1 EVALUATION OF HATCHERY REARED LARGEMOUTH BASS USING RADIO TELEMETRY By BRANDON C. THOMPSON A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2012
2 2012 Brandon C. Thompson
3 To my patient and loving wife whose support was critical in the writing of my thesis
4 ACKNOWLEDGMENTS Many thanks to my advisor and committee chair, Dr. Mike Allen; Dr. Tom Frazier and Wesley Porak for serving as my committee members; and Bill Johnson, Nick Trippel, and Steve Crawford for assistance in designing the study and providing advice throughout. Rick Stout, Michael Matthews, Derek Piotrowicz, Josh Sakmar, and the rest of the staff at the Florida Bass Conservation Center (FBCC) raised all the hatchery largemouth bass for this study. Kris Nault, Scott Bisping and Brad Fontaine spent long days in extreme conditions tracking fish on Lake Carlton during the ra dio telemetry study. Erin Leone provided assistance with data analysis while John Benton and Ryan Butrym helped with mapping and analysis of GIS data. Numerous other Florida Fish and Wildlife agency staff helped with various field and hatchery components of this study.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 6 LIST OF FIGURES ................................ ................................ ................................ .......... 7 ABSTRACT ................................ ................................ ................................ ..................... 8 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 10 2 TAG EFFECTS ON HATCHERY LARGEMOUTH BASS ................................ ....... 13 Intro ................................ ................................ ................................ ......................... 13 Methods ................................ ................................ ................................ .................. 15 Results ................................ ................................ ................................ .................... 19 Discussion ................................ ................................ ................................ .............. 21 3 EVACUATION RATES OF RADIO TRANSMITTERS BY PREDATORS ................ 30 Intro ................................ ................................ ................................ ......................... 30 Methods ................................ ................................ ................................ .................. 32 Results and Discussion ................................ ................................ ........................... 34 4 MOVEMENT, SURVIVAL, AND HABITAT USE OF HATCHERY REARED VERSUS WILD JUVENILE LARGEMOUTH BASS ................................ ................ 42 Intro ................................ ................................ ................................ ......................... 42 Methods ................................ ................................ ................................ .................. 44 Results ................................ ................................ ................................ .................... 53 Discussion ................................ ................................ ................................ .............. 62 5 CONCLUSION ................................ ................................ ................................ ........ 88 LIST OF REFERENCES ................................ ................................ ............................... 90 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 97
6 LIST OF TABLES Table page 2 1 Summary of initial fish length (mm TL), weight (g), sample size (N), day of growth measurement, and percent length and weight gained per day. .............. 26 2 2 Mean total length (mm), weight (g), standard error, and number of fish eaten in predator avoidance tests. ................................ ................................ ................ 27 3 1 Comparison of Prey Size, Predator size, and Evacuation times. ........................... 38 3 2 Parameter estimates for a model that predicts the probability of a predator evacuating a tag at a given time after consumption. ................................ ........... 39 4 1 Number of fish radio tagged, total length (mm), weight (g) and tag to body weight ratio (%) for hatchery reared fish (2009 and 2010) and wild fish. ............ 70 4 2 Fate of radio tagged hatchery and wild stocked juvenile largemouth bass. ........... 71 4 3 Dispersal and movement for radio tagged hatchery (2009 and 2010) and wild (2010) advanced fi ngerling largemouth bass ................................ ...................... 72 4 4 Initial and ending total lengths (mm) and weights (g) for radio tagged hatchery and wild largemouth bass. ................................ ................................ .................. 73
7 LIST OF FIGURES Fig ure page 2 1 Relative growth rate (percent body weight gained per day) at 7 and 21 days in 2009 and 30 days in 2010 for surgically implanted radio tagged fish. ................ 28 2 2 Proportion of tagged and control (untagged) advanced fingerling hatchery largemouth bass that were consumed in predator avoidance trials. ................... 29 3 1 Perc ent of radio tags remaining in the gut of a predator. ................................ ....... 40 3 2 A photo illustrating the curling effect on the antennae of a radio transmitter .......... 41 4 1 Illustration of primary study area depicting the geographical location and schematic of the water bodies that comprise the Oklawaha Chain of Lakes. ..... 74 4 2 Satellite image showing the canals connecting Lake Carlton to Lake Beauclair. ... 75 4 3 Example of the detailed Lake Carlton vegetation map developed using ................................ ................................ ........................... 76 4 4 Photograph of the surgical procedure used to implant a radio transmitter in an advanced fingerling largemouth bass. ................................ ................................ 77 4 5 Proportion of tags that remained o perational over a 34 day period ....................... 78 4 6 Persistence of 30 control radio tags allowed to expire in the lab. ........................... 79 4 7 Fate of radio tagged hatchery la rgemouth bass during the 2009 study period. ..... 80 4 8 Survival propabilities over 14 days for radio tagged hatchery largemouth bass stocked in Lake Carlton in 2009. ................................ ................................ ........ 81 4 9 Fate of radio tagged hatchery largemouth bass throughout the 2010 study period after being stocked into Lake Carlton. ................................ ..................... 82 4 10 Survival propabilities over 30 days for radio tagged hatchery and wild largemouth bass stocked in Lake Carlton in 2010. ................................ ............. 83 4 11 Dispersal of radio tagged hatchery and wild largemouth bass 84 4 12 Average movement per day of radio tagged hatchery and wild largemouth bass. ................................ ................................ ................................ ................... 85 4 13 Habitat use of hatchery and wild radio tagged largemouth bass. ......................... 86 4 14 Comparison of percent body weight (a) and length (b) gained per day between radio tagged wild (n = 7) and hatchery bass. ................................ ...................... 87
8 Abstract of Thesis Presented to the Gradua te School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science EVALUATION OF HATCHERY REARED LARGEMOUTH BASS USING RADIO TELEMETRY By Brandon C. Thompson August 2012 Chair: Name: Mi ke Allen Major: Fisheries and Aquatic Sciences The use of supplemental stocking of largemouth bass is a common management practice to augment poor natural recruitment. W e have a very poor understanding about how changes in behavior from living in an art ificial culture envir onment at the behavior after being released into a natural lake environment. Miniaturization of radio transmitters have recently allowed researchers to study stocked juvenile fish and make inferences ab out the entire stocked population based on tagged individuals. The short term effects of surgically inserting dummy microradio transmitters (1.2 to reared largemouth bass was investigated by evaluating growth a nd predator avoidance differences with untagged control fish. Some mortality of tagged fish and slight impairment of growth was observed for experiments run to 21 and 30 d. At the conclusion of each growth experiment, almost all fish shed their suture incisions were completely healed and no transmitter s were expelled. During predation trials, tagged fish were eaten at the same rate as controls, suggesting that vulnerability to predation was not affected by the presence of radio tags. Although some mo rtality and growth
9 impairment can be expected, this study suggests that surgical implantation of radio transmitters can be a valid tagging technique to examine the behavior of juvenile largemouth bass in short term field studies. A laboratory study was c onducted to determine how radio transmitters (0.3 g) implanted in juvenile largemouth bass that were eaten by adult largemouth bass were evacuated relative to predator size, prey size and water temperature. All transmitters ingested by the predators were evacuated within 84 h. The tags passed through the intestine independently of the size of predators or prey, whereas water temperature was directly related to the timing of evacuation. A field study using radio telemetry investigated the post stocking su rvival and behavior of hatchery reared advanced fingerling (90 120 mmTL) largemouth bass in Lake Carlton, Florida. We also tested the hypothesis that behavior and survival of domesticated hatchery stocked fish differs from wild fish. In 2009, we inserted radio transmitters in 50 hatchery stocked fish and in 2010, 30 hatchery fish were tagged along with 20 wild reared fish. When compared directly to tagged wild fish (n = 20), tagged hatchery stocked fish (n = 30) exh ibited lower survival, higher movement hig her proportion of offshore use, and slower growth All mortality observed for hatchery fish resulted from fish and bird predation and the majority occurred in the first seven days. Results of this study suggest that domestication affects from the hat chery rearing process resulted in reduced fitness of hatchery largemouth bass immediately following their release into the lake. I recommend acclimating advanced fingerling largemouth bass to predators and live prey prior to release to reduce initial mora lity after they are stocked.
10 CHAPTER 1 INTRODUCTION Largemouth bass Micropterus salmoides are one of the most popular and most intensively managed sport fish in North America. Use of supplemental stocking of largemouth bass is a common management pract ice to augment poor natural recruitment (Boxrucker 1986; Buynak and Mitchell 1999; Mesing et al. 2008). To provide a benefit to the recipient lake, stocked fish must survive and contribute to the natural population. When stocking evaluations have been co nducted for largemouth bass, contribution to the natural population has commonly been low (Boxrucker 1986; Crawford and Wicker 1987; Buynak and Mitchell 1999; Buckmeier and Betsill 2002; Hoxmeier and Wahl 2002; Porak et al. 2002; Hoffman and Bettoli 2005). Several studies have investigated factors influencing stocking success and a variety of factors such as predation, starvation and health issues have contributed to low survival (Hoxmeier and Wahl 2002; Porak et al. 2002; Diana and Wahl 2009). Hoxmeier a nd Wahl (2002) found that predation rates from adult largemouth bass were an important source of mortality on stocked advanced fingerling bass. Mesing et al. (2008) also found that advanced fingerling bass reared on natural prey had contributions up to 40 % the following fall in Lake Talquin, Florida. Stocking advanced size fingerlings has been shown to improve stocking success (Buynak and Mitchell 1999; Colvin et al. 2008; Mesing et al. 2008; Diana and Wahl 2009) as larger fish are less vulnerable to pred ation (Loska 1982). Hatchery reared fish exhibit deficits in survival behaviors, such as predator avoidance, foraging, and migration, because of the unnatural conditions in which they are reared (Brown and Laland 2001). Domestication effects of pellet reared hatchery
11 largemouth bass raised at the Florida Bass Conservation Center (FBCC) have included altered feeding kinematics and strike modes; difficulty transitioning from artificial diets to live prey; excessive vulnerability to angling; health and fit ness issues; and poor post release survival (Porak et al. 2002, Patel Wintzer 2004, Pouder et al. 2010). Improvements in the culture of advanced fingerlings prior to release include transition to live prey, improved formulated diet (Cardeilhac et al. 2008 ), and early spawned fish to expedite the timing of release (2010 stocking). These improvements could potentially improve survival and anticipation of a more successful stocking program and has warranted additional research. Limited studies have foc used on advanced fingerling survival immediately after release (Pouder et al. 2010) and we have a very poor understanding about how changes in behavior from living in an artificial culture environment at the hatchery affects their behavior after being rele ased into a natural lake environment. Technological advances in radio telemetry equipment have made it possible to study the behavior of very small animals due to development and production of very small transmitters. For the first time ever, we had the opportunity to study the initial dispersal, movement, habitat use, and survival of hatchery largemouth bass after their release into a lake using radio telemetry techniques. We were also able to compare their behavior and survival to that of wild fish. T his information will help scientists, fishery managers, and hatchery personnel understand some of the mechanisms that affect post stocking surviv al of hatchery fish, which may lead to improvements in fish culture, handling, and release protocol.
12 Studies c onducted within this thesis are divided into three chapters (Chapter 2 Chapter 4). In Chapter 2, we determined the effects of surgically implanting transmitters on growth, survival, and predator avoidance of small bass to validate whether or not tagged an imals adequately represent untagged animals in field telemetry studies. In Chapter 3, we developed an evacuation model which can be used to predict the the radio ta gged juvenile bass. This information was useful during field telemetry studies for determining whether a fish location was a small radio tagged bass or a predator that had eaten the radio tagged. In Chapter 4, our study objectives were to compare the beh avior and survival of hatchery reared largemouth bass to similar size wild juvenile bass after being stocked into a lake.
13 CHAPTER 2 TAG EFFECTS ON HATCH ERY LARGEMOUTH BASS Intro A major concern in any biotelemetry study is the assumption that res ults obtained using tagged animals accurately represents untagged animals, which allows inferences to the entire population. The presence of the tag or the surgical process could potentially affect the movement, behavior, or the chance of predation, which would bias behavior should be evaluated. With the miniaturization of radio transmitters, extensive research has been conducted on juveniles to test the effects of trans mitter presence on fish. Prior to equipped with transmitters that weigh more than 2% o 1996). Some recent studies, however, have successfully tagged fish outside this range, up to 10% in some species, without adverse affects (Brown et al. 1999; Jepsen et al. 2002). Swimming performance, growth rates, an d predator avoidance are commonly used to evaluate the effects of radio and acoustic transmitters on fish behavior and biology. Many studies have shown no significant differences in swimming performance of fish implanted with transmitters and controls (Co te et al. 1999; Robertson 2003; Anglea et al. 2004). Transmitter effect on juvenile fish growth rates has yielded variable results. After 30 days, Frost et al. (2010) found that growth rates in Chinook salmon Oncorhynchus tshawytscha were significantly l ess in surgery fish than in controls when
14 transmitters were used that weighed 2.6 growth was not significantly different between controls and tagged rainbow trout Oncorhynchus mykiss (Lucas 1989), subyearling Ch inook salmon (Martinelli et al. 1998), and juvenile Atlantic cod Gadus morhua (Cote et al. 1999). Anglea et al. (2004) found that that juvenile Chinook salmon tagged with transmitters did not result in greater predation susceptibility than untagged fish i n a laboratory experiment. Despite the importance of studying the early life history of hatchery and wild largemouth bass Micropterus salmoides the majority of the research associated with tag effects on juvenile fish has concentrated on salmonids. Di fferences in physiological characteristics in largemouth bass and the environment in which they inhabit could give significantly different results than that found for other species. Largemouth bass occupy much warmer systems, and elevated water temperatur e has been shown to have increased incision healing time which poses a greater risk of infection when fish were tagged in relatively warm water (Knight and Lasee 1996; Walsh et al. 2006). Cooke et al. (2003) studied the effects of suture material on incis ion healing of juvenile largemouth bass (140 0.8 mm total length [TL]), and found that successful implantation of micro transmitters can be achieved. I investigated tagging advanced fingerling largemouth bass produced at the Florida Bass Conservation Center (FBCC), which are defined as fish reared to a mean of 100 mm (TL) before stocking. These fish represent some of the smallest largemouth bass to be implanted with radio transmitters. My objective was to determine the effects of surgically implantin g radio transmitters on the growth and predator avoidance of advanced fingerling largemouth bass.
15 Methods Fish acquisition and surgical procedures We conducted two independent studies starting July 24, 2009 and March 28, 2010 to evaluate tagged bass at the FBCC during both study years. In order to avoid excessive tag to body mass ratio for surgery fish, largemouth bass greater than 90 mm were chosen at random from hatchery raceways for each experiment. In 2009 and 2010, test fish were held in tanks w here a constant flow of water was supplied at 22 to each trial by feeding to satiation three times per week. Feed was withheld 48 hours prior to and 24 hours after surgeries. Fish tagged in this study were similar in size to fish that would typically be stocked as advanced fingerlings. Dummy tags used were exact replicates (but no working parts) of the A2414 transmitter (Advanced Telemetry Systems, Isanti, Minnesota). The tag weight was 0.24 g in air, meas ured 5 x 12 mm and represented between 1.2% and 2.7% (mean = 1.7%) of the body weight of the fish. The trailing antenna wire length was 10 cm and consisted of a flexible braided metal. Surgical procedures were similar to Adams et al. (1998b), although incision and antenna placement were modified for largemouth bass (Cooke et al. 2003). Prior to surgeries, each fish was anesthetized in a 70 mg L 1 tricaine methanesulfonate (MS 222) bath, measured (mm TL), and weighed (0.1 gram). The fish was then trans ferred ventral side up to a surgical foam pad customized to fit the form of an advanced continuous water flow of 30 mg L 1 MS 222. A 9 mm long incision was made just off th e mid ventral line posterior to the pelvic girdle using a miniature, 3 mm blade scalpel. A modified shielded needle technique (Ross and Kleiner 1982) provided a guide for the
16 transmitter antenna to pass through the body wall. Once the antenna was threade d through the catheter, the catheter was removed and the transmitter (or dummy tag) was gently inserted into the body cavity. Due to the small incision site, the incision was closed with one Ethicon (5 0 taper RB 1 needle) coated vicryl absorbable suture. Tagged fish were placed in a recovery bath until equilibrium was restored. Surgery time averaged 149 seconds from when the fish was removed from the anesthetic bath until transferred to the recovery bath Growth A 21 day growth experiment was initiate d on July 24, 2009 and similar experiment was conducted for 30 days on March 28, 2010. Fish were randomly selected for the surgery and control treatment groups and all fish were anaesthetized in a 70 mg L 1 MS 222 bath, measured (mm TL), weighed (0.1 g), and pl aced in research tanks (Table 2 1). Fish in 2009 were uniquely identified by clipping a separate or series of dorsal spines as a marker to identify individual fish so that growth for these individuals could be calculated during the study period. In the 2010 growth study, a mean growth rate was calculated for the control and treatment groups in each tank using the mean size (TL, mm) of all fish at the beginning and the end of the experiment. I used replicate tanks of fish as the experimental unit, with fish assigned randomly to tanks and growth measured as the mean change in length for each tank. To estimate growth differences between surgery and control fish, and to test for tank effects, 10 tagged and 10 control fish were placed in each of four separate, identical tanks (n = 80 fish per experiment each year). Feeding resumed 24 hours after all surgeries were completed, and then all fish were fed to satiation each day during the experiment each year. In the 2009 experiment, all fish were removed measured, and
17 weighed on day 7 and day 21. In the 2010 experiment, all fish were removed, measured, and weighed on day 30. Condition of surgery fish was assessed at the conclusion of the trial by examining the external structures, the incision site, an d the exit site of the antenna for redness and swelling. Abnormalities were graded as mild or severe as observed by visual examination. Mortality was recorded for both test groups and tag expulsion was monitored for tagged fish in both years. Predator avoidance We tested the effects of surgical procedures and radio transmitter presence on the vulnerability to predators of advanced fingerling largemouth bass. Predation experiments were conducted between June 22, 2010 and August 6, 2010 in an outdoor 9 ,085 L (3.7 m diameter, 0.84 m depth) circular tank. This tank was supplied with a continuous flow of aerated well water and cooled to a consistent temperature (22 24 o C) throughout the experiment. Four rings of artificial plants (80 x 60 cm) were added t o the tank, which provided test fish with approximately 15% escapement cover by volume. Six adult largemouth bass averaging 408 mm TL (range = 340 505 mm TL) were captured by electrofishing in Lake Eustis, Florida and were selected as the predators for th is experiment. They were acclimated to the tank for one month before the experiment began. Predators were fed juvenile largemouth bass and Seminole killifish Fundulus seminolis before and between trials, but they were not fed 48 hours prior to the introd uction of prey fish for each trial. Although the same predators were used for each trial, we assumed if any changes if predator efficiency occurred, it would not affect the proportion of each treatment consumed. We also tested the correlation between suc cessive trials and time of trial duration to determine if any learning or increase in feeding efficiency occurred.
18 For each trial, advanced fingerling hatchery largemouth bass (n = 20) were randomly assigned one of two treatments, where they were eithe r measured and had radio tags surgically implanted (tagged) or handled only (control). All fish were measured ( mm TL) and weighed (g ), and then given a 24 hour recovery period. To investigate differences in predator avoidance, we conducted 10 separate tr ials, each consisting of 10 tagged and 10 control fish to be released into the predator test tank. To begin a trial, predators were crowded to one side of the tank with a fine mesh seine and test fish (n = 20) were then added to the opposite side of the t ank and given a 5 minute acclimation period. Direct observations were made as often as necessary to allow near 50% predation of all test fish to occur. After sufficient predation was achieved, test fish were crowded to one end of the tank with a seine an d then removed. Survivors from each trial were transferred to an indoor 1060 L tank for 24 hours to be observed for delayed mortality due to predation attempts from the ad ult largemouth bass predators. Data analysis For all analyses in this study, a sig nificance level of P < 0.05 was used. Initial lengths and weights between treatments for all experiments were compared using t test assuming equal variance when appropriate. All analyses were performed using SAS v 9.2 (Cary, NC) and parametric model assu mptions of normality and homogeneity of variance were visually examined to ensure that they were properly met. Relative growth rates were expressed as percent body weight gained per day (Busacker et al. 1990). Relative growth rates for surviving tagged and control fish were calculated on day 7 and 21 in 2009, and day 30 in 2010. We built linear mixed models assuming a completely randomized block design to account for potential tank effects on
19 growth in both experiments. When tank effects were estimate d at 0, data were pooled and we used a t test to evaluate differences between tagged and control fish. In 2010, because control fish were not individually marked, mean relative growth rate was calculated for each tank and treatments were compared using a linear mixed model with treatment (tag, control) as the fixed factor and tank as the random factor (to account for both treatments being contained in the same tank). Differences in the proportion of fish consumed between both test groups for predation tr ials were compared using a generalized linear mixed model assuming a binary distribution and a random effect for trial. We tested for differences in the proportion of fish eaten based on tag treatment, length, weight, and the interaction of length and wei ght with tagging on predation of fish. Results Growth During growth experiments in both years, substantial mortality was observed. In 2009, surgically implanted fish experienced 32.5% mortality by day 7, but no additional mortality occurred by day 21. T wo control fish died (5%) from day 7 to day 21 and all mortalities were excluded from the growth analysis. In 2010, only 19 of 40 (47.5%) tagged fish and 24 of 40 (60%) control fish survived the 30 day experiment. The majority (89%) of the mortality occu rred in the first 7 days of the experiment and 100% occurred in the first 14 days. One of the four replicate tanks suffered 100% mortality of both tagged and untagged fish and therefore, only three tanks were used in the analysis of growth for 2010. Necro psies of all mortalities revealed columnaris Flexibacter columnaris, a fish disease common to cultured environments, and damage to the intestines from surgery of tagged fish to be the primary causes of death. All f ish surviving their respective experiment visually appeared healthy, robust, and unaffected
20 by surgery or disease. After 21 days in 2009 and 30 days in 2010, all incisions were completely healed and there was little to no evidence of internal or external infections based on gross examination. A ll sutures were completely absorbed by the end of each respective experiment and n o transmitter expulsion was observed. experiment. Fish ranged from 93 to 118 mm TL and 8.8 t o 20.1 g in weight for 2009. In 2010, fish ranged from 96 to 121 mm TL and 8.7 to 19.4 g in weight. The initial average length of tagged and control fish ( Table 2 1) that survived each experiment were not significantly different in 2009 (t 65 = 0.11, P = 0.91) or 2010 (t 47 = 1.7, P = 0.28). There was also no difference in initial weight of tagged and control fish in 2009 (t 65 = 0.33, P = 0.74) and 2010 (t 47 = 1.5, P = 0.28). The mean relative growth rate was lower for tagged than untagged fish in both years ( Table 2 1 ). In 2009, relative growth rates for fish that survived to 21 days ranged from 0.05 to 3.42%TL per day for tagged fish and 1.05 to 3.91% for untagged fish. In 2010, relative growth rates at day 30 ranged from 0.80 to 2.09% TL per day for tagged fish and averaged 1.70 to 2.01% for the three remaining tanks of control fish. The mean relative growth rate of tagged fish was significantly lower than controls at day 7 (t 65 = 2.57, P = 0.012) and day 21 (t 63 = 5.74, P < 0.01) in 2009, and day 3 0 (F 1,2 = 38.35, P = 0.025) in 2010 ( Figure 2 1 ). Predator avoidance Fish sizes were similar between tagged and untagged fish in the predation experiments (Table 2 2 ). Length of tagged advanced fingerling bass ranged from 98 to 121 mm TL (109.5 0 .5 mm, mean SE) and control fish ranged from 96 to 120 mm TL (109.0 0.5 mm). Comparison of initial size between tagged
21 and control fish showed no significant difference in lengths (t 198 = 0.91; P = 0.360) or weights (t 198 = 1.52; P = 0.129). In each of the 10 trials, all fish survived and swam actively following the 24 hour surgical recovery period, showing no apparent stress from tag implantation. After removal of the seine, time to trial completion ranged from 0.5 to 5.0 hours and averaged 2.5 0. 5 hours ( SE). Predators did not become more efficient at capturing the prey as the trials were not positively correlated to trial duration (R 2 = 0.008). After initial predation attempts, the majority of prey stayed close to artificial vegetation for co ver. Predator avoidance behavior observed for tagged fish visually appeared the same as that of control fish. No fish surviving the trial period died within the 24 hour post trial period. Advanced fingerling bass with surgically implanted tags were cons umed in similar proportions to untagged fish (Figure 2 2). Total fish consumed during each trial ranged from 5 to 15 and collectively among the 10 trials, 98 total prey fish were consumed. There were no significant difference in the percent consumed betw een the two treatment groups (F 1,187 = 0.00; P = 0.975) as the survival for each trial (0.51 0.06; mean SE) was the same for tagged and control fish. There was also no effect of prey size (TL: F 1,187 = 1.50; P = 0.222; wt: F 1,187 = 0.72; P = 0.398) or the effect of interaction of size and tagging (TL x tagging: F 1,185 = 0.36; P = 0.550; wt x tagging: F 1,185 = 0.21; P = 0.647). Thus, tagging the fish did not influence vulnerability of advanced fingerling bass to predators. Discussion This study repr esents some of the smallest Centrarchids to be tested with radio transmitters. Although advanced fingerling largemouth bass experienced tank mortality
22 to avoid pre dators. Results indicate that transmitters representing 1.2 2.7% of the term biotelemetry studies of juvenile largemouth bass. However, in field applications, caution should be used by researchers that us e these radio telemetry techniques when interpreting initial growth data and separating tagging mortality from predation mortality after tagged fish have been released into a lake or river system. Observed mortality during the growth experiments was higher than expected in laboratory studies. When investigating survival of tagged juvenile largemouth bass, Cooke et al. (2003) observed similar mortality to this study with up to 33% mortality in the first five days after implantation. High tagging as sociated mortality in 2009 could have resulted from inexperience of surgeon conducting surgical procedures (Cooke et (Knight and Lasee 1996) High mortality in 2010 for control and tagged treatments along with analysis of necropsies indicated columnaris to be the primary cause of death, although the stress from the surgical procedure could have suppressed the immune system and increased the susceptibility of tagged fish to colum naris. Further study of the effects of tagging on the mortality of advanced fingerling largemouth bass in the laboratory and field should be assessed for future use of these radio tags on juvenile bass. expel their radio tags. Although many short term laboratory studies show that radio and acoustic transmitters are not commonly expelled, Frost et al. (2010) studied growth and survival of subyearling Chinook salmon and partial to complete tag expulsion w as observed in 37% of the tagged fish after 30 d. Tags used in their experiment were a prototype acoustic
23 tags which consisted of 2.6% to 5.9% body weight ratio which could have affected tag expulsion. In field applications, an expelled tag can result in misinterpreted data and tag expulsion should be evaluated prior to field implementation. Both tagged and untagged individuals grew substantially in length and weight (mean >1% of their body weight per day) over the course of each respective experiment with tagged fish showing some impairment of growth when compared to untagged fish. Differences in growth may not be explained by inefficient feeding. Robertson et al. (2003) found negative results on growth up to day 36 for tagged Atlantic salmon parr, h owever, consumption rates were similar between tagged and control fish. Differences in growth were consistent with several other studies tagging juvenile fish with transmitters (Adams et al. 1998a; Martinelli et al. 1998; Robertson et al. 2003; Frost et a l. 2010). Adams et al. (1998a) found that growth rates of juvenile Chinook salmon (114 159 mm fork length) tagged with radio transmitters were slightly impaired at day 21, but by day 54, they were growing at rates comparable to control fish. Although tag ged fish exhibited slower growth than untagged fish, observed statistical differences may not be biologically pertinent for meeting the objectives of our field study based on the following rationale discussed by Johnson (1999). We argue that the slight ob served growth differences will not alter the survival or behavior of radio tagged, advanced fingerling largemouth bass in the field during the short battery life of the tag. In a longer field trial, however, slowed growth may influence survival by size de pendent mortality of smaller individuals (Ludsin and DeVries 1997; Post et al. 1998; Pine et al. 2000). This study showed no difference in predation rates in laboratory tanks between tagged and control fish. Many factors influence the vulnerability of prey to avoid
24 predators such as fast start performance, inability to school effectively, failure to detect predators, and prey conspicuousness (Mesa et al. 1994). No significant differences between treatments or prey size supports our hypothesis that i mplantation of radio transmitters in advanced fingerling largemouth bass does not increase vulnerability to predation. Anglea et al. (2004) studied the predator avoidance and swimming performance of tagged juvenile Chinook salmon (122 198 mm fork length) and found that surgical implantation did not significantly affect swimming performance or result in greater predation susceptibility. Although we did not test the affect of tagging on swimming performance, a relationship has been demonstrated between swim ming performance and vulnerability to predation (Bams 1967). Studies that have tested both predator avoidance and swimming performance on juvenile fish have either shown that both were either affected or both unaffected (Adams et al. 1998b; Anglea et al. 2004). Results indicate that short term mortality was higher in tagged fish, but elevated temperatures that induces stress and high frequency of bacterial infections during lab experiments made it difficult to assess the extent that surgically implantin g tags contributed to mortality. Transmitter slowed growth rates of advanced fingerling largemouth bass, but did not affect their ability to avoid predators. We assume that some impairment of growth should not significantly affect the movement and behavi or of tagged individuals, therefore inferences made from tagged largemouth bass should be adequately representative of untagged stocked fish in the field. Further research that evaluates a modified tag weight to fish ratio and surgeon experience should be investigated to assess the potential to reduce the effects of radio tags on growth or mortality of juvenile bass.
25 Conclusions from this study show that successful implantation of radio transmitters in juvenile warm water fish (90 120 mm TL) such as lar gemouth bass is possible. This information could have useful application for researchers studying the recruitment and early life history of juvenile Centrarchids and other warm water fish. As transmitter size continues to decrease with advances in techno logy, the ability to study the behavior of previously unstudied small fish will be possible. Future researchers should evaluate the impacts of implanted transmitters in the laboratory prior to using radio tags in the field to avoid misinterpreting data du e to their effect on the fish.
26 Table 2 1. Summary of initial fish length (mm TL), w eight (g), sample size (N), day of growth measurement, and percent length and weight gained per day of tagged and untagged (control) advanced finge rling largemouth bass for gro wth and mortality experiments. In 2009, (N) represen ts individual fish, whereas in 2010, (N) represents tanks. All measurements are mean SE. Group Day N Initial TL (mm) Weight (g) % TL gain per day % weight gain per day 20 09 Tagged 7 27 106.71.35 14.70.58 0.210.03 1.700.19 21 27 0.370.03 1.440.15 2009 Control 7 40 106.91.19 14.40.47 0.440.02 2.180.09 21 38 0.580.02 2.630.13 2010 Tagged 30 3 103.41.40 11.870.56 0.280.07 1.490.11 2010 Control 30 3 105.11.13 12.560.40 0.460.07 1.970.11
27 Table 2 2. Mean total length (mm), weight (g), standard error, and number of fish eaten by adult predators per trial of juvenile largemouth bass by treatment group in predator avoidance tests. Control Tag ged Mean TL (mm) Weight (g) Prey eaten n Mean TL (mm) Weight (g) Prey eaten n Trial 1 111 1.2 13.3 0.6 7 10 112 1.4 13.3 0.4 7 10 Trial 2 112 1.5 16.4 0.7 3 10 113 0.9 16.8 0.6 5 10 Trail 3 107 1.1 13.5 0.4 3 10 109 1.0 13.8 0.3 6 10 Trial 4 107 1.4 12.8 0.6 8 10 107 0.6 13.2 0.3 7 10 Trial 5 108 1.6 12.9 0.6 4 10 110 2.1 13.9 0.9 1 10 Trial 6 105 1.1 12.1 0.4 5 10 106 0.9 12.7 0.3 6 10 Trial 7 111 1.7 13.3 0.6 4 10 110 1.5 13.2 0.5 5 10 Trial 8 110 1.1 13.0 0.5 5 10 109 1.9 13.3 0.7 3 10 Trial 9 104 1.6 11.5 0.5 5 10 107 1.7 12.8 0.6 3 10 Trial 10 112 0.9 13.8 0.4 5 10 112 0.8 13.7 0.4 6 10 Overall 109 0.5 13.2 0.2 49 110 0.5 13.7 0.2 49
28 Figure 2 1. Relative growth rate (percent body weight gained per day) at 7 and 21 days in 2009 and 30 days in 2010 for surgically implanted radio tagged fish and control (untagged) hatchery largemouth bass.
29 Figure 2 2. Proportion of tagged and control (untagged) advanced fingerling hatchery largemouth bass that were consumed by adult largemouth bass in each of 10 predator avoidance trials.
30 CHAPTER 3 EVACUATION RATES OF RADIO TRANSMITTERS B Y PREDATORS Intro Radio and acoustic tran smitters can provide valuable information about the survival, movement, and habitat use of fish. Micro transmitters have expanded this research to include many species of juvenile fish (Bolland et al. 2008; Jepsen et al. 1998; Koed et al. 2002). Tracking juvenile fish with radio telemetry allows inferences about their early life history characteristics based on the movement of the tagged individual. An assumption made by scientists conducting radio telemetry studies is that the information obtained about locations and movements of a study animal is from the study animal itself and not from a predator that consumed the tagged study animal. Juvenile wild and hatchery reared largemouth bass are highly susceptible to predation during their early life stage s. For example, stocked largemouth bass in Texas experienced losses to predation as high as 27.5% within 12 hours of stocking (Buckmeier et al 2005). Thus, predators that eat radio tagged individuals can impact the results of a telemetry study, if the i nvestigators are unable to distinguish between the movement and behavior of radio tagged fish or a predator that might have eaten the tagged fish. Estimates of predator evacuation rates may be critical when conducting predation studies of wild or hatchery reared fish. Knowing how long different types of tags will remain in a predator may allow researchers to estimate the time when the study animal was consumed. Evacuation rates have been estimated for coded wire tags (Niva and Hyvarinen 2001) and passive integrated transponder (PIT) tags (Petersen and Barfoot 2003) to quantify predation of stocked fish. For example,
31 predation by double crested cormorants Phalacrocorax auritus on stocked cutthroat trout Oncorhynchus clarkii and rainbow trout Oncorhynchus mykiss has been successfully estimated utilizing coded wire tags (Lovvorn et al. 1999). Although body size and food size may play a role, research suggests that temperature is the most important factor affecting gastric evacuation rate (He and Wurtsbaugh 1993; Henson and Newman 2000; Petersen and Barfoot 2003). Researchers have studied the evacuation rates of coded wire tags (Niva and Hyvarinen 2001) and passive integrated transponder (PIT) tags (Petersen and Barfoot 2003), but little is known about the e vacuation rates of radio transmitters. The increased size and mass of radio transmitters could slow or impede transit through the intestines of a predator and bias results assumed to be from tagged individuals. With the increased use of micro transmitter s to estimate the survival and behavior of stocked fish, this information becomes critical. Previous studies in Florida indicated low survival within one year after stocking advanced fingerling largemouth bass (Porak et al. 2002), suggesting that loss es due to predation could be significant. To make accurate inferences about the mortality of the stocked population, we must know if radio transmitters accumulate in the intestine of adult largemouth bass and estimate when evacuation is likely to occur af ter predation. Using short lived batteries that are common in micro transmitters, a final fate for each tagged fish cannot be accurately determined without knowledge of evacuation time by predators. I conducted a laboratory experiment to determine evacua tion rates of radio transmitters in relation to three variables; water temperature, predator size, and prey size.
32 Methods Adult largemouth bass were used as the predator in our evacuation trials, because they are the primary predator in Lake Carlton, Fl orida (study site for radio telemetry). Thirteen adult largemouth bass (mean = 381 mm TL 6.1) were captured by electrofishing from Lake Eustis, Florida and were divided (six in one tank and seven in the other) between two 1,987 L (1.82 m diameter, 69 cm deep) circular tanks at the Eustis Fisheries Research Laboratory (EFRL), where an inflow of aerated well water (mm TL), weighed (g), and individually marked with a PIT ta g ( Table 3 1 ). Predators were acclimated to the tanks for one month and fed a maintenance diet of juvenile largemouth bass and Seminole killifish Fundulus seminolis prior to initiation of the trials. Advanced fingerling hatchery largemouth bass to be used as radio tagged prey in this study were acquired from the Florida Bass Conservation Center (FBCC) and transported to the EFRL. These hatchery fish were acclimated and held in 1,060 L (1.24 m diameter, 69 cm deep) circular tanks until evacuation trial s began. All holding hatchery largemouth bass were fed to satiation three times per week with a commercially produced petted feed. A surgical procedure similar to Adams et al. (1998) was used to implant radio transmitters (ATS, model A2414), weighing 0.24 g in air, into the peritoneal cavity of advanced fingerling largemouth bass. Tagged bass were measured (mm TL), weighed (0.1 g ; Table 3 1) and allowed to recover from ane sthesia. Six 1,060 L test tanks were simultaneously used for individual trials and included three heated, insulated tanks, and three unheated tanks at ambient temperature. Evacuation rates for each predator (n = 13) were evaluated at two water tempera tures;
33 of water temperatures at Lake Carlton, Florida during the time of hatchery bass stocking. Two predators from our experiments died and one became diseased after trial s were completed in the ambient temperature tanks; resulting in three fewer trials (n = 10) for heated tanks. Evacuation experiments were conducted between December 2, 2009 and January 18, 2010. Predators were starved for 48 hours prior to the introduct ion of radio tagged fish to ensure empty stomachs at time of feeding. Predators were moved from the 1,987 L holding tanks into individual test tanks (1,060 L) seven days prior to testing. The three heated tanks were at ambient temperature when the predat ors were moved into them. After the introduction of the predators, a heating element was used to slowly increase the water temperature of the heated tanks, allowing the predators to acclimate while the unheated tanks remained at ambient temperatures. Dur ing trials, each tank contained one predator and one introduced tagged prey fish. The first four trials consisted of six tanks, except where predators died (one tank in two different trials), became diseased, or did not consume tagged fish (two fish in tw o different trials). A fifth trial was conducted to complete the testing of four individual fish (using four tanks; two heated and two unheated) that had only been tested in one water temperature. Tagged juvenile bass were placed in each of the test tank s five minutes after tagging and in all trials, prey fish were not force fed to predators. Tanks were checked every three hours to determine when the prey was eaten. The time to evacuation was calculated from the time the prey fish was eaten until the ra dio tag was observed on the bottom of the tank.
34 The effects of water temperature, predator size, and prey size on evacuation rates were evaluated using an analysis of covariance (ANCOVA) with Proc MIXED in SAS v 9.2 (Cary, NC). Water temperature was i ncluded as a categorical fixed effect, while predator size and prey size were included as continuous fixed effects. In addition, a random fish effect was included because each predatory fish was used during both the ambient and heated temperature trials. Model assumptions of normality and homogeneity of variance were examined visually to ensure validity of the analyses. To predict the evacuation time of radio tags, we modeled the percent of tags evacuated at a given time after consumption for each water t emperature using the power function described in Peterson and Barfoot (2003): where Ptag t is the percent of tags evacuated at time t S is a parameter that controls the shape of the function, and H is a measure of the half life of the prey item being eva cuated. We parameterized the gastric evacuation process power function using non linear modeling (Proc NLIN) in SAS v 9.2 (Cary, NC) for both water temperature treatments. Significance for all statistical analyses was set at P < 0.05. Results and Disc ussion The predators in this experiment ( Table 3 1) ranged from 343 to 455 mm TL and 571 to 1,325 g in weight. Advanced fingerling largemouth bass used as prey ranged in length from 96 to 137 mm TL and 8 to 23 g in weight. Radio transmitters implanted in juvenile bass did not accumulate in the intestine of adult largemouth bass and evacuation times ranged from 15 to 84 hours (Table 3 1). The median evacuation time of radio transmitters for adult bass was 51 h in non heated tanks and 34 h in heated
35 tank s (Table 3 1). The effect due to repeated observations from using the same predator fish in multiple trials was estimated at zero in the ANCOVA. Results of the ANCOVA modeling showed no significant differences in evacuation time due to predator total len gth (F 1,17 = 0.02, P = 0.88) or prey total length (F 1,17 = 0.04, P = 0.85). However, evacuation rates increased significantly with water temperature (F 1, 21 = 5.33, P = 0.03). Using the power function described in the methods, the model was a significan t predictor of evacuation (F 4, 54 = 997.8, P < 0.001), which can be seen by the tight fit of the model predictions to the observed evacuation times (pseudo R2 = 0. 97; Figure 3 1). The half ( Table 3 2). From this model, we are able to predict the percent of tags remaining in a r e 3 1). In this study, all adult largemouth bass evacuated radio transmitters that were implanted in juvenile bass in less than 84 h. Our results were similar to Niva and Hyvarinen (2001) who found northern pike had evacuated coded wire tags within 3 d. Petersen and Barfoot (2003) studied the rates of northern pikeminnow evacuating PIT tags after ingesting juvenile Chinook salmon and found that median evacuation times were 40 h at 14C and 22 h at 18C. I anticipated faster evacuation rates for lar gemouth bass in high water temperatures compared to other studies that were conducted in cooler water temperatures. Results could be explained by differences in species evacuation rates or the larger mass of our transmitter slowing transit through the int estine. Hunt (1960) studied the digestion rate of largemouth bass (n = 18) at water temperatures between 23C and 26C and found that a mean of 99% of the food was digested within 27 hours. Beamish (1971) found similar results, studying
36 largemouth bass h eld in tanks at 25C that evacuated lake emerald shiners Notropis atherinoides i n les s than 24 h. In our study, at 24 hours, only 30% of the transmitters were evacuated at 30C and only one of 13 at 22C. This gives support to speculate that transmitters could have been temporarily impeded from evacuating, although, additional consumption of prey, which typically would occur in the wild, could have increased this evacuation time. Variation of evacuated transmitters was largely explained by water te mperature and not significantly affected by predator or prey size. These results were consistent with other studies documenting significantly faster evacuation times with higher water temperatures, owing to higher metabolic processes (Petersen and Barfoot 2003; Henson and Newman 2000; He and Wurtsbaugh 1993). Petersen and Barfoot (2003) saw significantly higher evacuation rates of PIT tags by northern pikeminnow at 18C compared to 14C. Cochran and Adelman (1982) used an exponential decay model to quant ify gastric evacuation of largemouth bass in and found that the exponent of gastric evacuation increased exponentially with temperature. In radio telemetry field applications, this model of evacuation probabilities could help predict if a located tag i s from a juvenile radio tagged fish or is within a predator that has consumed a radio tagged fish. Once the transmitter stops movement, the estimated timing of predation can be calculated. As a result, once a tag stops moving, investigators could exclude the location information from the three days prior to the cessation of movement, because those locations likely resulted from predator movement.
37 Evacuation rates of radio or acoustic transmitters have not been previously studied to determine if and whe n radio transmitters will be evacuated from a predators stomach. Transmitters are much heavier and rigid than previously studied coded wire tags or PIT tags and often possess a trailing antenna (10 cm in this study). In this study, we were able to confirm that miniature radio transmitters did not accumulate in a antenna ( Figure 3 2), which can be used to distinguish natural mortality and predation in field applicatio ns when radio tags are recovered. If evacuation rates are unknown by scientists using radio transmitters to study juvenile bass, then significant movement and habitat information could be erroneously assigned to the radio tagged juvenile fish instead of t he predator that had consumed a radio tagged bass (i.e., prior to evacuating the transmitter). This knowledge of the timing of evacuation will reduce bias for any type of radio telemetry experiment that studies the survival or behavior of juvenile fish or any small species of fish that has a high predation risk. Evaluation of evacuation rates and tag accumulation should be assessed with other predator species and when using larger transmitters.
38 Table 3 1. Comparison of Prey Size, P redator size, and Evacuation times (mean SE) at two temperatures for adult largemouth bass (predator) consuming advanced fingerling largemouth bass with surgically implanted radio tags (prey). Temp Prey size Predator size Evacuation times ( o C) TL (m m) Weight (g) TL (mm) Weight (g) Average Time (h) Min (h) Max (h) n 22 114.83.2 14.21.3 3838.5 83656.8 51.05.4 27 84 13 30 116.63.3 15.11.4 3799.0 80260.5 33.65.0 15 54 10 Overall 115.62.3 14.61.0 3816.1 82040.7 45.44.4 15 84 23
39 Table 3 2. Parameter estimates for a model that predicts the probability of a predator evacuating a tag at a given time after consumption; using the power function described in Peterson and Barfoot (2003): where Ptagt is the probability a tag is evacuated at time t. The parameter estimates (and 95% confidence intervals) in the table are for S (parameter that controls the shape function) and H (a measure of the half life of the prey item being evacuated) used in predicting the proportion o f radio transmitters evacuated by adult largemouth bass at time t after consumption. Parameters Treatment S H 2.56 (2.19, 2.93) 47.72 (45.62, 49.82) 2.02 (1.70, 2.34) 30.73 (28.80, 32.65)
40 Figure 3 1 Percent of radio tags rema ining in the gut of a predator (i.e., adult largemouth bass) after consumption of radio tagged prey at 22C and 30C; observed during laboratory experiments.
41 Figure 3 2 A photo illustrating the curling effect on the antennae of a radio transmitter after ingestion of tagged fish by a predator (i.e., largemouth bass) and evacuation of radio tag after the tagged bass was digested. Photo by Brandon Thompson
42 CHAPTER 4 MOVEMENT, SURVIVAL, AND HABITAT USE OF H ATCHERY REARED VERSUS WILD JUVENILE LARGEM OUTH BASS Intro Studi es have been conducted to determine the contribution and long term survival of stocked largemouth bass (Buynak and Mitchell 1999; Porak 2002), but little information exists on their immediate survival and behavior after release. For example, Porak et al. (2002) found less than three percent survival for stocked largemouth bass in Florida lakes one year post stocking, but mechanisms that cause the mortality were largely undetermined. It also remains unclear if low contribution of hatchery fish tot the popu lation results from higher mortality in stocked fish compared to wild bass or if we simply cannot stock high enough numbers of hatchery fish to gain sufficient contribution. Behavior is an important mechanism of survival. Hatchery reared fish stocked i n the wild must go through a critical post stocking period when feeding and anti predatory skills must be developed (Heggberget et al. 1992; Brown and Laland 2001). Pouder et al. (2010) found that stocked advanced fingerling largemouth bass had difficulty transitioning to natural prey within seven days post stocking; possibly contributing to high initial mortality. Buckmeier et al. (2005) showed high predation of stocked largemouth bass immediately after release. I am particularly interested in determini ng how the behavior of stocked hatchery largemouth bass compares to that of wild bass. Behavioral differences might lead to high predation and/or an inability to locate and consume prey during this critical period. Intensive research has focused on hatch ery reared salmonids and has identified domestication issues, including behavioral deficits such as naivety to predation (Huntingford 2004). Understanding how movement and habitat use influence the fate of stocked largemouth bass during the initial period
43 following release can help improve the effectiveness of stock enhancement programs for largemouth bass. Biotelemetry has been used extensively to provide researchers with valuable information about adult largemouth bass mortality (Hightower et al. 20 01), movements (Hanson et al. 2007), and habitat use (Mesing and Wicker 1986). Micro transmitters now allow scientists to research post stocking survival and behavior of juvenile fish (Jepsen et al. 1998; Dieperink et al 2001; Bolland et al. 2008). Resu lts from these studies have identified differences in survival (Dieperink et al 2001) and behavior (Bolland et al. 2008) between hatchery reared and wild fish. However, to this point, micro transmitter biotelemetry has not been utilized to evaluate survi val and/or behavior of hatchery reared largemouth bass during the critical period following release into the wild. The population of largemouth bass on the Oklawaha Chain of Lakes ( Figure 4 1) has been identified as recruitment limited due to loss of nu rsery habitat (Benton et al. 1991, Benton 1999, Wicker and Johnson 1987). Stocking of advanced fingerling hatchery largemouth bass has been attempted to increase bass abundance in this system and to improve angler catch rates. The objectives of this chap ter were (1) to assess the behavior, movements, habitat use, and survival of hatchery reared advanced fingerling largemouth bass after being stocked into the wild and (2) to examine differences in survival, movements, habitat use, and growth between hatche ry reared and wild juvenile largemouth bass. Comparison of wild fish and hatchery reared stocked fish should help inform managers if survival rates are affected by lake conditions and/or the fitness of hatchery fish.
44 Methods Study Area Lake Carlton ( 1 57 ha) is a part of the Oklawaha Chain of Lakes (30,756 ha ), located in central Florida ( Figure 4 2). These lakes are characterized as hypereutrophic, they have high densities of phytoplankton, and they have a narrow band of emergent vegetation that exist s around the shoreline. Lake Carlton was selected for this study, because its habitat was representative of the entire chain of lakes and its small size allowed for efficient tracking of radio tagged fish. The primary aquatic macrophytes include panic g rasses Panicum spp., bulrush Scirpus californicus cattails Typha latifolia and pickerelweed Pontederia spp., along with sparse patches of spatterdock Nuphar luteum Limited submersed aquatic vegetation existed in the lake and consists primarily of eelgr ass Vallisneria americana The perimeter of Lake Carlton is 4.49 km and two separate navigable canals ( Figure 4 2 ) connect to Horseshoe lake (13 ha) and Lake Beauclair (459 ha). Habitat mapping To evaluate habitat use of radio tagged largemouth bass, an extensive evaluation of the littoral aquatic habitat was conducted in Lake Carlton. We measured the lake edge (practical shoreline) and all emergent aquatic vegetation in the littoral zone of Lake Carlton during spring 2009. We completed the habitat sur vey in spring 2010, when we measured submersed species of aquatic plants (e.g., Vallisneria spp) using transects spaced at 50 m intervals around the lake and created polygons from those data. Primary plant communities were measured to sub meter resolution with a Trimble GeoXT truthed to verify community polygons. We redefined the boundary of Lake Carlton in spring 2010 to account for a rise in water levels, although due to time restraints, the veget ation community polygons were not redrawn. Many species encountered between
45 the former lake edge and the new shoreline were terrestrial plants that grew during the low water periods prior to and during 2009. As such, this creates an unknown source of bia s and inaccuracy of habitat use when radio tagged largemouth bass utilized shallow water areas during 2010. A 3 m buffer was added in Arc/View to quantify offshore locations by tagged fish; assuming that fish within this buffer were relating to inshore ve getation. A small section of the Lake Carlton vegetation map is shown in Figure 4 3 to illustrate the types of aquatic plant polygons and largemouth bass locations that were Radio tags and surgical procedure We used model A2414 radio transmitters (Advanced Telemetry Systems; Isanti, MN, 148 151 MHz), which were the smallest commercially available tags. These transmitters weighed 0.24 g in air (5 x 12 mm) and were equipped with a 10 cm flexible braided wire a ntenna. These radio tags had a pulse rate of 60 pulses per minute in 2009 and had an expected battery life of 24 days. Transmitter pulse rates were modified to 30 pulses per minute in 2010 in an attempt to achieve a 30 day tag life. Tag range and accu racy were assessed for the water depth and conductivity at Lake Carlton. To assess range, live transmitters were suspended under a float at differing water depths in Lake Carlton from a shallow depth (<1 m) to the maximum depth of the lake (4 m). Distanc e from the radio transmitter was recorded where the transmitter pulse could no longer be heard. To assess transmitter range in vegetated areas, we placed transmitters at mid water depth in varying vegetation densities and near artificial structures. High accuracy (3 m) of locating tags in vegetated areas was essential for estimating habitat use. Accuracy of detection was assessed by hiding
46 transmitters from the tracking biologist in dense vegetation, and having that biologist throw a buoy where they tho ught the transmitter was located. Accuracy was measured by the distance (m) from the buoy to the known location of the transmitter. Tag failure rates were assessed in 2010 due to higher than expected tag failure rates in 2009. We were assured by the m anufacturer in 2009 that tag failure would be negligible, so we did not assess tag failure during the first year of the study. However, high tag failure rates in 2009 made it difficult to differentiate whether the radio tagged bass died via avian predator s or if the tags had failed. Therefore, in 2010, we studied tag failure rates in the laboratory. We allowed 30 live transmitters to expire in the lab to estimate expected tag failure in the field. Tags were submersed in water in tanks at the Eustis Fish eries Research Laboratory (EFRL) in water temperatures similar to those in Lake Carlton. The same model transmitters were used in the field and lab studies. The 30 control tags were selected systematically (every 2 3 tags) from a batch of 80 tags and fai lure rates were analyzed by frequency (149 151 MHz). technique described in Adams et al. (1998) was used to implant transmitters into the peritoneal cavity of largemouth ba ss and provide a lateral wall exit for the antennae. Each individual fish was anesthetized in 70 mg/L tricaine methanesulfonate (MS 222), then placed ventral side up on a molded foam pad soaked in Stress Coat where water with 30 mg/L MS 222 was continua lly pumped over the gills ( Figure 4 4). Each transmitter was inserted through a 9 mm incision 3 mm away from and parallel to the mid ventral line anterior to the pelvic girdle. The incision was closed with a single individual coated vicryl absorbable sut ure (5 0). The surgery took between two and
47 three minutes to complete. When a surgically tagged fish recovered, it was placed in a recovery tub (wild fish tagged at the lake) or transferred to a raceway (hatchery fish tagged at FBCC). Hatchery bass sto cking 2009 Radio transmitters were inserted into 50 hatchery reared advanced fingerling largemouth bass (105 0.8 mm; mean TL SE) at the FBCC on May 19, 2009 ( Table 4 1). Hatchery bass were selected for tagging from a group of fish raised using stand ard FWC culture and stocking protocol for bass that were to be stocked in Lake Carlton. After surgeries, fish were given a 24 hour recovery period before being transported to Lake Carlton. On May 20, 2009, we stocked the 50 radio tagged fish evenly aroun d the shoreline of Lake Carlton simultaneously with untagged fish that were being stocked. Each radio tagged fish was released with a group of hatchery fish (~100 200) to avoid biasing the post stocking experience of the radio tagged hatchery largemouth b ass. Individual stocking locations of the 50 radio tagged fish were recorded for dispersal and movement analysis. At the time of stocking, four plastic mesh cages were deployed in Lake Carlton with tagged and untagged hatchery fish. Each cage contained 25 control stocked hatchery fish and 12 tagged fish and were distributed evenly among the four cages. Hatchery and wild bass stocking 2010 A total of 30 hatchery reared largemouth bass (106.2 1.62 mm; mean TL SE) and 20 wild bass (110.7 1.53 mm) were surgically implanted with radio transmitters to compare their behavior and survival ( Table 4 1). Hatchery bass were selected for tagging from a group of fish that were raised at the FBCC for stocking in Lake Carlton. A minimum size of 90 mm TL (~9.0 g) g tag (Winter 1996;
48 Table 4 1). Attempts were made to keep both the hatchery and wild bass sizes between 90 and 120 mm TL. Surgeries were conducted on juvenile hatchery bass on March 29, 2010 at the FBCC and fish recovered for 24 hours before they were transported and stocked into Lake Carlton the following day. Wild juvenile largemouth bass were collected from Starke Lake (Orange County, Florida) via electrofishing on March 30, 20 10, transported to Lake Carlton, and surgically implanted with radio transmitters. Wild radio tagged fish recovered for 1 hour before they were stocked into Lake Carlton. Fish were stocked at 10 random, evenly spaced locations around the lake, with each stocking site receiving 5 radio tagged fish (3 hatchery and 2 wild fish) and approximately 1,100 microwire tagged hatchery bass. Mortality caused by stocking and tag implantation used in the survival analysis was determined by field observations of tag recoveries and movement. At the time of stocking, six plastic mesh cages were deployed in Lake Carlton with tagged and untagged hatchery and wild radio tagged fish to identify differences that may have resulted from stocking or surgical implantation. Ele ven dummy tagged hatchery fish (105.6 2.2 mm; mean TL SE) were divided among three cages and 10 dummy tagged wild fish (113.7 1.9 mm; mean TL SE) were divided into three separate cages. Non tagged hatchery fish (n = 75) were spread evenly among th ree cages as controls for the dummy tagged hatchery fish. Thirty wild non tagged wild bass collected from Starke Lake were spread evenly among three cages as controls for the dummy tagged wild fish. Cages were checked at 24, 48 and 72 hours post stocking for missing or dead fish.
49 Tracking Radio tagged fish were manually tracked daily by boat with an ATS R410 scanning receiver and a three element yagi antennae until their signals were lost. Due to the high numbers of tagged fish at large and time req uired tracking them in 2009, one half of the tagged fish were tracked one day and the other half were tracked the next day until a reduction in the remaining fish numbers facilitated daily tracking of all remaining fish. Similar to 2009, daily tracking wa s done in 2010 with the exception that two tracking boats and a search boat (utilizing an omni directional dipole antennae) were deployed each day to ensure that all fish were located each day. The small size of Horseshoe Lake allowed us to manually searc h for fish that might have emigrated from Lake Carlton through the canal ( Figure 4 2). However the large size of Lake Beauclair made searching difficult and we experienced a relatively high number of fish moving into Lake Beauclair during 2009. Therefor e, in 2010, to detect fish movement out of the lake, two ATS R4500S stationary data logging receivers and antennas were deployed at the narrow canal connecting Lake Carlton and Lake Beauclair ( Figure 4 2). Each of these loggers scanned 25 frequencies sequ entially for six seconds (5 minute cycle), allowing two pulses to be recorded. Information was uploaded daily from data loggers and extensive searches were conducted in Lake Beauclair for a frequency that was identified passing through the canal. When a fish was located, coordinates were recorded with a TRIMBLE GPS unit along with temperature, depth, and habitat data. Tracking began each day at a randomly chosen radio tagged bass location from the previous day. Trackers flipped a coin to determine a random direction, that is, whether to go clockwise or counter clockwise around the lake. When a signal was not quickly heard from the previous
50 for missing fish were conducted at the end of the day. However, in 2010, the search boat was immediately notified. The biologist in the search boat entered the frequency into their receiver and began searching the lake to determine if each missing fish had moved extensive dis tances or if the signal had disappeared from the lake. Additional frequencies were added to the scan throughout the day and trackers were notified when a signal was identified. Extensive searches were conducted daily for previously missing frequencies. Daily tracking was critical to make fate determinations, because we were using radio transmitters that had a very short battery life. We assumed that if a transmitter prev ious location. We also assumed from laboratory studies (Chapter 3) that 95% of transmitters would be evacuated by a predator (adult largemouth bass) within 72 hours of consuming a radio tagged fish. When fish movement ceased for two consecutive days, att attempt to coerce the fish to move. If no movement was observed by the third day, electrofishing was conducted to determine the final fate of the individual. When the targeted fis h was not captured via electrofishing, we assumed the fish had died and attempted to recover the radio tag from the lake bottom utilizing a high strength magnet (2.4 x 28 cm) attached to a 3 m aluminum pole or end of a throw rope when deep water recovery w as necessary. When a tag was recovered with the magnet, we assumed that was observed (Chapter 3). If the transmitter antenna did not display curling, we
51 concluded th at the fish had died from tagging effects or natural causes. Predation was also assumed (but unconfirmed) if the tag could not be recovered or, when abnormal movements were followed by no movement two to three days later. For example, if a fish spent 10 days in shallow, dense vegetation with little movement and then moved 0.5 km per day offshore for two days followed by no movement offshore for 15 more days, we assumed this fish was eaten by a predator three days prior to cessation of movement. I assumed that 100% of the live transmitters in the lake were detected and transmitters due to unobserved avian predation or undetected emigration from the lake. Once transmitters began to expire near the end of the study (28 30 days post stocking), we attempted to collect all remaining fish via electrofishing, after which they were measured to compare growth between groups. Relative growth rates for wild and hatchery fish at the conclusion of the study were expressed in percent body weight and length gained per day (Busacker et al. 1990). Data analysis Survival was calculated as the number of days that hatchery or wild largemouth bass remained alive in the system. For survival analysis, fish that differentiate losses due to tag failures or avian predation. In 2009, we assessed the initial hatchery survival for the first 14 days post stocking due to limit ed tag life. In 2010, we compared daily survival rates between hatchery and wild released fish to 30 days post stocking using a parametric accelerated failure time regression in PROC LIFEREG (SAS v 9.2, Cary NC; Allison 2010) assuming a Weibull hazard fu nction. We then square test. We also
52 compared daily survival of the three tag frequencies in a controlled laboratory setting (PROC LIFEREG). Dispersal and movement of hatchery and wild f ish were calculated to the nearest meter using Arc/View. Dispersal was defined as the distance traveled from individual stocking sites by a fish that was alive and had a live transmitter at days 7 and 14. Dispersal was assessed in 2009 for hatchery fish and mean dispersal was compared between hatchery and wild fish in 2010 at days 7 and 14 using a non parametric Mann Whitney U test because the data was not normally distributed. Movement was defined as the sum of each distance traveled for all locations o f an individual fish. Only locations where the fish was determined to be alive were included in movement analysis. We assessed the movement per day for hatchery fish in 2009 and compared the movement of fish that were eaten by predators versus fish that survived the experiment using a Mann Whitney U test. We also compared the average movement per day between hatchery fish in 2009 and 2010, and movement of hatchery versus wild fish in 2010 using Mann Whitney U tests. Habitat use by hatchery and wild fish was described in 2009 and 2010 by taking the mean proportion of locations of individual fish found in each vegetation layer. Water level changes affecting our habitat map along with limited locations in each habitat made it difficult to compare specific habitat use for hatchery fish between years and hatchery to wild fish in 2010. The offshore edge of the vegetated zone was largely unaffected by water level changes and the number of locations in either offshore or inshore zones was sufficient to test for differences. Therefore, comparisons of offshore
53 use were made for hatchery fish between years and between hatchery and wild fish in 2010 using a Mann Whitney U test. Only live locations (where transmitter moved from previous location) were used in the a nalysis of movement and habitat use. We also excluded locations 2 4 days (Chapter 3) prior to when a tag stopped moving if predation was the assigned fate, because movement and habitat use during that time period could have resulted from a predator. Grow th in percent body weight gained per day was compared between hatchery and wild fish surviving the experiment using a two tailed t test assuming equal variance. All tests in this study were considered statistically significant at P < 0.05. Re sults Range and accuracy of radio tag locations Range and accuracy was first determined to learn the distances required to track each radio tagged fish and the accuracy of each location at various depths and habitats. Transmitter range detection was neg atively related to water depth. When the transmitter was placed in a known location at a 1 m depth (average depth of littoral zone), range varied from 300 to 600 m. When placed in a 2.5 m depth (average depth of offshore zone), range decreased and averag ed 150 to 300 m. When a transmitter was dropped to the bottom of the lake where it was 4 m deep (the deepest known location in the study lake), the range to detect sound from the tag further decreased to an average of 100 meters. The type and density of vegetation or artificial structures did not appear to affect the range and accuracy of the transmitters. Locations of transmitters in the vegetated zone (1 m) were more accurate than locations in the open water (3 m), presumably due to shallower water d epths and references (eg., plants and shoreline) useful to help
54 triangulate the location. This information was applied to the logistics, protocols, and strategies used for tracking radio tagged largemouth bass each day on Lake Carlton. Tag failure rates A portion of the radio transmitters expired early in both years of tracking. In 2009, some missing transmitters gave abnormal pulses 1 2 days prior to disappearing, indicating that tag failure was occurring. Although assured by the manufacturer that th e high rate of tag failure experienced in 2009 was rectified, early tag failure persisted in 2010, based on observations of control tags monitored in the lab. Two control tags failed to produce any sound as early as two days after they were activated in t he lab and 36% failed by day 21. Field tags resulted in similar failure over the 30 day study period, although field tags were missing in a higher proportion from day 20 to day 30 ( Figure 4 5). After analyzing the persistence of control tags in the lab b y frequency, we found a significantly higher failure rate of tags with the 151 MHz frequency when compared to 149 MHz and 150 MHz ( Figure 4 6). At day 21, control tags in the lab had low failure rates for frequencies of 149 MHz (10%) and 150 MHz (11%), wh ereas 82% of the 151 MHz frequency had failed. Transmitters with frequency 151 MHz comprised 36% (11 of 30) of the control tags and 18% (9 of 50) of the field tags in 2010 and thus, field tags should have resulted in an overall lower failure rate than obs erved in the lab. All of the 151 MHz tags during field studies had coincidentally been implanted into wild largemouth bass. The manufacturer (ATS) suggested that high failure rates of transmitters with frequency 151 MHz may have resulted from a flaw in t he construction of these particular tags.
55 Hatchery bass stocking 2009 All fish implanted with radio transmitters (n = 50) survived the 24 hour recovery period following surgery and appeared healthy and active upon stocking. In four test cages, only one of the 12 radio tagged bass and four of 100 control hatchery fish died during the 3 day trial. We recorded 379 locations radio tagged of fish determined to be alive during this study segment. About one half of the hatchery fish survived the 14 day period and the other one half were confirmed deaths or disappeared ( Table 4 2). Throughout the study period, predation was determined to be the cause of death for eight of the 50 tagged fish. Bird predation was confirmed for three of the eight mortalities. On e transmitter had been eaten by a great blue heron Ardea Herodias. Two other radio transmitters were found evacuated with a distinct curling effect on the antenna and on dry ground within 10 m of the lakeshore under trees commonly used by double crested c ormorants Phalacrocorax auritus for roosting. Of the remaining five observed mortalities, one tag was found on the lake bottom and displayed distinct curling of the antenna. The other four deaths were deduced to be a result of predation based on movement and behavior prior to cessation of movement. Eighteen tagged fish disappeared from the study area; resulting from either tag failure or bird predation (i.e., birds carrying radio tagged bass outside of study area). The remaining fish (n = 24) survived a t least 14 days post stocking and all tags expired by June 7, 2009 (19 days post stocking). Losses due to mortality and missing fish were highest during the first seven days post stocking ( Figure 4 7). Of the eight observed mortalities during this stud y, seven of the radio tagged largemouth bass died in the first week. Based on the survival curve and excluding missing fish, survival of radio tagged hatchery fish in 2009 was estimated
56 to be 83% by 14 days ( Figure 4 8). Because tag failure was not asses sed in 2009, a range of survival to 14 days post stocking was estimated from 48 84% depending on how missing fish were interpreted. If we assume that all missing fish were due to bird predation (i.e., tags were evacuated outside our study area), then only 48% (24 of 50) of hatchery fish survived the first 14 days post stocking. However, if we assume that all missing tags resulted from tag failures, hatchery survival could be nearly 83% at day 14 ( Figure 4 8). Hatchery and wild bass stocking 2010 All radio tagged hatchery largemouth bass (n = 30) and hatchery dummy tagged fish (n = 11) held in mortality test cages survived the 24 hour recovery period and appeared to be in excellent condition when stocked into Lake Carlton. Only one of the 21 (4.8%) du mmy tagged bass distributed throughout six mortality test cages died during the 3 day trial. This dead fish was one of the 11 dummy tagged hatchery fish and was found dead during the 48 hour check. It is likely that the cause of death for this fish was r elated to stress associated with the tag implantation procedure, based on our lab studies on tag effects (Chapter 2). All of the dummy tagged wild fish were alive after three days. Among the 75 controls (i.e., non radio tagged hatchery fish that had been spread evenly among three cages), no fish died. There were no mortalities among the 30 non tagged wild fish that were spread evenly among three cages and served as controls for the radio tagged wild largemouth bass. Thus, we assumed handling mortality f rom surgical procedures and stocking to be negligible, and did not influence results. All transmitters expired within 34 days of stocking.
57 Hatchery reared and wild bass ranged in total length from 93 to 121 mm and 97 to 122 mm, respectively ( Table 4 1). Hatchery reared largemouth bass were slightly shorter than wild fish (t 48 = 1.93; P = 0.031) but weighed the same as wild fish (t 48 = 0.31; P = 0.76), because wild fish were thinner than the pellet fed hatchery fish. We recorded a total of 678 (390 hatch ery and 288 wild) locations for radio tagged fish that were determined to be alive. Half of the wild fish and only about a third of the hatchery fish survived the 30 day experiment ( Table 4 2). Of the 50 radio tagged individuals, we determined that 16 fish died: 14 hatchery fish (47% mortality) and two wild fish (10% mortality). Survival through 30 days appeared to be much lower for hatchery bass than wild bass ( Figure 4 10 ). There were marginal differences in survival rates between wild and hatchery fish (Wald = 3.54, P = 0.0599). Survival rates at 14 days post stocking were 91% for wild fish and 64% for hatchery bass. Survival rates at 30 days post stocking for wild and hatchery bass was 82% and 39%, respectively. We deduced that two of the radio tagged wild bass died from tagging stressors, because both fish stopped moving within three days post release and neither antenna was curled when their transmitters were later recovered. If both wild fish mortalities due to tagging were remo ved from this analysis, survival of wild fish would be 100% through 30 days. In that case, survival differences between wild and hatchery bass become much more significant. Sixteen tagged fish disappeared from our study area, which could have been due to either tag failure or avian predation. All mortality for hatchery and wild fish occurred within 14 days of stocking and 69% (11 of 16) took place in the first week.
58 Predation was determined to be the cause of all mortality of radio tagged hatchery large mouth bass (n = 14), because eight tags (27%) were found on the lake 3 days prior to cessation of movement. The other six transmitters were not found because they were evacuated in deep water where tag recovery was virtually impossible and predation was determined based on movement and behavior prior to cessation of movment. Similar to 2009, bird predation was suspected in three (10%) of the 14 predation events. One r adio tagged largemouth bass was tracked in a live cormorant in flight. Another transmitter was found on land with a curled antenna. A third was found evacuated on shore near a cormorant roost adjacent to Lake Beauclair, but had not been detected by the s tationary receiver that had been placed to detect fish moving through the Carlton Beauclair canal. Tagging mortality was determined to be the cause of death for both of the wild fish that died. We came to this conclusion because both fish ceased movemen t within three days of being stocked and upon tag recovery, antennas were straight and did not display the signs of having been consumed and evacuated by predators. Despite numerous attempts to retrieve the tags with the high strength magnet and electrof ish the radio tagged fish if it was still alive, neither of the tags could be found until 10 days after movement stopped; suggesting that the tag was still inside of a fish until decomposition had fully taken place. Hatchery fish suffered high predation (14 of 30) in the first 30 days post stocking ( Table 4 2). We assumed that all missing radio tagged fish (n = 8) were tag failures, but lab studies indicate that only six should have failed ( Figure 4 5 ). Thus, undetected bird
59 predation may have removed m ore transmitters from our study area, which would have made the estimates of predation mortality even higher for radio tagged hatchery largemouth bass No predation was observed for radio tagged wild fish and it appeared that tags in the field failed in p roportion to control tags in the laboratory. Two of 11 wild fish that were tagged from frequency 150 MHz became missing and this frequency had an expected failure rate of 20%. Nine wild fish were tagged with frequency 151 MHz (80% failure rate) and six o f an expected seven became missing. Dispersal Dispersal was highly variable for individual fish in both years. The distance that hatchery radio tagged largemouth bass dispersed from their original stocking sites in 2009 ranged from 12 to 1,340 m after 7 days and 11 to 2,010 m at day 14 ( Table 4 3). Hatchery largemouth bass dispersed rapidly; 17 of 27 fish (63%) dispersed more than 400 meters from their stock site within the first week ( Figure 4 11). Hatchery tagged fish appeared to disperse much grea ter distances than wild tagged fish ( Table 4 3) but again, there were no significant differences at day 7 ( U = 174; P = 0.140) or day 14 ( U = 54; P = 0.488). Dispersal was commonly limited by the size of the lake and in both years, substantial numbers o f tagged fish emigrated from the lake. Considering the diameter of Lake Carlton is only 1,400 m, some individuals circled around the entire lake and ended near their stocking location. In 2009, four of the remaining 27 (15%) fish had dispersed into Lake Beauclair by day 7 and by day 14, five of the remaining 23 (22%) fish had moved into Lake Beauclair. In 2010, s even of the 50 (14%) fish were observed emigrating from Lake Carlton, all of which were hatchery reared fish (23%).
60 Movement Average individ ual movement of hatchery fish was variable among individuals and similar between years. Total movement for hatchery fish locations in 2009 ranged from 11 m for a radio tagged largemouth bass that only persisted four days (3 m per day) to 6,005 m for a fis h that persisted 18 days (334 m per day). Distances moved per day during the study in 2009 were not found to be statistically significant ( U = 78; P = 0.795) between fish t hat survived (mean = 113 m) and fish that died (mean = 155 m). Movement per day fo r radio tagged hatchery bass that were located at least once (n = 47) ranged from 3 m to 500 m in 2009 and 1 m to 456 m in 2010 ( Table 4 3). Average movement per day was not significantly different ( U = 830; P = 0.114) in 2010 between hatchery fish in 200 9 (124 m) and 2010 (75 m) In 2009, 49% of radio tagged fish moved over 100 m compared to only 21% in 2010 ( Figure 4 12). In 2010, hatchery fish displayed a high range of movement and wild fish consistently moved very little ( Figure 4 12). Hatchery fi sh had significantly higher ( U = 444; P = 0.002) movements per day than wild fish ( Table 4 3). Only one wild fish (5%) in our study moved over 50 m per day on average compared to 52% of the hatchery fish ( Figure 4 12). The maximum distance traveled by a hatchery fish was 3.03 km in a 24 hour period in 2009, which included swimming the distance of Lake Carlton. We confirmed this was a live hatchery bass (and not a tag being carried by a predator) by capturing it with an electrofishing boat. Habitat use Habitat use varied by individual hatchery fish in 2009 and 2010 commonly using offshore habitats whereas wild fish mostly used vegetation ( Figure 4 13). The offshore zone (excluding buffer) comprised 146.0 ha (92.7%) and the inshore vegetated zone (incl uding buffer) contained 11.4 ha (7.3%). Some fish were
61 consistently located in open water, while others were consistently located in dense littoral zone vegetation. Average percent of offshore locations were not significantly different ( U = 837; P = 0.06 3) between hatchery fish in 2009 (34.7% SE = 4.8) and 2010 (23.0% SE = 5.1). A Mann Whitney U test revealed a significant difference ( U = 370; P = 0.036 ) in offshore use between hatchery (23.0% SE = 5.1) and wild fish (7.9% SE = 4.1). In a direct compari son, 11 of 30 (37%) hatchery fish were located offshore on at least 30% of their locations, whereas only 1 of 20 (5%) wild fish displayed this behavior. Although the exact depth tagged fish were suspended at was unknown, the mean depth and water tempera ture were determined from locations where the fish was alive. In 2009, mean depth was 1.02 (SE = 0.08) and mean water temperature was 26.7C (SE = 0.12; range 22.9 (1.37 m; SE = 0.10) was significantly deeper (t 47 = 2.47; P = 0.026) than that used by wild fish (1.04 m; SE = 0.10). Water temperature in 2010 ranged from 19.8C to 25.2C and averaged 23.7C during the 30 day experiment. and of vegetation and thus, we have little ability to detect differences in habitat selection with a small number of fish and short battery life, which limited the number of locations. Although we only tracked fish for 30 days post stocking, we did observ e some fish begin to make habitat choices. For example, one fish resided for seven days within 50 m of its stock site and then proceeded to travel from 300 800 m per day for four days to an area of complex habitat (eelgrass adjacent to emergent vegetation ) in Lake Beauclair where it spent the final 19 days (all within 50 m) prior to being captured by electrofishing.
62 Growth Radio tagged hatchery and wild bass surviving the study in 2010 differed in growth rate ( Figure 4 14). Surviving radio tagged wild f ish used in this analysis had similar (t 12 = 1.5; P = 0.16) initial total lengths as surviving hatchery fish. Wild radio tagged fish grew an average of 1.73% of their body weight per day, which was significantly higher (t 12 = 3.83; P = 0.002) than hatcher y fish (0.41 %; Figure 4.14 ). Although tag presence was shown to impair growth of tagged individuals in 30 day laboratory trials (Chapter 2), wild fish growth rates in Lake Carlton were comparable to untagged fish growth seen in the lab that were fed pelle ts. This high rate of growth indicates that tagged wild fish were able to forage efficiently soon after the surgical implantation of radio tags. Discussion This study represents the first attempt to compare the behavior and survival of stocked hatcher y and wild largemouth bass, and it demonstrated the benefits of using radio telemetry. My findings suggest that domestication effects of hatchery reared bass can have a negative influence on behavior and survival after release into the wild. Post stockin g survival was lower for hatchery reared fish than wild bass during the first 14 days and predation was largely responsible for the mortality of hatchery bass. Further, we noted behavioral differences between wild and hatchery fish that could have contrib uted to high mortality of hatchery fish. After the initial high mortality, survival of hatchery fish equaled that of wild fish through 30 days. If domesticated hatchery fish could be acclimated to wild conditions to modify their behavior closer to that o f wild fish prior to their release, improvements in survival could be expected. Survival rates of juvenile fish surgically implanted with radio tags should be interpreted with caution; they should be considered the minimum survival rate of non
63 tagged individuals (i.e., due to the possibility of mortality related to tag implantation). However, stress due to tagging was nearly equal between treatments, and we demonstrated significant differences in survival between hatchery and wild fish due to predati on. Similar to our results in this study, Ebner and Thiem (2009) used radio telemetry to quantify large differences in survival for hatchery (9%) and wild (95%) trout cod Maccullochella macquariensis at 13 months post release. Dieperink et al. (2001) rad io tagged 50 wild and 50 hatchery sea trout and found that hatchery reared fish had significantly higher mortality than wild fish. Although our mortality at 30 d of 62% for radio tagged hatchery fish in 2010 seems high, Pouder et al. (2010) who studied fo und 95% mortality of advanced fingerling hatchery largemouth bass 90 d post stocking. Hatchery largemouth bass stocked in Lake Carlton (73 fish/ha) simultaneously with our radio tagged fish in 2010 had an estimated 98% mortality one year post stocking (FW C unpublished data). Predation by fish and avian predators was determined to be the sole source of mortality experienced by hatchery fish both years while wild fish appeared to avoid predation. Evidence suggests that antipredator behavior in hatchery re ared fish is not developed (Olla and Davis 1989). In 2010, fish predation was deduced to be responsible for 50% mortality and avian predators were responsible for 14%. This estimate of avian predation may have been higher if the rate of transmitter failu re was not so high. The transmitter located in a live great blue heron was not observed again, suggesting that the bird evacuated the tag outside the study area. Jepsen et al. (1998) researched the survival of radio tagged Atlantic salmon and found pike to be responsible for 56% of the observed mortality and avian predators 31%. Avian
64 predation was shown to be extremely high by Dieperink et al. (2001) when they observed 65% mortality by birds in hatchery reared sea trout. Stein et al. (1981) also found largemouth bass to be significant predators on stocked tiger muskellunge ( Esox masquinongy x E. lucius ), accounting for 26% and 45% mortality of muskellunge stocked in two lakes. In my study, the majority of the observed predation of hatchery fish occurre d during the first 7 days post stocking in both years and all predation occurred during the first 14 days. This suggests that vulnerability to predation is high immediately after release and then survival may reach levels near wild fish. We did not obser ve natural mortality of hatchery fish due to starvation, but starvation could influence vulnerability to predation. Pellet reared hatchery largemouth bass often have exceptional fat reserves upon stocking. If predation due to starvation had occurred, we expect we would have likely seen higher predation after 14 days post stocking when low reserves and fatigue would have rendered hatchery fish more vulnerable. Because initial predation was high and wild fish avoided predation, we suggest that naivety to p redators was the primary cause for the observed high predation. This study demonstrates that advanced fingerling hatchery largemouth bass are capable of large scale dispersal. In 2009, we found that 37% of stocked hatchery bass had dispersed over 700 m by 7 days post stocking. These results are similar to fingerling largemouth bass stocked in Tennessee (Hoffman and Bettoli 2005) where 31% had dispersed over 600 m in one embayment after 7 days. In our study, we also es from 7 to 14 days post stocking, which agrees with previous research suggesting that high initial dispersal will stabilize (Buckmeier and Betsill 2002; Hoffman and Bettoli 2005). Although dispersal for both hatchery and wild
65 fish was highly variable an d differences were not significant, only one wild fish (7%) dispersed over 300 m at 7 days post stocking compared to 26% of hatchery fish. Copeland and Noble (1994) found that age 0 wild largemouth bass in North Carolina tagged with coded wire tags rarely moved from coves where they were tagged. Other research studying wild and hatchery fish movement have shown higher dispersal for hatchery stocked fish in trout cod Maccullochella macquariensis (Ebner and Thiem 2009) and rainbow trout Oncorhynchus mykiss (Bettinger and Bettoli 2002). Our radio telemetry study indicates that hatchery largemouth bass are capable of traveling long distances from their stocking site within one or two weeks and stocking evaluations for largemouth bass should consider this in t heir sampling design. We observed marked differences between the post stocking movements made by hatchery largemouth bass and wild bass. The significantly greater daily movements of the hatchery fish in our study was similar to patterns demonstrated by stocked and wild juvenile chubs (Bolland et al. 2008). In 2010 of my study, higher movement was observed, although not significant due to small sample size, for hatchery fish that died (155 m) compared to fish that survived the study (113 m). This, comb ined with the lower movement observed for wild fish that avoided predation would suggest that higher movement results in higher mortality. Aarestrup et al. (2005) used radio telemetry to investigate the movement and mortality of stocked brown trout and fo und surviving fish had significantly lower mean movement per day than fish that died. I suspect that domestication in hatchery raceways, inability to capture prey, and naivety to predators could contribute to higher rates of movement of hatchery bass than wild bass. These raceway cultured hatchery fish have absolutely no experience living in a natural lake
66 environment, they very likely become disoriented, and they may instinctively attempt to find surroundings that they had become familiar with in the hat chery. Differences in habitat use were detected between wild and hatchery reared largemouth bass during this initial post stocking period. Hatchery fish tended to occupy areas in open water more often than wild fish. Wild fish selected areas of high com plexity along the shoreline in shallower water; possibly to utilize cover to avoid predators, which would also have reduced energy expenditure associated with open water. Bolland et al. (2008) found similar results for juvenile chubs, where hatchery reare d fish spent significantly more time offshore than wild fish (68.2% vs. 24.1%, respectively). Because stocked fish were often located in open water, and most evaluations of stocked hatchery largemouth bass rely on electrofishing in littoral zones, electro fishing could underestimate the contribution or survival of hatchery fish. The offshore habitat use and high movement exhibited by hatchery reared bass could result from the complete lack of habitat complexity that these fish experience while developing in hatchery raceways and/or other domestication issues. The foraging arena theory (Walters and Juanes 1993) describe the continual decisions a fish must make with inherent trade offs between preferred foraging areas and predation risk (Walters and Martell 2004). Fish using offshore habitats may increase vulnerability to predation with reduced available cover along with reduced capture efficiency of prey compared to the vegetated littoral zone. Savino and Stein (1982) found that largemouth bass were unabl e to capture bluegills Lepomis macroc in uniform stem densities of 250 or 100 stems m 2 and were eaten more when venturing into open water (Svino and
67 (Pouder et al. 20 observed high predation mortality. Radio tagged hatchery largemouth bass exhibited significantly slower growth rates compared to wild fish during the course of this experiment. Hatchery largemouth bass reared on artificial diets have been shown to have significantly reduced foraging efficiency (Colgan et al. 1986). This could hinder growth by reducing the total prey consumed during the study period and increasing the energy exp ended on foraging due to a high proportion of failed predation attempts. In a study comparing diets of advanced fingerling largemouth bass to wild bass, Pouder et al. (2010) observed differences in diet similarity and a significantly greater proportion o f hatchery fish had empty stomachs than wild fish at 7 d post stocking, suggesting an inability of the hatchery fish to transition to live prey. In a direct comparison, we observed slow growth for hatchery fish which could suggest foraging inefficiencies. Behavioral differences observed in this study such as high rates of movement and open water habitat used by hatchery bass could have conceivably furthered energetic loses and contributed to slower growth. Wintzer and Motta (2005) demonstrated difference s in skull morphology for largemouth bass raised on artificial versus live feed which could also influence successful predation in the initial stocking period. Predation appeared to be the primary cause of mortality during this study, however fish weakene d by hunger may be more likely to fall victim to predation (Brown and Laland 2001). Thus, an inability to transition to natural prey or lack of recognition of where prey was located may have indirectly contributed to predation. By day 14, 30 and 60, Poud er et al. (2010) saw similar diets
68 between hatchery and wild fish, again suggesting after 14 days, surviving hatchery fish behave and survive similar to wild fish. This study revealed survival and behavioral differences for hatchery largemouth bass comp ared to wild fish, suggesting domestication effects continue to affect stocked hatchery largemouth bass post release. Factors such as high rates of movement, use of open water habitats, reduced growth, and poor foraging efficiency could cause excessive en ergy expenditure and increased vulnerability to predation. Conversely, wild fish had low rates of movement, utilized complex habitats more consistently, and displayed high growth rates; resulting in no observed predation. The results of the wild bass sto cking demonstrated that advanced fingerling sized largemouth bass can have high short term survival after being stocked into a lake with high densities of predators. This suggests that low stocking survival of hatchery reared fish was due to domestication issues and not degraded lake conditions. If hatchery stocked fish survive as well as wild fish in this study, the effectiveness of largemouth bass stocking would likely be improved. We suspect that the experiences that wild fish had during the first yea r of life (awareness of predators, use of complex habitats to avoid predation, foraging efficiency, energy conservation) were responsible for their high survival relative to hatchery fish when stocked into Lake Carlton. Schlechte and Buckmeier (2006) foun d that habituation 60 min before releasing stocked largemouth bass into research from ponds studies to natural lakes, Brennan et al. (2006) found that acclimation (3 d) at release sites has the potential to significantly improve post release survival of juvenile hatchery reared common snook. Opportunities to provide hatchery reared largemouth
69 bass with similar learning experiences should be investigated and could re sult in higher initial survival (Suboski and Templeton 1989; Brown and Laland 2001).
70 Table 4 1. Number of fish radio tagged, total length (mm), weight (g) and tag to body weight ratio (%) for hatchery reared fish (2009 and 2010) and wi ld fish collected from Starke Lake in 2010. Number tagged Total Length (mm) [mean SD (range)] Weight (g) [mean SD (range)] Tag ratio [mean (range)] Hatchery 2009 50 105.05.6 (97 120) 13.62.9 (9.4 21.1) 1.76 (1.14 2.55) Hatchery 2010 30 105.88.1 (93 121) 13.73.5 (8.4 22.2) 1.76 (1.08 2.86) Wild 2010 20 110.76.8 (97 122) 13.42.8 (8.6 19.8) 1.80 (1.21 2.79)
71 Table 4 2. Fate of radio tagged hatchery and wild stocked juvenile largemouth bass at the end of the study period in 2009 and 2010. Actu al numbers are presented in parentheses. Percentage of total released Disappeared Died Survived 2009 Hatchery (14 d) 36% (18) 16% (8) 48% (24) 2010 Hatchery (30 d) 27% (8) 47% (14) 27% (8) 2010 Wild (30 d) 40% (8) 10% (2) 50% (10)
72 Table 4 3. Dispe rsal and movement for radio tagged hatchery (2009 and 2010) and wild (2010) advanced fingerling largemouth bass. Dispersal (7 and 14 day) and movement per day are presented as mean SE (m) with the range in parentheses. Dispersal Movement 7 day 14 day Per day Hatchery 2009 58288 (12 1340) 661121 (11 2010) 12418.1 (3 500) Hatchery 2010 27268 (7 1110) 400172 (7 1528) 7516 (1 456) Wild 2010 15351 (12 703) 18468 (12 685) 2510 (1 211)
73 Table 4 4. Initial and ending total lengths (mm) and weights (g) for radio tagged hatchery and wild largemouth bass stocked into Lake Carlton on 30 March 2010 and collected via electrofishing. Two fish had no weights (Unk). Fish origin Date Collected Initial Measure Ending Measure TL (mm) Weig ht (g) TL (mm) Weight (g) Hatchery 28 April 110 14.2 116 16.1 Hatchery 17 April 115 17.5 118 16.0 Hatchery 17 April 114 15.5 117 18.7 Hatchery 28 April 98 10.2 105 11.6 Hatchery 28 April 112 16.5 122 20.0 Hatchery 28 April 111 15.0 116 17.4 Hatchery 28 April 109 15.0 114 15.2 Wild 13 April 121 18.0 127 Unk Wild 28 April 108 13.3 116 17.8 Wild 28 April 115 13.5 131 23.4 Wild 17 April 112 14.2 120 16.7 Wild 28 April 110 11.6 117 17.7 Wild 27 April 122 19.8 141 31.8 Wild 16 April 112 14.3 120 Unk
74 Figure 4 1. Illustration of primary study area depicting the geographical location and schematic of the water bodies that comprise the Oklawaha Chain of Lakes. Lake Carlton (157 hectares) in Lake County, Florida was used as the st udy lake for radio telemetry.
75 Fig ure 4 2. Satellite image showing the canals connecting Lake Carlton to Lake Beauclair (located at the top of Lake Carlton in this image) and Horseshoe Lake (located at the bottom of Lake Carlton). Points on the m ap also show the locations from 50 stocked advanced fingerling bass as determined by radio telemetry in 2009. Lake Carlton Canal Canal
76 Figure 4 3. Example of the detailed Lake Carlton vegetation map developed using tagged bass are illustrated by black dots in the littoral zone. The vegetation map does not extend to the lake edge during 2010, because the vegetation community was mapped during low water conditions of 2009. The vegetation between the mapped communities and the lake edge, if any, were primarily terrestrial species.
77 Figure 4 4. Photograph of the surgical procedure used to implant a radio transmitter in an advanced fingerling largemouth bass. Photo by Brandon Thompson
78 Figure 4 5. Proportion of tags that remained operational over a 34 day period as measured by live control tags that were allowed to expire in the laboratory and the field tags used in Lake Carlton in 2010.
79 Figure 4 6. Persistence of 30 control radio tags allowed to expire in the lab. The y axis is the probability that a transmitter frequency will continue to emit a signal each day after the battery was started. Transmitter frequenc ies (MHz) tested were 149 MHz (n = 10), 150 MHz (n = 9), and 151 MHz (n = 11). Logrank P = 0.00 02
80 Figure 4 7. Fate of radio tagged hatchery largemouth bass during the 2009 study period. The reporting period began the first day after fish were stocked into Lake Carlt on.
81 Figure 4 8. Survival propabilities over 14 days for radio tagged hatchery largemouth bass stocked in Lake Carlton in 2009.
82 Figure 4 9. Fate of radio tagged hatchery largemouth bass throughout the 2010 study period after being stocked into La ke Carlton.
83 Figure 4 10. Survival propabilities over 30 days for radio tagged hatchery and wild largemouth bass stocked in Lake Carlton in 2010.
84 Figure 4 11. Dispersal of radio tagged hatchery and wild largemouth bass (number of individuals) at 7 and 14 days post stocking for: (a) hatchery 2009; (b) hatchery 2010; and (c) wild 2010 fish.
85 Figure 4 12. Average movement per day of radio tagged hatchery and wild largemouth bass for live locations that survived greater than two days after sto cking.
86 Figure 4 13. Habitat use of hatchery and wild radio tagged largemouth bass stocked in Lake Carlton in 2009 and 2010, reported as the mean proportion of observations that occurred in each habitat type.
87 Figure 4 14. Comparison of percen t body weight (a) and length (b) gained per day between radio tagged wild (n = 7) and hatchery bass (n = 7) upon recapture near the conclusion of telemetry study in 2010.
88 CHAPTER 5 CONCLUSION The objectives of this study were three fold: (1) to determin e the effects of surgically implanting transmitters on growth, survival, and predator avoidance of small bass to validate whether or not tagged animals adequately represent untagged animals; (2) to determine if radio transmitters evacuate through predators after consumption and develop an evacuation model which can be used to predict evacuation rates of tags; (3) to compare the behavior and survival of hatchery reared largemouth bass to similar size wild juvenile bass after being stocked into a lake This study successfully radio tagged some of the small juvenile fish to date. Prior to field application, we were able to document the effects of the surgical process and transmitter presence on the growth and predator avoidance for juvenile largemouth bass. This study confirmed that warm water fish such as largemouth bass could potentially be radio tagged with little effect of their behavior. Using radio tags to estimate predation on juvenile largemouth bass would be appropriate because we saw no difference in predator avoidance between tagged and untagged fish. Depending on weight may not be appropriate for estimating growth. Study design for field application should recognize the potential and limitations for this technology. As the number of telemetry studies increase for juvenile fish that estimate survival from tagged individuals, evacuation of transmitters by potential predators becomes a factor when inferring su rvival and behavior information. In this study, we were able to document that microtransmitters did not accumulate in the intestine of adult largemouth bass that consumed juvenile largemouth bass and that evacuation rates could be
89 estimated relatively acc urately. This allowed more accurate fate determination for largemouth bass that died in the field and allowed accurate behavior data to be applied to the tagged individual. This study document ed behavioral and survival differences between wild and hatch ery reared largemouth bass. Significant differences in predation, movement, habitat use, and growth suggests that fish raised in domestication are less fit to survive initially in the wild. Unless these deficiencies can be reversed by conditioning hatche ry fish to the wild, stocking pellet reared advanced fingerling largemouth bass has limited application to supplement existing populations
90 LIST OF REFERENCES Adams, N. S., D. W. Rondorf, S. D. Evans, and J. E. Kelly. 1998a. Effects of surgically an d gastrically implanted radio transmitters on growth and feeding behavior of juvenile Chinook salmon. Transactions of the American Fisheries Society 127:128 136. Adams, N.S, D. R. Rondorf, S. D. Evans, J. E. Kelly, and R. W. Perry. 1998b. Effects of surgically and gastrically implanted radio transmitters on swimming performance and predator avoidance of juvenile Chinook salmon. Canadian Journal of Fisheries and Aquatic Sciences 55:781 787. Allison, P. D. 2010. Survival analysis using SAS: A pract ical guide, second edition. Cary, NC: SAS Institute Inc. Anglea, A. M., D. R. Geist, R. S. Brown, and K. A. Deters. 2004. Effects of acoustic transmitters on swimming performance and predator avoidance of juvenile Chinook salmon. North American Jou rnal of Fisheries Management 24:162 170. Bams R. A. 1967. Differences in performance of naturally and artificially propagated sockeye salmon migrant fry, as measured with swimming and predation tests. Journal of the Fisheries Research Board of Canada 24:1117 1153. Beamish, F. W. H. 1972. Ration size and digestion in largemouth bass, Micropterus salmoides Lacepede. Canadian Journal of Zoology. 50:153 164. Benton, J. 1999. Lake Griffin Fisheries Improvement, Wallop Breaux Project 6150 Annual Re port, Florida Fish and Wildlife Conservation Commission, Eustis, FL 18 pp Benton, J., D. Douglas, and L. Prevatt. 1991. Fisheries studies of the Ocklawaha Chain of Lakes, Wallop Breaux Project F 30 18 Completion Report, Florida Fish and Wildlife Con servation Commission, Eustis, FL 38 pp Bettinger, J. M. and P. W. Bettoli. 2002. Fate, dispersal, and persistence of recently stocked and resident rainbow trout in a Tennessee tailwater. North American Journal of Fisheries Management 22:425 432. Bo lland J. D., I. G. Cowx, and M. C. Lucas. 2008. Movements and habitat use of wild and stocked juvenile chub, Leuciscus cephalus (L.), in a small lowland river. Fisheries Management and Ecology 15:401 407. Boxrucker, J. 1986. Evaluation of supplement al stocking of largemouth bass as a management tool in small impoundments. North American Journal of Fisheries Management 6:391 396.
91 Brennan, N. P., M. C. Darcy, and K. M. Leber. 2006. Predator free enclosures improve post release survival of stocked common snook. Journal of Experimental Marine Biology and Ecology 335:302 311. Brown, C. and K. Laland. 2001. Social learning and life skills training for hatchery reared fish. Journal of Fish Biology 59:471 493. Brown, R. S., S. J. Cooke, W. G. Anderson, and R. S. McKinley. 1999. Evidence to Management 19:867 871. Buckmeier, D. L. and R. K. Betsill. 2002. Mortality and dispersal of stocked fingerling largemou th bass a nd effects of cohort abundance. Pages 667 676 in D. P. Phili pp and M. S. Ridgway, editors. Black bass: ecology, conservation, and management. American Fisheries Society Symposium 31, Bethesda, Maryland. Buckmeier, D. L., R. K. Betsill, and W. J. Schlechte. 2005. Initial predation of stocked fingerling largemouth bass in a Texas reservoir and implications for improving stocking efficiency. North American Journal of Fisheries Management 25:652 659. Busacker, G. P., I. R., Adelman and E. M. Goolish. 1990. Growth. Pages 363 387 in C. B. Schreck and P. B. Moyle, editors. Methods for fish biology. American Fisheries Society, Bethesda, Maryland. Buynak, G. L. and B. Mitchell. 1999. Contribution of stocked advanced fingerling larg emouth bass to the population and fishery at Taylorsville Lake, Kentucky. North American Journal of Fisheries Management 19:494 503. Cardeilhac, P., K. Childress, H. Townsend, N. Szab o, D. Samuelson, and R. Stout. 2008. Pages 133 134 in T. Reidarson, ed itor. Dietary associated incidence of hepatic lesions and tumors in largemouth bass Micropterus salmoides floridanus 39 th Annual Conference Proceedings of the International Association for Aquatic Animal Medicine. Omnipress Publishing, Madison, Wiscons in. Cochran, P. A. and I. R. Adelman. 1982. Seasonal aspects of daily ration and diet of largemouth bass with an evaluation of gastric evacuation rates. Environmental Biology of Fishes 7:265 275. Colgan, P. W., J. A. Brown, and S. D. Orsatti. 1986. Role of diet and experience in the development of feeding behavior in largemouth bass. Journal of Fish Biology 28:161 170.
92 Collis, K., D. D. Roby, C. P. David, B. A. Ryan and R. D. Ledgerwood. 2001. Colonial waterbird predation on juvenile salmo nids tagged with passive integrated transponders in the Columbia River estuary: vulnerability of different salmonid species, stocks, and rearing types. Transactions of the American Fisheries Society 130:385 396. Cooke, S. J., F. D. S. Graeb, C. D. Susk i, K. G. Ostrand. 2003. Effects of suture material on incision healing, growth and survival of juvenile largemouth bass implanted with miniature radio transmitters: case study of a novice and experienced fish surgeon. Journal of Fish Biology 62:1366 1 380. Copeland, J. R. and R. L. Noble. 1994. Movements by young of year and yearling largemouth bass and their implications for supplemental stocking. 1994. North American Journal of Fisheries Management 14:119 124. Cote, D., Scruton, D. A. Scruton, Lloys Cole, and R. S. Mckinley. 1999. Swimming performance and growth rates of juvenile Atlantic cod intraperitoneally implanted with dummy acoustic transmitters. North American Journal of Fisheries Management 19:1137 1141. Crawford, S. and A. M. Wi cker. 1987. Recruitment of stocked largemouth bass fingerlings into a central Florida fishery. Florida Scientist 4:211 215. Diana, M. J. and D. H. Wahl. 2009. Growth and survival of four sizes of stocked largemouth bass. North American Journal of Fisheries Management 29:1653 1663. Dieperink, C., S. Pedersen, and M. I. Pedersen. 2001. Estuarine predation on radio tagged wild and domesticated sea trout smolts. Ecology Freshwater Fish 10:177 183. Ebner, B. C. and J. D. Thiem. 2009. Monitor ing by telemetry reveals differences in movement and survival following hatchery or wild rearing of an endangered fish. Marine and Freshwater Research 60:45 57. Ebner, B. C., J. D. Thiem, and M. Lentermans. 2007. Fate of 2 year old hatchery reared t rout cod Macullochella macquariensis (Percicthyidae) stocked into two upland rivers. Journal of Fish Biology 71:182 199. Frost, D. A., R. L. McComas, and B. P. Sandford. 2010. The effects of a surgically implanted microacoustic tag on growth and survi val in subyearling fall Chinook salmon. Transaction of the American Fisheries Society 139:1192 1197.
93 Hanson, K. C., S. J. Cooke, C. D. Suski, G. Niezgoda, F. J. S. Phelan, R. Tinline, and D. P. Philipp. 2007. Assessment of largemouth bass behavior and activity at multiple spatial and temporal scales utilizing a whole lake telemetry array. Hydrobiologia 582:243 256. He, E. and W. A. Wurtsbaugh. 1993. An empirical model of gastric evacuation rates for fish and an analysis of digestion in pisciv orous brown trout. Transactions of the American Fisheries Society 122:717 730. Heggberget, T. G., M. Staurnes, R. Strand, and J. Husby. 1992. Smoltification in salmonids. NINA (Norsk Institutt for Naturforskning) Forskningsrapport 31:3 42. Henson, F. G. and R. M. Newman. 2000. Effect of temperature on growth at ration and gastric evacuation rate of ruffe. Transactions of the American Fisheries Society 129:552 560. Hightower J. E., J. R. Jackson and K. H. Pollock. 2001. Use of telemetry meth ods to estimate natural and fishing mortality of striped bass in Lake Gaston, North Carolina. Transactions of the American Fisheries Society 130:557 567. Hoffman, K. J. and P. W. Bettoli. 2005. Growth, dispersal, mortality, and contribution of large mouth bass stocked into Chickamauga Lake, Tennessee. North American Journal of Fisheries Management 25:1518 1527. Hoxmeier, R. J. H., and D. H. Wahl. 2002. Evaluation of supplemental stocking of largemouth bass across reservoirs: effects of predatio n, prey availability, and natural recruitment. Pages 639 647 in D. P. Philipp and M. S. Ridgeway, editors. Black bass: ecology, conservation, and management. American Fisheries Society Symposium 31, Bethesda, Maryland. Hunt, B. P. 1960. Digestion rate and food consumption of Florida gar, warmouth, and largemouth bass. Transaction of the American Fisheries Society 89:206 211. Huntingford F. A. 2004. Implications of domestication and rearing conditions for the behavior of cultivated fish. Jour nal of Fish Biology 65:122 142. Jepsen N., K. Aarestrup, Finn Okland, and G. Rasmussen. 1998. Survival of radio tagged Atlantic salmon (Salmo salar L.) and trout (Salmo trutta L.) smolts passing a reservoir during seaward migration. Hydrobiologia 371 /372:347 353. Jepsen, N., K, A., Thorstad, E.B. and Baras, E. 2002. Surgical implanting of telemetry transmitters in fish. How much have we learned? Hydrobiologia 483:239 248. Johnson, D. H. 1999. The insignificance of statistical significance test ing. The Journal of Wildlife Management 63:763 772.
94 Knight, B. C. and B. A. Lasee. 1996. Effects of implanted transmitters on adult bluegills at two temperatures. Transactions of the American Fisheries Society 125:440 449. Koed A., N. Jepsen, K. Aar estrup, and C. Nielsen. 2002. Initial mortality of radio tagged Atlantic salmon (Salmo salar L.) smolts following release downstream of a hydropower station. Hydrobiologia 483:31 37. Loska, P. M. 1982. A literature review on the stocking of black ba sses (Micropterus spp.) in reservoirs and streams. Georgia Department of Natural Resources, Rederal Aid in Sport Fish Restoration, Projec t SW 1, Final Report, Atlanta. Lovvorn, J. R., D. Yule, and C. E. Derby. 1999. Greater predation by double crested cormorants on cutthroat trout versus rainbow trout fingerlings stocked in a Wyoming river. Canadian Journal of Zoology 77:1984 1990. Lucas, M. C. 1989. Effects of implanted dummy transmitters on mortality, growth and tissue reaction in rainbow trou t Journal of Fish Biology 35:577 587. Ludsin, S. A. and D. R. DeVries. 1997. First year r ecruitment of largemouth bass: the interdependency of early life stages. Ecological Applications 7:1024 1038. Martinelli, T. L., H. C. Hansel, and R. S. Shive ly. 1998. Growth and physiological responses to surgical and gastric radio transmitter implantation techniques in subyearling chinook salmon (Oncorhynchus tshawytscha). Hydrobiologia 371/372:79 87. Mesa, M. G., T. P. Poe, D. M. Gadomski, and J. H. Pet ersen. 1994. Are all prey created equal?: a review and synthesis of differential predation on prey in substandard condition. Journal of Fish Biology 45:81 96. Mesing C. L. and A. M. Wicker. 1986. Home range, spawning migration, and homing of radi o tagged Florida largemouth bass in two central Florida lakes. Transactions of the American Fisheries Society 115:286 295. Mesing, C. L., R. L. Cailteux, P. A. Strickland, E. A. Long, and M. W. Rogers. 2008. Stocking of advanced fingerling largem outh bass to supplement year classes in Lake Talquin, Florida. North American Journal of Fisheries Management 28:1762 1774. Nelson T. C., M. L. Rosenau, N. T. Johnston. 2005. Behavior and surviva l of wild and hatchery origin winter steelhead spawners caught and released in a recreational fishery. North American Journal of Fisheries Management 25:931 943.
95 Niva, T. and P. Hyvarinen. 2001. Evacuation rates of coded wire tags implanted in prey of northern pike. North American Journal of Fisheries Management 21:692 695. Olla, B. L. and M. W. David. 1989. The role of learning and stress in predator avoidance of hatchery reared coho salmon juveniles. Aquaculture 76:209 214. Patel Wintzer, A. 2004. Behavioral and morphological consequences of rearing Florida largemouth bass with non elusive prey. University of South Florida, Petersen, J. H. and C. A. Barfoot. 2003. Evacuation of passive integrated transponder (PIT) tags from northern pikeminnow c onsuming tagged juvenile Chinook salmon. North American Journal of Fisheries Management 23:1265 1270. Pine III, W. E., S. A. Ludsin and D. R. DeVries. 2000. First summer survival of largemout h bass cohorts: is early spawning really best? Transaction s of the American Fisheries Society 129:504 513. Porak, W. F., W. E. Joh n son, S. Crawford, D. J. Renfro, T. R. Schoeb, R. B. Stout, R. A. Krause, and R. A. DeMauro. 2002. Factors affecting survival of largemouth bass raised on artificial diets and s tocked into Florida lakes. American Fisheries Society, Symposium 31:649 665. Post, D. M., J. F. Kitchell, and J. R. Hodgson. 1998. Interactions among adult demography, spawning date, growth rate, predation, overwinter mortality, and the recruitment of largemouth bass in a northern lake. Canadian Jo urnal of Fisheries and Aquatic Sciences 55:2588 2600. Pouder, W. F., N. A. Trippel, J. R. Dotson. 2010. Comparison of mortality and diet composition of pellet reared advanced fingerling and early coh ort age 0 wild largemouth bass through 90 days post stocking at Lake Seminole, Florida. North American Journal of Fisheries Management 30:1270 1279. Robertson, M. J., D. A. Scruton, and J. A. Browns. 2003. Effects of surgically implanted transmitter s on swimming performance, food consumption and growth of wild Atlantic salmon parr. Journal of Fish Biology 62:673 678. Ross, M. J. and Kleiner, C.F. 1982. Shielded needle technique for surgically implanting radio frequency transmitters in fish. Pro gressive Fish Culturist 44:41 43. Savino, J. F. and R. A. Stein. 1982. Predator prey interacti on between largemouth bass and bluegills as influenced by simulated, submersed vegetation. Transactions of the American Fisheries Society 111:255 266.
96 Savino, J. F. and R. A. Stein. 1989. Behavior of fish predators and their prey: habitat choice between open water and dense vegetation. Environm ental Biology of Fishes 24:287 293. Schlechte, W. J. and D. L. Buckmeier. 2006. A pond evaluation of h abituation as a means to reduce initial mortality associated with poststocking predation of hatchery reared largemouth bass. North American Journal of Fisheries Management 26:119 123. Stein, R. A. R. F. Carline, and R. S. Hayward. 1981. Largemouth bass predation on stocked tiger muskellunge. Transactions of the American Fisheries Society 110:604 612. Suboski, M. D. and J. J. Templeton. 1989. Life skills training for hatchery fish: Social learning and survival. Fisheries Research 7: 343 352. Walsh, M. G., A. K. Bjorgo, and J. J. Isely. 2000. Effects of implantation methods and temperature on mortality and loss of simulated transmitters in hybrid striped bass. Transactions of the American Fisheries Society 129:539 544. Walters, C. J. and F. Juanes. 1993. Recruitment limitation as a consequence of natural selection for use of restricted feeding habitats and predation risk taken by juvenile fish. Canadian Journal of Fisheries and Aquatic Science 50:2058 2070. Walters, C. J., and S. J D. Martell. 2004. Fisheries Ecology and Management Princeton Publishing, Princeton, New Jersey. Wicker, A. M. and W. E. Johnson. 1987. Relationship among fat content, condition factor, and first year survival of Florida largemouth bass. Transact ions of the American Fisheries Society 116:264 271. Winter, J. D. 1996. Advances in underwater biotelemetry. Pages 555 590 in B. R. Murphy and D. W. Willis, editors. Fisheries Techniques, 2 nd edition. American Fisheries Society, Bethesda, Maryland. Wintzer, A. P. and P. J. Motta. 2005. Diet induced phenotypic plasticity in the skull morphology of hatchery reared Florida largemouth bass, Micropterus salmoides floridanus. Ecolog y of Freshwater Fish 14:311 318
97 BIOGRAPHICAL SKETCH Brandon Charles T hompson was born in Wausau, Wisconsin in 1979. His love of the outdoors, passion for fishing, and fear of an office job led him to seek a career in fisheries science. In 2003, he earned a BS. in Fish and Wildlife Sciences at the University of Wisconsin S tevens Point. He reinforced his drive to succeed in fisheries while working fisheries internships through college for the Wisconsin DNR and US Forest Service. After graduating from Stevens Point, Brandon put off graduate school to gain experience and ens ure this was the field he wanted to make a career. He took a temporary job with the Wisconsin DNR restoring trout habitat and he realized that he could not only work in a job he loved, but also make a positive difference in the natural resources around hi m. He then accepted a position with the U.S. Fish and Wildlife Service in California monitoring salmon and steelhead populations. He finally moved to Florida to work for the Florida Fish and Wildlife Conservation Commission as a fisheries biologist. Af ter working three years, Brandon decided that fisheries career was indeed what he desired and a graduate degree would help him conduct his work at a higher level. He was able to start taking courses in the fall of 2009 under Dr. Mike Allen while working f hatchery largemouth bass. Brandon completed his courses and research in 2012.