AN ASSESSMENT OF FIN RAYS AND FIN SPINES FOR USE IN NON LETHAL AGING OF LARGEMOUTH BASS MICROPTERUS SALMOIDES IN FLORIDA By SUMMER LINDELIEN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTI AL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2018
2018 Summer Lindelien
To my Dad Mom, and everyone that said I could
4 ACKNOWLEDGMENTS T hank you to my collaborators and mentors at the Florida Fish and Wildlife Cons ervation Commission (FWC) Partial f unding for my project was provided by FWC Individual thanks are due to : Eric Nagid, Jason Dotson, Chris topher Anderson, Drew Dutterer, and Travis Tuten ; t heir time and guidance allowed me to achieve this Master of Science I was given an amazing opportunity, and I wil l always be grateful to the FW C I thank my advisor Dr. Daryl Parkyn as well as Dr. Mike Allen for being a member of my commit tee. I thank all of my lab mate s and friends that helped me along the way, including but not limited to : Todd Van Natta, Mark Ho yer and Dr. Ed Camp I am extremely thankful for the love and endless knowledge my parents have provided me Lastly, I would like to thank Cher for her nearly constant positivity, and overarching encouragement.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURES ................................ ................................ ................................ ......................... 8 ABSTRACT ................................ ................................ ................................ ................................ ... 11 CHAPTER 1 I NTRODUCTION ................................ ................................ ................................ .................. 13 2 M ATERIALS AND METHODS ................................ ................................ ........................... 20 Fish Collections and Measurements ................................ ................................ ....................... 20 Processing of Stru ctures ................................ ................................ ................................ ......... 22 Sagitta Otoliths ................................ ................................ ................................ ................ 22 Fin Rays and Spines ................................ ................................ ................................ ........ 23 Aging of Structures ................................ ................................ ................................ ................. 26 Sagitta Otoliths ................................ ................................ ................................ ................ 26 Fin Rays and Fin Spines ................................ ................................ ................................ .. 27 Age and Growth ................................ ................................ ................................ ...................... 36 Precision of Age Estimates ................................ ................................ ................................ ..... 36 Accuracy of Age Estimates ................................ ................................ ................................ .... 38 Temporal Synchronicity of Annuli Deposition in Sagittae ................................ .................... 38 Temporal Synchronicity of Annuli Deposition in Dorsal Spines ................................ ........... 39 3 R ESULTS ................................ ................................ ................................ ............................... 42 Age and Length of LMB ................................ ................................ ................................ ......... 42 Processing and Clarity of Aging Structures ................................ ................................ ............ 44 Aging Precision ................................ ................................ ................................ ...................... 47 Aging Accuracy ................................ ................................ ................................ ...................... 52 Age Assignment Issues ................................ ................................ ................................ ........... 61 Growth of LMB ................................ ................................ ................................ ...................... 66 Marginal Increment Ana lysis ................................ ................................ ................................ 68 Sagitta Otoliths ................................ ................................ ................................ ................ 68 Dorsal Spines ................................ ................................ ................................ ................... 68 4 D ISCUSSION ................................ ................................ ................................ ......................... 70 Recommendations ................................ ................................ ................................ ................... 70 Removing and Processing Structures ................................ ................................ ..................... 71 Aging Fin Struc ture Sections ................................ ................................ ................................ .. 72
6 Comparative Studies ................................ ................................ ................................ ............... 78 Marginal Increment Analysis and Temporal Synchronicity ................................ .................. 84 Future Work and Non lethal Aging ................................ ................................ ........................ 86 Management Implications ................................ ................................ ................................ ...... 87 LIST OF REFERENCES ................................ ................................ ................................ ............... 91 BIOGRAPHIC AL SKETCH ................................ ................................ ................................ ......... 98
7 LIST OF TABLES Table page 3 1 Average clarity rankings (i.e., 1 3) of Largem outh Bass aging structu res for all Reader 1 aged fish. ................................ ................................ ................................ ............. 46 3 2 Precision in age estimates for Reader 1 (SL) between two aging sessions using Largemouth Bass otoliths, fin rays, and spines ................................ ................................ 48 3 3 Precision in age estimates for Reader 2 (DCP) between two aging sessions using Largemouth Bass otoliths, fin rays, and spines. ................................ ................................ 48 3 4 Between reader percent agreement (PA; 1 yr) for Largemouth Bass o tolith, fin ray, and spine ages ................................ ................................ ................................ ................... 48 3 5 Reader 1 precision of dorsal spine section ages from fem ale versus male Largemou th Bass ................................ ................................ ................................ ................................ ... 52 3 6 Reader 1 (SL) percent agreement and disagre ement for Largemouth Bass fin ray and spine ages (i .e., accuracy estimates). ................................ ................................ ................. 52 3 7 Percent agreement (ages exact and years different) between Largemouth Bass sagitta otolith and dorsal spine ages (accura cy estimates for Reader 1 = SL) ............................. 56 3 8 Accuracy o f dorsal s pine section ages from femal e versus male Largemouth Bass ........ 57 3 9 Reader 2 (DCP) percent agreement and disagreement with otolith ages for Largemouth Bass fin ray a nd spine ages (i .e., accuracy estimates) ................................ .. 60 4 1 Between structure age estimates for current and comparable studies using several Largemouth Bass fin structures. ................................ ................................ ........................ 83 4 2 Precision of age estimates (betw een reader PA = percent agreement; CV = coefficient of variation) for several comparable studies using fin structures. ................... 84
8 LIST OF FIGURES Figure page 2 1 Clipping several Largemouth Bass fins as close to the body as possible. ......................... 22 2 2 Thawing, excising, and drying of Larg emouth Bass fin rays and spin es ......................... 24 2 3 Labeling, embedding, and dry ing of Larg emouth Bass fin rays and spines ..................... 25 2 4 Sectioning and permanently mounting Larg emouth Bass fin rays and spines. ................. 25 2 5 Transverse section (0.5 mm thickness) of a Largemouth Bass sagitta otolith collected from Rodman Reservoir with 11 counted opaque bands. ................................ .................. 27 2 6 Examples of clarity rankings in different pelvic fin ray section s from the same Largemouth Bass ................................ ................................ ................................ .............. 28 2 7 Pelvic fin ray sections of various aged Largemouth Bass. ................................ ................ 29 2 8 Dorsal fin ray sections of various aged Larg emouth Bass. ................................ ................ 30 2 9 Dorsal fin spine sections of various aged Largemouth Bass. ................................ ............ 31 2 10 Pectoral fin ray sections o f various aged Largemouth Bass. ................................ ............. 32 2 11 Anal fin ray sections of various aged Largemouth Bass ................................ ................... 33 2 12 Anal fin spine sections of various aged Largemouth Bass. ................................ ............... 34 2 13 Pelvic fin s pine sections of variou s aged Largemouth Bass ................................ ............. 35 2 14 Transverse section (0.5 mm) of a sagitta otolith from a Largemouth Bass showing the core (i.e., C), four (i.e., 1 4) labeled opaque zones along the measurement a xis. ....... 39 2 15 Slow (white line next to inner groove) and fast (white line adjacent from outer lobe) growth axes for marginal increment analysis o f Largemouth Bass dors al spines ............ 41 3 1 Length frequency distribution of Largemouth Bass collected from Rodman Reservoir for age estimation via sagitta otoliths. Sample size is shown. .......................... 43 3 2 Age distribution from aged Largemouth Bass otoliths ( n = 124) and dorsal spines ( n = 122). Mean age and range of aging structures are shown. ................................ .............. 43 3 3 Pelvic fin ray sections from Larg emouth Bass (LMB; larger segments) with annuli visible in basal versus distal cuts from a 7 yr old individual. ................................ ............ 45 3 4 Largemouth Bass dorsal spine sections cut deep into the body versus cut alo ng the body. ................................ ................................ ................................ ................................ ... 45
9 3 5 Dorsal spine section clarity rankings (i.e., 1 3 shown in the legend) based on age of Largemouth Bass ( n = 122). ................................ ................................ ............................... 47 3 6 Age biplot and residual age difference p lot for between reader estimates of Largemouth Bass otolith ages ................................ ................................ ........................... 49 3 7 Precision of Largemouth Bass dorsal spine section ages by Reader 1 as a function of time (date of aging session; n = 122). ................................ ................................ ................ 51 3 8 Scatter plot comparisons of age estimates obtained from Largemouth Bass sagitta otoliths versus f in rays and spines for Reader 1 ................................ ............................... 53 3 9 Residual age difference plots, circle sizes represent sample sizes for a particular Largemouth Bass age combination relative to the largest sample siz e ............................. 54 3 10 Reader 1 age biplot comparing dorsal spine section mean modal ages to otolith section modal ages. Linear regression dashed line is shown. ................................ ............ 57 3 11 Average ages (years) and coefficient of variations (CV) as a function of the month the Largemouth Bass dorsal spine section was aged by Reader 1. ................................ .... 58 3 12 Scatter plot comparisons of age estimates obtained from Largemouth Bass sagitta otoliths versus fin rays and spines for Reader 2. ................................ ............................... 61 3 13 Pseudo annuli in Largemouth Bass dorsal spine sections. ................................ ................ 62 3 14 Largemouth Bass dorsal spine sections cut tangentially (i.e., at an angle other than 90). ................................ ................................ ................................ ................................ .... 62 3 15 Compaction of the edge in Largem outh Bass dorsal spine sections ................................ 63 3 16 Double bands in Largemouth Bass dorsal spine se ctions. ................................ ................. 63 3 17 Erosion of the first annulus in Largem outh Bass dorsal spine sectio ns. ............................ 64 3 18 Recognition of the first annulus is Largem outh Bass dorsal spine sections ..................... 64 3 19 Dorsal spine section from an age 7 Largemouth Bass showing annular radii measurement axis in white ................................ ................................ ................................ 65 3 20 First and second annular radii in Largemouth Bass dorsal spines ( n = 114) as a function of age. Horizontal lines = media n measurement for annulus 1 and 2. ................ 65 3 21 Von Bertalanffy growth curve s for aged Rodman Reservoir L argemouth Bass (LMB) sagitta otolith sections. ................................ ................................ ................................ ....... 67 3 22 Von Bertalanffy growth curve s for aged Rodman Reser voir L argemouth Bass (LMB) dorsal spine sections ................................ ................................ ................................ ......... 67
10 3 23 Age 4 Largemouth Bass mean marginal translucent growth incr ement ( 1 SE) of otoliths over a 12 month period. ................................ ................................ ........................ 68 3 24 Age 4 Largemouth Bass mean marginal opaque growth increment of dors al spines (DS; n = 81) over a 12 month period. ................................ ................................ ................ 69
11 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requir ements for the Degree of Master of Science AN ASSESSMENT OF FIN RAYS AND FIN SPINES FOR USE IN NON LETH AL AGING OF LARGEMOUTH BASS MICROPTERUS SALMOIDES IN FLORIDA By Summer Lindelien December 2018 Chair: Daryl C. Parkyn Major: Fisheries and Aquatic Sciences Aging of Largemouth Bass Micropterus salmoides (LMB) s agitta otoliths is the most common ly utili zed method by fisheries scientists in Florida Otolith removal requires sacrificing fish; thus it is not ideal particularly for age determination in large, trophy fish. Consequently, n on lethal aging methods are being further explored for LMB in Florida Verifying a fin ray and/or spine method would reduce or elimin ate mortality during age sampling, in populations where conservati on may be warranted, as well as enable age determination of angler caught LMB (e.g., during tournaments and/or TrophyCatch submi ssions). The refore, the purpose of my study was to evaluate th e utility of fin structures (i.e., anal, pelvic, dorsal, and pectoral fin rays, and anal, pelvic and dorsal s pines ) for aging LMB collected from Rodman Reservoir, F L as a potentially non lethal means of d etermining ages. Sagitta otoliths were the most prec ise structures for aging LMB based on high within reader and between reader percent agreement (PA), high ( c ), low coefficient of variation (CV) and low average percent error (APE) among ages Of the fin structures assessed, d orsal spines were the most precise and accurate for Reader 1 Anal spine, anal ray, and dorsal spine ages p rovided the highest PAs ( 7 0% ) ; therefore, they were the most precise fin structures for Reader 2 Between reader PA of fin st ructures was hig hest for pelvic spines ; however, d orsal spine and
12 pelvic ray ages were also precise between readers (PA = 65% and 67 % respectively). A nal ra y s were the most accurate fin structure for Reader 2 Partial temporal synchronicity of annuli deposition was seen in LMB sagittae and dorsal spines based on the marginal increment analyses Dorsal spine sections fr om young fish (i.e., age s 1 4 ) incurred a ge over estimation, whereas older f i sh (i.e., age s 5 8 ) had a higher percentage of ages underestimated. Identi fying the best aging structure wa s dep endent on proper processing and aging of structures, aging experience and substantial practice. Although sa gittae were the most precise aging structures, m ultiple fin s tructures could be utilized as potential non lethal aging tools for LMB populations in Florida
13 CHAPTER 1 INTRODUCTION The proper determination of the age of fishes enhances our assessment s of three vital demographic rates ( i.e., recruitment growth, and mort ality) which are essential for understanding the characteristics of a fish population (Strickland and Middaugh 2015). While many structures have been utilized s cientists prefer and most fr equently use an aging approach based on otoliths, fin rays, fin spines and /or scales from ind ividual fish; t hese hard structures provide a series of annual depositions tha t are view ed to determine ages (Jearld 1983). The sacculus and lagena are membra nou s structures located in t he pars inferior or ventral portion of inner ear ( Popper 1973 ). The sagitta otolith is in the sacculus, whereas the asteriscus otolith is in the lagena (Secor et al. 1992 ) These otoliths function in acoustic trans duction sound detection ( Popper and Lu 2000 ), and perception of linear acceleration s ( Kasumyan 2004) Individually, the sagittae move inside the surrounding membranous chamber causing hair cells tuned to different vibration frequencies to be mechanically stimulated ; t he brain then receives a signal about this event of sound or movement (Popper et al. 2005). The s agittae were u tilized in my project which are the largest otoliths in Perciform fishes, and the most commonly used structure s for a ging of teleos t fishes ( Secor et al. 199 2 ; Long and Stewart 2010 ). Annual growth bands (i.e., annuli ) of the sa gitta otolith consist of a pair of s uccessively formed opaque and translucent rings that occur yearly in most temperate zones (Jearld 1983 ) Although otoliths (Jearld 1983) removing them is lethal. If a non lethal method is desired, use of fin rays, fin spines, and/or scales for aging is an alternate option In bony fish, either fin ra ys (i.e., lepidotrichia) or spines support the fins internally (Moyle and Cech 2004 ; Dur n et al. 2011 ). Lepidotrichia are calcified branched, and segmented
14 ( Dur n et al. 2011) They begin as embry onic spines ( i.e., actinotrichia) that become scale pr imor dia which are successive ly replaced by distally added segments and ultimately become fin rays (Moyle and Cech 2004 ; Dur n et al. 2011 ). If th e s e developmental process es do not transpire true spines are formed (Moyle and Cech 2004) Both fin rays and spin es are oss ified ; however, there are key structural differences between the two. Fin rays are comprised of a paired set of segmented elements ( Goss and Stagg 1957; Moyle and Cech 2004) A cross section can be taken of each of the se two elements and can be u sed to estimate age (Beamish 1981) In contrast fin s pines are solid bony structures that are approximately round ed in cross section and distally unsegmented with a single element (Moyle and C ech 2004) Under the microscope, the spinous structures display two fused segments (one total part to age ) The fusion creates an internal c oelom (i.e., lumen ) that contain s blood vessels ( D. Parkyn, University of Florida, per sonal communication ). Growth and erosion of this chamber can result in obstruction of earlier annuli ( Casselman 1983; Isermann et al. 2010) ; however, th rough authentication techniques, readings can be adjusted to account for this discrepancy (Debicella 2005 ; Mur ie et al. 2009 ) To properly age a fish, it is critical that the first year of growth be incl uded in each fin structure, so clipping fins within the epiaxial musculature is not necessary i f the first annulus remains and is recognizabl e (Debicella 2005). T o keep the fin structure method non lethal, fins should be removed as close to the body as possible (Zymonas and McMahon 2009) Research has shown that the excision of some fin structures (e g. pelvic fin rays ; Zymonas and McMahon 2009 ) is not influential on growth or survival of certain species (Collins and Smith 1996); however, fish injur y is still appropriately questioned (Koch et al. 2008 ). Inappropriate removal techniques could result in fish having and/or developing lethal wounds, which reduces the u tility of fin structures for non lethally inferring growth patterns (Koch et al. 2008) Fin clipping
15 standardization is necessary, considering the var iance in information about cutting locations found in most fin ray and spine literatu re (Koch et al. 2008 ) Increasingly, non lethal methods using fin rays and/or fin spines have been employe d in age and growth studies to examine fish demographics (Debicella 2005; Murie et al. 2009). For more than 60 years fin rays and fin spines have been used to age both freshwater and marine fishes (Cass and Beamish 1983; Debicella 2005). The fin ray metho d was li kely first utilized in 1916 for aging sturgeon Acipenser spp. (Beamish 1981). Since then, r esearchers have ap plied these procedures to age a variety of other fishes including but not limited to: salmonids (Burnet 1969; Sikstrom 1983 ; Zymonas and Mc Mahon 2009 ), White Sucker Catostomus commersoni (Beamish and Harvey 1969 ; Sylvester and Berry 2006 ), Lake Whitefish C oregonus clupeaformis ( Mills and Beamish 1980 ; Herbst and Marsden 2011 ), percids (Erickson 1983 ; Niewinski and Ferreri 1999 ), Smallmouth Ba ss Micropterus dolomieu (Rude et al. 2013), Goliath Grouper Epinephelus itajara (Murie et al. 2009), Gag Grouper Myct eroperca microlepis (Debicella 2005), White Grunt Haemulon plumier (Murie and Parkyn 1999), and Gray Triggerfish Balistes capriscus ( Ingram 2001 ; Allman et al. 2015). However, current fin ray and spine methods require further investigation regarding applic ations for sub tropical and tropical fish species where annuli are not laid as discretely as they are for fish in temperate regions (Debice lla 2005) This method has yet to be validated for Largemouth Bass Micropterus salmoides (LMB) in Florida where it wo uld be particularly useful. L argemouth Bass are native to the St. Lawrence River, the Great Lakes (except for Lake Superior), and the Miss issippi River basi ns south to the Gulf of Mexico; t heir range also includes the Atlantic Slope drainages from North Carolina to Florida, and the Gulf Slope drainages from southern Florida in to northern Mexico (Boschung and Mayden 2004; Page and Burr 2011). In
16 Florida, the range of LMB overlaps with the new ly described Florida Largemouth Bass Micropterus floridanus (FLMB ; Barthel et al. 2015 ) Genetically pure populations of FLMB are restricted to systems south of the Suwannee River in the Florida Peninsula F LMB coexist with LMB from the Suwannee River n ort h in what is called an intergrade zone (Barthel et al. 2015) including the fish examined in the present study Additionally, LMB typically mature between 2 and 4 years and t he maximum lifespan of this sp ecies is usually around 10 years (Barthel et al. 20 15). Ages of LMB have been assessed and determined using a variety of methods (Maraldo and MacCrimmon 1979) H istorically, s cales were the preferred and most commonly used aging structure for LMB ( Maceina et al. 2007 ). Scales are viable non lethal aging st ructure s that can be utilized for fast growing, short lived fish; however, edge crowding of annuli that occurs in older specimens (> 10 years) and calcium resorption can lead to the underestimation of age s (Beamish and Harvey 1969; Chilton and Beamish 1982; Dutka Gianelli 1999; Debicella 2005). It can also be challenging to see contrasting patterns when aging scales from fish in non temperate regions (Huish 1954; Duti l and Power 1977; Taubert and Tranquill i 1982 ; Hoyer et al. 1985 ). Subsequently, sagitta otoliths have become the favored aging structure s for Micropterus spp. (Klein et al. 2017) and v alidation procedures have provided evidence of their annual band depos ition (Mil ler and Storck 1982 ; Taubert and Tranquilli 1982 ; Hoyer et al. 1985; Crawford et al. 1989; Buckmeier and Howells 2003 ). However, f in ray or spine excision is a much quicker field technique, and if validated for LMB, would be an extremely valuable non lethal tool that could be coupled with trophy bass tagging during field sampling, conservation efforts, citizen science programs and tournaments where a fin could be clipped for aging and the fish released.
17 Dorsal fin spines and pectoral fin rays wer e originally assessed for LMB from nort hern latitudes by Maraldo and MacCrimmon (1979) and were concluded to be unreliable aging tools due to their apparent inconsistent growth patterns Recently, use of fin rays for non lethal age determination was examin ed for LMB in Florida ( K. Nault K John son and B. Eisenhauer Florida Fish and Wildlife Conservation Commission and J. Kerns and coworkers, University of Florida, unpublished data ). The study concluded that fin structures underestimated ages for older bass in Florida, and ages were consistent ly variable among readers Morehouse et al. (2013), Sotola et al. (2014), and Klein et al. (2017) also assessed aging methods of LMB. Dorsal spines were th e most un biased aging structure and were recommended for aging bigger LMB, whereas use of scales was recommended for estimating ages of younger LMB if possible (Morehouse et al. 2013). Opercles were the best aging structure compared to dorsal spines and sc ales for Sotola Although precision was high for anal spines and dorsal spines, accuracy study. Understanding the basis of systematic bias and variability may provide an avenue for correction of ages and make many assessed structures usefu l for aging LMB. Fin structure aging errors can be a function of counting double bands as two separate bands in stead of one connected band, or counting pse udo annuli (i.e., checks) which are false rings not laid down on a yearly basis (e.g., during spawni ng, fluctuations in temperature, dissolved oxygen, and water level or other stressors; Jackson et al. 2007 ; Snow et al. 2018 ) The first annulus can be di fficult to locate due to erosion of the lumen in dorsal spines which leads to occluded annuli (Iserma nn et al. 2010) Additionally the edge annuli can become compacted as growth slows in older LMB age classes (Debicella 2005) Differences in age estimates among structures may arise for several reasons but principally occur when t he fin structure annuli are
18 not laid down concurrently with the otolith annuli. M is assignment often happens when the first annulus is not located, or there are seasonal differences in the timing of annulus deposition amongst the otolith and fin structures (Vander K ooy 2009 ). The former can be address ed through examination of fish of various ages, and the latter by evaluating aging structures collected serially throughout the year via marginal increment analysis (MIA; Debicella 2005). A MIA is a type of validation technique that h elps determine the timing of annual increment formation in adult fish which sho uld be used when growth is fast and increment s are wide (Geffen 1992) Validation of age is the primary means for determining that annular increment (i.e., ring) counts corresp ond to years of age (Geffen 1992). In some cases, there is a one to one relationship, but in other cases, more band s may be present per year (Bwanika et al. 2007). When plotted as a function of month the outermost increment should display a sinusoidal wav e with a frequency of one year in annuli if the mean marginal growth increment is formed on a yearly cycle (Campana 2001 ). Calculation of the marginal increme nt is usually measured from its state of completion, ranging from approximately 0 mm (an increment would be starting to form) to 1 mm (a complete increment would be formed; Campana 200 1). Given that otolith derived ages have previously been validated for LMB (Miller and Storck 1982; Taubert and Tranquilli 1982; Hoyer et al. 1985; Crawford et al. 1989; Buckmeier and Howells 2003 ) secondary validation of age s in other structures is possible. However, i f annulus deposition varies temporally among different structures validating the growth of a n aging structure is necessary to determine the utility of fin rays and/or spines for aging LMB The purpose of my study was to evaluate empirically the utility o f fin rays and/or spines for aging LMB in Florida as a potentially non lethal means of determining ages. Specifically my objectives were to 1) select a fi n structure both easy to sample and read with high precisio n and
19 accuracy ( i.e., fin structure ages c orrespond well to ages obtained from sagitta otoliths); 2) examine the temporal synchronicity of annulus deposition for sagitta otoliths and the identified fin structure to better understand potential aging biases and how to correct them; and 3) compare ag es obtained from paired samples of sagitta otoliths to the most appropriate fin structure for systematic bias in age assignment.
20 CHAPTER 2 MATERIALS AND METHODS Fish Collections and Measurements Largemouth Bass were collected from Rod man Reservoir in north central Florida during February 2017 January 2018 using daytime boat electrofishing. Rodman Reservoir (3,700 ha ) was chosen for the study because of it s abundant LMB population, accessibility from the University of Florida, and ongoing c omplementary studies conducted by the Florida Fish and Wildlife Conservation Commission ( FW C ) including availability of age and growth data ( E. Nagid, F lorida Fish and W ildlife Conservation Commission personal communication ) The reservoir is a premier L MB fishery, possibly one of the best in the state of Florida. TrophyCatch LMB submissions are comm on here, and it is a popular LMB tournament location ( C. Anderson, Flori da Fish and Wildlife Conservation Commission, pers onal communication ) During the ini tial methodological development a subsample of fish across a variety of sizes were captured for evaluation Thereafter, we attempted to collect f ive fish per centimeter (cm) length class from 20 55 c m to create a viable sample for aging techniques The LMB we re kept in a live well prior to work up, and w e recorded the maximum total length ( MTL ; cm ) of each fish in the field and kept the carcasses on ice until the bony str ucture s were removed In the lab, LMB leng ths were taken to the nearest millimeter (mm) to account for shrinkage out of the cm classes measured in field and fish were weighed on a balance to the nearest g ram (g) Fish were assigned a unique identification code (ID) and sexed prior to otolith extraction S agitta e, the largest otoliths in LMB, were removed from the larger individuals using a hack saw. Sawing began vertically, parallel to the preopercle and ended with breakage of the skull and th e top of the otic capsule (Tesch 1971; Chilton and Beamish 1982) Sagitta o toliths were also extracted from underneath. F ish were turned ventral side up, the opercles were flared
21 and the isthmus and lower gill arches were cut to expose the upper surface of the buccal cavity. T head was then bent dorsally to expose the otic capsule (posterior to the brain, directly behind the eye, and above and fo rward of the first gill arch; Schneidervin and Hubert 1986 ). T he otic capsule was broken using pliers to expose the sagitta otoliths in side. The sagittae were extracted wiped to remove adhering tissue, rinsed in deionized water and stored dry in DWK Life Sciences Wheaton 20 mL PET scintillation vials ( DWK Life Sciences Millville, NJ ) prior to sectioning. Whole fins from the right side of e ach sacrificed LMB were removed Using a sharp pair of Fiskars Softgrip Garden Multi Snip shears (Fiskars C orp. Helsinki, Finland), the pectoral, anal, dorsal, and pelvic fins were removed as close to the body as possible (Fig ure 2 1). T wo techniques were tested in consideration of the no n lethal aging application Fins from fish IDs 42 77 were cut deep into the body while f ins from f ish 1 41 and 78 686 were cut as close to the body as possible. These two techniques were compared for annul i enumeration (i.e., first annulus included or not) After removal, the fins were placed in separate labeled bags corr esponding to their name and fish ID The five separate fin structure bags were placed in one large labeled bag and stored frozen
22 A B C D Figure 2 1. Clipping several L argemouth Bass fins as close to the body as possible A ) Clipping the anal fin with a pair of gardening shears, B) cutting th e pelvic fin off the body C) clipping off the pecto ral fin, a nd D) a trophy size Largemouth Bass with all its fins displayed before the cutting process. Processing of Structures Sagitt a Otoliths D ried right sagitta e were removed from their vials (the left otolith was used if the right was broken) The co re was marked with a pencil on every otolith, and each str ucture was placed in the same location on a fully frosted glass slide MA) Otoliths from fish I Ds 1 41 were either imbedded in two part epoxy or mounted wi th hot glue. These otoliths were heat annealed onto fully frosted slides with C rystalb ond Redding, CA ). All mounted otoliths were cut through a transv erse plane at the core into 0.5 m m sections using a Buehler IsoMet low speed sectioning saw (Buehler Lake Bluff, IL) fitted with three Norton diamond wafering blades ( Saint Gobain, Northboro, MA ) and two spacers t o create two sections. These sections were permanently mounted onto Fishe semi frosted
23 slides ( Waltham, MA ) with Flo t exx ( Waltham, MA ) Otoliths from fish 42 686 were mounted directly on to clear or semi frosted slides using Loctite Super Glue Gel Control ( Henkel Stamford, CT ) cross sectioned, and permanently mounted with H istomount liquid coverslip (National Diagnostics GA) Sections were stored under a fume hood until the mounting medium was completely dry. Fin Rays and Spines In prep aration for mount ing, fins were thawed and then dipped in a simmering ( ~ 95C ) water bat h (Debicella 2005) With a fine bend forceps, the softened dermal tiss ue was peeled off the rays and spines. Using a straight sided scissors, rays 3 5 were excised from the p ectoral, dor sal, and anal fins. F in rays 2 4 were excised from the pelvic fin, and s pines III V we re excised from the dorsal fin. The pelvic spine (i.e., I) and anal spine III were also excised from a subsample of f ishes R ays 3 5, 2 4 and spines III V were chosen because they tend ed to be less damaged, not fused and longer than those preceding or following (Chilton and Beamish 1982 ; Niewinski and Ferreri 1999 ). If these rays or spines looked worn, then adjacent fin ray elements were removed. The e xcised str uctures were stored flat and para llel to one another with the clipped surface ( closest to the base) exposed in coin envelopes (Fig ure 2 2) Fin structures were then dried in a drying box with a fan for two days.
24 A B C Figure 2 2. Thawing, excising, and drying of Largemouth Bass fin rays and spines. A ) Dorsal fin rays thawing pr ior to the excising process, B) the seven properly excised fin structures (left to right: pectoral rays 3 5, anal spine III, anal rays 3 5, pelv ic spine I, pelvic rays 2 4, dorsal spines III V, and dorsal rays 3 5), and C) dorsal rays 3 5 that have been excised, cleaned, and prepped for the drying box. Dried fin structures were labeled, conveniently mounted in Hysol epoxy thermopl astic resin (Loc tite Rocky Hill, CT ; Koch and Quist 2007 ) due to their small breakable sizes and ha rdened for 48 hours in a fume hood (Fig ure 2 3) A ll parts of the fin structures were covered with the two part epoxy t o ensure that the most distal end would not vibrate or slip out of the saw chuck during sectioning and the chuck was tightened properly to avoid forceful fastening which could result in ultimately breaking the mounted fin rays or spines (D. Murie, University of Florida, personal communication) The most basal portion of the fin structures was coated lightly in epoxy to reduce cutting time and potential damage (e.g., chipping) to the blade (D. Murie, University of Florida, personal communication) The fin structures were placed in the saw chuck entirely straight to decrease likelihood of sectioning tangentially Imbedded fin structures were then cross sectioned at least t wo times with a Buehler Is omet low speed saw (Buehler Lake Bluff, IL) fitted with a single Buehler diamond waf ering blade and a set of Buehler flanges that supported the blade ( Buehler Lake Bluff, IL ; Fig ure 2 4) to det ermine an optimal thickness for reading Rays and spines were marked at the widest portion of their bases and then sectioned (~0.7 1.4 mm ) dista lly (two to four sections dependi ng on the size of the structure ).
25 A B C Figure 2 3. Labeling, e mbedding, and drying of Largemouth Bass fin rays and spines. A ) Cleaned and dried spines with the small white l abel that will be e mbedded with the dorsal spines for identification B) dorsal spines I II V imbedded in two part epoxy with the m ost basal portion marked with permanent marker, and C) a large batch of imbedded fin rays and spines drying in the fume hood f or 48 hours. A B C Figure 2 4. Sectioning and permanently mounting Largemouth Bass fin rays and sp ines. A ) Sectioning a fin structure with the low speed saw, B) the first section taken as close to the base as possi ble, and C) cross sections (~0.8 1.4 mm) of fin structures permanently mounted on semi frosted slides. C uts were made ~ 0.8 2.0 mm from wher e t he proximal end of the fin ray and /or spine start ed to curve away from the body of the fish toward the distal tip; t hi s distance varied based on the siz e of the fin structure and how deep the fin was clipped on the body T he general size of the fin stru cture base wa s usually larger and longer in bigger fish than in smaller ones. Th e second cut was made ~ 0.7 1.4 mm from th e first cut depending on how each se ction came out. If a section at ~ 0.8 mm was too thin, the structure was cut between ~0.9 and 1.4 mm S ection ing was
26 continued until a section that could be aged was obtained All s ections were optimally cut to ~1. 0 mm and perm anently mounted with Flo texx on semi frosted slides. Fin structure sections were mounted i n order of thickness for LMB 1 41 and mounted in order of sectioning location (base most distal section) for the remainder of the sample s When permanently mounting fin structure sections to slides with Flo were completely covered with this mounting medium an d any bubbles that formed were removed with a teasing needle Aging of Structures Sagitta Otolith s Transverse sec tions from the right sagitta otoliths were viewed under transmitted light (25 40x power) using a Leica dissecting microscope ( Leica MZ 6, Wet zlar, Germany ) and /or with reflected light using an AmScope microscope LED spot light (AmScope Irvine, CA) for clarification Under transmitted light, dark amber opa que zones were counted to determine ages (Figure 2 5 ) A n arrangement of one dark opaque zone with one translucent zone was int erpreted as a complete annulus ( Debicella 2005 ). A training set of otol ith sections were examined independently by two readers who initially worked together to define criteria for the aging methodology. After establis hing aging criteria, otoliths were read twice by each reader (SL and D C P). The c larity and spacing of annuli w ere assessed and recorded on a scale of 1 3 ( i.e., 3 = perfectly ageable sectioned otolith) LMB a ge classes ( i.e., cohorts) we re identified for o toliths using the birthdate of Januar y 1 (Chilton and Beamish 1982), and a ges were assigned to fish based on opaque rings enumerated and any plus translucent growth ( denoted as 1 4 depending on width of translucent zone ) on the edge of the structure A fis h with two opaque rings extending to the edge and no translucent growth was assigned an age of two y ears A fish with two opaque rings and translucent growth on the edge was assigned two rings and an edge growth
27 combination from 2 1 to 2 4 depending on the amount of translucent zone visible on the edge F ish assigned as 2 1 and 2 2 were given a final age of two years and those assigned as 2 3 and 2 4 were given a final age of three years Figure 2 5 Transverse section (0.5 mm thickness) of a L argemo uth Bass sagitta otolith collected from Rodman Reservoir with 11 counted opaque bands Fin Rays and Fin Spines F in ray and spine sections were viewed under transmitted light (25 100x power ), using a Zeiss compound microscope ( Model # 392560 9004 Jena, Ge rmany; Fig ures 2 7 2 1 3 ) A green 540 nm narrowband interference florescence filter (Olympus Scie ntific Solutions, Waltham, MA ) was used to enhance the visual difference of translucent and opaque zones. In fin ray and spin e sections, a pair of translucent and op aque zones was called an annulus Unlike sagitta e it was the narrow translucent zones that were counted to determine ages (Chilton and Beamish 1982 ; Debicella 2005 ) Fin structures were assigned a number of translucent rings and opaque plus growth ( 1 4) to provide an age estimate. To assess potential non lethal aging meth ods, multiple different fin types were inspected for LMB. The agers attempted to analyze which fin provided the most distinct and/or clear pattern of growth using clarity rankings of 1 3 ; 1 indicated that the section was n early impossible to age (Figure 2 6 ) and a section assig ned a ranking of 3 was c omparable to aging a perfectly
28 sectioned otolith The clarity rankings were based on recognizable annuli and spacing patterns ( e.g., no melding of annuli due to a thin section, edge crowding, doub le banding, checks or settlement effects) Each structure was read twice and then a mean clarity ranking was calculated An aging record was created to store age estimates between two agers (on e experienced ager = DCP and one semi experienced ager = SL ). A B Figure 2 6 Examples of clarity rankings in different pelvic fin ray sections from the same Largemouth Bass A) Largemouth Bass p elvic ray section taken too close to the base, this sec tion was ranked a s a 1 on the clarity scale (1 3) and B) pel vic ray section taken ~ 1 mm distally from the previous basal section, ranked a s a 3 for clarity.
29 A B C Figure 2 7 Pelvic fin ray sections of vario us aged L argemouth Bass A ) Pelvic fin ray section showing both large and small seg ments including a single translucent band on the edge represented by a white dot, B) pelvic ray section aged as 4 years and C) pelvic ray section aged as 7 years white poi nts represent enumerated translucent bands.
30 A B C D Figure 2 8 Dorsal fin ray sections of various aged L argemouth Bass A ) d orsal ray section aged as 1 year showing checks within first year of growth, B) dorsal ra y section aged as 2 years white points represent counted translucent bands, C) dorsal ray section aged as 7 years, showing 7 translucent bands as white dots and plus opaque growth on the edge and D) dorsal ray section aged as 4 years
31 A B C D E F Figure 2 9 Dorsal fin spine sections of various aged L argemouth Bass A ) Dorsal spine sections displaying the core and first translucent band on the edge, B) dorsal spine sections aged as 1 year C) dorsal spin e section aged as 2 years D) dorsal s pine sections aged as 3 years E) dorsal spine section aged as 7 years, white points represent counted translucent zones, and measurement axis is the white line, and F) dorsal spine section aged as 9 years, white point s represent enumerated translucent ban ds.
32 A B C D E F Figure 2 10 Pectoral fin ray sections of various aged L argemouth Bass A ) p ectoral ra y section s aged as 1 year showing just the core, B) pectoral ray s ection aged as 1 year with the first translucent band on the edge, C) pectoral ray section with a double band counted as a single translucent band and plus opaque growth on the edge, D) pectoral ray section with th ree white points that represent translucent bands, E) pectoral ray section (sma ller segment) with three whites points that represent counted bands, and F) pec toral ray section aged as 8 years, white points are enumerated bands
33 A B C D E F Figure 2 11 Anal fin ray section s of various aged L argemouth Bass A ) a nal ray sections with checks inside the first year of growth, B) anal ray sections aged as 1 year arrow shows the end of the first translucent band which is a double band, C) anal ray section aged as 1 year with a d ouble band, D) anal ray section aged as 4 years white points represent counted translucent bands, and white arrow represents the end of the first translucent zone which is a double band, E) anal ray section aged as 6 years with translucent bands represent ed as white points and F) anal ray section with 8 counted bands in white and a red point showing where R eader 1 failed to count a translucent band and the arrow represents the diameter of the core.
34 A B C Figure 2 1 2 Anal f in spine sections of various aged L argemouth Bass A ) Anal spine section, B) anal spine section, and C) anal spine section aged as 4 years white points represent counted bands.
35 A B C Figure 2 13 Pelvic fin spine sections of various aged L argemouth Bass A ) P elvic spine section aged as 1 year B) p elvic spine section aged as 2 years white points represent counted bands, and C) pelvic sp ine section assigned an age of 2 years A subsample of fins from fish 42 77 of the age s 1 9 years was used to assess changes in appearance and presence of annuli from deep cuts ( i.e., into the body/musculature ) to the remainder of samples that were cut as close to the surface of the body as possible Before aging the fin rays and spines some t raining was required. Pre concert readings (Morehouse et al. 2013) were used for training where readers had n o knowledge of fish length s weights, sexes, or collection dates ; however, otolith age s w ere used as aids during trainin g These referenced otolit hs helped guide the readers in learning how to correctly identify annuli in
36 the LMB fin structures. D uring this time, the fish ID s and numbers were not matched to their respective ages to avoid biases. Aging criteria were developed f rom a subsample of fish previously aged by otoliths representing ages 1 7. The second attempt to age fin rays and spines was blind ( i.e., wit hout prior knowledge of otolith based ages). Aging criteria were adjusted to recognize where the first annulus was located and which marks were true annuli versus pseudo annuli (i.e., false annular bands) A QI maging Micropublisher 3.3 digital camera ( QImaging, Surrey, BC, Canada ) was mounted to a Zeiss ( Jena, Germany) compound microscope to capture images o f the bony structure sections. Q C apture Pro 7 software was used to view, age, and measure the digital photographs Most aging disagreements were evaluated wit h the imaging system to decide on a consensus age and it also functioned as a measurement tool for the MIAs Age and Growth A leng th frequency distribution and age distribution were built for otoliths and dorsal spines to assess and compare the size and age composition of the LMB collected from Rodman Reservoir. Von Bertalanffy growth models were used to vi sualize and estimate the no nli near regression parameters: asymptotic length ( L ( K ), and the time in which LMB length would be 0 ( t 0 ; Allen and Hightower 2010) f or ages assigned from sagittae and ages assigned from dorsal spines for both males and females. Parameters were estimated using Excel Solver to calculate maximum likelihood (Allen and Hightower 2010). Precision of Age Estimates Within reader precision ( i.e., reproducibility) of t he fi rst and second ages was assessed for each LMB fin struc ture by calculating the percent agreement ( PA ; Sikstrom 198 3 ; 2 1 ), average percent e rror (APE; Beamish and Fournier 1981; Campana 2001 ; 2 2 ), coefficient of variation (CV; Chang 1982; 2 3), a nd concordance correlation coefficient ( c ; Lin 1989 ;
37 Lin et al. 2002; Lin et al. 2007 ; 2 4 ) Precision was calculated within reader s and between readers Read er 1 was S ummer Lindelien (SL ; moderately experienced ager ~ 4 years ) and Reader 2 was D aryl Parkyn (DCP ; experienced ager ) Low APEs and CVs repre sent ed greater precision of age s. High PAs and c s also indicated better precision of ages. c strength of agreement was classified as almost perfect ( > 0.99 ) substantial ( > 0.95 0.99 ) moderate ( 0.90 0.95 ), and poor (< 0.90) precision of ages (Lin 1989; Lin et al. 2002 ). ( 2 1) (2 2 ) (2 3 ) (2 4 ) Withi n reader bias was assessed with age biplots ( i.e., age bias plots; Campana et al. 1995 ; Debicella 2005; Rude et al. 2013 ). Biases were determined t hrough age comparisons by plotting reading (i.e., age) one against reading two P A between reading one and two was calculated to determine the frequency of complete agreement and agreement within one year for all structur es. Residual age difference plots w ere also used to assess potential aging biases. Between reader age estimates were analyzed using the same methods described above in addition to PA tables (exact and years different) and CV versus otolith mean modal age graphs (Rude et al. 2013) Male an d female LMB dorsal spine section ages were also compared for precision using PA and CV. Reader 1 dorsal spine ages were compared over time ( i.e., by date of aging session) to evaluate change s in aging precision ( i.e., CV) as Reader 1 became more experienc ed
38 Accuracy of Age Estimates Structure related bias es in age estimates w ere assessed using the methods d escribed in Rude et al. (2013), where t he mean modal age s ( i.e., consensus) for each fin structure encompassing data from two readings were computed f or each fish ; otolith modal age was then plotted against the fin structure mean modal age ( i.e., mean age calculated for all fish of a given oto lith modal age) This process assumed that otolith age was an unbiased estimate of the true age of the fish, whi ch was likely reasonable based on Miller and Storck ( 1982 ) Taubert and Tranquilli ( 1982 ) Hoyer et al. ( 1985 ) Crawford et al. ( 1989 ) and Buckmeier and Howells ( 2003 ) The closer the points were to a one to one ratio showed the potential appl icability for non lethal aging (Rude et al. 2013) To assess aging differences related to length, tables grouped by 50 mm MTL increments were created, then PA ex act and years different were calculated A PE CV c and PA were also calculated between structures ( i.e., otolith based ages versus fin structure based ages). Between structure age biases were assessed using age biplots ( Campana et al. 1995 ; Debicella 2005; Rude et al. 2013 ) and residual age difference plots. Male and female LMB dorsal spine section ages were compared for accuracy using PA and CV. Reader 1 dorsa l spine ages were compared over time (i.e., by date of aging session) to evaluate changes in aging accuracy (i.e., CV) as Reader 1 became more ex perienced. Temporal Synchronicity of Annuli Deposition in Sagittae The MIA function ed as the verification technique for the periodicity of growth increment formation in LMB sagitta otoliths. In this study, a n MIA was completed for LMB otoliths from the age 4 cohort To estimate ages and group fish accordingly for sectioning, whole otoliths were placed in a watch glass fi lled with deionized water, aged using a Leica dissecting microscope with reflective light at 25x power and then sectioned following the af orementioned methodology Sectioned otoliths were aged twice at 25x power and measured using the
39 Q Imaging system. The core was located and using the measuring tool, measurements were taken on the dorsal axis ( Debicella 2005; Figure 2 1 4 ) The first measur ement taken was the ultimate radius (mm ; beginning at the last visi ble opaque band to the edge). The second measurement taken was the penultimate (i.e., previous) radius The ultimate radius was divided by the penult imate to calculate the marginal incremen t (mm ; Dutka Gianelli and Murie 1999 ; Murie and Parkyn 2005 ). An av erage was taken for each month ( i.e., 12), and the means and standard errors (SEs) were plotted by month to determine when the smallest growth increment occurred A Figure 2 1 4 Transve rse section (0.5 mm) of a sagitta otolith from a Largemouth Bass showing the core (i.e., C) four (i.e., 1 4) labeled opaque zones along the measurement axis, and the marginal increment on the periphery. Temporal Synchronicity of Annuli Deposition in Dors al Spines S pecimens of the age 4 cohort w ere included in the dorsal spine MIA Fish ID s from the previous grouped age 4 LMB otoliths were matched with their respective dorsal spines, processed and aged. After aging, dorsal s pine sections were measured wit h the Q Imaging system. The core was ident ified by locating and measuring parallel from the inner groove across
40 the structure to the edge. An axis of measurement for dorsal spine sections has not been established for LMB; therefore, measur ements were taken from two designated axes. The first axis was identified as a slow growth axis along the inner concavity dorsal to the core. The second axis was identified as a fast growth axis which was posterior to the radii measurement axis, along the smaller and less elongated lobe ( Fig ure 2 1 5 ) The measurements taken were : an ultimate radius measurement (mm) and a penultimate radius me asurement (mm). The slow and fast axis marginal increments were t hen plotted by month ( i.e., 12) A nnular marks were compared between dorsal spines ( i.e., translucent rings) and otoliths ( i.e., opaque rings) which indicated if the annuli in spines were being laid down at the same time as the otoliths. M arginal incremen t plots helped determin e if there was variance in annuli deposition o f translucent or opaque bands based on the time of year ( i.e., monthly variation)
41 Figure 2 1 5 Slow (white line next to inner groove) and fast (white line adjacent from outer lobe) growth axes for marginal increment analysis of Largemouth Bass dorsal s pines, white line across structure represents radii measurement axis, the arrow represents the core, and the penultimate and ultimate annuli are la beled in black text
42 CHAPTER 3 RESULTS Age and Length of LMB Largemouth Bass ( n = 686; 92 622 mm MTL) were c aptured from Rodman Reservoir (Putnam County, FL ) fro m February 2017 January 2018 Mean MTL of LMB included in the aging subsample ( n = 124) was 37 3 mm MTL (variation = 92 603 mm MTL ; Fig ure 3 1 ). At least one LMB from every targeted cm group (i.e., 20 55 cm) was represented in the aging sub sample ( Figure 3 1 ) In total, 1,024 bony structure cro ss sections were aged by Reader 1 and 487 were aged by Reader 2 The otolith age distribution ranged from 1 to 11 years, and 56 % of the fis h were older than age 3 ( Fig ure 3 2 ). Strong LMB year classes were prevalent for ages 1, 4, and 7 (Figure 3 2). Anal and dorsal ray age es timates varied from 1 to 11 years Dorsal spine ages varied from 1 to 11 years, comparable to sagitta otolith ages ; however, fewer fish were pr esent in the age classes : 1, 4, 7, and 10 years when compared to the otolith age distribution (Figure 3 2). P ectoral ray ages varied from 1 to 10 years P elvic ray ag es varied from 1 to 12 years which represented the oldest estimated age in the study Pel vic spine ag es ranged from 1 7 years, and a nal spine ages ranged from 1 8 years. Pelvic and anal age distributions included younger maximum ages relative to all other subsamples due to the oldest fish sampled having known (i.e., otolith derived) age s of 7 and 8 Maximum age and the number of fish in each age class varied among the fin structures in comparison to sagittae ; therefore, age assignment differed depending on whi ch aging structure was used. G reater differences from known ages decreased the accurac y and p recision of certain fin rays and spines; thus, providing reasoning for choosing one structure over another.
43 Figure 3 1. Length frequency distribution of Largemouth Bass collected from Rodman Reservoir for age estimation via sagitta otoliths Sa mple size is shown. Figure 3 2. Age distribution from aged Largemouth Bass otoliths ( n = 124) and dorsal spines ( n = 122) Mean age and range of aging structures are shown. 0 1 2 3 4 5 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 Frequency (%) Length Group (cm) 0 5 10 15 20 25 30 35 40 1 2 3 4 5 6 7 8 9 10 11 12 Number of Fish Age (years) Otolith Dorsal Spine Mean Dorsal Spine Age = 4.13 Mean Otolith Age = 4.05 Range = 1 11 years n = 124
44 Processing and Clarity of Aging Structures Clipping LMB fins took less time than extracting the sagitta otoliths Specifically, the dorsal fin and pectoral fin took qualitatively the shortest amount of time to cut whereas cutting the anal and pelvic fins was more diffi cult. C ompared to sagitta otoliths, which are norma lly used fo r aging LMB by scientists in Florida, dorsal spines were easier to remove from LMB. Excision time and ease were the same for LMB rays and spines; however, the spin ous structures (anal, pelvic, and dorsal) took less effort to clean than the fin r ays (anal, pelvic, dorsal, and pectoral) Dorsal spines were the easiest fin structure to clean but sagitta otoliths were the easiest overall aging structure to clean. Sectioning sagitta otoliths took the least amount of time of any aging structure. It wa s more diff icult and time consuming to section all of the imbedded fin structures, particularly if they were large. Section thickness had a large influence on readability. Fin structures sectioned at ~1.0 mm provided the best contrast and age estimates, b ut sectioni ng at multiple thicknesses (~0.8 1.4 mm) eliminate d more error s Thicker sections (i.e., 1.0 1.4 mm) provided qualitatively clearer banding patterns and more contrast between growth zones in the fin structures The best contrast and most visible ring patte rns came from sections not t oo close to the base (Figure 2 6 ) or too distal from the base (Figure 3 3 ). The exact distance that provided the most ageable sections depended on the individual fish and each fin str ucture.
45 A B C Figure 3 3. Pelvic fin ray sections from Largemouth Bass (LMB; larger segments) with annuli visible in basal versus distal cuts from a 7 yr old individual. A ) A b asal section from a LMB pelvic fin clipped level with the back of t he fish, B) a section taken from ~1.55 mm dis tal on the fin ray, and C) ~2.55 mm distal cut. Secti oned sagitta otoliths (0.5 mm thickness) were the most consistently clear structure s to read ( mean clarity ranking = 2.87; Table 3 1 ) Annuli were easiest to read al ong the dorsal axis (Figure 2 14 ) wher e bands were not disconnected from the sulcus and were more evenly spaced and measurable The seven f in structures (i.e., pelvic rays and spines, dorsal rays and spines, anal rays and spines, and pectoral rays) varied greatly in the abil ity for their annul i to be discerned (Fig ures 2 7 to 2 1 3 ) ; however, i t was not necessary to cut deep into the body of the fish when clipping the fins. All annuli were retained, and the structures were ageable when the fins were cut as close to the body as possible (Figure 3 4 ) A B Figure 3 4. Largemouth Bass dorsal spine sections cut deep into the body ve rsus cut along the body. A) Section of a dorsal spine assigned an age of 1 year, cut deep into the body when removed, first translucent band i s visible on the edge of the structure, and B) section of a dorsal spine assigned an age of 1 year, cut as close t o the body as possible, first translucent band is also visible.
46 Pectoral rays, dorsal spines, pelvic rays, and pelvic spines provided the hig hest mean clarit y rankings (i.e., 1 3; Table 3 1 ), indicating they were the optimal structures for inferring growt h patterns. Dorsal spine sections provided some of the most identifiable annuli patterns for agin g among fin structures (mean clarity ranking = 2.20; Table 3 1). The qualitatively clearest sections were 1.0 1.4 mm thick. Approximately 36% of dorsal spines ( n = 122) were quantitatively ranked a s a 3 for clarity. Most dorsal spine sections aged as 2 and 4 11 years were given a clarity ranking of 2 (Figure 3 5 ); thus, dorsal spines were viable ag ing structures, especially for older fish. Anal rays, dorsal rays, and anal spines were the least cl ear structures to age (Table 3 1 ); thus, their use as aging tools was less ideal. None of the fin structure s provided an average clarity ranking of less than 2.02, indicating that all fin structures were relati vely clear to read with substantial practice (Table 3 1 ). Table 3 1 Average clarity rankings ( i.e., 1 3) of Largemouth Bass aging structures for all R eader 1 aged fish n = sample size. Bony Structure n Clarity Sagitta Otolith 316 2.87 Anal Ray 124 2.02 Anal Spine 49 2.04 Pectoral Ray 121 2.22 Pelvic Ray 123 2.16 Pelvic Spine 47 2.13 Dorsal Ray 122 2.02 Dorsal Spine 122 2.20
47 Figure 3 5 Do rsal spine section clarity rankings ( i.e., 1 3 ) based on age of Largemouth Bass ( n = 122) Aging Precision A ge estimates from s agitta otoliths provided the lowest CV s and APE s and the highest c s and P A s for Reader 1 and Reader 2 ( i.e., within readers; Table s 3 2 and 3 3 ). O tolith age estimates between readers also provided the lowest CV and APE the highest PA ( 100% of sections were assigned ages 1 year of each other ) and perfect stre ngth of agreement ( c = 1.00) relative to all aging structures (Table 3 4 ). T hese calculated estimates the residual age difference plot and age biplot provided evidence that sectioned sagittae were the most precise LMB aging structures assessed in the p resent study (Figure 3 6 ) 0 10 20 30 40 50 60 70 80 90 100 1 2 3 4 5 6 7 8 9 10 11 Percent of Age Group (%) Age Group (years) 1 2 3
48 Table 3 2 Precision in age estimates for Reader 1 (SL) between two aging sessions using Largemouth Bass otoliths, fin rays, and spines n = sample size APE = average percent error, CV = coefficient of variation and c = coefficient Exact and Years Differen t (% of Fish Aged ) Aging Structure n Exact 1yr 2yr 3yr APE CV c Sagitta Otolit h 316 95 5 0 0 0.88 1.24 1.00 Anal Ray 124 62 34 2 2 5.11 7.23 0.95 Anal Spine 49 59 41 0 0 5.8 8 8.32 0.96 Pelvic Ray 123 73 22 3 2 3.69 5.22 0.9 6 Pelvic Spine 47 64 21 11 4 5.88 8.32 0.91 Dorsal Ray 122 58 36 5 1 5.28 7.47 0.9 6 Dorsal Spine 122 79 17 4 0 2.79 3.95 0.9 8 Pectoral Ray 121 73 22 6 0 3.92 5.54 0.9 7 Table 3 3 Precision in age estimates for Reader 2 (DCP) between two aging sessions using Largemouth Bass otoliths, fi n rays, and spines. n = sample size APE = average percent error, CV = coefficient of variation and c coefficient Exact and Years Different (% of Fish Aged ) Aging Structure n Exact 1yr 2yr 3yr APE CV c Sagitta Otolith 108 92 8 0 0 1.60 2.27 0.99 Anal Ray 54 74 17 4 1 5.94 8.39 0.91 Anal Spine 26 77 19 4 0 4.62 6.53 0.97 Pelvic Ray 43 64 33 2 0 3.36 4.75 0.97 Pelvic Spine 24 63 33 4 0 3.88 5.58 0.96 Dorsal Ray 79 55 22 1 0 5.40 7.44 0.97 Dorsal Spine 91 70 12 2 0 2.86 3.78 0.95 Pectoral Ray 62 68 27 2 0 6.02 8.51 0.95 Table 3 4. Between reader percen t agreement (PA; 1 yr) for Largemouth Bass otolith, fin ray, and spine ages. n = sample size, APE = average percent error, CV = coefficient of c limits ( CLs; precision in age estimates between Reader 1 and Reader 2 ). Aging Structure n PA ( 1 yr; %) APE CV c (95% CLs) Sa gitta Otolith 115 93 (100) 0.90 1.27 0.99 (1.00, 0.99) Anal Ray 44 68 (91) 9.18 12.99 0.92 (0.86, 0.95) Pelvic Ray 43 67 (98) 3.79 5.36 0.92 (0.87, 0.95) Pelvic Spine 24 79 (92) 2.82 3.99 0.97 (0.94, 0.98) Dorsal Ray 78 53 (94) 8.43 11.92 0.94 (0.91, 0 .96) Dorsal Spine 91 65 (92) 5.34 7.56 0.96 (0.94, 0.97) Pectoral Ray 62 51 (82) 10.32 14.59 0.80 (0.71, 0.87)
49 A B Figure 3 6. Age biplot and residual age difference plot for between reader estimates of Largemouth Bass otolith ages. A) Age biplot with Reader 1 and Reader 2 otolith age estimates, diagonal line represents ages where Reader 1 had exact agreement with Rea der 2, and B) residual plot of differences in consensus ages between readers. Circle size represe nts sample size of each age combina tion relative to the largest subsample of otoliths ( n = 38 at age 1). Out of 122 dorsal spine sections 79% were aged identi cally the highest PA of all fin structures aged by Reader 1 and 9 6% of the sections were aged within 1 year of ea ch other APE and CV were the low est and c was the high est among all aged fin structures indicating the dorsal spine was the most precise fin aging structure for Reader 1 (Table 3 2 ). Most dorsal spine precision errors were attributed to age differences due to over agin g young fish and under aging old fish. 0 1 2 3 4 5 6 7 8 9 10 11 0 1 2 3 4 5 6 7 8 9 10 11 Otolith Age Estimated by Ager 1 (SL; years) Otolith Age Estimated by Ager 2 (DCP; years) n = 115 -2.5 -1.5 -0.5 0.5 1.5 2.5 0 2 4 6 8 10 12 Difference Between Ager 1 and Ager 2 (years) Otolith Age Estimated by Ager 1 (years)
50 Pectoral and pelvic ra ys were also precise LMB aging structures based on the high pe rcentage of exact ages (73% for both) ages within 1 yr (94% and 95%, respectively), and low APEs and CVs (Table 3 2 ). Reader 1 age d 64% of pelvic spine sections exact between two aging sessions with a low percent ( 85% ) of ages within 1 yr compared to t he other fin structures assessed (Table 3 2 ). Most pelvic spine aging errors resulted from under aging old fish (i.e., 7 and 8 year olds) by one year and over aging younger fish (i.e., 2, 3, 4, and 5 year old fish) by two or even three years Thus, pelvic spine sections were not precise LMB aging structures for Reader 1, and their sample size was low (Table 3 2 ). Reader 1 aged 62% of anal ray anal spine, and dorsal ray sections exact ly the same (Table 3 2 ). These fin structures were considered less precise for estimating LMB ages due to their high CV s and low APE s ( Table 3 2 ). All of the aged fin structures by Reader 1 were classifie d as m oderately precise (0.90 0.95) and all but the pelvic spine were substantially precise (> 0.95 0. 99) according to c strength of agreemen t (Table 3 2 ) The a nal spine was the most precise LMB fin structure for Reader 2 c and PA though the sample size was relatively low ( n = 26 ; Table 3 3 ). However, dorsal spine section ages provided the lowest CV a nd APE values for Reader 2 and 70% of the sections were aged exact with 93% aged 1 yr (Table 3 3) Dorsal spines had the highest sample size among aged fin structures; therefore, they should be considered the sam e if not better for aging LMB relative to all fin structures assessed by Reader 2 Pelvic rays and spines were also precise fin aging structures for Reader 2 Anal rays had a high PA of ages exact, but a relatively low PA of ages within 1 year; as thus, they were conside red only moderately preci se aging structures based on c strength of agreement (0.91 ; Table 3 3 ). Reader 2 aged less fish than Reader 1 relative to all subsamples in the study (Table s 3 2 and 3 3 ) and was less precise than Reader 1
51 when aging pectoral and dorsal ray sections, but more precise when aging anal ray anal spine and pelvic spine sections ( Table s 3 2 and 3 3 ). The two readers aged the other fin structu re sections relatively similar regarding reproduc ibility; however, Reader 1 had a higher propor tion of fish aged 3 yr compared to Reader 2 (Table s 3 2 and 3 3 ). Precision (i.e., CV; Figure 3 7 ) for Reader 1 dorsal spine ages varied as a function of the date of aging but did not show a distinguishable tr end based on month. Mean age of the subsamples and the number of fish aged al so varied by month. Mean age of dorsal spin e sections from female LMB was 1 year older t han males, and females were 99 mm MTL longer (Table 3 5 ). CV was lower and PA was higher fo r dorsal spine section ages from female LMB, indicating dorsal spines were more precise for aging fema les compared to males (Table 3 5 ). Figure 3 7 Precision based on coefficient of variation of Largem outh Bass dorsal spine section ages by Reader 1 as a function of time (date of aging session ; n = 122 ) Average ages and coefficient of variations are on the same y axis scale. 0 2 4 6 8 10 12 14 16 Date of Dorsal Spine Aging Session Dorsal Spine Average Age (years) Dorsal Spine Average Coefficient of Variation
52 Table 3 5 Reader 1 p recision of dorsal spine section ages from female versus male Largemouth Bass. Sample size = n percent agreement = PA, coefficient of variation = CV, and maximum total length = MTL. Sex n PA (%) CV Average Age (years) Average MTL (mm) Female 64 81 2.88 5 424 Male 58 76 5.12 4 325 P recision of fin structures was also calculated from between reader age estimates. Despite a low sample size ( n = 24) p elvic spines were the mo st precise fin aging structure due to their low between reader CV, low APE, high PA and high c ( Table 3 4 ) Because of the ir high c dorsal spines were also considered precise between readers (0.96; Table 3 4 ) Pectoral, dorsal and anal rays were the l east precise fin structures as eviden ced by high variability between reader ages ( Ta ble 3 4 ). Anal spines wer e not assessed for between reader precision A ging Accurac y D orsal spine sections were aged 77% exact and 98% 1 yr with ages from sectioned otoli ths ( Table 3 6 ) Mean CV values were highest for young fish ( i.e., ages 1, 2, and 3) The dor sal spine versus otolith age biplot showed a tight fit to the one to one ratio line relative to the other fin structure age biplots, indicating less aging errors a nd higher accuracy (Figure 3 8) Table 3 6 Reader 1 (SL) p ercent agreement and disagreement f or Largemouth Bass fin ray and s pine ages ( i.e., accuracy estimates). Sample size = n average percent error = APE, cordance correlation coefficient c Agreement and Disagreement with Otolith Ages (%) Fin Structure n 3yr 2yr 1yr Exact +1yr +2yr +3yr APE CV c Anal Ray 124 0 2 15 61 16 5 1 5.22 10.44 0.95 Anal Spine 49 2 2 20 67 8 0 0 5.44 10.89 0.95 Pelvic Ray 123 1 2 14 64 13 5 2 5.13 10.25 0.94 Pelvic Spine 47 0 6 9 64 15 0 4 6.07 12.14 0.91 Dorsal Ray 122 0 1 15 63 16 5 0 3.69 7.38 0.96 Dorsal Spine 122 0 0 11 77 11 2 0 3.27 6.54 0.98 Pectoral Ray 121 0 3 17 63 14 3 0 3.97 7.94 0.96
53 Figure 3 8 Scatter plot comparison s of age estimates obtained from Largemouth Bass sagitta otoliths versus fin rays and spines for Reader 1. Diagonal lines represent comparisons where otolith age = estimated fin structure age. Circle size represents sample size for e ach age combination rel ative to the largest subsample of each fin structure. 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Dorsal Ray Age (years) Otolith Age (years) n = 122 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Pectoral Ray Age (years) Otolith Age (years) n = 121 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Pelvic Ray Age (years) Otolith Age (years) n = 123 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Dorsal Spine Age (years) Otolith Age (years) n = 122 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Anal Ray Age (years) Otolith Age (years) n = 124 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Pelvic Spine Age (years) Otolith Age (years) n = 47 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Anal Spine Age (years) Otolith Age (years) n = 49
54 Figure 3 9. Residual age difference plots, circle sizes = sample sizes for each Largemouth Bass age combination relative to the largest sample size for each fin structure (Reader 1). -4 -2 0 2 4 0 2 4 6 8 10 12 Difference Between Otolith and Ray Age (years) Otolith Age (years) Anal Fin Rays n = 124 -4 -2 0 2 4 0 2 4 6 8 10 12 Difference Between Otolith and Ray Age (years) Otolith Age (years) Pelvic Fin Rays n = 123 -4 -2 0 2 4 0 2 4 6 8 10 12 Difference Between Otolith and Ray Age (years) Otolith Age (years) Dorsal Fin Rays n = 122 -4 -2 0 2 4 0 2 4 6 8 10 12 Difference Between Otolith and Ray Age (years) Otolith Age (years) Pectoral Fin Rays n = 121 -4 -2 0 2 4 0 2 4 6 8 10 12 Difference Between Otolith and Spine Age (years) Otolith Age (years) Anal Fin Spines n = 49 -4 -2 0 2 4 0 2 4 6 8 10 12 Difference Between Otolith and Spine Age (years) Otolith Age (years) Dorsal Fin Spines n = 122 -4 -2 0 2 4 0 2 4 6 8 10 12 Difference Between Otolith and Spine Age (years) Otolith Age (years) Pelvic Fin Spines n = 47
55 Diffe rences in ages fell along the zero line in the residual age difference plot more than other fin structures (Figure 3 9), indicating that the dorsal spine was the most accurate fin structure for Reader 1. Dorsal spines from smal ler and younger fish ( 151 20 0 mm MTL ; age 1 ) had the lowest percentage of sections aged exact with otoliths compared to other s izes of aged fish (Table 3 7 ) ; therefore, d orsal spine sections from smaller and younger LMB were less accurate to age This issu e w as resolved by using radi i measurements to locate the first annulus (Fig ures 3 19 and 3 20 ; Beamish and Chilton 1977; Murie et al. 2009 ) Dorsal spine s ections from fish of 151 200 mm MTL tended to be mis aged by 1 y ea r and improperly called 2 yr olds, making the proportion of agi ng error equal to 50%. When age overestimation occurred with an older fish (i.e., 6 yr old bass that was mis aged by 1 y ea r and called a 7 yr old ) the proportion of aging error was only equal to 14%. A high pr oportion of error in the age sample was caused by over aging dorsal spine sections from young LMB ( Table 3 7 ) Dorsal spine s ections from fish of 301 350 mm MTL (mostly 2 yr olds) were prone to overestimation of ages by 1 and 2 y ea rs; whereas, s ections fro m fish of 351 400 mm MTL were under aged but only by 1 y ea r (Table 3 7 ) Dor sal spine sections from LMB of 551 600 mm MTL provided the second most accurate ages by size ( Table 3 7 ) Overall, d orsal spine sections from LMB aged 1 4 years tended to be over aged, and LMB aged 5 8 were more likel y to be under age d (Table 3 7 ).
56 Table 3 7. Percent agreement (ages exact and years different) between L argemouth Bass sagitta otolith and dorsal spine ages (accuracy estimates for Reader 1 = SL ). n = sample size; MTL = maximum total length. Ages Exact an d Years Different (% of Fish Aged) MTL (mm) n 3 2 1 E xact 1 2 3 Mean Known Age (years) 0 150 4 0 0 0 100 0 0 0 1 151 200 5 0 0 0 60 40 0 0 1 201 250 13 0 0 0 77 23 0 0 1 251 300 14 0 0 0 93 7 0 0 1 301 350 13 0 0 0 69 23 8 0 2 351 400 18 0 0 17 72 6 6 0 4 401 450 20 0 0 5 85 10 0 0 6 451 500 10 0 0 20 70 10 0 0 7 501 550 15 0 0 27 73 0 0 0 7 551 600 9 0 0 33 67 0 0 0 8 601 650 1 0 0 0 100 0 0 0 9 The otolith dorsal spine linear regression showed ages estimated from both structures tightl y fit to the one to one line ( R 2 = 0.99 ) indicating dorsal spines were highly accurate when referenced to otolith ages (Figure 3 10 ). Dorsal spine ages were more accurate when aging sections from male LMB compared to female LMB as evidenced by the l ow er CV and high er PA (Table 3 8 ). Average age and average CV differed b y month of LMB dorsal spine section aging session ( Fig ure 3 11 ).
57 Figure 3 10 Reader 1 age biplot comparing dorsal spine section mean modal ages to otolith section modal ages. Linear regre ssion dashed line is shown, and 1 standard error around the mean ages assigned to otoliths for all ages assigned to Largemouth Bass dorsal spines are represented by error bars. Coefficient of determination (R 2 ) and regression equation is shown. Solid line represents equal assigned ages. n = 122. Table 3 8 Accuracy of dorsal spine section ages from female versus male Largemouth Bass relative to ot oliths Sample size = n ; p ercent agreement = PA, coefficient of variation = CV and maximum total length = MTL Sex n PA (%) CV Average Age (years) Average MTL (mm) Female 64 70 6.67 5 424 Male 58 85 4.23 4 325 y = 0.944x + 0.2332 R = 0.9942 0 1 2 3 4 5 6 7 8 9 10 11 12 0 1 2 3 4 5 6 7 8 9 10 11 12 Dorsal Spine Mean Modal Age (years) Otolith Modal Age (years)
58 Figure 3 11 Average ages (years) and coefficient of variations (CV s ) as a function of the month the Largemouth Bass dorsal spine section was aged by Reader 1 (a ccuracy of Largemouth Bass dorsal spine section assigned ages; n = 122) Average age and av erage CV are on the same scale along the y axis Dorsal rays and pectoral rays also provided more accurate ages th a n the other fin structures examined ( c = 0.96 for both ; PA = 63% for both ); however, both structures were variable i n years differe nt (Table 3 6 ; Figure 3 9 ). Age data waivered much more considerably from the one to one line and the zero line, indica ting lower accuracy (Figure s 3 8 and 3 9 ) Accuracy was 64% overall when comparing pelvic spine ages to otolith ages, and 87 % of the fish were mis aged by 1 year (Table 3 6 ). Over aging by 1 year ( i.e., age 4) and under aging by 1 or more years ( i.e., ages 5 8) was seen in pelvic spines of older fish; thus, fish 4 years and older were difficult to age via pelvic spines. Pelvic spine ages did not fall along the one to one ratio line as closely as other spinous structures and their sample size was much lower ; thus, they were not as accurate aging st ructures ( n = 47; Figure 3 9 ). 0 1 2 3 4 5 6 7 8 9 Apr-17 Jul-17 Jan-18 Feb-18 Mar-18 Apr-18 May-18 Jun-18 Jul-18 Date of Dorsal Spine Aging Session Average Age Average CV
59 Pelvic fin ray sections ( n = 123) had a PA of 64% with 91% 1 yr when comparing Reader 1 between st ructure age estimates (Table 3 6 ). Figures 3 8 and 3 9 showed that younger fish (age s 1 4 ) were prone to o ver aging, and olde r (age s 5 10 ) fish wer e prone to under aging. Fish sized 151 200 mm MTL were difficult to age accurately (20% exact ) as well as fish siz ed 551 600 mm MTL (33% exact ) Pelvic rays were less accurate than the other assessed fin rays and spines by Reader 1. Anal fin ray sections ( n = 124) had a PA of 61% with 92% 1 yr when comparing between structure (otolith versus anal ray) ages for Reader 1 (Table 3 6 ). APE and CV were hig hest for fish aged 1, 2, and 10 years old (CV = 22.5, 31.3, and 17.7, respectively ). Anal ray sections from f ish of 351 400 mm MTL had the lowest exact PA (33%); 7 of the 18 fish in thi s size group were ove raged by 1 year. S horter (MTL) and younger ( i.e., 1 4 years) fi sh tended to b e overaged by 1 3 years. Fish 6 years and older tended to be under aged by 1 or 2 years ( Figure 3 9 ) The second highest exact PA of a ll size groups was 86% for anal ray sections from fish 251 300 mm MT L Anal rays were not accurate aging structures for Reader 1. Anal spine sections aged by Reader 1 ( n = 49) w ere less accurate than dorsal spine sections based on CV, but they were more accurat e than pelvic spine sections (Table 3 6 ) Anal spines were also prone to under aging (Tabl e 3 6 ). Most age diffe rences were grouped into the minus 1 year category, and some sections were even under aged by 3 years (Figure s 3 8 and 3 9 ). The most accurate fin structure for Reader 2 wa s the anal ray, where ages had an extremely high PA c and low C V ( Table 3 9 ) Although sample sizes were lower for all aged structures, the anal ray age biplot showed near perfect overlap of the one to one line (Figure 3 12). The pelvic spine also prove d to be an accurate aging structure for Reader 2 ( Table 3 9 ) The
60 c of the other fin structures ranged from 0.85 0.94 for Reader 2 ; however, d orsal spines were relatively more accurate than pectoral rays, dorsal rays, or pe lvic rays based on strength of agreement ( c = 0.94; Table 3 9 ). Dorsal ray and pectoral ray age biplots showed noticeable devi ations from the one to one line (Figure 3 12). Anal spines were not assessed for accuracy of Reader 2 ages. Table 3 9 Reader 2 (DCP) p ercent agree ment and disagreement with otolith ages for Largemouth Bass fin ray and spine ages ( i.e., accuracy estimates). Sample size = n average percent error correlation coefficient = c Agreement and Disagreement (%) Fin Structure n 3yr 2yr 1yr Exact +1yr +2yr +3yr APE CV c Anal Ray 44 0 2 0 96 2 0 0 0.68 1.36 0.99 Pectoral Ray 62 3 0 10 68 15 0 3 5.80 11.59 0.85 Pelvic Ray 43 0 5 26 65 2 0 0 7.50 14.99 0.93 Pelvic Spin e 24 0 4 13 75 8 0 0 1.71 3.42 0.97 Dorsal Ray 78 1 3 15 62 13 4 0 6.30 12.61 0.91 Dorsal Spine 91 0 9 20 55 12 4 0 5.07 10.14 0.94
61 Figure 3 12. Scatter plot comparisons of age estimates obtained from Largemouth Bass sagitta otoliths ve rsus fin rays and spines for Reader 2. Diagonal lines represent comparisons where otolith age = estimated fin structure age. Circle size represents sample size of ea ch age combination relative to the largest subsample of each fin structure. Age Assignment Issues Identifying the first annulus in dorsal spine se ctions from young LMB (i.e., 1 4 years) was difficult Pseudo annuli, commonly called checks (Figure 3 13 ) wer e a problem when dorsal spines were sectioned tangentially (i.e., at an angle other than 90 ) rather than transversely ( Figure 3 1 4 ). This slight tilt resulted in false bands that were visible throughout the section (Figure 3 14 ). 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Dorsal Spine Age (years) Otolith Age (years) n = 91 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Dorsal Ray Age (years) Otolith Age (years) 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Pelvic Spine Age (years) Otolith Age (years) n = 24 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Pelvic Ray Age (years) Otolith Age (years) n = 43 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Anal Ray Age (years) Otolith Age (years) n = 44 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 Pectoral Ray Age (years) Otolith Age (years) n = 63 n = 78
62 A B C Figure 3 13 Pseudo annuli in Largemouth Bass dorsal spine sections. A ) Core counted as a translucent band, dorsal spine section assigned an age of 2 years instead of the known age of 1 year, B) dorsal spine section core miscounted as a translucent band, first annulus not formed yet, and C) checks visible within dorsal spine secti on core. Figure 3 14 Largemouth Bass dorsal spine sections cut tangentially (i.e., at an angle other than 90). Dorsal spine annuli were generally easier to enumerate in older fish, unless the first annulus was eroded or the edge was compacted (Figu re 3 15 ). Edge compaction was an issue with fish older than 7 years (Figure 2 9 ). Erosion that resulted in occlusion of the first annulus was addressed b y measurement of radii ( Beamish and Chilton 1977; Murie et al. 2009; Figure 3 20 ).
63 A B Figure 3 15 Compaction of the edge in Largemouth Bass dorsal spine sections. A ) Missed annulus, dorsal spine sectioned assigned an age of 6 years instead of the known age of 7 years, and B) compaction of the edge in a dorsal spine section, w hite points represent counted translucent bands. Interpreting double bands as single bands was a common occurrence when aging dorsal spine sections, and some specimens of certain structures (anal rays and spines, and pelvic rays and spines) formed double b anding patterns for all of their annuli. Often, double bands were counted as two separate annuli instead of just one connected annulus (Figure 3 16 ). A B C Figure 3 16 Double bands in Largemouth Bass dorsal spine sections. A ) Counted double band as two translucent bands instead of a single ban d, B) double banding, white points represent enumerated translucent bands, and C) double bands present. In dorsal spine sections of older fish, it was sometimes hard to recognize where the first annulus was. The first translucent band was hard to distinguish or locate when blood vessels spread outwards from the core and covered radius 1 (Figure 3 17 ). It was most noticeable in some of the dorsal spine sections of the oldest fish (i.e., 7 10 years) in the sample. Over time, the first translucent band sometimes began to fade into the core (Figure 3 18 ).
64 A B Figure 3 17 Erosion of the first annulus in Largemouth Bass dorsal spine sections. A ) First annulus hard to identify, occluded by lumen and dermal tissue, aged as 6 instead of 7, white line represents radii measurement axis, and B) missed first annulus, white points represent counted bands. A B Figure 3 18 Recognition of the first annulus is Largemo uth Bass dorsal spine sections. A) C ore misidentified as first translucent band, white line represe nts measurement axis, white points represent trans lucent bands, and B ) dorsal spine section, white points represent counted translucent bands. Dorsal spine s ections from LMB ( n = 114) were measured with the QImaging system t o locate the firs t and second annuli (Figure 3 1 9 ). Fish aged 1 11 years were incl uded, and the average measurement for annular radius 1 was 0.54 mm ( range of radius 1 = 0.29 0.82 mm). The average measurement for annular radius 2 was 0.69 mm (range = 0.38 0.95 mm). Overlap of radii in younger fish ( i.e., ages 1 5) was identified, indica ting that the position of the first annulus was not consistent during aging of these fish, leading to an ov erestimation of their ages (Figure
65 3 20 ). Radii measurements that were not clustered around their median were interpreted respectively as over or und erestimated ages from dorsal spine sections. Most overlap was seen in radius 2 overlapping radius 1, indica ting overestimate d ages (Figure 3 20 ). Figure 3 1 9 Dorsal spine section from an age 7 L argemouth Bass showing annular radii m easurement axis in white, 7 translucent bands indicated by white dots, and fast and slow growth axes in white annular ra dius 1 = 0.54 mm, annular radius 2 = 0.69 mm Figure 3 20 First and second annula r radii in Largemouth Bass dorsal spines ( n = 114 ) as a function of age. Horizontal lines = median measurement for annulus 1 and 2. 0.00 0.20 0.40 0.60 0.80 1.00 1.20 0 2 4 6 8 10 12 Annular Radii (mm) Age (years) Radius 1 Radius 2 Median Radius 1 Median Radius 2
66 Growth of LMB The von Bertalanffy growth curve parameters: asymptotic length ( L ), Brody growth rate coefficient ( K ), and the age when LMB would have been zero length ( t 0 ) estimated according to Allen and Hightower (2010) were different among the sagittae and dorsal spines ( Figures 3 21 and 3 22 ). The parameter estimates were also different within structures for male and female LMB (Figures 3 21 and 3 22 ). The asymptotic length was highest for females aged via sagittae (723 mm; Fi gure 3 21 ). T he asymptotic length was higher for males a ged via dorsal spine sections when compared to males aged wi th sagitta otoliths (Figure 3 22 ). Males aged via sagitta otoliths approached the asymptotic length faster ( K = 0.38) than both males aged via dorsal spine sections ( K = 0.27) and females aged by both structures (Figure 3 22 ). Even though males aged via dorsal spine sections grew slower, they attained larger sizes than males aged via sagittae (Figure s 3 21 and 3 22 ) Growth curves for LMB in Rodman Reservoir differed depending on which aging struct ure was modeled and which sex was modeled. Male LMB were shorter at age compared to females, especially for older fish when aged using sagittae ( 4 years; Figure 3 21 ). Male LMB were shorter at age compared to females when aged us ing dorsal spines, but the growth curve began to differ between sexes at an earlier age than in fish aged using sagit tae ( 3 years; Figure 3 22 ). The growth curve for sa gitta otoliths from female LMB showed a higher predicted length (i.e., 250 mm MTL) for young fish (i.e., age 1) whereas the dorsal spine growth curve for female LMB showed a shorter predicted length at age for young LMB (i.e., age 1). The dorsal spine gro wth curve indicated that female LMB would be older at a smaller size than those aged via otoliths. Overall, the curves were comparable for observed and predicted growth of LMB for dorsal spines and sagit ta otoliths. Thus, d orsal spines appear viable for mo deli ng growth of LMB in Florida.
67 Figure 3 21 V on Bertalanffy growth cur ves for aged Rodman Reservoir Largemouth Bass (LMB) sagitta otolith sections n = 124; sexes combined LMB length at sagitta otolith age t = L t ), aged female LMB sagittae ( ; n = 64; female LMB length at sagitta otolith age t = L tf ), and aged male LMB sagitta otoliths ( ; n = 60; male LMB length at sagitta otolith age t = L tm ). Points represent observed values and lines represent predicted values. Figure 3 22 Von Bertal anffy growth curve s for aged Rodman Reservoir Largemouth Bass (LMB) dorsal spine sections n = 122; LMB length at dorsal spine age t = L t ), aged female LMB dorsal spines ( ; n = 64; female LMB length at dorsal spine age t = L tf ), and aged male LMB dorsal spines ( ; n = 58; male LMB length at dorsal spine age t = L tm ). Points represent observed values and lines represent predicted values. 0 100 200 300 400 500 600 700 0 2 4 6 8 10 12 Maximum Total Length (mm) Sagitta Otolith Age (years) 0 100 200 300 400 500 600 700 0 2 4 6 8 10 12 Maximum Total Length (mm) Dorsal Spine Ages (years) Lt = 594.83 [1 e 0.23( t + 1.00) ] L t = 586.14 [1 e 0.24( t + 1.09) ] L t f = 723.14 [1 e 0.14( t + 2.13) ] L t m = 470.75 [1 e 0.38( t + 0.70) ] L tm = 499.27 [1 e 0.27( t + 1.20) ] L tf = 709.65 [1 e 0.16( t + 1.49) ]
68 Marginal Increment Analysis Sagitta Otoliths The minimum mean marginal increment for sagittae sections ( n = 83) was 0.50 mm in April (Figure 3 23 ). April, May, and June had the smallest mean growth increments relative to the other months, indicating that the opaque bands in otoliths were laid down during these months (i.e., Mid April to June). The transluc ent zone formation began in July and was completed by March of the following year (Figure 3 23) Figure 3 23 Age 4 L argemouth Bass mean marginal translucent growth increm ent ( 1 SE) of otoliths over a 12 month period. Dorsal Spines The MIA for dorsal spine sections provided evidence that the translucent zones began forming in November and were completed by April of the following year (Figure 3 24 ). The average minimum marginal g rowth increment in dorsal spines was 0.19 mm in th e month of November (Fig ure 3 24 ). The translucent band was on the edge of the dorsal spine from 0.00 0.20 0.40 0.60 0.80 1.00 1.20 F M A M J J A S O N D J Mean Growth Increment (mm) Month n = 83
69 November to early April which was a relatively long time of slow growth deposition (Figure 3 24) This indica ted that the opaque bands were deposited in the end of April and completed in October (a smaller window of fast growth relative to otoliths). On average, the dorsal spine laid down its translucent bands (i.e., enumerated rings) five months prior to the oto lith laying down its opaque growth bands (i.e., enumerated rings). The dor sal spine and otolith were synchronous in laying down both translucent bands (i.e., November March) and opaque bands (i.e., April June). Figure 3 24 Age 4 L argemouth Bass mean marginal opaque growth increm ent of dorsal spines (DS; n = 81) over a 12 mo nth period showing fast ( 1 standard error; SE) and slow ( 1 SE) growth measured axes, and mean marginal translucent growth increment ( 1 SE) of otoliths ( n = 83). 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 J F M A M J J A S O N D Mean Marginal Increment (mm) Month Fast Growth DS Slow Growth DS Otoliths
70 CHAPTER 4 DISCUSSION Recommendations Although Largemouth Bass sagittae yielded the most precise ages for both readers, precision and accuracy of fin structures varied for each reader. This variability suggests that with substantial practice, multiple fin structu res could be used to non lethally age the Rodman Reservoir LMB population. H oweve r, the largest subsample of fin structures aged between readers and for Reader 2 was for dorsal spines, which were the second most precise fin structure between readers and the third most accurate fin structure for Reader 2 based on c However, dorsal spi nes were the most precise and accurate fin structure for Reader 1 across all calculations. Although processin g and aging times were not recorded dorsal spines were easiest to remove excise, and clean in the timeliest manner After sectioning they were t he second clearest fin structure to age. Due to these positive factors, the dorsal spine was identified for t he marginal increment analysis radii measurements, analyses of age related to sex and modeling of the von Bertalanffy growth curve s to further un derstand it s value as a n aging tool for LMB in Rodman Reservoir. Female LMB were separated from male LMB beca use females usually attain larger sizes and have faster growth rates than males (A llen et al. 2002) which may have been reflected in their respect ive dorsal spine growth patterns and the modeled growth curves Males were possibly less precise to age becau se of smaller growth increments and potential overlapping bands within dorsal spine sections of fish that had already reached the a symptotic length in which time, growth was slowed Understanding growth patterns and establishing aging criteria for dorsal spines was pertinent to properly assign ing ages (i.e., high accuracy and precision). T his methodology could be expanded to all other fin structures ; however, substantial practice would be necessary to obtain precise and accurate ages.
71 Removing and Processing Structures Clipping LMB fins was not as difficult or tim e consuming as removing sagittae Of all the fins removed, cutting off the anal and pel vic fins was the most difficult because of their shortness and reduced accessibility due to placement on the fi be harder to excise than other structures if the fish was alive The fins were all c ut from lifeless fish, but when applyi ng this method to live fish, removal would require two hands. Dorsal spines and pectoral rays were the easiest fins to cut off, and if this process was done on live specimens, these two fins would most likely be the least difficult to excise and clip. The dorsal spines are anterior on the dorsal fin and much more accessible than the posterior dorsal rays. Central rays and spines (e.g., 3 5 and III V, respectively) were assessed, howe ver a small number of dorsal spine I were also processed and aged. The fir st two spines were harder to work with because of their small si ze which increased the po tential for errors during excision It would be easier (i.e., two less cuts) to ask an angler or biologist to cut off the first three dorsal spines with cutting shears compared to making two knife cuts and th en excisin g the central spines The fins that we re retain ed all of their annuli but f isherme n and biologists wi ll need to learn proper clipping techniques so the fin cuts are not made too d orsally. A s we move distally from the base of the fin ray or spine there is a highe r risk of cutting off annuli Missing annuli are problematic during assignment of ages and can lead to improper estimates Proper removal techniques for fin structures are necessary and should be standardized to species just how otol ith removal has been. Cleaning and excising th e fins was laborious at first. During excision, time and ease were similar among spines and rays; however, the spinous structu res (anal, pelvic, and do rsal) took less effort to clean than the fin rays (anal, pe lvic, dorsal, and pectoral). Cleaning fin rays was most problematic because of their branched and segmented structure that created a space for
72 tissue which had to be meticu lously removed Fin s p ines were simpler to clean with the dorsal spine being the si mplest due to its unbranched structure Excision of rays and spines was the same once thawed; h owever, no e xcision was required of otoliths, and they were rubbed clean with water after a few seconds. Embedding otoliths was n ot necessary, whereas a substant ial amount of time was spent embedding the fin structures. However, i f this process was done in batches processing was more efficient Initially, all of the fin rays and spines were sectioned at ~0.8 mm using three blades and two spacers, which resulted in cutting the fins too long and polishing of the cut surface, causing annuli to appear blurry; this is why the structures were sectioned at differential widths for the rem ainder of the project Sections taken too close to the base were full of dermal tiss ue, and those too distal from the base showed cloudier, unapparent bands. If the fin sections were cut sideways (i.e., tangentially) annuli were also hard er to read due to the increased prevalence of double bands or checks Sectioning sagitta otoliths took the least amount of time of any aging struc ture. It was more time consuming to section all of the imbedded fin structures, particularly if they were large. Aging Fin Str ucture Sections Throughout the aging process, varied banding patterns we re seen within each sectioned structure. More practice was necessary to recognize true bands due to within structure and between structure differences in growth. When aging anal ray s ections, recognition of the first annulus and distinguishing false from true annuli was difficult, leading to over and under estimated LMB ages were cut too thin ( e.g., 0.7 mm), sing le or multiple translucent zone s (i.e., checks; Figure 2 11 ) were reveal ed within the first annulus. Double ban ds visible in the first years of growth also led to overestimated ages Additionally, annuli tended to stack along the inner concavity of the ana l
73 ray sections, directly across from the maximum convex point, making en umera tion more difficult The use of solid aging criteria, an understanding of anal ray growth patterns and previous experience aging similar structures allowed Reader 2 to interpret this structure more accurately than Reader 1 Although p elvic and pectoral ray sections may have provided comparably (i.e., anal and dorsal rays) more precise ages due to their annuli being easier to locate in both the small and large segments there were several factors that led to erroneous age estimates Pec toral ray ages were prone to overestimat ion when the first annulus was a double band (i.e., counted as two annuli instead of one), affecting how close the points were to the one to one age bias plot line If the core was recognized, the f irst year o f growth had relatively wide increments that were easy to identify Careful preparation, such as not sectioning or mounting tangentially was optimal for this structure. Failure to locate and/or recognize a n annulu s was the most difficult step, especially in young fish (i.e., ages 1 4) which resulted in variable ages for pectoral and pelvic ray sections Within pelvic rays, sometimes the first counted band was not a true annulus, but a check, resulting in fi sh being over aged. However, the succinct growth pattern ( i.e., distinctive double bands and annuli) commonly found in this str ucture made it easier to age Dorsal fin rays grew similar to anal rays in that numerous double bands and checks were present th at contr ibuted to overaged fish. These soft dorsal rays were troublesome for both readers as seen by the ir low precisio n and accuracy values. It appears that this structure requires more time and focus to age properly, and it may not b e the best option for aging L MB. H owever it w as considered precise for aging other fishes such as: Gag Grouper and Goliath Grouper (Debicella 2005 ; Murie et al. 2009 ). Assigned ages from dorsal rays w ere less than or equal to
74 59% agreement (2005) study and my present study, and more precise in Anal and pelv ic spine sections had large surface areas for growth where annuli deposition was hard to understand due to the inconsistent band widths among years. Edge grow th was particular ly hard to comprehend in pelvic spine sections due to the first few growth zones being wide, as well as multiple/double bands that caused over aging; this left little space for interpretation of annuli on the per iphery. Growth bands were e xcessively thick and difficult to follow completely around the structure leading to aging inaccuracies. Although the p elvic and anal spine sections had large surface areas, compacted edges were an issue that caused underestimated ages of older LMB. Based on my research, f ew to no studies have assessed the validity of the pelvic spine a s a non anal sp ines were not a suitable alternative to otoliths based on accuracy (Table 4 1). Compared to fin rays dorsal spin e sections had a larger surface area, but compared to all other fin structures they displayed more consistent ly distinct growth patterns similar to growth in sagittae The larger surface area within dorsal spine sections allowed for annu li to be deposite d with less crowding, especially on the edge in older specimens. Within these older fish, dorsal spines contained less checks compared to other fin structures, allowing higher precision and accuracy of age estimates. Although aging dorsal spine sections wa s regarded as less challenging, several aging imprecisions were key to understanding growth in this structure. F rom the start, distinguishing pseudo annuli (i.e., checks) from true annuli was one of the mo st difficult processes. T he core w as frequently mis i dentified as the first annulus, and more checks were found inside the core of dorsal spines in 1 year old fish. Commonly 1 year old fish were aged as 2 year olds when checks within the core were inappropriately called annuli. Many
75 young LMB were over age d due to these artifacts which could have formed from spawning, changes in tem perature, water levels, or other factors and/or stresso rs (Snow et al. 2018) Younger fish were prone to these deviations (e.g., settlement checks) hence it was often harder to assign their ages. Many non lethal aging studies have attributed over aging fish to checks being present (Porta et al. 2018). Some ch ecks were probably also visible because of processing error, either sectioning the dorsal spine tangentially, or not cleani ng the structure properly and then attempting to take sections from it. It was difficult at times to see all the translucent band s o n the edge of the dorsal spines for older individuals, yet growth was more patterned and easier to follow completely around these sections compared to younger fish, where Reader 1 struggled to locate the first annulus. Edge compaction was not as concern ing as reading checks when the proper equipment (i.e., QImaging system, compound microscope, and a filter) wa s used for viewin g the sections H owever, compaction of annuli usually occurred as fish grew older (i.e., age s 8 10) and was displayed as stacked band ing patterns on the edge of the structure. F ish with slowed growth were under aged due to translucent rings overlapping one another. If the dorsal spine section s were viewed with 100x power under a compound microscope, or projected onto a screen during agi ng, it was much less challenging to count these overlapping or indistinguishable bands and errors decreased Like other fi n structures, d ouble bands were common in the aged subsample of LMB dorsal spine sections. There was no particular age in which this imprec ision was more commonly found With substantial practice (i.e., repeatedly aging fish of various ages, recording deta iled aging criteria, and learning to pick out incongruities), a trained reader could distinguish double banding patterns from normal banding patterns and coun t them accurately However, when a
76 reader was unable to recognize a double b and from a single band the fish was g enerally over aged Erosion of the first annulus was another common aging issue found in dorsal spine sections of older LMB (i.e., age s 4+) expanding lumen (i.e., when it was fille d with a blood vessel) which made finding and measur ing the first annulus difficult Fish tended to be under aged due to this artifact (e.g., aged as 6 instead of 7). As the dorsal spine grew, it appeared that the lumen grew larger as well. Thi s should be kept in mind as it can lead to under aged fish by 1 or 2 years. E rosion could be recognized with experience A s growth slow ed (i.e., translucent bands grew closer together ) annuli were still mostly interpretable, and an 11 yr old LMB section was aged acc urately using the dorsal spine Additionally, older spine samples occasionally had an inconspicuous or faded first annulus. T his led to under aging of fish using spines. The radii measurement tool also assisted in locating an occluded first annulus in old er fish to prevent missing a growth band and under estimati ng ages ( Beamish and Chilton 1977; Murie et al. 2009 ). Radii measur ements helped validate readers understand ing of aging criteria and banding patterns so ages of dorsal spine sections could be prop erly assigned. Measurements from t his tool could be developed for other aging structures as well. Lastly, i nterpretation of th e edge was a difficult problem to assess, especially while conducting the MIA. Even when usin g a measuring tool, it was hard to un derstand where annuli began and ended on a sec tion which may have been due to the age of the fish being analyzed. It might be more appropriate to use a young er age class (e.g., age 2 or 3) when conducting the MIA so the band widths are larger and easier to see and measure on the edge of the LMB dorsal spine
77 sections. The age 4 cohort was chosen in this study because thes e LMB tended to be mature and closer to the size of trophy bass. T he edge could have also been obscured by a remaining folded layer of skin or slight breakage of the excised dorsal spine (Debicella 2005; M. Hoyer, University of Florid a, personal communicat ion). Regardless, t he periphery needs to be further studied and temporal synchroni city better understood regarding annuli formation in the dorsal spine. These aging issues associated with growth of dorsal spines were recognized through aging a large sample size of various ages and substantial practice with one structure at a time to prevent confusion of annuli placement between structures. Reader 1 aged a larger sample size of dorsal spines and beca me more familiar with their growth patterns and set of cr iteria as o pposed to Reader 2 This could be why dorsal spines were aged most accurately and precisely by Reader 1 relative to all fin stru ctures, yet they were n ot aged as accurately by Reader 2. Better establishment and communication of aging criteria wo uld undoubtedly increase accuracy and precision of a given structure for all readers. Dual aging with both otoliths and a fin structure (s) could enhance a scientist s understanding of the Rodman Reservoir LMB population. Otoliths could still be used to as sess the age structure of younger fish ( i.e., 1 4 years), and fin structures, specifically the dorsal spine could be used to age older fish (i.e., trophy bass; ages 5+ in this study). Both structure estimates could then be comp ared for validation Sagitta otoliths have their own potential aging biases and considering them as known ages poses some issues if a single annulus is not laid down on a yearly basis. This would affect the validity of structure a ges estimated for precision and accuracy ; h owever, high precision and accuracy estimates len t support for their use as a in this study The MIA showed a single annulus was laid in April, but validation for Rodman Reservoir otoliths has not been done before. Additionally, sagittae are
78 utilized for aging LMB by the FWC in Gainesville, FL. Ideally, fish with passive integrated transponder tags or kn o wn ages from hatcheries would have their ages compared to t he otoliths analyzed Comparative Studies The use of non lethal aging structures has a significant role in fisheries m anagement. As a result, many studies have explor ed the use of non lethal metho ds for determining fish age. Although a PA of 80% for precision or accuracy was only achieved for one o f the assessed fin structures (i.e., anal fin ray) multiple fin rays and spines could be used to help assess and validate LMB ages Several recent com parable LMB aging studies have also focused on utilizing fin structures ( K. Nault K Johnson, and B. Eisenhauer, Florida Fish and Wildlife Conservation Commission, and J. Kerns and coworkers, University of Florida, unpublished data ; Morehouse et al. 2013; Sotola e t al. 2014; Klein et al. 2017); wherea s many additional non lethal aging studies using fin rays and/or spi nes have been conducted on different species. In unpublished study, precision was low for all LMB fin structures (PA 50%; for anal rays, dorsal spines, and dorsal rays), and do rsal spines had low agreement with otolith a ges ( Table 4 2) In contrast, my study showed that all fin structures were semi precise (PA 55%) for aging LMB in a repeatable manner, perhaps because more type s of fin structures and larger sample sizes were a ged. Fin structure age estimations were variable among and within readers for both studies, possibly due to reader experience, something Rude et al (2013) assessed for Smallmouth Bass. Aging experience can affect the proper assignment of ages; thus, the id entification of one fin structure over a nother (Rude et al. 2013). In this study, Reader 1 was considered semi experienced and had aged Sander vitreus Walleye and Perca flavescens Yellow Perch dorsal spines prior to this research but no other fin structure s. Reader 2 was an experienced
79 ager and had aged many different fin structures, with emphasis on fin rays. Having a multitude of agers with differential experience could affect proper characterization of th e age structure of a fish population Preferably, e xperienced agers or trained scientists should develop and utilize non lethal aging techniques so ages are properly assigned to fish in a population. Skill is clearly a large factor when high accuracy or pr ecision of age estimates is desired Age estimate s from the anal and do rsal spines from LMB in a Georgia study also provided low concordance (i.e., low accuracy; Klein et al. 2017; Table 4 1) but high precision with known age, passive transponder tagged LMB. Sagitta otoliths exhibited the highest concor dance with known ages, and they were the most precise aging structure (CV = 0.0; Klein et al. 2017), comparable to my study. However, the present study did not include known age fish from a hatchery or tagg ing event, therefore the accuracy of sagitta otoli th ages was not assessed, just precision Sagitta should be used when stakeholders are not concerned with sacrifice or multiple structures are needed. Parr et al. (2018) even suggested reading structures co ncurrently to achieve the best age estimate for a fish. Even though Niewinski and Ferreri (1999) found otoliths to be the best aging structures based on high between reader agreement (96%) and low CV, dorsal spines were more precise than scales for older f ish ( i.e., ages 4+) and provided a CV of less than 10%, comparable to Reader 1 aged dorsal spines. This suggests that the use of more than one structure should be considered when estimating the ages of fish in a population (Niewinski and Ferreri 1999). However, otoliths (not just sagittae) are not always the most accurate aging structures for comparisons among species, for instance with Gar (King et al. 2018), so comparing CV or PA of otoliths to fin structures would not n ecessarily be a great es timator for how accurate dorsal spine s are relative to fin st ructures in other species
80 D orsal spines sometimes underestimated ages in old fish (i. e., ages 4+). unpublished study also identified ages from LMB older than 7 years as underestimated. U nderestimation of older ages has been documented often in oth er fin aging studies involving various fi sh species, such as Yellow Perch dorsal spines ( Niewinski and Ferreri 1999 ). Although this aging occurrence was present analyzing older fish seemed to be less difficu lt than identifying false annuli in young LMB F or instance, Zymonas and McMahon (2009) encountered high precision in Salvelinus confluentus Bull Trout pelvic fin rays in fish older than age 4. Granted, the sample sizes in their and my present study need to be increased for older specimens to fully unde rstand if older fish are easie r to age than younger fish when sample sizes of large fish are more robust. Scientists should seek a balance between selecting a structure with high precision and ac curacy and one with a high sample size. E ach fin structure ag ing study had differing sample sizes which ultimately affected the streng th of precision or accuracy of age estimates Dorsal spines were the most precise and mos t unbiased bony structure in an Indiana aging study examining LMB pectoral fin rays and dorsa l fin spines, but their precision estimates were much lower than those of my study ( Morehouse et al. 2013; Table 4 2 ) Morehouse et al. (2013) a ged more fin struct ures, but they agreed that edge crowding was less important when aging LMB compared to o ther aging imprecisions (i.e., erosion of the first annulus) ; a lthough both studies found identi fying the first annulus problem atic when it w as obstructed by the lumen Other studies also had issues with identifying the first few ann uli in fin structures for o ther species ( Zymonas and McMahon 2009 ; Iserman n et al. 2010; Rude et al. 2013; Koch et al. 2017 ).
81 Sotola et al. (2014) in New York also assessed LMB aging struct ures ( Table 4 1 ). O toliths were the least biased structure (i.e., average slope of 0.91 from the reader age bias regression ) comparable to my study The ir d orsal spine reader age bias regression provided a slope of 0.62, indicating between reader bias (Sotola et al. 2014) Bi ases in my study could have also be assessed using these pairwise age co mparisons by regressing reading 1 against reading 2, then conducting a t test to test the null hypothesis that the intercept ( 0 ) was not si gnificantly different from 0 and the slope ( 1 ) was not significantly different from 1, indicating one to one agreement between or within age estimates ( Maraldo and MacCrimmon 1979; Campana et al. 1995; Rude et al. 2013). Rejection of either hypothesis wo uld be interpreted as reader bias (Rude et al. 2013). This comparison technique was used by Maraldo and MacCrimmon (1979) who originally assessed LMB aging structures and found the clarity of some dorsal spine and pectoral ray section annuli to be atypical (i.e., diff iculty assigning first annulus and pseudo annuli present) Preparation was noted to have been hard, possi bly leading to aging difficulty as noted earlier However, unlike my study, t he opaque and translucent zones were lacking in definition whi ch made them hard to unde rstand and age ; these growth patterns were considered too inconsistent and inaccu rate for use in age and growth (Maraldo and MacCrimmon 1979). Other non lethal studies explored different preparation techniques for do rsal spines ( e.g., unsectioned and polished; Logsdon 2007) which could also be assessed for LMB aging precision and accuracy at a future date. Precision and accuracy issues have been documented for dorsal spines, though they are assumed to have annual mark s, bu t their use for age est imation has yet to be fully validated for LMB (Maraldo and MacCrimmon 1979). My study began the process of validation through a MIA.
82 Among the other LMB studies, our results fared well for precision of fin structures, indicatin g the pote ntial use of multiple fin structures for aging LMB in a repeatable, non lethal form (Table 4 2). Among the studies using dorsal spi nes, my study ranked as higher than average for precision Reader 2 aged LMB anal fin rays with the same a ccuracy ( PA = 96%) as Zymo nas and McMahon (2009) did for Bull Trout pelvic fin rays. Both soft rays had similar sampl e sizes and were influenced by r eader experience. Reader 2 had experience aging Goliath Grouper dorsal fin rays and found them to be accura te (PA = 71%) when using otoliths as known ages in another study (Murie et al. 2009). White Sucker pecto ral rays were aged with higher accuracy than my study and had a very robust sample size ( n = 229; PA = 79%; Sylvester and Berry 2006). Pelvic, pectoral, and dors al rays were less accurate for b oth readers in my current study compared to other non lethal studies. These fin structures were not commonly assessed for LMB, besides pectoral fin rays for Rude et al. ( 2013 ), where they were potentially more preci se but le ss accurate (PA = 46%) than my study Within reader PA was 91% f or dorsal spines in Lobotes surinamensis Tripletail and between reader precision was also high (Parr et al. 2018) D orsal spines proved to be more accurate than in my study (Parr et al. 2018 ; PA = 84%); however, there were a h igher number of age estimates compared Overall, both dorsal spines and the other assessed fin structures appeared t o be practical for assigning ages to LMB when compa red to several other simi lar studies (Table s 4 1 and 4 2 ). All of the LMB aging studies besides my research and conducted in latitudes north of Florida (e.g., Georgia; New Yo rk; Ontario; and Ind iana) where banding patterns were easier to identify and enumerate due to stronger seasonality in growth patterns This may also be why other species studies were able to achieve higher precision and accuracy (e.g., Niewinski and Ferrer i 1999; Zymonas and McMahon 2009 ; Parr et al. 2018 ).
83 Some of t he marine non lethal studies were more comparable location wise ( Debicella 2005; Murie et al. 2009; Allman et al. 2016 ) and provided similar accuracy and precision as LMB fin structures. Unders tanding the basis for age and growth of fin structures in LMB from sub tropi cal Florida provides a pathway for further research to be completed. The American Fisheries Society (AFS) conducted a survey in 2007 and found that only 5% of 45 Canadian and Ameri can management agencies used spines and 2% used fin rays to age LMB (Maceina et al. 2007). Readers should be sure that the chosen aging structure provides p recise and accurate results through a rapid rate of analysis from experienced agers, adequate sample sizes, and proper processing and aging techniques for each species (Sotola et al. 2014). Table 4 1 Between structure age estimates ( i.e., fin structure versus otolith; n = sample size, PA = percent agreement; CV = coefficient of variat ion) for current and comparable studies using several Largemouth Bass fin structures. Fin St ructure n PA (%) CV Study Dorsal Spine 48 30 30.0 Nault et al. (2012) Dorsal Spine 85 34 12.0 Klein et al. (2017) Dorsal Spine 122 77 6.5 Current: Reader 1 Dorsal Spine 91 55 10.1 Current: Reader 2 Dorsal Ray 48 42 22.0 Nault et al. (2012) Dorsal Ra y 122 63 7.4 Current: Reader 1 Dorsal Ray 78 62 12.6 Current: Reader 2 Anal Ray 48 45 19.0 Nault et al. (2012) Anal Ray 124 61 10.4 Current: Reader 1 Anal Ray 44 96 1.4 Current: Reader 2 Anal Spine 82 12 19.0 Klein et al. (2017) Anal Spine 49 67 10.9 Current: Reader 1
84 Table 4 2 Precision of age estimates ( between reader PA = percent agreement; CV = coefficient of variation ) for several comparable studies using fin structures from various species Fin Structure n PA (%) CV Study Species Dor sal Spine 63 31 13.2 Sotola et al. (2014) Micropterus salmoides Dorsal Spine 85 81 2.1 Klein et al. (2017) Micropterus salmoides Dorsal Spine 91 65 7.6 Current Study Micropterus salmoides Dorsal Spine 286 27 15.8 Morehouse et al. (2013) Micropterus sa lmoides Dorsal Spine 107 80 6.4 Niewinski and Ferreri (1999) Perca flavescens Dorsal Spine 47 55 22.3 Porta et al. (2018) Sander vitreus x Sander canadensis Pectoral Ray 287 23 21.7 Morehouse et al. (2013) Micropterus salmoides Pectoral Ray 62 51 14.6 Current Study Micropterus salmoides Dorsal Ray 110 59 9.3 Debicella (2005) Mycteroperca microlepis Dorsal Ray 151 56 7.4 Herbst and Marsden (2011) Coregonus clupeaformis Dorsal Ray 21 67 4.2 Murie et al. (2009) Epinephelus itajara Dorsal Ray 78 53 11. 9 Current Study Micropterus salmoides Pelvic Ray 740 87 3.4 Zymonas and McMahon (2009) Salvelinus confluentus Pelvic Ray 43 67 5.4 Current Study Micropterus salmoides Marginal Increment Analysis and Temporal Synchronicity T he translucent bands on age 4 LMB dorsal spine sections must have initiated in November 2016. We first collected LMB dorsal spines in February 2017 and they were aged as 4 0s from February April 2017, whereas t he LMB sagitta otoliths were aged as 3 4s in February 2017 and March 2017 from the same cohort of fish The 4 th opaque b and was laid down in April 2017 for otoliths, where they were assigned an age of 4 0. This indicated that the enumerated bands in the dorsal spine sections and the sagitta otolith sections were not synchronized by month. The opaque zone in otolith sections denoted slow growth (Debicella 2005) but was lai d down quickly Four rings and translucent growth of 1 was seen in most sagitta samples only a month after the opaque band was completed, whereas t h e translucent zone in sagittae wa s visible in various increments ( i.e., 1 4 margin codes) throughout the months May 2017 January 2018 ( i.e., the final month of collection). The t ranslucent zone in dorsal spine sections was visible on the edge from February 2017 April 2 017 ( i.e., 4 0 fish), and then again in t he months N ovember
85 2017 January 2018. For six months of the sampling period the translucent band was o n the edge of most dorsal spine sections T emporal synchronicity of translucent growth bands was recorded in th e months of February, March, and November January for dorsal spine and otolith sections, and the deposition of the opaque ring (i.e., fast growth) in spine sections and (i.e., slow growth) in sagitta otolith sections showed some overlap in April, May, and June. When assigning LMB to year classes using dorsal spine sections the counted translucent band is visible before the enumerated (i.e., opaque band) is in otol ith sections Dorsal spines collected in the early months of November April will already show the otoliths will not show the th translucent band on the edge of their dorsal spine sections but were rounded down into the previous year class ( i.e., age 4) because we used a January 1 birthdate The otolith and dorsal spine were not laying down their enumerated annuli in the same months (~5 months apart), but they were partially synchronous in laying their opaque and transl uc ent growth zones Without the MIA, this w oul d have not been recognized and w ould have affected fish placement into age classes. Finding the core within dorsal spine section s was one of the first steps in the MIA ; however, it was difficult to locate a sta ndard measure ment point for each sample c onsidering th e core was often partly eroded particularly for older specimens This has been noticed in other studies, and if an axis of measurement were standardized for each fin structure (e.g., locating the centra l portion of the core) accuracy and precision of measurements could be enhanced. This can be difficult when sections are not equal ly sized or shaped (Zymonas and McMahon 2009 ) and could have affected the integrity of the measurements taken. Dorsal s pines III V each grew
86 differen tly so every spine within a section had variable banding pattern s for measuring increments. This was difficult to see without using 100x power and/or the QImaging system Agencies should also consider the type and price of equipmen t necessary for aging fi n structures and conducting analyses on them as it is higher than for otoliths Dorsal spine growth may be linked to somatic growth of LMB ( K. Nault, K Johnson, and B. Eisenhauer, Florida Fish and Wildlife Conservation Commission, and J. Kerns and coworke rs, University of Florida, unpublished data ) considering the counted ring was laid before that of the sagitta otolith, and the opaque band was being laid as fast growth during the summer and fall months. The slow growth tran slucent band was being laid in t he winter and wa s apparent on the edge in the previously noted months. Otolith sections had a short window of time that the o paque (i.e., slow growth) band wa s deposited, whereas they were laying down a fast growth band the res t of the year. This was diffe rent from dorsal spine sections and might suggest that LMB do r sal spine band deposition mimics that of body growth whereas otoliths had continual deposition throughout the year similar to metabolic growth ( Debicella 200 5 ; D. Parky n, University of Florida, personal communication ; T. Tuten, Florida Fish and Wildlife Conservation Commission, personal communication ) Future Work and Non lethal Aging At a future date 20 LMB with clipped fins and 10 LMB with no fins clipped ( i.e., contr ols) of various s izes ( i.e., 200 550 mm MTL ) will be held at FW C Bass Conservation Center. The f ish will be acclimated for three weeks to standardize any health issues resulting from sho cking, handling, or transport T his should allow researchers to be mostly cer tain that health problems or mortality have occurred due to clipped fins/experimental treatments and not collection. The f ish will be in the same reproductive state ( i.e., pre sp awn), and 5 8 fish will be held per tank depending on their s ize Fish will be grouped in tanks based on
87 size to avoid predation of smaller fish by larger individuals. LMB dorsal spines will be clipped following acclimation, and survival after excision will be assessed. Specimens will be held for 30 days, and the wo unds will be evaluated. The h atchery staff will monitor tanks daily to see if fish have perished If mortality occurs, the length, weight, and treatment group (i.e., clipped or control) will be recorded This acute survival study may provide further justif icat ion for utilizing dorsal spines as non lethal aging structures for LMB. More large fish ( n = 25) need to be sampled for their dorsal spine ages. We can then add c, and other precision and accuracy measu res to see how they change. Larger fish s hould be the ultimate focus now that dorsal spine growth for a wide range of sizes is better understood. The sample in this study included > 25 fish that were equal to or greater than 500 mm MTL, but even more fish shou ld be aged and compared to their respectiv e otolith ages for further validation of this technique. It is important to increase the aged subsample of trophy bass (i.e., > 3.63 kg) because they are the main focus of this conservation application Trophy bass contribute many eggs to th eir respective water bodies, and these large individuals are rarely encountered during standard field sampling (Allen et al. 2002; Dutterer et al. 2014). Both factors emphasize the need to evaluate the fin structure aging tec hniq ue and its lethality for eliminating mortality of trophy sized LMB during age and growth assessments. Management Implications To manage different populations of LMB in Florida, other lakes will need to be assessed using the dorsal spine method To test the validity of this method, i t should be applied to slow and other fast growing LMB populations Different lakes have variable average lengths of the oldest fish in their populations ( L ; Allen et al. 2002) and the rate at which this length is attained ( K ) also differs ( E. Camp, University of Florida, personal communication ) F in spine technique s are not universal across all water bodies and the best aging structures can differ even among
88 populations (Zymonas and McMahon 2009); thus, the method ology wil l ne ed to be adjusted for LMB populations with different growth rates For instance, between reader accuracy ( PA ) for Walleye dorsal spines was 80%, 55%, and 59%, depending on the lake sampled (Erickson 1983). This may explain the necessity for cl ipping th e dorsal spines of smaller fish in some lakes that are slow growing, instead of strictly implementing a n established size range of LMB that structures will be taken from in every lake. Size ranges of fish that particular fin structures or otoliths are take n from should be lake specific based on the age of the fish in the lake a s well as LMB growth rates Applications to s mall populations and continuous collection of fin structures for age data when a water body is closed for exploitation might be applied to other Micropterus spp. in Florida, such as Micropterus notius Suwannee Bass or Micropterus punctulatus Choctaw Bass. Most scientists and managers aim to de velop sustainable management practices ; in this case, aging dorsal spines may also w ork particular ly well for the conservation of rare, endemic centrarchids (Quist et al. 2012; King et al. 2018) Non lethal ly obtained a ge data from fin structures could also be used to better estimate growth mortality, and recruitment of trophy sized LMB However, i f incorrect ly characterized (e.g., low sample size; improper age estimates) these parameters could result in inappropriate regulations or other unnecessary management actions As it stands, trophy LMB are rarely captured du ring field sampling, and the relea se rate by fishers is high, so numbers sacrificed would not be significantly altered with the implementati on of a non lethal fin structure aging method; how ever the value to stakeholders of those few giant LMB that are ca ptured is enormous in Florida. I mp lementing an aging technique that does not require the sacrifice of fish
89 for age data allows anglers and scientists to work together and form an age database that helps them make informed decisions about their fishery ( Por ta et al. 2018 ). TrophyCatch rewa rds bass anglers for catching and releasing LMB that are 3.63 kilograms (kg) via the submission of photos of the fish on a scale with the weight (g) visible (Dutterer et al. 2014). Some anglers provide additional information such as MTL (mm) and/or girth (mm) measurements. The monitoring program relies on angl ers for both collection and reporting of trophy bass data (Dutterer et al. 2014). LMB are extremely valuable sp ort fish, and many can obtain trophy status ( 3.63 kg), especially in Florida (Allen et al. 2002). Clipping/excising dorsal spines could be inco rporated into TrophyCatch submissions, similar to the already existing fin clip kit for genetics or implemented during weigh ins at tournaments. When an angler takes a length, weight, girth measurement, and a picture, they could also cut dorsal spines and send them to a local FWC office for aging. The same type of scenario could be used during other no kill activit ies when larger numbers of bigger fish are available relative to field sampling. Stakeholders could then p romote the con servation of trophy LMB b y collecting and submitting non lethal ly sampled fin structures for aging while researchers are tag ging and rel easing trophy fish in the field These methods would g ive oth ers a chance to catch big LMB know how old they are, and the fish could be recaptur e d for growth and other pertinent studies The development of these potent ially non lethal methods would reduce the necessity for killing LMB to obtain viable population age structure information, and provide a mechanism for determining fish ages, especia lly for long lived specimens captured during no kill activities (e.g., fish ing tournaments) and c itizen science programs like FW C Biologists could greatly increase the amount of age data for large fish via angler submitted fin clips. This would help researchers better understand LMB population dynamics and fill in the data that are
90 lacking for populations around Florida. Ultimately, age data from dorsal spines or other valid LMB fin rays and spines could help fis heries managers furthe r inform and evaluate stock assessments and/or management actions (e.g., stocking or habitat restoration). If removing dorsal spines proves to be non lethal and the validation efforts are continued across other lakes, sizes, and ages of LMB, the technique could be taug ht to fisheries biologists and citizens who handle trophy bass frequently, and it could provide an avenue for determining ages of LMB (particularly big fish) ; thus collecting more age data without sacrificing more and larger fish Ultimately, the use of m ultiple different fin structures and sagitta otoliths could be implemented once a LMB population has been fully assessed (i.e., young fish aged via otoliths; spines or other structures aged for old fish).
91 LIST OF REFERENCES Allen, M. S., and J E. Hightowe r. 2010. Pages 43 79. Fish population dynamics: mortality, growth, and recruitment. Inland fisheries management in North America, 3rd edition. American Fisheries Society, Bethesda, Maryland. Allen, M. S., W. Scheaffer, W. F. Porak, and S. Cra wford. 2002. Growth and mortality of Largemouth Bass in Florida waters: implications for use of length limits. Pages 559 566 in D. P. Philipp and M. S. Ridgway, editors. Black bass: ecology, conservation, and management. American Fisheries Society, Symposi um 31, Bethes da, Maryland. Allman, R. J., C. L. Fioramonti, W. F. Patterson III, and A. E Pacicco. 2016. Validation of annual growth zone formation in Gray Triggerfish Balistes capriscus dorsal spines, fin rays, and vertebrae. Gulf of Mexico Science 33(1 ) article 6:6 8 76. Anderson, R. O., and S. J Gutreuter. 1983. Length, weight, and associated structural indices. Pages 283 300 in L. A. Nielsen and D. L. Johnson, editors. Fisheries t echniques, 2 nd edition American Fisheries Society, Bethesda, Maryland. Bailey, R. M., and C. L. Hubbs. 1949 The black basses ( Micropterus ) of Florida, with description of a new species. University of Michigan Museum of Zoology Occasional Papers 516:1 40. Barthel, B. L., M. S. Allen, W. F. Porak and J. Kerns. 2015. Florid a Bass Micropterus floridanus (LeSueur, 1822). Pages 43 53 in M. D. Tringali, J. M. Long, T. W. Birdsong, and M. S. Allen, editors. Black bass diversity: multidisciplinary science for conservation. American Fisheries Society, Sy mposium 82, Bethesda, Maryla nd. Beamish, R. J. 1981. Use of fin ray sections t o age Walleye Pollock, Pacific c od, and albacore, and the importance of this method. Transactions of the American Fisheries Society 110 : 287 299. Beamish, R. J., and D. Chilton. 1977. Age determination of Lingcod ( Ophiodon elongatus ) using dorsal fin rays and scales. Journal of the Fisheries Research Board of Canada 34:1305 1313. Beamish, R. J., and D. A. Fournier. 1981. A method for comparing the precision of a set of age deter minations. Canadian Journal of Fisheries and Aquatic Sciences 38:982 983. Beamish, R. J., and H. H Harvey. 1969 Age determination s in the White S ucker. Journal of the Fisheries Research Board of Canada 26 : 633 638. Beamish, R. J., and G. A. McFarlane. 1983. The forgotten requiremen t for age validation in fisheries biology. Transactions of the American Fisheries Society 112:735 743. Boschung, H. T., and R. L. Mayden. 2004. Fishes of Alabama. Smithsonian Institution Press, Washington D.C.
92 Buckm eier, D. L., and R. G. Howells. 2003. Val idation of otoliths for estimating ages of Largemouth Bass to 16 years. North American Journal of Fisheries Management 23:590 593. Burnet, A. M. R. 1969. An examination of the use of scales and fin rays for age det ermination of Brown Trout ( Salmo trutta L .). New Zealand Journal of Marine and Freshwater Research 3:147 151. Bwanika, G. N., D. J. Murie, and L. J. Chapman. 2007. Comparative age and growth of Nile Tilapia ( Oreochromis niloticus L.) in lakes Nabugabo and Wamala, Uganda. Hydrobiologia 589:287 301 Campana, S. E. 2001. Accuracy, precision and quality control in age determination, including a review of the use and abuse of age validation methods. Journal of Fish Biology 59: 197 242. Campana, S. E., M. C. Annand, and J. I. McMillan. 1995 Graphical a nd statistical methods for determining the c onsistency of age determinations Transactions of the American Fisheries Society 124 : 131 138. Cass, A. J., and R. J. Beamish. 1983. First evidence of validity of the fin ray method of age determination for marine fishes. North American Journal of Fisherie s Management 3:182 188. Casselman, J. M. 1983. Age and growth assessment of fish from their calcified structures techniques and tools. NOAA (National Oceanic and Atmospheric Administration) Technical Report NMFS ( National Marine Fisheries Service) 8:1 17. Chang, W. Y. 1982 A statistical method for evaluating the reproducibility of age determination. Canadian Journal of Fisheries and Aquatic Sciences 39 : 1208 1210. Chilton, D. E., and R. J. Beamish. 1982. Age dete rmination methods for fishes studied by the groundfish program at the Pacific Biological Station. Canadian Special Publication of Fisheries and Aquatic Sciences 60. Collins, M. R., and T. I. J. Smith. 1996. Sturgeon fin ray removal is nondeleterious. Nort h American Journal of Fisheries Management 16:939 941. Crawford, S., W. S. Coleman, and W. F. Porak. 1989. Time of annulus formation in otoliths of Florida Largemouth Bass. North American Journal of Fisheries Management 9:231 233. Debicella, J. M. 2005. A ccuracy and precision of fin ray aging for gag ( Mycteroperca microlepis ). Durn, I., M. Mar Beffa, J. A. Santamara J. Becerra, and L. Santos Ruiz. 2011. Actinotrichia collagens and their role in fin formation. Developmental Biology 354:160 172.
93 Dutil, J. D., and G. Power. 1977 Validity of reading otoliths compared to reading scales for the determination of the age of Brook T rout ( Salvelinus fontinalis ). Naturaliste Canadien (Quebec) 104 :361 368 Dutka G ianelli, J., and D. J. Murie. 1999 Age and growth of Sheepshead, Archosargus probatocephalus (Pisces: Sparidae), from th e northwestern coast of Florida. Bulletin of Marine Science 68:69 83. Dutterer, A. C., C. Wiley, B. Wattendorf J. R. Dotson, and W. F. Pouder. 2014. TrophyCatch: a conservation program for trophy bass in Florida. Florida Scientist 77:167 183 Erickson, C M. 1983. Age determinations of Manitoban W alleyes using otoliths, dorsal spines, and scales. North American Journal of Fisheries Management 3:176 181. Geffen, A. J. 1992. Validation of otolith increment deposition rate. Canadian Special Publication of Fi sheries and Aquatic Sciences 117:101 113. Go ss, R. J., and M. W. Stagg. 1957. The regeneration of fins and fin rays in Fundulus heteroclitus Journal of Experimental Zoology 136:487 507. Herbst, S. J., and J. E. Marsden. 2011. Comparison of precision and b ias of scale, fin ray, and otolith age estim ates for Lake Whitefish ( Coregonus clupeaformis ) in Lake Champlain. Journal of Great Lakes Research 37:386 389. Hoyer, M. V., J. V. Shireman, and M. J. Maceina. 1985. Use of otoliths to determine age and growth of Largemouth Bass in Florida. Transactions of the American Fisheries Society 114:307 309. Huish, M. T. 1954. Life history of Black C rappie s of Lake George, Florida. Transactions of the American Fisheries Society 83 : 176 193. Ingram, G. W. Jr. 2001. Stock structure of Gray Triggerfish, Ballistes cap riscus on multiple spatial scales in the Gulf of Mexico. Docto ral dissertation, University of South Alabama, Mobile. Isermann, D. A., M. H. Wolter, and J. J. Breeggemann. 2010. Estimating Black Crappie age: an assessment of dorsal spines and scales as no nlethal alternatives to otoliths. North American Journal of Fisheries Management 30:1591 1598. Jackson, Z. J., M. C. Quist, J. G. Larscheid, E. C. Thelen, and M. J. Hawkins. 2007. Precision of scales and dorsal spines for estimating age of Common Carp. Jo urnal of Freshwater Ecology 22:231 239. Jearld, A. Jr. 1983. Age determination. Pages 301 324 in L. A. Nielsen and D. L. Johnson, editors. Fisheries t echniques, 2nd edition American Fisheries Society Bethesda, Maryland.
94 Kasumyan, A. O. 2004. The vestib ular system and sense of equilibrium in fish. Journal of Ichthyology 44:S224 S268. King, S. M., S. R. David, and J. A. Stein. 2018. Relative bias and precision of age estimates among calcified structures of Spotted Gar, Shortnose Gar, and Longnose Gar. Tr ansactions of the American Fisheries Society DOI: 10.1002/tafs.10042. Klein, Z. B., T. F Bonvechio, B. R. Bowen, and M. C. Quist. 2017. Precision and accuracy of age estimates obtained from anal fin spin es, dorsal fin spines, and sagittal otoliths for kno wn age Largemouth Bass. Southeastern Naturalist 16:225 234. Koch, J. D., and M. C. Quist. 2007. A technique for preparing fin rays and spines for age and growth analysis. North American Journal of Fisheri es Management 27:782 784. Koch, J., B. Neely, and B. Sowards. 2017. Precision of three structures for Saugeye age estimation. North American Journal of Fisheries Management 38:31 38. DOI: 10.1002/nafm.10019 Koch, J. D., W. J. Schreck, and M. C. Quist. 200 8. Standardised removal and sectioning locations for Shovelnose Sturgeon fin rays. Fisheries Management and Ecology 15:139 145. Lin, L. I. 1989. A concordance correlation coefficient to evaluate reproducibility. Biometrics 45: 255 268. Lin, L., A. S. Hedayat, B. Sinha, and M. Yang. 2002. Statistical method s in assessing agreement: models, issues and tools. Journal of the American Statistical Association 97:257 270. Lin, L., A. S. Hedayat, and W. Wu. 2007. A unified approach for assessing agreement for continuous and categorical data. Journal of Biopharmaceu tical Statistics 17:629 652. Logsdon, D. E. 2007. Use of unsectioned dorsal spines for estimating Walleye ages. North American Journal of Fisheries Management 27:1112 1118. Long, J. M., and D. R. Stewart. 2010. Verification of otolith identity used by fis heries scientists for aging Channel Catfish. Transactions of American Fisher ies Society 139:1775 1779. Maceina, M. J., J. Boxrucker, D. L. Buckmeier R. S. Gangl D. O. Lucchesi D. A. Isermann, J. R. Jackson, and P. J. Martinez. 2007 Current status and r eview of freshwater fish aging procedures used by state and provincial fishe ries agencies with recommendations for future directions. Fisheries 32 : 329 340. Maraldo, D. C., and H. R. MacCrimmon. 1979 Comparison of agein g methods and growth rates for Large mouth B ass, Micropterus salmoides Lac p de from northern latitudes. Environmental Biology of Fishes 4: 263 271. Miller, S. J., and T. Storck. 198 2 Daily growth rings in otoliths of young of the year Largemouth B ass. Transactions of the American Fisheries Society 111 : 527 530.
95 Mills, K. H., and R. J. Beamish. 1980 Comparison of fin ray an d scale age determinations for Lake W hitefish ( Coregonus clupeaformis ) and their implications for estim ates of growth and annual survival. Canadian Journal of Fisheri es and Aquatic Sciences 37: 534 544. Morehouse, R. L., S. B. Donabauer, and A. C. Grier. 2013. Estimating Largemouth Bass age: precision and comparisons among scales, pectoral fin rays, and dor sal fin spines as nonlethal methods. Fisheries and Aquaculture Journal 74:1 7. Moyle, P. B., and J. J. Cech Jr 2004. Form an d movement. Pages 11 35. Fishes: an introduction to ichthyology, 5th ed ition Pearson Education, Inc., Upper Saddle River, New Jer sey. Moyle, P. B., and J. J. Cech Jr. 2004. Perciformes: Snooks to Sn akeheads. Pages 377 403. Fishes: an intro duction to ichthyology, 5th ed ition Pearson Education, Inc., Upper Saddle River, New Jersey. Murie, D. J., and D. C. Parkyn. 2005. Age and grow th of White Grunt ( Haemulon plumieri ): a comparison of two populations along the west coast of Florida. Bulleti n of Marine Science 76:73 93. Murie, D. J., D. C. Parkyn, C. C. Koenig, F. C. Coleman, J. Schull, and S. Frias Torres. 2009. Evaluation of finray s as a non lethal ageing method for protected Goliath Grouper Epinephelus itajara Endangered Species Research 7:213 220. Nagid, E. J., T. Tuten, and K. G. Johnson. 2015. Effects of reservoir drawdowns and the expansion of hydrilla coverage on year class s trength of Largemouth Bass. North American Journal of Fisheries Management 35:54 61. Niewinski, B. C., and C. P. Ferreri. 1999. A comparison of three structures for estimating the age of Yellow Perch. North American Journal of Fisheries Management 19:872 877. Page, L. M., and B. M. Burr. 2011. Peterson field guide to freshwater fishes of North America north of Mexico, 2 nd edition. Houghton Mifflin Harcourt, Boston. Parr, R. T., R. B. Bringolf, and C. A. Jennings. 2018. Efficacy of otoliths and first dors al spines for preliminary age and growth determination in Atlantic Tripletails. Marine and Coastal Fisheries 10:71 79. Popper, A. N., and R. R. Fay. 1973. Sound detection and processing by teleost fishes: a critical review. The Journal of the Acoustical S ociety of America 53:1515 1529. Popper, A. N., and Z. Lu. 2000. Structure function relationships in fish oto lith organs. Fisheries Research 46:15 25. Popper A. N., J. Ramcharitar, and S. E. Campana. 2005. Why otoliths? Insights from inner ear physiology and fisheries biology. Marine and Freshwater Research 56:497 504.
96 Porta, M. J., R. A. Snow, and D. E. Shoup. 2018. Comparison of Saugeye age estimates and population characteristics using otoliths and dorsal spines. Journal of the Southeastern Association of Fish and Wildlife Agencies. 5:23 29. Quist M. C., M. A. Pegg, and D. R. DeVries. 2012. Age and growth. Pages 677 732 in A. V. Zale, D. L. Parrish, and T. M. Sutton, editors. Fisheries techniques, 3rd edition. American Fisheries Society, Bethesda, Mary land. Rude, N. P., W. D. Hintz, J. D. Norman, K. L. Kanczuzewski, A. J. Yung, K. D. Hofer, and G. W. Whitledge. 2013. Using pectoral fin rays as a non lethal aging structure for Smallmouth Bass: precision with otolith age estimates and the importance of r eader experience. Journal of Freshwater Ecology 28:199 210. Schneid ervin, R. W., and W. A. Hubert. 1986. A rapid technique for otolith removal from salmonids and catostomids. North American Journal of Fisheries Management 6:287. Scott, W. B., and E. J. Cro ssman. 1973. Freshwater fishes of Canada. Fisheries Research Board of Canada Bulletin 184. Secor, D. H., J. M., Dean, and E. H. Laban. 1992. Otolith removal and preparation for microstructural examination. Pages 64 71 in D. K. Stevenson and S. E. Campana, editors. Otolith microstructure examination and analysis. Canadian Special Publication of Fisheries and Aquatic Sciences 117. Sikstrom, C. B. 1983. Otolith, pectoral fin ray, and scale age deter minations for Arctic Grayling. P rogressive Fish Culturist 45: 220 223. Snow, R. A., M. J. Porta, and D. R. Stewart. 2018. Otolith annulus validation and variables influencing false annuli formation in dorsal spines of Saugeye. Journal of Applied Ichthyology 34:1153 1159. Sotola, V. A., G. A. Maynard, E. M. Hayes Pon tius, T. B. Mihuc, M. H. Malchoff, and J. E. Marsden. 2014. Precisi on and bias of using opercles as compared to otoliths, dorsal spines, and scales to estimate ages of Largemouth and Smallmouth Bass. Northeastern Naturalist 21:565 573. Strickland, P. A., a nd C. R. Middaugh. 2015. Validation of annulus formation in Spotted Sucker otoliths. Journal of Fish and Wildlife Management 6:208 212. Sylvester, R. M., and C. R. Berry. 2006. Comparison of White Sucker age estimates from scales, pectoral fin rays, and ot oliths. North American Journal of Fisheries Management 26:24 31. T aubert, B. D., and J. A. Tranquilli. 1982. Verification of the formation of annuli in otoliths of Largemouth Bass. Transactions of the American Fisheries Society 111:531 534. Tesch, F. W. 1 971. Age and growth. Pages 98 130 in W. E. Ricker, editor. Methods for assessment of fish production in freshwaters. Blackwell Scientific Publications, London.
97 VanderKooy. S. J. 2009 A practical handbook for determining the age s of Gulf of Mexico fishes Gulf Stat es Marine Fisheries Council, Oceans Spring, Mississippi. Zymonas N. D., and T. E. McMahon. 2009. Comparison of pelvic fin rays, scales, and otoliths for estimating age and growth of Bull Trout, Salvelinus confluentus Fisheries Management and E cology 16:155 164.
98 BIOGRAPHICAL SKETC H Summer Lindelien grew up in Minnesota and attended the University of Wisconsin Superior in Superior, WI, where she received her Bachelor of Science in biology. Summer worked as an aquatic ecology field and laborator y technician at Natural Resources Research Institute (NRRI) in Duluth, MN, as a research technician with Lake Superior Research Institute (LSRI) in Superior, WI, and as a fisheries biologist I for the Florida Fish and Wildl ife Conservation Commission (FW C ) in Gainesville, FL, prior to becoming a graduate research assistant with the University of Florida Fisheries and Aquatic Sciences program.