<%BANNER%>

Relations between Hydrological Variables and Year-Class Strength of Sportfish in Eight Florida Waterbodies


PAGE 1

RELATIONS BETWEEN HYDROLOGICA L VARIABLES AND YEAR-CLASS STRENGTH OF SPORTFISH IN EIGHT FLORIDA WATERBODIES By TIMOTHY F. BONVECHIO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2003

PAGE 2

Copyright 2003 by Timothy F. Bonvechio

PAGE 3

This document is dedicated to to my parents, Robert and Constance.

PAGE 4

ACKNOWLEDGMENTS This thesis would not have been completed without the hard work and dedication of many people. I thank R. Burnes, R. Cailteux, P. Cooney, R. Crook, K. Dockendorf, G. Kaufman, J. Krummrich, S. Larsen, C. Mwatela, G. Ripo, M. Rogers, T. Thompson, N. Trippel, K. Tugend and B. Whitfield for providing help with the field sampling. I thank J. Berg, R. Crook, P. Cooney, K. Dockendorf, M. Duncan, D. Dutterer, K. Henry, K. Jacoby, G. Kaufman, S. Larsen, C. Mwatela, M. Rogers, T. Thompson, K. Tugend, T. Tuten, N. Trippel, B. Whitfield, and J. Wingate for the help in the lab processing and preparing any presentation. I thank J. Donze, M. Hoyer, R. Varner, D. Watson, and D. Willis for their help in generating bathymetric maps. I thank S. Crawford and E. Nagid for their help in aging problematic sunfish species. I thank J. Hill and C. Cichra for their MFL literature review. I thank my committee members, M. Allen, C. Cichra, J. Estes, and D. Murie, for the many helpful comments they offered throughout this study. I especially thank my committee chair, M. Allen, for giving me the chance to fish and work under him, and hopefully, absorb some of his knowledge. This research was funded by the Wildlife Foundation of Florida, Inc. Finally, I thank my fiance, Kimberly Tugend, for dealing with me during this study. I thank my parents, Robert and Constance Bonvechio, for teaching me the right way to get the job done in life and giving me the confidence to tackle any task in life. iv

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES............................................................................................................vii LIST OF FIGURES.........................................................................................................viii ABSTRACT.......................................................................................................................ix INTRODUCTION...............................................................................................................1 METHODS..........................................................................................................................4 Site Selection................................................................................................................4 Species Selected............................................................................................................5 Data Collection and Analyses.......................................................................................5 RESULTS..........................................................................................................................13 Hydrological Fluctuations..........................................................................................13 Sampling Effort and Age Distribution........................................................................13 Verification Analyses.................................................................................................14 System-Specific Correlation Analyses of Residuals with Hydrological Variables....16 Lake Annie..........................................................................................................16 Lake Bonny.........................................................................................................16 Crooked Lake......................................................................................................17 Lake Disston........................................................................................................17 Ochlockonee River..............................................................................................18 Santa Fe River.....................................................................................................19 Withlacoochee River North.................................................................................19 Withlacoochee River South.................................................................................19 Multiple Regression Analysis of Combined Residuals..............................................20 Lakes....................................................................................................................20 Rivers...................................................................................................................21 Correlation Analysis of Combined Recruitment Variability Indexes.........................22 DISCUSSION....................................................................................................................43 v

PAGE 6

MANAGEMENT IMPLICATIONS FOR SETTING MINIMUM FLOWS AND LEVELS.....................................................................................................................53 APPENDIX A INADEQUATE CATCH CURVES...........................................................................56 B AGE FREQUENCIES................................................................................................57 LIST OF REFERENCES...................................................................................................59 BIOGRAPHICAL SKETCH.............................................................................................68 vi

PAGE 7

LIST OF TABLES Table page 1. Characteristics of lakes and rivers sampled.............................................................30 2. Historical lake elevation (m) means, the coefficient of variation (CV = SD/*100%), number of water level observations (N), and ranges of each year for the four study lakes.............................................................................................31 3. Historical river flow rate (m 3 /sec) means, coefficient of variation (CV = SD/*100%), number of flow rate observations (N), and ranges of each year for the four study rivers..................................................................................................33 4. Waterbody, species, season and year of sampling...................................................35 5. Description of otolith reader agreement across system, species, season, and year...........................................................................................................................37 6. Catch-curve linear regression equations for each waterbody and species with r 2 P, ages, and N...........................................................................................................39 7. Significant correlation relationships between measures of bluegill, black crappie, largemouth bass, and redear sunfish year-class strength and stage for four lakes...40 8. Significant correlations between measures of bluegill, largemouth bass, redbreast sunfish, and Suwannee bass year-class strength and flow rate for the four rivers.........................................................................................................................41 9. Significant multiple regression equations predicting year-class strength (i.e., residual) from flow rates for the four rivers.............................................................42 vii

PAGE 8

LIST OF FIGURES Figure page 1. Location of eight sample sites..................................................................................23 2. Determination of year-class strength displaying the hypothetical catch-curve (a) and residual plot (b)..................................................................................................24 3. Florida LAKEWATCH personnel generated this map of Lake Annie by using differentially corrected global positioning equipment (GPS)..................................25 4. Florida LAKEWATCH personnel generated this map of Lake Bonny by using differentially corrected global positioning equipment (GPS)..................................26 5. Florida LAKEWATCH personnel generated this map of Crooked Lake by using differentially corrected global positioning equipment (GPS)..................................27 6. Florida LAKEWATCH personnel generated this map of Lake Disston by using differentially corrected global positioning equipment (GPS)..................................28 7. Hypsographic curves displaying the change in surface area with hypothetical declines in water level (m) for each of the four lakes..............................................29 viii

PAGE 9

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science RELATIONS BETWEEN HYDROLOGICAL VARIABLES AND YEAR-CLASS STRENGTH OF SPORTFISH IN EIGHT FLORIDA WATERBODIES By Timothy F. Bonvechio December 2003 Chair: Micheal S. Allen Major Department: Fisheries and Aquatic Sciences Hydrological variables, such as water level (m) and flow rate (m 3 /s), have influenced recruitment of sportfishes in lakes, reservoirs, and rivers. I evaluated how annual and seasonal flows and water levels were related to year-class strength of selected sportfishes across four rivers and four lakes in Florida. Species investigated included black crappie Pomoxis nigromaculatus, bluegill Lepomis macrochirus, largemouth bass Micropterus salmoides, redbreast sunfish, L. auritus, redear sunfish L. microlophus, and Suwannee bass M. notius. Hydrological data were obtained from the U.S. Geological Survey and the appropriate Florida Water Management Districts. The residuals from catch-curve regressions developed from otoliths of fish caught by electrofishing were used to assess relationships between hydrological variables and year-class strength of sportfish. The multiple regression model computed for the Suwannee bass residuals combined from two rivers indicated that year-class strength was negatively related to spring median flow rates (r 2 = 0.38). Similarly, largemouth bass ix

PAGE 10

residuals combined from four rivers indicated that year-class strength was negatively related to spring median flow rates and positively related to winter median flow rates (r 2 = 0.24). Conversely, bluegill residuals combined from two rivers indicated that year-class strength was positively related to the previous fall median flow rates (r 2 = 0.38). Redbreast sunfish residuals combined from three rivers indicated that year-class strength was positively related to the previous fall median flow rates before spawning (r 2 = 0.31). On a broader scale, Micropterus spp. (i.e., notius and salmoides) residuals were combined from four rivers and a multiple regression model indicated that year-class strength was negatively related to spring median flow rates and positively related to winter median flow rates (r 2 = 0.22). Lepomis spp. (i.e., auritus and macrochirus) residuals were combined from three rivers and the multiple regression equation indicated that year-class strength was positively related to fall median flow rates before spawning and negatively related to fall median flow rates (r 2 = 0.47). No significant regression models were detected between water level in lakes and year-class strength of sportfish. Management implications of this work include regulation changes pertaining to minimum flows and levels (MFLs). Detecting impacts of flow on year-class strength of sportfish across lakes and rivers were variable but relationships were easier to detect in rivers. High flows, at least once every three years in the fall, may allow inundation of floodplain habitat, thus providing favorable environmental conditions for Lepomis spp. in rivers. Setting MFLs during periods of prolonged drought (i.e., three years or more) should consider impacts to short-lived species such as Lepomis spp. I contend that low flows for three or more consecutive years should be prevented and thus, MFLs should consider biological impacts to short-lived fishes. x

PAGE 11

INTRODUCTION Water demand due to human population expansion in central Florida has increased ground-water removal, resulting in reduced ground water and surface water levels, particularly during periods of low rainfall. Reduced surface-water levels may have consequences for important sport fisheries in rivers and lakes. Previous studies have shown that low water levels may reduce habitat availability and substantially alter fish and invertebrate communities (Travnichek et al. 1995; Petts 1996). Recruitment can be defined as the number of fish born in a given year that survive to reproductive or harvestable size (Willis and Murphy 1996). Recruitment into a fishery can be influenced by a number of density-dependent and independent factors (Everhart and Youngs 1981; Royce 1996). Possible density-dependent factors include cannibalism, disease, and predation (Houde 1987). Density-independent factors may include variations in temperature, turbidity, flow rate, or water level changes (Everhart and Youngs 1981; Sigler and Sigler 1990; Royce 1996). Increased water level may increase the amount of spawning habitat and food resources available to sportfish by inundating shoreline vegetation (Jenkins 1970; Aggus and Elliot 1975; Keith 1975; Timmons et al. 1980; Miranda et al. 1984; Meals and Miranda 1991). Alternately, above average flow rates in lotic systems have both positively and negatively influenced recruitment of fish, depending on the system and species (Filipek et al. 1991; Mason et al. 1991; Sallee et al. 1991; Kriksunov and Mamina 1995; Raibley et al. 1997). Thus, variations in flows and water levels may influence fish recruitment. 1

PAGE 12

2 The extent of drought and flood periods will influence the magnitude, frequency, duration, and timing of changes in water flow (Grossman et al. 1990; Reice et al. 1990), which can influence fish community structure in rivers (Poff and Ward 1989; Walker et al. 1995; Richter et al. 1996; Poff et al. 1997). Stream flow is considered a major variable that affects the abundance and distribution of many riverine species (Resh et al. 1988; Power et al. 1995). Catch-curve analysis has been used by fishery biologists to estimate total annual mortality (Ricker 1975). Residuals from catch curves can be used as an index of relative year-class strength, with positive residuals indicating relatively strong year classes and negative residuals indicating relatively weak year classes (Maceina and Bettoli 1998). Residuals can then be used to relate year-class strength to environmental variables (Wrenn et al. 1996; Maceina 1997). Wrenn et al. (1996) used a catch-curve multiple regression with environmental variables and determined that reservoir retention time over 16 days produced strong year classes of largemouth bass Micropterus salmoides. This technique was also used to relate recruitment of largemouth bass to flow conditions in two Alabama reservoirs (Maceina 1997). In addition, Sammons et al. (2002) used residuals to index crappie recruitment in several Tennessee reservoirs. Minimum flows and levels (MFLs) for Florida lakes and rivers are set by the appropriate Florida Water Management District governing board to prevent significant ecological harm to waterbodies as a result of permitted water withdrawals (SJRWMD 2001). The state legislature requires the establishment of MFLs under Subsection 373.042(2), Florida Statutes (F.S.) (SJRWMD 2001). Based on previous evaluations of topography, soils, historical ground and surface water data, and vegetation data, MFLs

PAGE 13

3 are mandated to prevent harm to that ecological system (SJRWMD 2001). The MFLs are also reviewed periodically by the appropriate water management district and the Florida Department of Environmental Protection (SJRWMD 2001). However, criteria for establishing MFLs in Florida have not included biological factors such as ecological harm to fish recruitment. Knowledge of potential hydrological impacts on sportfishes could be used when MFLs are established and reviewed. I evaluated the influence of water level and flow rate on year-class strength of six important sportfish species. My objectives were to (1) estimate the age frequency of selected sportfish species across eight Florida waterbodies; (2) use residuals around a catch curve to index relative year-class strength for each species and (3) relate relative year-class strength of each species to historic water-level fluctuations within each waterbody and among waterbodies for lakes and rivers.

PAGE 14

METHODS Site Selection Waterbodies were chosen for sampling based on two requirements: 1) historical stream-flow (m 3 /sec) or water-level (m) data dating back to at least 1992, which experienced significant changes in the flow or water level; and 2) historically low levels of aquatic vegetation. Water-level data back to 1992 were needed to have a minimum of nine years included for analyses with residuals. Aquatic vegetation can influence the amount of habitat that is gained or lost due to fluctuations in water level. High percentages of aquatic vegetation in waterbodies influence fish communities (Shireman et al. 1985; Smith and Orth 1990; Bettoli et al. 1993; Hoyer and Canfield 1996, Tate et al. 2003); thus fluctuating water levels may have a small impact on the amount of available habitat when abundant macrophytes are present. To reduce the confounding effect of aquatic plants on water level effects, I selected lakes that had historically low plant coverage (i.e., <20% coverage). Electrofishing transects were 20 minutes and included samples from all available habitat types. All sampling took place in the fall (September) through spring (March) due to potential sampling biases in the summer (Pope and Willis 1996). Lakes Annie, Bonny, Crooked, and Disston were chosen because of their fluctuating water levels and low to moderate levels of aquatic vegetation (Figure 1). Lake characteristics are shown in Table 1. I performed electrofishing at Lake Annie from January to February 2002 and January 2003, at Lake Bonny in November 2001 and from 4

PAGE 15

5 October to November 2002, at Crooked Lake from January to February 2002 and January 2003, and at Lake Disston in March 2002 and February 2003. The four rivers targeted for sampling were the Santa Fe, Ochlockonee, Withlacoochee North, and Withlacoochee South (Figure 1, Table 1). I performed electrofishing in the middle Santa Fe River in October 2001 and September 2002, at the upper Ochlockonee River in November 2001 and from September to October 2002, at the middle Withlacoochee River South in March 2002 and from February to March 2003, and at the middle Withlacoochee River North from November to December 2002. Species Selected The species chosen varied across systems but included black crappie Pomoxis nigromaculatus, bluegill Lepomis macrochirus, largemouth bass, redbreast sunfish L. auritus, redear sunfish L. microlophus, and Suwannee bass M. notius. All are sought by anglers as sportfishes. The species chosen for collection differed among lakes and rivers based on presence and abundance. Suwannee bass and redbreast sunfish are usually present only in rivers, whereas bluegill, black crappie, largemouth bass, and redear sunfish are present in both lakes and rivers. Data Collection and Analyses Three boats equipped for electrofishing were used in 2001 and 2002. I used two, 4.9-m jon boats with either a 50-hp or 60-hp outboard and a 4.3-m jon boat with a 15-hp outboard. Electrofishing equipment consisted of the same generator (5000-W AC) and either a Coffelt model VVP 15 or Smithroot model VI-A pulsator and similar cathode probes with electrical output ranging from 5-8 amps of DC current. All large individuals of each species were kept for age analyses due to possible growth differences between male and females (e.g., largemouth bass; Carlander 1977).

PAGE 16

6 The number of fish to be sacrificed was a concern, so all species were subsampled in smaller size groups where gender-specific growth was expected to be similar (Bettoli and Miranda 2001). Five largemouth bass per centimeter group < 35 cm total length (TL) and all largemouth bass > 35 cm TL were kept. Five Suwannee bass per centimeter group < 25 cm TL and all Suwannee bass > 25 cm TL were kept. For black crappie, bluegill, redear, and redbreast sunfish, five fish per centimeter group < 15 cm TL and all fish > 15 cm TL were kept unless 20 fish were obtained per cm group > 15 cm TL. Fish were placed on ice and returned to the laboratory where they were measured and weighed, gender determined, and sagittal otoliths removed. Age and growth for all species was determined from annuli on the sagittal otoliths (Schramm and Doerzbacher 1982; Taubert and Tranquilli 1982; Hoyer et al. 1985; Maceina and Betsill 1987; Crawford et al. 1989; Schramm 1989; Hales and Belk 1992; Mantini et al. 1992; Buckmeier and Howells 2003). The annuli on otoliths from all fish were counted by observing the whole otolith under a microscope. Otoliths from fish > age-1 were mounted on frosted glass microscope slides with super glue and then sectioned transversely along the dorsoventral plane into 0.5-mm sections using a South Bay Tech. Inc., Model 650, diamond wheel saw. Sections were mounted on glass microscope slides using Thermo Shandon synthetic mountant cement and the annuli were counted under a dissecting microscope. The total number of rings or annuli were recorded. The date of collection was considered when determining the age of the otolith (Hoyer et al. 1985; Crawford et al. 1989). Ages were determined by number of rings alone or the number of rings +1 for those fish caught before or after January 1 st respectively. Two independent readers were used, and in the instance a discrepancy

PAGE 17

7 occurred between the two readings, a third reader was used to resolve the age or the otolith was discarded. To remain consistent and potentially increase precision, the same three readers were used for the duration of the study. An age-length key (Ricker 1975) was used to estimate the age frequency of subsampled fish below the specified length, and in some cases above the specified length (i.e., if > 20 fish/cm group were collected above the specified length). The fraction of fish in the aged subsample (N = 5 fish/cm group) was extrapolated to the unaged fish below the specified length to assign fish age based on length. The age frequency of each species of fish below and above the specified length was combined to estimate the total number of fish at each age. Sampling biases due to short-term weather events, flow conditions, or water level could cause a biased age structure. To assess the potential for sampling biases within a single year, I verified the age structure of fishes in three rivers and four lakes in the second sampling year. Fish were collected and aged as described above, and the age structure of each species was compared to data from the previous year for each waterbody. The Withlacoochee River North was not sampled in the second year. Bathymetric maps were generated for lakes Annie, Bonny, Crooked, and Disston by Florida LAKEWATCH (2000, 2001, and 2003) personnel using differentially corrected global positioning equipment (GPS). A planimeter (Keuffel and Esser Co.) was used to trace the shoreline contour of each lake in planimeter units. The planimeter units were then compared to the scale of each map and converted into the actual surface area for each lake. Surface areas of different depths in each lake were plotted on hypsographic curves. A hypsographic curve indicates the change in lake surface area per

PAGE 18

8 change in water level. The hypsographic curves were generated to assess potential influence of water level on lake surface area and thus potential changes in inshore habitat availability. Using the estimated age-frequency for each species and system, I performed catch-curve analyses: log e (NUMBER) = b 0 b 1 (AGE) (1) where NUMBER is the estimated total number of fish of age x and AGE is fish age in years. Because sample size varied among species and waterbodies, the estimated total number of fish of each age was standardized into percent frequency at each age prior to the log e transformation. Catch-curve residuals (i.e., observed deviation from expected values from the regression line) display variation in recruitment among years, and therefore index relative-year class strength (Maceina 1997; Figure 2). Maceina (1997) defined strong and weak year classes as residuals greater than 0.50 and less than -0.50, respectively. I expected positive residuals to correspond to strong year classes and negative residuals to correspond to weak year classes. The magnitude of positive and negative residuals affect the amount of variation explained (r 2 ) in a simple linear regression model such as a catch curve. For example, I would expect relatively constant recruitment if the r 2 is high (i.e., > 0.95). Variability in recruitment of black crappie has been indexed with the recruitment variability index (RVI) (Guy and Willis 1995). The RVI was calculated as described by Guy and Willis (1995): RVI = [S N /(N m + N p )] N m /N p (2) where S N is the summation of the cumulative relative frequency distribution based on the number of fish in each age-class, N m is the number of age-groups missing from the

PAGE 19

9 sample that should be present (which does not include ages past the last age obtained), N p is the number of age-groups present in the sample, and N p > N m The RVI ranges from to 1. Recruitment is more stable as RVI increases from to 1. All assumptions described by Guy and Willis (1995) were met in order for estimates to be obtained. Thus, the recruitment coefficient of determination (RCD = r 2 ), RVI, and residual analysis are three options for assessing recruitment patterns of infrequently sampled populations (Isermann et al. 2002). I used all three indices to assess variability of recruitment for each species. The RVI has previously not been applied to sportfish populations other than black crappie. Capture efficiency of boat electrofishing is size selective (Bayley and Austen 2002). Bayley and Austen (2002) determined catchability of largemouth bass and bluegill across a range of fish sizes. Based on their estimates, I used minimum catchable size groups to determine what age groups to include in the catch curves. Because electrofishing captures small fish at lower rates than their true abundance (Anderson 1995; Reynolds 1996), a catchable size was determined for each species across systems. Lepomis spp. and Pomoxis spp. were assigned a 100 mm-TL catchable size and Micropterus spp. were assigned a 200 mm-TL catchable size. Bayley and Austen (2002) found that fish above these sizes did not vary greatly in catchability. I assumed that Lepomis spp. and Pomoxis spp. catchabilities varied with fish size in a similar manner to bluegill in Bayley and Austen (2002) and that Suwannee bass catchability would be similar to largemouth bass. If at least 60% of the fish in an age group were assigned catchable size, that year class was kept in the catch curve. If less than 60% of the fish were above catchable size or there were less than three fish in the age group (Ricker

PAGE 20

10 1975), then the year class was deleted from the catch curve. Due to variable growth rates across systems and species, the age range of catchable-sized fish used in the catch curves fluctuated across systems and species. A sign test (Conover 1980) was performed on the residuals using the UNIVARIATE procedure (SAS 2000) to verify that similar strong and weak year classes in a subsequent year did occur and were not the result of sampling biases within a single year. Residuals for each year class sampled in consecutive years were scored as a + if both were above or below the regression line, which indicated agreement between the two sample years. Conversely, if a yearclass residual was above the line in year one but below the line in year two, or vice versa, the year-class was scored as a -. The sign test assessed whether + scores were significantly higher than - scores, which would indicate that residuals coincided between years across systems and species ( = 0.10). To perform the sign test, the species were grouped according to rivers and lakes, and separate sign tests were performed on the two groups. I selected the best catch-curve regression for each system and year to relate hydrological data using three criteria. First, the catch curve was selected to contain ages obtained during years when extreme water level or flow fluctuations occurred (i.e., minimum or maximum flow rates). Second, the catch curve that contained the most ages that were fully recruited to the gear was used, given the first criteria was met. If both catch curves incorporated extreme environmental conditions and contained the same number of ages, then the sample with the larger number of fish was chosen. Correlation analyses was used to assess the relationship between residuals and environmental variables for the year the fish in each age were born for each system.

PAGE 21

11 The water level (m) and flow data (m 3 /sec) were separated into four periods: 1) January through March was considered winter: 2) April through June was considered spring; 3) July through September was considered summer; and 4) October through December was considered fall. Previous researchers (Maceina and Stimpert 1998; Sammons et al. 2002) have found hydrological conditions before spawning was related to crappie year-class strength, so the fall periods before the spawn were also evaluated in the correlation analyses for each sportfish species and were referred to as pre-spawn fall. Correlation analyses was performed on the residuals with the levels of minimum, maximum, and mean flow rates and/or stages for the annual, fall pre-spawn, winter, spring, summer, and fall periods. To assess trends across systems, catch-curve residuals were standardized as percent frequencies and hydrological data (i.e., stream flow, m 3 /sec, and water level, m) were also standardized as a percent of the median on a seasonal basis. Residuals were combined according to genus (e.g., Lepomis), species (e.g., macrochirus), and system type (i.e., river) for multiple regression with hydrological variables. I only obtained one black crappie catch curve on one lake, and thus black crappie were not included in the lake analyses of combined residuals. I used multiple linear regression models to determine if year-class strength (i.e., residuals) were related to the five seasonal variables (i.e., median flow or stage for all seasons) from the equation (Myers 1990). Y = 0 + 1 (x 1 ) + 2 (x 2 ) + + k (x k ) + (3) Where Y is the year-class strength (residual), 0 is the intercept value, k is parameter estimate, and x k is the independent variable. Independent variables evaluated for flow and stage included pre-fall, winter, spring, summer, and fall percent of median values. To compare across systems, median values were used in the multiple liner regression

PAGE 22

12 models due to the influence of extreme values on the mean. Backward selection was used and the procedure terminated when the removal of no other independent variables could significantly (P < 0.10) improve the overall model. Multiple linear regression was performed on residuals across systems to provide some insight on regional responses of fish year-class strength in relation to flow or stage. The multiple regression models also allowed for multiple independent variables to be evaluated concurrently. Multicollinearity is a common problem in multiple regression models, and I checked for this problem by evaluating the Variance Inflation Factor (VIF) (Myers 1990). Regression models with a VIF > 10 were not observed and as a result, multicollinearity was not considered a problem in the analyses. To assess whether variation in flow or stage was related to overall recruitment variability, RVI values were correlated with the overall coefficient of variation (CV=SD/.*100%) in flow or stage across systems for each species. All statistical analyses were conducted with SAS (2000) and statistical tests were considered significant when P < 0.10 and marginally significant relationships when P = 0.11. A P-value of 0.10 was chosen due to low power associated with most correlations (i.e., N = 5 age classes), and because I considered Type-II error to be important (Peterman 1990).

PAGE 23

RESULTS Hydrological Fluctuations The overall historical range of water levels across the four lakes fluctuated from < 1 to > 3 meters among years (Table 2). The overall range in water level was highest at Lake Crooked (3.8 m) and lowest at Lake Disston (1.2 m)(Table 2). For the four rivers, the overall historical range of flow rates varied from 0 to 1,062 m 3 /sec among years (Table 3). Overall flow ranges were highest in the Withlacoochee North (2.2,062 m 3 /sec) and lowest in the Withlacoochee South (0.9 m 3 /sec) (Table 3). The bathymetric maps of Lakes Annie, Bonny, Crooked and Disston were generated at 97, 99, 96, and 97% of the maximum water level recorded, respectively (Figures 3, 4, 5, and 6). Bathymetric maps display the outermost contour line representing the shoreline of each lake at the time of map construction. These outermost contour lines were subject to change depending on the lake levels. Hypsographic curves revealed the potential influence of water level on lake surface area (Figure 7). Sampling Effort and Age Distribution A combined total of 576 electrofishing transects were conducted on the eight waterbodies. The Florida Fish and Wildlife Conservation Commission (FFWCC) contributed an additional 16 electrofishing transects on the Ochlockonee River in 2002, increasing the total electrofishing transects to 592. Most transects (> 97%) were sampled for 20 minutes in both years, but transects other than 20 minutes (10 min) were 13

PAGE 24

14 included in the analyses. To obtain an adequate age sample, electrofishing effort differed according to the system, species, and year (Table 4). The number of fish collected and aged differed among systems, species and year (Table 4). A total of 19,301 sportfish were collected during the study: 8,298 bluegill; 5,558 redbreast sunfish; 3,675 largemouth bass; 733 Suwannee bass; 529 redear sunfish; and 508 black crappie (Table 4). For all populations combined, bluegill ranged from 22 to 263 mm TL, black crappie ranged from 109 to 350 mm TL, largemouth bass ranged from 48 to 635 mm TL, redbreast sunfish ranged from 33 to 244 mm TL, redear sunfish ranged from 56 to 312 mm TL, and Suwannee bass ranged from 67 to 406 mm TL (Table 4). A total of 6,327 sportfish were aged during the study, which included 1,918 bluegill, 463 black crappie, 2,238 largemouth bass, 802 redbreast sunfish, 434 redear sunfish, and 472 Suwannee bass (Table 4). For all samples combined, there was 96% between-reader agreement resulting from 6,063 reader agreements out of 6,327 otoliths (Table 5). The 264 otolith disagreements were read by a third independent reader. A total of five otoliths were thrown out from the 264 disagreements because the third independent reader determined an age that was different than either of the first two readers. For all systems and sample years combined, between-reader agreement among species ranged from 91 to 98%. Ages ranged from 0 to 10 years for bluegill, 0 to 9 for black crappie, 0 to 12 for largemouth bass, 0 to 7 for redbreast sunfish, 1 to 10 for redear sunfish, and 0 to 12 for Suwannee bass. Verification Analyses A total of 36 usable catch-curves, 20 from lakes and 16 from rivers, were generated on sportfish population age samples obtained from seven systems. Depending on the

PAGE 25

15 species and the system, some samples of aged fish were smaller than catchable size and could not be verified with the coinciding age sample. These under represented age samples were not included in the sign test. The sign test on the residuals from the lakes and rivers revealed that similar strong and weak year classes in subsequent years, N = 43, P = 0.05, and N = 32, P = 0.02, respectively. Residuals matched 65% (28/43) of the time for lakes and matched 72% (23/32) of the time for rivers. Thus, strong and weak year classes were evident in both years in most cases. Changes in the age frequencies caused year classes with low residuals to change from positive to negative or vice versa (4/15 lake residual disagreements and 2/11 river residual disagreements) but fell within 0.20 of matching with each other. As a result, only 26% and 22% of lake and river residual disagreements were extreme values (i.e., > 0.2), respectively. These arbitrary residuals that fluctuated around zero were left in the two separate sign tests but did not significantly influence the results. Nevertheless, strong and weak year-classes were generally consistent between sample years. The verification process determined which systems were adequately sampled for two years in a row and could then be included in the correlation and multiple regression analysis of residuals with the hydrological variables. A specific system and a species were included if at least 50% of the residuals matched in consecutive years (Table 6). I assumed samples that did not meet this criteria, were not related to water levels or flow rates and were not used in the analyses. Species samples that were not verified due to a small age sample during one year or the species was only sampled one time included Lake Annie bluegill, Lake Bonny largemouth bass, Crooked Lake redear sunfish, Lake Disston redear sunfish, and the Withlacoochee River North largemouth bass, redbreast

PAGE 26

16 sunfish, and Suwannee bass. These samples were used in the correlation and multiple regression analyses. System-Specific Correlation Analyses of Residuals with Hydrological Variables Lake Annie No significant relationships between stage and relative year-class strength were found for Lake Annie, and most of the variation (r 2 ) was explained in the catch-curve regressions (Table 6). Bluegill and largemouth bass residuals were not correlated with any annual or seasonal water levels on a minimum, maximum or mean level (all P > 0.27). Thus, relatively constant recruitment of bluegill and largemouth bass residuals were found on Lake Annie and year-class strength was not related to water level. Lake Bonny Year-class strength of sportfish was variable in relation to water levels at Lake Bonny. Black crappie residuals from the catch curve indicated variable recruitment was present (i.e., low r 2 value, Table 6). Black crappie residuals from the catch curve were negatively correlated with maximum pre-spawn fall water levels (Table 7). Thus, low maxium water levels prior to the spawn may be an indicator of potentially strong year classes of black crappie in the following year. Largemouth bass residuals from the catch curve revealed relatively constant recruitment from 1995-1998. Although most of the variation (r 2 ) was explained in the catch curve regression (Table 6), largemouth bass residuals from the catch curve were positively correlated with minimum, maximum, and mean fall water levels (all P < 0.06; Table 7). As a result, higher water levels may provide young-of-year largemouth bass. Black crappie and largemouth bass residuals were not correlated with any other annual or any other seasonal water levels (all P > 0.12). Bluegill residuals from the catch curve were not correlated with any annual

PAGE 27

17 or seasonal water level variables (all P > 0.15). Thus, year class strength of bluegill was not related to water level, but high water level in the fall was positively related to largemouth bass and negatively related to black crappie in the pre-spawn fall season. Crooked Lake No significant relationships were found at Crooked Lake except for largemouth bass. Largemouth bass residuals from the catch curve indicated variable recruitment was not present (i.e., high r 2 value, Table 6). But, largemouth bass residuals were positively correlated with minimum, maximum, and mean fall water levels (all P < 0.07; Table 7). Largemouth bass residuals were not correlated with any other annual or seasonal water level on a minimum, maximum or mean level (all P > 0.25). Bluegill and redear sunfish residuals from the catch curve regressions revealed relatively constant recruitment from 1995, and much of the variation (r 2 ) was explained (Table 6). Bluegill and redear sunfish residuals were not correlated with any annual or seasonal water level variables (all P > 0.19). Thus, year-class strength of bluegill and redear sunfish was not related to water level, but high water level in the fall was positively related to largemouth bass year-class strength. Lake Disston Relations between stage and residuals varied among species at Lake Disston. Most of the variation (r 2 ) was explained in the bluegill catch curve (Table 6), but residuals were negatively correlated with pre-spawn fall water levels (P < 0.07; Table 7). Most of the variation (r 2 ) was also explained in the largemouth bass catch curve (Table 6), but largemouth bass residuals were positively correlated with minimum annual water levels (Table 7). Most of the variation (r 2 ) was left unexplained in the redear sunfish catch curve (Table 6), and redear sunfish residuals were positively correlated with pre-spawn

PAGE 28

18 fall minimum and mean water levels (Table 7). Bluegill, largemouth bass, and redear sunfish residuals were not correlated with any other annual or seasonal water level variables (all P > 0.13). Results indicate an increase in water level during the fall before the spawn were positively related to year-class strength for redear sunfish and negatively related to for bluegill. Also, an increase in the minimum water level on an annual basis was positively related to largemouth bass year-class strength. Ochlockonee River The relationships between flow rates and year-class strength of sportfish at the Ochlockonee River were variable. Bluegill residuals from the catch curve were positively correlated with flow variables with the exception of pre-spawn minimum fall flow rates (all P < 0.07; Table 8). Variable bluegill relationships with flow rates may indicate the species ability to persist during multiple flow regimes. Largemouth bass residuals from the catch curve indicated variable recruitment was present (i.e., low r 2 value, Table 6), but no significant relationships were found between flow and relative largemouth bass year class strength in this system (all P > 0.12). Redbreast sunfish residuals from the catch curve were positively related to flow variables at this system (all P < 0.11; Table 8). Redbreast sunfish year-class strength was positively associated with flow in several seasons including winter, summer, and fall seasons. Thus, high flow favored strong redbreast sunfish year classes. No other sportfish residuals were significantly correlated with any other annual or seasonal flow rate variables (all P > 0.12). Results indicate strong year classes of bluegill and redbreast sunfish were associated with higher flow rates but no significant relationships were detected for largemouth bass. Thus, relationships were variable across species and seasons in the Ochlockonee River.

PAGE 29

19 Santa Fe River Conversely, black bass year-class strength was negatively correlated to flow in the Santa Fe River (both P < 0.09; Table 8). Both largemouth bass and Suwannee bass year-class strength were negatively related to flow, particularly during the spring (Table 8). No other black bass residuals were correlated with any other annual or seasonal flow rate variables (all P > 0.12). Strong year classes of largemouth bass and Suwannee bass were associated with low flow rates, indicating potential persistence of these species in the event of a water withdrawal from the river. Withlacoochee River North Similarly, largemouth bass residuals from the catch curve were negatively correlated to flow variables at the Withlacoochee River North (all P < 0.10; Table 8) for several seasons, suggesting that high flow reduced largemouth bass year-class strength (Table 8). Largemouth bass residuals were not correlated with any other annual or seasonal flow rate variables (all P > 0.12). Redbreast sunfish and Suwannee bass residuals were not correlated with any annual or seasonal flow rate variables (all P > 0.13). Thus, low flow appeared to be favorable for strong largemouth bass year-class strength in the Withlacoochee River North and the Santa Fe River. Withlacoochee River South Generally, sunfish (L. auritus and L. macrochirus) year-class strength from the Withlacoochee River South was positively correlated with flow rates and no relationships were detected for largemouth bass. Bluegill residuals from the catch curve, when significant, were positively related to flow variables (all P < 0.11; Table 8). Largemouth bass residuals were not correlated with any annual or seasonal flow rate variables (all P > 0.40). No other sportfish residuals were correlated with any other annual or seasonal

PAGE 30

20 flow rates (all P > 0.12). Thus, high flow favored strong bluegill and redbreast sunfish year classes in this system. Historical hydrological data for both lakes and rivers indicated high water levels and flow rates during 1997 and 1998 (i.e., high rainfall), whereas extremely low water levels and flow rates were observed in late 1999 through 2001 (i.e., drought conditions). The fluctuations in water level and/or flow rate data were related to year-class strength (i.e., residuals) among some of the systems. For example, largemouth bass from Crooked Lake displayed a relatively strong year-class in 1998 (i.e., high annual mean stage) (Table 2) and a weak year class in 2000 (i.e., lower annual mean stage) (Table 2), as indicated by positive and negative residuals (i.e., +0.58, -0.49). Also, largemouth bass from Lake Disston displayed a potential strong year-class in 1998 (i.e., high mean annual water level) (Table 2), as indicated by a positive residual (i.e., +0.55). Other positive residuals (i.e., strong year classes) > +0.40 in 1998 included Lake Bonny largemouth bass; Ochlockonee River bluegill, largemouth bass, and redbreast sunfish; Withlacoochee River North redbreast sunfish and the Withlacoochee River South bluegill. Thus, I found evidence that high water in 1998 caused relatively strong year classes. Multiple Regression Analysis of Combined Residuals Lakes Multiple regression models for each species combined across lakes revealed no significant relationships between seasonal stage values and year-class strength (all P > 0.13). The number of residuals included in the models ranged from ten for redear sunfish to 34 for Lepomis spp. combined. Thus, although within-lake relationships were found for several species at three of four lakes evaluated, I found no trends in year-class

PAGE 31

21 strength across the lakes using multiple regression, suggesting that regional trends were not evident. Rivers In general, multiple regression models combined across rivers indicated year-class strength of sunfish (i.e., redbreast sunfish and bluegill) was positively related to flow rates and black bass year-class strength was negatively related to flow rates (all P < 0.10; Table 9). The number of residuals included in the models ranged from eight for Suwannee bass to 32 for Micropterus spp. combined. The best multiple regression equation for bluegill indicated that year-class strength was positively related with pre-spawn fall median flow rates and the model explained 38% of the variation (Table 9). Similar to bluegill, the best model for redbreast sunfish indicated that year-class strength was positively related with pre-spawn fall median flow rates and the model explained 31% of the variation (Table 9). Conversely, the best model for largemouth bass indicated that year-class strength was negatively related with spring median flow rates and positively related to winter median flow rates, and the model explained 24% of the variation (Table 9). Similar to largemouth bass, the best Suwannee bass model indicated that year-class strength was negatively related with spring median flow rates and the model explained 38% of the variation (Table 9). The best multiple regression equation for sunfish (L. auritus and L. macrochirus) indicated that year-class strength was positively related to pre-spawn fall median flow rates and negatively related to fall median flow rates, and the model explained 47% of the variation (Table 9). The best model for black basses (M. salmoides and M. notius) indicated that year-class strength was negatively related to spring median flow rates and positively related to winter median flow rates and the model explained 22% of the

PAGE 32

22 variation (Table 9). Thus, regression equations revealed variable responses between sunfish and black bass populations and flow conditions across systems. Correlation Analysis of Combined Recruitment Variability Indexes Recruitment variability indexes were combined across system type (i.e., lakes or rivers) for each species. In order to perform a correlation analysis, a minimum of three data points were needed, so pooled RVI species samples that contained less than three data points were discarded from the analysis (i.e., black crappie were only obtained from one system). These discarded samples included black crappie and redear sunfish in lakes and bluegill and Suwannee bass in rivers. The coefficient of variation (CV) was obtained for the annual water level of each lake and the annual water level and flow rate from each river for correlation analysis with RVI. Combined bluegill RVIs in four lakes were not correlated with the CV of water level (P = 0.65). Combined largemouth bass RVIs in four lakes were not correlated with the CV of water level (P = 0.63). Combined largemouth bass RVIs in four rivers were not correlated with any flow rate (P = 0.41). Combined redbreast sunfish RVIs in three rivers were not correlated with any flow rate (P = 0.44). Thus, overall recruitment variation indexed with the RVI was not significantly related to overall flow or stage variation among these systems.

PAGE 33

23 Figure 1. Location of eight sample sites. Arrows are pointing to approximate sample locations. All locations were sampled for two consecutive years during the same season (e.g., fall or winter), except the Withlacoochee River North which was only sampled in one year.

PAGE 34

24 Figure 2. Determination of year-class strength displaying the hypothetical catch-curve (a) and the associated residual plot (b).

PAGE 35

25 Figure 3. Florida LAKEWATCH personnel generated this map of Lake Annie by using differentially corrected global positioning equipment (GPS). Data were collected March 20, 2003. Scale and map contours are in meters and were generated using kriging technique in Surfer software package (Golden, CO). The center of the lake is located at Latitude 27 59 32 and Longitude 81 36 25. On this date, the lake surface area was calculated at 180 hectares.

PAGE 36

26 Figure 4. Florida LAKEWATCH personnel generated this map of Lake Bonny by using differentially corrected global positioning equipment (GPS). Data were collected March 20, 2003. Scale and map contours are in meters and were generated using kriging technique in Surfer software package (Golden, CO). The center of the lake is located at Latitude 28 02 16 and Longitude 81 55 29. On this date, the lake surface area was calculated at 108 hectares.

PAGE 37

27 Figure 5. Florida LAKEWATCH personnel generated this map of Crooked Lake by using differentially corrected global positioning equipment (GPS). Data were collected August 8, 2001. Scale and map contours are in meters and were generated using kriging technique in Surfer software package (Golden, CO). The center of the lake is located at Latitude 27 48 27 and Longitude 81 34 42. On this date, the lake surface area was calculated at 1356 hectares.

PAGE 38

28 Figure 6. Florida LAKEWATCH personnel generated this map of Lake Disston by using differentially corrected global positioning equipment (GPS). Data were collected July 1, 1999. Scale and map contours are in meters and were generated using kriging technique in Surfer software package (Golden, CO). The center of the lake is located at Latitude 29 17 2" and Longitude 81 23 31". On this date, the lake surface area was calculated at 1060 hectares.

PAGE 39

29 Figure 7. Hypsographic curves displaying the change in surface area with hypothetical declines in water level (m) for each of the four lakes. Florida LAKEWATCH bathymetric maps were traced using a planimeter to construct the curves.

PAGE 40

30 Table 1. Characteristics of lakes and rivers sampled. Surface area (ha) was obtained from Florida LAKEWATCH (2000, 2001, and 2003). Trophic state was calculated according to criteria established by Forsberg and Ryding (1980). Species collected are abbreviated as follows: blc = black crappie; bg = bluegill; lmb = largemouth bass; rbsu = redbreast sunfish; resu = redear sunfish; swb = Suwannee bass. Water data sources are abbreviated as follows: SWFWMD = Southwest Florida Water Management District; SJRWMD = St. Johns River Water Management District. Sample location displays the furthest North and South GPS points sampled on the river. U.S.G.S. Gauge # indicates the United States Geological Surveys water monitoring station. Lake County Surface Area (ha) Trophic State Species Collected Water Data Source Annie Polk 180 Eutrophic bg, lmb, resu SWFWMD Bonny Polk 108 Hypereutrophic blc, bg, lmb SWFWMD Crooked Polk 1,356 Mesotrophic blc, bg, lmb, resu SWFWMD Disston Flagler 1,060 Mesotrophic bg, lmb, resu SJRWMD Sample Location River County North South Species Collected U.S.G.S. Gauge # Ochlockonee Gadsden/Leon 30.59N, 84.36W 30.49N, 84.40W bg, lmb, rbsu, swb 02339000 Santa Fe Alachua/Columbia/ Gilchrist 29.83N, 82.65W 29.85N, 82.63W lmb, rbsu, swb 02321975 Withlacoochee (North) Hamilton/Madison 30.62N, 83.27W 30.50N, 83.24W lmb, rbsu, swb 02319000 Withlacoochee (South) Citrus/Marion 29.01N, 82.39W 28.98N, 82.33W bg, lmb, rbsu 02313000

PAGE 41

31 Table 2. Historical lake elevation (m) means, the coefficient of variation (CV = SD/*100%), number of water level observations (N), and ranges of each year for the four study lakes. Lake Year N Mean Elevation (m) CV Overall Range (m) Annie 1990 11 33.2 0.28 33.033.3 1991 9 33.3 0.51 33.133.5 1992 0 . 1993 12 34.2 0.19 34.134.3 1994 12 34.4 0.69 34.134.8 1995 12 35.0 0.84 34.735.4 1996 12 35.5 0.17 35.335.5 1997 12 35.1 0.27 35.035.3 1998 12 35.5 0.44 35.335.8 1999 12 35.0 0.37 34.835.2 2000 12 34.3 0.77 34.034.8 2001 10 33.7 0.77 33.434.1 Bonny 1988 4 40.0 0.13 40.040.1 1989 12 39.6 0.58 39.339.9 1990 12 39.3 0.63 38.839.6 1991 12 39.1 1.14 38.639.7 1992 12 39.6 0.76 39.240.0 1993 11 40.0 0.35 39.740.1 1994 12 39.8 0.57 39.540.1 1995 12 39.9 0.53 39.640.3 1996 12 39.7 0.23 39.639.9 1997 12 39.6 0.91 39.140.2 1998 12 40.0 0.67 39.640.4 1999 12 39.6 0.46 39.239.8 2000 12 38.8 0.93 38.4 39.5 2001 10 38.1 1.14 37.538.9 Crooked 1990 12 32.7 0.35 32.632.9 1991 12 32.8 1.21 32.433.3 1992 12 33.5 1.53 32.934.1 1993 12 34.1 0.36 33.934.3 1994 12 34.1 1.04 33.734.8 1995 12 35.0 1.06 34.535.5 1996 12 35.3 0.27 35.235.5 1997 12 35.1 0.50 34.835.4 1998 12 35.9 0.42 35.636.2 1999 12 35.8 0.70 35.436.2 2000 12 35.6 0.68 35.336.1 2001 10 34.9 0.67 34.635.2

PAGE 42

32 Table 2. Continued. Lake Year N Mean Elevation (m) CV Overall Range (m) Disston 1992 271 4.03 2.96 3.744.31 1993 307 3.95 3.40 3.724.24 1994 330 4.06 3.66 3.764.55 1995 330 3.98 4.56 3.674.43 1996 321 4.08 2.55 3.874.32 1997 273 4.06 2.76 3.814.40 1998 273 3.97 5.56 3.584.36 1999 272 3.88 4.70 3.604.23 2000 293 3.87 3.56 3.654.14 2001 291 4.01 5.60 3.684.93 2002 47 4.05 0.53 4.014.08

PAGE 43

33 Table 3. Historical river flow rate (m 3 /sec) means, coefficient of variation (CV = SD/*100%), number of flow rate observations (N), and ranges of each year for the four study rivers. River Year N Mean Flow Rate (m 3 /sec) CV Overall Range (m 3 /sec) Ochlockonee 1990 365 18.7 162 0.57151 1991 365 58.9 142 1.67634 1992 366 28.9 113 2.61229 1993 365 30.1 169 0.62289 1994 365 65.5 129 4.79770 1995 365 25.1 146 1.67286 1996 366 23.6 111 2.55145 1997 365 32.7 104 1.39174 1998 365 41.5 158 0.57504 1999 365 10.5 102 1.2744.2 2000 366 7.00 111 0.5439.4 2001 365 17.8 142 1.22174 2002 273 8.50 130 0.9659.8 Santa Fe 1992 92 50.3 103 18.8239 1993 365 17.2 78.2 2.8049.8 1994 365 16.6 66.9 4.1362.6 1995 365 15.9 49.6 4.9843.6 1996 366 17.6 123 5.52195 1997 365 18.5 66.7 6.5487.2 1998 365 37.8 112 14.7259 1999 365 8.22 48.1 3.7418.7 2000 366 2.95 48.2 1.169.06 2001 365 1.27 115 0.027.53 2002 15 0.14 21.6 0.100.19 Withlacoochee 1990 365 29.3 150 2.80188 North 1991 365 124 150 4.561062 1992 366 49.4 126 4.76388 1993 365 54.1 176 2.38538 1994 365 99.6 93.7 8.24530 1995 365 47.6 173 3.74535 1996 366 30.0 139 3.74238 1997 365 71.3 108 3.94351 1998 365 89.4 152 4.59974 1999 365 13.5 118 2.1580.4 2000 366 16.8 133 2.58154 2001 273 42.7 132 3.51283

PAGE 44

34 Table 3. Continued. River Year N Mean Flow Rate (m 3 /sec) CV Overall Range (m 3 /sec) Withlacoochee 1990 365 8.90 39.4 3.7021.9 South 1991 365 15.0 85.2 3.7049.6 1992 366 6.50 71.2 1.7037.1 1993 365 10.3 37.8 5.6021.4 1994 365 21.1 75.3 4.6062.0 1995 365 31.6 77.9 8.3097.1 1996 366 24.8 39.9 11.644.7 1997 365 11.6 104 3.1084.1 1998 365 54.6 87.5 9.40150 1999 365 10.3 41.6 4.2027.8 2000 366 3.10 49.2 1.107.30 2001 273 4.10 96.8 0.9024.9

PAGE 45

35 Table 4. Waterbody, species, season and year of sampling. Number of fish aged, total fish collected (Total), electrofishing effort in minutes (Effort), and minimum total length collected (Min TL, mm), and maximum total length collected (Max TL, mm). Waterbody Species Season Year Age d Total Effort Min TL Max TL Lake Annie Bluegill Winter 2002 73 73 400 35 207 2003 5 5 84 360 54 200 Largemouth Bass 2002 23 2 376 400 90 585 2003 17 6 370 360 105 635 Redear Sunfish 2002 12 5 151 400 85 312 2003 8 2 141 360 65 234 Lake Bonny Black Crappie Fall 2001 22 7 254 940 111 350 2002 18 8 206 810 109 346 Bluegill 2001 23 6 1,721 940 26 263 2002 20 0 2,722 810 22 226 Largemouth Bass 2001 7 2 72 940 236 567 2002 71 82 810 142 579 Crooked Lake Black Crappie Winter 2002 2 7 27 520 180 341 2003 21 21 580 131 253 Bluegill 2002 153 499 520 37 249 2003 14 2 198 580 45 253 Largemouth Bass 2002 211 408 520 106 635 2003 133 221 580 107 562 Redear Sunfish 2002 3 0 30 520 56 223 2003 11 5 125 580 57 272 Lake Disston Bluegill Winter 2002 21 5 573 820 38 263 2003 233 317 900 49 256 Largemouth Bass 2002 133 169 820 148 589 2003 10 7 108 900 153 596 Redear Sunfish 2002 5 2 52 820 72 244 2003 3 0 30 900 119 246 Ochlockonee Bluegill Fall 2001 20 9 688 760 37 218 River 2002 15 0 914 1,228 66 242 Largemouth Bass 2001 13 7 206 760 75 597 2002 18 2 335 1,228 57 556 Redbreast Sunfish 2001 16 9 923 760 48 228 2002 12 8 2,653 1,228 33 222 Suwannee Bass 2001 2 4 24 760 80 406 2002 3 7 38 1,228 67 392 Santa Fe River Largemouth Bass Fall 2001 14 9 254 810 74 535 2002 16 5 256 710 48 501 Redbreast Sunfish 2001 6 9 403 810 62 244 2002 5 7 162 710 90 190 Suwannee Bass 2001 151 275 810 75 383 2002 17 5 305 710 72 383

PAGE 46

36 Table 4. Continued. Waterbody Species Season Year Age d Total Effort MinTL Max TL Withlacoochee Largemouth Bass Fall 2002 13 7 203 1,220 52 588 River North Redbreast Sunfish 2002 19 8 1,061 420 45 241 Suwannee Bass 2002 8 5 91 1,220 67 386 Withlacoochee Bluegill Winter 2002 143 353 620 52 230 River South 2003 10 9 156 997 66 242 Largemouth Bass 2002 15 8 350 620 109 546 2003 17 5 265 997 97 629 Redbreast Sunfish 2002 10 2 252 620 57 212 2003 7 9 97 997 40 201 Sum 6,32 7 19,301

PAGE 47

37 Table 5. Description of otolith reader agreement across system, species, season, and year. Discrepancies indicate the number of discrepancies between two readers, and % Agreement indicates the percent agreement between two readers for each sample. Waterbody Species Season Year Number Aged Discrepancies % Agreement Lake Annie Bluegill Winter 2002 73 0 100 2003 55 2 96 Largemouth Bass 2002 232 0 100 2003 176 1 99 Redear Sunfish 2002 125 1 99 2003 82 9 89 Lake Bonny Black Crappie Fall 2001 227 30 87 2002 188 11 94 Bluegill 2001 236 5 98 2002 200 17 92 Largemouth Bass 2001 72 4 94 2002 71 3 96 Crooked Lake Black Crappie Winter 2002 27 0 100 2003 21 1 95 Bluegill 2002 153 5 97 2003 142 8 94 Largemouth Bass 2002 211 0 100 2003 133 0 100 Redear Sunfish 2002 30 0 100 2003 115 11 90 Lake Disston Bluegill Winter 2002 215 20 91 2003 233 7 97 Largemouth Bass 2002 133 7 95 2003 107 0 100 Redear Sunfish 2002 52 3 94 2003 30 1 97 Ochlockonee Bluegill Fall 2001 209 2 99 River 2002 150 5 97 Largemouth Bass 2001 137 6 96 2002 182 1 99 Redbreast Sunfish 2001 169 1 99 2002 128 4 97 Suwannee Bass 2001 24 0 100 2002 37 1 97 Santa Fe River Largemouth Bass Fall 2001 149 18 88 2002 165 6 96 Redbreast Sunfish 2001 69 4 94 2002 57 9 84 Suwannee Bass 2001 151 6 96 2002 175 15 91

PAGE 48

38 Table 5. Continued. Waterbody Species Season Year Number Aged Discrepancies % Agreement Withlacoochee River North Largemouth Bass Fall 2002 137 0 100 Redbreast Sunfish 2002 198 20 90 Suwannee Bass 2002 85 0 100 Withlacoochee Bluegill Winter 2002 143 4 97 River South 2003 109 3 97 Largemouth Bass 2002 158 4 98 2003 175 0 100 Redbreast Sunfish 2002 102 8 92 2003 79 1 99 Sum 6,327 264 Mean 129 5.4 96

PAGE 49

39 Table 6. Catch-curve linear regression equations for each waterbody and species with r 2 P, ages, and N. Ages is the number of fish ages used in each catch curve analysis used to generate residuals for correlation analyses with hydrological variables. N is the total number of fish collected by electrofishing used to generate residuals. Waterbody Species Year Equation r 2 P Ages N Lake Annie Bluegill 2002 Log (% Freq) = 4.39 0.36 age 0.94 0.03 4 66 Largemouth Bass 2002 Log (% Freq) = 5.36 0.70 age 0.97 0.00 6 270 Lake Bonny Black Crappie 2001 Log (% Freq) = 5.15 0.66 age 0.79 0.02 6 222 Bluegill 2001 Log (% Freq) = 4.97 0.84 age 0.82 0.01 6 790 Largemouth Bass 2001 Log (% Freq) = 5.74 0.61 age 0.98 0.01 4 47 Crooked Lake Bluegill 2003 Log (% Freq) = 4.39 0.42 age 0.91 0.00 7 144 Largemouth Bass 2002 Log (% Freq) = 5.17 0.65 age 0.88 0.01 6 314 Redear Sunfish 2003 Log (% Freq) = 4.07 0.32 age 0.83 0.01 7 111 Lake Disston Bluegill 2002 Log (% Freq) = 5.32 0.72 age 0.91 0.00 7 371 Largemouth Bass 2002 Log (% Freq) = 4.63 0.48 age 0.87 0.01 6 113 Redear Sunfish 2002 Log (% Freq) = 2.94 0.07 age 0.20 0.56 4 39 Ochlockonee Bluegill 2001 Log (% Freq) = 6.39 1.10 age 0.88 0.06 4 225 River Largemouth Bass 2001 Log (% Freq) = 4.58 0.46 age 0.79 0.04 5 59 Redbreast Sunfish 2001 Log (% Freq) = 7.15 1.40 age 0.86 0.07 4 434 Santa Fe River Largemouth Bass 2001 Log (% Freq) = 4.68 0.52 age 0.83 0.03 5 166 Suwannee Bass 2001 Log (% Freq) = 4.79 0.55 age 0.91 0.01 5 193 Withlacoochee Largemouth Bass 2002 Log (% Freq) = 3.44 0.16 age 0.50 0.12 6 46 River North Redbreast Sunfish 2002 Log (% Freq) = 6.53 1.14 age 0.94 0.04 5 613 Suwannee Bass 2002 Log (% Freq) = 4.37 0.31 age 0.46 0.52 3 37 Withlacoochee Bluegill 2003 Log (% Freq) = 5.48 0.74 age 0.95 0.01 5 125 River South Largemouth Bass 2003 Log (% Freq) = 4.06 0.33 age 0.79 0.01 7 137 Redbreast Sunfish 2002 Log (% Freq) = 7.13 1.38 age 0.95 0.14 3 211

PAGE 50

40 Table 7. Significant correlation relationships between measures of bluegill, black crappie, largemouth bass, and redear sunfish year-class strength and stage for four lakes. Minimum, maximum, and mean stages were measured for the annual, pre-spawn fall, fall, winter, spring, and summer periods. N indicates the number of residuals obtained from the catch curve. R indicates whether the relationship was positively or negatively correlated. r 2 indicates the percent variation explained in the model. No significant relationships were detected for Lake Annie. Water Body Species Water Variable Period Level N R P r 2 Bonny Black Crappie Stage Pre-Spawn Fall Max 6 -0.74 0.09 0.47 Largemouth Bass Stage Fall Min 4 0.94 0.06 0.88 Largemouth Bass Stage Fall Max 4 0.99 0.01 0.98 Largemouth Bass Stage Fall Mean 4 0.99 0.01 0.99 Crooked Largemouth Bass Stage Fall Min 6 0.81 0.05 0.08 Largemouth Bass Stage Fall Max 6 0.77 0.07 0.23 Largemouth Bass Stage Fall Mean 6 0.79 0.06 0.17 Disston Bluegill Stage Pre-Spawn Fall Min 7 -0.72 0.07 0.51 Bluegill Stage Pre-Spawn Fall Max 7 -0.88 0.01 0.77 Bluegill Stage Pre-Spawn Fall Mean 7 -0.75 0.05 0.56 Largemouth Bass Stage Annual Min 6 0.74 0.10 0.54 Redear Sunfish Stage Pre-Spawn Fall Mean 4 0.98 0.02 0.96 Redear Sunfish Stage Pre-Spawn Fall Min 4 0.99 0.02 0.97

PAGE 51

41 Table 8. Significant correlations between measures of bluegill, largemouth bass, redbreast sunfish, and Suwannee bass year-class strength and flow rate for the four rivers. Levels of minimum, maximum, and mean flow rates were measured for the annual, pre-spawn fall, fall, winter, spring, and summer periods. N indicates the number of residuals obtained from the catch curve. R indicates whether the correlation was positive or negative. P* indicates a marginally significant relationship. Waterbody Species Water Variable Period Level N R P r 2 Ochlockonee Bluegill Flow Pre-Spawn Fall Min 4 -0.93 0.07 0.87 River Bluegill Flow Summer Max 4 0.99 0.02 0.97 Bluegill Flow Fall Min 4 0.94 0.07 0.88 Redbreast Sunfish Flow Annual Max 4 0.99 0.00 0.99 Redbreast Sunfish Flow Annual Mean 4 0.89 0.11* 0.79 Redbreast Sunfish Flow Winter Max 4 0.99 0.00 0.99 Redbreast Sunfish Flow Winter Min 4 0.95 0.05 0.91 Redbreast Sunfish Flow Winter Mean 4 0.99 0.01 0.98 Redbreast Sunfish Flow Summer Min 4 0.92 0.08 0.85 Redbreast Sunfish Flow Summer Max 4 0.92 0.08 0.85 Redbreast Sunfish Flow Pre-Spawn Fall Max 4 0.98 0.02 0.96 Santa Fe River Largemouth Bass Flow Spring Max 5 -0.94 0.02 0.89 Largemouth Bass Flow Spring Mean 5 -0.84 0.08 0.71 Suwannee Bass Flow Annual Mean 5 -0.83 0.08 0.69 Suwannee Bass Flow Spring Min 5 -0.82 0.09 0.68 Suwannee Bass Flow Spring Max 5 -0.86 0.06 0.75 Suwannee Bass Flow Spring Mean 5 -0.96 0.01 0.92 Withlacoochee Largemouth Bass Flow Annual Min 6 -0.75 0.09 0.56 River North Largemouth Bass Flow Fall Max 6 -0.88 0.02 0.78 Largemouth Bass Flow Fall Mean 6 -0.78 0.07 0.61 Largemouth Bass Flow Summer Mean 6 -0.83 0.04 0.69 Largemouth Bass Flow Summer Min 6 -0.74 0.10 0.55 Withlacoochee Bluegill Flow Spring Max 5 0.82 0.09 0.67 River South Bluegill Flow Winter Min 5 0.79 0.11* 0.63 Bluegill Flow Winter Mean 5 0.80 0.11* 0.64 Redbreast Sunfish Flow Pre-Spawn Fall Min 3 0.99 0.03 0.99

PAGE 52

42 Table 9. Significant multiple regression equations predicting year-class strength (i.e., residual) from flow rates for the four rivers. Model r 2 DF P-value Bluegill residual = -0.2142 + 0.0009 pre-spawn fall median flow 0.38 7 0.10 Largemouth bass residual = 0.2438 0.0028 spring median flow + 0.0009 winter median flow 0.24 21 0.08 Redbreast sunfish residual = -0.1967 + 0.0006 pre-spawn fall median flow 0.31 11 0.06 Suwannee bass residual = 0.3656 0.0032 spring median flow 0.38 7 0.10 Lepomis spp. residual = 0.0570 + 0.0006 pre-spawn fall median flow 0.0004 fall median flow 0.47 19 0.00 Micropterus spp. residual = 0.2502 0.0028 spring median flow + 0.0008 winter median flow 0.22 29 0.03

PAGE 53

DISCUSSION In general, year-class strength was more strongly related to system hydrology in rivers than in lakes. Changes in water flow and stage influence fish community structure in rivers (Poff and Ward 1989; Grossman et al. 1990; Walker et al. 1995; Richter et al. 1996; Poff et al. 1997). According to Resh et al. (1988) and Power et al. (1995), stream flow is considered a major variable that affects the abundance and distribution of many riverine species. Variation in flow/stage was much higher for rivers than for lakes in this study, and I found stronger relationships between year-class strength and flow/stage in rivers than in lakes. Thus, rapid and extreme changes in flow/stage in rivers may more strongly influence fish populations than the relatively minor and less variable stage changes in lakes. Relationships between year-class strength and hydrology in rivers appear to be stronger and easier to detect than in lakes. Results indicated that black bass year-class strength was often negatively related to spring median flow rates in rivers. For example, combined Suwannee bass residuals revealed year-class strength was negatively related with spring median flow rates in rivers. Largemouth bass year-class strength was also negatively related with spring median flow rates and positively related to winter median flow rates in rivers. Similarly, combined black bass (M. salmoides and M. notius) residuals were negatively related with spring median flow rates and positively related to winter median flow rates in rivers. Mason et al. (1991) and Sallee et al. (1991) found that smallmouth bass M. dolomieu, year-class strength was negatively correlated with flow rates. Mason et al. (1991) 43

PAGE 54

44 surmised that high flow events caused low dissolved oxygen concentrations which may have negatively influenced smallmouth bass year class strength. High flow during spring may reduce nest success due to turbulence or increased juvenile smallmouth bass mortality due to displacement (Sallee et al. 1991; Filipek et al. 1991). Filipek et al. (1991) found that largemouth bass and spotted bass M. punctulatus year-class strength was negatively correlated with flow rates in rivers. Similarly, largemouth bass and Suwannee bass exhibited strong year classes in the Santa Fe River during low flow periods in the spring, similar to smallmouth bass year-class strength in three Virginia rivers (Scott Smith, VDGIF, Personal Communication). In Florida rivers, periods of low flow often result in clear water conditions which may increase aquatic macrophyte abundance, potentially improving habitat for young black basses and providing increased survival. Conversely, largemouth bass relative year-class strength was positively correlated with water levels in three of four lakes. Specifically, fall water levels after the spawn were positively correlated with largemouth bass residuals from Lakes Bonny and Crooked, and residuals were positively correlated with annual water levels at Lake Disston. Lake Bonny electrofishing samples in November 2001 yielded no age-0 and six age-1 fish (i.e., low mean annual water level in 2000 and 2001). Electrofishing samples in November 2002 indicated a strong 2002 year class, with 52 out of 71 fish being age-0 (i.e., high summer and fall water level) and no age-1 or age-2 fish were obtained, suggesting potential year class failure during low water years (2000 and 2001). Personal observation during electrofishing at Lake Bonny in the fall of 2001 revealed growth of terrestrial vegetation due to two consecutive years of low water. I surmise that, as a

PAGE 55

45 result of increased rainfall, terrestrial vegetation became inundated in 2002, producing a strong year-class of largemouth bass. Potential mechanisms for a strong year class could include increased food resources, increased available spawning habitat and decreased predation (Bross 1969; Jenkins 1970; Aggus and Elliot 1975; Keith 1975; Shirley and Andrews 1977; Aggus 1979; Timmons et al. 1980; Miranda et al. 1984; Ploskey 1986; Meals and Miranda 1991). Thus, I found evidence that largemouth bass year-class strength was positively related to stage in lakes but negatively related to flow in rivers. Trends were more evident in rivers as multiple regression models for combined systems were significant in rivers but no trends were found among lakes. Recruitment variability among black crappie populations has been characterized as a boom or bust fishery, with a strong year class forming every 3-5 years (Hooe 1991; Guy and Willis 1995; Allen and Miranda 1998; Maceina and Stimpert 1998). Black crappie residuals were negatively correlated with pre-spawn fall maximum water levels at Lake Bonny. The results dispute what has been found with stronger year classes of crappie Pomoxis spp. being positively correlated with higher water levels in the winter or spring (Beam 1983; Willis 1986; Miller et al. 1990; Maceina 2003). However, my results were similar to Maceina and Stimpert (1998), who also found that crappie year-class strength was influenced by hydrological variables prior to the spawn. Lower water levels prior to the spawning period concentrates prey items (i.e., higher food availability) (Jenkins 1970; Aggus 1979) which may provide more favorable conditions for a successful spawn (Liston and Chubb 1985) due to increased growth (Aggus 1979) and potentially higher fecundity of adults (Crim and Glebe 1990). Springate et al. (1985) found that food deprivation reduced the rate of maturation and the total fecundity of

PAGE 56

46 rainbow trout Oncorhynchus mykiss. Furthermore, Ploskey (1986) surmised that spawning success of largemouth bass did not necessarily insure a strong year class but increased the likelihood of producing one if environmental conditions were favorable. Thus, low water levels prior to the spawn may be an indicator of potentially strong year classes of black crappie in the following spring. However, I caution that this relation was found only in one lake where I obtained enough black crappie for assessment. Unlike black bass, combined bluegill and redbreast sunfish residuals were positively related with pre-fall median flow rates in rivers. Combined Lepomis spp. (auritus and macrochirus) residuals were positively related with pre-fall median flow rates and negatively related with fall median flow rates in rivers. Fish yield and production are strongly related to the extent of accessible floodplain (Junk et al. 1989). Aggus (1979) indicated that under conditions of an expanding physical environment, opportunistic species may increase greatly in abundance and biomass. High flow in the fall before a spawn could increase Lepomis spp. year-class strength in rivers through increased food availability and possible increases in fecundity (Springate et al. 1985; Crim and Glebe 1990). Access to the floodplain before the spawn may provide higher fecundity due to higher densities of invertebrate prey in floodplain habitats (Neckles et al. 1990), which may influence Lepomis spp. year-class strength in rivers. My results were variable among species and system types, but I detected a positive relationship between Lepomis spp. (auritus and macrochirus) year-class strength and fall median flow rates in three rivers combined. Redbreast sunfish may serve as indicator species for setting MFLs, because I found redbreast sunfish to be positively correlated with high flow rates

PAGE 57

47 in two rivers. In general, high flow rates in the fall prior to spawning was related to strong-year classes of Lepomis spp. in Florida rivers. Much is known about reservoirs and the relations of hydrology to sportfish year-class strength (Bross 1969; Jenkins 1970; Aggus and Elliot 1975; Keith 1975; Shirley and Andrews 1977; Aggus 1979; Timmons et al. 1980; Miranda et al. 1984; Ploskey 1986; Meals and Miranda 1991; Ozen and Noble 2002), but little is known about natural lakes (Tate et al. 2003). Few studies have assessed effects of flow on year-class strength of Lepomis spp. Meals and Miranda (1991) found that age-0 centrarchid abundance generally increased with water levels in Mississippi reservoirs. Bluegill and redear sunfish year-class strength were negatively and positively correlated with fall water level before the spawn and in spring and summer at Lake Disston, respectively. The lack of relationships between sunfish species and stage in lakes may result because of protracted spawning periods (Mettee et al. 1996), which could allow year-class success if environmental conditions become favorable for some period during the spring or summer. My sample sizes for redear sunfish may have been too low in lakes, which could have precluded detecting differences. Thus, no general trends were evident across lakes for Lepomis spp. Catch-curve regressions for sportfish revealed relatively constant recruitment (r 2 ) for many of the lakes. Due to relatively stable recruitment, correlations of residuals with hydrological variables were often not found. Despite fluctuations in water level, sportfish populations may have experienced constant recruitment due to lake morphometry (Figure 7). Guy and Willis (1995) suggested that recruitment patterns of black crappie may be influenced by the morphometry of the system. Lakes with shallow sloping shorelines can

PAGE 58

48 encounter drastic reductions of lake surface area during drought conditions (Hutchinson 1957). Conversely, lakes with steep sloping shorelines can encounter less than significant reductions of lake surface area during drought conditions (Hutchinson 1957). For example, a 1 m (meter) water level decrease at Lake Annie results in 17% less surface area (hectares) (Figure 2). In contrast, surface areas (hectares) of Lakes Bonny, Crooked, and Disston decrease 6, 9, and 8 % with a 1-m drop in water level. Lake Annies surface area varied the most of any lake but no significant correlations were found between year class strength of sportfish and water levels. Alternately, significant correlations were found with year-class strength of sportfish and water levels for Lakes Bonny, Crooked, and Disston despite smaller percent surface area decreases. The hypsographic curves were generated at stages below the maximum observed water level during this study, and as a result, surface area (hectares) estimates were up to 4% less than the actual maximum water level for the four lakes. Nevertheless, I expected year-class strength of sportfish to vary in lakes with large changes in surface area per change in water level. This did not occur possibly due to relatively few relationships found between stage and year-class strength at lakes. Differences in river morphology may explain why largemouth bass were not correlated with flow rates on the upper Ochlockonee River but negatively correlated with flow at the Santa Fe and Withlacoochee North Rivers. Bain et al. (1988) found that stream species respond differently to variable water levels and have specific microhabitat preferences. In addition, Talmage et al. (2002) used the habitat variables of percent overhanging vegetation, percent woody debris, and substrate characteristics to explain up to 50% of the variability in warmwater stream-fish abundance and species richness.

PAGE 59

49 Although measuring changes in riverine habitat and the extent of floodplain connection with changes in stage was not an objective of this study, effects of flow and stage on available habitat should be considered in future studies in Florida rivers. I found no relation between overall recruitment variability of sportfish indexed with the RVI and variation in stage or flow for either lakes or rivers. The main limitation in determining RVI is the proportion of missing year classes within an age-structure sample (Isermann et al. 2002). Furthermore, as the proportion of missing year classes in a sample increases, RVI decreases (Isermann et al. 2002). This limitation has proved to be problematic in our study because most of the sportfish populations had no missing year classes. The RVI may not be useful in assessing sportfish populations if missing year classes do not occur, potentially limiting the value of the RVI for these species. As with any scientific study, violation of assumptions during analysis can be a problem. I performed multiple catch curve analyses, assuming mortality was constant (Ricker 1975). In all likelihood, fishing mortality (F) differed among species and systems. Systems incurring high fishing mortality may demonstrate reduced age groups in the catch curve analysis (i.e., residuals). Also, F may have been variable among different ages within catch-curve regressions, which could have influenced my residuals. Use of residuals from catch-curve regressions to assess year-class strength in fish is best suited for longer-lived fishes (Maceina 1997). This may explain the low number of significant relationships, because my sample size and statistical power were low due to relatively short-lived fish, such as bluegill and redear sunfish. Previous studies (Schramm and Doerzbacher 1982; Hoyer et al. 1985; Maceina and Betsill 1987; Crawford et al. 1989; Schramm 1989; Hales and Belk 1992; Mantini et al.

PAGE 60

50 1992; Buckmeier and Howells 2003) have validated the formulation of one annulus per year for otoliths on all sportfish that I examined, except Suwannee bass. I assumed that rings on Suwannee bass otoliths were annuli, similar to largemouth bass in the Southeast. Although I found relations between sportfish year-class strength and hydrological variables in lakes and rivers, these correlations may not indicate a causal relationship. Spurious correlations could have resulted in my study, because a relatively large number of comparisons were made (Jackson and Somers 1991). Similar to other studies (Beam 1983; Willis 1986; Wrenn et al. 1996, Maceina and Stimpert 1998; Maceina 2003), my results indicated relations between flow/stage and fish year-class strength. Nevertheless, I caution that some of the relationships may not have been cause and effect. Future studies should quantitatively assess habitat changes with flow and stage changes to better elucidate mechanisms between system hydrology and fish recruitment. There are many factors that can affect electrofishing catch rates including fish behavior, fish size, fish species, population density, sampling crew, water clarity, water conductivity, water level, water temperature, and weather conditions (Hardin and Connor 1992; Hilborn and Walters 1992; Reynolds 1996; Bayley and Austen 2002). Electrofishing for this project took place during two separate hydrological regimes. Electrofishing in the fall of 2001 and spring of 2002 occurred when water levels and flow rates were near extreme minimum lows (i.e., drought conditions). Conversely, samples in the fall of 2002 and spring of 2003 occurred when water levels and flow rates were near mean or above historic mean values. Electrofishing catch rates of sportfish in the first year of sampling were much higher than in the second year of sampling for the same system. As a result, more ages (i.e., year classes) were generally obtained for the catch

PAGE 61

51 curves in the first year of sampling. The results may be due to sampling biases because my electrofishing equipment was not able to reach deeper areas in both the lakes and rivers during the second year of sampling (i.e., reduced catchability). Conversely, under low water conditions (i.e., low stage) sportfish became more concentrated and may not have been able to avoid the electric field, hence catch rates increased (Pierce et. al 1985). Nevertheless, strong and weak year-classes were usually corroborated in the catch curves. Substantial difficulty and disagreement was observed during the aging process of redbreast sunfish. I surmise that the input of water from natural springs in rivers coupled with a potential protracted spawning period (Davis 1972; Bass and Hitt 1974; Lukas and Orth 1993) occasionally produced false annuli (i.e., opaque bands) in redbreast sunfish otoliths. Hales and Belk (1992) encountered a similar problem with bluegill in a South Carolina cooling reservoir. However, after discussion with experienced otolith readers from the FFWCC (i.e., Steve Crawford and Eric Nagid, personal communication), the faintness and incompleteness of these false annuli distinguished them from true annuli. Therefore, I believe that redbreast sunfish were aged with adequate accuracy. Periodic natural fluctuations of water levels and flow rates occur on a regular basis due to rainfall. Despite these fluctuations, fish assemblages persist over time (Bass 1990; Paller 1997). Aumen and Gray (1995) supported a more natural hydrologic variability because they found that sustained high water levels may cause a decline in vegetated habitats resulting in lower sportfish abundance. Fish species that persist over time may be long-lived or may produce multiple cohorts within a year (Chesson and Warner 1981). As a result, once environmental and hydrological conditions are favorable (i.e., the ideal water level or flow rate), a strong year class is produced (Warner and Chesson 1985).

PAGE 62

52 Factors influencing year-class strength of sportfish can be ambiguous to fisheries biologists. Fish recruitment is a complex process that involves both density-dependent and independent factors (Everhart and Youngs 1981; Royce 1984), so I did not expect all the variation to be explained by hydrological variables. Studies have examined fish assemblage changes after the impoundment of a river (Quinn and Kwak 2003) or the institution of a minimum flow (Travnichek et al. 1995; Bowen et al. 1998), but I attempted to relate year-class strength of six recreationally important sportfish to hydrology in some of Floridas rivers and lakes. By examining the relationships of sportfish populations to hydrological influences, I related year-class strength of some sportfishes in Florida to the water levels and stream flow rates. Although it is easier from an agency perspective to manage multiple systems with the same hydrological regime, the variability in results indicate the need for system-specific management, particularly in lakes (Guy and Willis 1995; Isermann et al. 2002; Sammons et al. 2002) where I found few among-system trends. These results will possibly help natural resource managers develop effective MFLs to protect against damage to sportfish populations.

PAGE 63

MANAGEMENT IMPLICATIONS FOR SETTING MINIMUM FLOWS AND LEVELS Minimum Flows and Levels (MFLs) have been implemented by various methods and models including the Montana Method (Tennant 1976), the Instream Flow Incremental Methodology (IFIM) (Bovee 1982), the Physical Habitat Simulation (PHABISM) (Milhous et al. 1989), and the Range Variability Approach (RVA) (Richter et al. 1996). The Montana Method simulates flow regimes based on the average daily discharge or the mean annual flow. The IFIM predicts curves based on habitat preferences at different flow levels, such that these curves are based on field sampling of fish locations with associated measurements of habitat conditions (i.e., depth, velocity). The PHABISM is an intricate component of the IFIM that analyzes habitat availability for fishes. The RVA approach uses hydrologic variability such as timing, frequency, duration, and rates of change to sustain and protect natural ecosystem functions. All models have their advantages and shortcomings, but no particular model has become the norm for identifying the correct method for setting a MFL. The methods described above relate specific habitat changes to fish assemblages. Rather than making predictions of specific habitat changes and their influences on fish assemblages, I evaluated broad-scale relations between water level/flow and sport fish year-class strength in both lakes and rivers. My results suggest that among-system relationships are common for rivers but not for lakes. Thus, MFLs would be more likely to be regionally applicable in Florida rivers than in lakes. Although I evaluated impacts 53

PAGE 64

54 to sportfish populations, other portions of the fish assemblage (e.g., threatened or endangered fishes) were not evaluated and should be considered as part of biological criteria in setting MFLs. Currently, MFLs are set by the appropriate Florida Water Management District and the Florida Department of Environmental Protection according to previous evaluations of topography, soils, historical ground and surface water data, and vegetation data. Thus far, biological factors have not been considered when MFLs are set. Criteria used by establishing MFLs in Florida should include biological factors that prevent ecological harm. This study provided fish recruitment data that can be used for setting MFLs. Fish recruitment is one biological factor that should be taken into account when MFLs are set because freshwater sportfishing in Florida is a one billion dollar industry in Florida (USFWS 1996). Thus, MFLs should consider impacts to sportfish populations. Missing year classes occur naturally in fish populations due to adverse environmental conditions, but sportfish populations seem to be resilient and persist in multiple system types (Bass 1990; Bayley and Osborne 1993; Paller 1997). Despite reductions in number and abundance of species, Paller (1997) found the persistence of multiple species, including bluegill and largemouth bass, following 3.5 years of low water in a South Carolina reservoir. According to Lowe et al. (1994) and Neubauer et al. (2003), adverse environmental conditions (e.g., low flow) will not cause significant harm unless these conditions persist for more than five years. My results indicated that Lepomis spp. year-class strength was positively related to flow rate, and five years of low flow rates is probably too much time for these species to persist in a riverine ecosystem due to their life history. Lepomis spp. are short lived and mortality could eliminate

PAGE 65

55 species if complete year-class failure occurred for five years. Maximum age for redbreast sunfish collected in this study was age seven, but the majority of the individuals collected did not exceed age four. Conversely, life history strategies for Micropterus spp. include being longer lived, with a maximum age of 12 in this study. Thus, low flows for five or more years may cause weak year classes and reduced abundances of Lepomis spp. in rivers; potentially impacting sport fisheries. High flows at least once every three years in the fall may allow inundation of floodplain habitat, producing favorable environmental conditions for Lepomis spp. Setting MFLs during periods of prolonged drought (i.e., three years or more) should consider impacts to short lived species such as Lepomis spp. I contend that low flows for three or more consecutive years should be prevented, and thus, MFLs should consider biological impacts to short-lived fishes. Detecting impacts of flow on year-class strength of sportfish across lakes and rivers were variable but relationships were easier to detect in rivers. For example, in rivers, redbreast sunfish year-class strength was positively related to pre-fall median flow rates, whereas largemouth bass were negatively related to spring median flow rates and positively related to winter median flow rates. Thus, setting low flows could have detrimental effects on some fishes but positive on others. Nevertheless, fish have high fecundity and have adapted to highly variable conditions, particularly in rivers, where fish tend to persist through short periods of unfavorable conditions (Bayley and Osborne 1993). Thus, extreme high or low flow/stage events occur naturally, but minimum flows and levels should be set to prevent substantial alteration of this natural variability such as three or more consecutive years of low levels in lakes and rivers.

PAGE 66

APPENDIX A INADEQUATE CATCH CURVES Catch-curve linear regression equations that were not used in correlation analyses with hydrological variables. A ^ next to the species (i.e., redear sunfish ^) denotes < 50% agreement in residuals or inadequate sample size collected. Age is the number of fish ages used in each catch curve analysis. N is the total number of fish collected by electrofishing used to generate residuals and the associated r 2 and P values. Waterbody Species Year Equation r 2 P Ages N Lake Annie Bluegill 2003 Log (% Freq) = 7.24 1.45*age 0.99 0.01 3 70 Largemouth Bass 2003 Log (% Freq) = 5.20 0.64*age 0.95 0.01 5 270 Redear Sunfish ^ 2002 Log (% Freq) = 3.45 0.13*age 0.01 0.96 3 128 2003 Log (% Freq) = 5.82 0.86*age 0.94 0.03 4 118 Lake Bonny Black Crappie 2002 Log (% Freq) = 6.87 0.92*age 0.73 0.15 4 96 Bluegill 2002 Log (% Freq) = 4.17 0.61*age 0.65 0.10 5 520 Largemouth Bass 2002 Low Sample Size Crooked Lake Black Crappie ^ 2002 Low Sample Size 2003 Low Sample Size Bluegill 2002 Log (% Freq) = 5.83 0.91*age 0.97 0.00 6 428 Largemouth Bass 2003 Log (% Freq) = 4.76 0.52*age 0.72 0.03 6 166 Redear Sunfish 2002 Low Sample Size Lake Disston Bluegill 2003 Log (% Freq) = 4.82 0.52*age 0.90 0.00 7 258 Largemouth Bass 2003 Log (% Freq) = 4.62 0.48*age 0.93 0.00 6 93 Redear Sunfish 2003 Low Sample Size Ochlockonee Bluegill 2002 Log (% Freq) = 7.77 1.66*age 0.99 0.00 3 401 River Largemouth Bass 2002 Log (% Freq) = 5.02 0.62*age 0.87 0.00 4 119 Redbreast Sunfish 2002 Log (% Freq) = 7.20 1.52*age 0.88 0.22 3 1486 Suwannee Bass ^ 2001 Low sample size 2002 Low sample size Santa Fe River Largemouth Bass 2002 Log (% Freq) = 4.38 0.43*age 0.57 0.14 5 115 Redbreast Sunfish ^ 2001 Log (% Freq) = 7.31 1.47*age 0.99 0.00 4 330 2002 Log (% Freq) = 6.33 1.06*age 0.93 0.04 4 155 Suwannee Bass 2002 Log (% Freq) = 5.16 0.63*age 0.93 0.01 5 213 Withlacoochee Bluegill 2002 Log (% Freq) = 6.70 1.18*age 0.91 0.05 4 281 River (South) Largemouth Bass 2002 Log (% Freq) = 5.68 0.81*age 0.83 0.03 5 243 Redbreast Sunfish 2003 Log (% Freq) = 4.91 0.53*age 0.46 0.53 3 59 56

PAGE 67

APPENDIX B AGE FREQUENCIES Age-frequency distributions and recruitment variability index (RVI) values for each species, system, and year. An asterisk next to the year (i.e., 2001*) denotes the sample used for relation to the hydrological variables. A ^ next to the species (i.e., rbsu^) denotes < 50% agreement in residuals or inadequate sample size, therefore the sample was not used in any correlation analyses. Samples with less than 3 fish obtained of a particular age were used to index RVI values but were not included in catch-curve analyses. In addition, year-classes were removed from the catch-curves and RVI index, if they were below the assigned catchable size. Species abbreviations are as follows: bg = bluegill; blc = black crappie; lmb = largemouth bass; rbsu = redbreast sunfish; resu = redear sunfish; swb = Suwannee bass. Age Water Body Species Year RVI 0 1 2 3 4 5 6 7 8 9 10 11 12 Lake bg 2002* 0.73 9 25 18 15 8 . . . Annie bg 2003 0.91 12 54 13 3 . . . . lmb 2002* 0.86 101 119 83 32 25 7 4 1 . . lmb 2003 0.85 92 120 75 39 28 8 1 1 . . resu^ 2002 0.73 0 37 53 38 2 2 2 0 1 . resu 2003 0.72 5 56 45 12 5 2 . . . Lake blc 2001* 0.77 19 11 59 52 89 14 4 4 . . Bonny blc 2002 0.75 57 29 23 48 19 27 2 1 . . bg 2001* 0.86 763 327 130 206 113 9 5 . . . bg 2002 0.87 2398 369 31 64 41 15 . . . lmb 2001* 0.77 0 6 12 23 14 6 4 1 1 . . lmb 2002 0.43 52 0 0 3 4 3 3 4 1 1 . Crooked blc^ 2002 0.61 1 3 8 10 1 1 3 0 1 . Lake blc 2003 0.29 2 0 4 6 4 0 2 3 . . bg 2002 0.70 48 53 35 17 13 17 5 4 0 1 . bg 2003* 0.91 52 240 140 23 16 6 3 1 . lmb 2002* 0.82 86 93 114 74 19 7 7 2 . . lmb 2003 0.77 50 31 52 46 27 7 3 1 . . resu 2002 0.78 4 8 10 5 1 1 . . . resu 2003* 0.65 19 26 29 16 14 17 5 4 0 1 . Lake Disston bg 2002* 0.87 88 173 114 46 25 5 3 5 . . bg 2003 0.81 25 68 67 59 37 17 6 4 2 1 . lmb 2002* 0.81 54 46 31 10 12 10 3 1 . . lmb 2003 0.83 5 32 27 19 6 5 4 1 1 1 . resu 2002* 0.66 11 10 7 11 11 2 . . . resu 2003 0.53 2 6 8 5 1 6 1 . . 57

PAGE 68

58 Appendix B. Continued. Age Water Body Species Year RVI 0 1 2 3 4 5 6 7 8 9 10 11 12 Ochlockonee bg 2001* 0.68 0 447 109 100 10 6 . . . River bg 2002 0.94 167 287 325 63 16 2 . . . lmb 2001* 0.30 87 54 19 25 9 3 2 3 0 0 0 0 1 lmb 2002 0.86 122 85 57 20 31 11 2 2 2 2 1 . rbsu 2001* 0.89 0 468 179 233 18 4 1 1 . . rbsu 2002 0.72 456 623 1313 110 63 0 2 . . . swb^ 2001 0.36 4 4 2 2 1 2 1 2 0 0 0 0 1 swb 2002 0.48 9 10 6 3 3 2 3 0 0 2 . Santa Fe River lmb 2001* 0.88 29 53 100 31 10 22 1 1 3 . . lmb 2002 0.80 84 55 40 48 7 9 11 2 . . rbsu^ 2001 0.91 0 70 249 63 15 31 1 1 . . rbsu 2002 0.93 6 76 64 11 4 1 . . . swb 2001* 0.70 25 49 123 28 19 20 1 2 3 1 0 1 swb 2002 0.85 27 61 82 75 36 11 9 1 2 . . Withlacoochee lmb 2002* 0.55 108 28 11 15 5 4 7 2 1 4 0 0 1 River (North) rbsu 2002* 0.90 205 149 369 163 68 7 6 2 . . swb 2002* 0.33 31 18 13 17 7 0 0 2 . . Withlacoochee bg 2002 0.88 67 164 74 39 4 2 . . . River (South) bg 2003* 0.83 17 66 41 14 12 3 . . . lmb 2002 0.88 104 116 69 23 32 3 1 1 1 . lmb 2003* 0.55 123 24 45 24 21 12 5 6 0 0 1 rbsu 2002* 0.90 34 141 61 9 2 . . . rbsu 2003 0.74 37 20 32 7 . . . .

PAGE 69

LIST OF REFERENCES Aggus, L. R. 1979. Effects of freshwater fish predatorprey dynamics. Pages 47-56 in H. Clepper, editor. Predator-Prey Systems in Fisheries Management. Sport Fishing Institute, Washington, D.C., USA. Aggus, L. R., and G. V. Elliot. 1975. Effects of cover and food on year-class strength of largemouth bass. Pages 317-322 in R. H. Stroud and H. Clepper, editors. Black Bass Biology and Management. Sport Fishing Institute, Washington, D. C., USA. Allen, M. S., and L. E. Miranda. 1998. An age-structured model for crappie fisheries. Ecological Modeling 107:289-303. Anderson, C. S. 1995. Measuring and correcting for size selection in electrofishing mark-recapture experiments. Transactions of the American Fisheries Society 124:663-676. Aumen, N. G., and S. Gray. 1995. Research synthesis and management recommendations from a five-year ecosystem-level study of Lake Okeechobee, Florida (USA). Archiv fur Hydrobiologie 45:343-356. Bain, M. B., J. T. Finn, and H. E. Booke. 1988. Streamflow regulation and fish community structure. Ecology 69:382-392. Bass, D. G. 1990. Stability and persistence of fish assemblages in the Escambia River, Florida. Rivers 1:296-306. Bass, D. G., and V. G. Hitt. 1974. Ecological aspects of the redbreast sunfish, Lepomis auritus, in Florida. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 28:296-307. Bayley, P. B., and D. J. Austen. 2002. Capture efficiency of a boat electrofisher. Transactions of the American Fisheries Society 131:435-451. Bayley, P. B., and L. L. Osborne. 1993. Natural rehabilitation of stream fish populations in an Illinois catchment. Freshwater Biology 29:295-300. Beam, J. H. 1983. The effect of annual water level management on population trends of white crappie in Elk City Reservoir, Kansas. North American Journal of Fisheries Management 3:34-40. 59

PAGE 70

60 Bettoli, P. W., and L. E. Miranda. 2001. Cautionary note about estimating mean length at age with subsampled data. North American Journal of Fisheries Management 21:425-428. Bettoli, P. W., M. J. Maceina, R. L. Noble, and R. K. Betsill. 1993. Response of a reservoir fish community to aquatic vegetation removal. North American Journal of Fisheries Management 13:110-124. Bovee, K. D. 1982. A guide to stream habitat analysis using the instream flow incremental methodology. Instream flow information paper no. 12. U.S. Fish and Wildlife Service, Washington, D. C., USA. Bowen, Z. H., M. C. Freeman., and K. D. Bovee. 1998. Evaluation of generalized habitat criteria for assessing impacts of altered flow regimes on warmwater fishes. Transactions of the American Fisheries Society 127:455-468. Bross, M. G. 1969. Fish samples and year-class strength (1965-1967) from Canton Reservoir. Proceedings of the Oklahoma Academy of Science 48:194-199. Buckmeier, D. L., and R. G. Howells. 2003. Validation of otoliths for estimating ages of largemouth Bass to 16 years. North American Journal of Fisheries Management 23:590-593. Carlander, K. D. 1977. Handbook of freshwater fish biology, Volume 2. Iowa State University Press, Ames, Iowa, USA. Chesson, P. L., and R. R. Warner. 1981. Environmental variability promotes existence in lottery competitive systems. The American Naturalist 117:923-943. Conover, W. J. 1980. Practical nonparametric statistics. Second Edition, John Wiley and Sons, New York, New York, USA. 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. Crim, L.W., and B. D. Glebe. 1990. Reproduction. Pages 529-553 in C. B. Schreck and P. B. Moyle, editors. Methods of fish biology. American Fisheries Society, Bethesda, Maryland. Davis, J. R. 1972. The spawning behavior, fecundity rates, and food habits of the redbreast sunfish in southeastern North Carolina. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 25:556-560. Everhart, W. H., and W. D. Youngs. 1981. Principles of Fisheries Science. Second Edition, Cornell University Press, London.

PAGE 71

61 Filipek, S. P., M. A. Armstrong, and L. G. Claybrook. 1991. Effects of a hundred-year flood on the smallmouth bass population of the Upper Caddo River, Arkansas. Pages 84-89 in D. C. Jackson, editor. The First International Smallmouth Bass Symposium. Mississippi Agricultural and Forestry Experiment Station, Mississippi State University, Starkville, Mississippi, USA. Florida LAKEWATCH. 2000. Florida LAKEWATCH Data 1986-1999. Department of Fisheries and Aquatic Sciences, University of Florida/Institute of Food and Agricultural Sciences. Library, University of Florida. Gainesville, Florida, USA. Forsberg, C., and S. O. Ryding. 1980. Eutrophication parameters and trophic state indices in 30 Swedish waste-receiving lakes. Archives fur Hydrobiologie 88: 189-207. Grossman, G. D., J. F. Dowd, and M. Crawford. 1990. Assemblage stability in stream fishes: review. Environmental Management 14:661-671. Guy, C. S., and D. W. Willis. 1995. Population characteristics of black crappies in South Dakota Waters: A case for ecosystems-specific management. North American Journal of Fisheries Management 15:754-765. Hales, L. S., and M. C. Belk. 1992. Validation of otolith annuli of bluegills in a southeastern thermal reservoir. Transactions of the American Fisheries Society 121:823-830. Hardin, S., and L. L. Connor. 1992. Variability of electrofishing crew efficiency and sampling requirements for estimating reliable catch rates. North American Journal of Fisheries Management. 12:612-617. Hilborn, R., and C. J. Walters. 1992. Quantitative fisheries stock assessment: choice, dynamics, and uncertainty. Chapman and Hall, New York. Hooe, M. L. 1991. Crappie biology and management. North American Journal of Fisheries Management 11:483-484. Houde, E. D. 1987. Fish early life dynamics and recruitment variability. American Fisheries Society Symposium 2:17-29. Hoyer, M. V., and D. E. Canfield. 1996. Largemouth bass abundance and aquatic vegetation in Florida Lakes: an empirical analysis. Journal of Aquatic Plant Management 34:23-32. 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.

PAGE 72

62 Hutchinson, G. E. 1957. Hydrological cycle and water balance. Pages 221-249 in J. Wiley and Sons, editors. A treatise on limnology, Volume 1:geography, physics and chemistry, New York, New York, USA. Isermann, D. A., W.L. McKibbin, and D. W. Willis. 2002. An analysis of methods for quantifying crappie recruitment variability. North American Journal of Fisheries Management 22:1124-1135. Jackson, D. A., and K. M. Somers. 1991. The spectre of spurious correlations. Oecologia 86:147-151. Jenkins, R. M. 1970. The influence of engineering design and operation and other environmental factors on reservoir fishery resources. Water Resources Bulletin 6:110-119. Junk, W. J., P. B. Bayley, and R. E. Sparks. 1989. The flood pulse concept in river-floodplain systems. Proceedings of the International Large River Symposium (LARS). Canadian Special Publications in Fisheries and Aquatic Sciences 106:110-127. Keith, W. E. 1975. Management by water level manipulation. Pages 489-497 in H. Clepper, editor, Black bass biology and management. Sport Fishing Institute, Washington, D. C., USA. Kriksunov, Ye. A., and K. M. Mamina. 1995. Effects of flows in the Ural River on recruitment of stellate sturgeon, Acipensar stellatus. Journal of Ichthyology 34:662-665, 35:52-58. Liston, C. R., and S. Chubb. 1985. Relationships of water level fluctuations and fish. Pages 121-140 in H. H. Prince and F. D. DItri. Coastal Wetlands. Lewis Publishers, Chelsea, Michigan. Lowe, E. F., S. Mortellaro, R. Mattson, S. Flannery, and B. Epting. 1994. Final Report, District Water Management Plan-Conventions Subcommittee on Impacts to Natural Systems. Florida Water Management Districts and Florida Department of Environmental Regulation Joint Subcommittee Report, Palatka, Florida, USA. Lukas, J. A., and D. J. Orth. 1993. Reproductive ecology of redbreast sunfish Lepomis auritus in a Virginia stream. Journal of Freshwater Ecology 8:235-244. Maceina, M. J. 2003. Verification of the influence of hydrologic factors on crappie recruitment in Alabama reservoirs. North American Journal of Fisheries Management 23:470-480. Maceina, M. J. 1997. Simple application of using residuals from catch-curve regressions to assess year-class strength in fish. Fisheries Research 32:115-121.

PAGE 73

63 Maceina, M. J., and R. K. Betsill. 1987. Verification and use of whole otoliths to age white crappie. Pages 267-278 in R. C. Summerfelt and G. E. Hall, editors. The age and growth of fish. The Iowa State University Press, Ames, Iowa, USA. Maceina, M. J., and P. W. Bettoli. 1998. Variation in largemouth bass recruitment in four mainstream impoundments of the Tennessee River. North American Journal of Fisheries Management 18:998-1003. Maceina, M. J., and M. R. Stimpert. 1998. Relations between reservoir hydrology and crappie recruitment in Alabama. North American Journal of Fisheries Management 18:104-113. Mantini, L., M. V. Hoyer, J. V. Shireman, and D. E. Canfield. 1992. Annulus validation, time of formation, and mean length at age of three sunfish species in north central Florida. Proceedings of the Annual Conference of Southeast Association of Fish and Wildlife Agencies 46:357-367. Mason, J. W., D. J. Graczyk, and R. A. Kerr. 1991. Effects of runoff on smallmouth bass populations in four southwestern Wisconsin streams. Pages 28-35 in D. C. Jackson, editor. The First International Smallmouth Bass Symposium. Mississippi Agricultural and Forestry Experiment Station, Mississippi State University, Starkville, Mississippi, USA. Meals, K. O., and L. E. Miranda. 1991. Variability in abundance of age-0 centrarchids among littoral habitats of flood control reservoirs in Mississippi. North American Journal of Fisheries Management 11:298-304. Mettee, M. F., P. E. ONeil, and J. M. Pierson. 1996. Fishes of Alabama and the Mobile basin. Oxmoor House, Birmingham, Alabama, USA. Miller, S. J., D. D. Fox, L. A. Bull, and T. D. McCall 1990. Population dynamics of black crappie in Lake Okeechobee, Florida, following suspension of commercial harvest. North American Journal of Fisheries Management 10:98-105. Milhous, R. T., M. A. Updike, and D. M. Schneider. 1989. Physical habitat simulation system reference manual, version 2. Instream flow information paper 26. U.S. Fish and Wildlife Service Biological Report 89(16). Miranda, L. E., W. L. Shelton, and T. D. Bryce. 1984. Effects of water level manipulation on abundance, mortality, and growth of young-of-year largemouth bass in West Point Reservoir, Alabama-Georgia. North American Journal of Fisheries Management 4:314-320. Mitzner, L. R. 1995. Effect of environmental factors and harvest regulations upon the crappie ( Pomoxis ) sportfishery at Rathbun Lake. Technical Bulletin No. 5. Iowa Department of Natural Resources, Ames, Iowa, USA.

PAGE 74

64 Myers, R. H. 1990. Classical and modern regression with applications. PWS-Kent, Boston, Massachusetts, USA. Neckles, H. A., H. R. Murkin, and J. A. Cooper. 1990. Influences of seasonal flooding on macroinvertebrate abundance in wetland habitats. Freshwater Biology 23: 311-322. Neubauer, C. P., G. B. Hall, E. F. Lowe, R. B. Hupalo, and L. W. Keenan. 2003. The multiple minimum flows and levels method of the St. Johns River Water Management District. St. Johns River Water Management District, Palatka, Florida, USA. Ozen, O., and R. L. Noble. 2002. Relationship between water level fluctuations and largemouth bass spawning in a Puerto Rico Reservoir. Pages 213-220 in D. P. Philip and M. S. Ridgway, editors. Black Bass: ecology, conservation, and management. American Fisheries Society, Bethesda, Maryland, USA. Paller, M. H. 1997. Recovery of a reservoir fish community from drawdown related impacts. North American Journal of Fisheries Management 17:726-733. Peterman, R. M. 1990. Statistical power analysis can improve fisheries research and management. Canadian Journal of Fisheries and Aquatic Sciences 47:2-15. Petts, G. E. 1996. Water allocation to protect river ecosystems. Regulated Rivers: Research and Management 12:353-365. Pierce, R. B., D. W. Coble, and S. D. Corley. 1985. Influence of river stage on shoreline electrofishing catches in the upper Mississippi River. Transactions of the American Fisheries Society 114:857-860. Ploskey, G. R. 1986. Effects of water-level changes on reservoir ecosystems, with implications for fisheries management. Pages 86-97 in G. E. Hall and M. J. Van Den Avyle, editors. Reservoir Fisheries Management: Strategies for the 80's. American Fisheries Society, Southern Division Reservoir Committee, Bethesda, Maryland. Poff, N. L., and J. V. Ward. 1989. Implications of streamflow variability and predictability for lotic community structure: a regional analysis of streamflow patterns. Canadian Journal of Fisheries and Aquatic Sciences 46:1805-1818. Poff, N. L., and seven coauthors. 1997. The natural flow regime: A paradigm for river conservation and restoration. BioScience 47:769-784. Pope, K. L., and D. W. Willis. 1996. Seasonal influences on freshwater fisheries sampling data. Reviews in Fisheries Science 4:57-73.

PAGE 75

65 Power, M. E., A. Sun, M. Parker, W. E. Dietrich, and J. T. Wooten. 1995. Hydraulic food-chain models: an approach to the study of food web dynamics in large rivers. Bioscience 45: 159-167. Quinn, J. W., and T. J. Kwak. 2003. Fish assemblage changes in an Ozark River after impoundment: A long-term perspective. Transactions of the American Fisheries Society 132:110-119 Raibley, P. T., T. M. OHara, K. S. Irons, K. D. Blodgett, and R. E. Sparks. 1997. Largemouth bass size distributions under varying annual hydrological regimes in the Illinois River. Transactions of the American Fisheries Society 126: 850-856. Reice, S. R., R. C. Wissmar, and R. J. Najman. 1990. Disturbance regimes, resilience, and recovery of animal communities and habitats in lotic systems. Environmental Management 14:647-659. Resh, V. H., and nine coauthors. 1988. The role of disturbance in stream ecology. Journal of the North American Benthological Society 7:433-455. Reynolds, J. B. 1996. Electrofishing. Pages 221-253 in B. R. Murphy and D. W. Willis, editors. Fisheries Techniques, 2nd edition. American Fisheries Society, Bethesda, Maryland, USA. Richter, B. D., J. V. Baumgartner, J. Powell, and D. P. Braun. 1996. A method for assessing hydrologic alteration with ecosystems. Conservation Biology 10:1163-1174. Ricker, W. E. 1975. Computation and interpretation of biological statistics of fish populations. Bulletin 191, Fisheries Research Board of Canada, Ottawa. Royce, W. F. 1996. Introduction to the practice of fishery science. Academic Press, San Diego, California, USA. Sallee, R. D., J. L. Langbein, H. Brown, and J. Ferencak. 1991. Effects of discharge fluctuations on survival of smallmouth bass in the Kankakee River, Illinois. Pages 90-95 in D. C. Jackson, editor. The First International Smallmouth Bass Symposium. Mississippi Agricultural and Forestry Experiment Station, Mississippi State University, Starkville, Mississippi, USA. Sammons, S. M., P. W. Bettoli, D. A. Isermann, and T. N. Churchill. 2002. Recruitment variation of crappies in response to hydrology of Tennessee reservoirs. North American Journal of Fisheries Management 22:1393-1398. SAS. 2000. Users guide, version 8. SAS Institute Inc., Cary, North Carolina, USA. Schramm, H. J., and J. F. Doerzbacher. 1982. Use of otoliths to age black crappie from Florida Proceedings of the Annual Conference of Southeast Association of Fish and Wildlife Agencies 36:95-105.

PAGE 76

66 Schramm, H. J. 1989. Formation of annuli in otoliths of bluegills. Transactions of the American Fisheries Society 118:546-555. Shireman, J. V., M. V. Hoyer, M. J. Maceina, and D. E. Canfield. 1985. The water quality of Lake Baldwin, Florida: Four years after vegetation removal by grass carp. Proceedings of the Fourth Annual Conference and International Symposium, North American Lake Management Society 4:201-206. Shirley, K. E., and A. K. Andrews. 1977. Growth, reproduction, and mortality of largemouth bass during the first year of life in Lake Carl Blackwell, Oklahoma. Transactions of the American Fisheries Society 106:590-595. Sigler, W. F., and J. W. Sigler. 1990. Recreational Fisheries: management, theory, and application. University of Nevada Press, Reno, Nevada, USA. St. Johns River Water Management District (SJRWMD). 2001. St. Johns River Water Management District: minimum flows and levels. (Available on the World Wide Web at http://sjr.state.fl.us). Smith, S. M., and D. J. Orth. 1990. Distributions of largemouth bass in relation to submerged aquatic vegetation in Flat Top Lake, West Virginia. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 44:36-44. Springate, J. R. C., N. R. Bromage, and P. R. T. Cumaranatunga. 1985. The effects of different ration on fecundity and egg quality in the rainbow trout (Salmo gairdneri). Pages 371-393. in C. Cowey, editor. Feeding and nutrition in fish. Academic Press, London. Talmage, P. J., J. A. Perry, and R. M. Goldstein. 2002. Relation of instream habitat and physical conditions to fish communities of agricultural streams in the northern Midwest. North American Journal of Fisheries Management 22:825-833. Tate, W. B., M. S. Allen, R. A. Myers, E. Nagid, and J. Estes. 2003. Relation of age-0 largemouth bass abundance to hydrilla coverage and water level at Lochloosa and Orange Lakes, Florida. North American Journal of Fisheries Management. 23:251-257. Taubert, 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. Tennant. D. L. 1976. Instream flow regimens for fish wildlife, recreation and related environmental resources. Pages 359-373 in J. F. Orsborn and C. H. Allman, editors. Instream Flow Needs. American Fisheries Society, Bethesda, Maryland, USA.

PAGE 77

67 Timmons, T. J., W. L. Shelton, and W. D. Davies. 1980. Differential growth of largemouth bass in West Point Reservoir, Alabama-Georgia. Transactions of the American Fisheries Society 109:176-186. Travnichek, V. H., M. B. Bain, and M. J. Maceina. 1995. Recovery of a warmwater fish assemblage after initiation of a minimum-flows release downstream from a hydroelectric dam. Transactions of the American Fisheries Society 124:836-844. U.S. Department of the Interior, Fish and Wildlife Service. 1996. National Survey of Fishing, Hunting, and Wildlife Associated Recreation, Washington, D.C., USA Walker, K. F., F. Sheldon, and J. T. Puckridge. 1995. A perspective on dryland river ecosystems. Regulated Rivers: Research and Management 11:85-104. Warner, R. R., and P. L. Chesson. 1985. Coexistence mediated by recruitment fluctuations: a field guide to the storage effect. The American Naturalist 125:769-787. Willis, D. W. 1986. Review of water level management on Kansas reservoirs, Pages 110-114 in G. E. Hall and M. J. Van Den Avyle, editors. Reservoir Fisheries Management: Strategies for the 80's. American Fisheries Society, Southern Division Reservoir Committee, Bethesda, Maryland, USA. Willis, D. W., and B. R. Murphy. 1996. Planning for sampling. Pages 1-15 in B. R. Murphy and D. W. Willis, editors. Fisheries Techniques, 2nd edition. American Fisheries Society, Bethesda, Maryland, USA. Wrenn, W. B., D. R. Lowery, M. J. Maceina, and W. C. Reeves. 1996. Relationships between largemouth bass and aquatic plants in Guntersville Reservoir, Alabama. Multidimensional approaches to reservoir fisheries management. American Fisheries Society, Special Publication 16:382-393.

PAGE 78

BIOGRAPHICAL SKETCH Timothy Frederick Bonvechio was born October 29, 1977, in West Palm Beach, Florida, the son of Robert Bonvechio, Jr. and Constance Bonvechio. He graduated from Palm Beach Gardens Community High School in 1996. He received his Associate of Arts degree in Biology from Palm Beach Community College in May of 1998. He received his Bachelor of Science degree in Forest Resources and Conservation from the University of Florida in December of 2000. While finishing undergraduate studies, he gained part-time employment in May of 2000, as a freshwater fisheries technician under the direction of Dr. Mike Allen. The day after Tim received his Bachelor of Science degree, Dr. Allen recruited him on as a full-time technician. During his tenure as technician, he gained valuable experience in fisheries management. As a result, he began his graduate work at the University of Florida in the spring of 2002 to pursue a Master of Science degree in fisheries management. After graduation in December 2003, he plans on pursuing a career in fisheries management as a state freshwater fisheries biologist. Upon departure, he leaves an old saying which has no author, When God created the earth, He made it two-thirds water and one-third land, surely he meant for man to spend more time fishing than plowing. 68


Permanent Link: http://ufdc.ufl.edu/UFE0001460/00001

Material Information

Title: Relations between Hydrological Variables and Year-Class Strength of Sportfish in Eight Florida Waterbodies
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0001460:00001

Permanent Link: http://ufdc.ufl.edu/UFE0001460/00001

Material Information

Title: Relations between Hydrological Variables and Year-Class Strength of Sportfish in Eight Florida Waterbodies
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0001460:00001


This item has the following downloads:


Full Text












RELATIONS BETWEEN HYDROLOGICAL VARIABLES AND YEAR-CLASS
STRENGTH OF SPORTFISH IN EIGHT FLORIDA WATERBODIES
















By

TIMOTHY F. BONVECHIO


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

UNIVERSITY OF FLORIDA


2003

































Copyright 2003

by

Timothy F. Bonvechio

































This document is dedicated to to my parents, Robert and Constance.















ACKNOWLEDGMENTS

This thesis would not have been completed without the hard work and dedication of

many people. I thank R. Burnes, R. Cailteux, P. Cooney, R. Crook, K. Dockendorf, G.

Kaufman, J. Krummrich, S. Larsen, C. Mwatela, G. Ripo, M. Rogers, T. Thompson, N.

Trippel, K. Tugend and B. Whitfield for providing help with the field sampling. I thank

J. Berg, R. Crook, P. Cooney, K. Dockendorf, M. Duncan, D. Dutterer, K. Henry, K.

Jacoby, G. Kaufman, S. Larsen, C. Mwatela, M. Rogers, T. Thompson, K. Tugend, T.

Tuten, N. Trippel, B. Whitfield, and J. Wingate for the help in the lab processing and

preparing any presentation. I thank J. Donze, M. Hoyer, R. Varner, D. Watson, and D.

Willis for their help in generating bathymetric maps. I thank S. Crawford and E. Nagid

for their help in aging problematic sunfish species. I thank J. Hill and C. Cichra for their

MFL literature review. I thank my committee members, M. Allen, C. Cichra, J. Estes,

and D. Murie, for the many helpful comments they offered throughout this study. I

especially thank my committee chair, M. Allen, for giving me the chance to fish and

work under him, and hopefully, absorb some of his knowledge. This research was

funded by the Wildlife Foundation of Florida, Inc.

Finally, I thank my fiancee, Kimberly Tugend, for dealing with me during this

study. I thank my parents, Robert and Constance Bonvechio, for teaching me the right

way to get the job done in life and giving me the confidence to tackle any task in life.
















TABLE OF CONTENTS

page

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

LIST OF TABLES ....................................................... ............ .............. vii

L IST O F FIG U R E S .............. ............................ ............. ........... .......... viii

ABSTRACT ........ ........................... ...... ........... .......... ix

INTRODUCTION .............. ................................... ..............

M E T H O D S ............................................................................ . 4

S ite S e le c tio n ................................................................................................................ 4
Species Selected................................................... 5
D ata C collection and A nalyses.............................................................. ............... 5

R E S U L T S ................................................................................13

H ydrological Fluctuations ................................................. ............................... 13
Sampling Effort and Age Distribution......................................... 13
V verification A naly ses ......................................................................... ............. 14
System-Specific Correlation Analyses of Residuals with Hydrological Variables.... 16
L ak e A n n ie ................................................................16
L a k e B o n n y .................................................................................................... 1 6
C ro ok ed L ak e ...............................................................17
L ak e D issto n ................................ ......... .. ................................... ................... 17
Ochlockonee River ....................... ................ .. ...............18
S a n ta F e R iv e r ................................................................................................ 1 9
Withlacoochee River North ..................................................... 19
W ithlacoochee R iver South ............................................................. .... ........ 19
Multiple Regression Analysis of Combined Residuals ........................................ 20
L akes ................................................................... .. ........ 20
R iv ers ............... .... ........ ... ...... ........ ......... ............... ............... ....2 1
Correlation Analysis of Combined Recruitment Variability Indexes........................22

D ISCU SSIO N ........................................................................... ........ ........ 43




v









MANAGEMENT IMPLICATIONS FOR SETTING MINIMUM FLOWS AND
L E V E L S ........................................................................... 5 3

APPENDIX

A INADEQUATE CATCH CURVES ........................................ ....................... 56

B A G E FR E Q U E N C IE S ........................................................................ .................. 57

L IST O F R EFER EN CE S ............................................................................ ..............59

B IO G R A PH IC A L SK E TCH ..................................................................... ..................68
















LIST OF TABLES


Table pge

1. Characteristics of lakes and rivers sampled. ................................. .................30

2. Historical lake elevation (m) means, the coefficient of variation (CV =
SD/0* 100%), number of water level observations (N), and ranges of each year
for the four study lakes .............................................. .. .. .... .. ........ .... 31

3. Historical river flow rate (m3/sec) means, coefficient of variation (CV =
SD/0*100%), number of flow rate observations (N), and ranges of each year for
the four study rivers............. .... .......................................................... .. .... .... .. 33

4. Waterbody, species, season and year of sampling. ...............................................35

5. Description of otolith reader agreement across system, species, season, and
year. ...................... ...................................... 37

6. Catch-curve linear regression equations for each waterbody and species with r2,
P ages, and N .........................................................................39

7. Significant correlation relationships between measures of bluegill, black crappie,
largemouth bass, and redear sunfish year-class strength and stage for four lakes...40

8. Significant correlations between measures of bluegill, largemouth bass, redbreast
sunfish, and Suwannee bass year-class strength and flow rate for the four
riv ers ................. ..................................... ...........................4 1

9. Significant multiple regression equations predicting year-class strength (i.e.,
residual) from flow rates for the four rivers. ....................................................... 42















LIST OF FIGURES


Figure pge

1. Location of eight sam ple sites. ........................................ ............................ 23

2. Determination of year-class strength displaying the hypothetical catch-curve (a)
and residual plot (b).................... ..................... ........... 24

3. Florida LAKEWATCH personnel generated this map of Lake Annie by using
differentially corrected global positioning equipment (GPS). ................................25

4. Florida LAKEWATCH personnel generated this map of Lake Bonny by using
differentially corrected global positioning equipment (GPS). ................................26

5. Florida LAKEWATCH personnel generated this map of Crooked Lake by using
differentially corrected global positioning equipment (GPS). ................................27

6. Florida LAKEWATCH personnel generated this map of Lake Disston by using
differentially corrected global positioning equipment (GPS). ................................28

7. Hypsographic curves displaying the change in surface area with hypothetical
declines in water level (m) for each of the four lakes. .............................................29















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

RELATIONS BETWEEN HYDROLOGICAL VARIABLES AND YEAR-CLASS
STRENGTH OF SPORTFISH IN EIGHT FLORIDA WATERBODIES

By

Timothy F. Bonvechio

December 2003

Chair: Micheal S. Allen
Major Department: Fisheries and Aquatic Sciences

Hydrological variables, such as water level (m) and flow rate (m3/s), have

influenced recruitment of sportfishes in lakes, reservoirs, and rivers. I evaluated how

annual and seasonal flows and water levels were related to year-class strength of selected

sportfishes across four rivers and four lakes in Florida. Species investigated included

black crappie Pomoxis nigromaculatus, bluegill Lepomis macrochirus, largemouth bass

Micropterus salmoides, redbreast sunfish, L. auritus, redear sunfish L. microlophus, and

Suwannee bass M. notius. Hydrological data were obtained from the U.S. Geological

Survey and the appropriate Florida Water Management Districts.

The residuals from catch-curve regressions developed from otoliths of fish caught

by electrofishing were used to assess relationships between hydrological variables and

year-class strength of sportfish. The multiple regression model computed for the

Suwannee bass residuals combined from two rivers indicated that year-class strength was

negatively related to spring median flow rates (r2 = 0.38). Similarly, largemouth bass









residuals combined from four rivers indicated that year-class strength was negatively

related to spring median flow rates and positively related to winter median flow rates

(r2 = 0.24). Conversely, bluegill residuals combined from two rivers indicated that year-

class strength was positively related to the previous fall median flow rates (r2 = 0.38).

Redbreast sunfish residuals combined from three rivers indicated that year-class strength

was positively related to the previous fall median flow rates before spawning (r2 = 0.31).

On a broader scale, Micropterus spp. (i.e., notius and salmoides) residuals were

combined from four rivers and a multiple regression model indicated that year-class

strength was negatively related to spring median flow rates and positively related to

winter median flow rates (r2 = 0.22). Lepomis spp. (i.e., auritus and macrochirus)

residuals were combined from three rivers and the multiple regression equation indicated

that year-class strength was positively related to fall median flow rates before spawning

and negatively related to fall median flow rates (r2 = 0.47). No significant regression

models were detected between water level in lakes and year-class strength of sportfish.

Management implications of this work include regulation changes pertaining to

minimum flows and levels (MFLs). Detecting impacts of flow on year-class strength of

sportfish across lakes and rivers were variable but relationships were easier to detect in

rivers. High flows, at least once every three years in the fall, may allow inundation of

floodplain habitat, thus providing favorable environmental conditions for Lepomis spp. in

rivers. Setting MFLs during periods of prolonged drought (i.e., three years or more)

should consider impacts to short-lived species such as Lepomis spp. I contend that low

flows for three or more consecutive years should be prevented and thus, MFLs should

consider biological impacts to short-lived fishes.















INTRODUCTION

Water demand due to human population expansion in central Florida has increased

ground-water removal, resulting in reduced ground water and surface water levels,

particularly during periods of low rainfall. Reduced surface-water levels may have

consequences for important sport fisheries in rivers and lakes. Previous studies have

shown that low water levels may reduce habitat availability and substantially alter fish

and invertebrate communities (Travnichek et al. 1995; Petts 1996).

Recruitment can be defined as the number of fish born in a given year that survive

to reproductive or harvestable size (Willis and Murphy 1996). Recruitment into a fishery

can be influenced by a number of density-dependent and independent factors (Everhart

and Youngs 1981; Royce 1996). Possible density-dependent factors include cannibalism,

disease, and predation (Houde 1987). Density-independent factors may include

variations in temperature, turbidity, flow rate, or water level changes (Everhart and

Youngs 1981; Sigler and Sigler 1990; Royce 1996). Increased water level may increase

the amount of spawning habitat and food resources available to sportfish by inundating

shoreline vegetation (Jenkins 1970; Aggus and Elliot 1975; Keith 1975; Timmons et al.

1980; Miranda et al. 1984; Meals and Miranda 1991). Alternately, above average flow

rates in lotic systems have both positively and negatively influenced recruitment of fish,

depending on the system and species (Filipek et al. 1991; Mason et al. 1991; Sallee et al.

1991; Kriksunov and Mamina 1995; Raibley et al. 1997). Thus, variations in flows and

water levels may influence fish recruitment.









The extent of drought and flood periods will influence the magnitude, frequency,

duration, and timing of changes in water flow (Grossman et al. 1990; Reice et al. 1990),

which can influence fish community structure in rivers (Poff and Ward 1989; Walker et

al. 1995; Richter et al. 1996; Poff et al. 1997). Stream flow is considered a major

variable that affects the abundance and distribution of many riverine species (Resh et al.

1988; Power et al. 1995).

Catch-curve analysis has been used by fishery biologists to estimate total annual

mortality (Ricker 1975). Residuals from catch curves can be used as an index of relative

year-class strength, with positive residuals indicating relatively strong year classes and

negative residuals indicating relatively weak year classes (Maceina and Bettoli 1998).

Residuals can then be used to relate year-class strength to environmental variables

(Wrenn et al. 1996; Maceina 1997). Wrenn et al. (1996) used a catch-curve multiple

regression with environmental variables and determined that reservoir retention time over

16 days produced strong year classes of largemouth bass Micropterus salmoides. This

technique was also used to relate recruitment of largemouth bass to flow conditions in

two Alabama reservoirs (Maceina 1997). In addition, Sammons et al. (2002) used

residuals to index crappie recruitment in several Tennessee reservoirs.

Minimum flows and levels (MFLs) for Florida lakes and rivers are set by the

appropriate Florida Water Management District governing board to prevent significant

ecological harm to waterbodies as a result of permitted water withdrawals (SJRWMD

2001). The state legislature requires the establishment of MFLs under Subsection

373.042(2), Florida Statutes (F.S.) (SJRWMD 2001). Based on previous evaluations of

topography, soils, historical ground and surface water data, and vegetation data, MFLs









are mandated to prevent harm to that ecological system (SJRWMD 2001). The MFLs are

also reviewed periodically by the appropriate water management district and the Florida

Department of Environmental Protection (SJRWMD 2001). However, criteria for

establishing MFLs in Florida have not included biological factors such as ecological

harm to fish recruitment. Knowledge of potential hydrological impacts on sportfishes

could be used when MFLs are established and reviewed.

I evaluated the influence of water level and flow rate on year-class strength of six

important sportfish species. My objectives were to (1) estimate the age frequency of

selected sportfish species across eight Florida waterbodies; (2) use residuals around a

catch curve to index relative year-class strength for each species and (3) relate relative

year-class strength of each species to historic water-level fluctuations within each

waterbody and among waterbodies for lakes and rivers.















METHODS

Site Selection

Waterbodies were chosen for sampling based on two requirements: 1) historical

stream-flow (m3/sec) or water-level (m) data dating back to at least 1992, which

experienced significant changes in the flow or water level; and 2) historically low levels

of aquatic vegetation. Water-level data back to 1992 were needed to have a minimum of

nine years included for analyses with residuals. Aquatic vegetation can influence the

amount of habitat that is gained or lost due to fluctuations in water level. High

percentages of aquatic vegetation in waterbodies influence fish communities (Shireman et

al. 1985; Smith and Orth 1990; Bettoli et al. 1993; Hoyer and Canfield 1996, Tate et al.

2003); thus fluctuating water levels may have a small impact on the amount of available

habitat when abundant macrophytes are present. To reduce the confounding effect of

aquatic plants on water level effects, I selected lakes that had historically low plant

coverage (i.e., <20% coverage). Electrofishing transects were 20 minutes and included

samples from all available habitat types. All sampling took place in the fall (September)

through spring (March) due to potential sampling biases in the summer (Pope and Willis

1996).

Lakes Annie, Bonny, Crooked, and Disston were chosen because of their

fluctuating water levels and low to moderate levels of aquatic vegetation (Figure 1).

Lake characteristics are shown in Table 1. I performed electrofishing at Lake Annie from

January to February 2002 and January 2003, at Lake Bonny in November 2001 and from









October to November 2002, at Crooked Lake from January to February 2002 and January

2003, and at Lake Disston in March 2002 and February 2003. The four rivers targeted

for sampling were the Santa Fe, Ochlockonee, Withlacoochee North, and Withlacoochee

South (Figure 1, Table 1). I performed electrofishing in the middle Santa Fe River in

October 2001 and September 2002, at the upper Ochlockonee River in November 2001

and from September to October 2002, at the middle Withlacoochee River South in March

2002 and from February to March 2003, and at the middle Withlacoochee River North

from November to December 2002.

Species Selected

The species chosen varied across systems but included black crappie Pomoxis

nigromaculatus, bluegill Lepomis macrochirus, largemouth bass, redbreast sunfish

L. auritus, redear sunfish L. microlophus, and Suwannee bass M. notius. All are sought

by anglers as sportfishes. The species chosen for collection differed among lakes and

rivers based on presence and abundance. Suwannee bass and redbreast sunfish are

usually present only in rivers, whereas bluegill, black crappie, largemouth bass, and

redear sunfish are present in both lakes and rivers.

Data Collection and Analyses

Three boats equipped for electrofishing were used in 2001 and 2002. I used two,

4.9-m jon boats with either a 50-hp or 60-hp outboard and a 4.3-m jon boat with a 15-hp

outboard. Electrofishing equipment consisted of the same generator (5000-W AC) and

either a Coffelt model VVP 15 or Smithroot model VI-A pulsator and similar cathode

probes with electrical output ranging from 5-8 amps of DC current.

All large individuals of each species were kept for age analyses due to possible

growth differences between male and females (e.g., largemouth bass; Carlander 1977).









The number of fish to be sacrificed was a concern, so all species were subsampled in

smaller size groups where gender-specific growth was expected to be similar (Bettoli and

Miranda 2001). Five largemouth bass per centimeter group < 35 cm total length (TL)

and all largemouth bass > 35 cm TL were kept. Five Suwannee bass per centimeter

group < 25 cm TL and all Suwannee bass > 25 cm TL were kept. For black crappie,

bluegill, redear, and redbreast sunfish, five fish per centimeter group < 15 cm TL and all

fish > 15 cm TL were kept unless 20 fish were obtained per cm group > 15 cm TL. Fish

were placed on ice and returned to the laboratory where they were measured and

weighed, gender determined, and sagittal otoliths removed.

Age and growth for all species was determined from annuli on the sagittal otoliths

(Schramm and Doerzbacher 1982; Taubert and Tranquilli 1982; Hoyer et al. 1985;

Maceina and Betsill 1987; Crawford et al. 1989; Schramm 1989; Hales and Belk 1992;

Mantini et al. 1992; Buckmeier and Howells 2003). The annuli on otoliths from all fish

were counted by observing the whole otolith under a microscope. Otoliths from fish

> age-1 were mounted on frosted glass microscope slides with super glue and then

sectioned transversely along the dorsoventral plane into 0.5-mm sections using a South

Bay Tech. Inc., Model 650, diamond wheel saw. Sections were mounted on glass

microscope slides using Thermo Shandon synthetic mountant cement and the annuli were

counted under a dissecting microscope. The total number of rings or annuli were

recorded. The date of collection was considered when determining the age of the otolith

(Hoyer et al. 1985; Crawford et al. 1989). Ages were determined by number of rings

alone or the number of rings +1 for those fish caught before or after January 1st,

respectively. Two independent readers were used, and in the instance a discrepancy









occurred between the two readings, a third reader was used to resolve the age or the

otolith was discarded. To remain consistent and potentially increase precision, the same

three readers were used for the duration of the study.

An age-length key (Ricker 1975) was used to estimate the age frequency of

subsampled fish below the specified length, and in some cases above the specified length

(i.e., if> 20 fish/cm group were collected above the specified length). The fraction of

fish in the aged subsample (N = 5 fish/cm group) was extrapolated to the unaged fish

below the specified length to assign fish age based on length. The age frequency of each

species offish below and above the specified length was combined to estimate the total

number of fish at each age.

Sampling biases due to short-term weather events, flow conditions, or water level

could cause a biased age structure. To assess the potential for sampling biases within a

single year, I verified the age structure of fishes in three rivers and four lakes in the

second sampling year. Fish were collected and aged as described above, and the age

structure of each species was compared to data from the previous year for each

waterbody. The Withlacoochee River North was not sampled in the second year.

Bathymetric maps were generated for lakes Annie, Bonny, Crooked, and Disston

by Florida LAKEWATCH (2000, 2001, and 2003) personnel using differentially

corrected global positioning equipment (GPS). A planimeter (Keuffel and Esser Co.) was

used to trace the shoreline contour of each lake in planimeter units. The planimeter units

were then compared to the scale of each map and converted into the actual surface area

for each lake. Surface areas of different depths in each lake were plotted on

hypsographic curves. A hypsographic curve indicates the change in lake surface area per









change in water level. The hypsographic curves were generated to assess potential

influence of water level on lake surface area and thus potential changes in inshore habitat

availability.

Using the estimated age-frequency for each species and system, I performed catch-

curve analyses:

loge (NUMBER) = bo bi (AGE) (1)
where NUMBER is the estimated total number of fish of age x and AGE is fish age in

years. Because sample size varied among species and waterbodies, the estimated total

number of fish of each age was standardized into percent frequency at each age prior to

the loge transformation.

Catch-curve residuals (i.e., observed deviation from expected values from the

regression line) display variation in recruitment among years, and therefore index

relative-year class strength (Maceina 1997; Figure 2). Maceina (1997) defined strong

and weak year classes as residuals greater than 0.50 and less than -0.50, respectively. I

expected positive residuals to correspond to strong year classes and negative residuals to

correspond to weak year classes. The magnitude of positive and negative residuals affect

the amount of variation explained (r2) in a simple linear regression model such as a catch

curve. For example, I would expect relatively constant recruitment if the r2 is high

(i.e., > 0.95). Variability in recruitment of black crappie has been indexed with the

recruitment variability index (RVI) (Guy and Willis 1995).

The RVI was calculated as described by Guy and Willis (1995):

RVI = [SN/(Nm + Np)] Nm/Np (2)
where SN is the summation of the cumulative relative frequency distribution based on the

number of fish in each age-class, Nm is the number of age-groups missing from the









sample that should be present (which does not include ages past the last age obtained),

Np is the number of age-groups present in the sample, and Np > Nm. The RVI ranges

from -1 to 1. Recruitment is more stable as RVI increases from -1 to 1. All assumptions

described by Guy and Willis (1995) were met in order for estimates to be obtained. Thus,

the recruitment coefficient of determination (RCD = r2), RVI, and residual analysis are

three options for assessing recruitment patterns of infrequently sampled populations

(Isermann et al. 2002). I used all three indices to assess variability of recruitment for

each species. The RVI has previously not been applied to sportfish populations other

than black crappie.

Capture efficiency of boat electrofishing is size selective (Bayley and Austen

2002). Bayley and Austen (2002) determined catchability of largemouth bass and

bluegill across a range of fish sizes. Based on their estimates, I used minimum catchable

size groups to determine what age groups to include in the catch curves. Because

electrofishing captures small fish at lower rates than their true abundance (Anderson

1995; Reynolds 1996), a catchable size was determined for each species across systems.

Lepomis spp. and Pomoxis spp. were assigned a 100 mm-TL catchable size and

Micropterus spp. were assigned a 200 mm-TL catchable size. Bayley and Austen (2002)

found that fish above these sizes did not vary greatly in catchability. I assumed that

Lepomis spp. and Pomoxis spp. catchabilities varied with fish size in a similar manner to

bluegill in Bayley and Austen (2002) and that Suwannee bass catchability would be

similar to largemouth bass. If at least 60% of the fish in an age group were assigned

catchable size, that year class was kept in the catch curve. If less than 60% of the fish

were above catchable size or there were less than three fish in the age group (Ricker









1975), then the year class was deleted from the catch curve. Due to variable growth rates

across systems and species, the age range of catchable-sized fish used in the catch curves

fluctuated across systems and species.

A sign test (Conover 1980) was performed on the residuals using the

UNIVARIATE procedure (SAS 2000) to verify that similar strong and weak year classes

in a subsequent year did occur and were not the result of sampling biases within a single

year. Residuals for each year class sampled in consecutive years were scored as a "+" if

both were above or below the regression line, which indicated agreement between the

two sample years. Conversely, if a year- class residual was above the line in year one but

below the line in year two, or vice versa, the year-class was scored as a "-". The sign test

assessed whether "+" scores were significantly higher than "-" scores, which would

indicate that residuals coincided between years across systems and species (a = 0.10). To

perform the sign test, the species were grouped according to rivers and lakes, and

separate sign tests were performed on the two groups.

I selected the best catch-curve regression for each system and year to relate

hydrological data using three criteria. First, the catch curve was selected to contain ages

obtained during years when extreme water level or flow fluctuations occurred (i.e.,

minimum or maximum flow rates). Second, the catch curve that contained the most ages

that were fully recruited to the gear was used, given the first criteria was met. If both

catch curves incorporated extreme environmental conditions and contained the same

number of ages, then the sample with the larger number of fish was chosen.

Correlation analyses was used to assess the relationship between residuals and

environmental variables for the year the fish in each age were born for each system.









The water level (m) and flow data (m3/sec) were separated into four periods: 1) January

through March was considered winter: 2) April through June was considered spring;

3) July through September was considered summer; and 4) October through December

was considered fall. Previous researchers (Maceina and Stimpert 1998; Sammons et al.

2002) have found hydrological conditions before spawning was related to crappie year-

class strength, so the fall periods before the spawn were also evaluated in the correlation

analyses for each sportfish species and were referred to as pre-spawn fall. Correlation

analyses was performed on the residuals with the levels of minimum, maximum, and

mean flow rates and/or stages for the annual, fall pre-spawn, winter, spring, summer, and

fall periods.

To assess trends across systems, catch-curve residuals were standardized as percent

frequencies and hydrological data (i.e., stream flow, m3/sec, and water level, m) were

also standardized as a percent of the median on a seasonal basis. Residuals were

combined according to genus (e.g., Lepomis), species (e.g., macrochirus), and system

type (i.e., river) for multiple regression with hydrological variables. I only obtained one

black crappie catch curve on one lake, and thus black crappie were not included in the

lake analyses of combined residuals. I used multiple linear regression models to

determine if year-class strength (i.e., residuals) were related to the five seasonal variables

(i.e., median flow or stage for all seasons) from the equation (Myers 1990).

Y = Po + 1l(xi) + 02(x2)+ ... + Pk(xk) + i (3)
Where Y is the year-class strength (residual), Po is the intercept value, Pk is parameter

estimate, and xk is the independent variable. Independent variables evaluated for flow

and stage included pre-fall, winter, spring, summer, and fall percent of median values.

To compare across systems, median values were used in the multiple liner regression









models due to the influence of extreme values on the mean. Backward selection was

used and the procedure terminated when the removal of no other independent variables

could significantly (P < 0.10) improve the overall model. Multiple linear regression was

performed on residuals across systems to provide some insight on regional responses of

fish year-class strength in relation to flow or stage. The multiple regression models also

allowed for multiple independent variables to be evaluated concurrently.

Multicollinearity is a common problem in multiple regression models, and I checked for

this problem by evaluating the Variance Inflation Factor (VIF) (Myers 1990). Regression

models with a VIF > 10 were not observed and as a result, multicollinearity was not

considered a problem in the analyses.

To assess whether variation in flow or stage was related to overall recruitment

variability, RVI values were correlated with the overall coefficient of variation

(CV=SD/0.*100%) in flow or stage across systems for each species. All statistical

analyses were conducted with SAS (2000) and statistical tests were considered significant

when P < 0.10 and marginally significant relationships when P = 0.11. A P-value of 0.10

was chosen due to low power associated with most correlations (i.e., N = 5 age classes),

and because I considered Type-II error to be important (Peterman 1990).















RESULTS

Hydrological Fluctuations

The overall historical range of water levels across the four lakes fluctuated from

< 1 to > 3 meters among years (Table 2). The overall range in water level was highest

at Lake Crooked (3.8 m) and lowest at Lake Disston (1.2 m)(Table 2). For the four

rivers, the overall historical range of flow rates varied from 0 to 1,062 m3/sec among

years (Table 3). Overall flow ranges were highest in the Withlacoochee North

(2.2-1,062 m3/sec) and lowest in the Withlacoochee South (0.9-150 m3/sec) (Table 3).

The bathymetric maps of Lakes Annie, Bonny, Crooked and Disston were generated at

97, 99, 96, and 97% of the maximum water level recorded, respectively (Figures 3, 4, 5,

and 6). Bathymetric maps display the outermost contour line representing the shoreline

of each lake at the time of map construction. These outermost contour lines were subject

to change depending on the lake levels. Hypsographic curves revealed the potential

influence of water level on lake surface area (Figure 7).

Sampling Effort and Age Distribution

A combined total of 576 electrofishing transects were conducted on the eight

waterbodies. The Florida Fish and Wildlife Conservation Commission (FFWCC)

contributed an additional 16 electrofishing transects on the Ochlockonee River in 2002,

increasing the total electrofishing transects to 592. Most transects (> 97%) were sampled

for 20 minutes in both years, but transects other than 20 minutes (10-17 min) were









included in the analyses. To obtain an adequate age sample, electrofishing effort differed

according to the system, species, and year (Table 4).

The number of fish collected and aged differed among systems, species and year

(Table 4). A total of 19,301 sportfish were collected during the study: 8,298 bluegill;

5,558 redbreast sunfish; 3,675 largemouth bass; 733 Suwannee bass; 529 redear sunfish;

and 508 black crappie (Table 4). For all populations combined, bluegill ranged from 22

to 263 mm TL, black crappie ranged from 109 to 350 mm TL, largemouth bass ranged

from 48 to 635 mm TL, redbreast sunfish ranged from 33 to 244 mm TL, redear sunfish

ranged from 56 to 312 mm TL, and Suwannee bass ranged from 67 to 406 mm TL (Table

4).

A total of 6,327 sportfish were aged during the study, which included 1,918

bluegill, 463 black crappie, 2,238 largemouth bass, 802 redbreast sunfish, 434 redear

sunfish, and 472 Suwannee bass (Table 4). For all samples combined, there was 96%

between-reader agreement resulting from 6,063 reader agreements out of 6,327 otoliths

(Table 5). The 264 otolith disagreements were read by a third independent reader. A

total of five otoliths were thrown out from the 264 disagreements because the third

independent reader determined an age that was different than either of the first two

readers. For all systems and sample years combined, between-reader agreement among

species ranged from 91 to 98%. Ages ranged from 0 to 10 years for bluegill, 0 to 9 for

black crappie, 0 to 12 for largemouth bass, 0 to 7 for redbreast sunfish, 1 to 10 for redear

sunfish, and 0 to 12 for Suwannee bass.

Verification Analyses

A total of 36 usable catch-curves, 20 from lakes and 16 from rivers, were generated

on sportfish population age samples obtained from seven systems. Depending on the









species and the system, some samples of aged fish were smaller than catchable size and

could not be verified with the coinciding age sample. These under represented age

samples were not included in the sign test. The sign test on the residuals from the lakes

and rivers revealed that similar strong and weak year classes in subsequent years, N = 43,

P = 0.05, and N = 32, P = 0.02, respectively. Residuals matched 65% (28/43) of the time

for lakes and matched 72% (23/32) of the time for rivers. Thus, strong and weak year

classes were evident in both years in most cases. Changes in the age frequencies caused

year classes with low residuals to change from positive to negative or vice versa (4/15

lake residual disagreements and 2/11 river residual disagreements) but fell within 0.20 of

matching with each other. As a result, only 26% and 22% of lake and river residual

disagreements were extreme values (i.e., > 0.2), respectively. These arbitrary residuals

that fluctuated around zero were left in the two separate sign tests but did not

significantly influence the results. Nevertheless, strong and weak year-classes were

generally consistent between sample years.

The verification process determined which systems were adequately sampled for

two years in a row and could then be included in the correlation and multiple regression

analysis of residuals with the hydrological variables. A specific system and a species

were included if at least 50% of the residuals matched in consecutive years (Table 6).

I assumed samples that did not meet this criteria, were not related to water levels or flow

rates and were not used in the analyses. Species samples that were not verified due to a

small age sample during one year or the species was only sampled one time included

Lake Annie bluegill, Lake Bonny largemouth bass, Crooked Lake redear sunfish, Lake

Disston redear sunfish, and the Withlacoochee River North largemouth bass, redbreast









sunfish, and Suwannee bass. These samples were used in the correlation and multiple

regression analyses.

System-Specific Correlation Analyses of Residuals with Hydrological Variables

Lake Annie

No significant relationships between stage and relative year-class strength were

found for Lake Annie, and most of the variation (r2) was explained in the catch-curve

regressions (Table 6). Bluegill and largemouth bass residuals were not correlated

with any annual or seasonal water levels on a minimum, maximum or mean level (all

P > 0.27). Thus, relatively constant recruitment of bluegill and largemouth bass residuals

were found on Lake Annie and year-class strength was not related to water level.

Lake Bonny

Year-class strength of sportfish was variable in relation to water levels at Lake

Bonny. Black crappie residuals from the catch curve indicated variable recruitment was

present (i.e., low r2 value, Table 6). Black crappie residuals from the catch curve were

negatively correlated with maximum pre-spawn fall water levels (Table 7). Thus, low

maximum water levels prior to the spawn may be an indicator of potentially strong year

classes of black crappie in the following year. Largemouth bass residuals from the catch

curve revealed relatively constant recruitment from 1995-1998. Although most of the

variation (r2) was explained in the catch curve regression (Table 6), largemouth bass

residuals from the catch curve were positively correlated with minimum, maximum, and

mean fall water levels (all P < 0.06; Table 7). As a result, higher water levels may

provide young-of-year largemouth bass. Black crappie and largemouth bass residuals

were not correlated with any other annual or any other seasonal water levels (all

P > 0.12). Bluegill residuals from the catch curve were not correlated with any annual









or seasonal water level variables (all P > 0.15). Thus, year class strength of bluegill was

not related to water level, but high water level in the fall was positively related to

largemouth bass and negatively related to black crappie in the pre-spawn fall season.

Crooked Lake

No significant relationships were found at Crooked Lake except for largemouth

bass. Largemouth bass residuals from the catch curve indicated variable recruitment was

not present (i.e., high r2 value, Table 6). But, largemouth bass residuals were positively

correlated with minimum, maximum, and mean fall water levels (all P < 0.07; Table 7).

Largemouth bass residuals were not correlated with any other annual or seasonal water

level on a minimum, maximum or mean level (all P > 0.25). Bluegill and redear sunfish

residuals from the catch curve regressions revealed relatively constant recruitment from

1995-2001, and much of the variation (r2) was explained (Table 6). Bluegill and redear

sunfish residuals were not correlated with any annual or seasonal water level variables

(all P > 0.19). Thus, year-class strength of bluegill and redear sunfish was not related to

water level, but high water level in the fall was positively related to largemouth bass

year-class strength.

Lake Disston

Relations between stage and residuals varied among species at Lake Disston. Most

of the variation (r2) was explained in the bluegill catch curve (Table 6), but residuals were

negatively correlated with pre-spawn fall water levels (P < 0.07; Table 7). Most of the

variation (r2) was also explained in the largemouth bass catch curve (Table 6), but

largemouth bass residuals were positively correlated with minimum annual water levels

(Table 7). Most of the variation (r2) was left unexplained in the redear sunfish catch

curve (Table 6), and redear sunfish residuals were positively correlated with pre-spawn









fall minimum and mean water levels (Table 7). Bluegill, largemouth bass, and redear

sunfish residuals were not correlated with any other annual or seasonal water level

variables (all P > 0.13). Results indicate an increase in water level during the fall before

the spawn were positively related to year-class strength for redear sunfish and negatively

related to for bluegill. Also, an increase in the minimum water level on an annual basis

was positively related to largemouth bass year-class strength.

Ochlockonee River

The relationships between flow rates and year-class strength of sportfish at the

Ochlockonee River were variable. Bluegill residuals from the catch curve were

positively correlated with flow variables with the exception of pre-spawn minimum fall

flow rates (all P < 0.07; Table 8). Variable bluegill relationships with flow rates may

indicate the species ability to persist during multiple flow regimes. Largemouth bass

residuals from the catch curve indicated variable recruitment was present (i.e., low

r2 value, Table 6), but no significant relationships were found between flow and relative

largemouth bass year class strength in this system (all P > 0.12).

Redbreast sunfish residuals from the catch curve were positively related to flow

variables at this system (all P < 0.11; Table 8). Redbreast sunfish year-class strength was

positively associated with flow in several seasons including winter, summer, and fall

seasons. Thus, high flow favored strong redbreast sunfish year classes. No other

sportfish residuals were significantly correlated with any other annual or seasonal flow

rate variables (all P > 0.12). Results indicate strong year classes of bluegill and redbreast

sunfish were associated with higher flow rates but no significant relationships were

detected for largemouth bass. Thus, relationships were variable across species and

seasons in the Ochlockonee River.









Santa Fe River

Conversely, black bass year-class strength was negatively correlated to flow in the

Santa Fe River (both P < 0.09; Table 8). Both largemouth bass and Suwannee bass year-

class strength were negatively related to flow, particularly during the spring (Table 8).

No other black bass residuals were correlated with any other annual or seasonal flow rate

variables (all P > 0.12). Strong year classes of largemouth bass and Suwannee bass were

associated with low flow rates, indicating potential persistence of these species in the

event of a water withdrawal from the river.

Withlacoochee River North

Similarly, largemouth bass residuals from the catch curve were negatively

correlated to flow variables at the Withlacoochee River North (all P < 0.10; Table 8) for

several seasons, suggesting that high flow reduced largemouth bass year-class strength

(Table 8). Largemouth bass residuals were not correlated with any other annual or

seasonal flow rate variables (all P > 0.12). Redbreast sunfish and Suwannee bass

residuals were not correlated with any annual or seasonal flow rate variables (all P >

0.13). Thus, low flow appeared to be favorable for strong largemouth bass year-class

strength in the Withlacoochee River North and the Santa Fe River.

Withlacoochee River South

Generally, sunfish (L. auritus and L. macrochirus) year-class strength from the

Withlacoochee River South was positively correlated with flow rates and no relationships

were detected for largemouth bass. Bluegill residuals from the catch curve, when

significant, were positively related to flow variables (all P < 0.11; Table 8). Largemouth

bass residuals were not correlated with any annual or seasonal flow rate variables (all

P > 0.40). No other sportfish residuals were correlated with any other annual or seasonal









flow rates (all P > 0.12). Thus, high flow favored strong bluegill and redbreast sunfish

year classes in this system.

Historical hydrological data for both lakes and rivers indicated high water levels

and flow rates during 1997 and 1998 (i.e., high rainfall), whereas extremely low water

levels and flow rates were observed in late 1999 through 2001 (i.e., drought conditions).

The fluctuations in water level and/or flow rate data were related to year-class strength

(i.e., residuals) among some of the systems. For example, largemouth bass from Crooked

Lake displayed a relatively strong year-class in 1998 (i.e., high annual mean stage) (Table

2) and a weak year class in 2000 (i.e., lower annual mean stage) (Table 2), as indicated

by positive and negative residuals (i.e., +0.58, -0.49). Also, largemouth bass from Lake

Disston displayed a potential strong year-class in 1998 (i.e., high mean annual water

level) (Table 2), as indicated by a positive residual (i.e., +0.55). Other positive residuals

(i.e., strong year classes) > +0.40 in 1998 included Lake Bonny largemouth bass;

Ochlockonee River bluegill, largemouth bass, and redbreast sunfish; Withlacoochee

River North redbreast sunfish and the Withlacoochee River South bluegill. Thus, I found

evidence that high water in 1998 caused relatively strong year classes.

Multiple Regression Analysis of Combined Residuals

Lakes

Multiple regression models for each species combined across lakes revealed

no significant relationships between seasonal stage values and year-class strength (all

P > 0.13). The number of residuals included in the models ranged from ten for redear

sunfish to 34 for Lepomis spp. combined. Thus, although within-lake relationships were

found for several species at three of four lakes evaluated, I found no trends in year-class









strength across the lakes using multiple regression, suggesting that regional trends were

not evident.

Rivers

In general, multiple regression models combined across rivers indicated year-class

strength of sunfish (i.e., redbreast sunfish and bluegill) was positively related to

flow rates and black bass year-class strength was negatively related to flow rates (all

P < 0.10; Table 9). The number of residuals included in the models ranged from eight for

Suwannee bass to 32 for Micropterus spp. combined.

The best multiple regression equation for bluegill indicated that year-class strength

was positively related with pre-spawn fall median flow rates and the model explained

38% of the variation (Table 9). Similar to bluegill, the best model for redbreast sunfish

indicated that year-class strength was positively related with pre-spawn fall median flow

rates and the model explained 31% of the variation (Table 9). Conversely, the best model

for largemouth bass indicated that year-class strength was negatively related with spring

median flow rates and positively related to winter median flow rates, and the model

explained 24% of the variation (Table 9). Similar to largemouth bass, the best Suwannee

bass model indicated that year-class strength was negatively related with spring median

flow rates and the model explained 38% of the variation (Table 9).

The best multiple regression equation for sunfish (L. auritus and L. macrochirus)

indicated that year-class strength was positively related to pre-spawn fall median flow

rates and negatively related to fall median flow rates, and the model explained 47% of the

variation (Table 9). The best model for black basses (M salmoides and M. notius)

indicated that year-class strength was negatively related to spring median flow rates and

positively related to winter median flow rates and the model explained 22% of the









variation (Table 9). Thus, regression equations revealed variable responses between

sunfish and black bass populations and flow conditions across systems.

Correlation Analysis of Combined Recruitment Variability Indexes

Recruitment variability indexes were combined across system type (i.e., lakes or

rivers) for each species. In order to perform a correlation analysis, a minimum of three

data points were needed, so pooled RVI species samples that contained less than three

data points were discarded from the analysis (i.e., black crappie were only obtained from

one system). These discarded samples included black crappie and redear sunfish in lakes

and bluegill and Suwannee bass in rivers. The coefficient of variation (CV) was obtained

for the annual water level of each lake and the annual water level and flow rate from each

river for correlation analysis with RVI.

Combined bluegill RVIs in four lakes were not correlated with the CV of water

level (P = 0.65). Combined largemouth bass RVIs in four lakes were not correlated with

the CV of water level (P = 0.63). Combined largemouth bass RVIs in four rivers were

not correlated with any flow rate (P = 0.41). Combined redbreast sunfish RVIs in three

rivers were not correlated with any flow rate (P = 0.44). Thus, overall recruitment

variation indexed with the RVI was not significantly related to overall flow or stage

variation among these systems.













Ochlockonee River


Lake Disston

Lake Annie

Crooked Lake

Lake Bonny


Santa Fe River


Withlacoochee River South


N


A


0 100 200 300 Kilometers
| !


Figure 1. Location of eight sample sites. Arrows are pointing to approximate sample
locations. All locations were sampled for two consecutive years during the
same season (e.g., fall or winter), except the Withlacoochee River North
which was only sampled in one year.













Catch Curve





Z= -0.66
R2= 0.79
P 0.02


o 2
2000


1999 199 1997 1996I I
1999 1998 1997 1996


1995


1994


Year Class


1.5

1.0

0.5

0

-0.51

-1.0


Residual Plot






1998 1997 1996 1995
9 1998 1997 1996 1995 1S



Year Class


Figure 2. Determination of year-class strength displaying the hypothetical catch-curve
(a) and the associated residual plot (b).









Annie (Polk County)
Florida LAKEWATCH Bathymetric Map


0 200


800 m


Figure 3. Florida LAKEWATCH personnel generated this map of Lake Annie by using
differentially corrected global positioning equipment (GPS). Data were
collected March 20, 2003. Scale and map contours are in meters and were
generated using kriging technique in Surfer software package (Golden, CO).
The center of the lake is located at Latitude 270 59' 32" and Longitude 810
36' 25". On this date, the lake surface area was calculated at 180 hectares.


I I 1









Bonny (Polk County)
Florida LAKEWATCH Bathymetric Map


0 200 400 600 800 m


Figure 4. Florida LAKEWATCH personnel generated this map of Lake Bonny by using
differentially corrected global positioning equipment (GPS). Data were
collected March 20, 2003. Scale and map contours are in meters and were
generated using kriging technique in Surfer software package (Golden, CO).
The center of the lake is located at Latitude 280 02' 16" and Longitude 810
55' 29". On this date, the lake surface area was calculated at 108 hectares.








Crooked (Polk County)
Florida LAKEWATCH Bathymetric Map


tN


1000 2000 3000


4000 m


Figure 5. Florida LAKEWATCH personnel generated this map of Crooked Lake by
using differentially corrected global positioning equipment (GPS). Data were
collected August 8, 2001. Scale and map contours are in meters and were
generated using kriging technique in Surfer software package (Golden, CO).
The center of the lake is located at Latitude 270 48' 27 and Longitude 810 34'
42. On this date, the lake surface area was calculated at 1356 hectares.


I I







Disston (Flagler County)
Florida LAKEWATCH Bathymetric Map





tN


0 500 1000 m


Figure 6. Florida LAKEWATCH personnel generated this map of Lake Disston by
using differentially corrected global positioning equipment (GPS). Data were
collected July 1, 1999. Scale and map contours are in meters and were
generated using kriging technique in Surfer software package (Golden, CO).
The center of the lake is located at Latitude 290 17' 2" and Longitude 810 23'
31". On this date, the lake surface area was calculated at 1060 hectares.






































Lake Crooked Lake Disston

Area (ha) Area (ha)

102 402 702 1002 1302 10 120 230 340 450 560 670 780 890 100
0 0 a








aO 7* W 7






Figure 7. Hypsographic curves displaying the change in surface area with hypothetical
declines in water level (m) for each of the four lakes. Florida LAKEWATCH
bathymetric maps were traced using a planimeter to construct the curves.


Lake Annie

Area (ha)

30 50 70 90 110 130 150 170 190



g -
3 -
0>


a 1 7
-c


Lake Bonny

Area (ha)

10 30 50 70 90 110

7


2 -
S4 -
5

n 7










Table 1. Characteristics of lakes and rivers sampled. Surface area (ha) was obtained
from Florida LAKEWATCH (2000, 2001, and 2003). Trophic state was
calculated according to criteria established by Forsberg and Ryding (1980).
Species collected are abbreviated as follows: blc = black crappie;
bg = bluegill; Imb = largemouth bass; rbsu = redbreast sunfish;
resu = redear sunfish; swb = Suwannee bass. Water data sources are
abbreviated as follows: SWFWMD = Southwest Florida Water Management
District; SJRWMD = St. Johns River Water Management District. Sample
location displays the furthest North and South GPS points sampled on the
river. U.S.G.S. Gauge # indicates the United States Geological Survey's
water monitoring station.


County

Polk
Polk
Polk
Flagler


County


Surface
Area (ha)
180
108
1,356
1,060


Trophic State

Eutrophic
Hypereutrophic
Mesotrophic
Mesotrophic


Species Collected

bg, Imb, resu
blc, bg, Imb
blc, bg, Imb, resu
bg, Imb, resu


Sample Location Spes
NorthSpecies Collected
North South


Water Data
Source
SWFWMD
SWFWMD
SWFWMD
SJRWMD

U.S.G.S.
Gauge #


Ochlockonee Gadsden/Leon 30.590N, 30.490N, bg, Imb, rbsu, swb 02339000
84.360W 84.40W
Santa Fe Alachua/Columbia/ 29.830N, 29.850N, 1mb, rbsu, swb 02321975
Gilchrist 82.650W 82.63W
Withlacoochee Hamilton/Madison 30.620N, 30.500N, 1mb, rbsu, swb 02319000
(North) 83.270W 83.24W
Withlacoochee Citrus/Marion 29.010N, 28.980N, bg, 1mb, rbsu 02313000
(South) 82.390W 82.33W


Lake


Annie
Bonny
Crooked
Disston


River











Table 2. Historical lake elevation (m) means, the coefficient of variation
(CV = SD/0* 100%), number of water level observations (N), and ranges
of each year for the four study lakes.


Lake


Annie














Bonny


1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001

1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001


Crooked 1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001


Mean Elevation
(m)

33.2
33.3

34.2
34.4
35.0
35.5
35.1
35.5
35.0
34.3
33.7

40.0
39.6
39.3
39.1
39.6
40.0
39.8
39.9
39.7
39.6
40.0
39.6
38.8
38.1

32.7
32.8
33.5
34.1
34.1
35.0
35.3
35.1
35.9
35.8
35.6
34.9


CV


0.28
0.51

0.19
0.69
0.84
0.17
0.27
0.44
0.37
0.77
0.77

0.13
0.58
0.63
1.14
0.76
0.35
0.57
0.53
0.23
0.91
0.67
0.46
0.93
1.14

0.35
1.21
1.53
0.36
1.04
1.06
0.27
0.50
0.42
0.70
0.68
0.67


Overall Range
(m)

33.0-33.3
33.1-33.5


34.1-
34.1-
34.7-
35.3-
35.0-
35.3-
34.8-
34.0-
33.4-

40.0-
39.3-
38.8-
38.6-
39.2-
39.7-
39.5-
39.6-
39.6-
39.1-
39.6-
39.2-
38.4-
37.5-

32.6-
32.4-
32.9-
33.9-
33.7-
34.5-
35.2-
34.8-
35.6-
35.4-
35.3-
34.6-


34.3
34.8
35.4
35.5
35.3
35.8
35.2
34.8
34.1

40.1
39.9
39.6
39.7
40.0
40.1
40.1
40.3
39.9
40.2
40.4
39.8
39.5
38.9

32.9
33.3
34.1
34.3
34.8
35.5
35.5
35.4
36.2
36.2
36.1
35.2







32


Table 2. Continued.

Mean Elevation Overall Range
Lake Year N CV
(m) (m)

Disston 1992 271 4.03 2.96 3.74-4.31
1993 307 3.95 3.40 3.72-4.24
1994 330 4.06 3.66 3.76-4.55
1995 330 3.98 4.56 3.67-4.43
1996 321 4.08 2.55 3.87-4.32
1997 273 4.06 2.76 3.81-4.40
1998 273 3.97 5.56 3.58-4.36
1999 272 3.88 4.70 3.60-4.23
2000 293 3.87 3.56 3.65-4.14
2001 291 4.01 5.60 3.68-4.93
2002 47 4.05 0.53 4.01-4.08











Table 3. Historical river flow rate (m3/sec) means, coefficient of variation
(CV = SD/0* 100%), number of flow rate observations (N), and ranges of
each year for the four study rivers.

Mean Flow Rate
River Year N Mn Fw Re CV Overall Range (m3/sec)
(m3/sec)

Ochlockonee 1990 365 18.7 162 0.57-151
1991 365 58.9 142 1.67-634
1992 366 28.9 113 2.61-229
1993 365 30.1 169 0.62-289
1994 365 65.5 129 4.79-770
1995 365 25.1 146 1.67-286
1996 366 23.6 111 2.55-145
1997 365 32.7 104 1.39-174
1998 365 41.5 158 0.57-504
1999 365 10.5 102 1.27-44.2
2000 366 7.00 111 0.54-39.4
2001 365 17.8 142 1.22-174
2002 273 8.50 130 0.96-59.8

Santa Fe 1992 92 50.3 103 18.8-239
1993 365 17.2 78.2 2.80-49.8
1994 365 16.6 66.9 4.13-62.6
1995 365 15.9 49.6 4.98-43.6
1996 366 17.6 123 5.52-195
1997 365 18.5 66.7 6.54-87.2
1998 365 37.8 112 14.7-259
1999 365 8.22 48.1 3.74-18.7
2000 366 2.95 48.2 1.16-9.06
2001 365 1.27 115 0.02-7.53
2002 15 0.14 21.6 0.10-0.19

Withlacoochee 1990 365 29.3 150 2.80-188
North 1991 365 124 150 4.56-1062
1992 366 49.4 126 4.76-388
1993 365 54.1 176 2.38-538
1994 365 99.6 93.7 8.24-530
1995 365 47.6 173 3.74-535
1996 366 30.0 139 3.74-238
1997 365 71.3 108 3.94-351
1998 365 89.4 152 4.59-974
1999 365 13.5 118 2.15-80.4
2000 366 16.8 133 2.58-154
2001 273 42.7 132 3.51-283











Table 3. Continued.


Mean Flow Rate
River Year N en Fw Re CV Overall Range (m3/sec)
(m3/sec)

Withlacoochee 1990 365 8.90 39.4 3.70-21.9
South 1991 365 15.0 85.2 3.70-49.6
1992 366 6.50 71.2 1.70-37.1
1993 365 10.3 37.8 5.60-21.4
1994 365 21.1 75.3 4.60-62.0
1995 365 31.6 77.9 8.30-97.1
1996 366 24.8 39.9 11.6-44.7
1997 365 11.6 104 3.10-84.1
1998 365 54.6 87.5 9.40-150
1999 365 10.3 41.6 4.20-27.8
2000 366 3.10 49.2 1.10-7.30
2001 273 4.10 96.8 0.90-24.9











Table 4. Waterbody, species, season and year of sampling. Number of fish aged, total
fish collected (Total), electrofishing effort in minutes (Effort), and minimum
total length collected (Min TL, mm), and maximum total length collected
(Max TL, mm).


Waterbody


Species


Lake Annie Bluegill

Largemouth Bass

Redear Sunfish

Lake Bonny Black Crappie

Bluegill

Largemouth Bass

Crooked Lake Black Crappie

Bluegill

Largemouth Bass

Redear Sunfish

Lake Disston Bluegill

Largemouth Bass

Redear Sunfish

Ochlockonee Bluegill
River
Largemouth Bass

Redbreast Sunfish

Suwannee Bass

Santa Fe River Largemouth Bass

Redbreast Sunfish

Suwannee Bass


Season Year

Winter 2002
2003
2002
2003
2002
2003
Fall 2001
2002
2001
2002
2001
2002
Winter 2002
2003
2002
2003
2002
2003
2002
2003
Winter 2002
2003
2002
2003
2002
2003
Fall 2001
2002
2001
2002
2001
2002
2001
2002
Fall 2001
2002
2001
2002
2001
2002


Aged Total

73 73
55 84
232 376
176 370
125 151
82 141
227 254
188 206
236 1,721
200 2,722
72 72
71 82
27 27
21 21
153 499
142 198
211 408
133 221
30 30
115 125
215 573
233 317
133 169
107 108
52 52
30 30
209 688
150 914
137 206
182 335
169 923
128 2,653
24 24
37 38
149 254
165 256
69 403
57 162
151 275
175 305


Min Max
Effort TL TL
TL TL
400 35 207
360 54 200
400 90 585
360 105 635
400 85 312
360 65 234
940 111 350
810 109 346
940 26 263
810 22 226
940 236 567
810 142 579
520 180 341
580 131 253
520 37 249
580 45 253
520 106 635
580 107 562
520 56 223
580 57 272
820 38 263
900 49 256
820 148 589
900 153 596
820 72 244
900 119 246


760
1,228
760
1,228
760
1,228
760
1,228
810
710
810
710
810
710


37 218
66 242
75 597
57 556
48 228
33 222
80 406
67 392
74 535
48 501
62 244
90 190
75 383
72 383











Table 4. Continued.


Species


Min Max
Season Year Aged Total Effort M
TL TL


Withlacoochee Largemouth Bass
River North
Redbreast Sunfish


Suwannee Bass


Withlacoochee Bluegill
River South
Largemouth Bass


Redbreast Sunfish


Fall 2002


2002

2002


Winter 2002
2003
2002
2003
2002
2003


Sum


137 203 1,220


52 588


198 1,061 420 45 241


85 91 1,220


143 353
109 156
158 350
175 265
102 252
79 97
6.327 19.301


67 386


620 52 230
997 66 242
620 109 546
997 97 629
620 57 212
997 40 201


Waterbody










Table 5. Description of otolith reader agreement across system, species, season, and
year. Discrepancies indicate the number of discrepancies between two
readers, and % Agreement indicates the percent agreement between two
readers for each sample.

Number
Waterbody Species Season Year Aed Discrepancies % Agreement

Lake Annie Bluegill Winter 2002 73 0 100
2003 55 2 96
Largemouth Bass 2002 232 0 100
2003 176 1 99
Redear Sunfish 2002 125 1 99
2003 82 9 89
Lake Bonny Black Crappie Fall 2001 227 30 87
2002 188 11 94
Bluegill 2001 236 5 98
2002 200 17 92
Largemouth Bass 2001 72 4 94
2002 71 3 96
Crooked Lake Black Crappie Winter 2002 27 0 100
2003 21 1 95
Bluegill 2002 153 5 97
2003 142 8 94
Largemouth Bass 2002 211 0 100
2003 133 0 100
Redear Sunfish 2002 30 0 100
2003 115 11 90
Lake Disston Bluegill Winter 2002 215 20 91
2003 233 7 97
Largemouth Bass 2002 133 7 95
2003 107 0 100
Redear Sunfish 2002 52 3 94
2003 30 1 97
Ochlockonee Bluegill Fall 2001 209 2 99
River 2002 150 5 97
Largemouth Bass 2001 137 6 96
2002 182 1 99
Redbreast Sunfish 2001 169 1 99
2002 128 4 97
Suwannee Bass 2001 24 0 100
2002 37 1 97
Santa Fe River Largemouth Bass Fall 2001 149 18 88
2002 165 6 96
Redbreast Sunfish 2001 69 4 94
2002 57 9 84
Suwannee Bass 2001 151 6 96
2002 175 15 91











Table 5. Continued.


Waterbody Species

Withlacoochee Largemouth Bass
River North
Redbreast Sunfish

Suwannee Bass
Withlacoochee Bluegill
River South
Largemouth Bass

Redbreast Sunfish


Sum
Mean


Number
Season Year ged Discrepancies % Agreement
Aged
Fall 2002 137 0 100

2002 198 20 90

2002 85 0 100
Winter 2002 143 4 97
2003 109 3 97
2002 158 4 98
2003 175 0 100
2002 102 8 92
2003 79 1 99

6,327 264
129 5.4 96











Table 6. Catch-curve linear regression equations for each waterbody and species with
r2, P, ages, and N. Ages is the number of fish ages used in each catch curve
analysis used to generate residuals for correlation analyses with hydrological
variables. N is the total number of fish collected by electrofishing used to
generate residuals.


Waterbody


Species


Lake Annie Bluegill
Largemouth Bass


Lake Bonny Black Crappie
Bluegill
Largemouth Bass


Crooked Lake




Lake Disston




Ochlockonee
River



Santa Fe River



Withlacoochee
River North



Withlacoochee
River South


Bluegill
Largemouth Bass
Redear Sunfish


Bluegill
Largemouth Bass
Redear Sunfish


Bluegill
Largemouth Bass
Redbreast Sunfish


Largemouth Bass
Suwannee Bass


Largemouth Bass
Redbreast Sunfish
Suwannee Bass


Bluegill
Largemouth Bass
Redbreast Sunfish


Year Equation
2002 Log (% Freq)
2002 Log (%Freq)


2001 Log (%Freq)
2001 Log (%Freq)
2001 Log (%Freq)


2003 Log (% Freq)
2002 Log (%Freq)
2003 Log (% Freq)


2002 Log (% Freq)
2002 Log (%Freq)
2002 Log (% Freq)


2001 Log (%Freq)
2001 Log (%Freq)
2001 Log (%Freq)


2001 Log (%Freq)
2001 Log (%Freq)


2002 Log (%Freq)
2002 Log (%Freq)
2002 Log (% Freq)


2003 Log (% Freq)
2003 Log (% Freq)
2002 Log (%Freq)


r'
4.39 0.36 x age 0.94
5.36- 0.70 x age 0.97


5.15 0.66
4.97 0.84
5.74 0.61


4.39 -0.42
5.17 0.65
4.07 0.32


5.32 -0.72
4.63 0.48
2.94 0.07


6.39- 1.10
4.58 0.46
7.15 1.40


age 0.79
age 0.82
age 0.98


age 0.91
age 0.88
age 0.83


age 0.91
age 0.87
age 0.20


age 0.88
age 0.79
age 0.86


4.68 0.52 x age 0.83
4.79- 0.55 x age 0.91


3.44 0.16
6.53 1.14
4.37 -0.31


5.48 0.74
4.06 0.33
7.13- 1.38


age 0.50
age 0.94
age 0.46


age 0.95
age 0.79
age 0.95


Ages N
4 66
6 270


P
0.03
0.00


0.02
0.01
0.01


0.00
0.01
0.01


0.00
0.01
0.56


0.06
0.04
0.07


0.03
0.01


0.12
0.04
0.52


0.01
0.01
0.14







40


Table 7. Significant correlation relationships between measures of bluegill, black
crappie, largemouth bass, and redear sunfish year-class strength and stage for
four lakes. Minimum, maximum, and mean stages were measured for the
annual, pre-spawn fall, fall, winter, spring, and summer periods. N indicates
the number of residuals obtained from the catch curve. R indicates whether
the relationship was positively or negatively correlated, r2 indicates the
percent variation explained in the model. No significant relationships were
detected for Lake Annie.


Species

Black Crappie
Largemouth Bass
Largemouth Bass
Largemouth Bass


Crooked Largemouth Bass
Largemouth Bass
Largemouth Bass


Bluegill
Bluegill
Bluegill
Largemouth Bass
Redear Sunfish
Redear Sunfish


Water
Variable
Stage
Stage
Stage
Stage


Stage
Stage
Stage

Stage
Stage
Stage
Stage
Stage
Stage


Period


Pre-Spawn Fall
Fall
Fall
Fall


Pre-Spawn Fall
Pre-Spawn Fall
Pre-Spawn Fall
Annual
Pre-Spawn Fall
Pre-Spawn Fall


Level N R


Max
Min
Max
Mean

Min
Max
Mean

Min
Max
Mean
Min
Mean
Min


-0.74
0.94
0.99
0.99


0.81 0.05 0.08
0.77 0.07 0.23
0.79 0.06 0.17


-0.72
-0.88
-0.75
0.74
0.98
0.99


Water
Body
Bonny


Disston











Table 8. Significant correlations between measures of bluegill, largemouth bass,
redbreast sunfish, and Suwannee bass year-class strength and flow rate for the
four rivers. Levels of minimum, maximum, and mean flow rates were
measured for the annual, pre-spawn fall, fall, winter, spring, and summer
periods. N indicates the number of residuals obtained from the catch curve.
R indicates whether the correlation was positive or negative. P* indicates a
marginally significant relationship.


Waterbody


Ochlockonee
River











Santa Fe River







Withlacoochee
River North





Withlacoochee
River South


Species


Bluegill
Bluegill
Bluegill
Redbreast Sunfish
Redbreast Sunfish
Redbreast Sunfish
Redbreast Sunfish
Redbreast Sunfish
Redbreast Sunfish
Redbreast Sunfish
Redbreast Sunfish

Largemouth Bass
Largemouth Bass
Suwannee Bass
Suwannee Bass
Suwannee Bass
Suwannee Bass

Largemouth Bass
Largemouth Bass
Largemouth Bass
Largemouth Bass
Largemouth Bass

Bluegill
Bluegill
Bluegill
Redbreast Sunfish


Water
Variable
Flow
Flow
Flow
Flow
Flow
Flow
Flow
Flow
Flow
Flow
Flow

Flow
Flow
Flow
Flow
Flow
Flow

Flow
Flow
Flow
Flow
Flow

Flow
Flow
Flow
Flow


Period Level N R

Pre-Spawn Fall Min 4 -0.93 0.07 0.87
Summer Max 4 0.99 0.02 0.97
Fall Min 4 0.94 0.07 0.88
Annual Max 4 0.99 0.00 0.99
Annual Mean 4 0.89 0.11* 0.79
Winter Max 4 0.99 0.00 0.99
Winter Min 4 0.95 0.05 0.91
Winter Mean 4 0.99 0.01 0.98
Summer Min 4 0.92 0.08 0.85
Summer Max 4 0.92 0.08 0.85
Pre-Spawn Fall Max 4 0.98 0.02 0.96


Spring
Spring
Annual
Spring
Spring
Spring

Annual
Fall
Fall
Summer
Summer


Max 5 -0.94 0.02
Mean 5 -0.84 0.08
Mean 5 -0.83 0.08
Min 5 -0.82 0.09
Max 5 -0.86 0.06
Mean 5 -0.96 0.01

Min 6 -0.75 0.09
Max 6 -0.88 0.02
Mean 6 -0.78 0.07
Mean 6 -0.83 0.04
Min 6 -0.74 0.10


0.89
0.71
0.69
0.68
0.75
0.92

0.56
0.78
0.61
0.69
0.55


Spring Max 5 0.82 0.09 0.67
Winter Min 5 0.79 0.11* 0.63
Winter Mean 5 0.80 0.11* 0.64
Pre-Spawn Fall Min 3 0.99 0.03 0.99










Table 9. Significant multiple regression equations predicting year-class strength (i.e.,
residual) from flow rates for the four rivers.


Model r2 DF P-value


Bluegill residual
= -0.2142 + 0.0009 x pre-spawn fall median flow 0.38 7 0.10
Largemouth bass residual
= 0.2438 0.0028 x spring median flow + 0.0009 x winter median flow 0.24 21 0.08
Redbreast sunfish residual
= -0.1967 + 0.0006 x pre-spawn fall median flow 0.31 11 0.06
Suwannee bass residual
= 0.3656 0.0032 x spring median flow 0.38 7 0.10
Lepomis spp. residual
= 0.0570 + 0.0006 x pre-spawn fall median flow 0.0004 x fall median flow 0.47 19 0.00
Micropterus spp. residual
= 0.2502 0.0028 x spring median flow + 0.0008 x winter median flow 0.22 29 0.03















DISCUSSION

In general, year-class strength was more strongly related to system hydrology in

rivers than in lakes. Changes in water flow and stage influence fish community structure

in rivers (Poff and Ward 1989; Grossman et al. 1990; Walker et al. 1995; Richter et al.

1996; Poff et al. 1997). According to Resh et al. (1988) and Power et al. (1995), stream

flow is considered a major variable that affects the abundance and distribution of many

riverine species. Variation in flow/stage was much higher for rivers than for lakes in this

study, and I found stronger relationships between year-class strength and flow/stage in

rivers than in lakes. Thus, rapid and extreme changes in flow/stage in rivers may more

strongly influence fish populations than the relatively minor and less variable stage

changes in lakes. Relationships between year-class strength and hydrology in rivers

appear to be stronger and easier to detect than in lakes.

Results indicated that black bass year-class strength was often negatively related to

spring median flow rates in rivers. For example, combined Suwannee bass residuals

revealed year-class strength was negatively related with spring median flow rates in

rivers. Largemouth bass year-class strength was also negatively related with spring

median flow rates and positively related to winter median flow rates in rivers. Similarly,

combined black bass (M salmoides and M notius) residuals were negatively related with

spring median flow rates and positively related to winter median flow rates in rivers.

Mason et al. (1991) and Sallee et al. (1991) found that smallmouth bass M dolomieu,

year-class strength was negatively correlated with flow rates. Mason et al. (1991)









surmised that high flow events caused low dissolved oxygen concentrations which may

have negatively influenced smallmouth bass year class strength. High flow during spring

may reduce nest success due to turbulence or increased juvenile smallmouth bass

mortality due to displacement (Sallee et al. 1991; Filipek et al. 1991). Filipek et al.

(1991) found that largemouth bass and spotted bass M. punctulatus year-class strength

was negatively correlated with flow rates in rivers. Similarly, largemouth bass and

Suwannee bass exhibited strong year classes in the Santa Fe River during low flow

periods in the spring, similar to smallmouth bass year-class strength in three Virginia

rivers (Scott Smith, VDGIF, Personal Communication). In Florida rivers, periods of low

flow often result in clear water conditions which may increase aquatic macrophyte

abundance, potentially improving habitat for young black basses and providing increased

survival.

Conversely, largemouth bass relative year-class strength was positively correlated

with water levels in three of four lakes. Specifically, fall water levels after the spawn

were positively correlated with largemouth bass residuals from Lakes Bonny and

Crooked, and residuals were positively correlated with annual water levels at Lake

Disston. Lake Bonny electrofishing samples in November 2001 yielded no age-0 and six

age-1 fish (i.e., low mean annual water level in 2000 and 2001). Electrofishing samples

in November 2002 indicated a strong 2002 year class, with 52 out of 71 fish being age-0

(i.e., high summer and fall water level) and no age-1 or age-2 fish were obtained,

suggesting potential year class failure during low water years (2000 and 2001). Personal

observation during electrofishing at Lake Bonny in the fall of 2001 revealed growth of

terrestrial vegetation due to two consecutive years of low water. I surmise that, as a









result of increased rainfall, terrestrial vegetation became inundated in 2002, producing a

strong year-class of largemouth bass. Potential mechanisms for a strong year class could

include increased food resources, increased available spawning habitat and decreased

predation (Bross 1969; Jenkins 1970; Aggus and Elliot 1975; Keith 1975; Shirley and

Andrews 1977; Aggus 1979; Timmons et al. 1980; Miranda et al. 1984; Ploskey 1986;

Meals and Miranda 1991). Thus, I found evidence that largemouth bass year-class

strength was positively related to stage in lakes but negatively related to flow in rivers.

Trends were more evident in rivers as multiple regression models for combined systems

were significant in rivers but no trends were found among lakes.

Recruitment variability among black crappie populations has been characterized as

a "boom or bust" fishery, with a strong year class forming every 3-5 years (Hooe 1991;

Guy and Willis 1995; Allen and Miranda 1998; Maceina and Stimpert 1998). Black

crappie residuals were negatively correlated with pre-spawn fall maximum water levels at

Lake Bonny. The results dispute what has been found with stronger year classes of

crappie Pomoxis spp. being positively correlated with higher water levels in the winter or

spring (Beam 1983; Willis 1986; Miller et al. 1990; Maceina 2003). However, my results

were similar to Maceina and Stimpert (1998), who also found that crappie year-class

strength was influenced by hydrological variables prior to the spawn. Lower water levels

prior to the spawning period concentrates prey items (i.e., higher food availability)

(Jenkins 1970; Aggus 1979) which may provide more favorable conditions for a

successful spawn (Liston and Chubb 1985) due to increased growth (Aggus 1979) and

potentially higher fecundity of adults (Crim and Glebe 1990). Springate et al. (1985)

found that food deprivation reduced the rate of maturation and the total fecundity of









rainbow trout Oncorhynchus mykiss. Furthermore, Ploskey (1986) surmised that

spawning success of largemouth bass did not necessarily insure a strong year class but

increased the likelihood of producing one if environmental conditions were favorable.

Thus, low water levels prior to the spawn may be an indicator of potentially strong year

classes of black crappie in the following spring. However, I caution that this relation was

found only in one lake where I obtained enough black crappie for assessment.

Unlike black bass, combined bluegill and redbreast sunfish residuals were

positively related with pre-fall median flow rates in rivers. Combined Lepomis spp.

(auritus and macrochirus) residuals were positively related with pre-fall median flow

rates and negatively related with fall median flow rates in rivers. Fish yield and

production are strongly related to the extent of accessible floodplain (Junk et al. 1989).

Aggus (1979) indicated that under conditions of an expanding physical environment,

opportunistic species may increase greatly in abundance and biomass. High flow in the

fall before a spawn could increase Lepomis spp. year-class strength in rivers through

increased food availability and possible increases in fecundity (Springate et al. 1985;

Crim and Glebe 1990). Access to the floodplain before the spawn may provide higher

fecundity due to higher densities of invertebrate prey in floodplain habitats (Neckles et al.

1990), which may influence Lepomis spp. year-class strength in rivers. My results were

variable among species and system types, but I detected a positive relationship between

Lepomis spp. (auritus and macrochirus) year-class strength and fall median flow rates in

three rivers combined. Redbreast sunfish may serve as indicator species for setting

MFLs, because I found redbreast sunfish to be positively correlated with high flow rates









in two rivers. In general, high flow rates in the fall prior to spawning was related to

strong-year classes of Lepomis spp. in Florida rivers.

Much is known about reservoirs and the relations of hydrology to sportfish year-

class strength (Bross 1969; Jenkins 1970; Aggus and Elliot 1975; Keith 1975; Shirley and

Andrews 1977; Aggus 1979; Timmons et al. 1980; Miranda et al. 1984; Ploskey 1986;

Meals and Miranda 1991; Ozen and Noble 2002), but little is known about natural lakes

(Tate et al. 2003). Few studies have assessed effects of flow on year-class strength of

Lepomis spp. Meals and Miranda (1991) found that age-0 centrarchid abundance

generally increased with water levels in Mississippi reservoirs. Bluegill and redear

sunfish year-class strength were negatively and positively correlated with fall water level

before the spawn and in spring and summer at Lake Disston, respectively. The lack of

relationships between sunfish species and stage in lakes may result because of protracted

spawning periods (Mettee et al. 1996), which could allow year-class success if

environmental conditions become favorable for some period during the spring or

summer. My sample sizes for redear sunfish may have been too low in lakes, which

could have precluded detecting differences. Thus, no general trends were evident across

lakes for Lepomis spp.

Catch-curve regressions for sportfish revealed relatively constant recruitment (r2)

for many of the lakes. Due to relatively stable recruitment, correlations of residuals with

hydrological variables were often not found. Despite fluctuations in water level, sportfish

populations may have experienced constant recruitment due to lake morphometry (Figure

7). Guy and Willis (1995) suggested that recruitment patterns of black crappie may be

influenced by the morphometry of the system. Lakes with shallow sloping shorelines can









encounter drastic reductions of lake surface area during drought conditions (Hutchinson

1957). Conversely, lakes with steep sloping shorelines can encounter less than

significant reductions of lake surface area during drought conditions (Hutchinson 1957).

For example, a 1 m (meter) water level decrease at Lake Annie results in 17% less

surface area (hectares) (Figure 2). In contrast, surface areas (hectares) of Lakes Bonny,

Crooked, and Disston decrease 6, 9, and 8 % with a 1-m drop in water level. Lake

Annie's surface area varied the most of any lake but no significant correlations were

found between year class strength of sportfish and water levels. Alternately, significant

correlations were found with year-class strength of sportfish and water levels for Lakes

Bonny, Crooked, and Disston despite smaller percent surface area decreases. The

hypsographic curves were generated at stages below the maximum observed water level

during this study, and as a result, surface area (hectares) estimates were up to 4% less

than the actual maximum water level for the four lakes. Nevertheless, I expected year-

class strength of sportfish to vary in lakes with large changes in surface area per change

in water level. This did not occur possibly due to relatively few relationships found

between stage and year-class strength at lakes.

Differences in river morphology may explain why largemouth bass were not

correlated with flow rates on the upper Ochlockonee River but negatively correlated with

flow at the Santa Fe and Withlacoochee North Rivers. Bain et al. (1988) found that

stream species respond differently to variable water levels and have specific microhabitat

preferences. In addition, Talmage et al. (2002) used the habitat variables of percent

overhanging vegetation, percent woody debris, and substrate characteristics to explain up

to 50% of the variability in warmwater stream-fish abundance and species richness.









Although measuring changes in riverine habitat and the extent of floodplain connection

with changes in stage was not an objective of this study, effects of flow and stage on

available habitat should be considered in future studies in Florida rivers.

I found no relation between overall recruitment variability of sportfish indexed with

the RVI and variation in stage or flow for either lakes or rivers. The main limitation in

determining RVI is the proportion of missing year classes within an age-structure sample

(Isermann et al. 2002). Furthermore, as the proportion of missing year classes in a

sample increases, RVI decreases (Isermann et al. 2002). This limitation has proved to be

problematic in our study because most of the sportfish populations had no missing year

classes. The RVI may not be useful in assessing sportfish populations if missing year

classes do not occur, potentially limiting the value of the RVI for these species.

As with any scientific study, violation of assumptions during analysis can be a

problem. I performed multiple catch curve analyses, assuming mortality was constant

(Ricker 1975). In all likelihood, fishing mortality (F) differed among species and

systems. Systems incurring high fishing mortality may demonstrate reduced age groups

in the catch curve analysis (i.e., residuals). Also, F may have been variable among

different ages within catch-curve regressions, which could have influenced my residuals.

Use of residuals from catch-curve regressions to assess year-class strength in fish is best

suited for longer-lived fishes (Maceina 1997). This may explain the low number of

significant relationships, because my sample size and statistical power were low due to

relatively short-lived fish, such as bluegill and redear sunfish.

Previous studies (Schramm and Doerzbacher 1982; Hoyer et al. 1985; Maceina and

Betsill 1987; Crawford et al. 1989; Schramm 1989; Hales and Belk 1992; Mantini et al.









1992; Buckmeier and Howells 2003) have validated the formulation of one annulus per

year for otoliths on all sportfish that I examined, except Suwannee bass. I assumed that

rings on Suwannee bass otoliths were annuli, similar to largemouth bass in the Southeast.

Although I found relations between sportfish year-class strength and hydrological

variables in lakes and rivers, these correlations may not indicate a causal relationship.

Spurious correlations could have resulted in my study, because a relatively large number

of comparisons were made (Jackson and Somers 1991). Similar to other studies (Beam

1983; Willis 1986; Wrenn et al. 1996, Maceina and Stimpert 1998; Maceina 2003), my

results indicated relations between flow/stage and fish year-class strength. Nevertheless,

I caution that some of the relationships may not have been cause and effect. Future

studies should quantitatively assess habitat changes with flow and stage changes to better

elucidate mechanisms between system hydrology and fish recruitment.

There are many factors that can affect electrofishing catch rates including fish

behavior, fish size, fish species, population density, sampling crew, water clarity, water

conductivity, water level, water temperature, and weather conditions (Hardin and Connor

1992; Hilborn and Walters 1992; Reynolds 1996; Bayley and Austen 2002).

Electrofishing for this project took place during two separate hydrological regimes.

Electrofishing in the fall of 2001 and spring of 2002 occurred when water levels and flow

rates were near extreme minimum lows (i.e., drought conditions). Conversely, samples

in the fall of 2002 and spring of 2003 occurred when water levels and flow rates were

near mean or above historic mean values. Electrofishing catch rates of sportfish in the

first year of sampling were much higher than in the second year of sampling for the same

system. As a result, more ages (i.e., year classes) were generally obtained for the catch









curves in the first year of sampling. The results may be due to sampling biases because

my electrofishing equipment was not able to reach deeper areas in both the lakes and

rivers during the second year of sampling (i.e., reduced catchability). Conversely, under

low water conditions (i.e., low stage) sportfish became more concentrated and may not

have been able to avoid the electric field, hence catch rates increased (Pierce et. al 1985).

Nevertheless, strong and weak year-classes were usually corroborated in the catch curves.

Substantial difficulty and disagreement was observed during the aging process of

redbreast sunfish. I surmise that the input of water from natural springs in rivers coupled

with a potential protracted spawning period (Davis 1972; Bass and Hitt 1974; Lukas and

Orth 1993) occasionally produced false annuli (i.e., opaque bands) in redbreast sunfish

otoliths. Hales and Belk (1992) encountered a similar problem with bluegill in a South

Carolina cooling reservoir. However, after discussion with experienced otolith readers

from the FFWCC (i.e., Steve Crawford and Eric Nagid, personal communication), the

faintness and incompleteness of these false annuli distinguished them from true annuli.

Therefore, I believe that redbreast sunfish were aged with adequate accuracy.

Periodic natural fluctuations of water levels and flow rates occur on a regular basis

due to rainfall. Despite these fluctuations, fish assemblages persist over time (Bass 1990;

Paller 1997). Aumen and Gray (1995) supported a more natural hydrologic variability

because they found that sustained high water levels may cause a decline in vegetated

habitats resulting in lower sportfish abundance. Fish species that persist over time may

be long-lived or may produce multiple cohorts within a year (Chesson and Warner 1981).

As a result, once environmental and hydrological conditions are favorable (i.e., the ideal

water level or flow rate), a strong year class is produced (Warner and Chesson 1985).









Factors influencing year-class strength of sportfish can be ambiguous to fisheries

biologists. Fish recruitment is a complex process that involves both density-dependent

and independent factors (Everhart and Youngs 1981; Royce 1984), so I did not expect all

the variation to be explained by hydrological variables. Studies have examined fish

assemblage changes after the impoundment of a river (Quinn and Kwak 2003) or the

institution of a minimum flow (Travnichek et al. 1995; Bowen et al. 1998), but I

attempted to relate year-class strength of six recreationally important sportfish to

hydrology in some of Florida's rivers and lakes. By examining the relationships of

sportfish populations to hydrological influences, I related year-class strength of some

sportfishes in Florida to the water levels and stream flow rates. Although it is easier from

an agency perspective to manage multiple systems with the same hydrological regime,

the variability in results indicate the need for system-specific management, particularly in

lakes (Guy and Willis 1995; Isermann et al. 2002; Sammons et al. 2002) where I found

few among-system trends. These results will possibly help natural resource managers

develop effective MFLs to protect against damage to sportfish populations.















MANAGEMENT IMPLICATIONS FOR SETTING
MINIMUM FLOWS AND LEVELS

Minimum Flows and Levels (MFLs) have been implemented by various methods

and models including the 'Montana Method' (Tennant 1976), the 'Instream Flow

Incremental Methodology' (IFIM) (Bovee 1982), the 'Physical Habitat Simulation'

(PHABISM) (Milhous et al. 1989), and the 'Range Variability Approach' (RVA)

(Richter et al. 1996). The 'Montana Method' simulates flow regimes based on the

average daily discharge or the mean annual flow. The IFIM predicts curves based on

habitat preferences at different flow levels, such that these curves are based on field

sampling of fish locations with associated measurements of habitat conditions (i.e., depth,

velocity). The PHABISM is an intricate component of the IFIM that analyzes habitat

availability for fishes. The RVA approach uses hydrologic variability such as timing,

frequency, duration, and rates of change to sustain and protect natural ecosystem

functions. All models have their advantages and shortcomings, but no particular model

has become the norm for identifying the correct method for setting a MFL.

The methods described above relate specific habitat changes to fish assemblages.

Rather than making predictions of specific habitat changes and their influences on fish

assemblages, I evaluated broad-scale relations between water level/flow and sport fish

year-class strength in both lakes and rivers. My results suggest that among-system

relationships are common for rivers but not for lakes. Thus, MFLs would be more likely

to be regionally applicable in Florida rivers than in lakes. Although I evaluated impacts









to sportfish populations, other portions of the fish assemblage (e.g., threatened or

endangered fishes) were not evaluated and should be considered as part of biological

criteria in setting MFLs.

Currently, MFLs are set by the appropriate Florida Water Management District and

the Florida Department of Environmental Protection according to previous evaluations of

topography, soils, historical ground and surface water data, and vegetation data. Thus

far, biological factors have not been considered when MFLs are set. Criteria used by

establishing MFLs in Florida should include biological factors that prevent ecological

harm. This study provided fish recruitment data that can be used for setting MFLs. Fish

recruitment is one biological factor that should be taken into account when MFLs are set

because freshwater sportfishing in Florida is a one billion dollar industry in Florida

(USFWS 1996). Thus, MFLs should consider impacts to sportfish populations.

Missing year classes occur naturally in fish populations due to adverse

environmental conditions, but sportfish populations seem to be resilient and persist in

multiple system types (Bass 1990; Bayley and Osborne 1993; Paller 1997). Despite

reductions in number and abundance of species, Paller (1997) found the persistence of

multiple species, including bluegill and largemouth bass, following 3.5 years of low

water in a South Carolina reservoir. According to Lowe et al. (1994) and Neubauer et al.

(2003), adverse environmental conditions (e.g., low flow) will not cause significant harm

unless these conditions persist for more than five years. My results indicated that

Lepomis spp. year-class strength was positively related to flow rate, and five years of low

flow rates is probably too much time for these species to persist in a riverine ecosystem

due to their life history. Lepomis spp. are short lived and mortality could eliminate









species if complete year-class failure occurred for five years. Maximum age for redbreast

sunfish collected in this study was age seven, but the majority of the individuals collected

did not exceed age four. Conversely, life history strategies for Micropterus spp. include

being longer lived, with a maximum age of 12 in this study. Thus, low flows for five or

more years may cause weak year classes and reduced abundances of Lepomis spp. in

rivers; potentially impacting sport fisheries. High flows at least once every three years in

the fall may allow inundation of floodplain habitat, producing favorable environmental

conditions for Lepomis spp. Setting MFLs during periods of prolonged drought (i.e.,

three years or more) should consider impacts to short lived species such as Lepomis spp.

I contend that low flows for three or more consecutive years should be prevented, and

thus, MFLs should consider biological impacts to short-lived fishes.

Detecting impacts of flow on year-class strength of sportfish across lakes and rivers

were variable but relationships were easier to detect in rivers. For example, in rivers,

redbreast sunfish year-class strength was positively related to pre-fall median flow rates,

whereas largemouth bass were negatively related to spring median flow rates and

positively related to winter median flow rates. Thus, setting low flows could have

detrimental effects on some fishes but positive on others. Nevertheless, fish have high

fecundity and have adapted to highly variable conditions, particularly in rivers, where

fish tend to persist through short periods of unfavorable conditions (Bayley and Osborne

1993). Thus, extreme high or low flow/stage events occur naturally, but minimum flows

and levels should be set to prevent substantial alteration of this natural variability such as

three or more consecutive years of low levels in lakes and rivers.
















APPENDIX A
INADEQUATE CATCH CURVES

Catch-curve linear regression equations that were not used in correlation analyses
with hydrological variables. A A next to the species (i.e., redear sunfish A) denotes
< 50% agreement in residuals or inadequate sample size collected. Age is the
number of fish ages used in each catch curve analysis. N is the total number of fish
collected by electrofishing used to generate residuals and the associated r2 and P
values.

Waterbody Species Year Equation r2 P Ages N
Lake Annie Bluegill 2003 Log (% Freq) = 7.24 1.45*age 0.99 0.01 3 70
Largemouth Bass 2003 Log (% Freq) = 5.20 0.64*age 0.95 0.01 5 270
Redear Sunfish A 2002 Log (% Freq) = 3.45 0.13*age 0.01 0.96 3 128
2003 Log (% Freq) = 5.82 0.86*age 0.94 0.03 4 118
Lake Bonny Black Crappie 2002 Log (% Freq) = 6.87 0.92*age 0.73 0.15 4 96
Bluegill 2002 Log (% Freq) = 4.17 0.61*age 0.65 0.10 5 520
Largemouth Bass 2002 Low Sample Size -
Crooked Lake Black Crappie A 2002 Low Sample Size
2003 Low Sample Size
Bluegill 2002 Log (% Freq) = 5.83 0.91*age 0.97 0.00 6 428
Largemouth Bass 2003 Log (% Freq) = 4.76 0.52*age 0.72 0.03 6 166
Redear Sunfish 2002 Low Sample Size -
Lake Disston Bluegill 2003 Log (% Freq) = 4.82 0.52*age 0.90 0.00 7 258
Largemouth Bass 2003 Log (% Freq) = 4.62 0.48*age 0.93 0.00 6 93
Redear Sunfish 2003 Low Sample Size -
Ochlockonee Bluegill 2002 Log (% Freq) = 7.77 1.66*age 0.99 0.00 3 401
River Largemouth Bass 2002 Log (% Freq) = 5.02 0.62*age 0.87 0.00 4 119
Redbreast Sunfish 2002 Log (% Freq) = 7.20 1.52*age 0.88 0.22 3 1486
Suwannee Bass A 2001 Low sample size -
2002 Low sample size -
Santa Fe River Largemouth Bass 2002 Log (% Freq) = 4.38 0.43*age 0.57 0.14 5 115
Redbreast Sunfish A 2001 Log (% Freq) = 7.31 1.47*age 0.99 0.00 4 330
2002 Log (% Freq) = 6.33 1.06*age 0.93 0.04 4 155
Suwannee Bass 2002 Log (% Freq) = 5.16 0.63*age 0.93 0.01 5 213
Withlacoochee Bluegill 2002 Log (% Freq) = 6.70 1.18*age 0.91 0.05 4 281
River (South) Largemouth Bass 2002 Log (% Freq)= 5.68 -0.81*age 0.83 0.03 5 243
Redbreast Sunfish 2003 Log (% Freq) = 4.91 0.53*age 0.46 0.53 3 59
















APPENDIX B
AGE FREQUENCIES

Age-frequency distributions and recruitment variability index (RVI) values for each
species, system, and year. An asterisk next to the year (i.e., 2001*) denotes the
sample used for relation to the hydrological variables. A A next to the species (i.e.,
rbsuA) denotes < 50% agreement in residuals or inadequate sample size, therefore
the sample was not used in any correlation analyses. Samples with less than 3 fish
obtained of a particular age were used to index RVI values but were not included in
catch-curve analyses. In addition, year-classes were removed from the catch-
curves and RVI index, if they were below the assigned catchable size. Species
abbreviations are as follows: bg = bluegill; blc = black crappie; Imb = largemouth
bass; rbsu = redbreast sunfish; resu = redear sunfish; swb = Suwannee bass.


Species Year RVI Ag
0 1 2 3 4 5 6
bg 2002* 0.73 9 25 18 15 8
bg 2003 0.91 12 54 13 3
Imb 2002* 0.86 101 119 83 32 25 7
Imb 2003 0.85 92 120 75 39 28 8
resu^ 2002 0.73 0 37 53 38 2 2
resu 2003 0.72 5 56 45 12 5 2
blc 2001* 0.77 19 11 59 52 89 14 4
blc 2002 0.75 57 29 23 48 19 27 2
bg 2001* 0.86 763 327 130 206 113 9 5
bg 2002 0.87 2398 369 31 64 41 15
Imb 2001* 0.77 0 6 12 23 14 6 4
Imb 2002 0.43 52 0 0 3 4 3 3
blcA 2002 0.61 1 3 8 10 1 1
blc 2003 0.29 2 0 4 6 4 0
bg 2002 0.70 48 53 35 17 13 17
bg 2003* 0.91 52 240 140 23 16 6
Imb 2002* 0.82 86 93 114 74 19 7
Imb 2003 0.77 50 31 52 46 27 7
resu 2002 0.78 4 8 10 5 1 1
resu 2003* 0.65 19 26 29 16 14 17
bg 2002* 0.87 88 173 114 46 25 5
bg 2003 0.81 25 68 67 59 37 17
Imb 2002* 0.81 54 46 31 10 12 10
Imb 2003 0.83 5 32 27 19 6 5
resu 2002* 0.66 11 10 7 11 11 2
resu 2003 0.53 2 6 8 5 1 6


7 8 9 10 11 12


1
1
0 1





1
1 1
0 1
3
4 0 1
1
2
1

4 0 1
5
4 2 1
1
1 1 1


Water
Body
Lake
Annie




Lake
Bonny




Crooked
Lake







Lake Disston











Appendix B. Continued.


Species Year RVI
0 1 2 3 4


Water
Body
Ochlockonee
River







Santa Fe River


2001* 0.68 0 447 109 100 10
2002 0.94 167 287 325 63 16
2001* 0.30 87 54 19 25 9
2002 0.86 122 85 57 20 31
2001* 0.89 0 468 179 233 18
2002 0.72 456 623 1313 110 63
2001 0.36 4 4 2 2 1
2002 0.48 9 10 6 3 3
2001* 0.88 29 53 100 31 10
2002 0.80 84 55 40 48 7
2001 0.91 0 70 249 63 15
2002 0.93 6 76 64 11 4
2001* 0.70 25 49 123 28 19
2002 0.85 27 61 82 75 36


1mb 2002* 0.55 108 28 11 15 5
rbsu 2002* 0.90 205 149 369 163 68
swb 2002* 0.33 31 18 13 17 7


bg
bg
Imb
Imb
rbsu
rbsu


2002 0.88
2003* 0.83
2002 0.88
2003* 0.55
2002* 0.90
2003 0.74


67 164 74 39
17 66 41 14
104 116 69 23
123 24 45 24
34 141 61 9
37 20 32 7


Age
5 6
6
2
3 2
11 2
4 1
0 2
2 1
2 3
22 1
9 11
31 1
1
20 1
11 9
4 7
7 6
0 0
4 2
12 3
32 3
21 12
2


7 8 9 10 11 12


0 0 1
1


0 0 1


bg
bg
Imb
Imb
rbsu
rbsu
swb^
swb
Imb
Imb
rbsu^
rbsu
swb
swb


0 1


3 1 0 1
2
1 4 0 0 1


Withlacoochee
River (North)

Withlacoochee
River (South)
















LIST OF REFERENCES


Aggus, L. R. 1979. Effects of freshwater fish predator- prey dynamics. Pages 47-56 in
H. Clepper, editor. Predator-Prey Systems in Fisheries Management. Sport
Fishing Institute, Washington, D.C., USA.

Aggus, L. R., and G. V. Elliot. 1975. Effects of cover and food on year-class strength of
largemouth bass. Pages 317-322 in R. H. Stroud and H. Clepper, editors. Black
Bass Biology and Management. Sport Fishing Institute, Washington, D. C., USA.

Allen, M. S., and L. E. Miranda. 1998. An age-structured model for crappie fisheries.
Ecological Modeling 107:289-303.

Anderson, C. S. 1995. Measuring and correcting for size selection in electrofishing
mark-recapture experiments. Transactions of the American Fisheries Society
124:663-676.

Aumen, N. G., and S. Gray. 1995. Research synthesis and management
recommendations from a five-year ecosystem-level study of Lake Okeechobee,
Florida (USA). Archiv fur Hydrobiologie 45:343-356.

Bain, M. B., J. T. Finn, and H. E. Booke. 1988. Streamflow regulation and fish
community structure. Ecology 69:382-392.

Bass, D. G. 1990. Stability and persistence offish assemblages in the Escambia River,
Florida. Rivers 1:296-306.

Bass, D. G., and V. G. Hitt. 1974. Ecological aspects of the redbreast sunfish, Lepomis
auritus, in Florida. Proceedings of the Annual Conference of the Southeastern
Association of Fish and Wildlife Agencies 28:296-307.

Bayley, P. B., and D. J. Austen. 2002. Capture efficiency of a boat electrofisher.
Transactions of the American Fisheries Society 131:435-451.

Bayley, P. B., and L. L. Osborne. 1993. Natural rehabilitation of stream fish populations
in an Illinois catchment. Freshwater Biology 29:295-300.

Beam, J. H. 1983. The effect of annual water level management on population trends of
white crappie in Elk City Reservoir, Kansas. North American Journal of Fisheries
Management 3:34-40.









Bettoli, P. W., and L. E. Miranda. 2001. Cautionary note about estimating mean length
at age with subsampled data. North American Journal of Fisheries Management
21:425-428.

Bettoli, P. W., M. J. Maceina, R. L. Noble, and R. K. Betsill. 1993. Response of a
reservoir fish community to aquatic vegetation removal. North American Journal
of Fisheries Management 13:110-124.

Bovee, K. D. 1982. A guide to stream habitat analysis using the instream flow
incremental methodology. Instream flow information paper no. 12. U.S. Fish and
Wildlife Service, Washington, D. C., USA.

Bowen, Z. H., M. C. Freeman., and K. D. Bovee. 1998. Evaluation of generalized
habitat criteria for assessing impacts of altered flow regimes on warmwater fishes.
Transactions of the American Fisheries Society 127:455-468.

Bross, M. G. 1969. Fish samples and year-class strength (1965-1967) from Canton
Reservoir. Proceedings of the Oklahoma Academy of Science 48:194-199.

Buckmeier, D. L., and R. G. Howells. 2003. Validation of otoliths for estimating ages of
largemouth Bass to 16 years. North American Journal of Fisheries Management
23:590-593.

Carlander, K. D. 1977. Handbook of freshwater fish biology, Volume 2. Iowa State
University Press, Ames, Iowa, USA.

Chesson, P. L., and R. R. Warner. 1981. Environmental variability promotes existence
in lottery competitive systems. The American Naturalist 117:923-943.

Conover, W. J. 1980. Practical nonparametric statistics. Second Edition, John Wiley
and Sons, New York, New York, USA.

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.

Crim, L.W., and B. D. Glebe. 1990. Reproduction. Pages 529-553 in C. B. Schreck and
P. B. Moyle, editors. Methods of fish biology. American Fisheries Society,
Bethesda, Maryland.

Davis, J. R. 1972. The spawning behavior, fecundity rates, and food habits of the
redbreast sunfish in southeastern North Carolina. Proceedings of the Annual
Conference of the Southeastern Association of Fish and Wildlife Agencies
25:556-560.

Everhart, W. H., and W. D. Youngs. 1981. Principles of Fisheries Science. Second
Edition, Comell University Press, London.









Filipek, S. P., M. A. Armstrong, and L. G. Claybrook. 1991. Effects of a hundred-year
flood on the smallmouth bass population of the Upper Caddo River, Arkansas.
Pages 84-89 in D. C. Jackson, editor. The First International Smallmouth Bass
Symposium. Mississippi Agricultural and Forestry Experiment Station, Mississippi
State University, Starkville, Mississippi, USA.

Florida LAKEWATCH. 2000. Florida LAKEWATCH Data 1986-1999. Department of
Fisheries and Aquatic Sciences, University of Florida/Institute of Food and
Agricultural Sciences. Library, University of Florida. Gainesville, Florida, USA.

Forsberg, C., and S. O. Ryding. 1980. Eutrophication parameters and trophic state
indices in 30 Swedish waste-receiving lakes. Archives fur Hydrobiologie 88:
189-207.

Grossman, G. D., J. F. Dowd, and M. Crawford. 1990. Assemblage stability in stream
fishes: review. Environmental Management 14:661-671.

Guy, C. S., and D. W. Willis. 1995. Population characteristics of black crappies in South
Dakota Waters: A case for ecosystems-specific management. North American
Journal of Fisheries Management 15:754-765.

Hales, L. S., and M. C. Belk. 1992. Validation of otolith annuli of bluegills in a
southeastern thermal reservoir. Transactions of the American Fisheries Society
121:823-830.

Hardin, S., and L. L. Connor. 1992. Variability of electrofishing crew efficiency and
sampling requirements for estimating reliable catch rates. North American Journal
of Fisheries Management. 12:612-617.

Hilborn, R., and C. J. Walters. 1992. Quantitative fisheries stock assessment: choice,
dynamics, and uncertainty. Chapman and Hall, New York.

Hooe, M. L. 1991. Crappie biology and management. North American Journal of
Fisheries Management 11:483-484.

Houde, E. D. 1987. Fish early life dynamics and recruitment variability. American
Fisheries Society Symposium 2:17-29.

Hoyer, M. V., and D. E. Canfield. 1996. Largemouth bass abundance and aquatic
vegetation in Florida Lakes: an empirical analysis. Journal of Aquatic Plant
Management 34:23-32.

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.









Hutchinson, G. E. 1957. Hydrological cycle and water balance. Pages 221-249 in J.
Wiley and Sons, editors. A treatise on limnology, Volume 1:geography, physics
and chemistry, New York, New York, USA.

Isermann, D. A., W.L. McKibbin, and D. W. Willis. 2002. An analysis of methods for
quantifying crappie recruitment variability. North American Journal of Fisheries
Management 22:1124-1135.

Jackson, D. A., and K. M. Somers. 1991. The spectre of "spurious correlations."
Oecologia 86:147-151.

Jenkins, R. M. 1970. The influence of engineering design and operation and other
environmental factors on reservoir fishery resources. Water Resources Bulletin
6:110-119.

Junk, W. J., P. B. Bayley, and R. E. Sparks. 1989. The flood pulse concept in river-
floodplain systems. Proceedings of the International Large River Symposium
(LARS). Canadian Special Publications in Fisheries and Aquatic Sciences
106:110-127.

Keith, W. E. 1975. Management by water level manipulation. Pages 489-497 in H.
Clepper, editor, Black bass biology and management. Sport Fishing Institute,
Washington, D. C., USA.

Kriksunov, Ye. A., and K. M. Mamina. 1995. Effects of flows in the Ural River on
recruitment of stellate sturgeon, Acipensar stellatus. Journal of Ichthyology
34:662-665, 35:52-58.

Liston, C. R., and S. Chubb. 1985. Relationships of water level fluctuations and fish.
Pages 121-140 in H. H. Prince and F. D. D'Itri. Coastal Wetlands. Lewis
Publishers, Chelsea, Michigan.

Lowe, E. F., S. Mortellaro, R. Mattson, S. Flannery, and B. Epting. 1994. Final Report,
District Water Management Plan-Conventions Subcommittee on Impacts to Natural
Systems. Florida Water Management Districts and Florida Department of
Environmental Regulation Joint Subcommittee Report, Palatka, Florida, USA.

Lukas, J. A., and D. J. Orth. 1993. Reproductive ecology of redbreast sunfish Lepomis
auritus in a Virginia stream. Journal of Freshwater Ecology 8:235-244.

Maceina, M. J. 2003. Verification of the influence of hydrologic factors on crappie
recruitment in Alabama reservoirs. North American Journal of Fisheries
Management 23:470-480.

Maceina, M. J. 1997. Simple application of using residuals from catch-curve regressions
to assess year-class strength in fish. Fisheries Research 32:115-121.









Maceina, M. J., and R. K. Betsill. 1987. Verification and use of whole otoliths to age
white crappie. Pages 267-278 in R. C. Summerfelt and G. E. Hall, editors. The age
and growth of fish. The Iowa State University Press, Ames, Iowa, USA.

Maceina, M. J., and P. W. Bettoli. 1998. Variation in largemouth bass recruitment in
four mainstream impoundments of the Tennessee River. North American Journal
of Fisheries Management 18:998-1003.

Maceina, M. J., and M. R. Stimpert. 1998. Relations between reservoir hydrology and
crappie recruitment in Alabama. North American Journal of Fisheries Management
18:104-113.

Mantini, L., M. V. Hoyer, J. V. Shireman, and D. E. Canfield. 1992. Annulus validation,
time of formation, and mean length at age of three sunfish species in north central
Florida. Proceedings of the Annual Conference of Southeast Association of Fish
and Wildlife Agencies 46:357-367.

Mason, J. W., D. J. Graczyk, and R. A. Kerr. 1991. Effects of runoff on smallmouth
bass populations in four southwestern Wisconsin streams. Pages 28-35 in D. C.
Jackson, editor. The First International Smallmouth Bass Symposium. Mississippi
Agricultural and Forestry Experiment Station, Mississippi State University,
Starkville, Mississippi, USA.

Meals, K. 0., and L. E. Miranda. 1991. Variability in abundance of age-0 centrarchids
among littoral habitats of flood control reservoirs in Mississippi. North American
Journal of Fisheries Management 11:298-304.

Mettee, M. F., P. E. O'Neil, and J. M. Pierson. 1996. Fishes of Alabama and the Mobile
basin. Oxmoor House, Birmingham, Alabama, USA.

Miller, S. J., D. D. Fox, L. A. Bull, and T. D. McCall 1990. Population dynamics of
black crappie in Lake Okeechobee, Florida, following suspension of commercial
harvest. North American Journal of Fisheries Management 10:98-105.

Milhous, R. T., M. A. Updike, and D. M. Schneider. 1989. Physical habitat simulation
system reference manual, version 2. Instream flow information paper 26. U.S.
Fish and Wildlife Service Biological Report 89(16).

Miranda, L. E., W. L. Shelton, and T. D. Bryce. 1984. Effects of water level
manipulation on abundance, mortality, and growth of young-of-year largemouth
bass in West Point Reservoir, Alabama-Georgia. North American Journal of
Fisheries Management 4:314-320.

Mitzner, L. R. 1995. Effect of environmental factors and harvest regulations upon the
crappie (Pomoxis) sportfishery at Rathbun Lake. Technical Bulletin No. 5. Iowa
Department of Natural Resources, Ames, Iowa, USA.









Myers, R. H. 1990. Classical and modern regression with applications. PWS-Kent,
Boston, Massachusetts, USA.

Neckles, H. A., H. R. Murkin, and J. A. Cooper. 1990. Influences of seasonal flooding
on macroinvertebrate abundance in wetland habitats. Freshwater Biology 23: 311-
322.

Neubauer, C. P., G. B. Hall, E. F. Lowe, R. B. Hupalo, and L. W. Keenan. 2003. The
multiple minimum flows and levels method of the St. Johns River Water
Management District. St. Johns River Water Management District, Palatka,
Florida, USA.

Ozen, 0., and R. L. Noble. 2002. Relationship between water level fluctuations and
largemouth bass spawning in a Puerto Rico Reservoir. Pages 213-220 in D. P.
Philip and M. S. Ridgway, editors. Black Bass: ecology, conservation, and
management. American Fisheries Society, Bethesda, Maryland, USA.

Paller, M. H. 1997. Recovery of a reservoir fish community from drawdown related
impacts. North American Journal of Fisheries Management 17:726-733.

Peterman, R. M. 1990. Statistical power analysis can improve fisheries research and
management. Canadian Journal of Fisheries and Aquatic Sciences 47:2-15.

Petts, G. E. 1996. Water allocation to protect river ecosystems. Regulated Rivers:
Research and Management 12:353-365.

Pierce, R. B., D. W. Coble, and S. D. Corley. 1985. Influence of river stage on shoreline
electrofishing catches in the upper Mississippi River. Transactions of the
American Fisheries Society 114:857-860.

Ploskey, G. R. 1986. Effects of water-level changes on reservoir ecosystems, with
implications for fisheries management. Pages 86-97 in G. E. Hall and M. J. Van
Den Avyle, editors. Reservoir Fisheries Management: Strategies for the 80's.
American Fisheries Society, Southern Division Reservoir Committee, Bethesda,
Maryland.

Poff, N. L., and J. V. Ward. 1989. Implications of streamflow variability and
predictability for lotic community structure: a regional analysis of streamflow
patterns. Canadian Journal of Fisheries and Aquatic Sciences 46:1805-1818.

Poff, N. L., and seven coauthors. 1997. The natural flow regime: A paradigm for river
conservation and restoration. BioScience 47:769-784.

Pope, K. L., and D. W. Willis. 1996. Seasonal influences on freshwater fisheries
sampling data. Reviews in Fisheries Science 4:57-73.









Power, M. E., A. Sun, M. Parker, W. E. Dietrich, and J. T. Wooten. 1995. Hydraulic
food-chain models: an approach to the study of food web dynamics in large rivers.
Bioscience 45: 159-167.

Quinn, J. W., and T. J. Kwak. 2003. Fish assemblage changes in an Ozark River after
impoundment: A long-term perspective. Transactions of the American Fisheries
Society 132:110-119

Raibley, P. T., T. M. O'Hara, K. S. Irons, K. D. Blodgett, and R. E. Sparks. 1997.
Largemouth bass size distributions under varying annual hydrological regimes in
the Illinois River. Transactions of the American Fisheries Society 126: 850-856.

Reice, S. R., R. C. Wissmar, and R. J. Najman. 1990. Disturbance regimes, resilience,
and recovery of animal communities and habitats in lotic systems. Environmental
Management 14:647-659.

Resh, V. H., and nine coauthors. 1988. The role of disturbance in stream ecology.
Journal of the North American Benthological Society 7:433-455.

Reynolds, J. B. 1996. Electrofishing. Pages 221-253 in B. R. Murphy and D. W. Willis,
editors. Fisheries Techniques, 2nd edition. American Fisheries Society, Bethesda,
Maryland, USA.

Richter, B. D., J. V. Baumgartner, J. Powell, and D. P. Braun. 1996. A method for
assessing hydrologic alteration with ecosystems. Conservation Biology 10:1163-
1174.

Ricker, W. E. 1975. Computation and interpretation of biological statistics of fish
populations. Bulletin 191, Fisheries Research Board of Canada, Ottawa.

Royce, W. F. 1996. Introduction to the practice of fishery science. Academic Press, San
Diego, California, USA.

Sallee, R. D., J. L. Langbein, H. Brown, and J. Ferencak. 1991. Effects of discharge
fluctuations on survival of smallmouth bass in the Kankakee River, Illinois. Pages
90-95 in D. C. Jackson, editor. The First International Smallmouth Bass
Symposium. Mississippi Agricultural and Forestry Experiment Station, Mississippi
State University, Starkville, Mississippi, USA.

Sammons, S. M., P. W. Bettoli, D. A. Isermann, and T. N. Churchill. 2002.
Recruitment variation of crappies in response to hydrology of Tennessee reservoirs.
North American Journal of Fisheries Management 22:1393-1398.

SAS. 2000. User's guide, version 8. SAS Institute Inc., Cary, North Carolina, USA.

Schramm, H. J., and J. F. Doerzbacher. 1982. Use of otoliths to age black crappie from
Florida Proceedings of the Annual Conference of Southeast Association of Fish and
Wildlife Agencies 36:95-105.









Schramm, H. J. 1989. Formation of annuli in otoliths of bluegills. Transactions of the
American Fisheries Society 118:546-555.

Shireman, J. V., M. V. Hoyer, M. J. Maceina, and D. E. Canfield. 1985. The water
quality of Lake Baldwin, Florida: Four years after vegetation removal by grass
carp. Proceedings of the Fourth Annual Conference and International Symposium,
North American Lake Management Society 4:201-206.

Shirley, K. E., and A. K. Andrews. 1977. Growth, reproduction, and mortality of
largemouth bass during the first year of life in Lake Carl Blackwell, Oklahoma.
Transactions of the American Fisheries Society 106:590-595.

Sigler, W. F., and J. W. Sigler. 1990. Recreational Fisheries: management, theory, and
application. University of Nevada Press, Reno, Nevada, USA.

St. Johns River Water Management District (SJRWMD). 2001. St. Johns River Water
Management District: minimum flows and levels. (Available on the World Wide
Web at http://sjr.state.fl.us).

Smith, S. M., and D. J. Orth. 1990. Distributions of largemouth bass in relation to
submerged aquatic vegetation in Flat Top Lake, West Virginia. Proceedings of the
Annual Conference of the Southeastern Association of Fish and Wildlife Agencies
44:36-44.

Springate, J. R. C., N. R. Bromage, and P. R. T. Cumaranatunga. 1985. The effects of
different ration on fecundity and egg quality in the rainbow trout (Salmo gairdneri).
Pages 371-393. in C. Cowey, editor. Feeding and nutrition in fish. Academic
Press, London.

Talmage, P. J., J. A. Perry, and R. M. Goldstein. 2002. Relation of instream habitat and
physical conditions to fish communities of agricultural streams in the northern
Midwest. North American Journal of Fisheries Management 22:825-833.

Tate, W. B., M. S. Allen, R. A. Myers, E. Nagid, and J. Estes. 2003. Relation of age-0
largemouth bass abundance to hydrilla coverage and water level at Lochloosa and
Orange Lakes, Florida. North American Journal of Fisheries Management. 23:251-
257.

Taubert, 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.

Tennant. D. L. 1976. Instream flow regimens for fish wildlife, recreation and related
environmental resources. Pages 359-373 in J. F. Orsborn and C. H. Allman,
editors. Instream Flow Needs. American Fisheries Society, Bethesda, Maryland,
USA.









Timmons, T. J., W. L. Shelton, and W. D. Davies. 1980. Differential growth of
largemouth bass in West Point Reservoir, Alabama-Georgia. Transactions of the
American Fisheries Society 109:176-186.

Travnichek, V. H., M. B. Bain, and M. J. Maceina. 1995. Recovery of a warmwater fish
assemblage after initiation of a minimum-flows release downstream from a
hydroelectric dam. Transactions of the American Fisheries Society 124:836-844.

U.S. Department of the Interior, Fish and Wildlife Service. 1996. National Survey of
Fishing, Hunting, and Wildlife Associated Recreation, Washington, D.C., USA

Walker, K. F., F. Sheldon, and J. T. Puckridge. 1995. A perspective on dryland river
ecosystems. Regulated Rivers: Research and Management 11:85-104.

Warner, R. R., and P. L. Chesson. 1985. Coexistence mediated by recruitment
fluctuations: a field guide to the storage effect. The American Naturalist 125:769-
787.

Willis, D. W. 1986. Review of water level management on Kansas reservoirs, Pages
110-114 in G. E. Hall and M. J. Van Den Avyle, editors. Reservoir Fisheries
Management: Strategies for the 80's. American Fisheries Society, Southern
Division Reservoir Committee, Bethesda, Maryland, USA.

Willis, D. W., and B. R. Murphy. 1996. Planning for sampling. Pages 1-15 in B. R.
Murphy and D. W. Willis, editors. Fisheries Techniques, 2nd edition. American
Fisheries Society, Bethesda, Maryland, USA.

Wrenn, W. B., D. R. Lowery, M. J. Maceina, and W. C. Reeves. 1996. Relationships
between largemouth bass and aquatic plants in Guntersville Reservoir, Alabama.
Multidimensional approaches to reservoir fisheries management. American
Fisheries Society, Special Publication 16:382-393.















BIOGRAPHICAL SKETCH

Timothy Frederick Bonvechio was born October 29, 1977, in West Palm Beach,

Florida, the son of Robert Bonvechio, Jr. and Constance Bonvechio. He graduated from

Palm Beach Gardens Community High School in 1996. He received his Associate of

Arts degree in Biology from Palm Beach Community College in May of 1998. He

received his Bachelor of Science degree in Forest Resources and Conservation from the

University of Florida in December of 2000. While finishing undergraduate studies, he

gained part-time employment in May of 2000, as a freshwater fisheries technician under

the direction of Dr. Mike Allen. The day after Tim received his Bachelor of Science

degree, Dr. Allen recruited him on as a full-time technician. During his tenure as

technician, he gained valuable experience in fisheries management. As a result, he began

his graduate work at the University of Florida in the spring of 2002 to pursue a Master of

Science degree in fisheries management. After graduation in December 2003, he plans

on pursuing a career in fisheries management as a state freshwater fisheries biologist.

Upon departure, he leaves an old saying which has no author, "When God created

the earth, He made it two-thirds water and one-third land, surely he meant for man to

spend more time fishing than plowing."