOTE 1: It is the responsibility of the Principal Investigator to ensure that all personnel who have animal contact be enrolled in the FSU Medical Monitoring Program for Vertebrate Animal Users before they begin working with animals. This requirement must be addressed or the ACUC will not approve the protocol. Enrolling in the program can be accomplished by filling out the Medical Monitoring for Vertebrate Animal Users Form (EHS 7-2) (http://www.safety.fsu.edu/forms.html) or contacting either Laboratory Animal Resources (644- 4262) or Environmental Health and Safety (644-9117) for assistance. NOTE 2: All Personnel must have completed the ACUC required training prior to beginning work with animals. This requirement must be addressed or the ACUC may either not approve the

65 protocol or may suspend approval of the protocol. Please contact Laboratory Animal Resources, 101 BRF, or phone at 644-4262 for assistance in scheduling training.

2. PURPOSE OF RESEARCH (NOTE: This statement is used for public inquiries. Using non-technical language, briefly address the points below. Scientific abstracts are not acceptable. Define all abbreviations the first time they are used.):

ñ Objective: The primary objectives of this work are: 1. to research the basic biology (ecology, evolution, anatomy, physiology, life history, habitat use, sensory etc.) of economically and ecologically important fishes of the southeastern US so that appropriate conservation measures may be taken. This includes capture of both juvenile and adult fish. 2. to instruct students in the biology of fishes at the FSU Coastal and Marine Laboratory (FSUCML). 3. to understand how predatory fishes and oyster reefs reciprocally benefit each other in Florida estuaries

ñ Background: Many species of marine and estuarine fishes are over- fished and/or threatened by loss of critical habitat. The work we propose is to evaluate critical habitat associations, ecological roles, basic life history parameters, and habitat loss. This will involve capture of juvenile fish (which are sensitive to capture) and adult fish (which typically are relatively insensitive to capture when returned rapidly to the water). Based on this information, to inform state and federal management agencies of practicable approaches necessary for the recovery of these fish populations to sustainable levels.

ñ Relevance: This work is important because: 1. Most of the economically important fish species of the southeastern US are considered by the National Marine Fisheries Service (NMFS) and The Florida Fish and Wildlife Commission (FWC) to be undergoing over-fishing, already over-fished, or threatened. Our research will provide the necessary detail in life history and habitat to effectively manage fisheries for these species. 2. In our course, Biology of Fishes, taught at the FSUCML we will use fish specimens to demonstrate characteristics of the basic biology so that future generations will have the knowledge to continue to conserve these valuable resources. 3. Oyster reefs historically helped maintain estuarine health by filtering enormous volumes of water, cycling nutrients, and increasing

66 biodiversity. Because this critical habitat has been globally degraded to only ~15% of its historical abundance, considerable effort has been devoted towards restoring oyster reefs. These restoration efforts, however, have failed to consistently improve diversity and ecosystem functioning. By linking the effects of predatory fishes with the survivorship and filtration behavior of oysters and with processes in the water column and the sediments on oyster reefs in a biogeographic context (collaborative project with researchers from NC, SC, and GA), our research will provide valuable information to resource managers charged with restoring oyster reefs and recovering ecosystem services.

ñ Species: Our objectives are to investigate the basic biology (ecology, evolution, physiology, anatomy, behavior, etc) of fishes that are economically important, poorly researched, or threatened to inform management agencies of our findings, and to teach this information to students. Information derived from this research is absolutely necessary to the continued survival of these species and to our continued ability to sustainably exploit their populations for the purposes of food and recreation. We humans have the technology and capacity to obliterate targeted fish populations. We have done this primarily through our ignorance of the basic biological needs of exploited fishes which translates to poor management of both fisheries and critical habitat. Over-fishing has occurred and is occurring in many other parts of the world where management is mis-informed, minimal, or non-existent. Here in the southeastern US, we do not want to perpetuate this pattern of over-fishing and habitat loss, so we support effective management. However, for management to be effective we must have accurate information about the biology of these species, the impacts of exploitation, and the loss of critical habitat. This is the information we provide to management and pass on to our students. The main groups of economically important species we study are: reef fishes (several species of groupers and snappers), estuarine fishes (ecologically important species such as catfish, drum, pinfish, pigfish, some sharks, garfish, and the juvenile stages of several reef fish), elasmobranch fishes (coastal sharks and endangered smalltooth sawfish) and pelagic fishes (amberjack, and several species of herring and scad).

Note: Our investigations are primarily non-destructive. We generally biopsy, tag, and release the fish we capture. When doing biopsies and tagging we always use triple antibiotic salve to minimize the chances of infection. Live specimens are occasionally brought back to the lab for short-term research or education purposes, where they are kept in

67 conditions similar to their natural habitats, before being released from the location captured. In the Koenig-Coleman research and Grubbs research, the only fish economically important species that die are those caught for food and recreation by cooperating commercial and for-hire fishermen who fish according to state and federal regulations on their boats and on their regular fishing trips. When we accompany them on these trips we only sample dead fish which were caught and killed by them. Some mortality is unavoidable during standardized fishery-independent gillnet and longline sampling. Mortality rates vary between gear types and species captured. Overall mortality rates are approximately 10% for sharks, 0% for batoids, and <5% for bony fishes. The fishes that die are sampled for gut contents and tissue samples taken for isotopic analyses. For the oyster reef research, sampling will be of short duration and great care will taken to reach the sampling gear prior to the onset of ebb tide conditions (see methods below). As a result, we are able to maintain sampling mortality below 10% of the catch. In the summer, we expect one sampling event to generate forty fish with less than 4 mortalities.

3. ANIMAL USAGE: The ACUC can approve protocols for 3 years. The number of animals asked for should be for that 3-year period. List the total number of animals of each species to be used for the proposed project. Of the total number of animals to be used, estimate the number expected to fall in the USDA pain categories B through E (categories explained below).

TOTAL # SPECIES OF B C D E (Common and Scientific Name) ANIMALS TO BE USED

Reef fishes and seagrass-associated fishes

Goliath grouper Epinephelus itajara 200 200

Red grouper Epinephelus morio 200 200 Scamp Mycteroperca phenax 100 100

68 Gag Mycteroperca microlepis 200 200

Red snapper Lutjanus campechanus 200 200

Gray snapper Lutjanus griseus 100 100

Greater amberjack Seriola dumerili 50 50

Alligator gar Atractosteus spatula 25 25

Pinfish Lagodon rhomboides 200 200

Pigfish Orthopristis chrysoptera 100 100

Black sea bass Centropristis striatus 50 50

Gulf flounder Paralichthys albigutta 10 10

Vermilion snapper Rhomboplites aurorubens 200 200

Hardhead catfish Ariopsis felis 200 200

Red drum Sciaenops ocellatus 50 50

Toadfish Opsanus tau 100 100

Sharks and Rays

Atlantic sharpnose Rhizoprionodon terraenovae 400 400

Bonnethead shark Sphyrna tiburo 400 400

Blacktip shark Carcharhinus limbatus 200 200

Blacknose shark Carcharhinus acronotus 100 100

Spinner shark Carcharhinus brevipinna 25 25

Bull shark Carcharhinus leucas 25 25

Tiger shark Galeocerdo cuvier 25 25

Lemon shark Negaprion brevirostris 10 10

Nurse shark Ginglymostoma cirratum 10 10

Great hammerhead shark Sphyrna mokarron 10 10

Finetooth shark Carcharhinus isodon 10 10

Cownose ray Rhinoptera bonasus 25 25

Atlantic stingray Dasyatis sabina 25 25

Southern stingray Dasyatis americana 25 25

69 Smalltooth sawfish Pristis pectinata 25 25

Lesser electric ray Narcine bancrofti 25 25

Bony fishes captured in fishery-independent sampling

Orange filefish Aluterus schoepfi 10 10

Scrawled filefish Aluterus scriptus 10 10

Sheepshead Archosargus probatocephalus 10 10

Hardhead catfish Arius felis 200 200

Gaftopsail catfish Bagre marinus 200 200

Silver perch Bairdiella chrysoura 25 25

Gulf menhaden Brevoortia patronus 500 500

Yellowfin menhaden Brevoortia smithi 500 500

Grass porgy Calamus arctifrons 25 25

Blur runner Caranx chrysos 10 10

Crevalle jack Caranx hippos 10 10

Black seabass Centropristis striata 25 25

Spadefish Chaetodipterus faber 10 10

Striper burrfish Chilomycterus schoepfi 10 10

Atlantic bumper Chloroscombrus chrysurus 25 25

Spotter seatrout Cynoscion nebulosus 10 10

Sand seatrout Cynoscion arenarius 10 10

Sand perch Diplectrum formosum 10 10

Spottail pinfish Diplodus holbrooki 10 10

Sharksucker Echeneis naucrates 100 100

Ladyfish Elops saurus 25 25

White grunt Haemulon plumieri 10 10

Scrawled cowfish Lactophrys quadricornis 10 10

Pinfish Lagodon rhomboides 270 270

Spot Leiostomus xanthurus 10 10

70 Longnose gar Lepisosteus osseus 25 25

Mangrove snapper Lutjanus griseus 10 10

Gulf kingfish Menticirrhus littoralis 10 10

Atlantic croaker Micropogonias undulatus 10 10

Striped mullet Mugil cephalus 10 10

Gag Mycteroperca microlepis 10 10

Leatherjacket Oligoplites saurus 10 10

Pigfish Orthopristis chrysoptera 25 25

Gulf flounder Paralichthys albigutta 10 10

Gulf butterfish Perprilus burti 25 25

Harvestfish Peprilus paru 25 25

Black drum Pogonias cromis 10 10

Bluefish Pomatomus saltatrix 50 50

Searobin Prionotus sp. 10 10

Cobia Rachycentron canadum 10 10

Red drum Sciaenops ocellatus 10 10

Spanish mackerel Scomberomorus maculatus 200 200

Bigeye Scad Selar crumenophthalmus 10 10

Lookdown Selene vomer 10 10

Moonfish Selene setapinnis 10 10

Inshore lizardfish Synodus foetens 10 10

Florida pompano Trachinotus carolinus 10 10

Houndfish Tylosurus crocodilus 10 10

Category B: # of animals being bred, conditioned, or held for use in teaching, testing, experiments, research, or surgery but not yet used for such purposes. For most protocols this total will be 0 for the entire project period. Category C: # of animals upon which teaching, research, experiments, or tests will be conducted involving no pain, distress, or use of pain-relieving drugs.

71 Category D: # of animals upon which experiments, teaching, research, surgery, or tests will be conducted involving accompanying pain or distress to the animals for which appropriate anesthetic, analgesic, or tranquilizing drugs will be used. Category E: # of animals upon which teaching, experiments, research, surgery or tests will be conducted involving accompanying pain or distress to the animals and for which the use of appropriate anesthetic, analgesic, or tranquilizing drugs will adversely affect the procedures, results, or interpretation of the teaching, research, experiments, surgery, or tests. The form Explanation for Category E, form must be submitted with the AUD Protocol. (An explanation of the procedures producing pain or distress in these animals and the justification for not using appropriate analgesic, anesthetic or tranquilizing drugs must also be addressed in question 15.)

4. Justify the number of animals to be used. Briefly describe how the estimated number of animals needed for the experiments was determined. The number of animals used is based on the study objectives. With the type of studies we are doing, it is usually difficult to get enough fish to do the work. The numbers that are indicated beside each species indicate how many we would like to get to accomplish our objectives (based on our past experience), but we will likely fall short of those numbers. For example, we did a study in Madison Swanson Marine Protected Area in 2003 to 2005 and found that the numbers proposed for gag, red grouper, red snapper, scamp, and amberjack are the likely number we will catch, tag and release alive for the present project in the same reserve. It is important to remember that these species are all heavily fished, caught by both commercial and recreational fisheries. Our work will decrease mortality, pain, and suffering in these species through our research, not increase it.

i. For the work in the Madison Swanson Marine Protected Area in the northeastern Gulf of Mexico we would like to capture, tag, and release as many fish species as possible to determine movement patterns relative to the boundaries of the reserve. The reserves are over 100 square miles in area and were designed to protect these fish, their spawning, and their spawning habitat. The more specimens we can tag and monitor, the better the understanding we will have of home range and therefore the effectiveness of these marine protected areas.

ii. In studies of vastly over-fished and/or threatened fish such as the alligator gar, smalltooth sawfish, and goliath grouper, we are very careful not to

72 injure captured individuals and we use the utmost caution not to cause distress or pain in these fish. In the case of goliath grouper we have submitted an alternative proposal to that of NMFS and FWC to counter their proposal to kill 800 of this recovering species (classified by the World Conservation Union [IUCN] as critically endangered). We will use our non- destructive methods to provide them with the necessary information for stock assessment. However, again, we must sample a sizable number to get accurate estimates of movement patterns, reproductive conditions, survival rates, growth rates, and diet. Such information will be used in mathematical models developed by NMFS and FWC to allow effective management of recovering fish populations.

iii. In studies of juvenile fishes throughout coastal waters, we typically capture only a dozen or fewer fish per site. The number of fish captured will be low, probably less than 100, so the 200 limit for gag grouper will be sufficient

iv. For the research on sensory mechanisms of coastal fishes, we have a target sample size of 10 (i.e., experimental rounds) for each of the seven experimental treatments. We will use live pinfish in only a single experimental treatment, in a single experimental round (i.e., each trial will involve a different individual). We therefore need 70 live pinfish (7 treatments * 10 rounds) that will be temporarily held and subsequently released and 200 pinfish (5 per olfactory treatment * 4 olfactory treatments per round * 10 rounds) that will be sacrificed. We will use a single flounder for each round of the six treatments involving it, so will need 10 (1 * 10).

v. For monitoring research conducted on oyster reefs, we deploy very small traps and nets for a very short duration and only 3 times per year. The goal of this research is not to collect or harvest a specific number of fishes. Rather it is to observe how many fishes utilize a particular reef habitat and how this usage changes through time and condition of the habitat.

5. DESCRIPTION OF ANIMAL EXPERIMENTAL PROCEDURES: (Briefly explain the experimental design. The description should permit the ACUC to understand the experimental progress of an animal from birth or arrival at the facility to the endpoint of the study and animal disposition. Diagrams, flowcharts, tables or timelines are preferable. These should indicate test and control groups, the number of animals in each group, the sequence of the experimental manipulations in each group and the time between each manipulation.

NOTE: Examples are available. Experimental details should be addressed in answer to questions 6-13, not in this section. Please provide details of all other work not covered in questions 6-13 in answer to question #14.

Capturing Fish at Depth The first and most difficult problem we solved in past studies was how to catch a fish at depths greater than about 50 ft and have them survive. For capture we decided on chevron traps (dimensions: 6 ft x 4.5 ft x 2 ft) with a plastic coated wire mesh of 1.5 x 1.5 inches. These traps are baited with commercial mackerel bait and are allowed to remain on the bottom for several hours. Traps are attached to surface floats by a 3/8 inch polypropylene line. Trapped fish are retrieved after several hours and held at a pre-determined mid depth where they are vented by divers (see below).

Fish are also captured by hook-and-line (defined in the Magnuson-Stevens Fisheries Conservation and Management Act 50 C.F.R. Wildlife and Fisheries§ 600.10 as one or more hooks attached to one or more lines (can include a troll). Fish are captured using large circle hooks, 600 lb test monofilament leader, a 2 kg lead weight, and cut or live bait. The gear is attached to a 1.0 cm diameter braided nylon handline and suspended above the bottom by a 60 cm diam. float. The fish are allowed to fight the float until exhaustion (about 3 to 5 minutes) before being hauled to the surface. We vent captured fish (see procedural details below) either at the surface if caught at depths < 25 m, or in situ at about 10 m if caught at depths > 25 m. Venting protocol We have found that if the depth of capture for fish with a closed gas bladder (all fish listed except for sharks and garfish) is greater than 50 ft, the captured fish should be vented at a depth where the volume of gas bladder gas is double that on the bottom. (Note that most reef fish have closed gas bladders, that is, an internal bladder containing gas with no connection to the outside of the fish, which they use to regulate their buoyancy at various depths. With this type of gas bladder, the fish must re-dissolve the gas in the blood to reduce the amount of gas in the gas bladder; this is a slow process. All else being equal, the volume of a gas increases as the pressure decreases (Boyles Law), so a closed gas bladder becomes a liability to a fish if it is being hauled to the surface at the end of a line or in a trap because the gas expands and if the degree of expansion is too great, the bladder tears and the fish will likely hemorrhage and die.) We mitigate the hooked (or trapped) fish’s liability by venting the fish and providing an escape path for the gas. This is done by puncturing the body wall while the fish is at a depth that is not injurious (a depth where the gas bladder gas has expanded to only 2 to 2.5 x its initial volume.) If fish are not vented, they will die with a high degree of certainty, especially at capture depths greater than 200 ft. (We determined the 50% mortality depth of capture to be about 140 ft.) After venting, captured fish will be hauled to the surface either in the trap or on the line. The diver will ascend to the surface prior to raising the fish either in the trap or on the line.

74 Venting is accomplished using a 2 cm long point that is 0.5 cm in diameter attached to the end of a pole spear. The depth of penetration of the point is controlled by a washer-shaped stop that allows the point to only penetrate the body wall and gas bladder wall, but not impact the other internal organs. The body wall is punctured at just below the midline at a point just behind the pectoral fin (location of the gas bladder.) Puncturing of the body wall of the fish provides an escape route for the gas so that further gas expansion and hemorrhage does not occur as the fish is raised to the surface. Fish are vented at a depth of water that is based on the depth of capture so that the gas is allowed to expand to only 2 to 2.5 times the volume of that on the bottom (e.g., a fish trapped at 300 ft would be vented at about 110 ft.) Gas expansion to 2 to 2.5 times that on the bottom is equivalent to catching a fish in 30 to 45 ft of seawater. Survival of released fish is very high at this depth of capture based on our NOAA-funded studies of the past.

When fish reach the surface, they are guided onto a stretcher that is then hoisted above the gunwale with two davits. The captured fish’s gills are bathed with seawater by a hose attached to an overboard submersible 12 VDC pump; the eyes are covered to protect them from direct sunlight while the fish is held in place with velcro straps for sample collection (Figure 1).

Tissue sampling Once on board, the fish is held in a flowing seawater bath, then transferred to a cloth-lined cradle and its eyes are protected from direct sunlight. Once in the cradle, the gills are irrigated with a hose that supplies seawater of high volume and low pressure (this maintains normal respiration in the fish). We first measure the fish, then we clip a spine and a ray (Goliath grouper fin rays are about 10-20 grams each, depending on the size of the fish. Fin ray length is 3-4 inches and width is 1/4-1/2 inch.) from the dorsal fin (both are used to age the fish; annual rings are laid down concentrically, like the rings of a tree with each ring equaling a year’s growth). Spines and rays grow back within a year.

There is no overt reaction by the fish to this clipping of fins, and they do grow back in a relatively short period of time, so we can safely assume that the procedure does not cause pain or at least not more than momentary discomfort, like fingernails being clipped on a human.

Gonad biopsy

75 We then take a gonad biopsy by inserting a catheter into the gonoduct and drawing out some gonad tissue by using a partial vacuum. As an added precaution against infection, the plastic tube used for gonad biopsy is coated with triple antibiotic just prior to insertion into the gonoduct. This gonad tissue is then preserved in 10% formalin, labeled and stored for later histological examination. These tissues will tell us the sex and the reproductive condition of the fish.

The gonad biopsy procedure is apparently not painful because the fish does not flinch or show overt signs of pain. We try very hard not to cause pain in 300 to 500 lb goliath grouper (see image above), or any other species for that matter, because they are solid muscle and can hurt us (or themselves) if they start thrashing about in response to some painful procedure. Additionally, innervation of higher teleost (bony fish) gonads is in the smooth muscle surrounding the gonad tissue (muscle functions in expelling sperm and eggs), not in the oogenic or spermatogenic tissue (Uematsu 1986), which is what we sample via the gonoduct (i.e. no incision is made and no needles penetrate the body wall). So without nerves in the gonogenic tissue (the tissue we sample), and without overt signs of pain, it is highly unlikely that the fish feel pain while their ovaries/testes are being biopsied.

Tagging fish The last procedure is to tag the fish with a dart tag, which is a 1/8 inch diameter plastic string about 6 inches in length with a barbed nylon head that is inserted in the dorsal aspect of the fish. We always use triple antibiotic salve on the tag before inserting it so as to avoid infection. On some fish we insert a transmitter that identifies the fish and emits that identification to an in situ receiver placed at the capture site. We implant the receiver by making a small incision in the body wall of the fish, then pushing the 2 x ½ inch long cylindrical transmitter into the body cavity. Antibiotic salve is then applied to the wound and the incision (and any other wounds) is closed by suturing or by the use of titanium surgical staples. Fish are then released on the site where they were captured. No anesthetics are used and no harsh chemicals are applied to these fish. (Note, it is unlawful to use systemic chemicals such as anesthetics on fish that could be caught again and used for human food, but we are very gentle and fish do not behave as if they experience pain when making or closing the incision). Our objective is to impact them the least amount as possible so that their behavior subsequent to release is as normal as possible.

Collection of fertilized grouper eggs Verification of spawning sites is made by collecting fertilized goliath grouper eggs (as we have done in the past with passive downstream plankton nets and actively towed nets, (Koenig and Coleman 2009). Collected eggs will be allowed to develop to at least the neurula stage and then are preserved in 95% EtOH for

76 later genetic confirmation of identity. We will also estimate relative egg production and percent fertilization. Exact spawning times will be estimated by back-calculating from known embryonic developmental rates at specific temperatures. Estimated spawning times will be compared with sound production and the number of spawners present to evaluate a possible relationship.

Passive acoustics Passive acoustics have also been demonstrated to be a noninvasive technique which can provide important reproductive data on soniferous species (Walters 2005, Rountree et al. 2006). Goliath grouper emit low frequency, high intensity sounds correlated with reproductive timing on a seasonal scale (Mann et al. 2009). Passive acoustics can therefore be used to evaluate the temporal and spatial distribution of spawning activity.

Permanent recording devices (digital spectrum recorders, DSG) have allowed us to monitor soniferous aggregations of goliath grouper continuously for prolonged time periods. The DSG will be custom built by co-PI, Mann (USF). Acoustic data will be sampled at 10,000 Hz for ten continuous seconds every 10 minutes and recorded to 32 GB Secure Digital flash memory. With this recording schedule, the devices can run for three months. The Long Term Acoustic Recording Systems (LARS at both sites will be contained in underwater housings and connected to an external hydrophone. Each DSG will be anchored 0.5 m off the bottom and held vertically in the water. Data will be downloaded to a PC. Sounds will be compared to known goliath grouper sounds using MATLAB Signed processing to evaluate frequency range and pulse duration. Because goliath grouper appear to avoid overlap of calls it may be possible to count the number of individuals from recordings. Sounds also will be evaluated for the number of fish calling based on Signed processing (evaluating dB level).

Preliminary analysis of the association between sound production and courtship/ spawning will include evaluating both the diel (a 24-hour period that usually includes a day and the adjoining night) and seasonal pattern of sound and diver and/or ROV (remote operated vehicle) or remote camera observations on possible courtship behavior/sound interactions. FSU owns a Deep Ocean Engineering “mini-phantom” ROV and USF owns a “Video Ray” ROV with sonar both of which are available for this project. Custom built remote cameras with hydrophones, which have been used successfully to document mutton snapper and yellowfin grouper spawning, will be constructed and used in this project also to document spawning behavior and associated sound production. Potential courtship behavior, including sound production, was observed by Colin (1994) and consisted of coloration changes, prodding (the male’s snout nudging the female’s ventral region), and what appeared to be small groups made up of both sexes rising and turning together in the water column. The latter is consistent with videoed observations of mutton snapper and yellowfin grouper spawning, (Locascio personal communication).

77 Diet studies. To determine diet in the captured fish, we use the lavage method. This entails the flushing of the stomach contents into a net through the use of a seawater pump providing just enough pressure to expel the contents. Bait is discarded, but other items are retained, preserved in 10% formalin, and later identified, weighed and enumerated. This procedure does not hurt the fish because the seawater flow is relatively gentle and marine fish constantly drink seawater in the natural environment to maintain osmotic balance.

The maximum time of restraint of a fish for all procedures is about 5 minutes, but because several people are performing the procedures simultaneously on the same fish, the time is usually only about 2 minutes.

After fish are released we monitor their movements in two ways: (1) by monitoring our in situ receivers which record the presence of the fish within ¼- mile radius, and (2) by recording the time and place of tag returns from fishermen who catch the fish in their normal fishing operations. If tagged fish are recaptured by us(about 10% of the fish are recaptured only once over the course of the study), we do the same procedure again so that we can determine growth rates and if the fish have undergone sex change (the vast majority of groupers change sex from female to male during the course of their lives) and if so, under what circumstances.

Some fish specimens (usually only one of two of each species) are retained for educational purposes. These fish are euthanized with the fish anesthetic MS222 before being placed either in 10% formalin or frozen, depending on the intended use. If we already have certain fish species in our reference collections at the lab, we will not collect additional specimens.

Juveniles Capture of small juveniles allows us to determine (1) what the fish larvae and early juveniles are eating in their late pelagic and early benthic stages and (2) will provide an estimate of early mortality that could be related to weather conditions. At this early stage of development the fish are about ½ inch long to 2.0 inches long and are very sensitive to capture and are either dead or moribund upon capture. However, if they are not dead before preserving them, we will euthanize them in an overdose of the anesthetic MS222. All juveniles will be captured with nets and all will be preserved in either EtOH or formalin.

Shark Surveys We employ two fishery-independent survey methods to investigate the abundance, diversity, and habitat use of sharks and rays from Apalachee Bay to the Florida Keys. These methods are approved under Special Activity License 08SR-1092 through the Florida Fish and Wildlife Conservation Commission. The first survey employs experimental gillnets consisting of six panels (30.5m long by 3.05m high) with 7.62cm

78 (3.0”), 8.89cm (3.5”), 10.16cm (4.0”), 11.43cm (4.5”), 12.70cm (5.0”), and 13.97cm (5.5”) stretched mesh. The nets are anchored on each end and marked with end floats and a central buoy. Soak times are less than one hour (usually 30-45 minutes) to minimize mortality of target species as well as bycatch. The gillnet survey is only efficient in assessing abundance of small sharks (<150cm) and does not efficiently sample batoids (batoids are flat-bodied cartilaginous marine fish and include stingrays, guitarfishes, electric rays, skates and sawfishes). Therefore, we will create a second fishery-independent longline survey to assess abundance, diversity, and seasonal habitat use of adult and juvenile sharks as well as adult rays in the Big Bend region. The mainline is 3.5mm monofilament or 6.4mm tarred nylon. Depending on habitat, 6.4mm tarred nylon may be used as mainline instead of monofilament. The mainline is anchored at each end and marked by a buoy and highflier. Additional buoys mark the line at 20-hook intervals. A standard set consists of 100 gangions. Gangions are placed at 15-meter intevals, therefore, the length of a standard set is 1,500 meters from anchor to anchor. Two primary gangion configurations are used. The first configuration is designed to target small coastal species and juveniles of large coastal species and consists of a stainless steel tuna clip with an 8/0 stainless steel swivel attached to 2.5m of 250KG monofilament followed by an 8/0 swivel, 1.5m of 1.6mm stainless aircraft cable, and terminated by a 10/0, 12/0, or 14/0 circle hook. The second configuration is of the same design but for larger sharks, therefore these include 400KG monofilament, 2.2mm aircraft cable, and 16/0 circle hooks. Hooks are baited with Boston mackerel (Scomber scombrus) or comparable species. The bait fish for the longline survey are either purchased from a bait dealer or they are dead bycatch from the gillnet survey. Nothing will be sacrificed specifically for bait. Soak times do not exceed two hours. The PI has made hundreds of sets using this or similar configurations and captured nearly 10,000 sharks from more than 30 species. Capture mortality is relatively low using both gear types though this varies between species.

All handling and tagging protocols are consistent with the Guidelines for the Use of Fishes in Research (2004) published by the American Fisheries Society, American Institute of Fishery Research Biologists and American Society of Ichthyologists and Herpetologists. All captured sharks, rays and teleost bycatch are placed in a measuring trough or measured on deck. All animals are identified as to species and sex. Precaudal length, fork length and stretch total length is measured for sharks, disk width is measured for batoids, and only fork length is recorded for bony fish bycatch. Sharks less than 100 cm TL are tagged with nylon (Hallprint) dart tags beneath the first dorsal fins into the dorsal musculature. Larger sharks are tagged similarly with M-type dart tags. The tags are inserted into the dorsal musculature at the base of the first dorsal fin using a stainless steel application needle approximately 3 mm in diameter. The head of the nylon dart is anchored behind the medial cartilage of the dorsal fin which greatly increases retention. The M-type darts are designed to be anchored only in the musculature and do not reach the medial cartilage. After insertion of the tag, the tagging needle is retracted. These are the standard shark tags and

79 standard tagging procedures used by the National Marine Fisheries Service and most researchers studying shark movements. The PI (D. Grubbs) has tagged more than 10,000 sharks using these methods and has examined more than 100 recaptures, including sharks at liberty more than ten years. Necrosis is extremely rare and long term growth rates suggest tagging has no effect on biological function. A small sample of tissue is collected for genetic and stable isotope analyses (fin clip or muscle biopsy). Fin clips approximately (~1cm2) are collected from the left pelvic fin using stainless steel snips. A stainless steel biopsy punch (3-6 mm diameter depending on fish size) is used to collect muscle samples from the dorsal musculature. The biopsy punch extracts via suction a muscle sample 2-5 mm deep. The small hole remaining heals naturally. Antibiotics are not applied to the wound for two reasons. First, topical antibiotics have not been shown to be effective in fish surgery and may be deleterious by harming natural bacterial growth (Mulcahey 2010). In addition, the only antibiotic currently approved by the FDA for use with fishes that will be released and may enter the human food chain is oxytetracycline. Nevertheless, a small amount of triple antibiotic in mineral oil is applied directly to the wound immediately following the biopsy procedure. All animals are released immediately following tagging and total handling time is less than one minute.

All batoids except smalltooth sawfish will be tagged with looped cinch tags or Peterson disk tags which are the standard tags used by the National Marine Fisheries Service and the PI has tagged more than 1,000 batoids using these methods. A small hole is made in the anterior edge of the spiracle using a stainless steel tagging needle. The tag is then inserted into this hole and the cinch snap closed. Smalltooth sawfish are a federally listed endangered species. All sawfish must be handled and tagged in accordance with protocols established in NMFS ESA Permit 13330. Dr. John Carlson of the National Marine Fisheries Service is the primary permit holder and Dr. Dean Grubbs of FSU is sub- permitted on this ESA protocol. All sawfish are measured using the same methods as with sharks with additional morphometric and meristic measurements specific to sawfish (e.g., rostral length, tooth counts). A fin clip (~1cm2) will be taken for genetics and stable isotope analysis. All sawfish are tagged with a Passive Integrated Transponder (PIT tag) at the base of the first dorsal fin. External tags (rototag or nylon dart tags) are applied to the first dorsal fin. Dart tags are applied as in sharks. The barb of Roto tags (or cattle ear tags) is simply pushed through the anterior edge of the first dorsal fin near the base. This region consists only of connective tissue and is not vascularized. The tagging procedure results in no bleeding. Once pushed through the dorsal fins, the back plate of the tag is snapped onto the barb, locking it in place. Acoustic transmitters are attached to the rototag for a subset of juvenile sawfish and pop- off archiving satellite transmitters are attached to the dorsal fin of adult smalltooth sawfish.

Sharks and teleost bycatch that are dead at capture are retained for studies of life history and trophic ecology.

Prey sensory experiments Background and Significance: As part of our ongoing research on predator-prey interactions, which can have strong effects on the abundance of populations and the structure of communities, we wish to investigate the sensory tactics employed by fishes. Specifically, we will investigate the processes underlying prey being able to detect their predators. Visual and olfactory (i.e., chemical) cues have both received considerable attention from previous studies (e.g., Helfman 1989, Murphy and Pitcher 1997, Smith and Belk 2001, Brown 2003, Brown and Chivers 2006). For example, many freshwater fishes have been shown to have strong abilities to detect chemical cues, often better than they can detect visual ones, presumably due to the low visibility found in turbid river waters and often high structural complexity (i.e., an additional impedance to visual detection) associated with aquatic habitats. Far less research has been conducted in habitats of high visibility, where the expectation has been that visual detection is far more important. However, recent work on coral reefs, where water clarity is high, has shown chemical cues to be surprisingly important (McCormick and Larson 2007, McCormick and Manassa 2008, Holmes and McCormick 2010).

Marine habitats of the Big Bend, FL present several combinations of water clarity (from clear to murky) and habitat complexity (from low to high), thus presenting species that inhabit them with a range of environments that can affect their ability to detect predators visually. Some species (e.g., pinfish) are abundant across all habitat combinations (Stallings et al. 2010), suggesting they may have strong abilities for both visual and olfactory detection or some level of plasticity depending on the specific environment they are in. Moreover, one of the more important predators of pinfish in these shallow habitats, and that used in previous predator-prey experiments we have conducted at the FSU Coastal and Marine Lab, is the Gulf flounder which itself relies on highly cryptic feeding behaviors. Controlled laboratory experiments are required to better understand the relative importance of visual versus olfactory detection abilities of pinfish on Gulf flounder.

Methods: To examine the relative importance of chemical vs. visual cues for pinfish, we will conduct controlled laboratory experiments in closed seawater aquaria (i.e., using aquarium grade sea salt such as “instant ocean” mixed with deionized water). Closed seawater aquaria will be used to eliminate uncontrolled chemicals (i.e., from other individuals or species) likely present in the ambient water and flow-through seawater system. Pinfish and Gulf flounder will be captured using standard otter trawling methods1, immediately placed in aerated

1 Otter trawls are towed along the sea floor. The mouth of the net is kept open with large rectangular otter boards positioned in such a way that the hydrodynamic forces acting on them when the net is towed pushes them outwards and prevents the mouth of the net from closing and allowing capture of fishes.

81 containers aboard the research vessel, brought back to the FSUCML, and placed in acclimation tanks consisting of 50% ambient water from their collection site and 50% “instant ocean” seawater. Water salinity and temperature will exactly mimic that from which fish were captured; salinity can be controlled by altering the mix of instant ocean and water, while temperature can be controlled with either heaters or coolers as needed. Our previous work with both pinfish and Gulf flounder has shown them to be incredibly hardy species, but we will allow them to acclimate in the ambient/instant ocean mix for one hour before transferring to aquaria with 100% instant ocean water, and then will allow them to acclimate an additional hour before beginning experimental trials. Densities of pinfish will not exceed 10 individuals per 76 liter acclimation tank and one individual per 38 liter experimental tank. Only juvenile stages of pinfish with standard lengths less than 90mm (Nelson 2002) will be used. Densities of Gulf flounder will not exceed two individuals per 76 liter acclimation tanks and one individual per 38 liter experimental tank. Flounder sizes will not exceed 250mm standard length.

Experimental aquaria will contain a single structure for the fish to hide in. Following the one hour acclimation period in the 50/50 mixture water and the additional one hour in the experimental tank, we will use a video camera to measure the proportion of time a single pinfish spends hiding in the absence of the predator cue (see below for different cues) over a short period of time (e.g., 5-10 min). Then we will introduce the cue and again measure the proportion of time hiding over the same period of time. The measured variable becomes the difference in time hiding between pre- and post-cue. Seven different cues will be tested:

1. Visual only 2. Olfactory only a. Predator alone b. Conspecific capture release c. Predator + capture release 3. Visual and olfactory

The following methods will be used for stimulus preparation and experimentation: ñ Cue 1 (visual only): a single Gulf flounder will be placed in a separate 38 liter aquarium adjacent to the experimental aquaria containing the pinfish. ñ Cue 2a (olfactory – predator alone): 60ml of water will be drawn from the aquarium containing a Gulf flounder being held out of sight of the experimental aquaria. The “predator water” will be injected into a plastic tube attached to another tube with an air stone and air supply. The air stone/supply will both aerate the water and help disperse the cue. ñ Cue 2b (olfactory – capture release): Prey often release alarm cues

82 associated with their epidermis when harmed/captured/etc. We will rapidly euthanize five pinfish of similar size as the live experimental individuals using an overdose of MS-222 (minimum concentration of 250 mg/L; http://www.research.cornell.edu/care/documents/SOPs/CARE306.pdf). We will ensure each pinfish is dead before removing it from the solution by making sure there is no opercular movement for at least 3 minutes. Such small numbers of individuals to sacrifice represents a miniscule portion of the local population (which we have estimated at over 6 billion individuals) and the euthanasia method will be rapid and humane. Once sacrificed, the pinfish will be placed on a clean petri dish, given 25 superficial cuts to the skin, rinsed with 60ml of seawater and filtered to remove any solid materials. The filtered water will then be introduced to the experimental aquarium using the techniques described above in 2a. ñ Cue 2c (olfactory – predator + capture release): Water will be drawn as in Cue 2a and 2b (60 ml from each), mixed and 60ml will be administered to the experimental aquarium. ñ Cue 3 (visual and olfactory): These are three treatments representing a combination of the visual cue (1) combined with each of the individual olfactory cues (2a, 2b, 2c).

Each experimental pinfish will be used only once and will be released unharmed to the location from which they were captured. Total holding times for experimental pinfish will not exceed one week (and will often be just 1-2 days) and they will be fed ad libitum with commercial-grade fish food while in capture. Because experimental aquaria must use new water at the onset of each trial, tanks will be cleaned after each trial. Cleaning of experimental tanks will proceed as: 1) removal of experimental animal, 2) complete draining of water in tank via siphoning, which will also pick up and remove any solid waste (e.g., feces) from experimental animal, 3) rinsing of the tanks and single habitat structure three times with deionized water. Flounder will not be held for more than three weeks, will be fed ad libitum with commercial-grade fish food while in capture, and will be released unharmed at the location from which they were captured. Temperature and salinity will be closely monitored and adjusted as needed to mimic ambient conditions for all aquaria.

Oyster reef ecology Oyster reefs historically helped maintain estuarine health by filtering enormous volumes of water, cycling nutrients, and increasing biodiversity. Because this critical habitat has been globally degraded to only ~15% of its historical abundance, considerable effort has been devoted towards restoring oyster reefs. These restoration efforts, however, have failed to consistently improve diversity and ecosystem functioning (i.e., cycling nutrients and filtering water). By linking predator effects and ecosystem services provided by oysters (nutrient cycling and water filtration) in a biogeographic context, our research will provide valuable

83 information to resource managers charged with restoring oyster reefs and recovering ecosystem services.

Intertidal oyster reefs (Crassostrea virginica) are a model system to address linkages between predator effects and ecosystem services for several reasons: they occur throughout the mid-Atlantic and Gulf coasts (1,750 km); they contain a similar food-web assemblage across latitudinal gradients in predation, resource supplies, and environmental conditions; they are strongly influenced by predator effects; and they influence sediment and nutrient cycles through water filtration and waste excretion. Here, we propose to examine whether the consumptive- and non-consumptive effects (hereafter CEs and NCEs) of predators differentially influence oyster reef water filtration services, and whether and why these predator-ecosystem linkages differ throughout the oyster’s biogeographic range. Specifically, this proposed research involves a series of standardized sampling and experimental studies to: (1) investigate biogeographic patterns in oyster food web structure, resource supplies, environmental conditions, and sediment properties associated with reef function (2) determine how the vital rates of oysters (i.e., growth and survivorship), which can influence benthic-pelagic coupling, vary geographically; and (3) examine experimentally the relative importance of consumptive and non-consumptive predator effects on oyster reef communities and the ecosystem processes they provide, and how these effects vary latitudinally. This project will provide a mechanistic understanding of the basis for biogeographical shifts in valuable ecosystem services performed by an important marine foundation species, and it will also advance our understanding of the interactions between predator effects in food webs and the ecosystem processes that depend on them.

Objective #1 – Monitoring: To understand how estuarine fishes are by supported by and structure the food web on oyster reefs, we will monitor the distribution and abundance of predatory fishes every four months. This monitoring is approved under Special Activity License 10-1223-SR through the Florida Fish and Wildlife Conservation Commission and depends on sampling 5 intertidal oyster reefs (3 x 3 m) per site (n=4). This sampling employs small monofiliament nets (30 feet long x 4 feet high; 3 inch square mesh openings) that are wrapped around each reef, thereby reducing the maximum continuous length of the net to fifteen feet. These nets are anchored on each end and marked with end floats and a central buoy. For each sampling period, gillnets will be erected for two hours at peak high tide when fish utilization of shallow intertidal reefs is typically highest (< 2 hour soak time). When retrieving monofilament nets from each reef, a research team will work quickly to remove all fish from nets. Upon freeing each fish from the net, we will quickly identify and measure the size of each fish before releasing them into the water. Fishes that are dead at capture will be retained for studies of trophic ecology. Targeted fishes include: drum, hardheaded catfish, Gulf flounder, croaker, and toadfish.

84 Objective 2 -- Field experiment: We predict that hardhead catfish (Ariopsis felis) and oyster toadfish (Opsanus tau) will be the most abundant fishes on intertidal oyster reefs. As a result, we will examine if these fishes benefit the survivorship of intertidal oysters by both consuming and altering the foraging behavior of oyster consumers (mud crabs). To address our study questions, we will conduct a field experiment in oyster reefs adjacent to the Florida State University Coastal and Marine Laboratory (FSUCML), located in St. Teresa, FL as well as those adjacent to the University of Florida, Whitney Marine Lab. At each site, this experiment will consist of four treatments (n=5): (1) oysters only, (2) oysters and mudcrabs (no fish), (3) oysters, mudcrabs and 1 toadfish, and (4) oysters, mudcrabs, 1 toadfish and 1 catfish. Before beginning the experiment, we will quantify the condition of oysters (density and size of living oysters) and mudcrabs (density and size). These data will be collected at the end of the experiment to examine how the densities and traits of oysters and mudcrabs were altered by experimental treatments (i.e., fish predators present or absent).

The four experimental treatments will be maintained by enclosing replicate oyster patches (1.5 x 1.5 m) within square cages (2 x 2 x 2 m). The cages will be constructed from aquaculture netting (Vexar), which allows all enclosed marine animals to experience the natural, free flow of seawater while also be constrained by a non-abrasive surface. All cages will be secured in place a minimum of 1 week prior to adding fishes, which will be collected via the monitoring protocol described above. Upon capture, each fish (10 toadfish and 5 catfish) will be measured, placed in aerated containers and transported to the experimental cage. Total handling time from capture to release inside the cages will not exceed 5 minutes. In addition to allowing flow-through of seawater, these cages will also contain ambient levels of natural prey of toadfish and catfish (e.g., small crustaceans and especially mudcrabs) that are highly abundant in oyster reefs (Grabwoski et al. 2005). Because individual fish may exhibit stress symptoms despite the natural flow and food conditions within the cage, we will monitor the caged fishes in situ on a weekly basis by noting their color and activity level. If a fish displays any adverse effects of being caged (e.g., unusually pale coloration or reduced activity level), we will release it back to the location from which it was collected and replace it with a different individual. This experiment will last one month and will be repeated twice during the summer of 2011.

At the conclusion of each experiment, we will use a seine in each cage to retrieve all experimental fish. Each fish will be placed in a container with aerated seawater, where we will measure standard length via a wet, fish measuring board to calculate fish growth (growth = final size – initial size). After measuring each fish, we will quickly transfer the fish from a holding to a recovery tank, which will contain ambient water aerated with air stones. Upon measuring all fishes, we will then quickly release them into the location from which they were first obtained. In addition to collecting data on these fish and releasing them, we will

85 also be collecting data on mudcrabs, oysters, and nutrient cycling in sediments underneath oyster reef as a function of predator treatment.

Relocating fish Red grouper have been observed actively manipulating habitat in Florida Bay. In order to determine the ecological effects that this manipulation has on the surrounding fish community, some juvenile red grouper will be removed from their habitat and relocated to similar habitat some distance away. This is done to ensure that the same individual fish does not return to the experimental habitat and confound the experimental results. Fish caught by hook-and-line or commercial fish trap are immediately transferred to an aerated holding tank. Relocation habitats are of same type (limestone hardbottom surrounded by seagrass), depth, and quality, and are more than 5km from the capture site. Divers will evaluate suitable relocation sites before the experiment commences in order to reduce total holding time. Total holding time (from capture to release) will be less than 30 minutes. While on the boat, the fish is kept in a well aerated holding tank, covered to minimize sun exposure. Relocated individuals will be tagged with dart tags (procedure described above) prior to release.

6. Will Monoclonal/Polyclonal Antibody Production Be Performed? Yes____ No X (If no, proceed to Question#7)

7. Will Surgery Be Performed? Yes __X___ No ______(If no, proceed to Question #8)

a) Survival Surgery _X__ Non-Survival Surgery____ Multiple Survival Surgeries _____ (Check all that apply.)

b) List Room(s) where surgery will be performed : offshore on research vessel

c) List Procedures:

Implantation of acoustic transmitter in the body cavity: We will implant transmitters in body cavity for telemetry-based survival studies. The surgery entails making a small incision with a sterile scalpel in the body wall at the linea alba just anterior to the anus and slipping a small antibiotic-covered (2-inch long, 0.5-inch diameter diameter) transmitter into the opening. We then apply triple

86 antibiotic salve to the wound and use sterile surgical staples (titanium) or sterile sutures (nylon monofilament) to close the opening. Although the procedure is considered painful by some authorities, the fish do not exhibit any overt signs of pain or stress as the procedure is being done.

Gonad biopsy: Gonads are biopsied non-surgically (i.e., no incisions are made; the sample is taken through the gonoduct) with a small tube (1.5 to 3 mm internal diam depending on size of the fish and the gonoduct) and inserted into the gonoduct. A vacuum is applied with a hand-operated vacuum pump and a small amount of gonad tissue (1 to 10 grams depending on the size, sex, and reproductive condition of the fish) is removed from the internal lumen of the gonad. This procedure is apparently painless because the fish shows no signs of distress or pain—no flinching and no moving about. Again, triple antibiotic salve is used at the site of entry for any invasive biopsy method to minimize chances of infection.

Muscle biopsy: Muscle biopsies will be taken from large fish for genetic studies and for trophic studies (stable isotope analyses). In all cases, sterile instruments will be used. Biopsy punches are sold sterilized in sterile packaging—a new sterile punch will be used for each individual sample and triple antibiotic will be applied to each biopsy wound to minimize chances of infection.

d) List Multiple Survival Surgery Procedures and Time Frames Between Each Procedure: No multiple surgeries

Procedure 1 Procedure 2 Procedure 3 Time Between Each Procedure

e) Scientific justification for Multiple Survival Surgery:

e) Describe pre-anesthetic protocol (If any): N/A

f) Anesthetic Agents: Species Agent Dose Route of Administration

g) Will paralytic agents be used (list any and method of monitoring if yes): N/A

Describe aseptic surgical procedures: All surgical instruments that are to be re-used (e.g., retractors, forceps, etc.) will be autoclaved onshore prior to use offshore and stored in sterile plastic bags after autoclaving and prior to use. Sterile latex gloves will be worn by all those performing the surgery. So if 10 surgeries are to be done, then 10 non-disposable autoclaved instruments will be taken for each fish to be sampled. Disposable sterile surgical instruments purchased in sterile packaging (e.g., scalpel blades, biopsy punches, etc.) will be used whenever possible, one set of surgical instruments for each specimen and each procedure. Triple antibiotic salve will be applied to all wounds and points of biopsy to minimize chances of infection.

h) List methods for intra-operative monitoring of anesthetic depth:

j) Describe post-operative monitoring and care of animals, including provision of analgesic agents (agent, dose, route and frequency of administration): Tagging. – (1) Transmitters. Fish will be released immediately after implanting the transmitters. We cannot hold the fish after surgery because of their large size and because we will be up to 50 miles offshore in a research vessel where space and facilities are limiting. More importantly, the quicker the fish can be released to the wild, the better their chances of recovery and survival. These are wild fish and returning them to their natural environment as soon as possible is the best approach. Survival is then monitored among selected individuals through the implanted transmitters. (2)Dart tags: Conventional tagging with dart tags has been used for nearly a century to track the movements of fish. We buy our tags from Floy Tag Company (http://www.floytag.com/) which has been supplying tags to researchers for over 50 years. Nylon tag heads of the plastic dart tags are inserted with an applicator needle and the tag head is inserted through the pterygiophores (fin support bones) so that the tag is firmly anchored. We apply triple-antibiotic on all dart tags prior to insertion. Tagged fish are immediately returned to the sea. We have never observed an infection associated with tagging using our methods on any of the hundreds of recaptured adult or juvenile fish we’ve observed over the years (see example for goliath grouper juveniles in Koenig et al. 2007, attached)

8. Does This Protocol Include Food And/Or Water Restriction? Yes____ No__X__ (If no, proceed to Question #9)

88 9. Does this protocol include behavioral experiments? Yes _X___ No _____ (If no, proceed to Question #10)

a) Please list and describe in detail all behavioral experiments to be performed: The sensory experiment involving pinfish will measure the relative hiding versus exposed behaviors in the absence and presence of different predatory cues (visual and chemical) from one of their natural predators, the Gulf flounder (detailed description of methods given above). The cues we will be exposing the pinfish to are experienced by them in their natural environments. Moreover, measurements of behavior will not exceed a total of 20 minutes for each individual pinfish

b) If more than one procedure will be performed per animal, please list combinations and time frame: More than one procedure will not be performed per animal.

c) Will any behavioral experiment listed above have the potential for inducing pain or prolonged distress? If yes, list procedure, provide justification for its use and explain what methods will be taken (if any) to mitigate any pain or prolonged distress: The procedure will neither induce pain nor will it induce stress beyond natural predator-prey conditions the animals are exposed to in the wild.

10. Will live capture of wild animals be performed? Yes _X__ No ____ (If no, proceed to Question #11) a) Check all that apply: __X__Live capture and release at site of capture

____ Live capture and release at different location __X__Non-survival collection

b) Study Locations: Study will be in coastal waters of Florida out to the shelf edge (50-120 m deep)

c) Applicable permits. List all applicable federal and state permits (and permit numbers) required for this project; include approval and expiration

89 dates. Copies of permits do not need to be provided but should be available upon request:

Reef fish research permits KOENIG and COLEMAN - FFWCC Special Activity License SAL-10-1244-SRP (08/10/2010-08/9/2011) KIMBRO – State permit number 10-1223-SR (July 2010-June 2013) STALLINGS - SAL-10-1206-SR (4/13/2010 - 4/12/2013) and SAL-10-1221-SR (5/25/2010 - 5/24/2013)

GRUBBS - Permit for fishery independent shark surveys – SAL-08SR-1092 (07/909/2008-07/08/2011

Permits for sawfish research Federal Endangered Species Permit #13330, expires 31 October 2013 State of Florida Department of Environmental Protection Permit 11-0302874- 001 U.S. Army Corp of Engineers Permit SAJ-2010-02812 (Ten Thousand Islands) Federal Permit EVER-2010-SCI-0007 (Everglades National Park) d) Describe procedure(s); include method of capture, frequency of trap monitoring, marking procedures, transport methods, euthanasia and potential injury or mortality risks: These methods are described in detail under 5. above. All have a low potential for injury (i.e., < 5%).

Note: The FSU ACUC requires that a log describing all species captured as well as injury and mortality rates be maintained for field/wild animal studies and be available for review upon request.

11. List all test substances (drugs or pharmaceutical agents, chemicals, infectious agents, radioactive agents, recombinant nucleic acid, etc.) to be used in the course of the proposed work. Do not include anesthetics, analgesics or sedatives used as part of surgery or

90 postoperative care that have been previously included in Question #7. (If no agents are to be used, proceed to Question #12)

Complete this section for each animal species and substance/agent. Species Agent Route Site Volume Number Interval of Doses Between Doses NA NA NA

If toxic or other adverse reactions may occur in animals, describe possible reactions and methods to deal with the reactions.

12. This research requires the following safety procedures: Please review the information on hazardous agents and standard animal research precautions below. In addition, if the use of hazardous agents is planned, also review the FSU EH&S Use of Hazardous Agents In Animals Policy and complete Appendix B. The ACUC will not grant approval of animal use protocols using hazardous agents until this appendix has been completed, reviewed and approved.

Yes No

Standard Animal Research Precautions X

Special Animal Husbandry or Staff Precautions

If Yes, please complete and submit Appendix B, the Use of Hazardous Agents X in Animals Supplemental Safety Operating Procedures Form with this protocol.

If you have questions call Environmental Health and Safety, 644-5374 or 644-9117

NOTE: Hazards can be chemical (e.g. disinfectants, bleach, formalin, drugs), physical (including ionizing and non-ionizing radiation, bites, kicks, scratches, needle sticks, dust, noise, heavy lifting), or biological (zoonotic agents, recombinant DNA, infectious agents) in nature. The use of all drugs, chemicals or other potentially hazardous agents must undergo a risk assessment prior to protocol approval to determine if special standard operating procedures need to be put into place. Risk assessment is a science used to evaluate the risk for both carcinogenic and non-carcinogenic effects to human and

91 animal health from drugs, chemicals or other agents that may be used in the course of live vertebrate animal research. These risk assessments include the evaluation of both direct and indirect risks from the time an agent is used in an individual animal through the waste steam from that manipulated animal (dirty cages, carcasses, etc.). If uncertain whether any proposed drug, chemical or other agent for this protocol may be a hazardous agent, or if you need help in formulating a standard operating procedure for use of the above agent(s), please contact FSU Environmental Health and Safety at 644- 5374 or 644-9117.

If use of hazardous agents is planned, researchers are responsible to ensure that a copy of any approved SOPs and all required signage are properly displayed or available in the room. In addition, particularly if work is not continuous, written notification of pending hazardous work must be made to LAR at least two business days prior to working with the agent in an animal use area.

Note that health status, including pregnancy, allergies or some medications or medical conditions may increase risks associated with potential exposure to animal allergens, chemicals, drugs or other hazards. Individuals may contact EH&S with questions related to added precautions that may be employed.

Standard Animal Research Precautions ñ All individuals working with animals should follow routine use of personal protective equipment, good work practices and use of engineering controls appropriate to their project or animal husbandry requirements. ñ Mandatory PPE for work with animals include disposable gloves and a laboratory coat (or other dedicated protective clothing such as a scrub suit). Foot hugging, closed toe and low heeled shoes should be worn when working in the animal facility. Masks are recommended. In some cases, protective eye wear is also indicated. ñ Good work practices include performing animal manipulations within well ventilated areas, minimizing disturbance of animal bedding and exposure to loose animal hair, keeping cages and animal areas clean and washing hands after removal of gloves or any animal handling. Do not eat, drink, or apply cosmetics while working in an animal use area; food and drink is forbidden in animal housing and use areas. Remember that unfixed tissues, blood, serum, urine, and other materials derived from animals may also pose a risk. Engineering controls include biosafety cabinets, chemical fume hoods, HEPA filtered vacuums, downdraft tables and other fume scavenging devices.

13. Will Prolonged Restraint be used in this protocol? Yes ____ No __X___ (If no, proceed to Question #14)

14. Euthanasia. All methods are required to be in accordance with the 2007 AVMA Guidelines on Euthanasia. List all agents and, if more than one species on a protocol, species per method to be used.

a) Inhalant Agent(s): MS222

b) Injectable Agent(s):

c) Physical Method(s): ____Cervical dislocation ____ Decapitation

d) Will animals be anesthetized for cervical dislocation and/or decapitation? If no, then provide scientific justification for performing the procedure without anesthesia.

e) Exsanguination (describe method and anesthetic agent):

f) Other Method of Euthanasia or Disposal:

15. List and describe all other procedures performed in animals not already listed in questions 8-13 above (Rodent Breeding should submit Appendix A).

N/A

16. POTENTIAL DISCOMFORT, DISTRESS OR PAIN: Procedures involving animals should avoid or minimize discomfort, distress or pain. If more than momentary pain or distress cannot be avoided, justification must be provided. Procedures requiring justification include (but are not limited to): electric shock, Freund's Adjuvant, paralyzing agents, food restrictions, multiple survival surgeries, surgeries involving non-aseptic procedures, pain or distress without anesthesia or analgesia, euthanasia not consistent with the 2007 AVMA Guidelines on Euthanasia, other painful/distressful procedures. Methods to minimize or alleviate pain or distress must be described in this section if not already outlined above.

Surgery is absolutely necessary for implanting transmitters in the fish. There is no other way to attach a transmitter to a fish and be certain of retention. This procedure cannot be avoided. Our approach to minimizing pain is to use aseptic techniques and minimize time out of water for these wild-caught fish. This means that we will be gentle but quick in our handling of the fish. While the use of anesthetics may be advisable, it is seriously problematic when working with food/game fish that are to be released immediately back into the water following procedures. Currently there are few alternatives approved for anesthesia in fish.

93 For those appropriate to use in the field, there is no mechanism by which this protocol can hold the fish over the prolonged period of time necessary for FDA withdrawal times (30-90 days) to avoid chemicals entering the human food chain. Several years of work using the described approach along with recapture indicates that fish do not appear to experience detrimental effects from the lack of anesthesia for the telemetry procedure.

17. HOUSING REQUIREMENTS (please indicate “yes” or “no” for each item, describe non-standard caging, and if “yes” to 17(E) and 17(F) please include the appropriate forms with protocol submission. If “no” to 17(D) please include appropriate form.

YE NO CAGING DESCRIPTION S

X A. Standard

X B. Modified Standard

X C. Test Chambers

There may be occasions where use of environmental enrichment devices may interfere with research objectives. In these cases, investigators may request an exemption for X D. Environmental animals on study. All requests must undergo Enrichment review and approval by the FSU Animal Care and Use Committee. If checking NO, please fill out and submit the ‘Exemption from Environmental Enrichment’ form. THESE FORMS MUST BE FILLED OUT AND E. Non LAR ATTACHED X Personnel will provide care for Form - "Guidelines for developing an Animal the Animals Care Standard Operating Procedure"

THESE FORMS MUST BE FILLED OUT AND F. Animals will be ATTACHED X Housed Outside the LAR Facilities Form - "Request to House Outside LAR Facilities

94 Form - "Guidelines for developing an Animal Care Standard Operating Procedure"

Fish will be temporarily (< 3 weeks) housed at the FSUCML in aquaria of either flow-through seawater or closed system seawater made from aquarium grade sea salt such as “instant ocean” mixed with deionized water. Since flow-through seawater will be pumped in from the area adjacent to the marine lab, where the animals to be collected live, it will by necessity involve the exact same seawater conditions (i.e., temperature, salinity). Closed system seawater will be closely monitored and adjusted as required to mimic ambient natural seawater conditions for temperature and salinity.

18. ASSURANCES: Please “x” to indicate “yes” or “no”

Do any of the proposed procedures cause more than momentary distress, pain, or are they YES_X__ NO__ potentially painful? (If no, proceed to Question #19)

NOTE: If any procedures employed in this Animal Use Description fall into USDA Pain Categories D or E, then a thorough literature search for alternatives must be performed. Alternatives include methods that (a) reduce the total number of animals utilized for an experiment, (b) refine experimental techniques to minimize or avoid animal distress or pain and (c) replacement of whole-animal use with in vitro methods or other non-animal alternatives.

You must provide adequate information to the ACUC to assure the committee that alternatives to painful, potentially painful or distressful procedures were considered and are either not available or cannot be used. Your response must include: ° The databases searched or other sources consulted ° The date of the search and the years covered by the search ° The key words and/or strategy used when considering alternatives or descriptions of other methods and sources used to determine that no alternatives were available to the painful or distressful procedure

95 Databases (place an ‘X’ next to those used): Agricola Biosis Medline Psyc First Article First Others (list):

Journals and/or other sources consulted (list):

Basgall, M. 1999. How to tag a tuna. Duke Magazine:45-46 Block BA, Dewar H, Williams T, Prince ED, Farwell C, Fudge D. 1998. Archival tagging of Atlantic bluefin tuna (Thunnus thynnus thynnus). MARINE TECHNOLOGY SOCIETY JOURNAL 32 (1): 37-46 SPR 1998

Block, B. A. H. Dewar, C. Farwell, E. D. Prince. 1998. A new satellite technology for tracking the movements of Atlantic bluefin tuna. Proc. Nat. Acad. Sci. 95, Issue 16, 9384-9389, August 4, 1998.

Bushon AM, Stunz GW, Reese MM. 2007. Evaluation of visible implant elastomer for marking juvenile red drum. NORTH AMERICAN JOURNAL OF FISHERIES MANAGEMENT 27 (2): 460-464.

Berejikian BA, Endicott RC, Van Doornik DM, et al. 2007. Spawning by female Chinook salmon can be detected by electromyogram telemetry. TRANSACTIONS OF THE AMERICAN FISHERIES SOCIETY 136 (3): 593- 605.

Domeier ML, Kiefer D, Nasby-Lucas N, et al. 2005. Tracking Pacific bluefin tuna (Thunnus thynnus orientalis) in the northeastern Pacific with an automated algorithm that estimates latitude by matching sea-surface-temperature data from satellites with temperature data from tags on fish. FISHERY BULLETIN 103 (2): 292-306

Kennedy BM, Gale WL, Ostrand KG. 2007. Evaluation of clove oil concentrations for use as an anesthetic during field processing and passive integrated transponder implantation of juvenile steelhead. NORTHWEST SCIENCE 81 (2): 147-154.

Mulcahy DM. 2010. Antibiotic use during the intracoelomic implantation of electronic tags into fish. Reviews in Fish Biology and Fisheries. Online publication date: 16-Dec-2010.

Nickum, J.G., Bart Jr., H.L., Bowser, P.R., Greer, I.E., Hubbs, C.,Jenkins, J.A., MacMillan, J.R., Rachlin, J.W., Rose, J.D., Sorensen,P.W., Tomasso, J.R., 2004. Guidelines for the Use of Fishes in Research. American Fisheries Society, Bethesda, Maryland.

Robertson, OH. 1945. A method for securing stomach contents of live fish. Ecology 26(1):95-96

Sigurdsson T, Thorsteinsson V, Gustafsson L. 2006. In situ tagging of deep-sea redfish: application of an underwater, fish-tagging system. ICES JOURNAL OF MARINE SCIENCE 63 (3): 523-531.

Sluka RD, M Chiappone, KMS Sealey. 2001. Influence of habitat on grouper abundance in the Florida Keys, USA. Journal of Fish Biology. 58 (3): 682-700 Shardlow TF, Hyatt KD, Pearson SR. 2007. A new microcontrolled fish tag for accurately estimating life span and survey life of spawning salmon. AQUATIC LIVING RESOURCES 20 (2): 197-203.

van der Kooij J, Righton D, Strand E, et al. 2007. Life under pressure: insights from electronic data-storage tags into cod swimbladder function. ICES JOURNAL OF MARINE SCIENCE 64 (7): 1293-1301.

Wilson SK, Wilson DT, Lamont C, et al. 2006. Identifying individual great barracuda Sphyraena barracuda using natural body marks. JOURNAL OF FISH BIOLOGY 69 (3): 928-932.

Tagging Studies Claverie, T. and Smith, I.P. 2007. A comparison of the effect of three common tagging methods on the survival of the galatheid Munida rugosa. Fisheries Research 86: 285-288.

Emlen J.M. 1966. The role of time and energy in food preference. American Naturalist 100(916):611.

Gillanders B.M. 2006. Seagrasses, fish, and fisheries. pp. 503-536. In: Larkum AWD, Orth RJ, Duarte CM (ed.) Seagrasses: biology, ecology and conservation, Spronger, Dordrecht, The Netherlands.

Goodman, L.R., Campbell, J.G. 2007. Lethal levels of hypoxia for gulf coast estuarine animals. Marine Biology 152:37-42.

Harter S.L. and Heck K.L.J. 2006. Growth rates of juvenile pinfish (Lagodon rhomboides):Effectsof habitat and predation risk. Estuaries and Coasts 29(2):318.

Hemminga M.A., Duarte C.M. 2000. Fauna associated with seagrass systems. pp. 199- 247 Seagrass Ecology, Cambridge University Press, Cambridge, United Kingdom.

Jordan F., Bartolini M., Nelson C., Patterson P.E., Soulen H.L. 1997. Risk of predation affects habitat selection by the pinfish Lagodon rhomboides. Journal of Experimental Marine Biology and Ecology 208(1-2):45.

Kerwath, SE, Gotz, A, Wilke, C, et al. 2006. A comparative evaluation of three methods used to tag South African linefish. African Journal of Marine Science 28: 637-643.

Knight, A.E. 1990. Cold branding techniques for estimating Atlantic salmon parr densities. Am. Fish. Soc. Symp. 7: 36-37.

Koenig C.C. and Coleman F.C. 1998. Absolute abundance of juvenile gags in seagrass beds in the northeastern Gulf of Mexico. Transactions of the American Fishery Society 127:44. Koenig, C.C, F.C. Coleman, A-M. Eklund, J. Schull, J. Ueland. 2007. Mangroves as essential nursery habitat for goliath grouper (Epinephelus itajara). Bull. Mar. Sci. 80(3):567-586.

Nelson, G.A. 2002. Age, growth, mortality, and distribution of pinfish (Lagodon rhomboides) in Tampa Bay and adjacent Gulf of Mexico waters. Fishery Bulletin 100: 582-592.

Skinner, MA, Courtenay, SC, Parker, WR, et al. 2006. Evaluation of techniques for the marking of mummichogs with emphasis on visible implant elastomer. North American Journal of Fisheries Management 26: 1003-1010.

Smith D.A., Smith S.A., Holladay S.D. 1999. Effect of previous exposure to TM methanesulfonate on time to anesthesia in hybrid tilapias. J Aquat Anim Health 11:183–186.

Wootton, R.J. 1990. Ecology of teleost fishes. Chapman and Hall, London.

Sensory studies Brown GE (2003) Learning about danger: chemical alarm cues and local risk assessment in prey fishes. Fish Fish 4:227–234

Brown GE, Chivers DP (2006) Learning about danger: chemical alarmvcues and predation risk assessment in fishes. In: Brown C, Laland K, Krause J (eds) Fish cognition and behaviour. Blackwell Science,vOxford, pp 49–69

Helfman GS (1989) Threat-sensitive predator avoidance in damsel-fish - trumpetfish interactions. Behav Ecol Sociobiol 24:47–58

Holmes, TH, McCormick, MI (2010) Smell, learn and live: The role of chemical alarm cues in predator learning during early life history in a marine fish. Behavioural Processes 83: 299-305.

McCormick MI, Larson JK (2007) Field verification of the use of chemical alarm cues in a coral reef fish. Coral Reefs 26: 571-576.

McCormick MI, Manassa, R (2008) Predation risk assessment by olfactory and visual cues in a coral reef fish. Coral Reefs 27: 105-113.

Murphy KE, Pitcher TJ (1997) Predator attack motivation influences the inspection behaviour of European minnows. J Fish Biol 50:407–417

Smith ME, Belk MC (2001) Risk assessment in western mosquitofish (Gambusia affinis): do multiple cues have additive effects? Behav Ecol Sociobiol 51:101– 107 Stallings, C.D., Koenig, C.C. (2010) Faunal communities of the Big Bend seagrass meadow. Progress Report submitted to the Florida Fish and Wildlife Conservation Commission. Project 08007.

Other sources: discussion with marine veterinarian at the Tampa Bay Aquarium who is knowledgeable of the techniques of handling marine fish. Review of articles and discussion with colleagues indicated that this research was not duplicative of on-going research.

Date of search: 01/05/2011

Years covered by search: 1994 to 2010

Keywords used / search strategy:

Key words used in 2005 search covering from 1994 to 2005 (number of records in parentheses) were jewfish (12), Epinephelus itajara (7), goliath grouper (3), caged fish (65), red grouper (54), Epinephelus morio (42), fish biopsy (0), fish tagging (48), gastric lavage (313). After scanning articles, we reviewed abstracts of 27. Of these, four entire articles were obtained and reviewed.

Key words used in 2011 search covering from 2006 to 2011 (number of records in parentheses) were jewfish (0), goliath grouper (6), caged fish (22), red grouper (5), fish biopsy (0), fish tagging (196), gastric lavage (76), fish sensory experiment (155). There were several articles that were relevant to our study (see below) but none provided any improvements to the methods we are using.

Describe the results of the search and whether alternatives were found:

Diet analysis: Most of the articles on gastric lavage pertained to fish much smaller than the ones that we are interested in sampling. The fish in articles ranged from less than a pound to one or two pounds. Discussions with colleagues on sampling stomach contents of large fish (e.g., goliath grouper we sample range from 10 to 450 pounds) indicated that the method we describe is the only feasible method of sampling stomach contents. Our 2007 search did not provide any additional significant information regarding techniques useful for other large fish we are studying. Lavage methods are gentle on the fish and appropriate for large fish and have been used by researchers for many years for sampling non-destructively. For fish with dangerous teeth (like sharks), lavage may be administered with a metal tube while the fish bites down on a large piece of wood.

Tagging methodologies: 2005 to 2011: Many papers described various methods of tagging, but these were highly specialized methods. For example, fish were tagged in deep-water trawls where it would be impossible to bring the fish to the surface. However, there were no new methods that would apply to our work.

If alternatives are available and not used, provide the ACUC with justification as to why alternatives cannot be used:

No alternatives exist. We use the best and least injurious and least stressful methods known. This is because we immediately release the fish and want them to survive and be in the best possible condition. Our research on juvenile goliath groupers has been published (Koenig et al. 2007 [Bull Mar Sci 80(3):567-586]). This work shows high recapture rates on over 1,100 tagged juveniles. We also demonstrated a survival rate of 95%, which is very high for any juvenile fish. Thus, this work supports our contention that our field tagging methods minimize stress and maximize survival.

Fish are different from land vertebrates in that distress comes from, among other things, being out of water. The surgical procedures undoubtedly cause some momentary discomfort, but compared to being out of their respiratory medium, the discomfort is likely relatively minor. It must be remembered that the surgery is being done at sea on commercial fishing vessels or research vessels. This venue introduces all sorts of variables (e.g., sea state, storms, water and air temperature, etc.) not present in a laboratory setting. We therefore concentrate on speed and aseptic technique to minimize any discomfort for the fish and to maximize survival.

100 19. ASSURANCE THAT THE TRAINING OF ANIMAL-RELATED PERSONNEL IS APPROPRIATE FOR THE EXPERIMENTS PROPOSED:

Please “x” to indicate “yes” or “no” All animal-using personnel on this project will YES__X__ NO*__ have received adequate training before beginning work.

*If no is indicated, please explain:

20. ASSURANCE OF OCCUPATIONAL HEALTH AND SAFETY:

All individuals who will have animal contact YES__ NO* under this Protocol have been enrolled in the *__ FSU Occupational Health and Safety Medical Monitoring Program for Vertebrate Animal X__ Users.

**If no is indicated, please explain: All individuals who will have contact with fish will take the necessary courses, but have not done so yet. All without proper training have been notified with the relevant information.

21. ASSURANCE ON REGULATIONS AND GUIDELINES: Please “x” to indicate “yes” or “no”

(a) The relevant portions of the Animal Welfare YES NO Act including the associated USDA regulations. __X __

__ (b) The NIH Guide for the Care and YES__X__ NO__ Use of Laboratory Animals.

101 Copies of the NIH Guide and the 2007 AVMA Guidelines on Euthanasia are available in the "Investigator's Guide for Animal Care and Use" distributed to all individuals using vertebrate animals. Additional copies are available from LAR (644-4262). Copies of the USDA regulations are available in the departmental offices of all animal users or in LAR.

* * NOTE * *

The Request for ACUC Review of Proposed Vertebrate Animal Use and Animal Use Description must be submitted to the ACUC Secretary ([email protected], 101 BRF, 32306-4341) with adequate lead-time to allow ACUC approval before animals are intended for use. The Committee meets monthly; the Animal Use Description must be submitted by the first working day of the month for ACUC action during that month. It is the PI's responsibility to provide all materials necessary for ACUC approval in time to meet any agency deadlines. All grants utilizing vertebrate animals must be approved by the ACUC prior to release of any funds by Sponsored Research. For those agencies with just-in-time animal use releases, investigators should take note that the ACUC is not under obligation to automatically grant approval; the ACUC approval process is independent and will undertake its normal course of action. If a funding agency needs a letter indicating approval of the proposed research, please submit SRS Animal Subjects Use Form (http://www.research.fsu.edu/contractsgrants/forms.html) to LAR along with a copy of the animal use section of the grant.

Dana M. Bethea, photographer and copyright owner, grants Lisa Hollensead the one-time usage rights of the two photos, 'Tagged Sawfish' and 'Mud Bay at Low Tide', for her MSc thesis, Monitoring movement patterns of juvenile smalltooth sawfish (Pristis pectinata) using acoustic monitoring and tracking in a nursery habitat in southwest Florida, with the photo credit of Dana M. Bethea, NOAA Fisheries Panama City Laboratory.

102

REFERENCES

Allison, P.D. (2001) Logistic Regression using the SAS System: Theory and Application. Cary, NC: SAS Institute Inc.

Bigelow, H.B. and Schroeder W.C. (1953) Fishes of the Western North Atlantic, Part 2. Sawfishes, Guitarfishes, Skates, Rays and Chimaeriods. Memoir of the Sears Foundation for Marine Research. (Yale University: New Haven, CT)

Burt, W.H. (1949) Territoriality and home range concepts as applied to mammals. Journal of Mammalogy 30(1):25-27

Carlson, H.R, and Straty, R.R. (1981) Habitat and Nursery grounds of Pacific Rockfish, Seabastes spp., in Rocky Coastal Areas of Southeastern Alaska. Marine Fisheries Review 43(7) 13-19

Carlson, J.K., Heupel, M.R., Bethea, D.M, Hollensead, L.D. (2008) Coastal habitat use and residency of juvenile Atlantic sharpnose sharks (Rhizoprionodon terraenovae). Estuaries and Coasts 31: 931-940

Cartamil, D.P., Vaudo, J.J., Lowe, C.G. Wetherbee, B.M., Holland, K.N. (2003) Diel movement patterns of the Hawaiian stingray, Dasyatis lata: implications for ecological interactions between sympatric elasmobranch species. Marine Biology 142: 841-847

Castro, J.I. (1993) The shark nursery of Bulls Bay, South Carolina, with a review of the shark nurseries of the southeastern coast of the United States. Environmental Biology of Fishes 38: 37-48

Chapman, D.D, Simpfendorfer, C.A., Wiley, T.R, Poulakis, G.R. Curtis, C., Tringali, M., Carlson, J.K., Feldheim, K.A. (2011) Genetic diversity despite population collapse in a critically endangered marine fish: the smalltooth sawfish (Pristis pectinata). Journal of Heredity 102(6):643-652

Cody, M.L. (1985) Habitat Selection in Birds. Academic Press Inc. New York, NY, USA 4-27

Curtis, T.H, Adams, D.H., Burgess, G.H. (2011) Seasonal Distribution and Habitat Associations of Bull Sharks in the Indian River Lagoon, Florida: A 30 year Synthesis. Transactions of the American Fisheries Society 140(5): 1213-1226

Dahlgren, C.P., Kellison, G.T., Adams, A.J., Gillanders, B.M., Kendall, M.S., Layman, C.A., Ley, J.A., Nagelkerken, I., Serafy, J.E. (2006) Marine nurseries and effective juvenile habitats: concepts and applications. Marine Ecology Progress Series 312: 291- 295

103 DiGirolamo, A.L., Gruber, S.H., Pomory, C., Bennett, W.A. (2012) Diel temperature patterns of juvenile lemon sharks, Negaprion brevirostris, in a shallow-water nursery. Journal of Fish Biology 80: 1436-1448

Endangered Species Act (1973) 16 USC Sec. 1531-1544

Farrugia, T.J., Espinoza, M., Lowe, C.G. (2011) Abundance, habitat use and movement patterns of shovelnose guitarfish (Rhinobatos productus) in a restored southern California estuary. Marine and Freshwater Research 62:648-657

Fortin, M.J. and Dale, M.R.T. (2005) Spatial Analysis A guide for ecologists. University Press, Cambridge, United Kingdom 1-31

Gitzen, R.A., Millspaugh, J.J., Kernohan, B.J. (2006) Bandwidth selection for fixed- kernel analysis of animal utilization distributions. Journal of Wildlife Management 70(5):1334-1344

Godin, J.J. (1997) Behavioural Ecology of Teleost Fishes. Oxford University Press. New York, NY, USA 35-44

Gosling, L.M., Sutherland, W.J. (2000) Behaviour and Conservation. Cambridge University Press Cambridge, UK

Grubbs, R.D, Musick, J.A., Conrath, C.L., Romine, J.G. (2007) Long-term movements, migration and temporal delineation of a summer nursery for juvenile sandbar sharks in the Chesapeake Bay region, American Fisheries Society Symposium 50:87-107

Heck, K.L., Hays, G., Orth, R.J. (2003) Critical evaluation of the nursery role hypothesis for seagrass meadows. Marine Ecology Progress Series 253: 123-136

Heithaus M.R., Dill, L.M., Marshall, G.J., Behleier B.M. (2002) Habitat use and foraging behaviour of tiger sharts (Galeocerdo cuvier) in a seagrass ecosystem. Marine Biology 140: 237-248

Heupel, M.R. and Simpfendorfer, C.A. (2002) Estimation of mortality of juvenile blacktip sharks, Carcharhinus limbatus, within a nursery area using telemetry data. Canadian Journal of Fisheries and Aquatic Sciences 59(4):624-632

Heupel, M.R, Simpfendorger, C.A., Lowe, C.G. (2005) Passive Acoustic Telemetry Techonology: Current Applications and Future Directions. Results of the VR2 workshop held on Catalina Island Nov 28 – Dec 1 2005. Mote Technical Report 1066

Heupel, M.R., Carlson, J.K., Simpfendorfer, C.A. (2007) Shark nursery areas: concepts, definition, characterization and assumptions. Marine Ecology Progress Series 337:287- 297

104 Holland, K.N., Wetherbee, B.M., Peterson, J.D., Lowe, C.G. (1993) Movements and Distribution of Hammerhead Shark Pups on their Natal Grounds. Copeia 2: 495-502

IUCN. (2012) IUCN Red List of Threatened Species version 2012.1

Johnson, D.H. (1980) The comparison of usage and availability measurements for evaluating resource preference. Ecology 61: 65-71

Knip, D.M., Heupel, M.R., Simpfendorfer, C.A. (2010) Sharks in nearshore environments: models, importance, and consequences. Marine Ecology Progress Series 402: 1-11

Knip, D.M., Heupel, M.R., Simpfendorfer, C.A., Tobin, A.J., Moloney, J. (2011). Ontogenetic shifts in movement and habitat use of juvenile pigeye sharks Carcharhinus amboinensis in a tropical nearshore region. Marine Ecology Progress Series 425: 233-246

Laegdsgaard, P. and Johnson, C.R. (1995) Mangrove habitats as nurseries: unique assemblages of juvenile fish in subtropical mangroves in eastern Australia. Marine Ecological Progress Series 126: 67-81

Layman, C.A. (2000) Fish Assemblage Structure of the Shallow Ocean Surf-Zone on the Eastern Shore of Virginia Barrier Islands. Estuarine, Coastal and Shelf Science 51: 210- 213

Manly, B., McDonald, L., Thomas, D. (1993) Resource selection by Animals. Chapmand and Hall. London, UK

Meyer, C.G., Holland, K.N., Wetherbee, B.M., Lowe, C.G. (2000) Movement patterns, habitat utilization, home range size and site fidelity of whitesaddle goatfish, Parupeneus porphreus, in a marine reserve

Morrissey, J.F and Gruber S.H. (1993a) Home range of juvenile lemon sharks, Negaprion brevirostris. Copeia 2:425-434

Morrissey J.F. and Gruber S.H. (1993b) Habitat selection by juvenile lemon sharks, Negaprion brevirostris. Environmental Biology of Fishes 38:311-319

National Marine Fisheries Service. (2009) Recovery Plan for Smalltooth Sawfish (Pristis pectinata). Prepared by the Smalltooth Sawfish Recovery Team for the National Marine Fisheries Service, Silver Spring, Maryland.

Pape, O.L, Holley, J., Guérault, D., Désaunay, Y. (2003) Quality of Coastal and estuarine essential fish habitat based on the size of juvenile common sole (Solea solea) Estuarine, Coastal and Shelf Science 58: 793-803

105 Phillips, N.M., Whitty, J.M., Morgan, D.L., Chaplin, J.A. (2009) Freshwater sawfish (Pristis microdon) movements and population demographics in the Fitzroy River, Western Australia and genetic analysis of P. microdon and Pristis zijsron. Murdoch University. Centre for Fish and Fisheries Research, Western Australia

Phillips, N.M., Chaplin, J.A., Morgan, D.L, Peverell, S.C. (2011) Population genetic structure and genetic diversity of three critically endangered Prisitis sawfishes in Australian waters. Marine Biology 158:903-915

Poulakis, G.H. and Seitz J.C.(2004) Recent occurrence of the Smalltooth Sawfish, Pristis pectinata (Elasmobranchimorphi: Pristidae) in Florida Bay and the Florida Keys, with comments on sawfish ecology. Florida Scientist 67:27-35

Poulakis, G.R., Stevens, P.W., Timmers, A.A., Wiley, T.R., Simpfendorfer, C.A. (2011) Abiotic affinities and spatiotemporal distribution of the endangered smalltooth sawfish, Pristis pectinata, in a south-western Florida nursery. Marine and Freshwater Research 62(10):1165-1177

Primack, R.B. (2006) Essentials of Conservation Biology Fourth Edition. Sinauer Associates, Inc. Publishers. Sunderland, Massachusetts U.S.A.

R Development Core Team (2011). R: A language and environment for statistical computing. R Foundation for Statistical Computing,Vienna, Austria. ISBN 3-900051-07- 0, URL http://www.R-project.org/

Rodgers, A.R., Carr, A.P., Beyer, H.L., Smith, L., Kie, J.G. (2007) HRT: Home Range Tools for ArcGIS. Version 1.1 Ontario Ministry of Natural Resources, Centre for Northern Forest Ecosystem Research, Thunder Bay, Ontario, Canada

Ross, S.W. and Lancaster, J.E. (2002) Movements and site fidelity of two juvenile fish species using surf zone nursery habitats along the southeastern North Carolina coast. Environmental Biology of Fishes 63: 161-172

SAS software, Version 8.1 of the SAS System for Windows. Copyright © 2000 SAS Institute Inc. SAS and all other SAS Institute Inc. product or service names are registered trademarks of trademarks of SAS Institute Inc., Cary, NC, USA

Seitz, J.C. and Poulakis, G.R. (2002) Recent occurrence of Sawfishes (Elasmobranchimorphi: Pristidae) along the Southwest coast of Florida (USA). Florida Scientist 65:256-266

Simpfendorfer, C.A. (2000) Predicting population recovery rates for endangered western Atlantic sawfishes using demographic analysis. Environmental Biology of Fishes 58:371- 377

106 Simpfendorfer, C.A. and Wiley, T.R. (2005) Determination of the distribution of Florida’s remnant sawfish population and identification of areas critical to their conservation. Final Report. Florida Fish and Wildlife Conservation Commission, Tallahassee, Florida.

Simpfendorfer, C.A. 2006. Movement and habitat use of Smalltooth Sawfish. Final Report. Mote Marine Laboratory Technical Report 1070

Simpfendorfer, C.A., Poulakis, G.R., O’Donnell, P.M, Wiley, T.R., (2008) Growth rates of juvenile smalltooth sawfish Pristis pectinata Latham in the western Atlantic. Journal of Fish Biology 72:711-723

Simpfendorfer, C.A. Wiley, T.R., Yeiser, B.G. (2010) Improving conservation planning for an endangered sawfish using data from acoustic telemetry. Biological Conservation 143:1460-1469

Simpfendorfer, C.A., Yeiser, B.G., Wiley, T.R., Poulakis, G.R., Stevens, P.W., Heupel, M.R. (2011). Environmental influences on the spatial ecology of juvenile smalltooth sawfish (Pristis pectinata): Results from acoustic monitoring. PLoS ONE 6(2): e16918. doi:10.1371/journal.pone.0016918

Snelson, F.F. and Williams, S.E. (1981) Notes on the Occurrence, Distribution, and Biology of Elasmobranch Fishes in the Indian River Lagoon System, Florida. Estuaries. 4(2):11-120

Springer, S. (1967) Social organization of shark populations. Sharks, Skates, and Rays. Gilbert, P.W., Mathewson, R.F. and Rall, D.P. eds.

Stevens, J.D., McAuley, R.B., Simpfendorfer, C.A., Pillans, R.D. 2008. Spatial distribution and habitat utilization of sawfish (Pristis spp) in relation to fishing in northern Australia. A report to Department of the Environment, Water, Heritage and the Arts

White, G.C and Garrott, R.A. (1990) Analysis of Wildlife Radio-Tracking Data. Academic Press, Inc. San Diego California

Whitty, J.M., Morgan, D.L., Peverell, S.C. Thorburn, D.C. Beatty, S.J. (2009) Ontogenetic depth partitioning by juvenile freshwater sawfish (Pristis microdon: Pristidae) in a riverine environment. Marine and Freshwater Research 60(4): 306-316

Wiley, T.R. and Simpfendorfer, C.A. (2007) The ecology of elasmobranches occurring in the Everglades National Park, Florida: implications for conservation and management. Bulleting of Marine Science 80(1):171-189

107 Wiley, T.R. and Simpfendorfer, C.A. (2010) Using public encounter data to direct recovery efforts for the endangered smalltooth sawfish Pristis pectinata. Inter-Research Endangered Species Research 12(3) 179-191

Worton, B.J. (1989) Kernel methods for estimating the utilization distribution in Home- Range Studies. Ecology 70:164-168

Worton, B.J. (1995) Using Monte Carlo simulation to evaluate kernel-based home range estimators. Journal of Wildlife Management 59:794-800

108 BIOGRAPHICAL SKETCH

I was born in 1983 in Louisville, Kentucky. After I graduated from high school I moved to Tallahassee, Florida where I enrolled as an undergraduate at Florida State University (FSU). I graduated from FSU with a bachelors of science in biology with a minor in chemistry and a certificate in marine science in 2004. After graduation I was hired as biological technician at the NOAA National Marine Fisheries Service lab in Panama City, Florida. While there, I worked on a variety of projects. For two years, I worked on a telemetry study of Gulf Sturgeon (Acipenser oxyrinchus desotoi) looking at habitat use after the land fall of Hurricane Ivan in Pensacola Bay. I also worked on the lab’s long term abundance survey of coastal juvenile sharks (GULFSPAN). After my work with the Gulf Sturgeon project was over, I began working as an assistant observer coordinator. I helped organize observers assigned to the shark bottom long line fishery and the shark gillnet fishery. This included working up scientific samples collected by observers as well as entering and helping analyze annual catch data for internal reports. After another two years working as the assistant observer coordinator I decided to go back to school. I started work towards my master’s degree at FSU under the direction of R. Dean Grubbs and Don Levitan in the fall of 2008. My project focus was examining movement patterns and habitat use of juvenile smalltooth sawfish (Pristis pectinata) within a relatively pristine nursery habitat in the northern everglades. Since this species was the first domestic fish species listed on the United States Endangered Species list, my project not only had an interest in basic ecological questions about juvenile P. pectinata behavior, but also a strong conservation component.

109