1 INTERACTIONS BETWEEN FISH AND ANGLERS: A SPATIAL ANALYSIS OF FISH VULNERABILITY TO ANG LING By BRYAN GLEN MATTHIAS A THESIS PRESENTED T O THE GRADUATE SCHOO L OF THE UNIVERSITY OF FLORIDA IN PART I AL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORID A 2012
2 2012 Bryan Glen Matthias
3 To my parents and fiance
4 ACKNOWLEDGMENTS I would like to thank my advisor, Dr. Mike Allen for help guiding me through my graduate career and for direction and assistance with this project. I would also like to thank my committee members, Dr. Rob Ahrens and Dr. T. Douglas Beard. I would also li ke to thank the Allen lab members, J Kerns, K Wilson, E Bradshaw Settevendemio, Z Slagle, and N Cole for their assistance conducting the field work. Additionally, I would like to thank D Gwinn and M Hangsleben for their assistance and guidance in h elping me develop the skills needed to complete the analysis. Funding for this project and my graduate research assistantship was provided by the Florida Fish and Wildlife Conservation Commission (FWC).
5 TABLE OF CONTENTS ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 6 LIST OF FIGURES ................................ ................................ ................................ .......... 7 ABSTRACT ................................ ................................ ................................ ..................... 8 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 10 2 METHODS ................................ ................................ ................................ .............. 14 Study Area ................................ ................................ ................................ .............. 14 Angler Distributio n ................................ ................................ ................................ .. 14 Fish Sampling ................................ ................................ ................................ ......... 15 Habitat Sampling ................................ ................................ ................................ .... 17 Analysis ................................ ................................ ................................ .................. 18 Tag Returns ................................ ................................ ................................ ............ 23 3 RESULTS ................................ ................................ ................................ ............... 27 4 DISCUSSION ................................ ................................ ................................ ......... 39 5 MANAGEMENT IMPLICATIONS ................................ ................................ ............ 44 LIST OF REFERENCES ................................ ................................ ............................... 46 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 52
6 LIST OF TABLES Table page 2 1 Delta AIC scores and model weights for binomial generalized linear models predicting angler distribution. ................................ ................................ .............. 31 2 2 Delta AICc table and model weights testing for exchange rates of fish between areas vulner able and invulnerable to a ngling using the program MARK ................................ ................................ ................................ ................. 32 2 3 Biweekly survey estimates of exchange rates between areas vulnerab le (Vul) and invulnerable (Invul) to angling derived from the MARK model incorporating habitat preference of the fish and heterogenic movement ............ 33 2 4 Within day survey estimates of exchange rates between areas vulnerable (Vul) and invulnerable (Invul) to angling from April, June, and August ............... 34 2 5 Catch data of the tagged fish divided into fish habitat preference categories ..... 35
7 LIST OF FIGURES Figure page 2 1 Bathymetric map of Lake Santa Fe, Florida ................................ ....................... 24 2 2 Representation of a recreational fishery ................................ ............................. 25 2 3 Proportion of time spent onshore for each fish broken up into habitat selection groups ................................ ................................ ................................ 26 3 1 Distribution of largemouth bass anglers (red dots) and non bass anglers (black dots) on Lake Santa Fe from November 2010 through October2011 ...... 36 3 2 Fitted general linearized model predicting the relative likelihood that largemouth bass anglers will target a given area based on distance from vegetation and rugosity sc ore of the onshore habitat ................................ ......... 37 3 3 All fish locations from October 2010 through October 2011 ............................... 38
8 A bstract of Thesis Presented to the Graduate School Of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science INTER ACTIONS BETWEEN FISH AND ANGLERS: A SPATI AL ANALYSIS OF FISH VULNERABILITY TO ANG LING By: Bryan G. Matthias August 2012 Chair: Micheal S. Allen Major : Fisheries and Aquatic Sciences Most stock assessment models assume fish populations are comprised of fish that are equally vulnerable to angling. However, vulnerability to angling is influenced by genetic traits, habitat selection, movement patterns and angler behavior. Ideal free distribution (IFD) predicts that angler effort will mirror fish distribution, but tests of this hypothesis are rare in the literature. A field study was conducted comparing spatial patterns in recreational angler effort to fish distribution. A total of 313 largemouth bass anglers were mapped with over 90% fish ing within 50 m from shore, indicating fish located offshore were relatively invulnerable to angling Using radio telemetry, 81 largemouth bass Micropterus salmoides were tracked from November 2010 through October 2011 Three types of habitat preference were observed in tagged fish, onshore, offshore, and generalist. Biweekly and within day e xchange rates of fish between areas vulnerable and invulnerable to angling were estimated using program MARK E xchange rate s estimated from biweekly surveys indicat e onshore fish remained vulnerable to angling for the majority of the year. Conversely, o ffshore fish remained invulnerable to angling for a significant portion of the year G eneralist fish
9 likely did not remain invulnerable to angling due to high exchange rates between vulnerable and invulnerable locations Exchange rates estimated from within day surveys indicated that fish spending most of their time in invulnerable areas were actually vulnerable to angling due to high exchange rates. R esults f rom tag return data show ed that anglers caught 59% of the onshore fish, 47% of the offshore fish, and 65% of the generalist fish. T he distribution of effort and exchange rates estimated from biweekly surveys predict ed that a portion of the population was invulnerable to angling However, exchange rates estimated from within day surveys and results from angler tag returns indicate d all largemouth bass were equally vulnerable to angling.
10 CHAPTER 1 INTRODUCTION In open access fisheries w h ere fishing effor t is not controlled by regulations (e.g. most recreational fisheries), assessing fish stocks is vital to fisheries management to prevent stocks from being overfished Fisheries managers use data on fish populations and computer models to perform stock ass essments, which evaluate the impact of fishing on fish population age structure and abundance. Aside from size selectivity, m ost stock assessment models assume that fish populations are comprised of fish that are equally vulnerable to angling (Cox and Walters 2002 ) However t here are many factors that influence vulnerability of fish including genetic traits, seasonal and diel movements, habitat selection, and angler behavior (Warden and Lorio 1975; Hilborn and Ledbetter 1979; Mesing and Wicker 1986; Nuhfer and Alexander 1994; Snedden et al. 1999; Scheuerell and Schindler 2003; Cooke et al. 2007; Kobler et al. 2009; Philipp et a l. 2009; Suski and Ridgway 2009). Vulnerability to angling is believed to be influenced by genetic traits such as aggression and growth (Bryan and Larkin 1972; Nuhfer and Alexander 1994; Cooke et al. 2007). In nest guarding species such as largemouth ba ss Micropterus salmoides vulnerability to angling was found to be positively related to aggression in nest guarding males (Cooke et al. 2007). Brook trout Salvelinus fontinalis and cutthroat trout Oncorhynchus clarkii with higher growth rates were also f ound to be more vulnerable to angling than slower growing fish (Dwyer 1990; Nuhfer and Alexander 1994). Additionally, the degree of domestication was found to influence the vulnerability to angling in cutthroat trout and rainbow trout O mykiss (Brauhn an d Kinciad 1982; Dwyer 1990). Individuals selected for higher growth rates were more vulnerable to angling
11 than other domesticated strains of cutthroat trout and rainbow trout (Brauhn and Kinciad 1982; Dwyer 1990). However, it has also been observed that largemouth bass selected for vulnerability to angling had lower growth rates and higher metabolic rates tha n largemouth bass selected against vulnerability to angling (Redpath et al. 2009). Because largemouth bass selected for vulnerability to angling had higher metabolic rates, they required higher consumption rates to attain growth rates similar to largemouth bass selected against vulnerability to angling (Brett and Groves 1979; Jobling 1985). Therefore fish spend ing more time foraging are more vulnerab le to angling because these fish have a greater chance of encountering lures and a greater chance of react ing to lures (Cooke et al. 2007; Biro and Post 2008 ; Redpath et al. 2009 ). H abitat selection can also influence vulnerability to angling. Various habitat selection traits have been observed in arctic charr S. alpinus (Sandlund et al. 1987), lake trout S. namaycush (Morbey et al. 2006), juvenile sockeye salmon O. nerka (Scheuerell and Schindler 2003), and brook trout (Bourke et al. 1997) that correspond with foraging areas. A portion of largemouth bass, northern pike Esox lucius and lake trout populations utilize d onshore habitat wh ereas other fish used offshore habitat (Mesing a nd Wicker 1986; Morbey et al. 2006; Kobler et al. 2009). If anglers predominantly target one type of habitat, a portion of some fish populations could be relatively invulnerable to angling Additionally, a ngler behavior can influence the vulnerability of fish. The ideal free distribution (IFD ) predicts that angling effort will mirror the distribution of fish (Fretwell and Lucas 1970; Fretwell 1972) The IFD has been used to model both commercial and
12 artisanal fisheries ( Gillis et al. 1993 ; Gillis 2003; S wain and Wade 2003; Voges et al. 2005; Abernethy et al 2007 ). The IFD has worked well at predicting the distribution of commercial fisheries (Gillis et al. 1993; Gillis 2003; Swain and Wade 2003; Voges et al. 2005 ). However, for artisanal fisheries the IFD has not worke d well because these fisheries did not meet key assumptions (Abernethy et al. 2007) such as anglers must have perfect knowledge of the fishery (Fretwell and Lucas 1970; Fretwell 1972). I n artisanal fisheries, lack of knowledge prevented fishers from targeting locations with highest fish densities (Abernethy et al. 2007). In many recreational fisheries c atch inequality occurs when a small portion of anglers catch most of the fish ( Baccante 1995; van Poorten and Post 2005; Seekell et al. 2 011). C atch inequality implies that not all of the anglers have the same knowledge of the fishery and thus, not all anglers would be fishing in areas of highest fish abundances The IF D also assumes that the primary goal of an angler is to catch fish ( Fretwell and Lucas 1970; Fretwell 1972) This assumption may be appropriate for modeling commercial fisheries, however may not be appropriate for recreational fisheries According to the economic theory, the distribution of effort can be characterized by the perceived economic returns to anglers from fishing in alternative locations (Holland and Sutinen 1999) where economic returns to anglers from fishing can be both catch related and non catch related. Catch related returns, such as total catch, trop hy catch or harvest (Post et al. 2008; Hunt et al. 2011) vary between angler demographics and target species (Arlinghaus et al. 2008). N on catch related returns, such as travel costs, aesthetics, or remoteness (Parkinson et al. 2004; Post et al. 2008; Hu nt et al. 2011) have been found to be more important to anglers than catch related returns (Driver and
13 Knopf 1976; Fedler and Ditton 1994; Ditton 2004; Arlinghaus 2006). These violations of the IFD could cause differences in the vulnerability of fish and depending on the knowledge of anglers and non catch related returns to anglers from fishing. Previous studies show that there are various habitat selection traits in fish population s but it is unknown whether t he spatial distribution of angler effort ac tually mirrors the distribution of fish. The primary objective of this study was to determine if there is a portion of a fish population invulnerable to angling due to either angler or fish behavior. The second objective was to determine the exchange rat e of fish between areas that are targeted by anglers and areas that are not targeted by anglers. High exchange rates would indicate that all fish may be vulnerable to fishing even if angler effort does not spatially overlap with fish distribution. Finall y, data from angler tag returns were used to empirically test if there was a portion of the population that remained invulnerable to angling.
14 CHAPTER 2 METHODS Study Area Lake Santa Fe (27.74N, 82.07W) is located in north central Florida. The lake is comprised of two basins, a main basin of 1,873 ha (Florida Lakewatch 2005a) and a 577 ha northern basin (Florida Lakewatch 2005b), which is also known as Little Lake Santa F e (29.77N, 82.09W). The main lake reaches a maximum depth of 8.1 m and has an average depth of 4.9 m, and the northern basin has a maximum depth of 6.3 m and an average depth of 3.6 m (Figure 2 1 ). There are eleven canals around the lake seven in the main basin and four in the northern basin There is a thin band of emergent vegetation around the perimeter consisting primarily of maidencane Panicum hemitomon, bald cypress Taxodium distichum spatterdock Nuphar luteum and giant bulrush Scirpus califor nicus Angler Distribution T he location of anglers was sampled to evaluate the spatial distribution of fishing effort from November 201 0 through October 2011 S urveys were generally conducted on two non random weekdays and one random weekend day every two weeks. Weekday surveys were usually conducted on the first Monday and following Wednesday every two weeks due to the fish tracking schedule. Weekend surveys were usually conducted on a random weekend day following th e Monday/Wednesday survey. S urveys were conducted at random times during the time period of one hour after sunrise until one hour before sunset. Specific sample times on each day were chosen using a stratified random design by randomly selecting morning, afternoon, or evening with equal pr obability (P = 0.33). Within each time of day stratum, specific angler count
15 time was selected randomly within that period Starting locations at the lake for the angler survey were also randomly selected Surveys were conducted by driving around the pe rimeter of the lake to locate anglers near shore O ffshore anglers were located by running transects through the middle of the lake. Once a boat or person along the shore line was found, I determined if they were fishing. For someone to be considered fi shing, they must have a fishing line in the water switching tackle, or be dealing with a fish. D istance and bearing to anglers were obtained from a TruPulse 360B laser rangefinder and was entered into the Trimble Recon using Tripod Data Systems SOLO Fiel d Software. All anglers on the same boat or dock were given the same location. Anglers were classified as bass anglers or non bass anglers depending on observed fishing techniques. Starting in March 2011 subsets of anglers were also interviewed to determine the accuracy of visual estimation of the species they were targeting. From November 2010 through May 2011 the canals around the lake were sampled for angler distribution three times a month, two random weekdays and one random weekend day. F ro m June through October the canals were not sampled due to low water levels that precluded both angling and surveys. Fish Sampling The s patial distribution of largemouth bass was measured with radio telemetry. Fish were captured using electrofishing and angling in the fall of 2010. Electrofishing was conducted along the shoreline and angling was used to collect fish from water greater than three meters deep or at least 50 m from the edge of the vegetation. Offshore fish were targeted in an attempt to ob tain fish that were invulnerable to electrofishing during the tagging process L argemouth bass >350 mm were implanted
16 with Advanced Telemetry Systems (ATS) F1835 transmitters following the recommendations of Winter (1996). The transmitters had a 502 day life expectancy and weighed 14 g. Radio tags and surgery equipment were sterilized prior to surgery using isopropyl alcohol and implanted into the body cavity through a ventral incision. Two or three sutures were used to close the incision after which c yanoacrylate adhesive was applied to the incision and exposed sutures following the procedures outlined in Dutka Gianelli et al. ( 2011) Once the adhesive dried, the incision and surrounding area was covered with an antibiotic ointment. The fish were als o tagged with a n external reward tag ($200) to obtain angler catch data on the tagged fish Fish were allowed to recover in an aerated holding tank before being released near the capture location. A survey was conducted on Mondays once every two weeks t o track every fish. If a fish was not found during this tracking event, a second survey usually on the following Wednesday, was conducted by scanning the whole lake to find the fish. Random subsets of 30 fish were also tracked an additional time every two weeks on a random weekend day. Fish not found within the previous 30 days were not included in the weekend survey. Start locations for the general fish survey and the weekend survey were chosen at random to avoid finding fish at the same time during every tracking event. All tagged fish were tracked using ATS R410 receivers with hand held yagi antennae and locations of the fish were recorded using GPS receivers. More detai led temporal scale sampling was also conducted to assess within day fish movements. B etween 8 and 1 4 fish were tracked once every two hours during hours of safe light approximately 1 hour before sunrise to 1 hour after sunset A total of three surveys w ere conducted in April, June, and August. Fish were not tracked at night
17 due to low levels of angling effort at night. By tracking fish on a detailed temporal scale, I was able to evaluat e how effective the biweekly surveys were at classifying fish habit at use and fish movement patterns Habitat Sampling H abitat characteristics were measured to explain the spatial distribution of largemouth bass anglers and largemouth bass. The bathymetry of Lake Santa Fe was surveyed using a L o wrance LCX 28cHD and mapped in ArcGIS 10 (Figure 2 1 ). The outside edge of vegetation was mapped by taking waypoints every 10 m along the outside edge of vegetation. The shoreline and vegetated areas were sampled every 50 to 500 m A t each of these sampling points, the slope of the shoreline was surveyed and width of the vegetated area, presence of bald cypress, presence of spatterdock, presence of giant bulrush rugosity of the vegetated area and the presence of manmade structure we re recorded Slope of the shoreline was calculated using: ( 2 1 ) where depth of water was surveyed at approximately 40 m from the edge of vegetation and D Shore was the distance from the shoreline to the location where de pth was surveyed. The width of the vegetated area was measured as the distance from the shoreline to the outside edge of vegetation. For vegetation not connected with the shoreline, t he width of vegetation was determined using the average distance across the vegetated area Rugosity a qualitative measure ment of the complexity of vegetat ed areas, was visually assessed on a scale of 1 to 10 Habitat with a rugosity score of 1 represent ed a straight line of vegetation and habitat with a rugosity score of 10 represent ed a complex mosaic of plants with many patches channels and undulations
18 Analysis Martin (1958) and more recently Cox and Walters (2002) hypothesized that there is a portion of a population that is not vulnerable to angling at any given t ime (Figure 2 2 ). Under this assumption catch is comprised only of the fish from the vulnerable portion of a population (Cox and Walters 2002). Cox and Walters (2002) proposed that fish can be either vulnerable V t or invulnerable I t to angling at time t as per ( 2 2 ) where N t is the number of fish in a population at time t Vulnerable fish may be found in locations that anglers target and are willing to respond to fishing gear. Invulnerable fish are either in areas not targeted by anglers or do not react to angling gear (Cox and Walters 2002). F ish can move between vulne rable and invulnerable states by moving to different areas or changing their reactivity to lures using the form: ( 2 3 ) ( 2 4 ) where k 1 is the rate at which fish move from a vulnerable state to an invu lnerable state; k 2 is the rate at which fish move from an invulnerable state to a vulnerable state; and k 3 is the harvest rate (Cox and Walters 2002). If k 1 and k 2 are high, then the whole population is essentially vulnerable to angling because individual invulnerable for long periods of time. However, if k 1 and k 2 are low, angler catch and the fishing mortality rate would be based on only a subset of the entire fish stock. To test if there were spatial differences in the vulnerability to an gling of fish based on the distribution of effort, general ized linear model s (GLM s ) were used to analyze the distribution of largemouth bass anglers. Binomial GLMs were constructed to predict the
19 probability that largemouth bass angler s will target a given area using p resence and absence of largemouth bass angler s Because multiple largemouth bass anglers fishing from one platform (e.g. boat, dock, or section of shoreline) were not independent observations, largemouth bass anglers on th e same platform were classified as a single presence observation. Additionally, I generated pseudo absence data points because I did not measure locations where largemouth bass anglers were not observed. One thousand pseudo absence locations were randoml y generated within Lake Santa Fe to represent locations where largemouth bass anglers were not found. Generating pseudo absence data is a typical approach in species distribution modeling where presence only data is common (Warton and Shepherd 2010). Inc luding pseudo absence points enables presence only data to be analyzed using techniques developed to analyze presence absence data (Pearce and Boyce 2006), such as binomial GLM s Continuous predictor variables used in the model were: distance from shore, distance from vegetation, depth, the width of the vegetated area and slope of the shoreline Discrete predictor variables used in the model were: rugosity of the vegetat ed areas presence of manmade structure, presence of cypress trees, presence of spatterdock, and presence of bulrush. Rugosity of the vegetated areas was classified as a discrete predictor variable because rugosity was on a scale from 1 to 10. Discrete variables were used to test if there were differences in vulnerability of fish to angling within the various vegetated habitat types If largemouth bass anglers or pseudo absence points were located within 50 m from t he outside edge of vegetation, they were classi fied as being located in onshore habitat L argemouth bass a nglers or pseudo absence points that were located greater than 50 m from the edge of vegetation were classified as
20 being in offshore habitat. Fifty meters was chosen as the cutoff for targeting o nshore habitat because it seem ed unlikely that an angler would consistently cast farther than 50 m to target an area. Presence and pseudo absence points in offshore habitat were given values of zero for the width of vegetated area slope of the shoreline and the rugosity score of the vegetated areas Using Akaike Information Criteria (AIC; Akaike 197 4 ), the full model was compared to a variety of reduced models Significant variables from the full model were used to create a set of r educed models The best model was selected s of less than 10 To determine the vulnerability of the fish to anglin g e ach fish location was classified using the same criteria to predict the distribution of largemouth bass anglers (e.g. distance from vegetation, distance from sh ore, depth, etc.). T he best GLM used to describ e angler distribution was applied to each fish location to predict the likelihood that a largemouth bass angler would target the fish location. B ecause pseudo absence points were used instead of actual absence points, the model did not provide the actual probability that largemouth bass anglers target a given location (Pearce and Boyce 2006). Within the binomial GLM model structure, t he number of pseudo absence points used in a model influences the probability that largemouth bass angler s will target a given location T herefore this technique will only provide a relative likelihood that largemouth bass angler s will target a location (Pearce and Boyce 2006). As a result predicted vulnerabilit ies of the fish were scaled so the highest vulnerability was equal to one F ish locations with a vulnerability greater than or equal to 0.5 were considered to be in areas where they were vulnera ble to angling F ish locations with vulnerabilit ies
21 less than 0.5 were considered to be areas that were invulnerable to angling The vulnerability score used to determine if fish locations were vulnerable or invulnerable to angling was arbitrarily select ed post hoc. Using the fish locations classified as vulnerable to angling or invulnerable to angling, capture histories for each fish were created. A c apture histor y is a string of characters or numbers represent ing a time series of where a fish was found during each sampling event throughout the study Fish locations were classified as either V or I Fish locations in areas vulnerable to angling were represented with a V Fish locations in areas invulnerable to angling were represented with an I Fish not found or not searched for during a survey were given a 0 For example, a fish with a capture history of VI0I represents a fish initially found in a vulnerable location, found on the following survey in an invulnerable location, not fo und during the third survey, and found again in an invulnerable location during the fourth survey. The c apture histories of each fish created from the biweekly surveys were used in m ultistate models within t he program MARK (version 6.1; White and Burnham 1999) to estimate daily exchange rates of fish between vulnerable and invulnerable areas of the lake (e.g. k 1 and k 2 from Figure 2 2 ) and evaluate factors influencing exchange rates. The model parameters include d ; heterogeneous exchange rates season effects and fish habitat preference Exchange rates were allowed to be heterogeneous since movement s of largemouth bass to areas targeted by largemouth bass anglers may not be the same as the exchange rate out of the areas targeted by largemouth bass an glers (e.g. k 1 k 2 from Figure 2 2 ). To test for seasonal differences in exchange rates, the study was divided into four seasons : fall, winter, spring, and summer. To test if there
22 were difference s in exchange rates based on habitat preference of fish, fish were divided into three categories based on the majority of their locations. Each time a fish was found, the location was classified as either an onshore or offshore location. If a fish was fou nd within 50 m from the outside edge of vegetation, the location was defined as an onshore location. If a fish was found greater than 50 m from the outside edge of vegetation, the location was defined as an offshore location. A f ish was classified as an offshore fish if less than 30% of its total locations were onshore, an onshore fish if greater than 70% of its total locations were onshore or a generalist if 30 70% of its total locations were onshore The categories used to divide the fish into habitat preference groups were selected based on natural breaks within the data (Figure 2 3). Offshore fish ranged in size from 366 to 559 mm with an average of 477 mm. The sizes of the generalist fish ranged from 359 to 567 mm with an average of 428 mm. Final ly, onshore fish ranged in size from 353 to 648 mm with an average of 432 mm. All possible model combinations used to predict the exchange rates were tested and ranked using small sample corrected Aka i ke Information Criteria (AIC c ). The most parsimonious model was selected based on fewest parameters and AIC c values of less than 10. In order to account for possible bias in the estimation of exchange rates using the biweekly surveys, e xchange rates were also estimated in the prog ram MARK using the within day surveys. To compare the estimates from the biweekly surveys and the within day surveys, t he best model characterizing biweekly exchange rates was used to assess within day exchange patterns. Average exchange rates were estim ated instead of estimating exchange rates for each within day survey b ecause only 8 to 14 fish were
23 tracked during each survey Because the survey in April had only 7 capture occasions ( 14 hour tracking period) and June and August had 8 capture occasions (16 hour tracking period), only the first 7 capture occasions from June and August were used to estimate exchange rates. Tag Returns Angler tag return data were analyzed to test if there was a portion of the population that was invulnerable to angling. Anglers that caught tagged fish were instructed to remove the external $200 reward tag and call the number on the tag to receive the reward I nstructions were posted at the boat ramps and on the tag. When a nglers called to claim the reward they complete d a short telephone survey to identify their target species capture location if they were fishing near the shoreline or in open water, capture date, Chi square test (Eq. 2 5) was used t o test if there was a difference in the number of fish ca ught by anglers between the offshore fish, generalist fish and onshore fish ( 2 5) 2 ) test O i was the observed number of fish caught in fish habitat preference group i and E i was the expected number of fish caught in fish habitat preference group i T he expected numbers of fish caught were calculated as : ( 2 6 ) where N i was the number of fish in habitat preference group i and the average portion of total tagged fish caught by anglers
24 Figure 2 1 Bathymetric map of Lake Santa Fe, Florida. The center of the lake is located at 27.74N and 82.07W. Contour lines represent depth in meters.
25 Figure 2 2 Representation of a recreational fishery from Cox and Walters (2002). Total population is comprised of a vulnerable portion V and an invulnerable portion ( N V ) of the population (Cox and Walters 2002). Constants k 1 and k 2 represent the exchange rates between the vulnerable and invulnerable portions of the population (Cox and Walters 2002). Removals from the vulnerable portion of the population occur at the rate k 3 (Cox and Walters 2002).
26 Figure 2 3 Proportion of time spent onshore for each fish broken up into habitat selection groups. Blue represents fish that spend most of their time offshore, green represents generalist fish that frequently move between onshore and offshore habitat, and black represents fish that spend most of the time onshore.
27 CHAPTER 3 RESULTS A total of 832 anglers were surveyed from November 2010 through O ctober 2011. Of these 313 were targeting largemouth bass (Figure 3 1 ) Largemouth bass anglers tended to congregate around the littoral zone of the lake whereas the non bass anglers tended to be located in open water (Figure 3 1 ). Many of the offshore anglers were targeting black crappie Pomoxis nigromaculatus A total of 44 anglers interviewed indicated that 91% of all angler s were correctly identified Largemouth bass a ngler distribution was best characterized by the model containing distance from vegetation and vegetation rugosity (Table 2 1 ). All vegetated habitat s had relative likelihood values g reater than 0.65 and offshore habitats had relative likelihood values less than 0.20, indicating largemouth bass anglers target vegetated areas more than offshore areas (Figure 3 2 ). Largemouth bass an glers were more likely to target vegetated habitat with higher rugosity scores than vegetated habitat with lower rugosity scores (Figure 3 2 ). Eighty one largemouth bass were tagged in October 2010 Sixty four were tagged near onshore habitat and 17 tagged in offshore habitat. None of the fish suffered from mortality due to the surgery process, so all fish were included in the MARK analysis. Each f ish w as found an average of 28 times over 88 tracking events. Due to memory limitations in the program MARK, only 61 of the trackin g events could be analyzed to predict the exchange rates. To compress the data set tracking events having less than 10 found fish were combined with the preceding or following tracking event, whichever was closer. During the study 38 fish were harvest ed, died from natural mortality or were lost due to unreported harvest or tag failure The remainder of the fish were
28 searched for during the entire study, h owever t he transmitters started to fail after day 300 and only 17 fish were tracked for the full 365 days On average, approximately one third of the tagged fish were located in offshore habitat at any given whereas most bass anglers fished along shore (Figure 3 3 ). Assuming the distribution of tag ged fish was representative of the bass population as a whole the distribution of largemouth bass anglers did not mirror the distributi on of the largemouth bass. Thus, about a third of the fish appeared to be invulnerable to angling based on their location at any one time. The best model predicting excha nge rates using the biweekly survey data from vulnerable to invulnerable areas included the interaction between fish habitat preference and heterogenic exchange rates (Table 2 2). Estimates of exchange rates from the b iweekly survey ranged from 0.0 01 to 0. 035 per day with highest average exchange rates for offshore fish and lowest average for onshore fish (Table 2 3). O ffshore fish located in invulnerable area s had a 0.3 % chance of moving to vulnerable area s whereas offshore fish located in vulnera ble area s had a 3.5 % chance of moving to invulnerable area s each day (Table 2 3). This suggests a general offshore movement for fish classified as the offshore behavior type For generalist fish, the probability of moving to either a vulnerable or invuln erable area was approximately 0.7 to 0.8 % each day (Table 2 3). Finally, onshore fish located in vulnerable area s had a 0.1 % chance of moving to invulnerable area s whereas onshore fish located in invulnerable area s had a 1.1 % chance of moving to vulnerable area s each day (Table 2 3). This suggests a general onshore movement for fish with onshore habitat preference. D ifferences in exchange rates of offshore fish indicated this subpopulation remained invulnerable for a
29 significant portion of the y ear. Conversely exchange rates of onshore fish indicate this subpopulation was vulnerable to angling for the majority of the year. Exchange rates of generalist fish were almost equal, indicating they did not remain invulnerable for a significant portion of the year Ho wever the generalist fish were potentially subjected to less fishing effort than the onshore fish. E xchange rates estimated from the within day sampling indicated higher exchange rates than the biweekly observations. Estimates of exchan ge rates from the within day surveys ranged from 0.33 to 1 .00 per day (Table 2 4). Similar to trends in exchange rates from biweekly surveys offshore fish had the highest average exchange rates and onshore fish had the lowest average exchange rates (Tabl e 2 4). However, exchange rates from within day surveys were one to two order s of magnitude larger than estimates from the biweekly surveys (Table s 2 3, 2 4). E xchange rates for the offshore fish appear to be inflated due to small sample sizes. Out of 33 fish surveyed, only five were offshore fish and only one of these was found every two hours during the within day surveys. Three of the offshore fish were foun d to move between vulnerable and invulnerable areas whereas two of the offshore fish were only located in invulnerable areas. The exchange rate of 1.00 indicated that if offshore fish were located in vulnerable areas, they did not stay there for more tha n a day. Additionally offshore fish had a high probability of moving to vulnerable areas within a given day (59%; Table 2 4). However, there was an offshore fish that was not included in the within day survey that moved less than 50 m between biweekly t racking events. Therefore, i t seem ed unlikely that this fish had a high probability of moving to a vulnerable area within a given day Furthermore there were offshore fish that had seasonal movement patterns,
30 where they would spend periods of time onsho re (e.g. during the spawning season) while spending the rest of the year offshore. These fish would have low overall exchange rates, but would spend a significant portion of the year onshore. Although the re were some offshore fish that had high exchange rates, there were still individuals that likely remained invulnerable to angling for a significant portion of the year. However, results from the within day surveys indicated that the whole population was at least occasionally vulnerable to angling due to the high exchange rates that occurred owning to daily movements. Results from the tag returns indicated that 58% of all radio tagged fish were caught at least once by anglers (Table 2 5 ). Forty seven percent of the offshore fish were caught, 65% of the generalist fish were caught, and 59% of the onshore fish were caught (Table 2 5 ). Of the offshore fish that were caught, approximately 3 8 % were reported as being caught onshore and approximately 6 3 % were reported as being caught offshore (Table 2 5 ). O f the generalist fish 92% were caught onshore and only 8% were caught offshore (Table 2 5 ). N one of the onshore fish wer e reported as being caught offshore (Table 2 5 ). Additionally, all largemouth bass were caught by anglers specifically targeting largemouth bass or by anglers were targeting largemouth bass along with other species. Results from the C hi square test indicate d there was not a significant difference between the portions of fish caught based on the habitat preference of the fish ( C hi square = 1.39, P = 0.45), i ndicat ing all fish had similar vulnerabilit ies to angling erence.
31 Table 2 1 Delta AIC scores and model weights for binomial generalized linear models predicting angler distribution. D Veg represents distance from the outside edge of vegetation, R1:R7 represents dummy variables f or rugosity scores 1:7, W Veg represents width of the vegetated area, slope represents slope of the shoreline, D Shore represents distance from the shoreline, MM represents the presence of manmade structure (e.g. docks), CY represents the presence of cypress trees, SP represents the presence of spatterdock, and BR represents the presence of giant bulrush. Model Weight Npar D Veg +R1+R2+R3+R4+R5+R6+R7 0.0 0.97 9 D Veg +W Veg +Slope+R1+R2+R3+R4+R5+R6+R7+MM+CY+SP+BR 7.9 0.02 15 D Veg +D Shore +Depth+W Veg +Slope+R1+R2+R3+R4+R5+R6+R7+MM+C Y+SP+ BR 9.4 0.01 17 D Shore +W Veg +Slope+ R1+R2+R3+R4+R5+R6+R7+MM+CY+SP+ BR 12.5 0.00 15 D Veg 38.0 0.00 2 Depth+W Veg +Slope+ R1+R2+R3+R4+R5+R6+R7+MM+CY+SP+ BR 42.5 0.00 15 R1+R2+R3+R4+R5+R6+R7 43.0 0.00 8
32 Table 2 2 Delta AICc table and model weights testing for exchange rates of fish between areas vulnerable and invulnerable to angling using the program MARK. Habitat represents the habitat preference of the fish, HM represents the heterogenic movement, and Npar represents the number of parameters. Model Model Weight Npar Habitat HM 0.0 1.00 8 Habitat HM Season 20.1 0.00 26 Habitat 220.0 0.00 5 Habitat Season 231.2 0.00 14 HM 355.3 0.00 4 Season HM 361.3 0.00 10 Null 363.5 0.00 3 Season 363.7 0.00 6
33 Table 2 3 Biweekly survey estimates of exchange rates between areas vulnerable (Vul) and invulnerable (Invul) to angling derived from the MARK model incorporating habitat preference of the fish and heterogenic movement. Estimates are in daily (24 hour) time steps. The LCI and UCI represent the lower and upper 95% confidence intervals around the estimate. Estimates for movement are divided into habitat preference. Main Loc ation Parameter Est. LCI UCI Offshore Vul to Invul 0.035 0.021 0.054 Offshore Invul to Vul 0.003 0.002 0.004 Generalist Vul to Invul 0.008 0.006 0.010 Generalist Invul to Vul 0.007 0.005 0.009 Onshore Vul to Invul 0.001 0.001 0.001 Onshore Invul to Vul 0.011 0.008 0.016
34 Table 2 4 Within day survey estimates of exchange rates between areas vulnerable (Vul) and invulnerable (Invul) to angling from April, June, and August. Estimates were derived from the MARK model incorporating habitat preference of the fish and heterogenic movement. Estimates are in daily (24 hour) time steps. The LCI and UCI represent the lower and upper 95% confidence intervals around the estimate. Estimates for movement are divided into habit at preference. Main Location Parameter Est. LCI UCI Offshore Vul to Invul 1.000 0.966 1.000 Offshore Invul to Vul 0.586 0.111 0.996 General Vul to Invul 0.767 0.414 0.976 General Invul to Vul 0.888 0.587 0.994 Onshore Vul to Invul 0.334 0.122 0.711 Onshore Invul to Vul 0.996 0.844 1.000
35 Table 2 5 Catch data of the tagged fish divided into fish habitat preference categories. Not all anglers reported where they caught the tagged fish. Fish Habitat Preference Number of Fish Percent Caught Percent Caught Onshore Percent Caught Offshore Offshore 19 47 38 63 Generalist 23 65 92 8 Onshore 39 59 100 0 Total 81 58 85 15
36 Figure 3 1 Distribution of largemouth bass anglers (red dots) and non bass anglers (black dots) on Lake Santa Fe from November 2010 through October2011. Green area around the perimeter of the lake represents the vegetated areas and blue represents open water. Ang ler locations outside of the lake are located in canals. N
37 Figure 3 2 Fitted general linearized model predicting the relative likelihood that largemouth bass anglers will target a given area based on distance from vegetation and rugosity score of the onshore habitat. The dark green line represents the rugosity score of 0 (open water), the grey line represents a rugosity score of 1, the red line represents a rugosity score of 2, the light blue line represents a rugosity score of 3, the light green line represents a rugosity score of 4, the dark blue line represents a rugosity score of 5, the orange line represents a rugosity score of 6, and the black line represents a rugosity score of 7.
38 Figure 3 3 All fish locations from October 2010 throu gh October 2011 (green dots). Areas where largemouth bass anglers target (black) and where they do not target (grey) based on binomial generalized linear model incorporating distance from shore and rugosity of the shoreline habitat. Fish locations outsid e of the lake were found in canals. N
39 CHAPTER 4 DISCUSSION Unlike predict ions of the IFD theory, the distribution of largemouth bass anglers did not mirror the distribution of largemouth bass. The differences in the distribution of largemouth bass anglers a nd largemouth bass lend support to Martin (1958) and Cox and Walters ( 2002 ) hypothesis that a population is comprised of fish that are either vulnerable to angling or invulnerable to angling. However data from the tag returns indicated that even though a subset of fish spent most of the time in areas invulnerable to angling, ultimately they were captured at similar rates to o nshore fish. Therefore, the observed exchange rates of largemouth bass could have been sufficiently h igh such that all fish did not remain invulnerable to angling or there were other factors influencing the ir vulnerability Understanding the exchange rates between vulnerable and invulnerable states is a need in recreational fisheries (Cox and Walters 200 2), and this study showed that exchange rates were sufficient to make all fish essentially vulnerable during the year. L argemouth bass anglers target ed only abou t two thirds of the tagged fish at any given time indicating that largemouth bass anglers were not distributed according to the IFD Although the IFD has performed well at predicting the distribution of commercial anglers (Gillis et al. 1993; Gillis 2003; Swain and Wade 2003; Voges et al. 2005), there are major differences between commercial and recreational anglers. Similar to the results of Abernethy et al. (2007), recreational anglers violate key assumptions of the IFD The IFD states that anglers will be distributed such that all anglers will have similar catch rates (Voges et al. 200 5 ). Abernethy et al. (2007) found that the degree of angler experience, including knowledge of the fishing areas and an understanding of the
40 environmental dynamics was positively related to catch rates. Therefore, th e differences in angler experience and knowledge could have resulted in the largemouth bass anglers not being ideally distributed. Furthermore, largemouth bass anglers fishing in offshore areas could have been more effective at catching fish than largemo uth bass anglers fishing in onshore areas S tudies have shown that a few recreational anglers catch a large portion of the fish (Hilborn 1985; van Poorten and Post 2005; Seekell et al. 2011). Catch inequality has been found to be common in recreational f isheries (Baccante 1995). Lake Santa Fe has many brush piles scattered throughout the pelagic zone, however they are generally not marked with a float and are therefore difficult for anglers to find. Through fall and winter of 2010 2011, there was a larg e population of threadfin shad Dorosoma petenense and numerous fish were tracked as they were chasing these schools of prey fish. Largemouth bass anglers that effectively target fish aggregating near offshore structure (e.g. brush piles) and following sch ools of threadfin shad could have higher catch rates than anglers fishing in other areas Studies analyzing the effectiveness of marine protected areas (MPAs) indicate that fish with high exchange rates into and out of protected areas were not protected from fishing ( Walters and Bonfil 1999; Botsfod et al. 2003; Grss et al. 2011). Results from the tag returns showed that a portion of the offshore fish were vulnerable to anglers fishing onshore. This indicate d the exchange rates were high enough that th is subpopulation was not protected from angling. Furthermore, exchange rates estimated from the biweekly surveys were likely biased low b ecause of the large time intervals between tracking events. Studies have found that large time intervals between trac king
41 events may miss the majority of fish movements (Demers et al. 1996; Ovidio et al. 2000; Lkkeberg et al. 2002; Hanson et al. 2007). E xchange rates estimated from the within day surveys were therefore less biased and likely resulted in the whole popul ation being vulnerable to angling. U nlike MPAs there was still effort directed at largemouth bass in the offshore areas of the lake and c onsequently, the majority of offshore fish were caught in offshore habitats. Therefore the exchange rates of the fi sh were not the only factor influencing the vulnerability of fish to angling. Factors other than spatial location and exchange rates between vulnerable and invulnerable areas influence fish vulnerability to angling such as fish learning and hook avoidance Fish learning and hook avoidance has been noted in some species (Beukema 1970; Young and Hayes 2004; Askey et al. 2006; Kuparinen et al. 2010 ; Klefoth et al. 2012 ) including largemouth bass (Aldrich 1939; Anderson and Heman 1969; Hackney and Linkous 197 8). Beukema (1970) found that the angling catchability of common carp Cyprinus carpio was three times higher for individuals that had not been caught and released than those that had been caught and released. In response to experimental angling, c ommon c arp were also found to avoid previously used feeding areas (Klefoth et al. 2012). Additionally, Anderson and Heman (1969) and Hackney and Linkous (1978) found a link between fish learning and angler effort. On Lake Santa Fe, the largemouth bass anglers c oncentrated their effort near o nshore vegetated habitat. Although this study did not analyze the differences in catchability between fish habitat preference the onshore fish could have been more reluctant to react to lures because they were exposed to mo re fishing effort Consequently, the offshore fish could be relatively naive to angling due to substantially lower angler effort. These d ifferences in
42 the levels of vulnerability to angling caused by fish learning and hook avoidance could be exacerbated by a concentration of angling effort and by fish habitat preference Although this study showed that the distribution of largemouth bass anglers did not mimic the distribution of largemouth bass, s everal major assumptions were made i n determining the distribution of largemouth bass anglers. While setting up the logistic GLM to quantify angler distribution, I assumed largemouth bass anglers fishing outside the peripheral 50 m buffer of vegetation were fishing in open water. P ost hoc examination indicated that approximately 91% of the largemouth bass anglers were within 50 m of the edge of vegetation. Furthermore, our ability to classify anglers based on fishing technique was imperfect (91% correct) I chose not to incorporate this u ncertainty into the models because of the differences in the distributions of the two angler groups. Even though 9% of the anglers were misclassified, only one of the misclassified anglers was located offshore and the majority of non bass anglers were loc ated offshore. Additionally, the black crappie fishing was exceptional from November 2010 through February 2011 with over half of the non bass anglers surveyed during those four months. Moreover I chose not to incorporate the non bass anglers in determ ining the vulnerability of the fish because all tagged largemouth bass were caught by anglers that were specifically targeting largemouth bass or were targeting largemouth bass along with other species. However, by n ot including uncertainty in these assum ptions the estimates of effort in the offshore areas were likely conservative is important to understand the dynamics of angler behavior T he spatial distribution of
43 effort can significantly impact both the fish population and the vulnerability of individual fish. Not only is it possible for recreational anglers to overfish populations (Post et al. 2002; Lewin et al. 2006), but with selective, non random targeting of individuals within a population it might also be possible to alter the genetic structure of the population by targeting individuals with certain life history traits (Philipp et al. 2009). Additionally, the catchability of fish could have been influenced b y fish habitat preference traits. At high levels of exploitation, differences in catchability could lead to the overexploitation of individuals with a certain life history trait without overfishing the entire population. Therefore, an u nderstanding of re creational angler dynamics in relation to fish behavior and fish population dynamics is imperative increase the sustainability of recreational fisheries
44 CHAPTER 5 MANAGEMENT IMPLICATIONS In systems where the distribution of anglers is not the same as the distribution of the targeted species, the non random distribution of fishing effort can essentially create protected areas for fish. T his can happen when the majority of a population cong regate in certain areas or when fish are distributed throughout a large area and anglers are limited by the number of access points and distance (such as many costal fisheries). In these fisheries there might be a gradual decrease in the vulnerability of fish as distance from access points increase, compared to a sudden change in the vulnerability of fish as was observed on Lake Santa Fe S imilar relationship s have been observed between proximity to urban centers and fishing effort in lake rich environments where effort decreases with distance from urban centers (Post et al. 2008). Incorporating these differences in vulnerability to angling into stock assessment models can allow for more accurate predictions of stock status Additionally, diff erences in vulnerability to angling may offer unique alternatives to the use of MPAs as management tools. Sampling design can also have large implications on the results studies quantifying movement rates for creating and designing MPA s Similar to many studies (see Demers et al. 1996; Ovidio et al. 2000; Lkkeberg et al. 2002; Hanson et al. 2007), this study demonstrated that sampling designs with large time intervals between tracking events did not adequately quantify the exchange rates of fish between areas where anglers targeted and did not target. This is important to consider when quantify ing movement rates to test the effectiveness of MPAs because MPAs are less effective for fish with high movement rates compared to fish with low movement rates (Bo tsford et al. 2003). Additionally, this is important when considering the size of MPAs
45 because fish with high movement rates require larger protected areas (Grss et al. 2011). However b y analyzing movement rates over short time steps, it is possible to miss seasonal movement patterns or migrations. It is essential to determine the objectives of your study and develop a sampling design that includes the appropriate temporal and spatial scales to address those objectives
46 LIST OF REFERENCES Abernethy K. E., E. H. Allison, P. P. Molloy, and I. M. Cote. 2007. Why do fishers fish where they fish? Using the ideal free distribution to understand behaviour of artisanal reef fishers. Canadian Journal of Fisheries and Aquatic Sciences 64:1595 1604. Akaike, H. 1974. A new look at the statistical identification model. IEEE Transactions on Automatic Control 19:716 723. impounded reservoir. Transactions of the American Fisheries Society 68:221 227. Anderson, R. O., and M. L. Heman. 1969. Angling as a factor influencing the catchability of largemouth bass. Transactions of the American Fisheries Society 98:317 320. Arlinghaus, R. 2006. On the apparently striking disconnect between motivation and satisfaction in recreational fishing: the case of catch orientation of German anglers. North American Journal of Fisheries Management 26:592 605. Arlinghaus, R., M. Bork, and E. Fladung. 2008. Understanding the heterogeneity of recr eational anglers across an urban rural gradient in a metropolitan area (Berlin, Germany), with implications for fisheries management. Fisheries Research 92:53 62. Askey, P. J., S. A. Richards, J. R. Post, and E. A. Parkinson. 2006. Linking angling catch r ates and fish learning under catch and release regulations. North American Journal of Fisheries Management 26:1020 1029. Baccante, D. 1995. Assessing catch inequality in walleye angling fisheries. North American Journal of Fisheries Management 15:661 665. Beukema, J. J. 1970. Angling experiments with carp ( Cyprinus carpio L.). II. Decreasing catchability through one trial learning. Netherlands Journal of Zoology 20, 81 92. Biro, P. and J. R. Post. 2008. Rapid depletion of genotypes with fast growth and bold personality traits from harvested fish populations. Ecology 105:2919 2922. Botsford, L. W., M. Fiorenza and A. Hastings. 2003 Principles for the design of marine reserves. Ecological Applications 13:25 31. Bourke, P., P. Magnan, and M. A. Rodrguez. 1997. Individual variations in habitat use and morphology in brook charr. Journal of Fish Biology 51: 783 794. Brauhn, J. L. and H. Kincaid. 1982. Survival, growth, and catchability of rainbow trout of four strains. North American Journal of Fisheries Managem ent 2:1 10
47 Brett, R. J. and T. D. D. Groves. 1979. Physiological energetics. Pages 280 352. In W. S. Hoar, D. J. Randall, and J. R. Brett editors. Fish Physiology: Bioenergetics and Growth. Academic Press, New York, NY. Bryan, J. E., and P. A. Larkin. 1972. Food specialization by individual trout. Journal of the Fisheries Research Board of Canada 29:1615 1624. Cooke, S. J., C. D. Suski, K. G. Ostrand, D. P. Philipp, and D. H. Wahl. 2007. Physiological and behavioral consequences of long term artificial selection for vulnerability to recreational angling in a teleost fish. Physiological and Biochemical Zoology 80:480 490. Cox, S. P. and C. Walters. 2002. Modeling exploitation in recreational fisheries and implications for effort management on British Col umbia rainbow trout lakes. North American Journal of Fisheries Management 22:21 34. Demers, E., R. S. McKinley, A. H. Weatherley, and D. J. McQueen. 1996. Activity patterns of largemouth and smallmouth bass determined with electromyogram biotelemetry. Tran sactions of the American Fisheries Society 125:434 439. Ditton, R. B. 2004. Human dimensions of fisheries. Pages 199 208 in M. J. Manfredo, J. J. Vaske, B. L. Bruyere, D. R. Field, and P. J. Brown, editors. Society and natural resources: a summary of knowl edge prepared for the 10 th International Symposium on Society and Resource Management. Modern Litho, Jefferson City Missouri. Driver, B. L. and R. C. Knopf. 1976. Temporary escape: one product of sport fisheries management. Fisheries 1:21 29. Dutka Gianel li, J., R. Taylor, E. Nagid, J. Whittington, and K. Johnson. 2011. Habitat utilization and resource partitioning of apex predators in coastal rivers of southern Florida. FWRI Library No. F2771 07 11 F. Dwyer, W. P. 1990. Catchability of three strains of cu tthroat trout. North American Journal of Fisheries Management 10:458 461. Fedler, A. J. and R. B. Ditton. 1994. Understanding angler motivations in fisheries management. Fisheries 19:6 13. Florida Lakewatch. 2005a. Santa Fe (Alachua County) Florida Lakewat ch bathymetric map. Florida Lakewatch, Department of Fisheries and Aquatic Sciences, University of Florida. Florida Lakewatch. 2005b. Little Santa Fe (Alachua County) Florida Lakewatch bathymetric map. Florida Lakewatch, Department of Fisheries and Aquat ic Sciences, University of Florida.
48 Fretwell, S. D. 1972. Populations in a seasonal environment. Princeton University Press, Princeton, NJ. Fretwell, S. D. and H. L. Lucas, Jr. 1970. On territorial behavior and other factors influencing habitat distribu tion in birds. ACTA Biotheoretica. 19:16 36. Gillis, D. M. 2003. Ideal free distribution in fleet dynamics: a behavioral perspective on vessel movement in fisheries analysis. Canadian Journal of Zoology 81:177 187. Gillis, D. M., R. M. Peterman., and A. V Tyler. 1993. Movement dynamics in a fishery: application of the ideal free distribution to spatial allocation of effort. Canadian Journal of Fisheries and Aquatic Sciences 50:323 333. Grss, A., D. M. Kaplan, S. Gunette, C. M. Roberts and L. W. Botsford 2011. Consequences of adult and juvenile movement for marine protected areas. Biological Conservation 144:692 702. Hackney, P. A., and T. E. Linkous. 1978. Striking behavior of the largemouth bass and use of the binomial distribution for its analysis. T ransactions of the American Fisheries Society 107:682 688. 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 ( Micropterus salmoides ) behaviour and activity at multiple spatial and temporal scales using a whole lake telemetry array. Hydrobiologia 582:243 256. Hilborn, R. 1985. Fleet dynamics and individual variation: why some people catch more fish than others. Canadian Journal of Fisheries and A quatic Sciences 42:2 13. Hilborn, R., and M. Ledbetter. 1979. Analysis of the British Columbia salmon purse seine fleet: dynamics of movement. Journal of the Fisheries Research Board of Canada 36 :384 391. Holland, D. S. and J. G. Sutinen. 1999. An empirica l model of fleet dynamics in New England trawl fisheries. Canadian Journal of Fisheries and Aquatic Sciences 56:253 264. Hunt, L. M., R. Arlinghaus, N. Lester, and R. Kushneriuk. 2011. The effects of regional angling effort, angler behavior, and harvesting efficiency on landscape patterns of overfishing. Ecological Applications 21:2555 2575 Jobling, M. 1985. Growth. Pages 213 230. In P. Tytler and P. Calow editors. Fish Energetics: New Perspectives. Johns Hopkins University Press, Baltimore, MD. Klefoth, T. C. Skov, J. Krause, and R. Arlinghaus. 2012. The role of ecological context and predation risk stimuli in revealing the true picture about the genetic basis of boldness evolution in fish. Behavioral Ecology and Sociobiology 66:547 559.
49 Kobler, A., T. Kle foth, T. Mehner, and R. Arlinghaus. 2009. Coexistence of behavioural types in an aquatic top predator: a response to resource limitation? Oecolgia 161:837 847. Kuparinen, A., T. Kelfoth, and R. Arlinghaus. 2010. Abiotic and fishing related correlates of an gling catch rates in pike ( Esox lucius ). Fisheries Research 105:111 117. Lewin, W., R. Arlinghaus, and T. Mehner. 2006. Documented and potential impacts of recreational fishing: insights for management and conservation. Reviews in Fisheries Science 14:305 367. Lkkeberg, S., A. Fern, and T. Jrgensen. 2002. Effect of position fixing interval on estimated swimming speed and movement pattern of fish tracked with a stationary positioning system. Hydrobiologia 483:259 264. Martin, R. G. 1958. Influence of fi shing pressure on bass fishing success. Proceedings of the Annual Conference of the Southeastern Association of the Game and Fish Commissioners 11:76 82. Mesing, C. L., and A. M. Wicker. 1986. Home range, spawning migrations, and homing of radio tagged Flo rida largemouth bass in two central Florida lakes. Transactions of the American Fisheries Society 115:286 295. Morbey, Y. E., P. Addison, B. J. Shuter, and K. Vascotto. 2006. Within population heterogeneity of habitat use by lake trout Salvelinus namaycush Journal of Fish Biology 69:1675 1696. Nuhfer, A. J., and G. R. Alexander. 1994. Growth, survival, and vulnerability to angling of three wild brook trout strains exposed to different levels of angler exploitation. North American Journal of Fisheries Manag ement 14:423 434. Ovidio, M., J. C. Philippart, and E. Baras. 2000. Methodological bias in home range and mobility estimates when locating radio tagged trout, Salmo trutta, at different time intervals. Aquatic Living Resources 13:449 454. Parkinson, E. A., J. R. Post, and S. P. Cox. 2004. Linking the dynamics of harvest effort to the recruitment dynamics in a multistock, spatially structured fishery. Canadian Journal of Fisheries and Aquatic Sciences 61: 1658 1670. Pearce, J. L. and M. S. Boyce. 2006. Mo deling distribution and abundance with presence only data. Journal of Applied Ecology 43:405 412. Philipp, D. P., S. J. Cooke, J. E. Claussen, J. B. Koppelman, C. D. Suski, and D. P. Burkett. 2009. Selection for vulnerability to angling in largemouth bass. Transactions of the American Fisheries Society 138:189 199.
50 Post, J. R., L. Persson, E. A. Parkinson, and T. vanKooten. 2008. Angler numerical response across landscapes and the collapse of freshwater fisheries. Ecological Applications 18:1038 1049. Post J. R., M. Sullivan, S. Cox, N. P. Lester, C. J. Walters, E. A. Parkinson, A. J. Paul, collapse? Fisheries 27:6 17. Redpath, T. D., S. J. Cooke, R. Arlinghaus, D. H. Wahl, and D. P. Philipp. 2009. Life history traits and energetic status in relation to vulnerability to angling in an experimentally selected teleost fish. Evolutionary Applications 2:312 323. Sandlund, O. T., B. Jonsson, H. J. Malmquist, R. Gydemo, T. Lindem, S Sklason, S. S. Snorrason and P. M. Jnasson. 1987. Habitat use by arctic charr Salvelinus alpinus in Thingvallavatn, Iceland. Environmental Biology of Fishes 20:263 274. Scheuerell, M. D. and D. E. Schindler. 2003. Diel vertical migration by juvenile sockeye salmon: empirical evidence for the antipredation window. Ecology 84:1713 1720. Seekell, D. A., C. J. Brosseau, T. J. Cline, R. J. Winchcombe, and L. J. Zinn. 2011. Long term changes in recreational catch inequality in a trout stream. North American Journal of Fisheries Management 31:1100 1105. Snedden, G. A., W. E. Kelso, and D. A. Rutherford. 1999. Diel and seasonal patterns of spotted gar movement and habit at use in the Lower Atchafalaya River Basin, Louisiana. Transaction of the American Fisheries Society 128:144 154. Suski, C. D. and M. S. Ridgway. 2009. Seasonal pattern of depth selection in smallmouth bass. Journal of Zoology 279:119 128. Swain, D. P. and E. J. Wade. 2003. Spatial distribution of catch and effort in a fishery for snow crab (Chionoecetes opilio): tests of predictions of the ideal free distribution. Canadian Journal of Fisheries and Aquatic Sciences 60:897 909. van Poorten, B. T., and J. R. Post. 2005. Seasonal fishery dynamics of a previously unexploited rainbow trout population with contrasts to established fisheries. North American Journal of Fisheries Management 25:329 345. Voges, E., A. Gordoa, and J. G. Field. 2005. Dynamics of Namib ian hake fleet and management connotations: application of the ideal free distribution. Scientia Marina 69:285 293. Walters, C. J. and R. Bonfil. 1998. Multispecies spatial assessment models for the British Columbia groundfish trawl fishery. Canadian J ournal of Fisheries and Aquatic Sciences 56:601 628.
51 Warden Jr., R. L. and W. J. Lorio. 1975. Movements of largemouth bass ( Micropterus salmoides ) in impounded waters as determined by underwater telemetry. Transactions of the American Fisheries Society 10 4:696 702. Warton, D. I. and L. C. Shepherd. 2010. Poisson point process models solve the only data in ecology. The Annals of Applied Statistics 4:1383 1402. White, G. C. and K. P. Burnham. 1999. Program MARK: survival estimation from populations of marked animals. Bird Study 46:120 139. Winter, J. D. 1996. Advances in underwater biotelemetry. Pages 555 590. In B.R. Murphy & D.W. Willis editors. Fisheries Techniques, 2nd edition. Bethesda, MD: American Fisheries Society Young, R. G., and J. W. Hayes. 2004. Angling pressure and trout catchability: Behavioral observations of brown trout in two New Zealand backcountry rivers. North American Journal of Fisheries Management 24:1203 1213.
52 BIOGRAPHICAL SKETCH Bryan G. Matthias was born in Manitowoc, Wisconsin to Tim and Linda Matthias. He was raised in a rural Wisconsin with his brother Brad. Bryan graduated from the University of Wisconsin Stevens Point in 2009 with a double major in Fisher ies and Aquatic Sciences and Biology and a minor in Natural Sciences. Following graduation Bryan moved out west to work as a fisheries technician with Utah State University. In January of 2010 he moved down to Florida to work as a technician in Dr. Allen the fall of 2010 Bryan began work on his Master of Science degree and he is set to Ph.D. under Dr. Ahrens at the University of Florida and getting married to his wonderful fiance in 2014.