Citation
Replacement of gizzard shad (Dorosoma cepedianum) by blue tilapia (Tilapia aurea) as a potential biomanipulation agent in Florida eutrophic lakes

Material Information

Title:
Replacement of gizzard shad (Dorosoma cepedianum) by blue tilapia (Tilapia aurea) as a potential biomanipulation agent in Florida eutrophic lakes
Creator:
Fernandes, Carlos A, 1950-
Publication Date:
Language:
English
Physical Description:
ix, 165 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Feces ( jstor )
Fish ( jstor )
Gizzard ( jstor )
Lakes ( jstor )
Nitrogen ( jstor )
Phosphorus ( jstor )
Phytoplankton ( jstor )
Shad ( jstor )
Tilapia ( jstor )
Zooplankton ( jstor )
Dissertations, Academic -- Environmental Engineering Sciences -- UF
Environmental Engineering Sciences thesis Ph. D
Eutrophication -- Florida ( lcsh )
Gizzard shad -- Florida ( lcsh )
Lakes -- Florida ( lcsh )
Lake Bonny ( local )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1995.
Bibliography:
Includes bibliographical references (leaves 142-156)
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Carlos A. Fernandes.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
021857803 ( ALEPH )
33472897 ( OCLC )

Downloads

This item has the following downloads:


Full Text











REPLACEMENT OF GIZZARD SHAD (Dorosoma cepedianum) BY BLUE TILAPIA (TilaPia aurea) AS A POTENTIAL BIOMANIPULATION AGENT
IN FLORIDA EUTROPHIC LAKES


















By

CARLOS A. FERNANDES


















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1995














ACKNOWLEDGMENTS



I wish to thank Dr. Thomas L. Crisman, chair of my committee, for his friendship and encouraging words of advice. I also wish to acknowledge the other members of my committee, Dr. G. Ronnie Best, Dr. Frank G. Nordlie, Dr. Edward J. Phlips, and Dr. Horst 0. Schwassmann, whose critical reviews and comments resulted in the final manuscript. Dr. Edward J. Phlips provided laboratory support for both bioassays.

Mr. Eugene Medley provided field and laboratory support in the City of Lakeland. His assistance and friendship will always be appreciated. Mr. Bal C. Sukhraj was an important asset in my field work in Lakeland. Many thanks go to Dr. John Moore and staff of the Animal Nutrition Laboratory, University of Florida, for their contribution to the fish feces analyses. Mr. John Funk assisted with the fish feces analyses. Dr. Daniel Canfield generously offered the use of an electrofishing boat. Dr. Francisco Zimmermann helped with the statistical analyses.

Throughout this work, I have benefitted from a

productive working relationship with staff members from the Florida Game and Fresh Water Fish Commission in Eustis, ii








and the Polk County Water Resources Division. Their field and laboratory support was invaluable.

I particularly wish to thank my parents, Mr. A.P.

Fernandes (deceased) and Mrs. M.D. Fernandes, for teaching me how to walk through life, maintaining the necessary endurance without loosing sensitivity and appreciation for nature's creations. My sons, Carlos Filho and Diego, deserve special recognition for their positive attitude, and their confidence throughout my work. They also provided loving field support. Special recognition goes to Margaret, my fiancee, friend and fellow biologist, for her important field and laboratory support during the bioassay. Most of all, their love and caring support gave the perseverance needed to complete this work.

This study was funded by CAPES--FundaCgo Coordenag&o de Aperfeigoamento de Pessoal de Nivel Superior of the Ministry of Education of Brazil.



















iii















TABLE OF CONTENTS



ACKNOWLEDGEMENTS . . . . . . . . . ii

ABSTRACT . . . . . . . . . . . vi

CHAPTER 1 INTRODUCTION . . . . . . . . 1

CHAPTER 2 LITERATURE REVIEW . . . . . . . 7

Introduction . . . . . . . . 7
Biomanipulation . . . . . . . 7
Shad Ecological Energetics . . . . 13 Tilapia Ecological Energetics . . . . 16

CHAPTER 3 RELATIVE IMPORTANCE OF BLUE TILAPIA AND
GIZZARD SHAD TO LAKE SEDIMENT AND WATER COLUMN
NUTRIENT CONCENTRATION--ANALYSIS OF FISH FECES . 22

Introduction . . . . . .. .. 22
Methods . . . . . . . . . 27
Results and Discussion . . . . . 31
Conclusions . . . . . . ... . 39

CHAPTER 4 SEDIMENTS AND HISTORICAL ECOLOGY OF TWO
CENTRAL FLORIDA LAKES . . . . . . . 53

Introduction ........ ......... 53
Study Sites . . . . ... . . 56
Methods . . . . . . .. . . 59
Results and Discussion . . . .... 60
Conclusions . . . . . . . . 78

CHAPTER 5 LIMNOLOGICAL ASSESSMENT . . .. . . 89

Introduction . . ...... . . 89
Methods . . .. .. . ... . . . 91
Description of Study Area . . . . . 93 Summary . . * ................ . 129






iv










CHAPTER 6 SUMMARY . . . . . . .... 133

Fish Feces Experiment . . .. . . 134
Paleolimnological Analysis . . . . 137 Lakes Assessment .......... ... 139
Conclusion . . . . . .. . 140

LITERATURE CITED ..... ... .......... 142

APPENDICES . . . . . . .. .. . 157

BIOGRAPHICAL SKETCH ......... . .. . .. 165













































v














Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

REPLACEMENT OF GIZZARD SHAD (Dorosoma cepedianum) BY BLUE TILAPIA (Tilapia aurea) AS A POTENTIAL BIOMANIPULATION AGENT
IN FLORIDA EUTROPHIC LAKES

By

Carlos A. Fernandes

May 1995

Chairperson: Dr. Thomas L. Crisman Major Department: Environmental Engineering Sciences
The effects of replacement of gizzard shad (Dorosoma cepedianum) as a dominant filter feeding fish by blue tilapia (Tilapia aurea) in six eutrophic Florida lakes were analyzed using the lakes' historical limnological characteristics and sedimentary records. The digestive physiology of gizzard shad and blue tilapia (with emphasis on fecal material) was studied in order to evaluate the relative impact of the feces produced by different fish species.

Five of the six lakes studied are classified as hypereutrophic systems. Lake Gibson is classified as eutrophic. All of the study sites are urban lakes subject to anthropogenic pressures in the form of industrial, residential and commercial development in the watershed,


Vi








some agricultural activities, and recreational uses such as boating, fishing, and skiing.

Cluster analysis performed grouping all six lakes

revealed that, with the exception of Lake Parker, all lakes could be aggregated by common limnological features, and that the relationships were independent of year, season and the combination of year and season.

Mean primary productivity for the studied lakes was 334.22 mgC.m-3.h 106.35. Primary productivity in Lake Parker (mean= 66.41 mgC.m-3.h) was five times less than the mean for the other five lakes.

Sedimentary records reveal that the lakes in this study have been eutrophic for at least the last century with a significant accumulation of organic material in the superficial mud. Superficial sediment content was always z40% of the organic matter for any lake. Lake Parker was the only lake investigated that displayed a progressive decline in water column total phosphorus in the last fifteen years, perhaps signaling some reversal of cultural enrichment.

There was a significant difference in the amount of feces hand-stripped from the fishes studied. Blue tilapia had a greater amount of fecal material than did gizzard shad. The bulk of the fecal material collected from both fish species consisted of small green and blue-green algae, which were the dominant algal taxa in the environment at the time the fishes were collected.


vii








There were also significant differences in the

composition of the fecal material produced by each fish species. Gizzard shad fecal material consisted of 37% more organic matter and 50% more protein than did blue tilapia fecal material. Caloric content, measured as gross heat, in gizzard shad feces was 60% greater than in blue tilapia fecal material.

Initial chlorophyll a values from the feces measured for blue tilapia were 10 to 15 times greater than those measured for gizzard shad feces. Two bioassays were conducted over a seven day period for each procedure. Final chlorophyll a values for gizzard shad feces presented a tenfold increase from the initial chlorophyll a concentration, regardless of the presence of blue-green or green algae as the inoculum. Blue tilapia, on the other hand, exhibited a three- to fivefold increase in samples where a green algal species was used as the inoculum (there was zero growth in 66% of the samples); and there was no apparent growth in samples inoculated with a blue-green algae.

The results indicate that blue tilapia fecal material completely suppressed blue-green algal chlorophyll a production and appears to have suppressed green algal chlorophyll a production in more than 60% of the samples. Gizzard shad feces increased chlorophyll a values in bluegreen and green algal groups (tenfold increase). This study suggests that in Florida systems with conditions similar to


viii








those of the lakes studied, biomanipulation techniques can mitigate cultural eutrophication to improve water clarity, if gizzard shad are replaced by blue tilapia.
















































ix














CHAPTER 1
INTRODUCTION



Management of planktonic food webs to improve lake

trophic state is still an experimental procedure, and many interactions are unknown or poorly understood (Crowder et al. 1988). There is now enough evidence that top-down interactions (fish introduction or removal) have a significant effect on planktonic communities (algal biomass reduction), particularly in low nutrient lakes (Cooke et al. 1993).

Chemical treatment, unless it involves an inactivation of nutrient release (i.e., from the bottom), has proved unsatisfactory for algal control. Mechanical treatments such as algal harvesting, artificial circulation, and bottom sealing have been shown not to be satisfactory due to either their ephemeral effects or their high cost. Biological control, even in its earliest stage, seems to be a promising approach for long term algal control.

Traditionally, limnologists have considered the

interactions in a lake ecosystem as an unidirectional flow of components consisting of a nutrient-phytoplanktonzooplankton-fish pathway. Manipulation of food webs is an area fertile with possibilities for understanding ecosystem

1








2

function and for developing techniques for lake management that are not dependent on chemical and mechanical means.

All research done in the field of biomanipulation

supports the conclusion that the presence of large-bodied zooplankton, Daphnia spp., is required for a strong control over phytoplankton populations. Unfortunately, not all geographic areas meet this requirement. Tropical and subtropical areas, namely Florida and other areas of the southeastern U.S., do not have large species of zooplankton. Crisman and Beaver (1990) described only the presence of small Cladocera (e.g., Eubosmina spp., Ceriodaphnia spp.) in Florida lakes which are not good candidates for top-down biomanipulation purposes.

Lacking the presence of the most studied and accepted

candidate for the top-down type of biomanipulation, tropical and subtropical researchers have had to seek another organism which could perform a corresponding role. Filterfeeding fish, which are very common in those areas, seem to be the most suitable candidates. Filter-feeders do not visually detect individual prey items, but engulf a volume of water containing the food organisms and retain the planktonic prey and particles by passing this volume over entrapment structures (Lazzaro, 1987).

The most common filter-feeding fishes in Florida

eutrophic lakes are gizzard shad (Dorosoma cepedianum), which can produce a total grazing pressure on algal








3

populations even greater than the large bodied zooplankton in temperate systems (cf. Drenner et al., 1982a). Another very common filter-feeder fish inhabiting Florida eutrophic lakes is blue tilapia (Tilapia aurea) which was introduced into the United States in 1961. Both of these fishes feed on small, particulate material. For example, gizzard shad 5 cm in length can filter algae >19p in size; those 15 cm in length can filter algae >40p; and those 25 cm in length can filter algae >63p. Blue tilapia can consume algae >25p in size (Opuszinsky and Shireman, unpubl.).

The filtering rates of these species can be very high. According to Drenner et al. (1982a), the gizzard shad population in Lake Barkley, Texas (85 ha), can effectively filter a volume equivalent to the entire lake every 2.3 days.

Undoubtedly, filter-feeding fishes can alter the

phytoplankton community of aquatic systems. Nevertheless, contrary to expectations, an increase in planktonic algae biomass and primary production has been observed (Drenner et al., 1984, 1986; Janusko, 1974,1978; Opuszynski, 1978). Some of the reasons for the failure of filter-feeding fish to control algal blooms and in some cases even worsen water quality are [1] elimination of larger zooplankton species; [2] more rapid cycling of plant nutrients (Opuszynski and Shireman, unpubl); and [3] differential release of fecal material by individual species of fish.








4

Digestion in animals has been an almost totally

neglected subject. Fish studies have been limited to gut content and selectivity (Halver, 1989). Very little is known of the digestive capabilities of fishes.

Fish, like all other animals, require energy to sustain life, and they are among the most efficient animals in converting food to body tissue (Halver, 1989). However, a rough generalization can be made that about one-third of the energy in the food offered to fish will be lost as combustible waste (Halver, 1989). This will consist of uneaten food, feces, urine, and gill excretions. From the lost combustible wastes, feces (25% of the total) constitute the majority of this part of the energy balance.

Fecal material produced from different fishes and

released into the environment can produce a differential impact and consequently a differential response. Blue tilapia feces, being encased in mucilage, are more coherent and will be broken down slowly, whereas shad feces are released practically as aqueous substance permitting an almost instantaneous availability of nutrients for the phytoplankton community.

Because little is known of their feeding ecology,

especially quantitative feeding under natural conditions, attempts to predict and efficiently manage these filterfeeding fish as a tool to alleviate eutrophication processes still is in its experimental stage. Some researchers (i.e.








5

Opuszynsky and Shireman, unpubl.) have developed a new approach to improve the use of filter-feeding fish to counteract eutrophication. They constructed an apparatus consisting of a cage where fish are kept that was equipped with funnels under the cage to collect fish feces for estimating food consumption via a quantification of the production of feces. According to these researchers, this cage combines two indispensable features for an effective use of filter-feeding fish to control planktonic algal growth and to reduce eutrophication: [1] it enables the coexistence of filter-feeding fish and zooplankton by eliminating the fish from the water column thus improving phytoplankton consumption; and [2] fish feces can easily be collected and removed from the bottom of the cage, eliminating a major source of nutrient for the phytoplankton community.

There is a high degree of trophic overlap between young blue tilapia and larval gizzard shad in Lake George, Florida (Zale, 1984). Beaver and Crisman (1989) reported phytoplankton and zooplankton community alterations in eutrophic Florida lakes when blue tilapia has an established population. In eutrophic central Florida lakes, blue tilapia is quickly replacing gizzard shad as a major filter-feeding fish (Beaver and Crisman, 1989).

The current study addresses gaps on current knowledge regarding fish as biomanipulation agents for phytoplankton.








6

I worked with the digestive physiology of blue tilapia and gizzard shad with an emphasis on their fecal material, as well as with the limnological characteristics and paleolimnological interpretations from six central Florida lakes. The research addressed the following hypotheses: [1] There is differential nutrient bioavailability from blue tilapia and gizzard shad feces; [2] different type of feces will produce a different effect on the lake ecosystem; [3] fundamental differences in the physiology and biochemistry of blue tilapia and gizzard shad feces will impact the sediment composition in lakes; and [4] interlake differences should exist in key limnological parameters between lakes of differing relative abundance of blue tilapia and gizzard shad. The goals of this study are to [1] learn something about the fecal material composition of filter-feeding fish to help understand their role in freshwater systems; [2] evaluate management techniques that couple physicochemical water information with biotic community components and the lake sedimentary record; and [3] gather new information on the effect of different fish species' fecal material on lake ecosystems.














CHAPTER 2
LITERATURE REVIEW



Introduction


This chapter reviews the literature on biomanipulation and shad and tilapia ecological energetics by examining planktivorous fish from the community standpoint with special emphasis on fecal material composition and fate in the system and the significance of different fish species as trophic linkages in freshwater food webs.

Biomanipulation

The two main strategies to control eutrophication of fresh waters are 1) reduction of external and internal loading of nutrients (Bjork, 1985) and 2) control of internal ecological processes. With respect to toxic substances, organic wastes and acid precipitation, the first strategy alone will provide acceptable solutions to the problems on a long-term scale (Benndorf, 1988). However, a combination of strategies 1 and 2 could lead to an improvement in water quality and to a lower cost/benefit ratio in the management of the water resource.

Until recently, eutrophication problems were tackled

primarily by reducing external nutrient loading (Hosper and


7








8

Meijer, 1986; Van Liere, 1986). In recent years, many studies have shown that food web manipulation (biomanipulation) can help restore eutrophied lakes (Andersson et al., 1978; Henrikson et al., 1980; Reinertsen and Olsen, 1984; Carpenter et al., 1985).

Biomanipulation, as it was originally defined by

Shapiro et al., (1975), refers to management of aquatic communities by controlling natural populations of organisms with the end goal being water quality improvement. This definition was reviewed later by Shapiro (1990), and he pointed out the considerably broader connotation that the term has taken on recently relative to that espoused by him and many others in the field. In the original idea of Shapiro (1990), biomanipulation is a series of manipulations of the biota of lakes and their habitats to facilitate certain interactions and results which we, as lake users, consider beneficial, namely reduction of algal biomass and, in particular, of blue-green species.

According to Gophen (1990), in a broad sense,

biomanipulation is equatable to top-down forces, trophic cascade interactions or food-web manipulation. All these terms refer to manipulation of secondary or tertiary aquatic producers and its impact on lacustrine community structure. Whole-lake food-web manipulation by fish-stock management, i.e., reduction of planktivorous fish, through enhancement of piscivorous fish, may accelerate the rate of the








9

restoration process (e.g. Shapiro and Wright, 1984; Edmondson and Abella, 1988).

For studying the effect of food-web manipulations on lake restoration and to examine the fundamental mechanisms underlying ecosystem regulation, application of whole-lake manipulation-experiments can have major advantages (van Donk et al., 1990). These experiments simulate or actually encompass the conditions that would be expected to occur naturally in lakes (Carpenter and Kitchell, 1988). Problems of enclosure-size and omitted members of communities are not relevant to whole-lake manipulations (Frost et al., 1988). It is, however, difficult to perform these manipulations on a large scale and to interpret their results (Hulbert, 1984). Interpretations may be greatly eased by results of small-scale manipulations on similar systems (Frost et al., 1988).

Nevertheless, to obtain a good overview of the

community's processes, investigations must be conducted simultaneously as small-scale in situ or laboratory experiments that can be replicated under controlled conditions. In this approach, whole-lake manipulations can be considered as generating as well as testing hypotheses (O'Neill et al., 1986).

Examination of trophic-level interactions has long been an integral part of limnology (Hrbacek et al., 1961; Nauwerck, 1963; Brooks and Dodson, 1965). However, Shapiro








10

et al. (1975) and Shapiro (1978) recognized that eutrophication problems are biological manifestations of nutrient availability, and were the first to suggest that manipulation of trophic interactions (biomanipulation) could be used as a lake management tool to alleviate the biological consequences of eutrophication without the need for costly controls on nutrient loading (Crisman and Beaver, 1990).

Since biomanipulation research began in temperate

regions, the emphasis has focused basically on alterations of the zooplankton community (enhancement of large cladocerans especially of the genus Daphnia), because, through these manipulations, phytoplankton has been reduced (blue-green algae) and water transparency has increased despite difficulties in maintaining such a condition for a long time. Only recently has research been conducted using fishes as a major element in the biomanipulation process, but still only under temperate conditions.

In subtropical and tropical regions, conditions are quite different. One particular element must be stressed; there is a complete absence of large zooplankton in tropical regions. This and numerous other physicochemical and biological differences between these climatic regions impedes the applicability of biomanipulation research from the temperate regions to warmer climates (Crisman and Beaver, 1990). Thus, the natural choice for biomanipulation








11

schemes of subtropical and tropical regions is the use of planktivorous fish. Despite the potential for utilization of filter-feeding fish in biomanipulation techniques to maintain algal control, especially in shallow eutrophic lakes and reservoirs characterizing subtropical and tropical regions, little information exists regarding the direct impact on the general planktonic community.

Florida lakes possess characteristics which

fundamentally differentiate them from most temperate systems. The relative shallowness of Florida lakes, combined with a moderate climatic regime, dictates that stable thermal stratification usually only occurs in deeper (>5-6 m) lakes of the region (Beaver et al., 1981). Such lakes are warm monomictic and circulate throughout the winter months, while shallower basins are more strongly affected by local meteorological events and are subjected to higher internal nutrient loading via sediment resuspension (Pollman, 1982).

Subtropical lakes of Florida are inherently different from temperate lakes in a number of respects that could affect the success of whole lake biomanipulation. The size structure of subtropical zooplankton communities is skewed toward smaller individuals, and large-bodied cladocerans are absent (Bays and Crisman, 1983). Despite the lack of largebodied crustaceans, intense grazing activities by gizzard shad may strongly determine plankton structural and functional characteristics (Bays and Crisman, 1983). These








12

subtropical systems maintain primary production all year while production in temperate systems is greatly depressed or absent during winter. Thus, if non-vegetative seasonal estimates of production are included in determining annual production, it is likely that subtropical systems would realize greater yearly production.

From the few papers about the use and application of

biomanipulation techniques in the subtropical zone, Crisman and Beaver (1990) noted that the increase in cladoceran abundance in Florida lakes following elimination of fish predation agrees with observations in temperate lakes (Lynch, 1979; Carpenter et al., 1987; Van Donk et al., 1989), but unlike the latter lakes, community species composition was not altered nor was there a marked increase in crustacean mean body size (Shapiro and Wright, 1984; Benndorf et al., 1988). Large-bodied daphnids, the focus of all temperate studies, are absent in Florida regardless of trophic state or predation intensity (Crisman and Beaver, 1990). The results of Crisman and Beaver (1988) from research conducted in Lake Apopka, Florida, show a fundamental disagreement with the biomanipulation of Round Lake, Minnesota (Shapiro and Wright, 1984), and suggest that zooplankton size, structure and standing crop have only minimal influence on phytoplankton biomass in Florida lakes.

Regarding fish composition, Crisman and Beaver (1990) noted that unlike the eutrophic temperate lakes








13

characterized by size-selective planktivorous fishes, in subtropical systems this composition shifts to pump-filter feeding fishes, reflecting faunal dominance by gizzard shad (Dorosoma cepedianum). Crisman and Beaver (1990) noted that in both temperate and subtropical systems, removal of planktivorous fish results in higher macrozooplankton populations.

Unlike temperate systems, however, algal biomass was not reduced in the presence of enhanced macrozooplankton abundance, but actually increased. Crisman and Beaver (1990) suggested that small-bodied macrozooplankton, even if freed from fish predation, are of questionable value as biomanipulation tools in eutrophic subtropical lakes. These authors also stated that if biomanipulation were to be successful in the subtropics, emphasis should be shifted from zooplankton to the role played by planktivorous fish (pump-filter feeding).



Shad Ecological Energetics

The gizzard shad (Dorosoma cepedianum) occurs commonly in many lakes and streams in North America (Scott and Crossman, 1973), where it is the principal phytophagous fish (Crisman and Kennedy, 1982). This species is predominately found in eutrophic lakes, being reportedly a key link in the food chain between primary producers and the top carnivores








14

which are important to commercial and sport fishermen (Lewis, 1953; Jester and Jensen, 1972).

Despite its role as a forage fish, it has been

suggested that in warm-water, shallow lakes with soft mud bottoms, high turbidity, and relatively few predators, the gizzard shad may become a nuisance from an ecological and economic standpoints (Miller, 1960). Kutkuhn (1958) and Cramer and Marzolf (1970) reported that gizzard shad longer than 35 mm are primarily herbivorous, with some selection for zooplankton when sighted, while young gizzard shad (< 30 mm) feed primarily on zooplankton (Bodola, 1966). Organic detritus at times can also be an important food for the gizzard shad (Dalquest and Peters, 1966; Baker and Schmitz, 1971).

Apparently, owing to the high reproductive capacity and rapid growth of gizzard shad in shallow lakes and reservoirs, predators cannot effectively crop young-of-theyear fish and hence provide a control on population numbers (Miller, 1960; Bodola, 1966). It has also been suggested that gizzard shad may inhibit growth of more desirable fish and/or limit their numbers through interspecific competition (Berry, 1958; Miller, 1960).

The gizzard shad is a dominant native grazer in Florida eutrophic lakes. Crisman and Kennedy (1982), based on mesocosm experiments, showed that gizzard shad did not impact chlorophyll a values, productivity or phytoplankton








15

densities, and that shad caused a significant increase in both the concentration of orthophosphate and its ratio to total phosphorus under natural stocking densities. The fish was also responsible for a significant decrease in copepod density during that study.

Crisman and Beaver (1988) and Threlked and Drenner

(1987) noted that the overall effect of shad grazing was to stimulate the phytoplankton community and to decrease Secchi disk transparency. Shad do not effectively graze more evasive zooplankton such as Diaptomus (Drenner et al., 1978), particularly during the summer months, although other evasive herbivores may be simultaneously depressed (Drenner et al., 1982a). Many common algal taxa remain viable after gut passage through the digestive tract of shad, especially blue-greens (Velasquez, 1939; Smith, 1963; Crisman and Kennedy, 1982).

It has been assumed that the enhancement of

phytoplankton populations by shad grazing is an indirect effect caused by suppression of herbivorous zooplankton (Threlkeld, 1987). Shad may also enhance phytoplankton populations directly by providing a larger quantity of highly assimilable nutrients from their digestive products (Crisman and Kennedy, 1982).

Gizzard shad and other clupeids are very sensitive fish that can be easily stressed and killed. In systems where gizzard shad is considered a nuisance, as in all Florida








16

eutrophic lakes, the death of large numbers of shad, besides any aesthetic consequences, will represent an enormous return of phosphorus and nitrogen into the water column that is readily available to stimulate the growth of already overly productive algae, resulting in oxygen depletion that could lead to another fish kill.

The results of the Crisman and Kennedy (1982)

investigation suggest that the presence of gizzard shad can promote lake eutrophication both through elevation of orthophosphate concentrations and differential digestion of diatoms and green algae, thus increasing the competitive advantage of blue-green algae. Finally, they emphasize that this fish species does not appear to be a suitable candidate for use as a biocontrol agent for phytoplankton in eutrophic subtropical lakes.

Considerable literature is available concerning the

life history of Dorosoma cepedianum (Drenner, 1977; Lazzaro, 1987). Only a few investigators, however, have attempted to describe and quantify energy relationships in this species (Smith, 1971; Garland, 1972; Pierce, 1977; Crisman and Kennedy, 1982).



Tilapia Ecological Energetics

Blue tilapia (Tilapia aurea), a species native to

Africa and the Middle East, was introduced into the United States in 1961 in a Hillsborough County phosphate pit (Ware,








17

1973) as a means of controlling nuisance aquatic macrophytes (Courtenay and Robins, 1973; McBay, 1961). After its original introduction, the fish spread rapidly throughout the southeast, particularly Florida, but have not controlled excessive macrophyte growth (Ware et al., 1975).

Qualitative and quantitative measurements of the gut

contents of blue tilapia in situ indicate that this species is an opportunistic omnivore, utilizing zooplankton (Spataru and Zorn, 1978; Mallin, 1986), phytoplankton (Hendricks and Noble, 1980; Mallin, 1986), and detritus (Hendricks and Noble, 1980; Mallin, 1986). Gophen et al. (1983) found in laboratory tests that tilapia greater than 7.6 cm utilized a series of rapid suctions to draw prey into their buccal cavity. This mechanism is undirected, and thus, fish in this size range function as filter feeders. Tilapia smaller than

7.6 cm also function as filter-feeders; however, they also feed as size selective predators on individual zooplankton specimens.

Drenner et al. (1984a) reported that grazing activities by blue tilapia depressed some large algae (Uroglenopsis and Ceratium), while the smallest phytoplankton taxa were enhanced (Rhodomonas, Chrysochromulina, Chlamydomonas and Cyclotella). This enhancement was ascribed both to nutrient regeneration during gut passage and to fish feces, as well as the accompanying compositional shifts in the herbivorous zooplankton community. Little information is available on








18

nutrient release and degree of algal digestion by blue tilapia (Crisman and Beaver, 1988), but Popma (1982) noted that some algal cells may remain viable following passage through the digestive tract of tilapia.

The zooplankton community is also modified by tilapia grazing activity, and Drenner et al.(1984a) showed in pond studies that the population of Keratella was suppressed, while copepodids and adult Diaptomus were enhanced. Gophen et al. (1983) reported that blue tilapia are selective feeders on Bosmina and Ceriodaphnia, taxa that have poor evasive capabilities when compared with more successful zooplankters such as Mesocyclops (Crisman and Beaver, 1988). Dickman and Nanne (1987) noted that in Central American fish ponds very high levels of tilapia (2.5 adults.m-2) suppressed zooplankton populations and increased the importance of the bluegreen alga Microcystis aeruginosa.

The spawning behavior of Tilapia aurea is typical of

many cichlids with all incubating duties being performed by the female (McBay, 1961). Reproductive behavior progresses from schooling, territorial establishment by the males, prespawning courtship, spawning, and parental care. McBay (1961) suggested that the mature female will spawn at a constant water temperature of 230C. The number of hatchlings per spawn of T.aurea was comparatively smaller than in most fishes native to the United States, but apparently compensated for by strong parental care behavior.








19

Foote (1977) considered temperature, predation and salinity to be the primary limiting factors in the distribution of blue tilapia. The sensitivity of blue tilapia to low temperature is apparently the most important factor affecting the potential range of the species in North America. Shafland (1978) tested lower lethal acclimation temperatures for blue tilapia, spotted tilapia and the Mozambique mouth brooder, and found that all tilapia tested died between 6 and 120C. Of these three tilapia species, blue tilapia was the most tolerant of cold water. According to McBay (1961), T. aurea will not tolerate temperatures as low as 90C. Based on these results and January isotherms, Shafland concluded that the blue tilapia has the potential of extending its range to include the entire state of Florida (Williams et al. 1985).

Salinity has a slight but significant effect on the

cold tolerance of blue tilapia, suggesting that the species may be expected to extend its range farthest north along the coast and that populations in estuarine systems may be able to withstand exceptionally cold weather better than inland populations (Zale, 1984). However, the fish is capable of finding thermal refugia during cold weather; the presence and location of these must be considered when assessing habitat suitability based on thermal criteria (Zale, 1984).

There is concern that the presence of blue tilapia

reduces largemouth bass populations through competition for








20

nesting sites or predation on bass eggs (Noble et al., 1975). Zale (1984) also found a high degree of trophic overlap between young tilapia and larval shad in Lake George, and perhaps the enhanced abundances of individuals documented following introductions of blue tilapia have resulted from exploitative competition for zooplankton during early life stages. He suggested this to be a more realistic explanation than the "competition for algae and detritus among adults" theory usually invoked.

These results are confirmed by Beaver and Crisman

(1990) who reported phytoplankton and zooplankton community alterations in subtropical Florida lakes when blue tilapia replaces gizzard shad. Chlorophytes and rotifers were proportionally more abundant in tilapia-dominated lakes (Crisman and Beaver, 1988), while cladocerans composed a greater percentage of the zooplankton population in shaddominated lakes. Limited zooplankton distribution data suggest that total zooplankton biomass is comparably depressed in systems dominated by tilapia. Empirical evidence suggests that little measurable improvement in water quality through differential grazing will be realized if blue tilapia displace shad as the dominant rough fish in subtropical Florida lakes (Beaver and Crisman, 1990).

A summary from previous investigations in Florida lakes is compiled in Table 2-1. The data reported are condensed from many different sources which were already cited.









21

Table 2-1. Summary of previous investigations with blue
tilapia and gizzard shad in Florida lakes.


GIZZARD SHAD


I Chlorophytes f Cyanophytes ? Chrysophytes Chrysophytes

T Rotifers ? Rotifers

& Cladocerans ( Cladocerans

( Zooplankton biomass ( Primary Productivity



Previous studies (Courtenay and Robins, 1973; Ware et al., 1975; Crisman and Kennedy, 1982; Bays and Crisman, 1983; Zale, 1984; Crisman et al., 1986; Crisman and Beaver, 1988) delineated the differential impact that each fish has on distinct groups of organisms in Florida lakes. It is not clear if any change can be deduced for chrysophytes in systems dominated by blue tilapia, whereas some increase in chlorophytes is reported. Gizzard shad increased cyanophytes and chrysophytes. Regarding zooplankton composition, blue tilapia is reported to increase rotifers while suppressing cladocerans. The gizzard shad impact on rotifers has not been determined, although it is known that cladoceran populations decrease. Finally, both species seem to suppress total zooplankton biomass while gizzard shad was shown to stimulate phytoplankton primary productivity.














CHAPTER 3
RELATIVE IMPORTANCE OF BLUE TILAPIA AND GIZZARD SHAD TO LAKE
SEDIMENT AND WATER COLUMN NUTRIENT CONCENTRATION-ANALYSIS OF FISH FECES


Introduction


Examination of trophic-level interactions has long been an integral part of limnology (Caird, 1945; Hrb&cek et al., 1961; Nauwerck, 1963; Brooks and Dodson, 1965). Elements such as herbivory, predation, and nutrient recycling by animals have always been considered to affect the biomass and species composition of prey populations in terrestrial as well as aquatic communities. Several investigators (e.g., Hairston et al., 1960; Wiegert and Owen, 1971; Patten, 1973; Porter, 1977; and Paine, 1980) have examined several aspects of food webs and the factors which control the biomass and productivity of various trophic levels. Applicability of these ideas to aid management practices and promote aquatic ecosystem recovery are now under investigation (Cooke et al. 1993).

Shapiro et al. (1975) and Shapiro (1978) recognized

that eutrophication problems are biological manifestations of nutrient availability and were the first to suggest that manipulation of trophic interactions (biomanipulation) could be used as a lake management tool to alleviate the 22








23

biological consequences of eutrophication without the need for often costly controls of nutrient loading (Crisman and Beaver, 1990). Biomanipulation, as stated by Shapiro and Wright (1984), is based on the prediction that increased piscivore abundance will result in decreased planktivore abundance, increased zooplankton abundance, and increased zooplankton grazing pressure leading to a reduction in phytoplankton abundance and improved water clarity.

This "biological" approach could make it possible to increase herbivore density in aquatic communities, thereby lowering algal biomass to levels less than expected for a given nutrient concentration. In addition, limiting the abundance or even the occurrence of certain fish could curtail the flux of nutrients from epilimnetic or littoral sediment to the pelagic zone. Improvement of the water quality of lakes with algal blooms by implementing a combination of these biological techniques could reduce or eliminate the need for the use of common chemical (i.e. cooper sulfate) and mechanical methods to deal with eutrophication (Cooke et al., 1993).

Caird (1945), with his experiment of adding largemouth bass to a 15-ha Connecticut lake, was one of the first to publish observations about the effect of increased biomass of piscivorous fish on the phytoplankton community. More recently, investigators such as Hrb&cek et al. (1961), Brooks and Dodson (1965), and Hulbert et al. (1972) have








24

demonstrated that planktivorous fish (mainly filter-feeder fish) can severely reduce or even eliminate large-bodied zooplankter Daphnia spp.

Planktivorous fishes use two distinct means to feed on plankton: particulate feeding and filter feeding. Particulate feeders attack single individual planktonic prey items which they visually select from the water column (Werner, 1977; Vinyard, 1980; Lazzaro, 1987). Filter feeders do not visually detect individual prey but swallow a volume of water containing the food organisms and entrap the planktonic forms in structures such as gill rakers and other filtering structures (see Lazzaro, 1987), using rhythmic suctioning actions to capture the prey, while either swimming slowly or remaining quite stationary (Drenner, 1977; Gophen et al., 1983).

Gizzard shad (Dorosoma cepedianum), a filter feeder, is a dominant native grazer in Florida eutrophic lakes. Crisman and Kennedy (1982) used mesocosm experiments to demonstrate that gizzard shad had no impact on chlorophyll a values, lake productivity, or phytoplankton density. Gizzard shad was responsible for a significant increase in both the concentration of orthophosphate and its ratio to total phosphorus under natural stocking conditions and for the decrease in copepod density (Crisman and Kennedy, 1982).

Later experiments, however, demonstrated that the overall effect of shad grazing was to stimulate the








25

phytoplankton community and to decrease Secchi disk transparency (Crisman and Beaver, 1988; Threlked and Drenner, 1987). Shad do not effectively graze on more evasive zooplankton such as Diaptomus (Drenner et al., 1978), and many common algal taxa remain viable after gut passage through the digestive tract of shad, especially blue-green algae (Velasquez, 1939; Smith, 1963; Crisman and Kennedy, 1982).

Blue tilapia, Tilapia (=Sarotherodon =Oreochromis)

aurea, a fish native to West Africa and Palestine (Trewavas, 1965), was introduced into the United States in 1957 by researchers at Auburn University who were investigating its potential as a food and sport fish (Swingle, 1960). In 1961, the Florida Game and Freshwater Fish Commission acquired juvenile fish from Auburn University to study the potential use of blue tilapia as a sport fish (Crittenden, 1962) and as an agent for weed control (McBay, 1961; Courtenay and Robins, 1973). After its introduction, the fish successfully invaded natural habitats throughout the Southeast, particularly Florida. Blue tilapia has not, however, been a successful candidate for use as a sport fish or as a biological control for excessive macrophyte growth (Ware et al., 1975).

Tilapia can be classified as an opportunistic omnivore, consuming zooplankton (Spataru and Zorn, 1978; Mallin, 1986), phytoplankton (Hendricks and Noble, 1980; Mallin,








26

1986), and detritus (Hendricks and Noble, 1980; Mallin, 1986). Gophen et al. (1983) reported that tilapia greater than 7.6 cm act as filter feeders, and tilapia smaller than

7.6 cm, in addition to filter feeding, also feed as size selective predators on individual zooplankton species (particulate feeder).

The grazing activities of blue tilapia depressed some large sized algal groups (Uroclenopsis and Ceratium), while the smallest phytoplankton taxa were enhanced (Rhodomonas, Chrysochromulina, Chlamydomonas, and Cyclotella) (Drenner et al., 1984a). Tilapia grazing did suppress the zooplankter Keratella, while copepodid and adult Diaptomus populations were enhanced (Drenner et al., 1984a). Tilapia also selectively fed on Bosmina and Ceriodaphnia (Gophen et al., 1983), taxa reportedly having poor evasive capabilities.

Little information is available on nutrient release and degree of algal digestion by blue tilapia. Popma (1982) noted that some algal cells may remain viable following passage through the digestive tract of tilapia. Dickman and Nanne (1987), found that high concentrations of adult tilapia (2.5 adults m-2) raised in some Central America ponds suppressed zooplankton populations and increased the blue-green alga Microcystis aeruginosa.








27

Methods

Fish Collection

Blue tilapia were collected by electroshocking in Lake Alice, Florida. A boom-style electrofishing boat was used (7.5 GPP, with a 7 KW generator, Smith-Root, Inc. Vancouver, Washington). This model was the most efficient electrofishing unit for the type of lake sampled. The collected fish were immediately placed on ice to minimize post-capture digestive processes.

Blue tilapia were measured for total length (mm) and

weight (g). All fish collected were separated into two size groups (TL), less than 380 mm and greater than 380 mm, which will subsequently be referred to as size groups T-1 and T-2, respectively. Thirty-six fish were collected. Thirteen of these were in group T-l; twenty-three were in group T-2 (fish length ranged from a minimum of 284 mm to a maximum of 450 mm). Fecal material, here defined as the collection of all digested material that could be identified by its brownish-dark coloration in the fish intestines, was handstripped from all the fishes and pooled into two different jars (T-1 and T-2). The sample jars were immediately placed on ice in a cooler for transport to the laboratory. All sample material was kept at 40C within four hours after collection.

Samples were subdivided into four different groups; groups 1 and 2 pulled from the T-1 jar and groups 3 and 4








28

from the T-2 jar. Three aliquots for analysis (replicates) were taken from each of the four groups.

Gizzard shad were collected using an experimental

monofilament gillnet. All fish sampled were collected as part of a population control project being conducted by the Florida Game and Freshwater Fish Commission, Eustis (FL). The sampling net had 6 panels ranging from 6.3 to 12.7 cm stretch mesh in 1.3 cm increments. Gillnetting was done in Lake Denham, Florida, in increments of two hours.

Gizzard shad were measured for total length (mm) and

weight (g). All fish collected were separated into two size groups (TL), <320-mm and >320-mm, which will subsequently be referred to as size groups S-I and S-2, respectively. A total of 218 fish were collected in five different gillnet settings. Minimum fish length was 203 mm, and the maximum length was 440 mm. Fecal material was hand-stripped in the field from all the captured fishes and poured into two jars (S-I and S-2). The sample jars were immediately placed on ice in a cooler for transport to the laboratory. All sample material was kept at 40C within four hours after collection.

Samples were subdivided into four different groups; groups 1 and 2 pulled from the T-1 jar and groups 3 and 4 from the T-2 jar. Three aliquots for analysis (replicates) were taken from each one of the four groups.








29

Feces Analysis

The digestive tract was removed through an incision

made from the anus anterior to the base of the pelvic fin, then dorsally to the posterior side of the pectoral girdle to the dorsal side of the coelom. Fecal material was handstripped from the beginning of the intestine to the base of the anus and placed into a 250 ml jar. The feces were then pooled according to size group (T-1, T-2, S-1 and S-2).

Laboratory analyses consisted of determination of

percent dry matter, percent organic matter, bulk density, heat content upon combustion, protein analysis, total phosphorus and total nitrogen content and two algal bioassays. Dry weight/water content of feces was measured by weighing samples before and after drying at 1050C for 16 hours; organic matter was measured using a cool muffle furnace before ashing the samples at 6000C for at least 3 hours and then weighing. Nitrogen was measured as total Kjeldahl nitrogen (TKN) using a Technicon-II semi-automated manifold following a modification of the methods described by Bremner and Mulvaney (1982) (selenium was not used as a catalyst). The digestate was also used for total phosphorus

(TP) determinations. Liberated orthophosphate was determined with the ascorbic acid method (APHA, 1985).

All samples for calorimetry analyses were freeze dried. Volatilization began at -300C. Total energy was determined by oxygen bomb calorimetry (Parr Model 1261 Isoperibol








30

Calorimeter). Pre-weighed (pre-burn weight, approximately one gram of dried material) fecal samples were placed in a microbomb. The bomb was placed in 2 liters of water at 30310C. The sample was ignited and burned in the presence of oxygen. After ignition, increases in water temperature were registered to calculate energy within the sample, expressed as gross heat (cal.g-1).

An algal bioassay was done using thirty-six 300 ml

flasks containing FWH+Si (a standard medium for maintenance culture of freshwater blue-green algae). Eighteen flasks were prepared without KNO3 (medium -N); and eighteen flasks prepared without K2HPO4 (medium -P). These two groups of eighteen flasks were each subdivided into three different subgroups of six; one subgroup containing 3 ml of shad feces; another containing 3 ml of tilapia feces; and the third subgroup without feces, to function as a control. Each group of six flasks was subdivided further into two subgroups: three flasks inoculated with a strain of bluegreen algae (Microcystis aeruginosa), and three flasks inoculated with a strain of green algae (Selenastrum capricornutum). These sets of flasks were incubated for seven days in a room under constant light and temperature conditions.

Samples from all experimental flasks were analyzed for chlorophyll a using extractive and fluorimetric methods (APHA, 1989). Cell counting of algae from samples collected








31

at the beginning and end of the experiment from each flask was performed, using a 2 ml chamber on an inverted Leitz microscope.



Results and Discussion

Bulk density calculated for tilapia feces had a mean

value of 0.2576 g.cm-3, with a minimum value of 0.2326 g.cm3, and maximum value of 0.2815 g.cm-3 (Table 3-1). For shad feces, the mean value was 0.1401 g.cm-3, with a minimum value of 0.1258 g.cm-3 and maximum value of 0.1493 g.cm-3 (Table 3-1).

Percent dry matter for tilapia feces had a mean value of 13.0 %, with a minimum value of 12.6 %, and a maximum value of 13.6 % (Table 3-1). For shad, the mean value was

8.0%, with a minimum value of 7.5 % and a maximum value of 8.6 % (Table 3-1). Percent organic matter calculated from tilapia feces had a mean value of 49.6%, with a minimum value of 48.6 %, and maximum of 51.6 % (Table 3-1). Shad had a mean value of 68.0 %, with a minimum of 67.0 %, and a maximum of 68.9 % (Table 3-1).

Percent protein was calculated for tilapia and had a mean value of 21.6 %, minimum of 21.0 %, and maximum of 23.2 % (Table 3-1). For shad, the mean value was 32.4 %, with a minimum of 30.25 %, and maximum of 33.75 % (Table 3-1).






Table 3-1. Physico and chemical characteristics of shad and tilapia excretions expressed as ml and g.

PARAMETER SHAD TILAPIA SHAD TILAPIA SHAD TILAPIA

Mean Mean St. Deviat. St. Deviat. Range(*) Range(*)
(ml) (g) (ml) (g) (ml) (g) (ml) (g) (ml) (g) (ml) (g) Bulk 1.0/ 0.1401 1.0/ 0.2576 0.72/ 0.010 0.08/ 0.022 (0.90-1.06) (0.9-1.09)
Density
( g. cm- I (0.12-0.15) (0.23-0.28) % Dry N/A /8.0366 N/A /12.988 N/A /0.5748 N/A /0.4856 (7.49-8.64) (12.6-13.66) Matter
% Org. N/A / 67.84 N/A / 49.66 N/A /0.7954 N/A /1.3929 (67.06-68.90) (48.6-51.63)
Matter
% Protein N/A /32.427 N/A /21.655 N/A /1.3467 N/A /1.0568 (30.25-33.75) (21.1-23.19) Calories 7516.35 ml 4531.18 ml 253.49 278.53 (7147-7852) (3805-5091) (cal.g-1) 4041.05 g 2604.13 g 136.28 160.08 (3842-4222) (2187-2926) TN N/A / 52.98 N/A / 34.65 N/A /0.9307 N/A /1.7055 (48.38-54.00) (33.4-37.14) (pg.1- )
TP N/A / 6.51 N/A / 6.63 N/A /0.3636 N/A /0.4708 (5.98 7.04) (6.05- 7.13) (pg.11)


(*) Numbers expressed in the 1st line are in ml.
Numbers expressed in the 2" line are in g.
When only one line presented, it is expressed in g.




(J3








33

Caloric content, expressed as gross heat, for tilapia feces had a mean value of 2604.13 cal.g-1, with a minimum value of 2186.88 cal.g-1, and maximum of 2926.08 cal.g-1 (Table 3-1). Shad had a mean value of 4041.05 cal.g- 1, and minimum and maximum values of 3842.40 cal.g-1 and 4212.81 cal.g-1, respectively (Table 3-1). Pierce (1977), from data obtained from 27 samples of shad fecal material, calculated its mean caloric content as 1953 cal.g-.

Total nitrogen for tilapia feces displayed a mean value of 34.6 pg.l-, with a minimum of 33.42 pg.1-1, and a maximum of 37.14 pg.l-1 (Table 3-1). For shad, the mean was 52.98 pg.1-1, with the minimum and maximum values being 48.38 pg.1-1, and 54.0 pg.1-, respectively (Table 3-1). Mean total phosphorus (TP) for tilapia was 6.63 pg.1-, with a minimum of 6.05 pg.l-1, and a maximum of 7.13 pg.1-l (Table 3-1). Shad had a mean of 6.51 pg.1-1, with a minimum of 5.98 pg.1-1, and a maximum of 7.04 pg.l-1 (Table 3-1).

Absolute values for total nitrogen, percent of

nitrogen, crude protein, total phosphorus and percent phosphorus composition for gizzard shad and blue tilapia fecal material are provided in Tables 3-2 and 3-3.

There were significant differences in the amount of feces hand-stripped from both fishes. Tilapia had the greatest amount of feces, and it was not possible to isolate any particular factor responsible for such an occurrence. It may be related to the length of the intestinal tract (which









34


Table 3-2. Total nitrogen, total phosphorus and protein
content for gizzard shad feces.



pg N % N Crude Protein mg P % P Group 1 A 48.38 4.84 30.25 5.98 .60
B 48.95 4.89 30.56 6.84 .68 Group 2 A 54.00 5.40 33.75 6.41 .64
B 53.65 5.36 33.50 6.19 .62 Group 3 A 51.43 5.14 32.12 7.04 .70
B 53.36 5.34 33.37 6.79 .68 Group 4 A 52.44 5.24 32.75 6.55 .65
B 53.04 5.30 33.12 6.26 .63




Table 3-3. Total nitrogen, total phosphorus and protein
content for blue tilapia feces.



_g N %N Crude Protein pg P %P Group 1 A 33.68 3.37 21.06 7.13 .71
B 33.42 3.34 20.87 6.86 .69 Group 2 A 37.14 3.71 23.19 6.47 .65
B 34.38 3.44 21.50 6.05 .60 Group 3 A 22.61 2.26 14.12 5.50 .55
B 22.44 2.24 14.00 5.30 .53 Group 4 A 27.40 2.74 17.12 12.66 1.27
B 28.48 2.85 17.81 11.54 1.15








35

in tilapia is longer) or the time of the year (i.e., spawning time). In gizzard shad, the length of the intestinal tract increases with increasing standard length. Schmitz and Baker (1969) found that the intestine-to-bodylength ratio for gizzard shad was about 2.8:1, while in threadfin shad the ratio averaged 1.8:1. Although the length of the intestine in a fish may be partially determined by its diet, Schmitz and Baker (1969) believe that the length and form of the intestine are probably the result of a complex of intrinsic and extrinsic factors that can not be explained only by diet.

Tilapia produces five times more feces than bighead

carp, whereas the growth rate of both species was comparable (Opuszynski and Shireman, unpubl.). In this study, I found that the average feces produced by the shad was 2.08 0.34g fresh feces per 100g fish body weight and 4.5 0.21g fresh feces per 100g fish body weight for the tilapia. The fact that blue tilapia feces' bulk density is twice as high as shad feces (Table 3-1), indicates that tilapia produces four times more feces as dry matter than shad. Therefore, blue tilapia seems to be more suitable than bighead carp (Opuszynski and Shireman, unpubl.) and shad for use in biological schemes for water quality improvement.

There are obvious macroscopic differences in the feces produced by these two fish taxa. Tilapia feces are released








36

as mucilaginous pellets, while shad feces are released into the environment as a flocculent material.

The bulk of both fish feces consisted of small green and blue-green algae species, dominating the phytoplankton in the assemblage at the time the fishes were collected. Many of the algal cells did not show any signs of digestion, or they were slightly digested but easily identifiable under a microscope. It is known that many algal taxa may be viable after passage through fish digestive tracts (Fish, 1951, 1955; Lowe, 1959; Smith, 1963). Fish (1951, 1955) reported that, of the algae eaten by tilapia, blue-green and green algae generally pass through the fish undigested and that diatoms form the main usable food (Cailteux, 1988). In this study, I found green algae of the genus Closterium, the blue-green Microcystis, and the diatoms Melosira, Fragilaria, Tabellaria, and Cyclotella in tilapia feces. Shad feces contained the green algal genus Ankistrodesmus, the diatom Melosira, the dinoflagellate Peridinium, and the blue-green Microcystis.

Mean values for chlorophyll a (final initial

measurements) in shad feces were [1] for the inoculum made in the medium deficient of nitrogen(-N) and the blue-green algae 541.57 mg.m-3 36.85; and 859.18 mg.m-3 17.93 with green algae; [2] for the inoculum made in the medium deficient of phosphorus (-P) and the blue-green algae 432.21 mg.m-3 22.02; and 1173.76 mg.m-3 13.96 with green algae.








37

Mean values in tilapia feces were [1] for the inoculum made in the medium deficient of nitrogen (-N) and the blue-green algae -364.05 mg.m-3 38.06; and 329.33 mg.m-3 31.38 with green algae; [2] for the inoculum made in the medium deficient of phosphorus (-P) and the blue-green algae 683.14 mg.m-3 38.62; and -335.58 mg.m-3 17.98 for the inoculum with green algae.

The experiment was harvested on the seventh day

following daily evaluation of fluorescence. As a major trend, it can be said that fluorescence values increased dramatically (in the shad samples) during the first four days, after which they leveled off and did not change very much. In the medium containing blue-green algae, the combination shad feces -N as well as shad feces -P displayed a steady increase in values, which became more noticeable after the fourth day (Figure 3-1)). For tilapia feces, the medium deficient of nitrogen (-N) had the same value from the beginning and declined after the fourth day; whereas the medium deficient of phosphorus (-P), remained steady and began to increase slightly after the fourth day (Figure 3-1).

In the medium containing green algae, the mixture of

shad feces -N increased progressively daily and reached its highest values after the fourth day, which was a ten-fold increase. The combination -P displayed the same trend, however, with higher values increasing thirteen times after








Microcystis aeruginosa
0.05


F 0.04- ... + + I + + TII -P STII -N S 0 .0 3 .... .... ..... .. .
r + Shad -q







--= __--- .- ~ Otri -N






Shad + Tiapia Control

Figure 3-1. Fluorimetric activity from shad and tilapia
feces in medium with Micronvstis aeruoinosa.
- --i- - C t r l N









feces in medium with Microcystis aeruginosa. C








39

the fourth day (Figure 3-2). For tilapia feces, the medium deficient of nitrogen (-N), as well as the medium deficient of phosphorus (-P), showed a daily decrease in fluorescent values and on the fourth day of the experiment they stabilized in a value 50% lower than the initial number. (Figure 3-2). Shad feces displayed a five fold increase in fluorometric values compared with the measured in the medium with tilapia feces.

Phytoplankton biomass in the fecal material was

calculated from chlorophyll a concentrations assuming 1 mg chlorophyll a = 67 mg dry weight (APHA, 1982). Blue tilapia fecal material suppressed phytoplankton biomass (mg.dry weight) in 75% of the samples of blue green and green algal prepared in -N and -P media. Mean suppression in these samples was approximately 50% of the initial values (Table 3-4). Gizzard shad fecal material, on the other hand, increased phytoplankton biomass in all of the samples of blue green and green algae. The mean concentration increase was greater than tenfold of the initial values (Table 3-4).



Conclusions

Adult fishes that employ filter-feeding as their

primary feeding method are typically visual particulatefeeding planktivores when juveniles. They switch from a diet









Selenastrum capricornutum
0.3
'------------Shad -P
0.F 25

u 0.2- ..... ... ........

r t Shad -N e 0 .1 5 -.. .... .......... ... .SCtrl -P SI TII -P
0.05
Otri -N


11 1- 1 1 2 122.2222.33.33.33*44.4444.6 4 4 5666666-66 Time (days)

Shad + Tilapia Control

Figure 3-2. Fluorimetric activity from shad and tilapia
feces in medium with Selenastrum capricornutum.









Table 3-4. Phytoplankton biomass (mg.dry weight-') in fish excretions calculated from chlorophyll a values.



FISH + ALGA MEDIUM INITIAL VALUE FINAL VALUE FIN-INIT (X) (X)-CTRL.
mg/dry weight mg/dry weight mg/dry weight mg/dry weight Til (-N)+ Bluegreen 68054 43663 -24391 -23939 Til (-N)+ Green 62232 84297 22065 23269 Til (-P)+ Bluegreen 79421 33651 -45770 -51642 Til (-P)+ Green 68756 46273 -22483 -43562 Shad (-N)+ Bluegreen 2710 38995 36285 36737 Shad (-N)+ Green 4818 62383 57565 58769 Shad (-P)+ Bluegreen 3061 32019 28958 23086 Shad (-P)+ Green 2810 81454 78644 57565 Ctrl (-N)+ Bluegreen 903 451 -452 Ctrl (-N)+ Green 1882 678 -1204 Ctrl (-P)+ Bluegreen 502 6374 5872 Ctrl (-P)+ Green 2308 23387 21079



,,-








42

composed of large zooplankton as juveniles to increasing dependence on phytoplankton and/or smaller and non-evasive zooplankton as adults (Kutkuhn, 1957; Cramer and Marzolf, 1970; Janssen, 1978; Durbin, 1979; Drenner et al., 1982b; Lazzaro, 1987; and Yowell and Vinyard, 1993). Although the feeding habits of filter-feeding fish have been extensively studied, information about the quantitative feeding under natural conditions is scarce.

The theoretical basis of fish bioenergetics has

received considerable attention in the fisheries literature. Sophistication in describing basic equations and the addition of computer analysis have enabled investigators to partition more accurately the energy requirements of a population. The goal of this study was to provide additional information on the feces produced by gizzard shad and blue tilapia including fecal composition, and nutrient and caloric content. Experiments were conducted to compare the potential impact of the fecal material of each respective fish species on the primary production in aquatic systems.

Feces composed the greatest constituent of the energy balance estimated from an experiment with bighead carp employing energy budget equations to calculate food consumption and food assimilation, i.e., C=P+R+F+U (1) and A=C-F+U (2), where C=consumption, A=assimilation, P=production, R=respiration, F=feces, and U=urea (Opuszynski and Shireman, unpubl.).








43

I found that bulk density for tilapia fecal material was twice the value for shad feces. Averaged shad feces analysis had a composition of 37% more organic matter, 50% more protein and 50% less dry matter than tilapia feces which makes shad feces more organic, more protein-rich and more easily dissolved in the environment. Caloric content of shad feces was 60% higher than tilapia feces.

In general, assimilation efficiencies, which are

calculated using the balance of total calories of ingestion and egestion material in fish, have been assumed to approximate 80 percent of ingested energy (Winberg, 1956; Ricker, 1968; Mann, 1969; Moriarty and Moriarty, 1973). Although little is known about the energetics of omnivorous fish such as shad and tilapia, an assimilation efficiency of 42 percent is reported for shad by Pierce (1977), and of 7080 percent for tilapia (Moriarty and Moriarty, 1973).

In this study, I found caloric content in shad fecal material to have a mean of 4041.05 cal.g-1 (Table 3-5), whereas tilapia had a mean of 2604.13 cal.g-1 (Table 3-6). Considering both fishes relative to a natural diet, tilapia is egesting material that has 60% less caloric content than shad. This is, consequently, in accordance with what is stated in the literature about the assimilation efficiency for both, where tilapia has roughly two times greater assimilation efficiency than shad, being expected to egest material with a lower caloric content.









44

Table 3-5. Caloric content of gizzard shad feces.


SHAD CALORIC CONTENT (Cal/g)
Group 1 (A) 4115.56
(B) 4134.18 (C) 4150.35
Group 2 (A) 4066.21
(B) 4221.81 (C) 4196.37
Group 3 (A) 4026.14
(B) 3935.90
Group 4 (A) 3894.76
(B) 3867.86 (C) 3842.40




Table 3-6. Caloric content of blue tilapia feces.


TILAPIA CALORIC CONTENT (Cal/g)
Group 1 (A) 2532.20
(B) 2490.92
Group 2 (A) 2722.98

(B) 2648.89 (C) 2926.08
Group 3 (A) 2533.76
(B) 2560.45 (C) 2417.74
Group 4 (A) 2312.26
(B) 2186.88








45

The diet of the gizzard shad is one of the most

controversial aspects in the ecology of this species. It is generally agreed that very young shad (<35 mm) feed almost exclusively on microcrustaceans (Kutkuhn, 1958; Cramer and Marzolf, 1970), and fish larger than 35 mm have been reported to consume a wide variety of food materials. Tilapia are primarily vegetative feeders, feeding on algae or aquatic plants, though occasionally zooplankters and insect larvae picked up with bottom debris are eaten (Lowe, 1959). However, despite the similarity in feeding habits, differing fecal composition and assimilation rates suggest that tilapia and shad play significantly different roles in the nutrient dynamics of aquatic systems.

Chlorophyll a values for both shad and tilapia fecal material were very different between the beginning and end of the bioassays. In both bioassays for blue-green and green algal groups, tilapia initial values were 10 to 15 times higher than for shad (tilapia mean value of 915 mg.m-3 and shad mean value of 67 mg.m-3); whereas at the end of bioassays, the situation was completely inverted, with shad displaying a tenfold increase in chlorophyll a (shad mean value of 685 mg.m-3) regardless of the algal group tested. (Figures 3-3 to 3-6) Tilapia chlorophyll a values at the end of bioassay varied with the algal group tested. In the Microcystis experiment, all values were negative indicating suppressed chlorophyll a values (mean of -691 mg.m-3)









mg/m^3 1200


1000


800 F
F

600 ....


400


200 F



Control Tilapia Shad

Time
Figure 3-3. Bioassay initial (I) and final (F) values of chl. a (mg.m-3) with M.aeruainosa in medium -N.









mg/m^3 1400 12 0 0 ...........

1000 8 0 0 .... ...... .............

600-F F 400 200- F..


Control Tilapia Shad

Time
Figure 3-4. Bioassay initial (I) and final (F) values of chi. a (mg.m-3) with M.aeruoinosa in medium -P.









mg.m^3
1400
F
1 2 0 0 ...... ..........

1 0 0 0 ................................. ......

800

600

400

200 I F

0
Control Tilapia Shad Time
Figure 3-5. Bioassay initial (I) and final (F) values of chl. a (mg.m-3)
with S.capricornutum in medium -N.








mg.m^3
1400
F
1 2 0 0 .................

1000

800... ...

600
F
4 0 0 ................

2 0 0 I..............

0
Control Tilapia Shad Time
Figure 3-6. Bioassay initial (I) and final (F) values of chl. a (mg.m-3)
with S.capricornutum in medium -P.








50

(Figures 3-3 and 3-4), while in the Selenastrum tests, about 66% chlorophyll a had negative values indicating suppression, with the remaining values being positive with a mean of 600 mg.m-3 (Figures 3-5 and 3-6).

McDonald (1985) found that blue tilapia enhanced the

growth of Ankistrodesmus cells. Ware et al.(1975) concluded that blue tilapia had displayed little potential for algal control in Florida waters, and in fact, had spread rapidly and become a nuisance. Phytoplankton biomass in the fish fecal material calculated from chlorophyll a values at the beginning and end of the current bioassay showed that blue tilapia suppressed phytoplankton biomass in 75% of the samples of blue green and green algae to a level exceeding 50% of the initial value. Gizzard shad increased phytoplankton biomass concentration in all samples of blue green and green algae more than tenfold from the initial values.

It has been reported (Crisman and Kennedy, 1982) that gizzard shad are not suitable for use as a biocontrol agent for phytoplankton because they have no impact on chlorophyll a values, productivity or phytoplankton densities, and they can promote lake eutrophication through elevation of orthophosphate concentrations and differential digestion of some algae groups, especially blue-greens and greens.

On the other hand, results of extensive work with tilapia are still inconclusive regarding the possible








51

importance of this species in biomanipulation schemes. It is known that adult blue tilapia employ filter-feeding as their primary feeding method, while they feed as visuallyoriented, particulate-feeding zooplanktivores as juveniles (Yowell and Vinyard, 1993). It is assumed that minimization of feeding costs while maximizing net energy return has been the basis for the filter-feeding strategy. The remaining question is what causes these fishes to switch their feeding behavior.

Facultative planktivores that use both feeding modes interchangeably (Leon and O'Connel, 1969; Crowder and Binkowski, 1983; and Helfman, 1990) usually filter-feed on small, abundant zooplankton and particulate-feed on large, less abundant forms. Yowell and Vinyard (1993), suggest that this is a stratagem which maximizes net energy return in response to a changing prey situation. The reduction in zooplankton density may then allow an increase in non-grazed phytoplankton (Drenner et al., 1987; Vinyard et al., 1988).

The results of this study indicate that blue tilapia

fecal material, either by its composition and/or the absence of some essential nutrient, completely suppressed Microcystis aeruginosa chlorophyll a production, and appears to have suppressed Selenastrum capricornutum chlorophyll a production in > 60% of the samples. Gizzard shad feces increased chlorophyll a production values in both bluegreens and greens (tenfold increase). This study suggests









52

that, in Florida systems with conditions similar to those of the lakes studied, biomanipulation techniques can mitigate cultural eutrophication if gizzard shad are replaced by blue tilapia.














CHAPTER 4
SEDIMENTS AND HISTORICAL ECOLOGY OF TWO CENTRAL FLORIDA URBAN LAKES


Introduction



In the absence of historical data, past lake conditions can be reconstructed based on interpretations of the sedimentary record. Assuming that lake sediments accumulate in an orderly fashion, paleolimnology may provide information about how the lake ecosystem has changed over time. Reconstructing the reason for past changes in the lake may be used to predict the lake's response to future management strategies (Smeltzer and Swain, 1985; Brenner et al., 1993).

Lake sediments originate from numerous sources. The

main sources are both the biochemical substances produced by organisms, or resulting from their degeneration, and morphological pieces of specific organisms. The sediment constitution is influenced primarily by the geomorphology of the lake basin and the drainage basin (Wetzel, 1983). Paleolimnology has as a principal objective a formulation of general principles about the way lakes change with time (Livingstone, 1981; Cohen and Nielson, 1986; Johnson et al., 1991).

53








54

Florida has about 7800 lakes varying in size from 0.4 ha to over 180,000 ha. (Canfield and Hoyer, 1988). Lakes in Florida serve many agricultural, domestic, industrial and recreational purposes (Canfield and Hoyer, 1992). Outside the glaciated regions in North America, Florida lakes constitute the largest group of natural lakes (Hutchinson, 1957) on the continent and the most important group of solution basins (Hutchinson, 1957; Crisman, 1992)

Limnological characteristics and productivity from these lakes vary widely and range from oligotrophic to hypereutrophic (Canfield and Hoyer, 1988; Brenner et al. 1990; Beaver et al. 1981). Although Florida's aquatic systems are large in number and have a significant economic impact, water quality data have been collected for few lakes (>10%) (Brenner et al., 1990), and routine data acquisition began fairly recently, in the 1960s and 1970s (Huber et al., 1982).

Considering Florida's aquatic system dimensions and

economic importance, little information has been reported on the sediments of Florida lakes, i.e. Flannery et al. (1982); Brenner and Binford (1988); Brenner et al. (1990); Stoermer et al. (1992); Gottgens (1992); Brenner et al. (1993). Here I provide information on the relationship between accumulation rates of sediment variables such as organic matter, water content, total nitrogen and total phosphorus








55

and estimated trophic status based on sedimentation rates (g.cm-2.yr) for the last century.

Sediment cores collected from Lake Bonny and Lake

Gibson (Polk County, Florida) were dated isotopically with 210Pb and 137Cs to estimate material accumulation rates in the sediment profile using radioactive decay of fallout ("unsupported") 21Pb. These techniques have been widely used to detect changes in sediment accumulation rates due to urban development in a watershed (Smeltzer and Swain, 1985), clear cutting of vegetation (Oldfield et al., 1980) and major climatic events (Robins et al., 1978).

In this study, these techniques were applied to identify lake sediment profile physicochemical characteristics and evaluate the potential impact of gizzard shad feces when it is the dominant filter feeder fish on lake sediment composition and net accumulation rates of organic matter, total nitrogen and total phosphorus for Lakes Bonny and Gibson. Whenever possible, this information will be correlated with the results from sediment cores for Lake Parker and Lake Hollingsworth reported by Schelske et al. (1992). For this study, both lakes were considered to be dominated by blue tilapia population as the main filter feeder fish.

According to the classification of Forsberg and Ryding (1980), Lake Bonny is hypereutrophic and Lake Gibson is an eutrophic lake. Mean values for all physicochemical and








56

biological parameters are provided in Chapter 5 of this study. Both are small urban lakes subject to anthropogenic pressures in the form of industrial, residential and commercial development in the area and recreational uses such as boating, fishing, and skiing. In addition, there are agricultural activities at Lake Gibson watershed.



Study Sites

Lake Bonny is a small (144 ha of surface area), shallow (Z.x= 2.5 m), urban lake bordered by the city of Lakeland, (Polk County, Florida). It has no known discharges or withdrawals (Polk County Water Resources Division, 1990).

The population of Lakeland has increased from less than 500 inhabitants in 1880 to nearly 75,000 people in 1993. As a result, all of the lake watershed was developed for commercial and residential uses. Lake Bonny has consistently had poor water quality (Polk County Water Resources Division, 1990) with an average 1992 TSI of 73.5. Urban runoff from commercial and residential development in the watershed and low lake levels resulting from seepage have been a major problem. Extremely poor water quality was recorded in 1986 at a time of extreme low lake levels (Polk County Water Resources Division, 1990). Based on current data, the TSI was 98; total nitrogen 9.0 mg.1-1, total phosphorus 0.7 mg.l-', Secchi 0.2 m and chlorophyll a was 220 mg.m-3.








57

Using the classification system of Forsberg and Ryding (1980), Lake Bonny is classified as hypereutrophic. Average total phosphorus concentration is 59 pg.1-' and average total nitrogen concentration is 1858 pg.l-1. Total chlorophyll a concentrations average 40 pg.1-1, the water clarity as measured by a Secchi disc averages 0.6 m, and average pH is 7.8. Table 5-10 exhibits water quality data for Lake Bonny.

Before 1985, the plant community of Lake Bonny was comprised mainly of Hydrilla verticillata. Due to the macrophyte chemical control program employed by the City of Lakeland, it has been replaced by Typha sp. as the dominant macrophyte. Tvyha sp. is responsible for a percent lake area coverage (PAC) of 10% (cf. Canfield and Hoyer, 1992).

The fish population in this lake has changed

dramatically. The species that survived the drought of 1983 are currently repopulating the lake. The most abundant openwater species collected in experimental gillnets are gizzard shad and Florida gar with 17.0 and 11.3 fish/net/24 hr, respectively (Canfield and Hoyer, 1992).

Lake Gibson is a small (192 ha of surface area),

shallow lake (zmx= 6.1 m) located on the outskirts of the City of Lakeland in a rapidly urbanizing area where citrus groves and pastures are giving way to large-scale commercial and moderate-density residential development. A domestic wastewater treatment plant from one local elementary school








58

discharges to a wetland north of Lake Gibson, which overflows into the lake. There are no known withdrawals (Polk County Water Resources Division, 1990).

Lake Gibson is classified as eutrophic according to

Forsberg and Ryding's (1980) classification system. The 1992 Florida TSI for Lake Gibson is 55 (Polk County Water Resources Division, 1990). The water is slightly tannic (color is 43 cpu) which lowers the Secchi disc value and may bias the TSI value. The average total phosphorus concentration is 94 pg.l-1, and the average total nitrogen concentration is 1570 pg.l-'. Total chlorophyll a concentration averages 23 pg.l-1 and the water clarity as measured by use of a Secchi disc averages 1.3 m. Lake's annual primary productivity (mean = 241.18 mgC.m-3.h 229.1) displaying noticeable seasonal variability and the presence of blooms of cyanobacteria every few years suggests an eutrophic status for Lake Gibson. Water quality data for Lake Gibson are shown in Table 5-8.

The dominant macrophytes of the lake are water primrose, alligatorweed, and pennywort, which are responsible for 5% lake area coverage (Canfield and Hoyer, 1992).

No recent fish population evaluation was obtainable for this lake. From the 1979 evaluation, ten species of fish were reported, with channel catfish being dominant.









59

Methods

An 89 cm mud-water interface core was collected from the middle of Lake Bonny using a piston corer with a 12 cm diameter, 1.83 m long cellulose acetate butyrate (CAB) core barrel. Externally, the core was characterized by a light color sediment for the first 10 cm and a typical black mud color below in the rest of the core.

A 60 cm core was taken from the central portion of Lake Gibson using the same piston corer described above. The presence of sand mixed with mud was noticed beginning at the 50 cm depth. Sand content increased deeper in the core. Due to the soupiness of the sediments, in both cores a spoon was used to collect the samples down to the 10-cm layer. From that point on, collection was possible using a spatula. The sediment from Lake Gibson was more consolidated on the bottom than the sediment from Lake Bonny.

Concentrations of 21Pb and 137Cs were measured by

direct y-assay using a P-type, intrinsic-germanium detector (Princeton Gamma Tech). The counting system used for spectral analysis is located in the University of Florida's Department of Environmental Engineering Sciences Low Background Counting Room. The electronics for the system include a preamplifier (RG11B/C, Princeton Gamma Tech,Inc.), amplifier (TC 242, Tennelec), bias supply (5 kV, TC 950, Tennelec), power supply (TC 909, Tennelec), and transformer (Sola).








60

Samples for isotope analysis were dried at 950C for 24 hours, pulverized by mortar and pestle, weighed, and placed in small plastic vials. Core sections were combined (up to 4 cm) to obtain an adequate sample weight (> 1 g). Vials were sealed with plastic cement and left for 14 days to equilibrate radon (222Rn) with radium (226Ra). Counting times were never less than 23 hours. Standards were counted to track efficiency (counts per y) and to calculate a 226Ra conversion factor (PCi per count per sec.). Blanks were counted to determine background radiation.

Bulk density and water content of sediments were

measured by weighing samples before and after drying at 950C. Percent organic matter was evaluated by loss on ignition (LOI) at 5500C for one hour (H&kanson and Jansson, 1983). Nitrogen was measured as total Kjeldahl nitrogen (TKN) using a Technicon-II semi-automated manifold after digestion following Bremner and Mulvaney (1982) but modified to exclude selenium as a catalyst. The digestate was also used for total phosphorus (TP) determinations. Liberated orthophosphate was determined with the ascorbic acid method (APHA, 1985).



Results and Discussion

The Lake Bonny Record

The 89 cm Lake Bonny sediment core had a mean bulk

density of 0.094 g.cm-3 0.042, mean dry matter of 9.16%








61

3.93, mean organic matter of 56.00% 12.69, mean total nitrogen of 22.32 mg.g-1 2.68 and mean total phosphorus of

4.10 mg.g-1 1.53 (Table 4-1).

Lake Bonny had high organic content in surface

deposits, but this declined somewhat with depth and age down to 19 cm of the core which coincides with the 1980s, and then increased again to the bottom where it reached highest values (dated at the beginning of this century, 1909) (Figure 4-1). Total nitrogen and total phosphorus plotted against time were constant throughout the whole core (Figure 4-2). Bulk density increased in the top 40 cm of the core which coincides with a 210Pb determination of sediment age of approximately 1950, i.e., about the time when much agriculture in the watershed was being converted to urban development (USGS Topographic Maps, Lakeland Quadrangle, 1944, 1975).

Total P reconstruction near the base of the core suggests that the lake has always been mesotrophic to eutrophic. However, the lake has experienced periods of much higher nutrient enrichment, and values for total P in excess of 5.0 mg.g-1 were computed for several sections that postdate the 137Cs peak of 1960s and 1970s. No change in total P concentration has been noticed since then, suggesting that the lake trophic state has remained the same for the last thirty years.








Table 4-1. Sediment core physical and chemical characteristics from two central Florida urban lakes.



PARAMETER LAKE LAKE BONNY GIBSON
Mean St. Dev. Range Mean St. Dev. Range Bulk Density 0.279 0.319 0.094 0.094 0.042 0.021 (g. cm-3) 1.644 0.218
Dry Matter 21.61 13.93 9.47 9.16 3.93 2.24
(%) 76.49 19.81 Organic Matter 38.17 14.85 1.73 56.0 12.69 32.72
(%) 59.56 76.71
Nitrogen 9.33 2.99 1.46 22.32 2.68 18.07 (mg. g-l) 13.46 28.00 Phosphorus 1.86 1.49 0 4.10 1.53 1.51 (mg. g-l) 4.57 6.38













IQ








Percent organic matter
80



60



40



20



0 I I I I lI Il' I 1909 21 37 48 63 68 63 70 74 78 80 82 83 86 87 89 91 1993
Time (year)
Figure 4-1. Lake Bonny percent organic matter sediment core content versus time (year).









mg/g
30

NITR GEN 25


20







.PH..OSPH...S



0
1909 21 37 48 63 68 63 70 74 78 80 82 83 86 87 89 .91 1993 Time (year)
Figure 4-2. Lake Bonny TN (mg.g-1) and TP (mg.g-1) sediment core content versus time (year).








65

Sedimentation rates expressed in Figure 4-3 for Lake Bonny are in agreement with the lake history. The first peak of 0.1 g.cm-2.yr coincides with the 1955 development boom in the area, and the second peak of 0.13 g.cm-2.yr in 1983 coincides with the time when the lake was almost completely dry.

Detectable 137Cs, which is an human-produced

radionuclide that was injected into the atmosphere as a consequence of nuclear weapons testing in the early 1950s, matched well with the determined 210Pb chronology (Table 42). This agreement is remarkable, considering the shallowness of the lake and the flocculence of the bottom substrate which makes these upper sediments vulnerable to physical disturbance. Eutrophic systems such as Lake Bonny, however, accumulate sediments rapidly, so these physical disturbances likely affect short time-intervals only. The Lake Gibson Record

The sediment core from Lake Gibson had a mean bulk

density of 0.279 g.cm-3 0.319, mean dry matter of 21.61% 13.93, mean organic matter of 38.17% 14.85, mean total nitrogen of 9.33 mg.g-1 2.99, and mean total phosphorus of

1.86 mg.g-1 1.49. (Table 4-1).

The organic content of Lake Gibson sediments was somewhat low in surface deposits (30% organic matter), increased with depth to about the midpoint of the core, matching with 1955, and declined to its lowest values of









g/cm^2/yr
0.14 0.12 0.1


0.08




0.04 0.02


1909 21 37 48 63 68 63 70 74 78 80 82 83 86 87 89 91 1993 Time (year)
Figure 4-3. Lake Bonny sedimentation rate (g.cm-2.yr) versus time (year).







Table 4-2. Lake Bonny sediment analysis.

Core Bulk 210Pb 137Cs Deposition TN TP Section Density pCi.g-1 pCi.g-1 Period Rate mg.g1 mg.g-1
(cm) g. cm-3 year g. cm-2.yr
0 4 0.029 18.72 1.5 1993 0.05 28.00 5.16 4 7 0.050 12.30 3.0 1991 0.08 23.11 5.29 7 9 0.061 12.50 4.13 1989 0.07 23.62 5.35 9 11 0.066 11.22 4.01 1987 0.07 27.15 6.03 11 13 0.074 8.46 3.65 1985 0.09 20.02 4.62 13 15 0.078 8.14 3.20 1983 0.09 20.45 4.81 15 17 0.079 5.27 4.02 1982 0.13 22.26 5.21 17 19 0.090 6.99 4.56 1980 0.10 21.58 3.53 19 22 0.092 8.92 4.49 1978 0.07 19.15 4.93 22 26 0.092 6.51 3.28 1974 0.08 20.12 5.86 26 30 0.090 7.00 2.67 1970 0.07 21.72 6.38 30 34 0.089 6.00 1.07 1963 0.07 23.63 5.07 34 38 0.091 3.80 0.85 1958 0.09 22.85 3.64 38 42 0.102 3.43 0.92 1953 0.08 24.67 3.68 42 46 0.164 3.48 0.87 1948 0.07 19.9 3.03 46 50 0.174 3.17 0.78 1937 0.06 18.07 2.42 50 54 0.142 1.91 0.91 1921 0.06 20.59 2.49 54 58 0.137 1.94 0.59 1909 0.04 20.36 2.22

(J








68

1.4% organic matter at the bottom of the core (approximately 1892) (Figure 4-4). Total nitrogen and total phosphorus displayed high variability. Total nitrogen peaked about 1942, declined after that and peaked again in 1990. Total phosphorus started with low values (1.0 mg.g-1) about 1892, increased steadily to peak after 1966, and, from that point on, kept increasing to a 1990 value of 5.0 mg.g-1 (Figure 4-5).

Total P reconstruction at the base of the core suggests that this lake, on the basis of phosphorus concentrations, has been oligotrophic in the past. Nevertheless, values for total P in excess of 3.0 mg.g-1 and greater were recorded for all levels corresponding to the last 25 years (since 1966), reflecting changes in watershed use.

Bulk density increased below the 24 cm core, which matched with a 210Pb determined sediment age of approximately 1950. Detectable 137Cs coincided with the determined 210Pb chronology (Table 4-3). The sedimentation rate for Lake Gibson corresponded with the lake's history. A significative peak of 0.12 g.cm-2.yr coincided with 1955, the beginning of the development boom for the area (Figure 4-6).

The organic matter content of dry bulk sediment in Florida lacustrine surface mud averaged 39.7% from 97 surveyed lakes (Brenner and Binford, 1988). Lake Gibson had a percent organic matter mean value of 38.2%, while Lake









Percent organic matter
70 60 50


40 30 20 10


0
1892 1914 1923 1935 1942 1946 1951 1955 1966 1973 1979 1986 1990 1993 Time (year)
Figure 4-4. Lake Gibson percent organic matter sediment core content versus time (year).









mg/g
14


12


10


8


6
PHOSPHOR
4

2


01
1892 1914 1923 1935 1942 1946 1961 1955 1966 1973 1979 1986 1990 1993 Time (year) Figure 4-5. Lake Gibson TN (mg.g-1) and TP (mg.g-1) sediment core
content versus time (year).







Table 4-3. Lake Gibson sediment analysis



Core Bulk 210Pb 137Cs Deposition TN TP Section Density pCi.g-1 pCi.g-1 Period Rate mg.g1 mg.g1
(cm) g. cm-3 year g.cm-2.yr
0 2 0.113 11.86 4.41 1993 0.07 10.50 4.45 2 4 0.167 9.48 5.02 1990 0.08 10.56 4.57 4 6 0.221 9.31 4.18 1986 0.08 9.11 4.10 6 8 0.233 6.76 3.63 1979 0.08 8.46 3.77 8 10 0.271 5.72 2.96 1973 0.08 7.83 3.28 10 12 0.284 6.07 2.96 1966 0.06 7.49 3.03 12*- 14 0.229 2.33 1.32 1955 0.12 10.38 2.07 14 16 0.178 2.63 1.28 1951 0.09 11.00 1.62 16 18 0.168 2.61 0.84 1946 0.08 11.70 1.70 18 20 0.160 3.34 0.75 1942 0.05 13.46 1.55 20 24 0.161 2.29 0.71 1935 0.06 13.21 1.50 24 28 0.185 1.11 0.75 1923 0.09 12.22 1.28 28 32 0.208 1.42 0.57 1914 0.05 7.26 0.70 32 36 0.183 2.00 0.30 1892 0.02 9.98 0.89





-J









g/cm^2/yr
0.14 0.12


0.1 0.08


0.06 0.04


0.02



1892 1914 1923 1935 1942 1946 1951 1955 1966 1973 1979 1986 1990 1993 Time (year)
Figure 4-6. Lake Gibson sedimentation rate (g.cm-2.yr) versus time (year).








73

Bonny had 56.0%. Total nitrogen for the same array of lakes displayed high variability, ranging from 0.6 to 42.4 mg.g1(Brenner and Binford, 1988). Nitrogen concentration for Lake Gibson was 9.33 mg.g-1 (Table 4-4), and for Lake Bonny was 22.32 mg.g-1 (Table 4-5). Total P concentrations recorded by Brenner and Binford (1988) varied from 0.07 to 8.09 mg.g-I. Values for Lake Gibson were 1.8 mg.g-1 (Table 4-6) and for Lake Bonny was 4.1 mg.g-1 (Table 4-7).

Considering that these lakes overlie phosphatic

limestone deposits and are located in urban settings, each one of those isolated factors or a combination of both could be responsible for the high total P values. The total P value registered for Lake Bonny of 4.1 mg.g-1 is higher than what was found in 93 of the 97 systems studied for Brenner and Binford (1988).

The Lake Parker Record

The Lake Parker sediment core displayed usually

elevated deposition rates for bulk density, organic matter and phosphorus (Brenner et al., 1993). Bulk sediment accumulation rates increased from 11 mg.cm-2.yr-1 in 1922 to a maximum of 131 mg.cm-2.yr-1 in the late 1950s (Brenner et al., 1993). Phosphorus net accumulation rates increased more than ten times during the last century, reaching the highest value of 600 pg.cm-2.yr-1 in the late 1970s (Brenner et al., 1993). Appendices D and E show physical and chemical properties of Lake Parker sediment core.








74



Table 4-4. Total nitrogen (mg.g-') for
Lake Gibson sediment core.



Interval (cm) TN (mg/g)
0-2 10.4995 2-4 10.5649 4-6 9.1080 6-8 8.4608 8-10 7.8340 10-12 7.4951 12-14 10.3830 14-16 11.0008 16-18 11.6984 18-20 13.4574 20-24 13.2149 24-28 12.2212
28-32 7.2597 32-36 9.9815
36-40 10.4658 40-44 10.6539
44-48 7.2012 48-52 3.3171 52-56 1.4618 56-60 10.4134

Mean 9.33








75

Table 4-5. Total nitrogen (mg.g-') for
Lake Bonny sediment core.



Interval (cm) TN (mg/g)
0-2 28.0057 2-4 26.5107 4-7 23.1092 7-9 23.6192 9-11 27.1484 11-13 20.0185 13-15 20.4545 15-17 22.2599 17-19 21.5829 19-22 19.1497 22-26 20.1250 26-30 21.7168 30-34 23.6266 34-38 22.8466 38-42 24.6686 42-46 19.9064 46-50 18.0703 50-54 20.5948 54-58 20.3690 58-64 24.4444 64-72 19.5555 72-84 23.2875

Mean 22.32








76

Table 4-6. Total phosphorus (mg.g-1) for
Lake Gibson sediment core.



Interval (cm) TP (mg/g)
0-2 4.4495 2-4 4.5700 4-6 4.1031 6-8 3.7696 8-10 3.2824 10-12 3.0312 12-14 2.0766 14-16 1.6263 16-18 1.6984 18-20 1.5514 20-24 1.5047 24-28 1.2811 28-32 0.7041 32-36 0.8948 36-40 0.9506 40-44 0.9809 44-48 0.6574 48-52 0.1171 52-56 0.0097 56-60 0.0000

Mean 1.8








77

Table 4-7. Total phosphorus (mg.g-') for
Lake Bonny sediment core.



Interval (cm) TP (mg/g)
0-2 5.1567 2-4 5.2729 4-7 5.2941 7-9 5.3534 9-11 6.0308 11-13 4.6168 13-15 4.8145 15-17 5.2071 17-19 3.5261 19-22 4.9261 22-26 5.8654 26-30 6.3821 30-34 5.0733 34-38 3.6455 38-42 3.6842 42-46 3.0309 46-50 2.4239 50-54 2.4907 54-58 2.2233 58-64 2.0648 64-72 1.5072 72-84 1.5842

Mean 4.1








78

The Lake Hollingsworth Record

Inferences from paleolimnological core studies indicate that Lake Hollingsworth has had high nutrient and chlorophyll a concentrations since the 19th century; however, the highest levels occurred between the 1950s and the 1970s, when much of the agriculture in the watershed was converted to urban development (Schelske et al., 1992). Lake sediments' topmost 10 cm is poorly consolidated (densities of <0.032 g.dry.cm-3.wet) (Schelske et al., 1992), although density generally increases with depth and is highest in sand-rich deposits.

Organic matter in the core generally ranges between 40 and 60% of dry weight but decreases in sandy deposits (Schelske et al., 1992). Total phosphorus concentrations have remained relatively constant during the last ten years, ranging from 59 to 71 pg.L-' (Schelske et al., 1992). Appendix F show physical and chemical properties of Lake Hollingsworth sediment core.



Conclusions

The data presented provide support for the conclusion of Flannery et al.(1982) that Florida lakes of higher trophic state have greater proportions of organic matter in their surface sediments e.g. Lake Bonny (TSI of 73.46) with 57% organic matter and Lake Gibson (TSI of 55.18) with 35% organic matter. Organic matter in the Lake Parker








79

superficial layers was >70.0%, and in Lake Hollingsworth was >51.0% (Schelske et al., 1992). Lake Bonny organic matter is positively correlated with depth and shows no correlation with any other parameter. Bulk density was positively correlated with depth and negatively correlated with total nitrogen and total phosphorus at the 5% level confidence (SAS, 1989) (Table 4-8).

Organic matter in Lake Gibson was negatively correlated to bulk density and showed a positive correlation to total nitrogen. Bulk density was positively correlated with depth, and depth was negatively correlated to total phosphorus at the 5% level confidence (SAS, 1989) (Table 4-9).

Using the reconstruction of historical limnological

conditions, paleolimnology may help address issues of lake management (Smeltzer and Swain, 1985) and restoration. Sedimentary records may reflect past trophic status of a lake and, thereby, assist in setting a goal for restoration. Models may then be employed to identify primary nutrient sources and to foresee whether mitigation actions can reduce loading sufficiently to improve water quality or if other management techniques such as biomanipulation may be needed.

When reduction of nutrient loading to a lake is

feasible, a cost/benefit evaluation should be completed to ascertain which restoration techniques are financially practical. Then, paleolimnological records can help in the decision making process.








80


Table 4-8. Correlation matrix for sediment variables for Lake Bonny.



SEDIMENT VARIABLES CORRELATION P VALUE Depth/ Bulk Density 0.7848 0.0000 Depth/ Dry Matter 0.8255 0.0000 Depth/ Organic Matter 0.5482 0.0123 Depth/ Total Phosphorus -0.8135 0.0000 Bulk Density/T Nitrogen -0.5724 0.0084 Bulk Density/T Phosphorus -0.6747 0.0011









Table 4-9. Correlation matrix for sediment variable for Lake Gibson.




SEDIMENT VARIABLES CORRELATION P VALUE Depth/ Bulk Density 0.5862 0.0066 Depth/ Total Phosphorus -0.9054 0.0000 Bulk Density/ Org. Matter -0.7335 0.0002 Org. Matter/T Nitrogen 0.6123 0.0041








81

Historical water-column total P data collected for the last four years at Lake Bonny showed little variation, with the exception of Spring 1991, when the values were double the mean (mean = 0.112 mg.l-1). An increase in chlorophyll a and total bacteria were also reported for the same period. This information is in disagreement with Maceina and Soballe (1990), who suggested that resuspension during wind events is probably responsible for the highly variable total P and chlorophyll a values measured in many shallow Florida lakes.

Lake Gibson historical records for total P showed no significant variation for the last four years, with a mean of 0.210 mg.l-1. Chlorophyll a, TSI and total bacteria also displayed no notable variations. Anthropogenic alterations in the watershed, especially increased urbanization, may be responsible for the presence of only one genus of blue-green algae in the lake, as well as for the occasional algal bloom.

The Lake Parker historical record for net phosphorus

accumulation rates greatly exceed dangerous P loadings to a basin of <5 m mean depth (13 pg.cm-2.yr-1) (Brenner et al., 1993). Total phosphorus measured in the water column during this study had a mean of 0.22 mg.L-1 (0.07 to 0.602 mg.L-1).

Lake Hollingsworth historical total phosphorus

concentration show an overall increase upwards in the sediment core (Schelske et al., 1992). Total P values








82

changed from 2.03 mg.g-1 at the base of the core to 6.60 mg.g-1 at the top (Schelske et al., 1992). Total phosphorus in the water column measured during this study had a mean of

0.27 mg.L-1 (0.081 to 0.61 mg.L-1)

Due to the geology in the area, Lake Bonny has been an eutrophic lake for much of its recent history, even with conditions of minimal human occupation in the watershed (Figure 4-7). This should be the foundation for any management plan for this lake. Any plans for water quality improvement in the lake must consider that the lake is edaphically phosphorus-rich, and even a great effort to reduce phosphorus concentrations in the water may provide little or unnoticeable improvement in water clarity.

Lake Gibson, meanwhile, had a different

paleolimnological record that displayed alterations during the last 25 years (Figure 4-8) in which the lake changed from mesotrophic to eutrophic, reflecting the switch in past largely agricultural watershed uses to large scale commercial and moderate density residential development. Any attempt to reduce nutrient concentration in this lake must consider control of point source external loading of nutrients as well as use of biomanipulation techniques as a means to achieve water quality improvement.

Figure 4-9 show the historic variation of organic matter content, nitrogen and phosphorus in the Lake Hollingsworth sediment core cf. Schelske et al. 1992. It is








80



60



40



20



0
1909 21 37 48 63 66 68 60 63 70 74 78 80 82 83 86 87 89 91 1993 Time (year)

% Dry matter % Organic matter
Nitrogen (mg/g) Phosphorus (mg/g)

Figure 4-7. Changes in some physicochemical characteristics in
Lake Bonny through time.








80



60



40



20



1892 14 23 29 36 38 42 46 48 51 66 60 66 69 73 79 81 86 90 1993 Time (year)

--% Dry matter % Organic matter
9 Nitrogen (mg/g) Phosphorus (mg/g)

Figure 4-8. Changes in some physicochemical characteristics in
Lake Gibson through time.









60


50


40





20




0I 1847 94 98 13 16 36 40 69 63 67 70 73 76 79 81 83 86 88 1990 Time (year)

Organic matter (%) -+- Nitrogen (%) -- Phosphorus (mg/g)

Figure 4-9. Changes in some physicochemical characteristics in
Lake Hollingsworth through time (after Shelske et al.,1992). U,








86

noted that then was an increase in phosphorus concentration after the 1950s, matching with the period of more intense occupation in the watershed for the whole area. Water quality measurements collected since 1966 demonstrate that the lake has been hypereutrophic for at least 25 years (Schelske et al., 1992). Although the lake is naturally productive, anthropogenic impacts have accelerated the rate of eutrophication. Eutrophication problems, especially from cultural enrichment, should be resolved in a number of ways, ranging from eutrophication prevention and lake rehabilitation, to learning to live with the problem.

From the sediment core analysis of Lakes Parker and Hollingsworth, lakes considered as having a dominant blue tilapia population in this study, and Lakes Bonny and Gibson, considered as having a dominant gizzard shad population, it is not clear that there are any significant differences in sediment organic matter content, nitrogen, phosphorus or any other analyzed parameter. All four lakes seem to have high organic matter content in the topmost layers, and all of them had a record in the sediments of the main environmental changes for the area e.g. the 1950's switch from totally agricultural practices in the area to watershed conversion to urban development, as well as the efforts been made to recover these systems since the mid 1980's.








87

Sediment cores from Lakes Bonny and Gibson collected during this study showed no noticeable changes relative to the improvement of any water quality parameter. However, Lakes Parker and Hollingsworth had what perhaps could be called a modification. According to Schelske et al. (1992), total phosphorus reconstructions in Lake Parker's topmost 15 cm (after 1960s) of sediment suggested a progressive decline in the water-column that may signal some reversal of cultural eutrophication. The Lake Hollingsworth core revealed that, for at least in the last ten years, total phosphorus inferences have remained relatively constant (Schelske et al., 1992).

Taking in consideration that blue tilapia came into the systems after 1961, it is reasonable to assume, first of all, a lag time for the establishment of the fish population in the new environment, and, secondly, time for the manifestation of any modification that those fishes would bring to the systems. Since more than thirty years have passed since fish introduction, it is possible to hypothesize that this could be responsible for changes recorded in Lakes Parker and Hollingsworth water quality, even though this might still provide a weak indication of water quality improvement. However, considering that all lakes included in this study are subject basically to the same kind of impacts, as well as the lake recovery program carried out by the City of Lakeland, the explanation for








88

this slight noticeable improvement and/or stabilization process in the water measured parameters in those lakes may be due to the either an altered biological community and/or the modifications resulting from the introduced species.














CHAPTER 5
LIMNOLOGICAL ASSESSMENT



Introduction


Florida can be divided into three major physiographic zones: the Northern, Central and Southern zones (Puri and Vernon, 1964; White, 1970). All lakes reported in this study are located in Polk County, Florida. Polk County lies within the Central Highlands physiographic province. The vast majority of the county lies within the Polk and Lake Uplands. The Polk Uplands is an area of continuous high ground located between the Gulf Coastal Lowlands and the western edge of the Lake Wales Ridge (White, 1970). Along the eastern ridge, deposits of the Fort Preston Formation occur, but most of the region is underlain by deposits of the phosphatic Hawthorn and Bone Valley Formations (Puri and Vernon, 1964). In the eastern half of the Polk Uplands, most lakes are in association with sand ridges. Polk County has 550 lakes covering 37.9 x 103 ha.

A majority of phosphatic sand deposits and clays of the Miocene Bone Valley and Hawthorne Formations are located in the Polk Uplands and Lake Uplands physiographic regions (Puri and Vernon, 1964). Florida has led the nation in 89








90

phosphate production for over 90 years; the bulk of this production is from Polk County (Boyle and Hendry, 1985). The trophic status for lakes of this region is mostly mesotrophic or eutrophic, and waters are moderately organically stained (Canfield, 1981).

Water quality on the Polk Upland is highly variable. Based on data from Florida Game and Fresh Water Fish Commission, Canfield (1981), Polk County Water Resources Division and this study, mean pH ranged from 5.4 to 9.4 and total alkalinity concentrations averaged between 1 and 76 mg.l-1 as CaCO3. Total hardness concentration averaged between 14 and 195 mg.l-1 as CaCO3, and mean specific conductance ranged from 82 to 281 pmhos.cm-. Calcium was the dominant cation and bicarbonate the dominant anion in Lake Parker, whereas in most of the other lakes in the area, sodium and chloride were the dominant anions. Total nitrogen concentration was between 306 and 1566 mg.m-3, mean total phosphorus concentration from 2.4 to 144 mg.m-3, and Secchi disk depth 0.2 to 2.7 m. The wide range of water chemistry may be directly related to different regional geology (Canfield, 1981; Crisman, 1993).

The following agencies have conducted studies or

developed information on the lakes in Polk County: United States Geological Survey (USGS); Environmental Protection Agency (EPA); Polk County Water Resources Division (PCWR); City of Lakeland (CL); Florida Game and Freshwater Fish








91

Commission (FGFWFC); Southwest Florida Water Management District (SWFWMD); Central Florida Regional Planning Council (CF208).



Methods

I selected six lakes in Polk County, Florida: Lakes Parker, Hollingsworth, Hancock, Gibson, Bonny and Hunter. For this study all lakes were sampled quarterly during the years of 1992 and 1993 following approved U.S. EPA methodology. Additional data for the years of 1989 to 1991 were obtained from the above listed agencies.

Three water quality sampling stations were established for Lakes Parker, Hollingsworth and Hancock, and two stations were established for Lakes Gibson, Bonny and Hunter. Water was collected from just bellow the surface (0.5 m) in acid-cleaned Nalgene bottles. Samples were placed on ice and returned to the laboratory for analysis. Water temperature (OC), dissolved oxygen (mg/L), pH and conductance (pS/cm @ 250C) were measured by using a Hydrolab Station. Secchi depth (m) was measured at each station where water was collected.

At the laboratory, total alkalinity (mg/L as CaCO3) was determined by titration with 0.02 N H2SO4 (APHA, 1985). Total phosphorus concentrations (mg/L) were determined using the methods of Murphy and Riley (1962) after persulfate oxidation (Menzel and Corwin, 1965). Total nitrogen




Full Text
2
function and for developing techniques for lake management
that are not dependent on chemical and mechanical means.
All research done in the field of biomanipulation
supports the conclusion that the presence of large-bodied
zooplankton, Daphnia spp., is required for a strong control
over phytoplankton populations. Unfortunately, not all
geographic areas meet this requirement. Tropical and
subtropical areas, namely Florida and other areas of the
southeastern U.S., do not have large species of zooplankton.
Crisman and Beaver (1990) described only the presence of
small Cladocera (e.g., Eubosmina spp., Ceriodaphnia spp.) in
Florida lakes which are not good candidates for top-down
biomanipulation purposes.
Lacking the presence of the most studied and accepted
candidate for the top-down type of biomanipulation, tropical
and subtropical researchers have had to seek another
organism which could perform a corresponding role. Filter
feeding fish, which are very common in those areas, seem to
be the most suitable candidates. Filter-feeders do not
visually detect individual prey items, but engulf a volume
of water containing the food organisms and retain the
planktonic prey and particles by passing this volume over
entrapment structures (Lazzaro, 1987).
The most common filter-feeding fishes in Florida
eutrophic lakes are gizzard shad (Dqrqsqma cepedianum),
which can produce a total grazing pressure on algal


Table 4-5. Total nitrogen (mg.g-1
Lake Bonny sediment core.
| Interval (cm)
TN (mg/g)
I 0-2
28.0057
I 2-4
26.5107
4-7
23.1092
7-9
23.6192
9-11
27.1484
11-13
20.0185
1 13-15
20.4545
15-17
22.2599
17-19
21.5829
19-22
19.1497
22-26
20.1250
26-30
21.7168
30-34
23.6266
34-38
22.8466
38-42
24.6686
42-46
19.9064
I 46-50
18.0703
| 50-54
20.5948
54-58
20.3690
58-64
24.4444
64-72
19.5555
72-84
23.2875
| Mean
22.32
for


141
biomanipulation techniques employing fish as a major element
could lead to a significant improvement in water clarity in
subtropical and tropical lakes. This study certainly
provides evidence that gizzard shad and blue tilapia feces
have a different composition and may stimulate different
plankton community dynamics.
The results of this investigation suggest intriguing
possibilities for the potential role of filter feeding fish
in the control of eutrophication in Florida lacustrine
systems. Gizzard shad feces clearly stimulated algal growth,
while blue tilapia feces appeared to suppress algal growth.
However, additional research is necessary in the use of
filter feeding fish in biomanipulation schemes.


i Prod (mgC/m3/h)
Phyto (NU/ml x 1CT3)
Resp (mgC/m3/h)
7nnn v 10*9^
Figure 5-1. Phytoplankton and zooplankton population for all study lakes.
Values for primary productivity and community respiration are shown.
130


105
by Canfield and Hoyer (1992). Past limnological data
indicate that this lake has been hypereutrophic for at least
the last 25 years (PCWR, 1992). Table 5-9 shows water
quality data for Lake Hollingsworth.
Lake Hollingsworth displayed seasonality for its gross
primary productivity; the highest values were recorded in
late fall and mid-summer. Overall, the surface layer had the
highest values, with an annual mean of 369.88 mgC.'3.h
141.2; the minimum recorded value was 178.9 mgC.-3.h, and
the maximum was 506.34 mgC.m~3.h (Table 5-3). Net
photosynthesis is reported in table 5-4, community
respiration in table 5-5, and table 5-6 shows the average
annual productivity values for all study lakes.
Historical phytoplankton data for Lake Hollingsworth
from 1968 and 1971 (USGS, 1968, 1971) show the Cyanophyta as
the dominant group, especially species of Oscillatoria sp.,
Lynqbya sp., and Raphidiopsis sp., followed by Chlorophyta
and Diatoms. During this study, the phytoplankton community
was almost completely dominated by the green algae
Ankistrodesmus sp. and Cosmarium sp. and the blue green alga
Spirulina sp. The latter was replaced sometimes by
Aphanizomenon sp. and Anabaena sp. The diatoms were
represented by the genus Navicula.
Rotifers were the dominant zooplankton group throughout
the year. Keratella sp., Brachionus sp., and Monostvla sp.,
were the dominant taxa. The only adult copepod identified


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Thomas L. Crisman, Chair
Professor of Environmental
Engineering Sciences
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
G. Ronnie Best
Scientist of Environmental
Engineering Sciences
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Edward J.
Associate Professor of
Forest Resources and Conservation
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Frank G. Nordlie
Professor of Zoology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Pjjilosophy.
Horst 0. Schwassmann
Professor Emeritus of Zoology


51
importance of this species in biomanipulation schemes. It is
known that adult blue tilapia employ filter-feeding as their
primary feeding method, while they feed as visually-
oriented, particulate-feeding zooplanktivores as juveniles
(Yowell and Vinyard, 1993). It is assumed that minimization
of feeding costs while maximizing net energy return has been
the basis for the filter-feeding strategy. The remaining
guestion is what causes these fishes to switch their feeding
behavior.
Facultative planktivores that use both feeding modes
interchangeably (Leon and OConnel, 1969; Crowder and
Binkowski, 1983; and Helfman, 1990) usually filter-feed on
small, abundant zooplankton and particulate-feed on large,
less abundant forms. Yowell and Vinyard (1993), suggest that
this is a stratagem which maximizes net energy return in
response to a changing prey situation. The reduction in
zooplankton density may then allow an increase in non-grazed
phytoplankton (Drenner et al., 1987; Vinyard et al., 1988).
The results of this study indicate that blue tilapia
fecal material, either by its composition and/or the absence
of some essential nutrient, completely suppressed
Microcystis aeruginosa chlorophyll a production, and appears
to have suppressed Selenastrum capricornutum chlorophyll a
production in > 60% of the samples. Gizzard shad feces
increased chlorophyll a production values in both blue-
greens and greens (tenfold increase). This study suggests


5
Opuszynsky and Shireman, unpubl.) have developed a new
approach to improve the use of filter-feeding fish to
counteract eutrophication. They constructed an apparatus
consisting of a cage where fish are kept that was equipped
with funnels under the cage to collect fish feces for
estimating food consumption via a quantification of the
production of feces. According to these researchers, this
cage combines two indispensable features for an effective
use of filter-feeding fish to control planktonic algal
growth and to reduce eutrophication: [1] it enables the
coexistence of filter-feeding fish and zooplankton by
eliminating the fish from the water column thus improving
phytoplankton consumption; and [2] fish feces can easily be
collected and removed from the bottom of the cage,
eliminating a major source of nutrient for the phytoplankton
community.
There is a high degree of trophic overlap between young
blue tilapia and larval gizzard shad in Lake George, Florida
(Zale, 1984). Beaver and Crisman (1989) reported
phytoplankton and zooplankton community alterations in
eutrophic Florida lakes when blue tilapia has an established
population. In eutrophic central Florida lakes, blue tilapia
is quickly replacing gizzard shad as a major filter-feeding
fish (Beaver and Crisman, 1989).
The current study addresses gaps on current knowledge
regarding fish as biomanipulation agents for phytoplankton.


39
the fourth day (Figure 3-2). For tilapia feces, the medium
deficient of nitrogen (-N), as well as the medium deficient
of phosphorus (-P), showed a daily decrease in fluorescent
values and on the fourth day of the experiment they
stabilized in a value 50% lower than the initial number.
(Figure 3-2). Shad feces displayed a five fold increase in
fluorometric values compared with the measured in the
medium with tilapia feces.
Phytoplankton biomass in the fecal material was
calculated from chlorophyll a concentrations assuming 1 mg
chlorophyll a = 67 mg dry weight (APHA, 1982). Blue tilapia
fecal material suppressed phytoplankton biomass (mg.dry
weight) in 75% of the samples of blue green and green algal
prepared in -N and -P media. Mean suppression in these
samples was approximately 50% of the initial values (Table
3-4). Gizzard shad fecal material, on the other hand,
increased phytoplankton biomass in all of the samples of
blue green and green algae. The mean concentration increase
was greater than tenfold of the initial values (Table 3-4).
Conclusions
Adult fishes that employ filter-feeding as their
primary feeding method are typically visual particulate
feeding planktivores when juveniles. They switch from a diet


120
density residential development. The most commonly
encountered plants are Typha sp., Ludwiqia repens, and
Pontederia cordata (Canfield and Hoyer, 1992). Macrophyte
chemical control measures are employed for water hyacinth,
water lettuce, and hydrilla.
Lake Bonny was hypereutrophic during this study
according to Forsberg and Ryding (1980). Historical data
indicate that water quality has been improving since
approximately 1986, a time when the lake almost dried
completely. Table 5-18 shows water quality data for Lake
Bonny.
Lake Bonny did not display any seasonality for gross
primary productivity. The only noticeable trend was that
phytoplankton productivity was always greater at the
surface, with mean annual value of 448.04 mgC.m_3.h + 73.2;
the minimum was 345.12 mgC.m'3.h. at April, and the maximum
was 515.6 mgC.m"3.h at July (Table 5-3). Net photosynthesis
is reported in table 5-4, community respiration in table
5-5, and average annual productivity in table 5-6.
Species of blue green algae including Anabaena sp.,
Spirulina sp., Merismopedia sp., and Lynqbya sp. dominated
the phytoplankton assemblage of Lake Bonny. Green algae
(Closterium sp., Cosmarium sp., and Scenedesmus sp.) were
present, as well as the dinoflagellate Peridinium sp., and
the diatom Stauroneis sp.


149
HOLANOV, S.H. and I.C. Tash. 1978. Particulate and filter
feeding in threadfin shad, (Dorosoma petenense). at
different light intensities. J. Fish. Biol. 13:619-625.
HOSPER, S.H. and M.-L. Meijer. 1986. Control of phosphorus
loading and flushing as restoration methods for Lake
Veluwe, The Netherlands. Hydrobiol. Bull. 20:183-194.
HRBACEK, J., M. Dvorakova, V. Korinek and L. Prochazkova.
1961. Demonstration of the effect of the fish stock on
the species composition of zooplankton and the
intensity of metabolism of the whole plankton
association. Verh. Int. Ver. Limnol. 14:192-195.
HUBER, W.C., P.L. Brezonik, J.P. Heaney, R.E. Dickinson,
S.D. Preston, D.S. Dwornik and M.A. DeMaio. 1982. A
classification of Florida lakes. Rep. ENV-05-82-1 to
Florida Dep. Environ. Reg., Tallahassee.
HURLBERT, S.M. 1984. Pseudoreplication and the design of
ecological field experiments. Ecol. Monogr. 54:187-211.
HUTCHINSON, G.E. 1957. A Treatise on Limnology.I Geography,
Physics, and Chemistry. John Wiley and Sons, Inc, New
York, 1015 pp.
JANUSZKO, M. 1974. The effect of three species of
phytophagous fish on algae development. Pol. Arch.
Hydrobiol. 21:431-454.
. 1978. The influence of silver carp
(Hypophthalmichthvs molitrix) on eutrophication of the
environment of carp ponds. III. Phytoplankton. Rocz.
Nauk Roln. ser. H 99:55-79.
JESTER, D.B. and B.L. Jensen. 1972. Life history and ecology
of the gizzard shad, Dorosoma cepedianum (Le Sueur),
with reference to Elephant Butte Lake, New Mexico State
University, Agricultural Experiment Station Research
Report 218:1-56.
JOHNSON, T.C., J.D. Halfman and W.J. Showers. 1991.
Paleoclimate of the past 4000 years at Lake Turkana,
Kenya, based on the isotopic composition of authigenic
calcite. Paleogeography, Paleoclimatology,
Paleoecology 85:19-198.
KAJAK, Z., J. Zawiska and A. Hillbricht-Ilkowska. 1976.
Effect of experimentally increased fish stock on
biocenosis and recovery processes of a pond type lake.
Limnologia, Berlin 10(2):595-601.


127
Diaptomus was the only copepod identified. Data from
Canfield and Hoyer (1992) show a rotifer density of 31,200
individuals .m'3, and a copepod density of 14,000
individuals .m-3.
Eighteen species of fish were collected in Lake Hunter
(Canfield and Hoyer, 1992). Historically, Lake Hunter was a
hypereutrophic lake with an overabundance of forage fish,
principally gizzard shad and threadfin shad (Huish, 1955;
Ware and Horel, 1971). After the lake restoration done in
1983 and 1984, 10,000 largemouth bass and 1,000 sunshine
bass were stocked to establish a predator population (Moxley
et al. 1984). After 8 years, the most abundant open-water
species collected in experimental gillnets were gizzard shad
and sunshine bass, with 163 and 8 fish.net-1.24hr,
respectively (Canfield and Hoyer, 1992).
ANOVAs were performed using the General Linear Model
(GLM) (SAS, 1989) in order to identify significant seasonal
limnological factors (p < 0.05). The following parameters
were analyzed: Secchi, temperature, pH, conductivity,
dissolved oxygen, total suspended solids, turbidity,
chlorophyll a, color, alkalinity, NH3, TKN, total nitrogen,
total phosphorus, orthophosphate, and total bacteria.
The data were collected between 1989 and 1993, and
grouped into four seasons: winter (December to February);
spring (March to May); summer (June to August); and fall
(September to November). The results in the Table 5-22 are


1400
1200
1000
800
600
400
200
0
Figur
.rrT3
F
Control Tilapia Shad
Time
3-5. Bioassay initial (I) and final (F) values of chi. a (mg.itf3)
with S.capricornutum in medium -N.


138
Lake Gibson had less organic matter in its superficial
layers than did Lake Bonny. However, values of up to 40%
organic content were registered throughout the core, which
coincides with the average organic matter content of dry
bulk sediment in Florida lakes of 39.7% (Brenner and
Binford, 1988). Total nitrogen and total phosphorus declined
at depths > 24 cm in the core. Below 48 cm in the core,
total phosphorus was undetectable.
Total phosphorus reconstructions at the base of the
core suggest that the lake was oligotrophic for at least the
last century. Nevertheless, values for TP exceeding 3.0
mg.g'1 were registered for the sediment corresponding to
the last 25 years, reflecting changes in the watershed use.
Using 210Pb detrminations, it appears that the lake had
much less organic sediment before the 1950s. Detectable
137Cs coincided with the determined 210Pb chronology.
Lakes Parker and Hollingsworth, both defined in this
study as having a blue tilapia dominant fish population,
were the subject of a sediment survey done by Schelske et
al. (1992). They stated that Lake Parker has always been
mesotrophic to eutrophic. Analyses of diatom assemblages
indicated the presence of alkaline conditions. Determination
of total phosphorus in the topmost 15 cm of sediment
suggested eutrophic conditions. However, these analyses
further indicated a progressive decline in water column
total phosphorus that may signal some reversal of cultural


145
COHEN, A.S. and C. Nielson. 1986. Ostracodes as indicators
of paleohydrochemistry in lakes: a late Quaternary
example from Lake Elementeita, Kenya. Palaios 1:601-
609.
COOKE, G.D., E.B. Welch, S.A. Peterson, and P.R. Newroth.
1993. Restoration and management of lakes and
reservoirs. Second edition. Lewis Publishers, Boca
Raton, Fl.
COURTENAY, W.R.Jr. and C.R. Robins. 1973. Exotic aquatic
organisms in Florida with emphasis on fishes: a review
and recommendations. Trans. Am. Fish. Soc. 102: 1-12.
, and J.R. Stauffer, Jr. 1984. Distribution, Biology,
and Management of Exotic Fishes. The Johns Hopkins
University Press, Baltimore. 430 pp.
COVENEY, M.F., G. Cronberg, M. Enell, K. Larson, and L.
Olofsson. 1977. Phytoplankton, zooplankton and bacteria
standing crop and production relationships in a
eutrophic lake. Oikos 29:5-21.
COWELL, B.C., C.W. Dye and R.C. Adams. 1975. A synoptic
study of the limnology of Lake Thonotosassa, Florida.
Part I: Effects of primary treated sewage and citrus
wastes. Hydrobiologia 46(2-3):301-345.
, and D.S. Vodopich. 1981. Distribution and
seasonal abundance of benthic macroinvertebrates in a
subtropical Florida lake. Hydrobiologia 78:97-105.
CRAMER, J.D. and G.R. Marzolf. 1970. Selective predation on
zooplankton by gizzard shad. Trans. Amer. Fish. Soc.
99:320-332.
CRISMAN, T.L. 1980. Chydorid Cladoceran assemblages from
subtropical Florida. Pages 599-608 in W.C. Kerfoot
(ed), Evolution and Ecology of Zooplankton
Communities. Univ.Press of New England, Hanover, N.H.
. 1992. Natural lakes of the southeastern
United States: Origin, structure, and function. Pages
387-405 in W.H.Martin (ed), Biotic Communities of the
Southeastern United States. Wiley Press, New York.
, and J.R. Beaver. 1988. Lake Apopka trophic
structure manipulation Phase I Final Project Report.
St. Johns River Water Management District,
Jacksonville, FL. 127 pp.


3
populations even greater than the large bodied zooplankton
in temperate systems (cf. Drenner et al., 1982a). Another
very common filter-feeder fish inhabiting Florida eutrophic
lakes is blue tilapia (Tilapia aurea) which was introduced
into the United States in 1961. Both of these fishes feed on
small, particulate material. For example, gizzard shad 5 cm
in length can filter algae >19ju in size; those 15 cm in
length can filter algae >40jj; and those 25 cm in length can
filter algae >63^. Blue tilapia can consume algae >25p in
size (Opuszinsky and Shireman, unpubl.).
The filtering rates of these species can be very high.
According to Drenner et al. (1982a), the gizzard shad
population in Lake Barkley, Texas (85 ha), can effectively
filter a volume equivalent to the entire lake every 2.3
days.
Undoubtedly, filter-feeding fishes can alter the
phytoplankton community of aquatic systems. Nevertheless,
contrary to expectations, an increase in planktonic algae
biomass and primary production has been observed (Drenner et
al., 1984, 1986; Janusko, 1974,1978; Opuszynski, 1978). Some
of the reasons for the failure of filter-feeding fish to
control algal blooms and in some cases even worsen water
quality are [1] elimination of larger zooplankton species;
[2] more rapid cycling of plant nutrients (Opuszynski and
Shireman, unpubl); and [3] differential release of fecal
material by individual species of fish.


140
Cluster analysis (Pearson Correlation Coefficient)
(SAS, 1989) divided the six lakes into 4 aggregations. Lakes
Hunter and Bonny, both with a prevalent population of
gizzard shad, were the most statistically similar lakes in
this study. Lake Parker, a blue tilapia dominant lake,
displayed limnological characteristics that differed
significantly from the mean values reported for the other
study lakes.
Cluster analyses of lakes by fish species yielded
similar relationships. Lake Parker again did not show any
correlation with the other study lakes. Seasonally, summer
and fall appeared to be the most stable seasons of the year
for the study lakes, and winter is the one which exhibited
the greatest statistical variability. This observation is
certainly a consequence of temperature, with minimum annual
temperatures occurring during the winter months.
Primary productivity determinations showed high gross
photosynthesis for most of the lakes, with a mean of 334.22
mgC.m'3.h + 106.35. The only deviation occurred in Lake
Parker, where a mean value of 66.41 mgC.m'3.h was recorded.
Conclusion
In view of the evidence presented in this study,
coupled with the findings of mesocosm experiments performed
by Crisman and Kennedy (1982), Drenner et al. (1986), and
Crisman and Beaver (1988), it seems possible that


Time (year)
+~ % Dry matter % Organic matter
Nitrogen (mg/g) ^ Phosphorus (mg/g)
Figure 4-7. Changes in some physicochemical characteristics in
Lake Bonny through time.
00
u>


60
Samples for isotope analysis were dried at 95C for 24
hours, pulverized by mortar and pestle, weighed, and placed
in small plastic vials. Core sections were combined (up to 4
cm) to obtain an adequate sample weight (> 1 g). Vials were
sealed with plastic cement and left for 14 days to
equilibrate radon (222Rn) with radium (226Ra) Counting times
were never less than 23 hours. Standards were counted to
track efficiency (counts per y) and to calculate a 226Ra
conversion factor (PCi per count per sec.). Blanks were
counted to determine background radiation.
Bulk density and water content of sediments were
measured by weighing samples before and after drying at
95C. Percent organic matter was evaluated by loss on
ignition (LOI) at 550C for one hour (Hkanson and Jansson,
1983). Nitrogen was measured as total Kjeldahl nitrogen
(TKN) using a Technicon-II semi-automated manifold after
digestion following Bremner and Mulvaney (1982) but modified
to exclude selenium as a catalyst. The digestate was also
used for total phosphorus (TP) determinations. Liberated
orthophosphate was determined with the ascorbic acid method
(APHA, 1985).
Results and Discussion
The Lake Bonny Record
The 89 cm Lake Bonny sediment core had a mean bulk
density of 0.094 g.cm-3 0.042, mean dry matter of 9.16%


at the beginning and end of the experiment from each flask
was performed, using a 2 ml chamber on an inverted Leitz
microscope.
31
Results and Discussion
Bulk density calculated for tilapia feces had a mean
value of 0.2576 g.cm'3, with a minimum value of 0.2326 g.cm"
3, and maximum value of 0.2815 g.cm'3 (Table 3-1). For shad
feces, the mean value was 0.1401 g.cm"3, with a minimum
value of 0.1258 g.cm'3 and maximum value of 0.1493 g.cm"3
(Table 3-1).
Percent dry matter for tilapia feces had a mean value
of 13.0 %, with a minimum value of 12.6 %, and a maximum
value of 13.6 % (Table 3-1). For shad, the mean value was
8.0%, with a minimum value of 7.5 % and a maximum value of
8.6 % (Table 3-1). Percent organic matter calculated from
tilapia feces had a mean value of 49.6%, with a minimum
value of 48.6 %, and maximum of 51.6 % (Table 3-1). Shad had
a mean value of 68.0 %, with a minimum of 67.0 %, and a
maximum of 68.9 % (Table 3-1).
Percent protein was calculated for tilapia and had a
mean value of 21.6 %, minimum of 21.0 %, and maximum of
23.2 % (Table 3-1). For shad, the mean value was 32.4 %,
with a minimum of 30.25 %, and maximum of 33.75 %
(Table 3-1).


125
primrose, and pennywort. There is no aquatic weed control
program for Lake Hunter.
According to Forsberg and Ryding (1980), Lake Hunter
was hypereutrophic during this study. In 1983, Lake Hunter
was completely drained, and water quality has continued to
decline since then. Table 5-21 show water quality data for
Lake Hunter.
Lake Hunter does not have much seasonality in its gross
primary productivity (Table 5-3). The most productive layer
is at 50 cm water depth. The mean for the 50 cm depth was
306.71 mgC.m"3.h 175.6 throughout the year, with a minimum
value of 115.62 mgC.m'3.h in December and a maximum of
484.37 mgC.m_3.h in April. Values for net photosynthesis are
shown in table 5-4, community respiration in table 5-5, and
average annual productivity in table 5-6.
The phytoplankton community in Lake Hunter had
Cyanophytes as dominants with Chlorophytes being co
dominant. Species of blue-green algae such as Lvnabva sp.,
Spirulina sp., Microcystis sp..Merismopedia sp., and
Anabaena sp. were present as well as green algae such as
Closterium sp., and Scenedesmus sp.; the dinoflagellate
Peridinium sp. were abundant. No diatoms were reported for
this lake.
Rotifers and copepods comprised the dominant
zooplankton community for Lake Hunter. Dominant rotifers
were Keratella sp., Brachionus sp., and Monostvla sp.


155
, R.D. Gulati and M.P. Grimm. 1989. Food web
manipulation in Lake Zwemlust: positive and negative
effects during the first two years. Hydrobiol. Bull.
23:19-34.
Van LIERE, E. 1986. Loosdrecht lakes, origin,
eutrophication, restoration and research program.
Hydrobiol. Bull. 20:9-15.
VELASQUEZ, G.T. 1939. On the viability of alga obtained from
the digestive tract of the gizzard shad, Dorosoma
cepedianum (LeSeur). Am. Midi. Nat. 22:376-412.
VINYARD, G.L., R.W. Drenner, M. Gophen, U. Pollingher, D.L.
Winkelman and K.D. Hambright. 1988. An experimental
study of the plankton community impacts of two
omnivorous filter-feeding cichlids, Tilapia qalilaea
and Tilapia aurea. Can.J. Fish. Aquat. Sci. 45:685-690.
VOLLENWEIDER, R.A. 1974. A Manual on Methods for Measuring
Primary Production in Aquatic Environments. IBP
Handbook no. 12. Blackwell Scientific Publications,
Oxford. 225 pp.
WATKINS, C.E.,II J. V. Shireman and W.T. Haller. 1983. The
influence of aquatic vegetation upon zooplankton and
benthic macroinvertebrates in Orange Lake, Florida. J.
Aquat. Plant Manage. 21:78-83.
WARE, F.J. 1973. Status and impact of Tilapia aurea after
twelve years in Florida. Unpubl. Rep. Florida Game
Fresh Water Fish Comm., Tallahassee, Florida.
, and G. Horel. 1971. 1970-71 annual progress report
for research and development project. Florida Game and
Fresh Water Fish Commission, Federal Aid in Fish
Restoration Dingell-Johnson Project F-12-12 Lake
Management, Research and Development, Tallahassee,
Florida.
, R.D. Gasaway, R.A. Martz and T.F. Drda. 1975.
Investigations of herbivorous fishes in Florida, pp.
79-84 In Water Quality Management Through Biological
Control. Report n ENV-07-75-1. Department of
Environmental Engineering Sciences, University of
Florida, Gainesville, FI.
WETZEL, R.G. 1983. Limnology. 2nd edition. Ed. Saunders
Coll., Philadelphia.
/ and G.E. Likens. 1991. Limnological Analyses. 2nd
edition. Springer-Verlag, New York.


110
been eliminated. Known discharges are coming from a Coca-
Cola Citrus processing plant, Adams Citrus processing plant
and Florida Distillers via Lake Lena Run. Agrico Chemical
discharges directly into the lake. There are no known
withdrawals.
Lake Hancock's drainage basin is extremely large and
includes portions of the City of Lakeland (75,000
inhabitants) and Auburndale (18,000 inhabitants). Non-urban
land-use areas is split between mined areas and agricultural
development. The dominant vegetation (defined as greater
than or equal to 1% of macrophyte cover cf. PCWR, 1992)
includes cattails, pickerelweed, and pennywort. Macrophyte
chemical control is employed for water hyacinth, water
lettuce, and hydrilla.
Lake Hancock was reported as an alkaline, eutrophic,
hard water lake by Canfield (1981). Using the criteria of
Forsberg and Ryding (1980), Lake Hancock was classified as
hypereutrophic during this study. Lake water quality has
been improving for at least the last 10 years, although the
lake is still the most eutrophic lake in Polk County (Polk
County Water Resources Division, 1990). Table 5-12 shows
water quality data for Lake Hancock.
Lake Hancock did not display much seasonality for gross
primary productivity. The highest values were always
measured at the surface with an annual mean of 409.42 mgC.m'
3.h + 184.4, minimum value of 187.5 mgC.m'3.h in April and


6
I worked with the digestive physiology of blue tilapia and
gizzard shad with an emphasis on their fecal material, as
well as with the limnological characteristics and
paleoliranological interpretations from six central Florida
lakes. The research addressed the following hypotheses: [1]
There is differential nutrient bioavailability from blue
tilapia and gizzard shad feces; [2] different type of feces
will produce a different effect on the lake ecosystem; [3]
fundamental differences in the physiology and biochemistry
of blue tilapia and gizzard shad feces will impact the
sediment composition in lakes; and [4] interlake differences
should exist in key limnological parameters between lakes of
differing relative abundance of blue tilapia and gizzard
shad. The goals of this study are to [1] learn something
about the fecal material composition of filter-feeding fish
to help understand their role in freshwater systems; [2]
evaluate management techniques that couple physicochemical
water information with biotic community components and the
lake sedimentary record; and [3] gather new information on
the effect of different fish species' fecal material on lake
ecosystems.


57
Using the classification system of Forsberg and Ryding
(1980), Lake Bonny is classified as hypereutrophic. Average
total phosphorus concentration is 59 pg.l'1 and average
total nitrogen concentration is 1858 pg.l"1. Total
chlorophyll a concentrations average 40 pg.l"1, the water
clarity as measured by a Secchi disc averages 0.6 m, and
average pH is 7.8. Table 5-10 exhibits water quality data
for Lake Bonny.
Before 1985, the plant community of Lake Bonny was
comprised mainly of Hvdrilla verticillata. Due to the
macrophyte chemical control program employed by the City of
Lakeland, it has been replaced by Typha sp. as the dominant
macrophyte. Typha sp. is responsible for a percent lake area
coverage (PAC) of 10% (cf. Canfield and Hoyer, 1992).
The fish population in this lake has changed
dramatically. The species that survived the drought of 1983
are currently repopulating the lake. The most abundant open-
water species collected in experimental gillnets are gizzard
shad and Florida gar with 17.0 and 11.3 fish/net/24 hr,
respectively (Canfield and Hoyer, 1992).
Lake Gibson is a small (192 ha of surface area),
shallow lake (zniax= 6.1 m) located on the outskirts of the
City of Lakeland in a rapidly urbanizing area where citrus
groves and pastures are giving way to large-scale commercial
and moderate-density residential development. A domestic
wastewater treatment plant from one local elementary school


65
Sedimentation rates expressed in Figure 4-3 for Lake
Bonny are in agreement with the lake history. The first
peak of 0.1 g.cm'2.yr coincides with the 1955 development
boom in the area, and the second peak of 0.13 g.cm'2.yr in
1983 coincides with the time when the lake was almost
completely dry.
Detectable 137Cs, which is an human-produced
radionuclide that was injected into the atmosphere as a
consequence of nuclear weapons testing in the early 1950s,
matched well with the determined 210Pb chronology (Table 4-
2). This agreement is remarkable, considering the
shallowness of the lake and the flocculence of the bottom
substrate which makes these upper sediments vulnerable to
physical disturbance. Eutrophic systems such as Lake Bonny,
however, accumulate sediments rapidly, so these physical
disturbances likely affect short time-intervals only.
The Lake Gibson Record
The sediment core from Lake Gibson had a mean bulk
density of 0.279 g.cm'3 0.319, mean dry matter of 21.61%
13.93, mean organic matter of 38.17% 14.85, mean total
nitrogen of 9.33 mg.g'1 2.99, and mean total phosphorus of
1.86 mg.g'1 1.49. (Table 4-1).
The organic content of Lake Gibson sediments was
somewhat low in surface deposits (30% organic matter),
increased with depth to about the midpoint of the core,
matching with 1955, and declined to its lowest values of


28
from the T-2 jar. Three aliquots for analysis (replicates)
were taken from each of the four groups.
Gizzard shad were collected using an experimental
monofilament gillnet. All fish sampled were collected as
part of a population control project being conducted by the
Florida Game and Freshwater Fish Commission, Eustis (FL).
The sampling net had 6 panels ranging from 6.3 to 12.7 cm
stretch mesh in 1.3 cm increments. Gillnetting was done in
Lake Denham, Florida, in increments of two hours.
Gizzard shad were measured for total length (mm) and
weight (g). All fish collected were separated into two size
groups (TL), <320-mm and >320-ram, which will subsequently be
referred to as size groups S-l and S-2, respectively. A
total of 218 fish were collected in five different gillnet
settings. Minimum fish length was 203 mm, and the maximum
length was 440 mm. Fecal material was hand-stripped in the
field from all the captured fishes and poured into two jars
(S-l and S-2). The sample jars were immediately placed on
ice in a cooler for transport to the laboratory. All sample
material was kept at 4C within four hours after collection.
Samples were subdivided into four different groups;
groups 1 and 2 pulled from the T-l jar and groups 3 and 4
from the T-2 jar. Three aliquots for analysis (replicates)
were taken from each one of the four groups.


80
Table 4-8. Correlation matrix for sediment variables
for Lake Bonny.
SEDIMENT VARIABLES
CORRELATION
P VALUE
| Depth/ Bulk Density
0.7848
0.0000
| Depth/ Dry Matter
0.8255
0.0000
Depth/ Organic Matter
0.5482
0.0123
Depth/ Total Phosphorus
-0.8135
0.0000
| Bulk Density/T Nitrogen
-0.5724
0.0084
| Bulk Density/T Phosphorus
-0.6747
0.0011
Table 4-9. Correlation matrix for sediment variable
for Lake Gibson.
SEDIMENT VARIABLES
CORRELATION
P VALUE
| Depth/ Bulk Density
0.5862
0.0066
Depth/ Total Phosphorus
-0.9054
0.0000
Bulk Density/ Org. Matter
-0.7335
0.0002
Org. Matter/T Nitrogen
0.6123
0.0041


82
changed from 2.03 mg.g'1 at the base of the core to 6.60
mg.g'1 at the top (Schelske et al., 1992). Total phosphorus
in the water column measured during this study had a mean of
0.27 rng.L"1 (0.081 to 0.61 mg.L"1)
Due to the geology in the area, Lake Bonny has been an
eutrophic lake for much of its recent history, even with
conditions of minimal human occupation in the watershed
(Figure 4-7). This should be the foundation for any
management plan for this lake. Any plans for water quality
improvement in the lake must consider that the lake is
edaphically phosphorus-rich, and even a great effort to
reduce phosphorus concentrations in the water may provide
little or unnoticeable improvement in water clarity.
Lake Gibson, meanwhile, had a different
paleolimnological record that displayed alterations during
the last 25 years (Figure 4-8) in which the lake changed
from mesotrophic to eutrophic, reflecting the switch in past
largely agricultural watershed uses to large scale
commercial and moderate density residential development. Any
attempt to reduce nutrient concentration in this lake must
consider control of point source external loading of
nutrients as well as use of biomanipulation techniques as a
means to achieve water quality improvement.
Figure 4-9 show the historic variation of organic
matter content, nitrogen and phosphorus in the Lake
Hollingsworth sediment core cf. Schelske et al. 1992. It is


Table 5-16. Correlation analyses of water quality data for Lake Gibson
using the General Linear Model procedure (GLM). Data were combined by
year, season and year*season.
PARAMETER
YEAR
SEASON
YEAR*SEASON
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
Secchi
X
X
X
Temperature
X
X
X
PH
X
X
X
Conductivity
X
X
X
D.O.
X
X
X
TSS
X
X
X
Turbidity
X
X
X
Chlorophyll a
X
X
X
Color
X
X
X
Alkalinity
X
X
X
NH,
X
X
X
TKN
X
X
X
TN
X
X
X
TP
X
X
X
ORT
X
X
X
BACTERIA
X
X
X
119


74
Table 4-4. Total nitrogen (mg.g-1)
Lake Gibson sediment core.
| Interval (cm)
TN (mg/g)
0-2
10.4995
2-4
10.5649
4-6
9.1080
6-8
8.4608
8-10
7.8340
10-12
7.4951
12-14
10.3830
14-16
11.0008
16-18
11.6984
18-20
13.4574
20-24
13.2149
24-28
12.2212
28-32
7.2597
32-36
9.9815
36-40
10.4658
40-44
10.6539
44-48
7.2012
48-52
3.3171
52-56
1.4618
1 56-60
10.4134
| Mean
9.33
for


Table 4-2. Lake Bonny sediment analysis.
Core
Section
(cm)
Bulk
Density
g. cm'3
2iopb
pCi. g'1
137Cs
pCi. g'1
Deposition
Period Rate
year g.cr2.yr
TN
mg. g'1
TP
mg. g'1
0-4
0.029
18.72
1.5
1993
0.05
28.00
5.16
4-7
0.050
12.30
3.0
1991
0.08
23.11
5.29
7-9
0.061
12.50
4.13
1989
0.07
23.62
5.35
9-11
0.066
11.22
4.01
1987
0.07
27.15
6.03
11 13
0.074
8.46
3.65
1985
0.09
20.02
4.62
13 15
0.078
8.14
3.20
1983
0.09
20.45
4.81
15 17
0.079
5.27
4.02
1982
0.13
22.26
5.21
17 19
0.090
6.99
4.56
1980
0.10
21.58
3.53
19 22
0.092
8.92
4.49
1978
0.07
19.15
4.93
22 26
0.092
6.51
3.28
1974
0.08
20.12
5.86
26 30
0.090
7.00
2.67
1970
0.07
21.72
6.38
30 34
0.089
6.00
1.07
1963
0.07
23.63
5.07
34 38
0.091
3.80
0.85
1958
0.09
22.85
3.64
38 42
0.102
3.43
0.92
1953
0.08
24.67
3.68
42 46
0.164
3.48
0.87
1948
0.07
19.9
3.03
46 50
0.174
3.17
0.78
1937
0.06
18.07
2.42
50 54
0.142
1.91
0.91
1921
0.06
20.59
2.49
54 58
0.137
1.94
0.59
1909
0.04
20.36
2.22


129
reported as significant at p < 0.05. Temperature,
conductivity, dissolved oxygen, TSS, chlorophyll a, and
orthophosphate exhibited significant seasonal variation.
Other parameters displayed either interannual or no
variation. When year and season were combined, Secchi, pH,
dissolved oxygen and turbidity displayed significant
variability.
Summary
For the scope of this study, the lakes were divided
into two groups: [1] lakes considered as having blue tilapia
as the major component of its filter-feeder fish population
(Lakes Parker, Hollingsworth, and Hancock), and [2] lakes
were gizzard shad is the dominant rough fish (Lakes Gibson,
Bonny and Hunter). Evaluating the limnology of the study
lakes, there seems to exist differences among them. Figure
5-1 has information on gross primary productivity (mgC.m'
3.h), community respiration (mgC.nf3.h), phytoplankton
(NU/ml) and zooplankton (individuals.m-3) for all lakes.
Lakes Parker and Hollingsworth have a significantly
larger (ANOVA, p <0.05) zooplankton population than Lakes
Bonny and Hunter, which could be indicative of the different
proportionality of filter feeder fish in the population.
Lakes where blue tilapia have proportionally high population
tend to have higher number of zooplankton, especially Lake


58
discharges to a wetland north of Lake Gibson, which
overflows into the lake. There are no known withdrawals
(Polk County Water Resources Division, 1990).
Lake Gibson is classified as eutrophic according to
Forsberg and Ryding's (1980) classification system. The 1992
Florida TSI for Lake Gibson is 55 (Polk County Water
Resources Division, 1990). The water is slightly tannic
(color is 43 cpu) which lowers the Secchi disc value and may
bias the TSI value. The average total phosphorus
concentration is 94 pg.l-1, and the average total nitrogen
concentration is 1570 pg.l'1. Total chlorophyll a
concentration averages 23 pg.l'1 and the water clarity as
measured by use of a Secchi disc averages 1.3 m. Lake's
annual primary productivity (mean = 241.18 mgC.m"3.h
229.1) displaying noticeable seasonal variability and the
presence of blooms of cyanobacteria every few years suggests
an eutrophic status for Lake Gibson. Water guality data for
Lake Gibson are shown in Table 5-8.
The dominant macrophytes of the lake are water
primrose, alligatorweed, and pennywort, which are
responsible for 5% lake area coverage (Canfield and Hoyer,
1992).
No recent fish population evaluation was obtainable for
this lake. From the 1979 evaluation, ten species of fish
were reported, with channel catfish being dominant.


Table 5-19. Correlation analyses of water quality data for Lake Bonny
using the General Linear Model procedure (GLM). Data were combined by
year, season and year*season.
PARAMETER
YEAR
SEASON
YEAR*SEASON
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
Secchi
X
X
X
Temperature
X
X
X
PH
X
X
X
Conductivity
X
X
X
O

Q
X
X
X
TSS
X
X
X
Turbidity
X
X
X
Chlorophyll a
X
X
X
Color
X
X
X
Alkalinity
X
X
X
NH,
X
X
X
TKN
X
X
X
TN
X
X
X
TP
X
X
X
ORT
X
X
X
BACTERIA
X
X
X
123


162
APPENDIX F
Physical and chemical properties of Lake
Hollingsworth sediment core (cf. Schelske et al., 1992).
Depth
(cm)
Density
(g. dry. cm'3. wet)
Organic
Matter
(%LOI)
Ctot
(%>
Ntot
(%)
^tot .
(mg. g'1)
0-2
0.01772
54.3
27.6
3.04
6.99
2-4
0.02082
51.8
27.4
3.35
7.24
4-6
0.02487
54.5
27.4
2.89
6.42
6-8
0.02693
52.8
26.0
3.02
7.32
8-10
0.02846
52.8
26.7
N/A
N/A
I 10-12
0.03139
52.2
26.1
2.92
7.03
12-14
0.03583
48.7
25.5
2.76
6.47
I 14-16
0.03796
48.8
25.4
2.71
6.58
16-18
0.04688
40.7
20.8
2.36
5.43
18-20
0.06550
30.5
14.9
2.17
5.13
20-22
0.09093
24.2
13.9
1.34
5.18
22-24
0.08989
25.8
13.8
1.66
6.47
24-26
0.09053
28.9
15.0
1.67
8.87
26-28
0.08030
34.3
17.3
1.97
9.78
28-30
0.08537
30.6
16.3
1.53
9.64
30-32
0.10669
25.7
14.8
1.35
7.52
40-42
0.07557
43.4
24.3
2.43
6.22
50-52
0.07180
49.8
27.8
2.70
4.30
| 60-62
0.07493
55.5
34.3
3.04
3.57
70-72
0.07412
56.7
33.8
3.16
3.96
80-82
0.06584
62.0
35.3
3.53
3.20
90-92
0.07100
55.5
34.4
2.99
2.29
98-100
0.08207
48.4
28.4
2.81
2.48


Table 4-3. Lake Gibson sediment analysis
Core
Section
(cm)
Bulk
Density
g. cm'3
210pb
pCi. g'1
137Cs
pCi. g'1
Deposition
Period Rate
year g.cr2.yr
TN
mg. g-1
TP
mg. g'1
0-2
0.113
11.86
4.41
1993
0.07
10.50
4.45
2-4
0.167
9.48
5.02
1990
0.08
10.56
4.57
4-6
0.221
9.31
4.18
1986
0.08
9.11
4.10
6-8
0.233
6.76
3.63
1979
0.08
8.46
3.77
8-10
0.271
5.72
2.96
1973
0.08
7.83
3.28
10 12
0.284
6.07
2.96
1966
0.06
7.49
3.03
12 14
0.229
2.33
1.32
1955
0.12
10.38
2.07
14 16
0.178
2.63
1.28
1951
0.09
11.00
1.62
16 18
0.168
2.61
0.84
1946
0.08
11.70
1.70
18 20
0.160
3.34
0.75
1942
0.05
13.46
1.55
20 24
0.161
2.29
0.71
1935
0.06
13.21
1.50
24 28
0.185
1.11
0.75
1923
0.09
12.22
1.28
28 32
0.208
1.42
0.57
1914
0.05
7.26
0.70
32 36
0.183
2.00
0.30
1892
0.02
9.98
0.89


124
Lake Hunter
Lake Hunter is located in Polk County, Florida
(Appendix A). The lake lies in the Bartow Embayment division
of the Central Lakes District (Canfield, 1981). Lake Hunter
is a small urban lake located within the city limits of
Lakeland, with a surface area of 41 ha, a shoreline length
of 2.6 km, and a mean depth of 1.5 m (Table 5-20).
Table 5-20. Morphometry
of Lake Hunterd)
Surface Area
41 ha
Maximum Depth
2.7 m
Mean Depth
1.5 m
Development of Shoreline 80 %
Drainage Basin Area
208 ha
Shoreline Length
2.6 km
Macrophyte Cover
0.2 %
(1) Data front PCWR (1992).
Development in the watershed has caused Lake Hunter to
be used as a stormwater retention area. Twelve stormwater
discharge inlets enter the lake, six of which can be
considered as a major source of pollution. The lake
watershed is almost entirely developed and consists of
commercial and moderate to large residential developments.
Most typical of urban systems, Lake Hunter experiences
excessive sedimentation, poor water guality, and rapid
growth of undesirable aquatic vegetation. The lake's
macrophyte dominant vegetation (defined as greater than or
equal to 5% macrophyte cover) includes elephant ear, water


29
Feces Analysis
The digestive tract was removed through an incision
made from the anus anterior to the base of the pelvic fin,
then dorsally to the posterior side of the pectoral girdle
to the dorsal side of the coelom. Fecal material was hand-
stripped from the beginning of the intestine to the base of
the anus and placed into a 250 ml jar. The feces were then
pooled according to size group (T-l, T-2, S-l and S-2).
Laboratory analyses consisted of determination of
percent dry matter, percent organic matter, bulk density,
heat content upon combustion, protein analysis, total
phosphorus and total nitrogen content and two algal
bioassays. Dry weight/water content of feces was measured by
weighing samples before and after drying at 105C for 16
hours; organic matter was measured using a cool muffle
furnace before ashing the samples at 600C for at least 3
hours and then weighing. Nitrogen was measured as total
Kjeldahl nitrogen (TKN) using a Technicon-II semi-automated
manifold following a modification of the methods described
by Bremner and Mulvaney (1982) (selenium was not used as a
catalyst). The digestate was also used for total phosphorus
(TP) determinations. Liberated orthophosphate was determined
with the ascorbic acid method (APHA, 1985).
All samples for calorimetry analyses were freeze dried.
Volatilization began at -30C. Total energy was determined
by oxygen bomb calorimetry (Parr Model 1261 Isoperibol


Organic matter (%) l Nitrogen (%) Phosphorus (mg/g)
Figure 4-9. Changes in some physicochemical characteristics in
Lake Hollingsworth through time (after Shelske et al.,1992).
00
in


CHAPTER 5
LIMNOLOGICAL ASSESSMENT
Introduction
Florida can be divided into three major physiographic
zones: the Northern, Central and Southern zones (Puri and
Vernon, 1964; White, 1970). All lakes reported in this study
are located in Polk County, Florida. Polk County lies within
the Central Highlands physiographic province. The vast
majority of the county lies within the Polk and Lake
Uplands. The Polk Uplands is an area of continuous high
ground located between the Gulf Coastal Lowlands and the
western edge of the Lake Wales Ridge (White, 1970). Along
the eastern ridge, deposits of the Fort Preston Formation
occur, but most of the region is underlain by deposits of
the phosphatic Hawthorn and Bone Valley Formations (Puri and
Vernon, 1964). In the eastern half of the Polk Uplands, most
lakes are in association with sand ridges. Polk County has
550 lakes covering 37.9 x 103 ha.
A majority of phosphatic sand deposits and clays of the
Miocene Bone Valley and Hawthorne Formations are located in
the Polk Uplands and Lake Uplands physiographic regions
(Puri and Vernon, 1964). Florida has led the nation in
89


104
where the bedrock is dominated by phosphatic deposits from
the Hawthorn and Bone Valley Formation. Lake Hollingsworth
is an urban lake situated within Lakeland, Florida, with a
surface area of 145 ha, shoreline length of 4.34 km and mean
depth of 1.5 m (Table 5-8).
able 5-8. Morphometry of Lake Hollingswortha:
Surface Area
145 ha
Maximum Depth
2.44
m
Mean Depth
1.13
m
Development of Shoreline
100 %
Drainage Basin Area
495 ha
Shoreline length
4.34
km
Macrophyte Cover
3 %
(1) Data from PCWR (1992).
No known discharges or withdrawals were reported for
Lake Hollingsworth. Its drainage basin is totally developed,
and land-use within the basin includes residential and
commercial development and Florida Southern College.
The dominant macrophytic vegetation (defined as greater than
or egual to 3% macrophyte cover cf. PCWR, 1992) includes
American lotus, elephant ear, and cattails. Chemical control
measures are employed for water hyacinth, water lettuce, and
hydrilla.
Using the Forsberg and Ryding (1980) classification,
Lake Hollingsworth was classified as hypereutrophic during
this study. The lake was also classified as hypereutrophic


CHAPTER 6 SUMMARY
133
Fish Feces Experiment 134
Paleolimnological Analysis 137
Lakes Assessment 139
Conclusion 140
LITERATURE CITED 142
APPENDICES 157
BIOGRAPHICAL SKETCH 165
v


118
chlorophyll a, color, alkalinity, NH3, TKN, total nitrogen,
total phosphorus, orthophosphate, and total bacteria.
Temperature, conductivity, dissolved oxygen and chlorophyll
a, exhibited significant seasonal variation at the 95%
confidence level (Table 5-16). Other parameters displayed
either significant interannual variations or no variation at
all. When year and season were combined, Secchi,
conductivity and chlorophyll a displayed significant
variability.
Lake Bonny
Lake Bonny is located in Polk County, Florida (Appendix
A). The lake lies in the Bartow Embayment division of the
Central Lake District (Canfield, 1981). Lake Bonny is an
urban lake with a surface area of 144 ha, a shoreline length
of 8.2 km and a mean depth of 2.0 m (Table 5-17).
Table 5-17. Morphometry
of Lake Bonnyd)
Surface Area
144 ha
Maximum Depth
2.5 m
Mean Depth
2.0 m
Development of Shoreline 50 %
Drainage Basin Area
729 ha
Shoreline Length
8.2 km
Macrophyte Cover
10 %
(1) Data from PCWR (1992).
There are no known outflows from Lake Bonny. Land-use
in the basin is a combination of commercial and moderate


CHAPTER 4
SEDIMENTS AND HISTORICAL ECOLOGY OF TWO CENTRAL
FLORIDA URBAN LAKES
Introduction
In the absence of historical data, past lake conditions
can be reconstructed based on interpretations of the
sedimentary record. Assuming that lake sediments accumulate
in an orderly fashion, paleolimnology may provide
information about how the lake ecosystem has changed over
time. Reconstructing the reason for past changes in the lake
may be used to predict the lake's response to future
management strategies (Smeltzer and Swain, 1985; Brenner et
al., 1993).
Lake sediments originate from numerous sources. The
main sources are both the biochemical substances produced by
organisms, or resulting from their degeneration, and
morphological pieces of specific organisms. The sediment
constitution is influenced primarily by the geomorphology of
the lake basin and the drainage basin (Wetzel, 1983).
Paleolimnology has as a principal objective a formulation of
general principles about the way lakes change with time
(Livingstone, 1981; Cohen and Nielson, 1986; Johnson et al.,
1991) .
53


94
Table 5-1. Morphometry
of Lake Parkerci)
Surface Area
924 ha
Maximum Depth
3.0 m
Mean Depth
2.2 m
Development of Shoreline 70 %
Drainage Basin Area
6.18 x 10J ha
Shoreline Length
19.8 km
Macrophyte Cover
5.0 %
(1) Data from PCWR (1992).
The drainage area north of the lake receives water from
Lake Gibson and a sinkhole basin on the west side. Overflow
from Lake Mirror also enters Lake Parker. The lake also
receives groundwater inputs. FMC Corporation discharges
effluent to the lake by a drainage ditch and the City of
Lakeland Power Plant withdraws and discharges water to the
lake. Lake Parker has an outflow to Saddle Creek through a
small canal and a water level control structure at the
outlet.
Land-use in the basin is dominated by a large
commercial and residential area. The entire northern section
of the Lake Parker watershed has been impacted by phosphate
mining. Dominant macrophytic plant species (defined as
greater than or egual to 5% macrophyte cover cf. PCWR, 1992)
include cattail, hydrilla, and water naiad. Macrophyte
chemical control is employed for water hyacinths, water
lettuce, and hydrilla.


CHAPTER 6
SUMMARY
Fish community impact on lake ecosystems was addressed
from the point of view of the effects of replacement of a
native filter feeder, gizzard shad, by the exotic blue
tilapia which has been present in central Florida eutrophic
lakes for the last thirty years. Evaluation of the impact of
gizzard shad and blue tilapia was through the
characterization and estimation of the fecal material
produced by each respective fish species. Analyses were
performed in order to determine how the different kinds of
feces could affect bioavailability of nutrients in the
systems.
A mesocosm experiment was performed to evaluate past
and present lake conditions. Paleolimnological data were
coupled with ambient water quality data in order to create
the necessary framework to permit further investigation of
system alterations due to modifications in the biological
community.
Several issues addressed in this study provided
information that led to a mosaic of interpretations.
Historical limnological data were analyzed for the six lakes
studied. Physicochemical and limnological data were
133


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
REPLACEMENT OF GIZZARD SHAD (Dorosoma cepedianum) BY BLUE
TILAPIA (Tilapia aurea) AS A POTENTIAL BIOMANIPULATION AGENT
IN FLORIDA EUTROPHIC LAKES
By
Carlos A. Fernandes
May 1995
Chairperson: Dr. Thomas L. Crisman
Major Department: Environmental Engineering Sciences
The effects of replacement of gizzard shad (Dorosoma
cepedianum) as a dominant filter feeding fish by blue tilapia
(Tilapia aurea) in six eutrophic Florida lakes were analyzed
using the lakes' historical limnological characteristics and
sedimentary records. The digestive physiology of gizzard shad
and blue tilapia (with emphasis on fecal material) was studied
in order to evaluate the relative impact of the feces produced
by different fish species.
Five of the six lakes studied are classified as
hypereutrophic systems. Lake Gibson is classified as
eutrophic. All of the study sites are urban lakes subject to
anthropogenic pressures in the form of industrial,
residential and commercial development in the watershed,
vi


115
Discharge from Padgett Elementary School passes through
a domestic wastewater treatment plant then goes into a
wetland that overflows north of Lake Gibson. No known
outflows were reported. The lake is located in a rapidly
urbanizing area. Citrus groves and pastures are giving way
to large-scale commercial and moderate-density residential
development. Dominant macrophytic vegetation (defined as
greater or equal to 5% macrophyte cover cf. PCWR, 1992)
includes water primrose, alligator weed, and pennywort.
Chemical controls are employed for water hyacinth, water
lettuce, and hydrilla.
During this study, Lake Gibson was classified as a
eutrophic lake according to the classification of Forsberg
and Ryding (1980). Canfield (1981) classified this lake as a
slightly acid, mesotrophic, soft-water lake. The Florida TSI
for this lake is 56 (the lowest for all lakes included in
this study), and it appears that water quality has remained
stable over the past decade. The water is slightly tannic
(color is 43 cpu), which lowers the Secchi disk values and
may bias the TSI value. Table 5-15 shows the water quality
data for Lake Gibson.
Lake Gibson displayed noticeable seasonality in its
phytoplankton gross primary productivity. As has been
reported by other investigators elsewhere in Florida
(Nordlie 1976; McDiffet 1980; Beaver and Crisman 1989),
maximum algal productivity occurred during summer. The


79
superficial layers was >70.0%, and in Lake Hollingsworth was
>51.0% (Schelske et al., 1992). Lake Bonny organic matter is
positively correlated with depth and shows no correlation
with any other parameter. Bulk density was positively
correlated with depth and negatively correlated with total
nitrogen and total phosphorus at the 5% level confidence
(SAS, 1989) (Table 4-8).
Organic matter in Lake Gibson was negatively correlated
to bulk density and showed a positive correlation to total
nitrogen. Bulk density was positively correlated with depth,
and depth was negatively correlated to total phosphorus at
the 5% level confidence (SAS, 1989) (Table 4-9).
Using the reconstruction of historical limnological
conditions, paleolimnology may help address issues of lake
management (Smeltzer and Swain, 1985) and restoration.
Sedimentary records may reflect past trophic status of a
lake and, thereby, assist in setting a goal for restoration.
Models may then be employed to identify primary nutrient
sources and to foresee whether mitigation actions can reduce
loading sufficiently to improve water quality or if other
management techniques such as biomanipulation may be needed.
When reduction of nutrient loading to a lake is
feasible, a cost/benefit evaluation should be completed to
ascertain which restoration techniques are financially
practical. Then, paleolimnological records can help in the
decision making process.


154
SMITH, A.D. 1971. Some aspects of trophic relations of
gizzard shad, Dorosoma cepedianum. Ph.D. thesis.
Virginia Polytechnic Institute and State University,
Blacksburg, 86 pp.
SMITH, W.L., 1963. Viable algal cells from the gut of the
gizzard shad Dorosoma cepedianum (LeSeur). Proc. Okla.
Acad. Sci. 43:148-149.
SOLOMON, D.J. and A.E. Brafield. 1972. The energetics of
feeding, metabolism and growth of perch
(Perca fluviatilis, L.) J. Anim. Ecol. 41:699-718.
SPATARU, P. and M. Zorn. 1978. Food and feeding habits of
Tilapia aurea (Steindachner) (Cichlidae) in Lake
Kinneret (Israel). Aquaculture 13: 67-79.
STEEMAN-NIELSEN, E. 1965. On the terminology concerning
productions in aquatic ecology, with a note about
excess production. Arch. Hydrob. 62(02):184-189.
STEWART, H.G. 1966. Ground-water resources of Polk County,
Florida. Fla. Geol. Survey Rept. Invest. No.44.
STOERMER, E.F., N.A. Andersen and C.L. Schelske. 1992.
Diatom succession in the recent sediments of Lake
Okeechobee, Florida, U.S.A. Diatom Research 7(2):367-
386.
THRELKELD, S.T. and E.M. Choinski. 1987. Rotifers,
cladocerans and planktivorous fish: what are the major
interactions? Hydrobiologia 147:239-243.
, and R.W. Drenner. 1987. An experimental mesocosms
study of residual and contemporary effects of an
omnivorous, filter-feeding, clupeid fish on plankton
community structure. Limnol. Oceanogr. 32(6):1331-1341.
TUNDISI, J. and T.M. Tundisi. 1976. Produco orgnica em
ecossistemas aquticos. Ciencia e Cultura
28(8) : 864-887.
U.S. Environmental Protection Agency. 1979. Handbook for
analytical quality control in water and wastewater
laboratories. EPA 600/4-79-019, Cincinnati, Ohio.
Van DONK, E., M.P. Grimm, R.D. Gulati and J.P.G. Klein
Breteler. 1990. Whole-lake food-web manipulation as a
means to study community interactions in a small
ecosystem. Hydrobiologia 200/201:275-289.


33
Caloric content, expressed as gross heat, for tilapia
feces had a mean value of 2604.13 cal.g"1, with a minimum
value of 2186.88 cal.g'1, and maximum of 2926.08 cal.g"1
(Table 3-1). Shad had a mean value of 4041.05 cal.g" 1, and
minimum and maximum values of 3842.40 cal.g'1 and 4212.81
cal.g'1, respectively (Table 3-1). Pierce ( 1977 ), from data
obtained from 27 samples of shad fecal material, calculated
its mean caloric content as 1953 cal.g"1.
Total nitrogen for tilapia feces displayed a mean value
of 34.6 pg.l"1, with a minimum of 33.42 pg.l"1, and a
maximum of 37.14 pg.l'1 (Table 3-1). For shad, the mean was
52.98 pg.l"1, with the minimum and maximum values being
48.38 pg.l"1, and 54.0 pg.l"1, respectively (Table 3-1).
Mean total phosphorus (TP) for tilapia was 6.63 pg.l'1, with
a minimum of 6.05 pg.l"1, and a maximum of 7.13 pg.l"1
(Table 3-1). Shad had a mean of 6.51 pg.l"1, with a minimum
of 5.98 jug. I"1, and a maximum of 7.04 jug.l"1 (Table 3-1).
Absolute values for total nitrogen, percent of
nitrogen, crude protein, total phosphorus and percent
phosphorus composition for gizzard shad and blue tilapia
fecal material are provided in Tables 3-2 and 3-3.
There were significant differences in the amount of
feces hand-stripped from both fishes. Tilapia had the
greatest amount of feces, and it was not possible to isolate
any particular factor responsible for such an occurrence. It
may be related to the length of the intestinal tract (which


12
subtropical systems maintain primary production all year
while production in temperate systems is greatly depressed
or absent during winter. Thus, if non-vegetative seasonal
estimates of production are included in determining annual
production, it is likely that subtropical systems would
realize greater yearly production.
From the few papers about the use and application of
biomanipulation techniques in the subtropical zone, Crisman
and Beaver (1990) noted that the increase in cladoceran
abundance in Florida lakes following elimination of fish
predation agrees with observations in temperate lakes
(Lynch, 1979; Carpenter et al., 1987; Van Donk et al.,
1989), but unlike the latter lakes, community species
composition was not altered nor was there a marked increase
in crustacean mean body size (Shapiro and Wright, 1984;
Benndorf et al., 1988). Large-bodied daphnids, the focus of
all temperate studies, are absent in Florida regardless of
trophic state or predation intensity (Crisman and Beaver,
1990). The results of Crisman and Beaver (1988) from
research conducted in Lake Apopka, Florida, show a
fundamental disagreement with the biomanipulation of Round
Lake, Minnesota (Shapiro and Wright, 1984), and suggest that
zooplankton size, structure and standing crop have only
minimal influence on phytoplankton biomass in Florida lakes.
Regarding fish composition, Crisman and Beaver (1990)
noted that unlike the eutrophic temperate lakes


101
greens such as Scenedesmus sp., Closterium sp., Pediastrum
sp., Cosmarium sp. and Ankistrodesmus; the dinoflagellate
Peridinium sp.; and diatoms of the genus Navicula sp.
A total of 41 zooplankton taxa were identified for Lake
Parker (Kolasa, 1993). Rotifers were the most abundant taxa
(27 out of 41), and comprised almost twice the number of
taxa as all other zooplankton combined. Rotifers comprised
94.7 percent of the zooplankton on an annual basis (Kolasa,
1993). Brachionus sp., Keratella sp., Collurella sp., and
Trichocerca sp., were the most numerically abundant
rotifers. The most abundant adult crustacean zooplankton was
Bosmina lonqirostris, a cladoceran (Kolasa, 1993).
Lake Parker's fish species richness of 21 species
(Canfield and Hoyer, 1992) is expected for a lake of this
size and is similar to that found in hypereutrophic Florida
lakes. Canfield and Hoyer (1992) reported data from nine
0.08 ha blocknets (six placed in littoral habitats and three
in open-water locations), and found that total fish biomass
averaged 71.1 kg.ha'1, with the average of harvestable
sportfish of 31.7 kg.ha"1. The greatest total fish biomass
was collected in littoral blocknets set in hydrilla and
tapegrass, whereas the lowest biomass was harvested in open-
water blocknets. The dominant fish species collected in
littoral nets were largemouth bass and bluegill, and the
dominant fish species collected in open-water nets were
gizzard shad and threadfin shad (Canfield and Hoyer, 1992).


42
composed of large zooplankton as juveniles to increasing
dependence on phytoplankton and/or smaller and non-evasive
zooplankton as adults (Kutkuhn, 1957; Cramer and Marzolf,
1970; Janssen, 1978; Durbin, 1979; Drenner et al., 1982b;
Lazzaro, 1987; and Yowell and Vinyard, 1993). Although the
feeding habits of filter-feeding fish have been extensively
studied, information about the quantitative feeding under
natural conditions is scarce.
The theoretical basis of fish bioenergetics has
received considerable attention in the fisheries literature.
Sophistication in describing basic equations and the
addition of computer analysis have enabled investigators to
partition more accurately the energy requirements of a
population. The goal of this study was to provide additional
information on the feces produced by gizzard shad and blue
tilapia including fecal composition, and nutrient and
caloric content. Experiments were conducted to compare the
potential impact of the fecal material of each respective
fish species on the primary production in aquatic systems.
Feces composed the greatest constituent of the energy
balance estimated from an experiment with bighead carp
employing energy budget equations to calculate food
consumption and food assimilation, i.e., C=P+R+F+U (1) and
A=C-F+U (2), where C=consumption, A=assimilation,
P=production, R=respiration, F=feces, and U=urea (Opuszynski
and Shireman, unpubl.).


Table 5-13. Correlation analyses of water quality data for Lake Hancock
using the General Linear Model procedure (GLM). Data were combined by
year, season, and year*season.
PARAMETER
YEAR
SEASON
YEAR*SEASON
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
Secchi
X
X
X
Temperature
X
X
X
pH
X
X
X
Conductivity
X
X
X
o

o
X
X
X
TSS
X
X
X
Turbidity
X
X
X
Chlorophyll a
X
X
X
Color
X
X
X
Alkalinity
X
X
X
NH^
X
X
X
TKN
X
X
X
TN
X
X
X
TP
X
X
X
ORT
X
X
X
BACTERIA
X
X
X
114


68
1.4% organic matter at the bottom of the core (approximately
1892) (Figure 4-4). Total nitrogen and total phosphorus
displayed high variability. Total nitrogen peaked about
1942, declined after that and peaked again in 1990. Total
phosphorus started with low values (1.0 mg.g'1) about 1892,
increased steadily to peak after 1966, and, from that point
on, kept increasing to a 1990 value of 5.0 mg.g-1 (Figure
4-5) .
Total P reconstruction at the base of the core suggests
that this lake, on the basis of phosphorus concentrations,
has been oligotrophic in the past. Nevertheless, values for
total P in excess of 3.0 mg.g-1 and greater were recorded
for all levels corresponding to the last 25 years (since
1966), reflecting changes in watershed use.
Bulk density increased below the 24 cm core, which
matched with a 210Pb determined sediment age of
approximately 1950. Detectable 137Cs coincided with the
determined 210Pb chronology (Table 4-3). The sedimentation
rate for Lake Gibson corresponded with the lake's history. A
significative peak of 0.12 g.cm_2.yr coincided with 1955,
the beginning of the development boom for the area (Figure
4-6) .
The organic matter content of dry bulk sediment in
Florida lacustrine surface mud averaged 39.7% from 97
surveyed lakes (Brenner and Binford, 1988). Lake Gibson had
a percent organic matter mean value of 38.2%, while Lake


92
concentrations (mg/L) were determined by a modified Kjeldahl
technique (Nelson and Sommers, 1975). Total suspended solids
(mg/L), organic suspended solids (mg/L) and inorganic
suspended solids (mg/L) were determined according to
standard methods (APHA, 1985). Water samples were analyzed
for color (Pt-Co units) using the platinum-cobalt method and
matched Nessler tubes (APHA, 1985). (The results of these
laboratory analyses were provided by the aforementioned
agencies).
Total chlorophyll a concentrations (pg/L) were
determined by filtering a measured portion of lake water
through a Gelman type A-E glass filter. Chlorophyll a was
determined by using the method of Yentsch and Menzel (1963)
and the equations of Parson and Strickland (1963). Primary
productivity (mgC.m'3.h) was measured using the light and
dark bottle method according to Wetzel and Likens (1991).
Phytoplankton samples for qualitative analysis were
collected quarterly at one station in each lake during the
1992/93 survey. Samples were collected at mid-Secchi depth
using a Van Dorn water bottle, and then dispensed into dark,
polypropylene bottles containing Lugol's solution (APHA,
1985). At the laboratory, samples were refrigerated in the
dark until examined. Zooplankton were sampled quarterly at
one station in each lake during the 1992/93 survey. Samples
were collected with an 8.0 L Kemmerer bottle and filtered
with a 28 pm net, and were preserved with 10 percent sugar-


91
Commission (FGFWFC); Southwest Florida Water Management
District (SWFWMD); Central Florida Regional Planning Council
(CF208).
Methods
I selected six lakes in Polk County, Florida: Lakes
Parker, Hollingsworth, Hancock, Gibson, Bonny and Hunter.
For this study all lakes were sampled quarterly during the
years of 1992 and 1993 following approved U.S. EPA
methodology. Additional data for the years of 1989 to 1991
were obtained from the above listed agencies.
Three water quality sampling stations were established
for Lakes Parker, Hollingsworth and Hancock, and two
stations were established for Lakes Gibson, Bonny and
Hunter. Water was collected from just bellow the surface
(0.5 m) in acid-cleaned Nalgene bottles. Samples were placed
on ice and returned to the laboratory for analysis. Water
temperature (C), dissolved oxygen (mg/L), pH and
conductance (pS/cm @ 25C) were measured by using a Hydrolab
Station. Secchi depth (m) was measured at each station where
water was collected.
At the laboratory, total alkalinity (mg/L as CaC03) was
determined by titration with 0.02 N H2S04 (APHA, 1985).
Total phosphorus concentrations (mg/L) were determined using
the methods of Murphy and Riley (1962) after persulfate
oxidation (Menzel and Corwin, 1965). Total nitrogen


Table 3-1. Physico and chemical characteristics of shad and
tilapia excretions expressed as ml and g.
PARAMETER
SHAD
TILAPIA
SHAD
TILAPIA
SHAD
TILAPIA
Mean
(ml) (g)
Mean
(ml) (g)
St. Deviat.
(ml) (g)
St. Deviat.
(ml) (g)
Range(*)
(ml) (g)
Range(*)
(ml) (g)
Bulk
Density
(g. cm-5)
1.0/ 0.1401
1.0/ 0.2576
0.72/ 0.010
0.08/ 0.022
(0.90-1.06)
(0.12-0.15)
(0.9-1.09)
(0.23-0.28)
% Dry
Matter
N/A /8.0366
N/A /12.988
N/A /0.5748
N/A /0.4856
(7.49-8.64)
(12.6-13.66)
% Org.
Matter
N/A / 67.84
N/A / 49.66
N/A /0.7954
N/A /1.3929
(67.06-68.90)
(48.6-51.63)
% Protein
N/A /32.427
N/A /21.655
N/A /1.3467
N/A /1.0568
(30.25-33.75)
(21.1-23.19)
Calories
(cal. g"1)
7516.35 ml
4041.05 g
4531.18 ml
2604.13 g
253.49
136.28
278.53
160.08
(7147-7852)
(3842-4222)
(3805-5091)
(2187-2926)
TN
N/A / 52.98
N/A / 34.65
N/A /0.9307
N/A /1.7055
(48.38-54.00)
(33.4-37.14)
TP
N/A / 6.51
N/A / 6.63
N/A /0.3636
N/A /0.4708
(5.98 7.04)
(6.05- 7.13)
(*) Numbers expressed in the 1st line are in ml.
Numbers expressed in the 2nd line are in g.
When only one line presented, it is expressed in g.
u>
to


19
Foote (1977) considered temperature, predation and
salinity to be the primary limiting factors in the
distribution of blue tilapia. The sensitivity of blue
tilapia to low temperature is apparently the most important
factor affecting the potential range of the species in North
America. Shafland (1978) tested lower lethal acclimation
temperatures for blue tilapia, spotted tilapia and the
Mozambique mouth brooder, and found that all tilapia tested
died between 6 and 12C. Of these three tilapia species,
blue tilapia was the most tolerant of cold water. According
to McBay (1961), T. aurea will not tolerate temperatures as
low as 9C. Based on these results and January isotherms,
Shafland concluded that the blue tilapia has the potential
of extending its range to include the entire state of
Florida (Williams et al. 1985).
Salinity has a slight but significant effect on the
cold tolerance of blue tilapia, suggesting that the species
may be expected to extend its range farthest north along the
coast and that populations in estuarine systems may be able
to withstand exceptionally cold weather better than inland
populations (Zale, 1984). However, the fish is capable of
finding thermal refugia during cold weather; the presence
and location of these must be considered when assessing
habitat suitability based on thermal criteria (Zale, 1984).
There is concern that the presence of blue tilapia
reduces largemouth bass populations through competition for


Percent organic matter
Time (year)
Figure 4-1. Lake Bonny percent organic matter sediment core
content versus time (year).


34
Table 3-2. Total nitrogen, total phosphorus and protein
content for gizzard shad feces.
pg N
% N
Crude Protein
pg P
% P
Group 1
A
48.38
4.84
30.25
5.98
.60
B
48.95
4.89
30.56
6.84
.68
Group 2
A
54.00
5.40
33.75
6.41
.64
B
53.65
5.36
33.50
6.19
.62
Group 3
A
51.43
5.14
32.12
7.04
.70
B
53.36
5.34
33.37
6.79
.68
Group 4
A
52.44
5.24
32.75
6.55
. 65
B
53.04
5.30
33.12
6.26
. 63
Table 3-3. Total nitrogen, total phosphorus and protein
content for blue tilapia feces.
pg N
%N
Crude Protein
pg P
%P
Group 1
A
33.68
3.37
21.06
7.13
.71
B
33.42
3.34
20.87
6.86
.69
Group 2
A
37.14
3.71
23.19
6.47
.65
B
34.38
3.44
21.50
6.05
.60
Group 3
A
22.61
2.26
14.12
5.50
.55
B
22.44
2.24
14.00
5.30
.53
Group 4
A
27.40
2.74
17.12
12.66
1.27
B
28.48
2.85
17.81
11.54
1.15


88
this slight noticeable improvement and/or stabilization
process in the water measured parameters in those lakes may
be due to the either an altered biological community and/or
the modifications resulting from the introduced species.


134
collected for more than twelve months in all the lakes.
Phytoplankton primary productivity and community respiration
were measured. Phytoplankton, zooplankton and fish
communities were identified. Sediment cores were taken from
two lakes, and sediment core information was gathered for
two of the others. Finally, the fecal composition of the two
major filter feeder fish in the systems was determined and
relative impact estimated.
All lakes except one were classified as hypereutrophic,
according to the classification of Forsberg and Ryding
(1980). Lake Gibson was classified as eutrophic. For
analytical purposes, I grouped the lakes into two
categories: lakes where the prevalent filter feeder fish
(defined as >50% of the filter-feeder fish population) was
gizzard shad and lakes where blue tilapia was the prevalent
filter feeder fish. Lakes Gibson, Bonny, and Hunter were
dominated by shad during this study, while Lakes Parker,
Hollingsworth and Hancock were tilapia dominated systems.
Fish feces experiment
There were significant differences in the amount of
feces hand-stripped from both species of fishes. Blue
tilapia releases its fecal material in pellet form enveloped
in mucilage, whereas gizzard shad releases its feces as a
flocculent material. Blue tilapia produced the greatest
amount of feces when compared with gizzard shad. The bulk


LITERATURE CITED
AMERICAN PUBLIC HEALTH ASSOCIATION. 1985. Standard methods
for the examination of water and wastewater. 16th
edition. APHA, Washington D.C.
ANDERSSON, G., H. Berggren, G. Cronberg and C. Gelin. 1978.
Effects of planktivorous and benthivorous fish on
organisms and water chemistry in eutrophic lakes.
Hydrobiologia 59 (1):9-15.
APPLEBY, P.G., P.J. Nolan, D.W. Gifford, M.J. Godfrey, F.
Oldfield, N.J. Anderson and R.W. Battarbee. 1986. 210Pb
dating by low background gamma counting. Hydrobiologia
143:21-27.
, and F. Oldfield. 1983. The assessment of 210Pb
data from sites with varying sediment accumulation
rates. Hydrobiologia 103:29-35.
BAGENAL, T. (ed.) 1978. Methods for assessment of fish
production in fresh waters. IBP Handbook no. 3.
Blackwell Scientific Publications, Oxford. 365 pp.
BAKER, C.D. and E.H. Schmitz. 1971. Food habits of adult
gizzard and threadfin shad in two Ozark reservoirs, p.
3-10. In G.E. Hall (ed) Reservoir Fisheries and
Limnology. Special Publ. #8, Amer. Fish. Soci. 511 p.
BAYS, J.S. and T.L. Crisman. 1983. Zooplankton and trophic
state relationships in Florida lakes. Can. J. Fish.
Aquat. Sci. 40:1813-1819.
BEAVER, J.R. 1990. Importance of organic color,
bacterioplankton and planktivore grazing in structuring
ciliated protozoan communities in subtropical lakes.
Ph.D. Dissertation. University of Florida. 227 pp.
r and T.L. Crisman. 1989. Replacement of an
endemic pump-filter feeding fish by an exotic omnivore:
Implications for water quality management in sub
tropical lakes.
142


APPENDIX G
Sediment core dating for Lake Hollingsworth
(cf. Schelske et al., 1992).
Depth
(cm)
Total 210Pb
(dpm.g-1)
214Bi
(dpm. g 1)
Excess 210Pb
(dpm.g-1)
137Cs
(dpm. g'1)
Age
(years)
Date
(AC)
Sed.rate
(g. cm-2. yr"1)
0-4
31.41
7.90
23.62
2.79
1.5
1990
0.051
4-8
32.98
8.25
24.84
3.06
3.8
1988
0.046
8-12
29.01
8.32
20.78
3.04
6.2
1986
0.051
12-16
28.63
7.46
21.26
3.28
9.4
1983
0.045
16-18
24.99
5.93
19.17
3.06
11.4
1981
0.046
18-20
18.24
6.54
11.77
2.84
13.3
1979
0.071
20-22
19.40
7.53
11.93
3.69
16.1
1976
0.065
22-24
20.12
9.35
10.82
2.89
18.8
1973
0.066
24-26
23.04
11.99
11.11
2.29
21.9
1970
0.059
26-28
26.71
13.99
12.80
3.17
25.4
1967
0.046
28-30
21.73
10.54
11.26
2.58
29.0
1963
0.047
30-32
18.20
9.14
9.11
1.57
33.2
1959
0.051
32-40
17.05
8.70
8.39
1.20
51.6
1940
0.040
40-42
15.90
8.26
7.67
0.82
56.7
1935
0.030
42-50
11.51
6.46
5.07
0.70
75.5
1916
0.031
a\
u>


43
I found that bulk density for tilapia fecal material
was twice the value for shad feces. Averaged shad feces
analysis had a composition of 37% more organic matter, 50%
more protein and 50% less dry matter than tilapia feces
which makes shad feces more organic, more protein-rich and
more easily dissolved in the environment. Caloric content of
shad feces was 60% higher than tilapia feces.
In general, assimilation efficiencies, which are
calculated using the balance of total calories of ingestion
and egestion material in fish, have been assumed to
approximate 80 percent of ingested energy (Winberg, 1956;
Ricker, 1968; Mann, 1969; Moriarty and Moriarty, 1973).
Although little is known about the energetics of omnivorous
fish such as shad and tilapia, an assimilation efficiency of
42 percent is reported for shad by Pierce (1977), and of 70-
80 percent for tilapia (Moriarty and Moriarty, 1973).
In this study, I found caloric content in shad fecal
material to have a mean of 4041.05 cal.g"1 (Table 3-5),
whereas tilapia had a mean of 2604.13 cal.g-1 (Table 3-6).
Considering both fishes relative to a natural diet, tilapia
is egesting material that has 60% less caloric content than
shad. This is, consequently, in accordance with what is
stated in the literature about the assimilation efficiency
for both, where tilapia has roughly two times greater
assimilation efficiency than shad, being expected to egest
material with a lower caloric content.


45
The diet of the gizzard shad is one of the most
controversial aspects in the ecology of this species. It is
generally agreed that very young shad (<35 mm) feed almost
exclusively on microcrustaceans (Kutkuhn, 1958; Cramer and
Marzolf, 1970), and fish larger than 35 mm have been
reported to consume a wide variety of food materials.
Tilapia are primarily vegetative feeders, feeding on algae
or aquatic plants, though occasionally zooplankters and
insect larvae picked up with bottom debris are eaten (Lowe,
1959). However, despite the similarity in feeding habits,
differing fecal composition and assimilation rates suggest
that tilapia and shad play significantly different roles in
the nutrient dynamics of aquatic systems.
Chlorophyll a values for both shad and tilapia fecal
material were very different between the beginning and end
of the bioassays. In both bioassays for blue-green and green
algal groups, tilapia initial values were 10 to 15 times
higher than for shad (tilapia mean value of 915 mg.nf3 and
shad mean value of 67 mg.nf3); whereas at the end of
bioassays, the situation was completely inverted, with shad
displaying a tenfold increase in chlorophyll a (shad mean
value of 685 mg.nf3) regardless of the algal group tested.
(Figures 3-3 to 3-6) Tilapia chlorophyll a values at the end
of bioassay varied with the algal group tested. In the
Microcystis experiment, all values were negative indicating
suppressed chlorophyll a values (mean of -691 mg.nf3)


30
Calorimeter). Pre-weighed (pre-burn weight, approximately
one gram of dried material) fecal samples were placed in a
microbomb. The bomb was placed in 2 liters of water at 30-
31C. The sample was ignited and burned in the presence of
oxygen. After ignition, increases in water temperature were
registered to calculate energy within the sample, expressed
as gross heat (cal.g'1).
An algal bioassay was done using thirty-six 300 ml
flasks containing FWH+Si (a standard medium for maintenance
culture of freshwater blue-green algae). Eighteen flasks
were prepared without KN03 (medium -N); and eighteen flasks
prepared without K2HP04 (medium -P). These two groups of
eighteen flasks were each subdivided into three different
subgroups of six; one subgroup containing 3 ml of shad
feces; another containing 3 ml of tilapia feces; and the
third subgroup without feces, to function as a control. Each
group of six flasks was subdivided further into two
subgroups: three flasks inoculated with a strain of blue-
green algae (Microcystis aeruginosa), and three flasks
inoculated with a strain of green algae (Selenastrum
capricornutunU These sets of flasks were incubated for
seven days in a room under constant light and temperature
conditions.
Samples from all experimental flasks were analyzed for
chlorophyll a using extractive and fluorimetric methods
(APHA, 1989). Cell counting of algae from samples collected


81
Historical water-column total P data collected for the
last four years at Lake Bonny showed little variation, with
the exception of Spring 1991, when the values were double
the mean (mean = 0.112 mg.I'1). An increase in chlorophyll a
and total bacteria were also reported for the same period.
This information is in disagreement with Maceina and Soballe
(1990), who suggested that resuspension during wind events
is probably responsible for the highly variable total P and
chlorophyll a values measured in many shallow Florida lakes.
Lake Gibson historical records for total P showed no
significant variation for the last four years, with a mean
of 0.210 mg.l-1. Chlorophyll a, TSI and total bacteria also
displayed no notable variations. Anthropogenic alterations
in the watershed, especially increased urbanization, may be
responsible for the presence of only one genus of blue-green
algae in the lake, as well as for the occasional algal
bloom.
The Lake Parker historical record for net phosphorus
accumulation rates greatly exceed dangerous P loadings to a
basin of <5 m mean depth (13 pg. cm'2.yr'1) (Brenner et al.,
1993). Total phosphorus measured in the water column during
this study had a mean of 0.22 rag.L"1 (0.07 to 0.602
mg. L'1) .
Lake Hollingsworth historical total phosphorus
concentration show an overall increase upwards in the
sediment core (Schelske et al.,
1992). Total P values


139
enrichment (Schelske et al., 1992). Diatom reconstructions
of historical water column total phosphorus concentrations
indicate that Lake Hollingsworth has been eutrophic to
hypereutrophic for the past 100 years (Schelske et al.,
1992). During the last ten years, however, total phosphorus
inferences (cf. Schelske et al., 1992) suggest that the rate
of eutrophication remained relatively constant.
Since both of these lakes were classified as having a
blue tilapia dominant fish population in this study, it is
possible that some of the trophic reversal can be credited
to a filter feeder fish replacement. Additional support for
this hypothesis can be drawn indirectly from the absence of
an improvement trend in the sediment nutrient concentration
in Lakes Bonny and Gibson, both gizzard shad dominant lakes.
However, additional study is required in order to confirm
this hypothesis.
Lakes assessment
The lakes were analyzed for the years 1989 to 1993 and
separated by season within each of these years. The seasons
were defined as winter (December, January and February),
spring (March, April and May), summer (June, July and
August), and fall (September, October and November). ANOVAs
were performed using General Linear Model (GLM) (SAS, 1989)
in order to identify significant seasonal limnological
factors (p < 0.05).


mg/g
Time (year)
Figure 4-2. Lake Bonny TN (mg.g-1) and TP (mg.g"1) sediment core
content versus time (year).


Prod (mgC/m3/h)
Zoop (#/m3 x 10*2)
Re8p (mgC/m3/h)
lilil Fish (kg/ha)
] Phyto (NU/ml x 10*3)
Figure 5-2. Fish, phytoplankton and zooplankton population for all study lakes.
Values for primary productivity and community respiration are shown.
132


137
suppressed (zero growth) in 73% of all samples containing
blue tilapia feces, all the samples containing gizzard shad
feces displayed a ten-fold increase.
Paleolimnolooical analysis
Analyses of sediment cores from Lakes Bonny and Gibson,
both defined in this study as having a prevalent filter-
feeder population of gizzard shad, were done to assess the
distribution of organic deposits throughout the lake basin
and for sediment age determination. Sediment
characterization was based on one core collected from the
deepest part of each lake.
Lake Bonny showed high organic matter content in the
superficial layers (58% of organic matter). TP and TN were
constant throughout the whole core. Bulk density and dry
matter began to decline after 1950, which coincides with
urban development of the watershed. Total phosphorus
reconstructions near the base of the core suggest that the
lake has always been mesotrophic to eutrophic or
hypereutrophic. However, the lake has experienced periods of
much higher nutrient enrichment, and values for total
phosphorus in excess of 5.0 mg.g'1 were common for layers
that post-dated the 137Cs peak after 1960. Detectable 137Cs,
injected into the atmosphere as a consequence of nuclear
weapons testing in the early 1950s, matched well with the
determined 210Pb chronology!


143
, T.L. Crisman and J.S. Bays. 1981. Thermal
regimes of Florida lakes. A comparison with biotic and
climatic transitions. Hydrobiologia. 83:267-273.
BENNDORF, J. 1988. Objectives and unsolved problems in
ecotechnology and biomanipulation:A preface.
Limnologica (Berlin) 19:5-8.
, H. Kneschke, K. Kossath and E. Penz. 1984.
Manipulation of the pelagic food web by stocking with
predacious fishes. Int. Revue ges. Hydrobiologia
69:407-428.
, H. Schultz, A. Benndorf, R. Unger, E. Penz, H.
Kneschke, K. Kossatz, R. Dunke, U. Hornig, R. Kruspe
and S. Reichel, 1988. Food-web manipulation by
enhancement of piscivorous fish stocks:Long-term
effects in the hypertrophic Bautzen Reservoir.
Limnologica 19:97-110.
BERRY, F.H. 1958. Age and growth of the gizzard shad (D.
lecepedi) (LeSueur) in Lake Newnan, Florida. Proc.
Eleventh Amer. Conf., Southeastern Assoc. Game and Fish
Commissioners. 318-331.
BJORK, S. 1985. Lake restoration technigues. In:Proceedings
International Congress Lakes Pollution and Recovery",
Rome:281-292.
BODOLA, A. 1966. Life history of the gizzard shad, Dorosoma
cepedianum (LeSueur), in western Lake Erie. Fishery
Bull. 65:391-425.
BOYLE, J.R. and C.W. Hendry, Jr. 1985. The mineral industry
of Florida, 1983. In Minerals Yearbook, 1983. U.S.
Bureau of Mines, Washington, v.2, pp. 137-148.
BREMNER, J.M. and C.S. Mulvaney. 1982. Nitrogen-total. Pages
599-609 in A.L. Page, R.H. Miller, and D.R. Keeney
(eds.), Methods of Soil Analysis, Part II, Chemical and
Microbiological Properties. Amer. Soc. Agro. & Soil
Sci. Soc. Amer., Madison, Wise. 599-609.
BRENNER, M. and M.W. Binford, 1988. Relationships between
concentrations of sedimentary variables and trophic
state in Florida lakes. Canadian J. Fish. Aquat. Sci.
45:294-300.
/ M.W. Binford and E.S. Deevey. 1990. Lakes. Pages
364-391 in R.L. Myers and J.J. Ewel (eds.), Ecosystems
of Florida. Univ. Central Florida Press, Orlando.


112
maximum of 625.0 mgC.m'3.h in July (Table 5-3). Values for
net photosynthesis are shown in table 5-4, community
respiration in table 5-5, and average annual productivity in
table 5-6.
During this study, blue-green algae including Spirulina
sp. and Anabaena sp. were the most abundant algae, followed
by Microcystis sp., and Merismopedia sp. A small number of
green algae such as Closterium sp. were also recorded.
The zooplankton community was dominated by rotifers,
including Brachionus sp. and Keratella sp., and Diaptomus
was the only copepod genus reported (Kolasa, 1993).
Fourteen species of fish were reported for Lake
Hancock. Field collections from fifteen-minute
electrofishing events in July 1991 and February 1992,
indicated that blue tilapia was responsible for 71.4 and
52.0 % of total fish weight, respectively, while values for
gizzard shad were 4.4 and 32.0 % by weight for the two
sampling periods. The dominant fish species collected were
blue tilapia, followed by black crappie in the 1991 sample
and gizzard shad in 1992.
ANOVAs were performed using the General Linear Model
(GLM) (SAS, 1989) in order to identify significant seasonal
limnological factors (p < 0.05). The following parameters
were analyzed: Secchi, temperature, pH, conductivity,
dissolved oxygen, total suspended solids, turbidity,
chlorophyll a, color, alkalinity, NH3, TKN, total nitrogen,


9
restoration process (e.g. Shapiro and Wright, 1984;
Edmondson and Abella, 1988).
For studying the effect of food-web manipulations on
lake restoration and to examine the fundamental mechanisms
underlying ecosystem regulation, application of whole-lake
manipulation-experiments can have major advantages (van Donk
et al., 1990). These experiments simulate or actually
encompass the conditions that would be expected to occur
naturally in lakes (Carpenter and Kitchell, 1988). Problems
of enclosure-size and omitted members of communities are not
relevant to whole-lake manipulations (Frost et al., 1988).
It is, however, difficult to perform these manipulations on
a large scale and to interpret their results (Hulbert,
1984). Interpretations may be greatly eased by results of
small-scale manipulations on similar systems (Frost et al.,
1988) .
Nevertheless, to obtain a good overview of the
community's processes, investigations must be conducted
simultaneously as small-scale in situ or laboratory
experiments that can be replicated under controlled
conditions. In this approach, whole-lake manipulations can
be considered as generating as well as testing hypotheses
(O'Neill et al., 1986) .
Examination of trophic-level interactions has long been
an integral part of limnology (Hrbacek et al., 1961;
Nauwerck, 1963; Brooks and Dodson, 1965). However, Shapiro


136
remained constant and a decrease was noticed after the
fourth day of experimentation. When the treatment group
feces plus Selenastrum capricornutum was evaluated, high
initial fluorescence values decreased after the first day of
the bioassay, reaching minimum values on the fourth day of
the experiment. (Figures 3-3 and 3-4). Control flasks in the
medium deprived of phosphorus and in the medium deprived of
nitrogen exhibited algal fluorescence readings of zero or
near zero.
Chlorophyll a readings for blue tilapia feces were very
high in the beginning of the experiment (ten-fold greater
than the values measured for gizzard shad feces). The
results showed that gizzard shad fecal material started the
bioassay at low chlorophyll a value in the flasks containing
blue green algae both in the nitrogen and in the phosphorus
deprived medium. At the end of the experiment, chlorophyll a
values increased ten-fold in the nitrogen deprived medium
and 13.5 times in the phosphorus deprived medium. Blue
tilapia feces displayed, on the other hand, under the same
conditions, increased chlorophyll a only at the end of the
experiment and only 5 to 6 times the initial value, an
increase which is only half of that produced by gizzard shad
feces.
Phytoplankton biomass in gizzard shad and blue tilapia
feces was indicative of the different impact these fishes
can have in aquatic systems. While phytoplankton biomass was


4
Digestion in animals has been an almost totally-
neglected subject. Fish studies have been limited to gut
content and selectivity (Halver, 1989). Very little is known
of the digestive capabilities of fishes.
Fish, like all other animals, reguire energy to sustain
life, and they are among the most efficient animals in
converting food to body tissue (Halver, 1989). However, a
rough generalization can be made that about one-third of the
energy in the food offered to fish will be lost as
combustible waste (Halver, 1989). This will consist of
uneaten food, feces, urine, and gill excretions. From the
lost combustible wastes, feces (25% of the total) constitute
the majority of this part of the energy balance.
Fecal material produced from different fishes and
released into the environment can produce a differential
impact and conseguently a differential response. Blue
tilapia feces, being encased in mucilage, are more coherent
and will be broken down slowly, whereas shad feces are
released practically as aqueous substance permitting an
almost instantaneous availability of nutrients for the
phytoplankton community.
Because little is known of their feeding ecology,
especially guantitative feeding under natural conditions,
attempts to predict and efficiently manage these filter
feeding fish as a tool to alleviate eutrophication processes
still is in its experimental stage. Some researchers (i.e.


mg/nT3
1200
1000
800
600
400
200
0
Figure 3-3. Bioassay initial (I) and final (F) values of chi. a (mg.nf3)
with M.aeruginosa in medium -N.
I
Control Tilapia Shad
Time


135
density of blue tilapia feces was two times the bulk density
for gizzard shad feces. On average, shad feces was composed
of 37% more organic matter, 50% more protein, and had a 60%
higher caloric content than did blue tilapia feces.
Furthermore, blue tilapia feces consisted of 50% more dry
matter than gizzard shad feces.
I carried out two bioassays to evaluate fecal nutrient
composition and algal content. I measured the change in
fluorescence for each treatment group. The three treatment
groups were: [1] feces plus Microcystis aeruginosa; [2]
feces plus Selenastrum capricornutum; and [3] control (no
feces). Each treatment group was inoculated with algae in
both a medium deprived of nitrogen and a medium deprived of
phosphorus.
Gizzard shad fecal samples always yielded greater
fluorescence values regardless of the algal strain or
chemical composition of the growth medium (-N or -P) used.
In treatment group feces plus Selenastrum capricornutum, a
significant increase in fluorescence occurred in the first
day of experiment; whereas in treatment group feces plus
Microcystis aeruginosa, a significant increase occurred on
the fourth day of the bioassay (Figures 3-3 and 3-4).
Results were different in the flasks inoculated with
blue tilapia fecal material. Despite high initial
fluorescence values, in the presence of treatment group
feces plus Microcystis aeruginosa, fluorescence values


23
biological consequences of eutrophication without the need
for often costly controls of nutrient loading (Crisman and
Beaver, 1990). Biomanipulation, as stated by Shapiro and
Wright (1984), is based on the prediction that increased
piscivore abundance will result in decreased planktivore
abundance, increased zooplankton abundance, and increased
zooplankton grazing pressure leading to a reduction in
phytoplankton abundance and improved water clarity.
This "biological" approach could make it possible to
increase herbivore density in aquatic communities, thereby
lowering algal biomass to levels less than expected for a
given nutrient concentration. In addition, limiting the
abundance or even the occurrence of certain fish could
curtail the flux of nutrients from epilimnetic or littoral
sediment to the pelagic zone. Improvement of the water
quality of lakes with algal blooms by implementing a
combination of these biological techniques could reduce or
eliminate the need for the use of common chemical (i.e.
cooper sulfate) and mechanical methods to deal with
eutrophication (Cooke et al., 1993).
Caird (1945), with his experiment of adding largemouth
bass to a 15-ha Connecticut lake, was one of the first to
publish observations about the effect of increased biomass
of piscivorous fish on the phytoplankton community. More
recently, investigators such as Hrbcek et al. (1961),
Brooks and Dodson (1965), and Hulbert et al. (1972) have


90
phosphate production for over 90 years; the bulk of this
production is from Polk County (Boyle and Hendry, 1985). The
trophic status for lakes of this region is mostly
mesotrophic or eutrophic, and waters are moderately
organically stained (Canfield, 1981).
Water quality on the Polk Upland is highly variable.
Based on data from Florida Game and Fresh Water Fish
Commission, Canfield (1981), Polk County Water Resources
Division and this study, mean pH ranged from 5.4 to 9.4 and
total alkalinity concentrations averaged between 1 and 76
mg.l'1 as CaC03. Total hardness concentration averaged
between 14 and 195 mg.l'1 as CaC03, and mean specific
conductance ranged from 82 to 281 pmhos.cm'1. Calcium was
the dominant cation and bicarbonate the dominant anion in
Lake Parker, whereas in most of the other lakes in the area,
sodium and chloride were the dominant anions. Total nitrogen
concentration was between 306 and 1566 mg.m'3, mean total
phosphorus concentration from 2.4 to 144 mg.m'3, and Secchi
disk depth 0.2 to 2.7 m. The wide range of water chemistry
may be directly related to different regional geology
(Canfield, 1981; Crisman, 1993).
The following agencies have conducted studies or
developed information on the lakes in Polk County: United
States Geological Survey (USGS); Environmental Protection
Agency (EPA); Polk County Water Resources Division (PCWR);
City of Lakeland (CL); Florida Game and Freshwater Fish


20
nesting sites or predation on bass eggs (Noble et al.,
1975). Zale (1984) also found a high degree of trophic
overlap between young tilapia and larval shad in Lake
George, and perhaps the enhanced abundances of individuals
documented following introductions of blue tilapia have
resulted from exploitative competition for zooplankton
during early life stages. He suggested this to be a more
realistic explanation than the "competition for algae and
detritus among adults" theory usually invoked.
These results are confirmed by Beaver and Crisman
(1990) who reported phytoplankton and zooplankton community
alterations in subtropical Florida lakes when blue tilapia
replaces gizzard shad. Chlorophytes and rotifers were
proportionally more abundant in tilapia-dominated lakes
(Crisman and Beaver, 1988), while cladocerans composed a
greater percentage of the zooplankton population in shad-
dominated lakes. Limited zooplankton distribution data
suggest that total zooplankton biomass is comparably
depressed in systems dominated by tilapia. Empirical
evidence suggests that little measurable improvement in
water guality through differential grazing will be realized
if blue tilapia displace shad as the dominant rough fish in
subtropical Florida lakes (Beaver and Crisman, 1990).
A summary from previous investigations in Florida lakes
is compiled in Table 2-1. The data reported are condensed
from many different sources which were already cited.


13
characterized by size-selective planktivorous fishes, in
subtropical systems this composition shifts to pump-filter
feeding fishes, reflecting faunal dominance by gizzard shad
(Dorpsoma cepedianum). Crisman and Beaver (1990) noted that
in both temperate and subtropical systems, removal of
planktivorous fish results in higher macrozooplankton
populations.
Unlike temperate systems, however, algal biomass was
not reduced in the presence of enhanced macrozooplankton
abundance, but actually increased. Crisman and Beaver (1990)
suggested that small-bodied macrozooplankton, even if freed
from fish predation, are of questionable value as
biomanipulation tools in eutrophic subtropical lakes. These
authors also stated that if biomanipulation were to be
successful in the subtropics, emphasis should be shifted
from zooplankton to the role played by planktivorous fish
(pump-filter feeding).
Shad Ecological Energetics
The gizzard shad (Dorosoma cepedianum) occurs commonly
in many lakes and streams in North America (Scott and
Crossman, 1973), where it is the principal phytophagous fish
(Crisman and Kennedy, 1982). This species is predominately
found in eutrophic lakes, being reportedly a key link in the
food chain between primary producers and the top carnivores


26
1986), and detritus (Hendricks and Noble, 1980; Mallin,
1986). Gophen et al. (1983) reported that tilapia greater
than 7.6 cm act as filter feeders, and tilapia smaller than
7.6 cm, in addition to filter feeding, also feed as size
selective predators on individual zooplankton species
(particulate feeder).
The grazing activities of blue tilapia depressed some
large sized algal groups (Uroqlenopsis and Ceratium), while
the smallest phytoplankton taxa were enhanced (Rhodomonas.
Chrvsochromulina, Chlamydomonas, and Cvclotella) (Drenner et
al., 1984a). Tilapia grazing did suppress the zooplankter
Keratella, while copepodid and adult Diaptomus populations
were enhanced (Drenner et al., 1984a). Tilapia also
selectively fed on Bosmina and Ceriodaphnia (Gophen et al.,
1983), taxa reportedly having poor evasive capabilities.
Little information is available on nutrient release and
degree of algal digestion by blue tilapia. Popma (1982)
noted that some algal cells may remain viable following
passage through the digestive tract of tilapia. Dickman and
Nanne (1987), found that high concentrations of adult
tilapia (2.5 adults itf2) raised in some Central America
ponds suppressed zooplankton populations and increased the
blue-green alga Microcystis aeruginosa.


102
ANOVAs were performed using the General Linear Model
(GLM) (SAS, 1989) in order to identify significant seasonal
limnological factors (p < 0.05). The following parameters
were analyzed: Secchi, temperature, pH, conductivity,
dissolved oxygen, total suspended solids, turbidity,
chlorophyll a, color, alkalinity, NH3, TKN, total nitrogen,
total phosphorus, orthophosphate, and total bacteria.
The data were collected between 1989 and 1993 and
grouped in four seasons: winter (December to February);
spring (March to May); summer (June to August); and fall
(September to November). The analyses of the data were
according to the year of collection, season and the
combination of year and season (Table 5-7).
Temperature, conductivity, dissolved oxygen, turbidity,
and total bacteria are the parameters that exhibited
significant seasonal variation at the 95% confidence level.
The other parameters' variability displayed either
interannual variations or no variation at all. When year and
seasonality were combined, Secchi, conductivity, turbidity,
NH3 and total bacteria are the parameters that displayed
significant variability.
Lake Hollingsworth
Lake Hollingsworth is located in Polk County, Florida
(Appendix A). The basin lies in the Bartow Embayment
division of the Central Lakes District (Canfield, 1981),


Table 5-6. Average annual productivity values (ragC.m'3.h) (+S.Dev.)
at surface (1); 50 cm (2); and 100 cm (3) depths
for all study lakes.
Parameter
Parker
Hollings.
Hancock
Gibson
Bonny
Hunter
(1)
66.41
(8.21)
369.88
(141.2)
409.42
(184.4)
214.18
(229.1)
448.04
(73.2)
229.56
(89.0)
Gross prim, product. (2)
154.87
(42.51)
197.25
(187.4)
101.6
(118.0)
167.21
(123.7)
177.5
(44.34)
306.71
(175.6)
(3)
70.09
(6.39)
102.92
(76.06)
N/A
94.74
(86.5)
85.08
(25.67)
201.02
(75.92)
(1)
19.92
(2.46)
110.96
(42.37)
122.82
(55.33)
75.71
(80.78)
156.81
(25.64)
80.35
(31.15)
Net primary product. (2)
46.46
(12.75)
59.17
(56.23)
30.48
(35.40)
58.52
(43.29)
62.12
(15.52)
107.35
(61.46)
(3)
21.03
(1.92)
30.87
(22.82)
N/A
33.16
(30.26)
29.78
(8.99)
70.35
(26.58)
(1)
46.48
(5.74)
258.92
(98.86)
286.6
(129.10)
139.22
(148.91)
291.23
(47.62)
149.21
(57.85)
Respiration (2)
108.41
(29.76)
138.08
(131.20)
71.12
(82.61)
108.69
(80.40)
115.38
(28.82)
199.36
(114.14)
(3)
49.06
(4.47)
72.05
(53.24)
N/A
61.59
(56.21)
55.30
(16.68)
130.66
(49.35
100


those of the lakes studied, biomanipulation techniques can
mitigate cultural eutrophication to improve water clarity,
if gizzard shad are replaced by blue tilapia.
IX


150
KOLASA, K.V. 1993. Lake Parker diagnostic feasibility study.
Final report. Southwest Florida Water Management
District, Brooksville, FL.
KUSHLAN, J.A. 1974. Effects of a natural fish kill on the
water quality, plankton, and fish population of a pond
in the Big Cypress Swamp, Fl. Trans. Amer. Fish. Soc.
2:235-243.
KUTKUHN, J.H. 1958. Utilization of plankton by juvenile
gizzard shad in a shallow prairie lake. Trans. Amer.
Fish. Soc. 87:80-103.
LAMARRA, V.A.,Jr. 1975. Digestive activities of carp as a
major contributor to the nutrient loading of lakes.
Verh. Int. Ver. Limnol. 19:2461-2468.
LANGELAND, A., J.I. Kolsvik, Y. Olsen and H. Reinertsen.
1987. Limnocorral experiments in a eutrophic lake -
Effects of fish on the planktonic and chemical
conditions. Pol. Arch. Hydrobiol. 34(l):51-65.
LAZZARO, X. 1987. A review of planktivorous fishes: their
evolution, feeding behaviors, selectivities, and
impacts. Hydrobiologia 146:97-167.
LEVENTER, H. 1981. Biological control of reservoirs by fish.
Bamidgeh 33(l):3-23.
LEWIS, W.M. 1953. Analysis of the gizzard shad population of
Crab Orchard Lake, Illinois. Ill. Acad. Sci. Trans.
46:231-234.
LIVINGSTONE, D.A. 1981. Paleolimnology. The ecology and
utilization of African inland waters. Pages 176-182 in
J.J. Symoens, M. Burgis, J.J. Gaudet (eds.), United
Nations Environmental Programme. Nairobi, Kenya.
LUND, J.W.G. & J.F. Tailing. 1957. Botanical limnological
methods with special reference to algae. Botanical
Review 23(8-9):489-583.
LYNCH, M., 1979. Predation, competition and zooplankton
community structure: an experimental study. Limnol.
Oceanogr. 24: 253-272.
McBAY, L.G., 1961. The biology of Tilapia nilotica Linneaus.
Proceedings of the Fifteenth Annual Conference,
Southeastern Association of Game and Fish
Commissioners, pp. 208-218.


59
Methods
An 89 cm mud-water interface core was collected from
the middle of Lake Bonny using a piston corer with a 12 cm
diameter, 1.83 m long cellulose acetate butyrate (CAB) core
barrel. Externally, the core was characterized by a light
color sediment for the first 10 cm and a typical black mud
color below in the rest of the core.
A 60 cm core was taken from the central portion of Lake
Gibson using the same piston corer described above. The
presence of sand mixed with mud was noticed beginning at the
50 cm depth. Sand content increased deeper in the core. Due
to the soupiness of the sediments, in both cores a spoon was
used to collect the samples down to the 10-cm layer. From
that point on, collection was possible using a spatula. The
sediment from Lake Gibson was more consolidated on the
bottom than the sediment from Lake Bonny.
Concentrations of 210Pb and 137Cs were measured by
direct y-assay using a P-type, intrinsic-germanium detector
(Princeton Gamma Tech). The counting system used for
spectral analysis is located in the University of Florida's
Department of Environmental Engineering Sciences Low
Background Counting Room. The electronics for the system
include a preamplifier (RG11B/C, Princeton Gamma Tech,Inc.),
amplifier (TC 242, Tennelec), bias supply (5 kV, TC 950,
Tennelec), power supply (TC 909, Tennelec), and transformer
(Sola).


87
Sediment cores from Lakes Bonny and Gibson collected
during this study showed no noticeable changes relative to
the improvement of any water quality parameter. However,
Lakes Parker and Hollingsworth had what perhaps could be
called a modification. According to Schelske et al. (1992),
total phosphorus reconstructions in Lake Parker's topmost 15
cm (after 1960s) of sediment suggested a progressive decline
in the water-column that may signal some reversal of
cultural eutrophication. The Lake Hollingsworth core
revealed that, for at least in the last ten years, total
phosphorus inferences have remained relatively constant
(Schelske et al., 1992).
Taking in consideration that blue tilapia came into the
systems after 1961, it is reasonable to assume, first of
all, a lag time for the establishment of the fish population
in the new environment, and, secondly, time for the
manifestation of any modification that those fishes would
bring to the systems. Since more than thirty years have
passed since fish introduction, it is possible to
hypothesize that this could be responsible for changes
recorded in Lakes Parker and Hollingsworth water quality,
even though this might still provide a weak indication of
water quality improvement. However, considering that all
lakes included in this study are subject basically to the
same kind of impacts, as well as the lake recovery program
carried out by the City of Lakeland, the explanation for


97
Table 5-3.
Gross
primary productivity for all study lakes
expressed as (mgC.m .h).
I LAKE
DEPTH
08/92
12/92
04/93
07/93
MEAN
S.DEV
0 cm
64.06
78.12
64.53
58.93
66.41
8.21
y Parker
50 cm
140.62
203.12
103.86
171.87
154.87
42.51
100 cm
70.31
62.51
69.43
78.12
70.09
6.39
0 cm
178.90
437.5
356.78
506.34
369.88
141.2
1 Holl a
50 cm
46.87
31.25
304.65
406.25
197.25
187.4
100 cm
46.87
30.98
146.35
187.5
102.92
76.06
0 cm
356.45
468.75
187.5
625.0
409.42
184.4
A Hanc b
50 cm
187.67
0
0
218.75
101.60
118.0
100 cm
N/A
N/A
N/A
N/A
N/A
N/A
0 cm
346.89
15.62
25.46
468.75
214.18
229.1
Gibson
50 cm
278.26
93.75
31.25
265.6
167.21
123.7
100 cm
156.78
0
43.75
178.46
94.74
86.5
0 cm
478.34
453.12
345.12
515.6
448.04
73.2
| Bonny
50 cm
156.89
187.5
131.25
234.37
177.50
44.34
100 cm
98.14
93.75
46.87
101.56
85.08
25.67
0 cm
303.85
109.37
289.56
215.47
229.56
89.0
| Hunter
50 cm
423.74
115.62
484.37
203.12
306.71
175.6
100 cm
205.63
98.45
281.25
218.75
201.02
75.92
(a) Hollingsworth.
(b) Hancock.


g/crrT2/yr
Time (year)
Figure 4-3. Lake Bonny sedimentation rate (g.cirf2.yr)
versus time (year).
a\
CTi


CHAPTER 2
LITERATURE REVIEW
Introduction
This chapter reviews the literature on biomanipulation
and shad and tilapia ecological energetics by examining
planktivorous fish from the community standpoint with
special emphasis on fecal material composition and fate in
the system and the significance of different fish species as
trophic linkages in freshwater food webs.
Biomanipulation
The two main strategies to control eutrophication of
fresh waters are 1) reduction of external and internal
loading of nutrients (Bjork, 1985) and 2) control of
internal ecological processes. With respect to toxic
substances, organic wastes and acid precipitation, the first
strategy alone will provide acceptable solutions to the
problems on a long-term scale (Benndorf, 1988). However, a
combination of strategies 1 and 2 could lead to an
improvement in water guality and to a lower cost/benefit
ratio in the management of the water resource.
Until recently, eutrophication problems were tackled
primarily by reducing external nutrient loading (Hosper and
7


95
Using the criteria of Forsberg and Ryding (1980), Lake
Parker was hypereutrophic during this study. Canfield (1981)
also classified Lake Parker as hypereutrophic. It is
suggested that the trophic status of this lake has remained
relatively stable for at least the last decade. Table 5-2
shows water quality data for Lake Parker.
Phytoplankton primary productivity was measured in
October, December, April and July at surface, 50cm and 100
cm depths each time, using the light and dark bottle method.
The results are expressed as gross photosynthesis
(mgC.m'3.h) cf. Wetzel and Likens (1991) (Table 5-3). Values
for net photosynthesis are presented in table 5-4, community
respiration in table 5-5, and table 5-6 shows the average
annual productivity values for all study lakes. Lake Parker
does not display great seasonality in primary productivity.
The highest values were always found at the 50 cm depth,
with a mean for that depth of 154.87 mgC.m"3.h 42.51
throughout the year. A minimum of 103.86 mgC.'3.h for April
and a maximum of 203.12 mgC.m'3.h for December were recorded
at the 50 cm depth. The lowest value recorded for the lake
was 58.93 mgC.nf3.h at surface during July, while the
maximum recorded for the lake was 203.12 mgC.nf3.h at 50 cm
during December.
The phytoplankton community in Lake Parker was
represented by blue-green algae including Anabaena sp.,
Microcystis sp., Merismopaedia sp., and Aphanizomenon;


98
Table 5-4. Net primary productivity for all study lakes
expressed as (mgC.m'3.h).
LAKE
DEPTH
08/92
12/92
04/93
07/93
MEAN
S.DEV
0 cm
19.22
23.44
19.36
17.68
19.92
2.46
Parker
50 cm
42.18
60.94
31.16
51.56
46.46
12.75
100 cm
21.09
18.75
20.83
23.44
21.03
1.92
0 cm
53.67
131.25
107.03
151.90
110.96
42.37
Holl a
50 cm
14.06
9.37
91.39
121.87
59.17
56.23
100 cm
14.06
9.29
43.90
56.25
30.87
22.82
0 cm
106.93
140.62
56.25
187.5
122.82
55.33
[ Hanc b
50 cm
56.30
0
0
65.62
30.48
35.40
100 cm
N/A
N/A
N/A
N/A
N/A
N/A
0 cm
124.41
5.46
8.91
164.06
75.71
80.78
1 Gibson
50 cm
97.39
32.81
10.94
92.96
58.52
43.29
100 cm
54.87
0
15.31
62.46
33.16
30.26
0 cm
167.42
158.59
120.79
180.46
156.81
25.64
II Bonny
50 cm
54.91
65.62
45.94
82.03
62.12
15.52
100 cm
34.35
32.81
16.40
35.55
29.78
8.99
0 cm
106.35
38.28
101.35
75.41
80.35
31.15
| Hunter
50 cm
148.31
40.47
169.53
71.09
107.35
61.46
100 cm
71.97
34.45
98.44
76.56
70.35
26.58
(a) Hollingsworth.
(b) Hancock.


37
Mean values in tilapia feces were [1] for the inoculum made
in the medium deficient of nitrogen (-N) and the blue-green
algae -3 64.05 mg.m'3 38.06; and 329.33 mg.nf3 31.38 with
green algae; [2] for the inoculum made in the medium
deficient of phosphorus (-P) and the blue-green algae 683.14
mg.m'3 38.62; and -335.58 mg.m"3 17.98 for the inoculum
with green algae.
The experiment was harvested on the seventh day-
following daily evaluation of fluorescence. As a major
trend, it can be said that fluorescence values increased
dramatically (in the shad samples) during the first four
days, after which they leveled off and did not change very
much. In the medium containing blue-green algae, the
combination shad feces -N as well as shad feces -P displayed
a steady increase in values, which became more noticeable
after the fourth day (Figure 3-1)). For tilapia feces, the
medium deficient of nitrogen (-N) had the same value from
the beginning and declined after the fourth day; whereas the
medium deficient of phosphorus (-P), remained steady and
began to increase slightly after the fourth day (Figure
3-1) .
In the medium containing green algae, the mixture of
shad feces -N increased progressively daily and reached its
highest values after the fourth day, which was a ten-fold
increase. The combination -P displayed the same trend,
however, with higher values increasing thirteen times after


Table 5-7. Correlation analysis of water quality data for Lake Parker
using the General Linear Model procedure (GLM). Data were combined by
year, season and year*season.
PARAMETER
YEAR
SEASON
YEAR*SEASON
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
Secchi
X
X
X
Temperature
X
X
X
PH
X
X
X
Conductivity
X
X
X
D. 0.
X
X
X
TSS
X
X
X
Turbidity
X
X
X
Chlorophyll a
X
X
X
Color
X
X
X
Alkalinity
X
X
X
nh3
X
X
X
TKN
X
X
X
TN
X
X
X
TP
X
X
X
ORT
X
X
X
BACTERIA
X
X
X
103


Table 5-15. Summary of water quality data for L. Gibson.
Numbers shown are mean values with ranges in parentheses
116
Parameter
Unit
Canfield
(1981)
Several
Agencies *
(1989/91)
This study
(1992/93)
Temperature
C
N/A
25.03
(16.4-30.7)
24.43
(17.6-29.7)
D. 0.
mg/L
N/A
7.09
(4.02-9.8)
7.69
(6.57-9.0)
PH
7.2
(6.8-7.6)
7.36
(5.9-10.84)
7.15
(6.4-8.12)
Alkalinity
mg/L CaC03
17
(9-21)
19.01
(4-44)
27.33
(22-32)
Conduct.
pmhos/cm
138
(120-160)
162.78
(109-205)
172.66
(148-200)
H Nitrogen
mg/L
N/A
1.206
(0.1-3.71)
1.183
(0.88-1.69)
I nh3-n
mg/L
N/A
0.076
(0.001-0.3)
0.057
(0.016-0.09)
I TKN
mg/L
N/A
1.087
(0.56-3.43)
0.926
(0.35-1.29)
1 Phosphorus
mg/L
0.650
(0.450-1.0)
0.259
(0.063-0.35)
0.183
(0.16-0.206)
1 Chi. a
mg/m3
15.4
(3.2-28.9)
15.37
(2.2-93.1)
8.83
(5.8-12.8)
| Color
Pt units
N/A
66.6
(40-133)
60.83
(37-79)
U Secchi
m
2.2
(1.4-3.1)
0.87
(0.5-1.4)
1.2
(1.1-1.3)
Bacteria
x 106
N/A
287.97
(1-3600)
23.4
(1-60)
I TSI
N/A
58.5
(46.7-79.4)
55.18
(53.06-57.3)
(*) AGENCY CODES:
USGS; EPA; PCWR; CL; UFC; FGFWFC; CF208; and SWFWMD


147
selective feeding of planktivorous fish. J. Fish. Res.
Bd. Can. 34:1370-1373.
, S.B. Taylor, X. Lazzaro and D. Kettle. 1984 b.
Particle grazing and plankton community impact of an
omnivorous cichlid. Trans. Am. Fish. Soc. 113:397-402.
, S.T. Threlkeld and M.D. McCracken. 1986.
Experimental analysis of the direct and indirect
effects of an omnivorous filter-feeding clupeid on
plankton community structure. Can. J. Fish. Aquat.
Sci. 43:1935-1945.
EDMONDSON, W.T. and S.E.B. Abella. 1988. Unplanned
biomanipulation in Lake Washington. Limnologica
(Berlin) 19:73-79.
ELLIOTT, J.M. 1976. Energy losses in the waste products of
brown trout (Salmo trutta.L). J. Anim. Ecol. 45:561-
580.
FOOTE, K.J. 1977. Annual performance report. Blue tilapia
investigations. Florida Game Fresh Water Fish Comm.,
Tallahassee, Florida.
FLANNERY, M.S., R.D. Snodgrass and T.J. Whitmore. 1982.
Deepwater sediments and trophic conditions in Florida
lakes. Hydrobiologia 92:597-602.
FLORIDA BOARD OF CONSERVATION. 1969. Florida lakes. Part 3.
Gazeteer. Div. Water Resour., Tallahassee.
FORSBERG, C. and S.O. Ryding. 1980. Eutrophication
parameters and trophic state indices in 30 Swedish
waste-receiving lakes. Arch, fur Hydrobiol. 88:189-207.
FROST, T.M., D.L. DeAngelis, S.M. Bartell, D.J. Hall and
S.H. Hulbert. 1988. Scale in the design and
interpretation of aquatic community research. Pages
229-261 in S.R. Carpenter (ed), Complex interactions in
lake communities. Springer- Verlag.
GARLAND, C.R. 1972. A comparative study of the trophic
relationships of the gizzard shad (Dqrqsgma cepedianuim
in Acton Lake and Four-Mile Creek. M.S. thesis, Miami
University, Oxford, Ohio. 51p.
GOTTGENS, J.F. 1992. Quantitative impacts of lake-level
stabilization on sediment and nutrient dynamics:
coupling limnology with modeling. Ph.D. dissertation,
University of Florida, Gainesville.


TABLE OF CONTENTS
ACKNOWLEDGEMENTS ii
ABSTRACT vi
CHAPTER 1 INTRODUCTION 1
CHAPTER 2 LITERATURE REVIEW 7
Introduction 7
Biomanipulation 7
Shad Ecological Energetics 13
Tilapia Ecological Energetics 16
CHAPTER 3 RELATIVE IMPORTANCE OF BLUE TILAPIA AND
GIZZARD SHAD TO LAKE SEDIMENT AND WATER COLUMN
NUTRIENT CONCENTRATION--ANALYSIS OF FISH FECES 22
Introduction 22
Methods 27
Results and Discussion 31
Conclusions 39
CHAPTER 4 SEDIMENTS AND HISTORICAL ECOLOGY OF TWO
CENTRAL FLORIDA LAKES 53
Introduction 53
Study Sites 56
Methods 59
Results and Discussion 60
Conclusions 78
CHAPTER 5 LIMNOLOGICAL ASSESSMENT 89
Introduction 89
Methods 91
Description of Study Area 93
Summary 129
IV


151
MACEINA, M.J. and D.M. Soballe. 1990. Wind-related
limnological variation in Lake Okeechobee, Florida.
Lake and Reserv. Manage. 6:93-100.
MACKERETH, F.J.H., J. Heron and J.F. Tailing. 1978. Water
analysis: some revised methods for limnologists.
Freshwater Biological Association Scientific
Publications no. 36, Ambleside. 120 pp.
MALLIN, M.A., 1986. The feeding ecology of the blue tilapia
(T. aurea) in a North Carolina reservoir. Pages 323-326
in Lake and Reservoir Management. Vol. 2.
MANN, K. 1969. The dynamics of aquatic ecosystems. Pages 1-
81 in J. Cragg (ed.), Advances in Ecological Research
Academic Press, London. Vol. 6 236 pp.
MENZEL, D.W. and N.Corwin. 1965. The measurement of total
phosphorus in seawater based on the liberation of
organically bound fractions by persulfate oxidation.
Limnol. Oceanogr. 10:280-282.
MILLER, R.R. 1960. Systematics and biology of the gizzard
shad (Dorosoma cepedianum) and related fishes. Fishery
Bull., Fish and Wildlife Ser. 60:371-392.
MIURA, T. 1989. Effects of fish plankton feeders on the
plankton community in a small eutrophic lake. Programme
and Abstracts of International Conference
Biomanipulation Tool for Water Management, Amsterdam,
The Netherlands. Aug. B-ll. 116 pp.
MOXLEY, D., V. Williams and C. Harris. 1984. Resource
restoration section 1983-84 annual report. Florida Game
and Fresh Water Fish Commission, Annual Report,
Tallahassee, Florida.
MURPHY, J. and J.P. Riley. 1962. A modified single solution
method for the determination of phosphate in natural
waters. Anal. Chim. Acta 27:31-36.
NAKAMOTO, N. and T. Okino. 1972. Activity of phytoplankton
excreted by fish. The Bulletin of Plankton Society of
Japan 19(1):1-4.
NAUWERCK, A., 1963. Die Beziehungen zwischen Zooplankton und
Phytoplankton in See Erken. Symb. Bot. Upsal. 17(5):163
PP-
NELSON, D.W. and L.E. Sommers. 1975. Determination of total
nitrogen in natural waters. J. Environ. Qual. 4:465-
468.


APPENDIX E
Physical and chemical properties of Lake Parker
sediment core (cf. Schelske et al., 1992) Cont.
Interval
(cm)
Mid-depth
(cm)
Rho
(g.dry. cm*3)
Organic
Matter (%)
Ctot
(%)
N.nt
tot 1
(mg.g*1)
- -
^tot .
(mg.g*1)
50-52
51
0.06073
40.4
23.4
18.8
5.4
54-56
55
0.05879
41.0
21.6
21.4
5.4
58-60
59
0.06078
49.1
28.8
26.8
5.7
60-62
63
0.06016
44.3
23.6
25.0
5.7
66-68
67
0.07233
49.6
28.3
25.6
5.5
70-72
71
0.07785
54.6
31.3
28.0
4.5
74-76
75
0.07166
54.3
30.9
29.2
4.1
78-80
79
0.06960
55.0
31.0
28.6
4.2
82-84
83
0.07834
56.6
32.4
28.3
4.1
86-88
87
0.07616
57.5
32.6
23.2
4.7
90-92
91
0.08900
55.8
31.3
25.1
3.4


APPENDIX H
Sediment core dating for Lake Hollingsworth
(cf. Schelske et al., 1992). Cont.
Depth
(cm)
Total 210Pb
(dpm. g~1)
214Bi
(dpm. g"1)
Excess 210Pb
(dpm.g-1)
137Cs
(dpm.g'1)
Age
(years)
Date
(AC)
Sed. rate
(g.cm"2.yr"1)
50-52
7.13
4.66
2.48
0.59
78.7
1913
0.045
52-60
5.83
3.68
2.16
0.47
93.7
1898
0.039
60-62
4.53
2.70
1.84
0.35
98.2
1894
0.034
62-70
5.22
2.85
2.38
0.33
144.8
1847
0.013
70-72
5.91
3.00
2.92
0.31
N/A
N/A
N/A
72-80
4.68
3.06
0.00
0.21
N/A
N/A
N/A
80-82
3.44
3.11
N/A
0.11
N/A
N/A
N/A
82-90
3.56
3.08
N/A
0.12
N/A
N/A
N/A
90-92
3.68
3.06
N/A
0.12
N/A
N/A
N/A
92-98
4.11
2.78
N/A
0.17
N/A
N/A
N/A
o>


Microcystis aeruginosa
F
I
u
o
r
e
s
c
e
n
c
e
0.05
0.04
0.03
0.02
0.01
4-
' ^ I I I I ^ I I I! ^ I I, I ^
Ctrl -N
Ctrl -P
i i 1 1 1 1 *
-P
1 1* 1 1* 1 1* 2 2* 2 2* 2 2* 3 3* 3 3* 3 3- 4 4* 4 4* 4 4* 6 5* 6 5* 5 5* 6 6- 6 6* 6 6*
Time (days)
Shad
+
Tilapia
* Control
Figure 3-1. Fluorimetric activity from shad and tilapia
feces in medium with Microcystis aeruginosa.
OJ
00
Z .


25
phytoplankton community and to decrease Secchi disk
transparency (Crisman and Beaver, 1988; Threlked and
Drenner, 1987). Shad do not effectively graze on more
evasive zooplankton such as Diaptomus (Drenner et al.,
1978), and many common algal taxa remain viable after gut
passage through the digestive tract of shad, especially
blue-green algae (Velasquez, 1939; Smith, 1963; Crisman and
Kennedy, 1982).
Blue tilapia, Tilapia (=Sarotherodon =Oregchromis)
aurea, a fish native to West Africa and Palestine (Trewavas,
1965), was introduced into the United States in 1957 by
researchers at Auburn University who were investigating its
potential as a food and sport fish (Swingle, 1960). In 1961,
the Florida Game and Freshwater Fish Commission acguired
juvenile fish from Auburn University to study the potential
use of blue tilapia as a sport fish (Crittenden, 1962) and
as an agent for weed control (McBay, 1961; Courtenay and
Robins, 1973). After its introduction, the fish successfully
invaded natural habitats throughout the Southeast,
particularly Florida. Blue tilapia has not, however, been a
successful candidate for use as a sport fish or as a
biological control for excessive macrophyte growth (Ware et
al., 1975).
Tilapia can be classified as an opportunistic omnivore,
consuming zooplankton (Spataru and Zorn, 1978; Mallin,
1986), phytoplankton (Hendricks and Noble, 1980; Mallin,


131
Parker, where the zooplankton population was two times
larger than the other lakes, data that are in agreement with
Crisman and Beaver (1988).
Lakes considered as having more blue tilapia also have
proportionally lower gross primary productivity (Figure 5-1)
than the group where gizzard shad is considered dominant,
which is consistent with the results of Crisman and Beaver
(1988).
Regarding to the fish population, the two extremes were
in the group where blue tilapia is prevalent. Lake Parker
has the smallest fish population of the six studied lakes
(total fish biomass = 71.1 kg.ha"1) (Canfield and Hoyer,
1991), and Lake Hollingsworth the largest fish population
(total fish biomass = 1050 kg.ha"1) (Canfield and Hoyer,
1992). The small fish population in Lake Parker could also
help to explain the presence of the large zooplankton
population in that lake. There were no noticeable
differences in the size of the fish populations for Lakes
Bonny and Hunter, both belonging to the group where gizzard
shad was dominant.


mg/rrT3
Time
Figure 3-4. Bioassay initial (I) and final (F) values of chi. a (mg.nf3)
with M.aeruginosa in medium -P.


Table 5-22. Correlation analyses of water quality data for Lake Hunter
using the General Linear Model procedure (GLM). Data were combined by
year, season and year*season.
PARAMETER
YEAR
SEASON
YEAR*SEASON
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
Secchi
X
X
X
Temperature
X
X
X
PH
X
X
X
Conductivity
X
X
X
o

Q
X
X
X
TSS
X
X
X
Turbidity
X
X
X
Chlorophyll a
X
X
X
Color
X
X
X
Alkalinity
X
X
X
NH-,
X
X
X
TKN
X
X
X
TN
X
X
X
TP
X
X
X
ORT
X
X
X
BACTERIA
X
X
X
128


153
PULLIN, R.S.V. and R.W. Lowe-McConnell. 1984. The biology
and culture of Tilapia. Pages 34-57 in ICLARM
Conference Proceedings 7, International Center for
Living Aquatic Resources.
PURI, H.S. and R.O. Vernon. 1964. Summary of the geology of
Florida and a guidebook to the classic exposures.
Florida Geological Survey Special Publication n. 5.
REINERTSEN, H. and Y.Olson. 1984. Effects of fish
elimination on the phytoplankton community of a
eutrophic lake. Verh. Internat. Verein. Limnol. 22:649-
657.
ROUND, F.E. 1977. The Biology of the Algae. Edward Arnold,
London. 278 pp.
SAS Institute, Inc. 1989. SAS user's guide: statistics. SAS
Institute, Inc.
SCHWOERBEL, J. 1975. Mtodos de Hidrobiologia. H. Blume
Ediciones, Madrid. 262 pp.
SCOTT, W.B. and E.J. Crossman. 1973. Freshwater fishes of
Canada. Fish. Res. Bd. Can. Bull 184R. 966 p.
SHAFLAND, P.L., 1978. A status report on tilapia in Florida.
Non-native Fish Research Laboratory, Contribution no.
16. 5 pp.
SHAPIRO, J., 1978. The need for more biology in lake
restoration. Pages 161-167 in Lake Restoration. EPA-
440/5-79-001, Washington, D.C.
, 1990. Biomanipulation: the next phase making
it stable. Hydrobiologia 200/201:13-27.
, V. Lamarra and M. Lynch, 1975. Biomanipulation:
An ecosystem approach to lake restoration. Pages 85-86
in P.L. Brezonik and J.L. Fox (eds.), Water Quality
Management Through Biological Control. Dept. Environ.
Engineering Sciences, University of Florida,
Gainesville, 164 pp.
, and D.I. Wright. 1984. Lake restoration by
biomanipulation: Round Lake, Minnesota, the first two
years. Freshwat. Biol. 14:371-383.
SMELTZER, E. and E.B. Swain. 1985. Answering lake management
questions with paleolimnology. Pages 268-274 in Lake
Reserv. Manage.- Practical Applications. Proc. 4th
Annu. Conf. and Symp. (NALMS) 390pp.


86
noted that then was an increase in phosphorus concentration
after the 1950s, matching with the period of more intense
occupation in the watershed for the whole area. Water
quality measurements collected since 1966 demonstrate that
the lake has been hypereutrophic for at least 25 years
(Schelske et al., 1992). Although the lake is naturally
productive, anthropogenic impacts have accelerated the rate
of eutrophication. Eutrophication problems, especially from
cultural enrichment, should be resolved in a number of ways,
ranging from eutrophication prevention and lake
rehabilitation, to learning to live with the problem.
From the sediment core analysis of Lakes Parker and
Hollingsworth, lakes considered as having a dominant blue
tilapia population in this study, and Lakes Bonny and
Gibson, considered as having a dominant gizzard shad
population, it is not clear that there are any significant
differences in sediment organic matter content, nitrogen,
phosphorus or any other analyzed parameter. All four lakes
seem to have high organic matter content in the topmost
layers, and all of them had a record in the sediments of the
main environmental changes for the area e.g. the 1950's
switch from totally agricultural practices in the area to
watershed conversion to urban development, as well as the
efforts been made to recover these systems since the mid
1980's.


27
Methods
Fish Collection
Blue tilapia were collected by electroshocking in Lake
Alice, Florida. A boom-style electrofishing boat was used
(7.5 GPP, with a 7 KW generator, Smith-Root, Inc. Vancouver,
Washington). This model was the most efficient
electrofishing unit for the type of lake sampled. The
collected fish were immediately placed on ice to minimize
post-capture digestive processes.
Blue tilapia were measured for total length (mm) and
weight (g). All fish collected were separated into two size
groups (TL), less than 380 mm and greater than 380 mm, which
will subsequently be referred to as size groups T-l and T-2,
respectively. Thirty-six fish were collected. Thirteen of
these were in group T-l; twenty-three were in group T-2
(fish length ranged from a minimum of 284 mm to a maximum of
450 mm). Fecal material, here defined as the collection of
all digested material that could be identified by its
brownish-dark coloration in the fish intestines, was hand-
stripped from all the fishes and pooled into two different
jars (T-l and T-2). The sample jars were immediately placed
on ice in a cooler for transport to the laboratory. All
sample material was kept at 4C within four hours after
collection.
Samples were subdivided into four different groups;
groups 1 and 2 pulled from the T-l jar and groups 3 and 4


113
total phosphorus, orthophosphate, and total bacteria.
Secchi, temperature, conductivity, turbidity, chlorophyll a,
color and total bacteria exhibited significant seasonal
variation at the 95% confidence level (Table 5-13).
Interannual variation or no variation at all was reported
for the other parameters. When year and seasonality were
combined, Secchi, conductivity, turbidity, color and total
bacteria were the parameters displaying significant
variability.
Lake Gibson
Lake Gibson is located in Polk County, Florida
(Appendix A). The basin lies in the Polk Uplands
physiographic region and is situated in sandy deposits of
the Hawthorne and Bone Valley Formations. Lake Gibson is on
the outskirts of the City of Lakeland, with a surface area
of 192 ha, a shoreline length of 6.9 km, and a mean depth of
3.5 m (Table 5-14).
Table 5-14. Morphometry of Lake Gibsoncu
Surface Area 192 ha
Maximum Depth 6.1 ra
Mean Depth 3.5 m
Development of Shoreline 65 %
Drainage Basin Area 1.1 x 103 ha
Shoreline Length 6.9 km
Macrophyte Cover 5 %
(1) Data from PCWR (1992).


17
1973) as a means of controlling nuisance aquatic macrophytes
(Courtenay and Robins, 1973; McBay, 1961). After its
original introduction, the fish spread rapidly throughout
the southeast, particularly Florida, but have not controlled
excessive macrophyte growth (Ware et al., 1975).
Qualitative and quantitative measurements of the gut
contents of blue tilapia in situ indicate that this species
is an opportunistic omnivore, utilizing zooplankton (Spataru
and Zorn, 1978; Mallin, 1986), phytoplankton (Hendricks and
Noble, 1980; Mallin, 1986), and detritus (Hendricks and
Noble, 1980; Mallin, 1986). Gophen et al. (1983) found in
laboratory tests that tilapia greater than 7.6 cm utilized a
series of rapid suctions to draw prey into their buccal
cavity. This mechanism is undirected, and thus, fish in this
size range function as filter feeders. Tilapia smaller than
7.6 cm also function as filter-feeders; however, they also
feed as size selective predators on individual zooplankton
specimens.
Drenner et al. (1984a) reported that grazing activities
by blue tilapia depressed some large algae (Uroqlenopsis and
Ceratium), while the smallest phytoplankton taxa were
enhanced (Rhodomonas. Chrvsochromulina. Chlamvdomonas and
Cyclotella). This enhancement was ascribed both to nutrient
regeneration during gut passage and to fish feces, as well
as the accompanying compositional shifts in the herbivorous
zooplankton community. Little information is available on


16
eutrophic lakes, the death of large numbers of shad, besides
any aesthetic consequences, will represent an enormous
return of phosphorus and nitrogen into the water column that
is readily available to stimulate the growth of already
overly productive algae, resulting in oxygen depletion that
could lead to another fish kill.
The results of the Crisman and Kennedy (1982)
investigation suggest that the presence of gizzard shad can
promote lake eutrophication both through elevation of
orthophosphate concentrations and differential digestion of
diatoms and green algae, thus increasing the competitive
advantage of blue-green algae. Finally, they emphasize that
this fish species does not appear to be a suitable candidate
for use as a biocontrol agent for phytoplankton in eutrophic
subtropical lakes.
Considerable literature is available concerning the
life history of Dorosoma cepedianum (Drenner, 1977; Lazzaro,
1987). Only a few investigators, however, have attempted to
describe and quantify energy relationships in this species
(Smith, 1971; Garland, 1972; Pierce, 1977; Crisman and
Kennedy, 1982).
Tilapia Ecological Energetics
Blue tilapia (Tilapia aurea), a species native to
Africa and the Middle East, was introduced into the United
States in 1961 in a Hillsborough County phosphate pit (Ware,


96
Table 5-2. Summary of water quality data for Lake Parker.
Numbers shown are mean values with ranges in parentheses.
Parameter
Unit
Canfield
(1981)
Several
Agencies *
(1989/91)
This study
(1992/93)
Temperature
C
N/A
24.95
(16.4-31.8)
24.81
(17.7-30.2)
D.O.
mg/L
N/A
8.5
(4.7-13.2)
9.1
(8.1-10.3)
PH
8.9
(8.8-9.1)
8.9
(5.4-10.9)
8.5
(7.5-9.3)
Alkalinity
rag/L CaC03
68
(63-76)
55
(40-64)
43.5
(34-50)
Conduct.
pmhos/cm
215
(200-260)
387
(207-470)
247
(207-271)
Nitrogen
mg/L
1.56
(0.98-1.97)
3.15
(0.63-11.4)
3.13
(2.19-3.69)
nh3-n
mg/L
N/A
0.06
(0.01-0.19)
0.05
(0.02-0.07)
TKN
mg/L
N/A
2.9
(0.6-11.0)
2.8
(1.3-3.2)
Phosphorus
mg/L
0.08
(0.06-0.11)
0.15
(0.02-.34)
0.22
(0.07-0.602)
Chi. a
mg/m3
33.7
(9.3-49.3)
75.7
(18.6-222.8)
53.8
(43.0-76.6)
Color
Pt units
22
(15-30)
24.2
(6-57)
42.6
(29-53)
Secchi
m
0.5
(0.4-0.6)
0.4
(0.25-1.5)
0.35
(0.25-0.4)
Bacteria
x 106
N/A
761.7
(1-7800)
18.7
(1-60)
TSI
N/A
79.6
(62.7-94.0)
82.7
(76.6-86.9)
(*) AGENCY CODES:
USGS; EPA; PCWR; CL; UFC; FGFWFC; CF208; and SWFWMD


APPENDIX D
Physical and chemical properties of Lake Parker
sediment core (cf. Schelske et al., 1992)
Interval
Mid-depth
Rho
Organic
Ctot
Ntot ,
^tot .
(mg. g'1)
(cm)
(cm)
(g. dry. cm'3)
Matter (%)
(%)
(mg. g"1)
0-2
1
0.01820
71.7
34.9
63.7
4.9
2-4
3
0.02575
71.4
35.3
N/A
N/A
4-6
5
0.02702
70.1
34.3
71.8
5.5
6-8
7
0.02895
63.4
31.9
N/A
N/A
8-10
9
0.03554
59.1
28.6
23.3
4.5
10-12
11
0.03667
54.0
26.3
N/A
N/A
12.-14
13
0.03652
52.1
27.2
23.5
4.5
14-16
15
0.04000
52.0
26.6
N/A
N/A
16-18
17
0.03919
52.3
27.7
25.4
4.9
18-20
19
0.04368
50.1
26.8
22.7
5.1
22-24
23
0.04846
48.6
24.9
24.8
5.5
26-28
27
0.04934
47.7
24.6
25.6
5.9
30-32
31
0.05047
48.8
26.3
22.7
6.7
34-36
35
0.05258
48.4
26.6
23.8
6.5
38-40
39
0.05625
41.5
20.7
22.3
6.0
42-44
43
0.05751
37.1
20.4
19.0
6.5
46-48
47
0.05741
37.4
21.7
19.6
6.8
o


61
3.93, mean organic matter of 56.00% 12.69, mean total
nitrogen of 22.32 mg.g'1 2.68 and mean total phosphorus of
4.10 mg.g'1 1.53 (Table 4-1).
Lake Bonny had high organic content in surface
deposits, but this declined somewhat with depth and age down
to 19 cm of the core which coincides with the 1980s, and
then increased again to the bottom where it reached highest
values (dated at the beginning of this century, 1909)
(Figure 4-1). Total nitrogen and total phosphorus plotted
against time were constant throughout the whole core (Figure
4-2). Bulk density increased in the top 40 cm of the core
which coincides with a 210Pb determination of sediment age
of approximately 1950, i.e., about the time when much
agriculture in the watershed was being converted to urban
development (USGS Topographic Maps, Lakeland Quadrangle,
1944, 1975).
Total P reconstruction near the base of the core
suggests that the lake has always been mesotrophic to
eutrophic. However, the lake has experienced periods of much
higher nutrient enrichment, and values for total P in excess
of 5.0 mg.g'1 were computed for several sections that post
date the 137Cs peak of 1960s and 1970s. No change in total P
concentration has been noticed since then, suggesting that
the lake trophic state has remained the same for the last
thirty years.


146
, and J.R. Beaver. 1990. Applicability of
planktonic biomanipulation for managing eutrophication
in the subtropics. Hydrobiologia 1(11):177-186.
, and H.M. Kennedy, 1982. The role of gizzard
shad (Dgrgsgma cepedianum) in eutrophic Florida lakes.
Publ. 64. Water Resources Research Center, University
of Florida, Gainesville, 83 pp.
CROWDER, L.B. 1985. Optimal foraging and feeding mode shifts
in fishes. Environ. Biol. Fishes 12(l):57-62.
CROWDER, C.B., R.W. Drenner, W.C. Kerfoot, D.J. McQueen,
E.L. Mills, V. Sommer, C.N. Spencer, and J.J. Vanni.
1988. Food web interactions in lakes. Pages 314-397 in
S.R. Carpenter (ed), in Complex Interactions in Lake
Communities. Springer Verlag, New York.
DALQUEST, W.N. and L.J. Peters. 1966. A life history study
of four problematic fish in Lake Diversion, Archer and
Baylor counties, Texas. Texas Park and Wildlife Dept.
I.F. Series, no. 6:87 p.
DICKMAN, M. and H. Nanne. 1987. Impact of tilapia grazing on
plankton composition in artificial ponds in Guanacaste
Province, Costa Rica. J. Freshwat. Ecol. 4: 93-100.
DRENNER, R.W. 1977. The feeding mechanism of the gizzard
shad (Dgrgsgma cepedianum). Ph.D. Thesis Univ. Kansas
Lawrence.
, F. de Noyelles Jr. and D. Kettle. 1982a. Selective
impact of filter feeding gizzard shad on zooplankton
community structure. Limnol. Oceanogr. 27:965-968.
, K.D. Hambright, G.L. Vinyard, M. Gophen and V.
Pollingher. 1987. Experimental study of size-selective
phytoplankton grazing by a filter-feeding cichlid and
the cichlid's effects on plankton community structure.
Limnol. Oceanogr. 32:1140-1146.
, J.R. Mummert, F. de Noyelles Jr. and D. Kettle.
1984a. Selective particle ingestion by a filter-feeding
fish and its impact on phytoplankton community
structure. Limnol. Oceanogr. 29(5):941-948.
, W.J. O'Brien and J.R. Mummert. 1982b. Filter
feeding rates of gizzard shad. Trans. Am. Fish. Soc.
111:210-215.
, J.R. Strickler and W.J. O'Brien. 1978. Capture
probability. The role of zooplankton escape in the


148
GOPHEN, M. 1990. Biomanipulation: retrospective and future
development. Hydrobiologia 200/201:1-11.
, R.W. Drenner and G.L. Vinyard. 1983. Cichlid
stocking and the decline of the Galilee Saint Peters
Fish (Sarotherodon galilaeus) in Lake Kinneret, Israel.
Can. J. Fish. Aquat. Sci. 40: 983-986.
HAITH, D.A., W. Rast, K.H. Reckhow, L. Somlydy and G. van
Straten. 1989. The use of models. Pages 85-113 in S.-O.
Ryding and W. Rast (eds.), The Control of
Eutrophication of Lakes and Reservoirs, Parthenon Publ.
Group Inc., Park Ridge, N.J. 314 pp.
HAKANSON, L. and M. Jansson. 1983. Principles of Lake
Sedimentology. Springer-Verlag, New York.
HALL, D.J., W.E. Cooper & E.E. Werner. 1970. An experimental
approach to the production dynamics and structures of
freshwater animal communities. Limnol. Oceanogr.
15:839-928.
HALVER, J.E. 1989. Fish Nutrition. 2nd edition. San Diego
Academic Press, New York. 798 pp.
HANAZATO, T., H. Hayashy, T. Ichikawa and Y. Watanabe. 1989.
Dynamics of zooplankton community in enclosures of
different types in a shallow eutrophic lake. Jpn. J.
Limnol. 50(l):25-37.
HANEY, J.F. and D.J. Hall. 1973. Sugar-coded Daphnia A
preservation technique for cladocera. Limnol. Oceanogr.
18:331-333.
HEAT, R.C. and C.S. Conover. 1981. Hydrologic Almanac of
Florida. U.S. Geological Survey, Tallahassee, Florida.
HENDRICKS, M.K. and R.L. Noble. 1980. Feeding interactions
of three planktivorous fishes in Trinidad Lake, Texas.
Proc. Annu. Conf. SE Ass. Fish. Wildl. Agencies 33:
324-330.
HENRIKSON, L., H.G. Nyman, H.G. Oscarson and J.A. Stenson.
1980. Trophic changes without changes in the external
nutrient loading. Hydrobiologia 68:257-263.
HOBBIE, J.E. and R.J. Daley. 1977. Use of nucleopore filters
for counting bacteria by epifluorescent microscopy.
App. Environ. Micro. 33:1225-1228.


1400
1200
1000
800
600
400
200
0
.m~3
Control Tilapia Shad
Time
3-6. Bioassay initial (I) and final (F) values of chi. a (mg.m'3)
with S. capricornutum in medium -P. 5


10
et al. (1975) and Shapiro (1978) recognized that
eutrophication problems are biological manifestations of
nutrient availability, and were the first to suggest that
manipulation of trophic interactions (biomanipulation) could
be used as a lake management tool to alleviate the
biological consequences of eutrophication without the need
for costly controls on nutrient loading (Crisman and Beaver,
1990).
Since biomanipulation research began in temperate
regions, the emphasis has focused basically on alterations
of the zooplankton community (enhancement of large
cladocerans especially of the genus Daphnia), because,
through these manipulations, phytoplankton has been reduced
(blue-green algae) and water transparency has increased
despite difficulties in maintaining such a condition for a
long time. Only recently has research been conducted using
fishes as a major element in the biomanipulation process,
but still only under temperate conditions.
In subtropical and tropical regions, conditions are
quite different. One particular element must be stressed;
there is a complete absence of large zooplankton in tropical
regions. This and numerous other physicochemical and
biological differences between these climatic regions
impedes the applicability of biomanipulation research from
the temperate regions to warmer climates (Crisman and
Beaver, 1990). Thus, the natural choice for biomanipulation


93
formalin (Haney and Hall, 1973) containing rose bengal
stain. Obligue tows were started as close to the bottom as
possible, and the sampler was raised to the surface at a
slow, constant rate; this procedure was repeated four times,
filtering up to 32 L of water. Fish populations were
estimated from data of Canfield and Hoyer (1992) and Florida
Game and Freshwater Fish Commission (FGFWFC).
Description of Study Area
Lake Parker
Lake Parker is located in Polk County, Florida
(Appendix A). The basin lies in the Polk Uplands
physiographic region and is situated in phosphatic deposits
from the Hawthorn and Bone Valley Formations. Due to its
location just east of the Lakeland Ridge, the lake receives
inputs of groundwater which have been in contact with
limestone (Stewart, 1966). Lake Parker is a large urban lake
with a surface area of 924 ha, shoreline length of 19.8 km
and mean depth of 2.2 m (Table 5-1).


121
Table 5-18. Summary of water quality data for Lake Bonny.
Numbers shown are mean values with ranges in parentheses.
| Parameter
Unit
Several Agencies*
(1989/91)
This study
(1992/93)
| Temperature
C
24.87 (16.08-30.55)
24.26 (17.48-29.58)
D.O.
mg/L
7.24 (3.85-11.34)
7.47 (6.24-8.63)
PH
8.14 (5.9-9.79)
7.26 (6.28-8.34)
Alkalinity
mg/L CaC03
57.88 (48-84)
47.5 (37-56)
Conduct.
pmhos/cm
2 1 9.7 (170-257)
194.5 (158-219)
Nitrogen
mg/L
2.94 (0.74-7.2)
2.48 (1.5-4.06)
I NH,-N
mg/L
0.0 6 6 (0.001-0.20)
0.079 (0.031-0.13)
I TKN
mg/L
2.83 (0.73-6.2)
2.29 (1.1-3.75)
1 Phosphorus
mg/L
0.14 (0.019-0.36)
0.11 (0.073-0.169)
8 Chi. a
mg/m3
62.9 (18.1-182.5)
44.35 (23.3-94.7)
Color
Pt units
37.5 (13-100)
53.66 (39-67)
1 Secchi
m
0.5 (0.2-1.1)
0.56 (0.3-0.75)
| Bacteria
x 106
1035.9 (1-13400)
165.83 (10-540)
I TSI
7 6.72 (61.84-96.61)
7 3.46 (68.06-86.28)
(*) AGENCY CODES:
USGS; EPA; PCWR; CL; UFC; FGFWFC; CF208; and SWFWMD.


108
variation at the 95% confidence level (Table 5-10). Other
parameters displayed either interannual variations or no
variation at all. When year and seasonality were combined,
conductivity, turbidity, color, alkalinity and total
bacteria were the parameters displaying significant
variability.
Lake Hancock
Lake Hancock is located in Polk County, Florida
(Appendix A). The basin lies in the Polk and Lake Uplands
physiographic region, and it is situated in phosphatic
deposits from the Hawthorne and Bone Valley Formations. Lake
Hancock is a large lake with a surface area of 1839 ha,
shoreline length of 16.7 km, and mean depth of 0.85 m (Table
5-11).
Table 5-11. Morphometry of
Lake Hancockd)
Surface Area
1839 ha
Maximum Depth
1.5 m
Mean Depth
0.85 m
Development of Shoreline
5 %
Drainage Basin Area
34.1 x 103 ha
Shoreline Length
16.7 km
Macrophyte Cover
1 %
(1) Data from PCWR (1992).
The lake has historically received wastewater
discharges via Banana Lake from the City of Lakeland, and
via Lake Lena Run from the City of Auburndale. Both have


Table 4-7. Total phosphorus (mg.g"1) for
Lake Bonny sediment core.
77
I Interval (cm)
TP (mg/g)
I 0-2
5.1567
I 2-4
5.2729
4-7
5.2941
7-9
5.3534
9-11
6.0308
I 11-13
4.6168
I 13-15
4.8145
I 15-17
5.2071
I 17-19
3.5261
I 19-22
4.9261
1 22-26
5.8654
26-30
6.3821
30-34
5.0733
34-38
3.6455
38-42
3.6842
42-46
3.0309
46-50
2.4239
50-54
2.4907
j 54-58
2.2233
I 58-64
2.0648
[ 64-72
1.5072
I 72-84
1.5842
H Mean 4.1
= a BBS J1


24
demonstrated that planktivorous fish (mainly filter-feeder
fish) can severely reduce or even eliminate large-bodied
zooplankter Daphnia spp.
Planktivorous fishes use two distinct means to feed on
plankton: particulate feeding and filter feeding.
Particulate feeders attack single individual planktonic prey
items which they visually select from the water column
(Werner, 1977; Vinyard, 1980; Lazzaro, 1987). Filter feeders
do not visually detect individual prey but swallow a volume
of water containing the food organisms and entrap the
planktonic forms in structures such as gill rakers and other
filtering structures (see Lazzaro, 1987), using rhythmic
suctioning actions to capture the prey, while either
swimming slowly or remaining quite stationary (Drenner,
1977; Gophen et al., 1983).
Gizzard shad (Dorpspma cepedianunH a filter feeder, is
a dominant native grazer in Florida eutrophic lakes. Crisman
and Kennedy (1982) used mesocosm experiments to demonstrate
that gizzard shad had no impact on chlorophyll a values,
lake productivity, or phytoplankton density. Gizzard shad
was responsible for a significant increase in both the
concentration of orthophosphate and its ratio to total
phosphorus under natural stocking conditions and for the
decrease in copepod density (Crisman and Kennedy, 1982).
Later experiments, however, demonstrated that the
overall effect of shad grazing was to stimulate the


99
Table 5-5. Community respiration for all study lakes
expressed as (mgC.m'3.h).
LAKE
DEPTH
08/92
12/92
04/93
07/93
MEAN
S.DEV
0 cm
44.84
54.68
45.17
41.25
46.48
5.74
1 Parker
50 cm
98.44
142.18
72.7
120.31
108.41
29.76
n~
100 cm
49.22
43.76
48.6
54.68
49.06
4.47
0 cm
125.23
306.25
249.75
354.44
258.92
98.86
Holl a
50 cm
32.81
21.88
213.26
284.38
138.08
131.2
100 cm
32.81
21.69
102.45
131.25
72.05
53.24
0 cm
249.52
328.13
131.25
437.5
286.6
129.1
Hanc b
50 cm
131.37
0
0
153.13
71.12
82.61
100 cm
N/A
N/A
N/A
N/A
N/A
N/A
0 cm
225.48
10.16
16.55
304.69
139.22
148.9
| Gibson
50 cm
180.87
60.94
20.31
172.64
108.69
80.40
100 cm
101.91
0
28.44
116.0
61.59
56.21
0 cm
310.92
294.53
224.33
335.14
291.23
47.62
Q Bonny
50 cm
101.98
121.88
85.31
152.34
115.38
28.82
100 cm
63.79
60.94
30.47
66.01
55.30
16.68
0 cm
197.5
71.09
188.21
140.06
149.21
57.85
Hunter
50 cm
275.43
75.15
314.84
132.03
199.36
114.1
100 cm
133.66
64.00
182.81
142.19
130.66
49.35
(a) Hollingsworth.
(b) Hancock.


18
nutrient release and degree of algal digestion by blue
tilapia (Crisman and Beaver, 1988), but Popma (1982) noted
that some algal cells may remain viable following passage
through the digestive tract of tilapia.
The zooplankton community is also modified by tilapia
grazing activity, and Drenner et al.(1984a) showed in pond
studies that the population of Keratella was suppressed,
while copepodids and adult Diaptomus were enhanced. Gophen
et al. (1983) reported that blue tilapia are selective
feeders on Bosmina and Ceriodaphnia, taxa that have poor
evasive capabilities when compared with more successful
zooplankters such as Mesocyclops (Crisman and Beaver, 1988).
Dickman and Nanne (1987) noted that in Central American fish
ponds very high levels of tilapia (2.5 adults.m'2)
suppressed zooplankton populations and increased the
importance of the bluegreen alga Microcystis aeruginosa.
The spawning behavior of Tilapia aurea is typical of
many cichlids with all incubating duties being performed by
the female (McBay, 1961). Reproductive behavior progresses
from schooling, territorial establishment by the males,
prespawning courtship, spawning, and parental care. McBay
(1961) suggested that the mature female will spawn at a
constant water temperature of 23C. The number of hatchlings
per spawn of T.aurea was comparatively smaller than in most
fishes native to the United States, but apparently
compensated for by strong parental care behavior.


21
Table 2-1. Summary of previous investigations with blue
tilapia and gizzard shad in Florida lakes.
GIZZARD SHAD
| Chlorophytes
f Cyanophytes
? Chrysophytes
f Chrysophytes
f Rotifers
? Rotifers
J, Cladocerans
1 Cladocerans
J, Zooplankton biomass
f Primary Productivity
Previous studies (Courtenay and Robins, 1973; Ware et
al., 1975; Crisman and Kennedy, 1982; Bays and Crisman,
1983; Zale, 1984; Crisman et al., 1986; Crisman and Beaver,
1988) delineated the differential impact that each fish has
on distinct groups of organisms in Florida lakes. It is not
clear if any change can be deduced for chrysophytes in
systems dominated by blue tilapia, whereas some increase in
chlorophytes is reported. Gizzard shad increased cyanophytes
and chrysophytes. Regarding zooplankton composition, blue
tilapia is reported to increase rotifers while suppressing
cladocerans. The gizzard shad impact on rotifers has not
been determined, although it is known that cladoceran
populations decrease. Finally, both species seem to suppress
total zooplankton biomass while gizzard shad was shown to
stimulate phytoplankton primary productivity.


78
The Lake Hollingsworth Record
Inferences from paleolimnological core studies indicate
that Lake Hollingsworth has had high nutrient and
chlorophyll a concentrations since the 19th century;
however, the highest levels occurred between the 1950s and
the 1970s, when much of the agriculture in the watershed was
converted to urban development (Schelske et al., 1992). Lake
sediments' topmost 10 cm is poorly consolidated (densities
of <0.032 g.dry.cm-3.wet) (Schelske et al., 1992), although
density generally increases with depth and is highest in
sand-rich deposits.
Organic matter in the core generally ranges between 40
and 60% of dry weight but decreases in sandy deposits
(Schelske et al., 1992). Total phosphorus concentrations
have remained relatively constant during the last ten years,
ranging from 59 to 71 pg.L'1 (Schelske et al., 1992).
Appendix F show physical and chemical properties of Lake
Hollingsworth sediment core.
Conclusions
The data presented provide support for the conclusion
of Flannery et al.(1982) that Florida lakes of higher
trophic state have greater proportions of organic matter in
their surface sediments e.g. Lake Bonny (TSI of 73.46) with
57% organic matter and Lake Gibson (TSI of 55.18) with 35%
organic matter. Organic matter in the Lake Parker


117
highest values recorded were at the surface; the annual mean
was 214.18 mgC.ra'3.h + 229.1, with the minimum value of
15.62 mgC.m~3.h during winter and 4 68.7 5 mgC.nf3.h during
summer (Table 5-3). Values for net photosynthesis are shown
in table 5-4, community respiration in table 5-5, and
average annual productivity in table 5-6.
The phytoplankton community in Lake Gibson was
dominated by the blue green algal genus Spirulina.
Throughout the year, other blue green genera were
represented including Merismopedia sp. and Lvnqbva sp., as
well as the dinoflagellate Peridinium sp.
The zooplankton assemblage was comprised mainly of
rotifers. Brachionus sp., Keratella sp., and Collurella sp.,
were the most numerically abundant rotifers. The only
copepod was Diaptomus sp.
No recent fish population survey was available for this
lake. Ten species of fish were reported from the last
assessment done in 1979. The dominant species at that time
was channel catfish, followed by blue tilapia, with 63.9 %
and 26.0 % of total fish weight, respectively.
ANOVAs were performed using the General Linear Model
(GLM) (SAS, 1989) in order to identify significant seasonal
limnological factors (p < 0.05). The following parameters
were analyzed: Secchi, temperature, pH, conductivity,
dissolved oxygen, total suspended solids, turbidity,


11
schemes of subtropical and tropical regions is the use of
planktivorous fish. Despite the potential for utilization of
filter-feeding fish in biomanipulation techniques to
maintain algal control, especially in shallow eutrophic
lakes and reservoirs characterizing subtropical and tropical
regions, little information exists regarding the direct
impact on the general planktonic community.
Florida lakes possess characteristics which
fundamentally differentiate them from most temperate
systems. The relative shallowness of Florida lakes, combined
with a moderate climatic regime, dictates that stable
thermal stratification usually only occurs in deeper (>5-6
m) lakes of the region (Beaver et al., 1981). Such lakes are
warm monomictic and circulate throughout the winter months,
while shallower basins are more strongly affected by local
meteorological events and are subjected to higher internal
nutrient loading via sediment resuspension (Pollman, 1982).
Subtropical lakes of Florida are inherently different
from temperate lakes in a number of respects that could
affect the success of whole lake biomanipulation. The size
structure of subtropical zooplankton communities is skewed
toward smaller individuals, and large-bodied cladocerans are
absent (Bays and Crisman, 1983). Despite the lack of large
bodied crustaceans, intense grazing activities by gizzard
shad may strongly determine plankton structural and
functional characteristics (Bays and Crisman, 1983). These


Table 5-10 Correlation analyses of water quality data for Lake Hollingsworth
using the General Linear Model procedure (GLM). Data were combined by
year, season and year*season.
PARAMETER
YEAR
SEASON
YEAR*SEASON
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
Secchi
X
X
X
Temperature
X
X
X
PH
X
X
X
Conductivity
X
X
X
D.O.
X
X
X
TSS
X
X
X
Turbidity
X
X
X
Chlorophyll a
X
X
X
Color
X
X
X
Alkalinity
X
X
X
NH-,
X
X
X
TKN
X
X
X
TN
X
X
X
TP
X
X
X
ORT
X
X
X
BACTERIA
X
X
X
109


CHAPTER 3
RELATIVE IMPORTANCE OF BLUE TILAPIA AND GIZZARD SHAD TO LAKE
SEDIMENT AND WATER COLUMN NUTRIENT CONCENTRATION--
ANALYSIS OF FISH FECES
Introduction
Examination of trophic-level interactions has long been
an integral part of limnology (Caird, 1945; Hrbcek et al.,
1961; Nauwerck, 1963; Brooks and Dodson, 1965). Elements
such as herbivory, predation, and nutrient recycling by
animals have always been considered to affect the biomass
and species composition of prey populations in terrestrial
as well as aguatic communities. Several investigators (e.g.,
Hairston et al., 1960; Wiegert and Owen, 1971; Patten, 1973;
Porter, 1977; and Paine, 1980) have examined several aspects
of food webs and the factors which control the biomass and
productivity of various trophic levels. Applicability of
these ideas to aid management practices and promote aguatic
ecosystem recovery are now under investigation (Cooke et al.
1993) .
Shapiro et al. (1975) and Shapiro (1978) recognized
that eutrophication problems are biological manifestations
of nutrient availability and were the first to suggest that
manipulation of trophic interactions (biomanipulation) could
be used as a lake management tool to alleviate the
22


55
and estimated trophic status based on sedimentation rates
(g.cm'2.yr) for the last century.
Sediment cores collected from Lake Bonny and Lake
Gibson (Polk County, Florida) were dated isotopically with
210Pb and 137Cs to estimate material accumulation rates in
the sediment profile using radioactive decay of fallout
("unsupported") 210Pb. These techniques have been widely
used to detect changes in sediment accumulation rates due to
urban development in a watershed (Smeltzer and Swain, 1985),
clear cutting of vegetation (Oldfield et al., 1980) and
major climatic events (Robins et al., 1978).
In this study, these techniques were applied to
identify lake sediment profile physicochemical
characteristics and evaluate the potential impact of gizzard
shad feces when it is the dominant filter feeder fish on
lake sediment composition and net accumulation rates of
organic matter, total nitrogen and total phosphorus for
Lakes Bonny and Gibson. Whenever possible, this information
will be correlated with the results from sediment cores for
Lake Parker and Lake Hollingsworth reported by Schelske et
al. (1992). For this study, both lakes were considered to be
dominated by blue tilapia population as the main filter
feeder fish.
According to the classification of Forsberg and Ryding
(1980), Lake Bonny is hypereutrophic and Lake Gibson is an
eutrophic lake. Mean values for all physicochemical and


159
APPENDIX C
Lake Gibson sedimentation rate (g.cm"2.yr)
by year.
LAKE GIBSON
Core
Sect
(cm)
Bulk
Density
(g/cm3)
Unsupp.
Pb-210
(pCi/g)
Unsupp.
Pb-210
(pCi/cm2
Cum.Res.
Uns.Pb-21
(pCi/cm2)
Age
(yrs BP)
(1993)
Sed.
Rate
(g/cm2.yr)
Cs-137
(pCi/g)
0
2
0.113
11.86
2.68
28.29
0.00
0.07
4.41
2
4
0.167
9.48
3.16
23.61
3.20
0.08
3.02
4
6
0.221
9.31
4.11
22.43
7.43
0.08
4.18
6
8
0.233
6.76
3.13
18.34
13.92
0.08
3.63
8
10
0.271
3.72
3.10
13.19
19.97
0.08
2.96
10
12
0.284
6.07
3.43
12.09
27.30
0.06
2.96
12
14
0.229
2.33
1.07
8.64
38.09
0.12
1.32
14
16
0.178
2.63
0.94
7.37
42.32
0.09
1.28
16
18
0.168
2.61
0.88
6.64
46.36
0.08
0.84
18
20
0.160
3.34
1.07
5.76
51.11
0.05
0.75
20
24
0.161
2.29
1.47
4.69
37.70
0.06
0.71
24
28
0.183
1.11
0.82
3.22
69.78
0.09
0.73
28
32
0.208
1.42
1.18
2.40
79.22
0.03
0.37
32
36
0.183
2.00
1.46
1.22
100.93
0.02
0.30
36
40
0.161
1.42
0.91
-0.24
ERR
-0.01
0.25
40
44
0.177
0.12
0.08
-1.13
ERR
-0.30
0.37
44
48
0.269
1.41
1.32
-1.23
ERR
-0.03
0.19
48
32
0.317
0.39
0.81
-2.73
ERR
0.22
0.03
32
36
1.167
0.00
-0.93
-3.36
ERR
ERR
0.04
36
60
1.644
0.00
-2.63
-2.63
ERR
ERR
-0.05


g/crrT2/yr
1892 1914 1923 1935 1942 1946 1951 1955 1966 1973 1979 1986 1990 1993
Time (year)
Figure 4-6. Lake Gibson sedimentation rate (g.cm'2.yr)
versus time (year).
-j
N>


CHAPTER 1
INTRODUCTION
Management of planktonic food webs to improve lake
trophic state is still an experimental procedure, and many
interactions are unknown or poorly understood (Crowder et
al. 1988). There is now enough evidence that top-down
interactions (fish introduction or removal) have a
significant effect on planktonic communities (algal biomass
reduction), particularly in low nutrient lakes (Cooke et al.
1993) .
Chemical treatment, unless it involves an inactivation
of nutrient release (i.e., from the bottom), has proved
unsatisfactory for algal control. Mechanical treatments such
as algal harvesting, artificial circulation, and bottom
sealing have been shown not to be satisfactory due to either
their ephemeral effects or their high cost. Biological
control, even in its earliest stage, seems to be a promising
approach for long term algal control.
Traditionally, limnologists have considered the
interactions in a lake ecosystem as an unidirectional flow
of components consisting of a nutrient-phytoplankton-
zooplankton-fish pathway. Manipulation of food webs is an
area fertile with possibilities for understanding ecosystem
1


Table 4-1. Sediment core physical and chemical characteristics
from two central Florida urban lakes.
PARAMETER
LAKE
GIBSON
LAKE BONNY
Mean
St. Dev.
Range
Mean
St. Dev.
Range
Bulk Density
0.279
0.319
0.094
0.094
0.042
0.021
(g. cm-3)
1.644
0.218
Dry Matter
21.61
13.93
9.47
9.16
3.93
2.24
(%)
76.49
19.81
Organic Matter
38.17
14.85
1.73
56.0
12.69
32.72
(%)
59.56
76.71
Nitrogen
9.33
2.99
1.46
22.32
2.68
18.07
(mg. g"1)
13.46
28.00
Phosphorus
1.86
1.49
0
4.10
1.53
1.51
(mg. g"1)
4.57
6.38


Selenastrum capricornutum
F
I
u
o
r
e
s
c
e
n
c
e
Time (days)
' Shad + Tilapia Control
Figure 3-2. Fluorimetric activity from shad and tilapia
feces in medium with Selenastrum capricornutum.
4*.
o


This dissertation was submitted to the Graduate Faculty
of the College of Engineering and to the Graduate School and
was accepted as partial fulfillment of requirements for
the degree of Doctor of Philosophy.
May 1995 /
yvwinfred M. Phillips
/ Dean,College of Engineering
Karen A. Holbrook
Dean, Graduate School


Ill
Table 5-12. Summary of water quality data for L. Hancock.
Numbers shown are mean values with ranges in parentheses.
Parameter
Unit
Canfield
(1981)
Several
Agencies *
(1989/91)
This study
(1992/93)
Temperature
C
N/A
25.71
(23.1-28.6)
23.65
(23.1-24.2)
1 D'*
mg/L
N/A
7.91
(2.48-10.9)
6.54
(2.48-10.6)
I PH
9.4
(8.8-10.0)
9.06
(7.84-9.66)
9.31
(9.03-9.59)
1 Alkalinity
mg/L CaC03
76
(36-106)
N/A
N/A
1 Conduct.
^/mhos/cm
281
(240-340)
302.8
(232-394)
263
(232-294)
I Nitrogen
mg/L
N/A
7.54
(4.07-12.55)
7.64
(4.07-11.22)
1 nh3-n
mg/L
N/A
0.029
(0.018-0.057)
0.024
(0.018-0.03)
TKN
mg/L
N/A
7.54
(4.06-12.55)
7.64
(4.06-11.22)
I Phosphorus
mg/L
2.44
(1.4-3.2)
0.56
(0.366-0.944)
0.65
(0.37-0.94)
I Chi. a
mg/m3
144
(40.1-217)
184.1
(66-349.5)
221.15
(131.8-310)
B Color
Pt units
28
(25-30)
51.25
(37-75)
60
(45-75)
Secchi
m
0.8
(0.5-1.3)
0.2
(0.1-0.3)
0.25
(0.2-0.3)
(*) AGENCY CODES:
USGS; EPA; PCWR; CL; UFC; FGFWFC; CF208; and SWFWMD.


126
Table 5-21. Summary of water quality data for Lake Hunter.
Numbers shown are mean values with ranges in parentheses.
I Parameter
Unit
Several Agencies*
(1989/91)
This study
(1992/93)
| Temperature
C
25.50 (16.50-31.51)
24.98 (17.87-30.09)
1 D.O.
mg/L
9.60 (4.72-13.90)
10.46 (9.88-11.8)
PH
9.48 (7.88-11.35)
8.74 (8.33-9.45)
| Alkalinity
mg/L CaC03
65.25 (46-99.9)
66.6 (62-71)
Conduct.
pmhos/cm
199.6 (139-253)
202.8 (160-276)
Nitrogen
mg/L
2.24 (0.51-4.92)
2.50 (1.58-3.56)
NH,-N
mg/L
0.0 6 7 (0.01-0.292)
0.0 4 2 (0.032-0.056)
TKN
mg/L
2.14 (0.51-4.81)
2.21 (1.42-3.05)
| Phosphorus
mg/L
0.150 (0.019-0.26)
0.18 (0.158-0.239)
B Chi. a
mg/m3
70.69 (25.6-152.8)
69.75 (43.3-112.3)
H Color
Pt units
28.6 (15-55)
38.66 (25-60)
| Secchi
m
0.38 (0.2-0.6)
0.36 6 (0.2-0.6)
I Bacteria
x 106
1236.3 d-20000)
342.5 (140-600)
] TSI
79.18 (66.36-94.52)
81.58 (71.94-88.88)
(*) AGENCY CODES:
USGS; EPA; PCWR; CL; UFC; FGFWFC; CF208; and SWFWMD.


14
which are important to commercial and sport fishermen
(Lewis, 1953; Jester and Jensen, 1972).
Despite its role as a forage fish, it has been
suggested that in warm-water, shallow lakes with soft mud
bottoms, high turbidity, and relatively few predators, the
gizzard shad may become a nuisance from an ecological and
economic standpoints (Miller, 1960). Kutkuhn (1958) and
Cramer and Marzolf (1970) reported that gizzard shad longer
than 35 mm are primarily herbivorous, with some selection
for zooplankton when sighted, while young gizzard shad
(< 30 mm) feed primarily on zooplankton (Bodola, 1966).
Organic detritus at times can also be an important food for
the gizzard shad (Dalquest and Peters, 1966; Baker and
Schmitz, 1971).
Apparently, owing to the high reproductive capacity and
rapid growth of gizzard shad in shallow lakes and
reservoirs, predators cannot effectively crop young-of-the-
year fish and hence provide a control on population numbers
(Miller, 1960; Bodola, 1966). It has also been suggested
that gizzard shad may inhibit growth of more desirable fish
and/or limit their numbers through interspecific competition
(Berry, 1958; Miller, 1960).
The gizzard shad is a dominant native grazer in Florida
eutrophic lakes. Crisman and Kennedy (1982), based on
mesocosm experiments, showed that gizzard shad did not
impact chlorophyll a values, productivity or phytoplankton


36
as mucilaginous pellets, while shad feces are released into
the environment as a flocculent material.
The bulk of both fish feces consisted of small green
and blue-green algae species, dominating the phytoplankton
in the assemblage at the time the fishes were collected.
Many of the algal cells did not show any signs of digestion,
or they were slightly digested but easily identifiable under
a microscope. It is known that many algal taxa may be viable
after passage through fish digestive tracts (Fish, 1951,
1955; Lowe, 1959; Smith, 1963). Fish (1951, 1955) reported
that, of the algae eaten by tilapia, blue-green and green
algae generally pass through the fish undigested and that
diatoms form the main usable food (Cailteux, 1988). In this
study, I found green algae of the genus Closterium, the
blue-green Microcystis, and the diatoms Melosira,
Fraqilaria, Tabellara. and Cyclotella in tilapia feces.
Shad feces contained the green algal genus Ankistrodesmus,
the diatom Melosira. the dinoflagellate Peridinium, and the
blue-green Microcystis.
Mean values for chlorophyll a (final initial
measurements) in shad feces were [1] for the inoculum made
in the medium deficient of nitrogen(-N) and the blue-green
algae 541.57 mg.m"3 36.85; and 859.18 mg.m"3 17.93 with
green algae; [2] for the inoculum made in the medium
deficient of phosphorus (-P) and the blue-green algae 432.21
mg.m"3 22.02; and 1173.76 mg.m'3
13.96 with green algae.


8
Meijer, 1986; Van Liere, 1986). In recent years, many
studies have shown that food web manipulation
(biomanipulation) can help restore eutrophied lakes
(Andersson et al., 1978; Henrikson et al., 1980; Reinertsen
and Olsen, 1984; Carpenter et al., 1985).
Biomanipulation, as it was originally defined by
Shapiro et al., (1975), refers to management of aquatic
communities by controlling natural populations of organisms
with the end goal being water quality improvement. This
definition was reviewed later by Shapiro (1990), and he
pointed out the considerably broader connotation that the
term has taken on recently relative to that espoused by him
and many others in the field. In the original idea of
Shapiro (1990), biomanipulation is a series of manipulations
of the biota of lakes and their habitats to facilitate
certain interactions and results which we, as lake users,
consider beneficial, namely reduction of algal biomass and,
in particular, of blue-green species.
According to Gophen (1990), in a broad sense,
biomanipulation is equatable to top-down forces, trophic
cascade interactions or food-web manipulation. All these
terms refer to manipulation of secondary or tertiary aquatic
producers and its impact on lacustrine community structure.
Whole-lake food-web manipulation by fish-stock management,
i.e., reduction of planktivorous fish, through enhancement
of piscivorous fish, may accelerate the rate of the


152
NOBLE, R.L., R.D. Germany and C.R. Hall. 1975. Interactions
of blue tilapia and largemouth bass in a power plant
cooling reservoir. Proc. Annu. Conf. SE Game Fish.
Comm. 29: 247-251.
O'NEIL, R.V., D.L. DeAngelis, J.B. Waide and T.F.H. Allen.
1986. A Hierarchical Concept of Ecosystems. Princeton
University Press, Princeton, N.J.
OPUSZYNSKI, K. 1978. The influence of the silver carp
Hypophthalmichthvs molitrix, on eutrophicaton of the
environment of carp ponds. Part VII. Recapitulation.
Roczn. Nauk Roln. ser. H. 99:127-150.
. (undated). Fish manipulation as a tool for
keeping the lakes clean. The theory of
ichthyoeutrophication. Inland Fisheries Institute,
Varsaw, Poland.
, and J.V. Shireman. (unpubl). A new approach to
quantitative evaluation of grazing by filter feeding
fish and to their use for water purification and
aqua/agriculture production enhancement.
, J.V. Shireman, and W. Opuszynski. (unpubl). Feces
collection of filter-feeding fish as a method for
wastewater treatment and enhancement of
aquaculture/agriculture production.
PARSONS, T.R. and J.D. Strickland. 1963. Discussion of
spectrophotometric determination of marine-plant
pigments, with revised equations of ascertaining
chlorophylls and carotenoids. J. Mar. Res. 21:155-163.
PIERCE, R.J. 1977. Life history and ecological energetics of
the gizzard shad (Dqrqsqma cepedianum) in Acton Lake,
Ohio. Ph.D. dissertation, Miami University, Oxford,
Ohio. 201 p.
POLK COUNTY WATER RESOURCES DIVISION. 1992. Lake water
quality report. Polk County, Florida.
POLLMAN, C.D. 1982. Internal loading in shallow lakes. Ph.D.
dissertation, University of Florida, Gainesville, Fl.
POPMA, T.J. 1982. Digestibility of selected feedstuffs and
naturally occurring algae by Tilapia. Ph.D.
dissertation, Auburn University, Auburn, Ala.
POST, J.R. and D.J. McQueen. 1987. The impact of
planktivorous fish on the structure of a plankton
community. Freshwater Biology 17:77-89.


70
60
50
40
30
20
10
0
Percent organic matter
92 1914 1923 1935 1942 1946 1951 1955 1966 1973 1979 1986 1990 1993
Time (year)
Figure 4-4
Lake Gibson percent organic matter sediment core
content versus time (year).
o>
3


There were also significant differences in the
composition of the fecal material produced by each fish
species. Gizzard shad fecal material consisted of 37% more
organic matter and 50% more protein than did blue tilapia
fecal material. Caloric content, measured as gross heat, in
gizzard shad feces was 60% greater than in blue tilapia
fecal material.
Initial chlorophyll a values from the feces measured
for blue tilapia were 10 to 15 times greater than those
measured for gizzard shad feces. Two bioassays were
conducted over a seven day period for each procedure. Final
chlorophyll a values for gizzard shad feces presented a ten
fold increase from the initial chlorophyll a concentration,
regardless of the presence of blue-green or green algae as
the inoculum. Blue tilapia, on the other hand, exhibited a
three- to fivefold increase in samples where a green algal
species was used as the inoculum (there was zero growth in
66% of the samples); and there was no apparent growth in
samples inoculated with a blue-green algae.
The results indicate that blue tilapia fecal material
completely suppressed blue-green algal chlorophyll a
production and appears to have suppressed green algal
chlorophyll a production in more than 60% of the samples.
Gizzard shad feces increased chlorophyll a values in blue-
green and green algal groups (tenfold increase). This study
suggests that in Florida systems with conditions similar to
VXll


ACKNOWLEDGMENTS
I wish to thank Dr. Thomas L. Crisman, chair of my
committee, for his friendship and encouraging words of
advice. I also wish to acknowledge the other members of my
committee, Dr. G. Ronnie Best, Dr. Frank G. Nordlie, Dr.
Edward J. Phlips, and Dr. Horst 0. Schwassmann, whose
critical reviews and comments resulted in the final
manuscript. Dr. Edward J. Phlips provided laboratory support
for both bioassays.
Mr. Eugene Medley provided field and laboratory support
in the City of Lakeland. His assistance, and friendship will
always be appreciated. Mr. Bal C. Sukhraj was an important
asset in my field work in Lakeland. Many thanks go to Dr.
John Moore and staff of the Animal Nutrition Laboratory,
University of Florida, for their contribution to the fish
feces analyses. Mr. John Funk assisted with the fish feces
analyses. Dr. Daniel Canfield generously offered the use of
an electrofishing boat. Dr. Francisco Zimmermann helped with
the statistical analyses.
Throughout this work, I have benefitted from a
productive working relationship with staff members from the
Florida Game and Fresh Water Fish Commission in Eustis,
n


54
Florida has about 7800 lakes varying in size from 0.4
ha to over 180,000 ha. (Canfield and Hoyer, 1988). Lakes in
Florida serve many agricultural, domestic, industrial and
recreational purposes (Canfield and Hoyer, 1992). Outside
the glaciated regions in North America, Florida lakes
constitute the largest group of natural lakes (Hutchinson,
1957) on the continent and the most important group of
solution basins (Hutchinson, 1957; Crisman, 1992)
Limnological characteristics and productivity from
these lakes vary widely and range from oligotrophic to
hypereutrophic (Canfield and Hoyer, 1988; Brenner et al.
1990; Beaver et al. 1981). Although Florida's aquatic
systems are large in number and have a significant economic
impact, water quality data have been collected for few lakes
(>10%) (Brenner et al., 1990), and routine data acquisition
began fairly recently, in the 1960s and 1970s (Huber et al.,
1982).
Considering Florida's aquatic system dimensions and
economic importance, little information has been reported on
the sediments of Florida lakes, i.e. Flannery et al. (1982);
Brenner and Binford (1988); Brenner et al. (1990); Stoermer
et al. (1992); Gottgens (1992); Brenner et al. (1993). Here
I provide information on the relationship between
accumulation rates of sediment variables such as organic
matter, water content, total nitrogen and total phosphorus


Time (year)
+- % Dry matter % Organic matter
Nitrogen (mg/g) Phosphorus (mg/g)
Figure 4-8. Changes in some physicochemical characteristics in
Lake Gibson through time.


35
in tilapia is longer) or the time of the year (i.e.,
spawning time). In gizzard shad, the length of the
intestinal tract increases with increasing standard length.
Schmitz and Baker (1969) found that the intestine-to-body-
length ratio for gizzard shad was about 2.8:1, while in
threadfin shad the ratio averaged 1.8:1. Although the length
of the intestine in a fish may be partially determined by
its diet, Schmitz and Baker (1969) believe that the length
and form of the intestine are probably the result of a
complex of intrinsic and extrinsic factors that can not be
explained only by diet.
Tilapia produces five times more feces than bighead
carp, whereas the growth rate of both species was comparable
(Opuszynski and Shireman, unpubl.). In this study, I found
that the average feces produced by the shad was 2.08 0.34g
fresh feces per lOOg fish body weight and 4.5 0.21g fresh
feces per lOOg fish body weight for the tilapia. The fact
that blue tilapia feces' bulk density is twice as high as
shad feces (Table 3-1), indicates that tilapia produces four
times more feces as dry matter than shad. Therefore, blue
tilapia seems to be more suitable than bighead carp
(Opuszynski and Shireman, unpubl.) and shad for use in
biological schemes for water guality improvement.
There are obvious macroscopic differences in the feces
produced by these two fish taxa. Tilapia feces are released


76
Table 4-6. Total phosphorus (mg.g-1) for
Lake Gibson sediment core.
Interval (cm)
TP (mg/g) |
0-2
4.4495
2-4
4.5700
4-6
4.1031
6-8
3.7696
8-10
3.2824
10-12
3.0312
12-14
2.0766
| 14-16
1.6263
I 16-18
1.6984
18-20
1.5514
20-24
1.5047
24-28
1.2811
28-32
0.7041
32-36
0.8948
36-40
0.9506
40-44
0.9809
44-48
0.6574
48-52
0.1171
52-56
0.0097
56-60
0.0000
Mean 1.8


157
APPENDIX A
Map of the study area


50
(Figures 3-3 and 3-4), while in the Selenastrum tests, about
66% chlorophyll a had negative values indicating
suppression, with the remaining values being positive with a
mean of 600 mg.nf3 (Figures 3-5 and 3-6).
McDonald (1985) found that blue tilapia enhanced the
growth of Ankistrodesmus cells. Ware et al.(1975) concluded
that blue tilapia had displayed little potential for algal
control in Florida waters, and in fact, had spread rapidly
and become a nuisance. Phytoplankton biomass in the fish
fecal material calculated from chlorophyll a values at the
beginning and end of the current bioassay showed that blue
tilapia suppressed phytoplankton biomass in 75% of the
samples of blue green and green algae to a level exceeding
50% of the initial value. Gizzard shad increased
phytoplankton biomass concentration in all samples of blue
green and green algae more than tenfold from the initial
values.
It has been reported (Crisman and Kennedy, 1982) that
gizzard shad are not suitable for use as a biocontrol agent
for phytoplankton because they have no impact on chlorophyll
a values, productivity or phytoplankton densities, and they
can promote lake eutrophication through elevation of
orthophosphate concentrations and differential digestion of
some algae groups, especially blue-greens and greens.
On the other hand, results of extensive work with
tilapia are still inconclusive regarding the possible


158
APPENDIX B
Lake Bonny sedimentation rate (g.cm"2.yr)
by year.
LAKE BONNY
Core
Sect
(cm)
Bulk
Density
(g/cm3)
Unsupp.
Pb-210
(pCi/g)
Unsupp.
Pb-210
(pCi/c m2
Cum.Res.
Uns.Pb-21
(pCi/cm2)
Age
(yrs BP)
(1993)
Sed.
Rate
(g/cm2.yr)
Cs-137
(PCi/g)
0
4
0.029
18.72
2.17
31.84
0.00
0.03
1.3
4
7
0.030
12.30
1.84
29.67
2.27
0.08
3
7
9
0.061
12.30
1.33
27.83
4.32
0.07
4.13
9
11
0.066
11.22
1.48
26.31
6.13
0.07
4.01
11
13
0.074
8.46
1.23
24.83
7.99
0.09
3.63
13
13
0.078
8.14
1.27
23.37
9.65
0.09
3.20
13
17
0.079
3.27
0.83
22.30
11.43
0.13
4.02
17
19
0.090
6.99
1.26
21.47
12.63
0.10
4.36
19
22
0.092
8.92
2.46
20.21
14.39
0.07
4.49
22
26
0.092
6.31
2.39
17.75
18.76
0.08
3.28
26
30
0.090
7.00
2.32
13.36
23.40
0.07
2.67
30
34
0.089
6.00
2.14
12.84
29.16
0.07
1.07
34
38
0.091
3.80
1.38
10.70
33.01
0.09
0.83
38
42
0.102
3.43
1.40
9.32
39.44
0.08
0.92
42
46
0.164
3.48
2.28
7.92
44.67
0.07
0.87
46
30
0.174
3.17
2.21
3.64
33.56
0.06
0.78
30
34
0.142
1.91
1.08
3.43
71.32
0.06
0.91
34
38
0.137
1.94
1.06
2.33
83.64
0.04
0.39
38
64
0.124
0.00
-0.10
1.29
102.86
ERR
0.14
64
72
0.123
0.13
1.39
100.56
ERR
0.33
72
84
0.126
1.24
1.24
104.23
ERR
0.37


156
WHITE, W.H. 1970. The geomorphology of the Florida
peninsula. Florida Department of Natural Resources
Geological Bulletin n. 51.
WILLIAMS, V.P., D.E. Canfield, Jr., M.M. Hale, W.E. Johnson,
R.S. Kautz, J.T. Krummrich, F.H. Langford, K. Langland,
S.P. McKinney, D.M. Powell, and P.L. Shafland. 1985.
Lake habitat and fishery resources of Florida. Pages
54-69 in Seaman, Jr., (ed.), Florida Aquatic Habitat
and Fishery Resources. American Fisheries Society, New
York, 543 pp.
WYNGAARD, G.A., J.L. Elmore and B.C. Cowell. 1982. Dynamics
of a subtropical plankton community, with emphasis on
the copepod Mesocvclops edax. Hydrobiologia 89:39-48.
YENTSCH, C.S. and D.W.Menzel. 1963. A method for the
determination of phytoplankton chlorophyll and
phaeophytin by fluorescence. Deep Sea Res. 10:221-231.
ZALE, A.V. 1984. Applied aspects of the thermal biology,
ecology and life history of the blue tilapia, Tilapia
aurea (Pisces: Cichlidae). Technical Report no. 12,
Florida Cooperative Fish & Wildlife Research Unit,
Gainesvile, FL. 196 pp.


some agricultural activities, and recreational uses such as
boating, fishing, and skiing.
Cluster analysis performed grouping all six lakes
revealed that, with the exception of Lake Parker, all lakes
could be aggregated by common limnological features, and
that the relationships were independent of year, season and
the combination of year and season.
Mean primary productivity for the studied lakes was
334.22 mgC.nf3.h + 106.35. Primary productivity in Lake
Parker (mean= 66.41 mgC.m'3.h) was five times less than the
mean for the other five lakes.
Sedimentary records reveal that the lakes in this study
have been eutrophic for at least the last century with a
significant accumulation of organic material in the
superficial mud. Superficial sediment content was always
40% of the organic matter for any lake. Lake Parker was the
only lake investigated that displayed a progressive decline
in water column total phosphorus in the last fifteen years,
perhaps signaling some reversal of cultural enrichment.
There was a significant difference in the amount of
feces hand-stripped from the fishes studied. Blue tilapia
had a greater amount of fecal material than did gizzard
shad. The bulk of the fecal material collected from both
fish species consisted of small green and blue-green algae,
which were the dominant algal taxa in the environment at the
time the fishes were collected.
Vll


mg/g
Time (year)
Figure 4-5. Lake Gibson TN (mg.g-1) and TP (mg.g~1) sediment core
content versus time (year).


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EN5HY4IUE_3HLKWZ INGEST_TIME 2015-03-31T18:49:43Z PACKAGE AA00029815_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


REPLACEMENT OF GIZZARD SHAD (Dorosoma cepedianum) BY BLUE
TILAPIA (Tilapia aurea) AS A POTENTIAL BIOMANIPULATION AGENT
IN FLORIDA EUTROPHIC LAKES
By
CARLOS A. FERNANDES
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1995


and the Polk County Water Resources Division. Their field
and laboratory support was invaluable.
I particularly wish to thank ray parents, Mr. A.P.
Fernandes (deceased) and Mrs. M.D. Fernandes, for teaching
me how to walk through life, maintaining the necessary
endurance without loosing sensitivity and appreciation for
nature's creations. My sons, Carlos Filho and Diego, deserve
special recognition for their positive attitude, and their
confidence throughout ray work. They also provided loving
field support. Special recognition goes to Margaret, my
fiancee, friend and fellow biologist, for her important
field and laboratory support during the bioassay. Most of
all, their love and caring support gave the perseverance
needed to complete this work.
This study was funded by CAPES--Fundago Coordenago de
Aperfeigoamento de Pessoal de Nivel Superior of the Ministry
of Education of Brazil.
in


15
densities, and that shad caused a significant increase in
both the concentration of orthophosphate and its ratio to
total phosphorus under natural stocking densities. The fish
was also responsible for a significant decrease in copepod
density during that study.
Crisman and Beaver (1988) and Threlked and Drenner
(1987) noted that the overall effect of shad grazing was to
stimulate the phytoplankton community and to decrease Secchi
disk transparency. Shad do not effectively graze more
evasive zooplankton such as Diaptomus (Drenner et al.,
1978), particularly during the summer months, although other
evasive herbivores may be simultaneously depressed (Drenner
et al., 1982a). Many common algal taxa remain viable after
gut passage through the digestive tract of shad, especially
blue-greens (Velasquez, 1939; Smith, 1963; Crisman and
Kennedy, 1982).
It has been assumed that the enhancement of
phytoplankton populations by shad grazing is an indirect
effect caused by suppression of herbivorous zooplankton
(Threlkeld, 1987). Shad may also enhance phytoplankton
populations directly by providing a larger quantity of
highly assimilable nutrients from their digestive products
(Crisman and Kennedy, 1982).
Gizzard shad and other clupeids are very sensitive fish
that can be easily stressed and killed. In systems where
gizzard shad is considered a nuisance, as in all Florida


122
The Lake Bonny zooplankton community was dominated by
rotifers and copepods with 274,000 and 85,200
individuals .m"3, respectively (Canfield and Hoyer, 1992).
The dominant rotifers were Brachionus sp. and Keratella sp.
Diaptomus was the most abundant copepod.
Sixteen species of fish were collected in Lake Bonny
(Canfield and Hoyer, 1992). Gizzard shad was the most
abundant fish collected in open-water using experimental
gillnets, followed by Florida gar, with 17.0 and 11.3
fish.net'1.24hr, respectively (Canfield and Hoyer, 1992 ).
ANOVAs were performed using the General Linear Model
(GLM) (SAS, 1989) in order to identify significant seasonal
limnological factors (p < 0.05). The following parameters
were analyzed: Secchi, temperature, pH, conductivity,
dissolved oxygen, total suspended solids, turbidity,
chlorophyll a, color, alkalinity, NH3, TKN, total nitrogen,
total phosphorus, orthophosphate, and total bacteria.
Temperature, conductivity, dissolved oxygen, turbidity,
chlorophyll a, color, TKN, total nitrogen and total
phosphorus exhibited significant seasonal variation at the
95% confidence level (Table 5-19). Interannual variation or
no variation was reported for the other parameters. When
year and season were combined, Secchi, pH, conductivity,
TSS, turbidity, chlorophyll a, color, and total phosphorus
displayed significant variability.


73
Bonny had 56.0%. Total nitrogen for the same array of lakes
displayed high variability, ranging from 0.6 to 42.4 mg.g"1
(Brenner and Binford, 1988). Nitrogen concentration for
Lake Gibson was 9.33 mg.g"1 (Table 4-4), and for Lake Bonny
was 22.32 mg.g"1 (Table 4-5). Total P concentrations
recorded by Brenner and Binford (1988) varied from 0.07 to
8.09 mg.g"1. Values for Lake Gibson were 1.8 mg.g'1 (Table
4-6) and for Lake Bonny was 4.1 mg.g'1 (Table 4-7).
Considering that these lakes overlie phosphatic
limestone deposits and are located in urban settings, each
one of those isolated factors or a combination of both could
be responsible for the high total P values. The total P
value registered for Lake Bonny of 4.1 mg.g"1 is higher than
what was found in 93 of the 97 systems studied for Brenner
and Binford (1988).
The Lake Parker Record
The Lake Parker sediment core displayed usually
elevated deposition rates for bulk density, organic matter
and phosphorus (Brenner et al., 1993). Bulk sediment
accumulation rates increased from 11 mg.cm"2.yr"1 in 1922 to
a maximum of 131 mg.cm"2.yr"1 in the late 1950s (Brenner et
al., 1993). Phosphorus net accumulation rates increased
more than ten times during the last century, reaching the
highest value of 600 pg.cm"2.yr"1 in the late 1970s (Brenner
et al., 1993). Appendices D and E show physical and chemical
properties of Lake Parker sediment core.


52
that, in Florida systems with conditions similar to those of
the lakes studied, biomanipulation techniques can mitigate
cultural eutrophication if gizzard shad are replaced by blue
tilapia.


REPLACEMENT OF GIZZARD SHAD (Dorosoma cepedianum) BY BLUE
TILAPIA (Tilapia aurea) AS A POTENTIAL BIOMANIPULATION AGENT
IN FLORIDA EUTROPHIC LAKES
By
CARLOS A. FERNANDES
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1995

ACKNOWLEDGMENTS
I wish to thank Dr. Thomas L. Crisman, chair of my
committee, for his friendship and encouraging words of
advice. I also wish to acknowledge the other members of my
committee, Dr. G. Ronnie Best, Dr. Frank G. Nordlie, Dr.
Edward J. Phlips, and Dr. Horst 0. Schwassmann, whose
critical reviews and comments resulted in the final
manuscript. Dr. Edward J. Phlips provided laboratory support
for both bioassays.
Mr. Eugene Medley provided field and laboratory support
in the City of Lakeland. His assistance, and friendship will
always be appreciated. Mr. Bal C. Sukhraj was an important
asset in my field work in Lakeland. Many thanks go to Dr.
John Moore and staff of the Animal Nutrition Laboratory,
University of Florida, for their contribution to the fish
feces analyses. Mr. John Funk assisted with the fish feces
analyses. Dr. Daniel Canfield generously offered the use of
an electrofishing boat. Dr. Francisco Zimmermann helped with
the statistical analyses.
Throughout this work, I have benefitted from a
productive working relationship with staff members from the
Florida Game and Fresh Water Fish Commission in Eustis,
n

and the Polk County Water Resources Division. Their field
and laboratory support was invaluable.
I particularly wish to thank ray parents, Mr. A.P.
Fernandes (deceased) and Mrs. M.D. Fernandes, for teaching
me how to walk through life, maintaining the necessary
endurance without loosing sensitivity and appreciation for
nature's creations. My sons, Carlos Filho and Diego, deserve
special recognition for their positive attitude, and their
confidence throughout ray work. They also provided loving
field support. Special recognition goes to Margaret, my
fiancee, friend and fellow biologist, for her important
field and laboratory support during the bioassay. Most of
all, their love and caring support gave the perseverance
needed to complete this work.
This study was funded by CAPES--Fundago Coordenago de
Aperfeigoamento de Pessoal de Nivel Superior of the Ministry
of Education of Brazil.
in

TABLE OF CONTENTS
ACKNOWLEDGEMENTS ii
ABSTRACT vi
CHAPTER 1 INTRODUCTION 1
CHAPTER 2 LITERATURE REVIEW 7
Introduction 7
Biomanipulation 7
Shad Ecological Energetics 13
Tilapia Ecological Energetics 16
CHAPTER 3 RELATIVE IMPORTANCE OF BLUE TILAPIA AND
GIZZARD SHAD TO LAKE SEDIMENT AND WATER COLUMN
NUTRIENT CONCENTRATION--ANALYSIS OF FISH FECES 22
Introduction 22
Methods 27
Results and Discussion 31
Conclusions 39
CHAPTER 4 SEDIMENTS AND HISTORICAL ECOLOGY OF TWO
CENTRAL FLORIDA LAKES 53
Introduction 53
Study Sites 56
Methods 59
Results and Discussion 60
Conclusions 78
CHAPTER 5 LIMNOLOGICAL ASSESSMENT 89
Introduction 89
Methods 91
Description of Study Area 93
Summary 129
IV

CHAPTER 6 SUMMARY
133
Fish Feces Experiment 134
Paleolimnological Analysis 137
Lakes Assessment 139
Conclusion 140
LITERATURE CITED 142
APPENDICES 157
BIOGRAPHICAL SKETCH 165
v

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
REPLACEMENT OF GIZZARD SHAD (Dorosoma cepedianum) BY BLUE
TILAPIA (Tilapia aurea) AS A POTENTIAL BIOMANIPULATION AGENT
IN FLORIDA EUTROPHIC LAKES
By
Carlos A. Fernandes
May 1995
Chairperson: Dr. Thomas L. Crisman
Major Department: Environmental Engineering Sciences
The effects of replacement of gizzard shad (Dorosoma
cepedianum) as a dominant filter feeding fish by blue tilapia
(Tilapia aurea) in six eutrophic Florida lakes were analyzed
using the lakes' historical limnological characteristics and
sedimentary records. The digestive physiology of gizzard shad
and blue tilapia (with emphasis on fecal material) was studied
in order to evaluate the relative impact of the feces produced
by different fish species.
Five of the six lakes studied are classified as
hypereutrophic systems. Lake Gibson is classified as
eutrophic. All of the study sites are urban lakes subject to
anthropogenic pressures in the form of industrial,
residential and commercial development in the watershed,
vi

some agricultural activities, and recreational uses such as
boating, fishing, and skiing.
Cluster analysis performed grouping all six lakes
revealed that, with the exception of Lake Parker, all lakes
could be aggregated by common limnological features, and
that the relationships were independent of year, season and
the combination of year and season.
Mean primary productivity for the studied lakes was
334.22 mgC.nf3.h + 106.35. Primary productivity in Lake
Parker (mean= 66.41 mgC.m'3.h) was five times less than the
mean for the other five lakes.
Sedimentary records reveal that the lakes in this study
have been eutrophic for at least the last century with a
significant accumulation of organic material in the
superficial mud. Superficial sediment content was always
40% of the organic matter for any lake. Lake Parker was the
only lake investigated that displayed a progressive decline
in water column total phosphorus in the last fifteen years,
perhaps signaling some reversal of cultural enrichment.
There was a significant difference in the amount of
feces hand-stripped from the fishes studied. Blue tilapia
had a greater amount of fecal material than did gizzard
shad. The bulk of the fecal material collected from both
fish species consisted of small green and blue-green algae,
which were the dominant algal taxa in the environment at the
time the fishes were collected.
Vll

There were also significant differences in the
composition of the fecal material produced by each fish
species. Gizzard shad fecal material consisted of 37% more
organic matter and 50% more protein than did blue tilapia
fecal material. Caloric content, measured as gross heat, in
gizzard shad feces was 60% greater than in blue tilapia
fecal material.
Initial chlorophyll a values from the feces measured
for blue tilapia were 10 to 15 times greater than those
measured for gizzard shad feces. Two bioassays were
conducted over a seven day period for each procedure. Final
chlorophyll a values for gizzard shad feces presented a ten
fold increase from the initial chlorophyll a concentration,
regardless of the presence of blue-green or green algae as
the inoculum. Blue tilapia, on the other hand, exhibited a
three- to fivefold increase in samples where a green algal
species was used as the inoculum (there was zero growth in
66% of the samples); and there was no apparent growth in
samples inoculated with a blue-green algae.
The results indicate that blue tilapia fecal material
completely suppressed blue-green algal chlorophyll a
production and appears to have suppressed green algal
chlorophyll a production in more than 60% of the samples.
Gizzard shad feces increased chlorophyll a values in blue-
green and green algal groups (tenfold increase). This study
suggests that in Florida systems with conditions similar to
VXll

those of the lakes studied, biomanipulation techniques can
mitigate cultural eutrophication to improve water clarity,
if gizzard shad are replaced by blue tilapia.
IX

CHAPTER 1
INTRODUCTION
Management of planktonic food webs to improve lake
trophic state is still an experimental procedure, and many
interactions are unknown or poorly understood (Crowder et
al. 1988). There is now enough evidence that top-down
interactions (fish introduction or removal) have a
significant effect on planktonic communities (algal biomass
reduction), particularly in low nutrient lakes (Cooke et al.
1993) .
Chemical treatment, unless it involves an inactivation
of nutrient release (i.e., from the bottom), has proved
unsatisfactory for algal control. Mechanical treatments such
as algal harvesting, artificial circulation, and bottom
sealing have been shown not to be satisfactory due to either
their ephemeral effects or their high cost. Biological
control, even in its earliest stage, seems to be a promising
approach for long term algal control.
Traditionally, limnologists have considered the
interactions in a lake ecosystem as an unidirectional flow
of components consisting of a nutrient-phytoplankton-
zooplankton-fish pathway. Manipulation of food webs is an
area fertile with possibilities for understanding ecosystem
1

2
function and for developing techniques for lake management
that are not dependent on chemical and mechanical means.
All research done in the field of biomanipulation
supports the conclusion that the presence of large-bodied
zooplankton, Daphnia spp., is required for a strong control
over phytoplankton populations. Unfortunately, not all
geographic areas meet this requirement. Tropical and
subtropical areas, namely Florida and other areas of the
southeastern U.S., do not have large species of zooplankton.
Crisman and Beaver (1990) described only the presence of
small Cladocera (e.g., Eubosmina spp., Ceriodaphnia spp.) in
Florida lakes which are not good candidates for top-down
biomanipulation purposes.
Lacking the presence of the most studied and accepted
candidate for the top-down type of biomanipulation, tropical
and subtropical researchers have had to seek another
organism which could perform a corresponding role. Filter
feeding fish, which are very common in those areas, seem to
be the most suitable candidates. Filter-feeders do not
visually detect individual prey items, but engulf a volume
of water containing the food organisms and retain the
planktonic prey and particles by passing this volume over
entrapment structures (Lazzaro, 1987).
The most common filter-feeding fishes in Florida
eutrophic lakes are gizzard shad (Dqrqsqma cepedianum),
which can produce a total grazing pressure on algal

3
populations even greater than the large bodied zooplankton
in temperate systems (cf. Drenner et al., 1982a). Another
very common filter-feeder fish inhabiting Florida eutrophic
lakes is blue tilapia (Tilapia aurea) which was introduced
into the United States in 1961. Both of these fishes feed on
small, particulate material. For example, gizzard shad 5 cm
in length can filter algae >19ju in size; those 15 cm in
length can filter algae >40jj; and those 25 cm in length can
filter algae >63^. Blue tilapia can consume algae >25p in
size (Opuszinsky and Shireman, unpubl.).
The filtering rates of these species can be very high.
According to Drenner et al. (1982a), the gizzard shad
population in Lake Barkley, Texas (85 ha), can effectively
filter a volume equivalent to the entire lake every 2.3
days.
Undoubtedly, filter-feeding fishes can alter the
phytoplankton community of aquatic systems. Nevertheless,
contrary to expectations, an increase in planktonic algae
biomass and primary production has been observed (Drenner et
al., 1984, 1986; Janusko, 1974,1978; Opuszynski, 1978). Some
of the reasons for the failure of filter-feeding fish to
control algal blooms and in some cases even worsen water
quality are [1] elimination of larger zooplankton species;
[2] more rapid cycling of plant nutrients (Opuszynski and
Shireman, unpubl); and [3] differential release of fecal
material by individual species of fish.

4
Digestion in animals has been an almost totally-
neglected subject. Fish studies have been limited to gut
content and selectivity (Halver, 1989). Very little is known
of the digestive capabilities of fishes.
Fish, like all other animals, reguire energy to sustain
life, and they are among the most efficient animals in
converting food to body tissue (Halver, 1989). However, a
rough generalization can be made that about one-third of the
energy in the food offered to fish will be lost as
combustible waste (Halver, 1989). This will consist of
uneaten food, feces, urine, and gill excretions. From the
lost combustible wastes, feces (25% of the total) constitute
the majority of this part of the energy balance.
Fecal material produced from different fishes and
released into the environment can produce a differential
impact and conseguently a differential response. Blue
tilapia feces, being encased in mucilage, are more coherent
and will be broken down slowly, whereas shad feces are
released practically as aqueous substance permitting an
almost instantaneous availability of nutrients for the
phytoplankton community.
Because little is known of their feeding ecology,
especially guantitative feeding under natural conditions,
attempts to predict and efficiently manage these filter
feeding fish as a tool to alleviate eutrophication processes
still is in its experimental stage. Some researchers (i.e.

5
Opuszynsky and Shireman, unpubl.) have developed a new
approach to improve the use of filter-feeding fish to
counteract eutrophication. They constructed an apparatus
consisting of a cage where fish are kept that was equipped
with funnels under the cage to collect fish feces for
estimating food consumption via a quantification of the
production of feces. According to these researchers, this
cage combines two indispensable features for an effective
use of filter-feeding fish to control planktonic algal
growth and to reduce eutrophication: [1] it enables the
coexistence of filter-feeding fish and zooplankton by
eliminating the fish from the water column thus improving
phytoplankton consumption; and [2] fish feces can easily be
collected and removed from the bottom of the cage,
eliminating a major source of nutrient for the phytoplankton
community.
There is a high degree of trophic overlap between young
blue tilapia and larval gizzard shad in Lake George, Florida
(Zale, 1984). Beaver and Crisman (1989) reported
phytoplankton and zooplankton community alterations in
eutrophic Florida lakes when blue tilapia has an established
population. In eutrophic central Florida lakes, blue tilapia
is quickly replacing gizzard shad as a major filter-feeding
fish (Beaver and Crisman, 1989).
The current study addresses gaps on current knowledge
regarding fish as biomanipulation agents for phytoplankton.

6
I worked with the digestive physiology of blue tilapia and
gizzard shad with an emphasis on their fecal material, as
well as with the limnological characteristics and
paleoliranological interpretations from six central Florida
lakes. The research addressed the following hypotheses: [1]
There is differential nutrient bioavailability from blue
tilapia and gizzard shad feces; [2] different type of feces
will produce a different effect on the lake ecosystem; [3]
fundamental differences in the physiology and biochemistry
of blue tilapia and gizzard shad feces will impact the
sediment composition in lakes; and [4] interlake differences
should exist in key limnological parameters between lakes of
differing relative abundance of blue tilapia and gizzard
shad. The goals of this study are to [1] learn something
about the fecal material composition of filter-feeding fish
to help understand their role in freshwater systems; [2]
evaluate management techniques that couple physicochemical
water information with biotic community components and the
lake sedimentary record; and [3] gather new information on
the effect of different fish species' fecal material on lake
ecosystems.

CHAPTER 2
LITERATURE REVIEW
Introduction
This chapter reviews the literature on biomanipulation
and shad and tilapia ecological energetics by examining
planktivorous fish from the community standpoint with
special emphasis on fecal material composition and fate in
the system and the significance of different fish species as
trophic linkages in freshwater food webs.
Biomanipulation
The two main strategies to control eutrophication of
fresh waters are 1) reduction of external and internal
loading of nutrients (Bjork, 1985) and 2) control of
internal ecological processes. With respect to toxic
substances, organic wastes and acid precipitation, the first
strategy alone will provide acceptable solutions to the
problems on a long-term scale (Benndorf, 1988). However, a
combination of strategies 1 and 2 could lead to an
improvement in water guality and to a lower cost/benefit
ratio in the management of the water resource.
Until recently, eutrophication problems were tackled
primarily by reducing external nutrient loading (Hosper and
7

8
Meijer, 1986; Van Liere, 1986). In recent years, many
studies have shown that food web manipulation
(biomanipulation) can help restore eutrophied lakes
(Andersson et al., 1978; Henrikson et al., 1980; Reinertsen
and Olsen, 1984; Carpenter et al., 1985).
Biomanipulation, as it was originally defined by
Shapiro et al., (1975), refers to management of aquatic
communities by controlling natural populations of organisms
with the end goal being water quality improvement. This
definition was reviewed later by Shapiro (1990), and he
pointed out the considerably broader connotation that the
term has taken on recently relative to that espoused by him
and many others in the field. In the original idea of
Shapiro (1990), biomanipulation is a series of manipulations
of the biota of lakes and their habitats to facilitate
certain interactions and results which we, as lake users,
consider beneficial, namely reduction of algal biomass and,
in particular, of blue-green species.
According to Gophen (1990), in a broad sense,
biomanipulation is equatable to top-down forces, trophic
cascade interactions or food-web manipulation. All these
terms refer to manipulation of secondary or tertiary aquatic
producers and its impact on lacustrine community structure.
Whole-lake food-web manipulation by fish-stock management,
i.e., reduction of planktivorous fish, through enhancement
of piscivorous fish, may accelerate the rate of the

9
restoration process (e.g. Shapiro and Wright, 1984;
Edmondson and Abella, 1988).
For studying the effect of food-web manipulations on
lake restoration and to examine the fundamental mechanisms
underlying ecosystem regulation, application of whole-lake
manipulation-experiments can have major advantages (van Donk
et al., 1990). These experiments simulate or actually
encompass the conditions that would be expected to occur
naturally in lakes (Carpenter and Kitchell, 1988). Problems
of enclosure-size and omitted members of communities are not
relevant to whole-lake manipulations (Frost et al., 1988).
It is, however, difficult to perform these manipulations on
a large scale and to interpret their results (Hulbert,
1984). Interpretations may be greatly eased by results of
small-scale manipulations on similar systems (Frost et al.,
1988) .
Nevertheless, to obtain a good overview of the
community's processes, investigations must be conducted
simultaneously as small-scale in situ or laboratory
experiments that can be replicated under controlled
conditions. In this approach, whole-lake manipulations can
be considered as generating as well as testing hypotheses
(O'Neill et al., 1986) .
Examination of trophic-level interactions has long been
an integral part of limnology (Hrbacek et al., 1961;
Nauwerck, 1963; Brooks and Dodson, 1965). However, Shapiro

10
et al. (1975) and Shapiro (1978) recognized that
eutrophication problems are biological manifestations of
nutrient availability, and were the first to suggest that
manipulation of trophic interactions (biomanipulation) could
be used as a lake management tool to alleviate the
biological consequences of eutrophication without the need
for costly controls on nutrient loading (Crisman and Beaver,
1990).
Since biomanipulation research began in temperate
regions, the emphasis has focused basically on alterations
of the zooplankton community (enhancement of large
cladocerans especially of the genus Daphnia), because,
through these manipulations, phytoplankton has been reduced
(blue-green algae) and water transparency has increased
despite difficulties in maintaining such a condition for a
long time. Only recently has research been conducted using
fishes as a major element in the biomanipulation process,
but still only under temperate conditions.
In subtropical and tropical regions, conditions are
quite different. One particular element must be stressed;
there is a complete absence of large zooplankton in tropical
regions. This and numerous other physicochemical and
biological differences between these climatic regions
impedes the applicability of biomanipulation research from
the temperate regions to warmer climates (Crisman and
Beaver, 1990). Thus, the natural choice for biomanipulation

11
schemes of subtropical and tropical regions is the use of
planktivorous fish. Despite the potential for utilization of
filter-feeding fish in biomanipulation techniques to
maintain algal control, especially in shallow eutrophic
lakes and reservoirs characterizing subtropical and tropical
regions, little information exists regarding the direct
impact on the general planktonic community.
Florida lakes possess characteristics which
fundamentally differentiate them from most temperate
systems. The relative shallowness of Florida lakes, combined
with a moderate climatic regime, dictates that stable
thermal stratification usually only occurs in deeper (>5-6
m) lakes of the region (Beaver et al., 1981). Such lakes are
warm monomictic and circulate throughout the winter months,
while shallower basins are more strongly affected by local
meteorological events and are subjected to higher internal
nutrient loading via sediment resuspension (Pollman, 1982).
Subtropical lakes of Florida are inherently different
from temperate lakes in a number of respects that could
affect the success of whole lake biomanipulation. The size
structure of subtropical zooplankton communities is skewed
toward smaller individuals, and large-bodied cladocerans are
absent (Bays and Crisman, 1983). Despite the lack of large
bodied crustaceans, intense grazing activities by gizzard
shad may strongly determine plankton structural and
functional characteristics (Bays and Crisman, 1983). These

12
subtropical systems maintain primary production all year
while production in temperate systems is greatly depressed
or absent during winter. Thus, if non-vegetative seasonal
estimates of production are included in determining annual
production, it is likely that subtropical systems would
realize greater yearly production.
From the few papers about the use and application of
biomanipulation techniques in the subtropical zone, Crisman
and Beaver (1990) noted that the increase in cladoceran
abundance in Florida lakes following elimination of fish
predation agrees with observations in temperate lakes
(Lynch, 1979; Carpenter et al., 1987; Van Donk et al.,
1989), but unlike the latter lakes, community species
composition was not altered nor was there a marked increase
in crustacean mean body size (Shapiro and Wright, 1984;
Benndorf et al., 1988). Large-bodied daphnids, the focus of
all temperate studies, are absent in Florida regardless of
trophic state or predation intensity (Crisman and Beaver,
1990). The results of Crisman and Beaver (1988) from
research conducted in Lake Apopka, Florida, show a
fundamental disagreement with the biomanipulation of Round
Lake, Minnesota (Shapiro and Wright, 1984), and suggest that
zooplankton size, structure and standing crop have only
minimal influence on phytoplankton biomass in Florida lakes.
Regarding fish composition, Crisman and Beaver (1990)
noted that unlike the eutrophic temperate lakes

13
characterized by size-selective planktivorous fishes, in
subtropical systems this composition shifts to pump-filter
feeding fishes, reflecting faunal dominance by gizzard shad
(Dorpsoma cepedianum). Crisman and Beaver (1990) noted that
in both temperate and subtropical systems, removal of
planktivorous fish results in higher macrozooplankton
populations.
Unlike temperate systems, however, algal biomass was
not reduced in the presence of enhanced macrozooplankton
abundance, but actually increased. Crisman and Beaver (1990)
suggested that small-bodied macrozooplankton, even if freed
from fish predation, are of questionable value as
biomanipulation tools in eutrophic subtropical lakes. These
authors also stated that if biomanipulation were to be
successful in the subtropics, emphasis should be shifted
from zooplankton to the role played by planktivorous fish
(pump-filter feeding).
Shad Ecological Energetics
The gizzard shad (Dorosoma cepedianum) occurs commonly
in many lakes and streams in North America (Scott and
Crossman, 1973), where it is the principal phytophagous fish
(Crisman and Kennedy, 1982). This species is predominately
found in eutrophic lakes, being reportedly a key link in the
food chain between primary producers and the top carnivores

14
which are important to commercial and sport fishermen
(Lewis, 1953; Jester and Jensen, 1972).
Despite its role as a forage fish, it has been
suggested that in warm-water, shallow lakes with soft mud
bottoms, high turbidity, and relatively few predators, the
gizzard shad may become a nuisance from an ecological and
economic standpoints (Miller, 1960). Kutkuhn (1958) and
Cramer and Marzolf (1970) reported that gizzard shad longer
than 35 mm are primarily herbivorous, with some selection
for zooplankton when sighted, while young gizzard shad
(< 30 mm) feed primarily on zooplankton (Bodola, 1966).
Organic detritus at times can also be an important food for
the gizzard shad (Dalquest and Peters, 1966; Baker and
Schmitz, 1971).
Apparently, owing to the high reproductive capacity and
rapid growth of gizzard shad in shallow lakes and
reservoirs, predators cannot effectively crop young-of-the-
year fish and hence provide a control on population numbers
(Miller, 1960; Bodola, 1966). It has also been suggested
that gizzard shad may inhibit growth of more desirable fish
and/or limit their numbers through interspecific competition
(Berry, 1958; Miller, 1960).
The gizzard shad is a dominant native grazer in Florida
eutrophic lakes. Crisman and Kennedy (1982), based on
mesocosm experiments, showed that gizzard shad did not
impact chlorophyll a values, productivity or phytoplankton

15
densities, and that shad caused a significant increase in
both the concentration of orthophosphate and its ratio to
total phosphorus under natural stocking densities. The fish
was also responsible for a significant decrease in copepod
density during that study.
Crisman and Beaver (1988) and Threlked and Drenner
(1987) noted that the overall effect of shad grazing was to
stimulate the phytoplankton community and to decrease Secchi
disk transparency. Shad do not effectively graze more
evasive zooplankton such as Diaptomus (Drenner et al.,
1978), particularly during the summer months, although other
evasive herbivores may be simultaneously depressed (Drenner
et al., 1982a). Many common algal taxa remain viable after
gut passage through the digestive tract of shad, especially
blue-greens (Velasquez, 1939; Smith, 1963; Crisman and
Kennedy, 1982).
It has been assumed that the enhancement of
phytoplankton populations by shad grazing is an indirect
effect caused by suppression of herbivorous zooplankton
(Threlkeld, 1987). Shad may also enhance phytoplankton
populations directly by providing a larger quantity of
highly assimilable nutrients from their digestive products
(Crisman and Kennedy, 1982).
Gizzard shad and other clupeids are very sensitive fish
that can be easily stressed and killed. In systems where
gizzard shad is considered a nuisance, as in all Florida

16
eutrophic lakes, the death of large numbers of shad, besides
any aesthetic consequences, will represent an enormous
return of phosphorus and nitrogen into the water column that
is readily available to stimulate the growth of already
overly productive algae, resulting in oxygen depletion that
could lead to another fish kill.
The results of the Crisman and Kennedy (1982)
investigation suggest that the presence of gizzard shad can
promote lake eutrophication both through elevation of
orthophosphate concentrations and differential digestion of
diatoms and green algae, thus increasing the competitive
advantage of blue-green algae. Finally, they emphasize that
this fish species does not appear to be a suitable candidate
for use as a biocontrol agent for phytoplankton in eutrophic
subtropical lakes.
Considerable literature is available concerning the
life history of Dorosoma cepedianum (Drenner, 1977; Lazzaro,
1987). Only a few investigators, however, have attempted to
describe and quantify energy relationships in this species
(Smith, 1971; Garland, 1972; Pierce, 1977; Crisman and
Kennedy, 1982).
Tilapia Ecological Energetics
Blue tilapia (Tilapia aurea), a species native to
Africa and the Middle East, was introduced into the United
States in 1961 in a Hillsborough County phosphate pit (Ware,

17
1973) as a means of controlling nuisance aquatic macrophytes
(Courtenay and Robins, 1973; McBay, 1961). After its
original introduction, the fish spread rapidly throughout
the southeast, particularly Florida, but have not controlled
excessive macrophyte growth (Ware et al., 1975).
Qualitative and quantitative measurements of the gut
contents of blue tilapia in situ indicate that this species
is an opportunistic omnivore, utilizing zooplankton (Spataru
and Zorn, 1978; Mallin, 1986), phytoplankton (Hendricks and
Noble, 1980; Mallin, 1986), and detritus (Hendricks and
Noble, 1980; Mallin, 1986). Gophen et al. (1983) found in
laboratory tests that tilapia greater than 7.6 cm utilized a
series of rapid suctions to draw prey into their buccal
cavity. This mechanism is undirected, and thus, fish in this
size range function as filter feeders. Tilapia smaller than
7.6 cm also function as filter-feeders; however, they also
feed as size selective predators on individual zooplankton
specimens.
Drenner et al. (1984a) reported that grazing activities
by blue tilapia depressed some large algae (Uroqlenopsis and
Ceratium), while the smallest phytoplankton taxa were
enhanced (Rhodomonas. Chrvsochromulina. Chlamvdomonas and
Cyclotella). This enhancement was ascribed both to nutrient
regeneration during gut passage and to fish feces, as well
as the accompanying compositional shifts in the herbivorous
zooplankton community. Little information is available on

18
nutrient release and degree of algal digestion by blue
tilapia (Crisman and Beaver, 1988), but Popma (1982) noted
that some algal cells may remain viable following passage
through the digestive tract of tilapia.
The zooplankton community is also modified by tilapia
grazing activity, and Drenner et al.(1984a) showed in pond
studies that the population of Keratella was suppressed,
while copepodids and adult Diaptomus were enhanced. Gophen
et al. (1983) reported that blue tilapia are selective
feeders on Bosmina and Ceriodaphnia, taxa that have poor
evasive capabilities when compared with more successful
zooplankters such as Mesocyclops (Crisman and Beaver, 1988).
Dickman and Nanne (1987) noted that in Central American fish
ponds very high levels of tilapia (2.5 adults.m'2)
suppressed zooplankton populations and increased the
importance of the bluegreen alga Microcystis aeruginosa.
The spawning behavior of Tilapia aurea is typical of
many cichlids with all incubating duties being performed by
the female (McBay, 1961). Reproductive behavior progresses
from schooling, territorial establishment by the males,
prespawning courtship, spawning, and parental care. McBay
(1961) suggested that the mature female will spawn at a
constant water temperature of 23C. The number of hatchlings
per spawn of T.aurea was comparatively smaller than in most
fishes native to the United States, but apparently
compensated for by strong parental care behavior.

19
Foote (1977) considered temperature, predation and
salinity to be the primary limiting factors in the
distribution of blue tilapia. The sensitivity of blue
tilapia to low temperature is apparently the most important
factor affecting the potential range of the species in North
America. Shafland (1978) tested lower lethal acclimation
temperatures for blue tilapia, spotted tilapia and the
Mozambique mouth brooder, and found that all tilapia tested
died between 6 and 12C. Of these three tilapia species,
blue tilapia was the most tolerant of cold water. According
to McBay (1961), T. aurea will not tolerate temperatures as
low as 9C. Based on these results and January isotherms,
Shafland concluded that the blue tilapia has the potential
of extending its range to include the entire state of
Florida (Williams et al. 1985).
Salinity has a slight but significant effect on the
cold tolerance of blue tilapia, suggesting that the species
may be expected to extend its range farthest north along the
coast and that populations in estuarine systems may be able
to withstand exceptionally cold weather better than inland
populations (Zale, 1984). However, the fish is capable of
finding thermal refugia during cold weather; the presence
and location of these must be considered when assessing
habitat suitability based on thermal criteria (Zale, 1984).
There is concern that the presence of blue tilapia
reduces largemouth bass populations through competition for

20
nesting sites or predation on bass eggs (Noble et al.,
1975). Zale (1984) also found a high degree of trophic
overlap between young tilapia and larval shad in Lake
George, and perhaps the enhanced abundances of individuals
documented following introductions of blue tilapia have
resulted from exploitative competition for zooplankton
during early life stages. He suggested this to be a more
realistic explanation than the "competition for algae and
detritus among adults" theory usually invoked.
These results are confirmed by Beaver and Crisman
(1990) who reported phytoplankton and zooplankton community
alterations in subtropical Florida lakes when blue tilapia
replaces gizzard shad. Chlorophytes and rotifers were
proportionally more abundant in tilapia-dominated lakes
(Crisman and Beaver, 1988), while cladocerans composed a
greater percentage of the zooplankton population in shad-
dominated lakes. Limited zooplankton distribution data
suggest that total zooplankton biomass is comparably
depressed in systems dominated by tilapia. Empirical
evidence suggests that little measurable improvement in
water guality through differential grazing will be realized
if blue tilapia displace shad as the dominant rough fish in
subtropical Florida lakes (Beaver and Crisman, 1990).
A summary from previous investigations in Florida lakes
is compiled in Table 2-1. The data reported are condensed
from many different sources which were already cited.

21
Table 2-1. Summary of previous investigations with blue
tilapia and gizzard shad in Florida lakes.
GIZZARD SHAD
| Chlorophytes
f Cyanophytes
? Chrysophytes
f Chrysophytes
f Rotifers
? Rotifers
J, Cladocerans
1 Cladocerans
J, Zooplankton biomass
f Primary Productivity
Previous studies (Courtenay and Robins, 1973; Ware et
al., 1975; Crisman and Kennedy, 1982; Bays and Crisman,
1983; Zale, 1984; Crisman et al., 1986; Crisman and Beaver,
1988) delineated the differential impact that each fish has
on distinct groups of organisms in Florida lakes. It is not
clear if any change can be deduced for chrysophytes in
systems dominated by blue tilapia, whereas some increase in
chlorophytes is reported. Gizzard shad increased cyanophytes
and chrysophytes. Regarding zooplankton composition, blue
tilapia is reported to increase rotifers while suppressing
cladocerans. The gizzard shad impact on rotifers has not
been determined, although it is known that cladoceran
populations decrease. Finally, both species seem to suppress
total zooplankton biomass while gizzard shad was shown to
stimulate phytoplankton primary productivity.

CHAPTER 3
RELATIVE IMPORTANCE OF BLUE TILAPIA AND GIZZARD SHAD TO LAKE
SEDIMENT AND WATER COLUMN NUTRIENT CONCENTRATION--
ANALYSIS OF FISH FECES
Introduction
Examination of trophic-level interactions has long been
an integral part of limnology (Caird, 1945; Hrbcek et al.,
1961; Nauwerck, 1963; Brooks and Dodson, 1965). Elements
such as herbivory, predation, and nutrient recycling by
animals have always been considered to affect the biomass
and species composition of prey populations in terrestrial
as well as aguatic communities. Several investigators (e.g.,
Hairston et al., 1960; Wiegert and Owen, 1971; Patten, 1973;
Porter, 1977; and Paine, 1980) have examined several aspects
of food webs and the factors which control the biomass and
productivity of various trophic levels. Applicability of
these ideas to aid management practices and promote aguatic
ecosystem recovery are now under investigation (Cooke et al.
1993) .
Shapiro et al. (1975) and Shapiro (1978) recognized
that eutrophication problems are biological manifestations
of nutrient availability and were the first to suggest that
manipulation of trophic interactions (biomanipulation) could
be used as a lake management tool to alleviate the
22

23
biological consequences of eutrophication without the need
for often costly controls of nutrient loading (Crisman and
Beaver, 1990). Biomanipulation, as stated by Shapiro and
Wright (1984), is based on the prediction that increased
piscivore abundance will result in decreased planktivore
abundance, increased zooplankton abundance, and increased
zooplankton grazing pressure leading to a reduction in
phytoplankton abundance and improved water clarity.
This "biological" approach could make it possible to
increase herbivore density in aquatic communities, thereby
lowering algal biomass to levels less than expected for a
given nutrient concentration. In addition, limiting the
abundance or even the occurrence of certain fish could
curtail the flux of nutrients from epilimnetic or littoral
sediment to the pelagic zone. Improvement of the water
quality of lakes with algal blooms by implementing a
combination of these biological techniques could reduce or
eliminate the need for the use of common chemical (i.e.
cooper sulfate) and mechanical methods to deal with
eutrophication (Cooke et al., 1993).
Caird (1945), with his experiment of adding largemouth
bass to a 15-ha Connecticut lake, was one of the first to
publish observations about the effect of increased biomass
of piscivorous fish on the phytoplankton community. More
recently, investigators such as Hrbcek et al. (1961),
Brooks and Dodson (1965), and Hulbert et al. (1972) have

24
demonstrated that planktivorous fish (mainly filter-feeder
fish) can severely reduce or even eliminate large-bodied
zooplankter Daphnia spp.
Planktivorous fishes use two distinct means to feed on
plankton: particulate feeding and filter feeding.
Particulate feeders attack single individual planktonic prey
items which they visually select from the water column
(Werner, 1977; Vinyard, 1980; Lazzaro, 1987). Filter feeders
do not visually detect individual prey but swallow a volume
of water containing the food organisms and entrap the
planktonic forms in structures such as gill rakers and other
filtering structures (see Lazzaro, 1987), using rhythmic
suctioning actions to capture the prey, while either
swimming slowly or remaining quite stationary (Drenner,
1977; Gophen et al., 1983).
Gizzard shad (Dorpspma cepedianunH a filter feeder, is
a dominant native grazer in Florida eutrophic lakes. Crisman
and Kennedy (1982) used mesocosm experiments to demonstrate
that gizzard shad had no impact on chlorophyll a values,
lake productivity, or phytoplankton density. Gizzard shad
was responsible for a significant increase in both the
concentration of orthophosphate and its ratio to total
phosphorus under natural stocking conditions and for the
decrease in copepod density (Crisman and Kennedy, 1982).
Later experiments, however, demonstrated that the
overall effect of shad grazing was to stimulate the

25
phytoplankton community and to decrease Secchi disk
transparency (Crisman and Beaver, 1988; Threlked and
Drenner, 1987). Shad do not effectively graze on more
evasive zooplankton such as Diaptomus (Drenner et al.,
1978), and many common algal taxa remain viable after gut
passage through the digestive tract of shad, especially
blue-green algae (Velasquez, 1939; Smith, 1963; Crisman and
Kennedy, 1982).
Blue tilapia, Tilapia (=Sarotherodon =Oregchromis)
aurea, a fish native to West Africa and Palestine (Trewavas,
1965), was introduced into the United States in 1957 by
researchers at Auburn University who were investigating its
potential as a food and sport fish (Swingle, 1960). In 1961,
the Florida Game and Freshwater Fish Commission acguired
juvenile fish from Auburn University to study the potential
use of blue tilapia as a sport fish (Crittenden, 1962) and
as an agent for weed control (McBay, 1961; Courtenay and
Robins, 1973). After its introduction, the fish successfully
invaded natural habitats throughout the Southeast,
particularly Florida. Blue tilapia has not, however, been a
successful candidate for use as a sport fish or as a
biological control for excessive macrophyte growth (Ware et
al., 1975).
Tilapia can be classified as an opportunistic omnivore,
consuming zooplankton (Spataru and Zorn, 1978; Mallin,
1986), phytoplankton (Hendricks and Noble, 1980; Mallin,

26
1986), and detritus (Hendricks and Noble, 1980; Mallin,
1986). Gophen et al. (1983) reported that tilapia greater
than 7.6 cm act as filter feeders, and tilapia smaller than
7.6 cm, in addition to filter feeding, also feed as size
selective predators on individual zooplankton species
(particulate feeder).
The grazing activities of blue tilapia depressed some
large sized algal groups (Uroqlenopsis and Ceratium), while
the smallest phytoplankton taxa were enhanced (Rhodomonas.
Chrvsochromulina, Chlamydomonas, and Cvclotella) (Drenner et
al., 1984a). Tilapia grazing did suppress the zooplankter
Keratella, while copepodid and adult Diaptomus populations
were enhanced (Drenner et al., 1984a). Tilapia also
selectively fed on Bosmina and Ceriodaphnia (Gophen et al.,
1983), taxa reportedly having poor evasive capabilities.
Little information is available on nutrient release and
degree of algal digestion by blue tilapia. Popma (1982)
noted that some algal cells may remain viable following
passage through the digestive tract of tilapia. Dickman and
Nanne (1987), found that high concentrations of adult
tilapia (2.5 adults itf2) raised in some Central America
ponds suppressed zooplankton populations and increased the
blue-green alga Microcystis aeruginosa.

27
Methods
Fish Collection
Blue tilapia were collected by electroshocking in Lake
Alice, Florida. A boom-style electrofishing boat was used
(7.5 GPP, with a 7 KW generator, Smith-Root, Inc. Vancouver,
Washington). This model was the most efficient
electrofishing unit for the type of lake sampled. The
collected fish were immediately placed on ice to minimize
post-capture digestive processes.
Blue tilapia were measured for total length (mm) and
weight (g). All fish collected were separated into two size
groups (TL), less than 380 mm and greater than 380 mm, which
will subsequently be referred to as size groups T-l and T-2,
respectively. Thirty-six fish were collected. Thirteen of
these were in group T-l; twenty-three were in group T-2
(fish length ranged from a minimum of 284 mm to a maximum of
450 mm). Fecal material, here defined as the collection of
all digested material that could be identified by its
brownish-dark coloration in the fish intestines, was hand-
stripped from all the fishes and pooled into two different
jars (T-l and T-2). The sample jars were immediately placed
on ice in a cooler for transport to the laboratory. All
sample material was kept at 4C within four hours after
collection.
Samples were subdivided into four different groups;
groups 1 and 2 pulled from the T-l jar and groups 3 and 4

28
from the T-2 jar. Three aliquots for analysis (replicates)
were taken from each of the four groups.
Gizzard shad were collected using an experimental
monofilament gillnet. All fish sampled were collected as
part of a population control project being conducted by the
Florida Game and Freshwater Fish Commission, Eustis (FL).
The sampling net had 6 panels ranging from 6.3 to 12.7 cm
stretch mesh in 1.3 cm increments. Gillnetting was done in
Lake Denham, Florida, in increments of two hours.
Gizzard shad were measured for total length (mm) and
weight (g). All fish collected were separated into two size
groups (TL), <320-mm and >320-ram, which will subsequently be
referred to as size groups S-l and S-2, respectively. A
total of 218 fish were collected in five different gillnet
settings. Minimum fish length was 203 mm, and the maximum
length was 440 mm. Fecal material was hand-stripped in the
field from all the captured fishes and poured into two jars
(S-l and S-2). The sample jars were immediately placed on
ice in a cooler for transport to the laboratory. All sample
material was kept at 4C within four hours after collection.
Samples were subdivided into four different groups;
groups 1 and 2 pulled from the T-l jar and groups 3 and 4
from the T-2 jar. Three aliquots for analysis (replicates)
were taken from each one of the four groups.

29
Feces Analysis
The digestive tract was removed through an incision
made from the anus anterior to the base of the pelvic fin,
then dorsally to the posterior side of the pectoral girdle
to the dorsal side of the coelom. Fecal material was hand-
stripped from the beginning of the intestine to the base of
the anus and placed into a 250 ml jar. The feces were then
pooled according to size group (T-l, T-2, S-l and S-2).
Laboratory analyses consisted of determination of
percent dry matter, percent organic matter, bulk density,
heat content upon combustion, protein analysis, total
phosphorus and total nitrogen content and two algal
bioassays. Dry weight/water content of feces was measured by
weighing samples before and after drying at 105C for 16
hours; organic matter was measured using a cool muffle
furnace before ashing the samples at 600C for at least 3
hours and then weighing. Nitrogen was measured as total
Kjeldahl nitrogen (TKN) using a Technicon-II semi-automated
manifold following a modification of the methods described
by Bremner and Mulvaney (1982) (selenium was not used as a
catalyst). The digestate was also used for total phosphorus
(TP) determinations. Liberated orthophosphate was determined
with the ascorbic acid method (APHA, 1985).
All samples for calorimetry analyses were freeze dried.
Volatilization began at -30C. Total energy was determined
by oxygen bomb calorimetry (Parr Model 1261 Isoperibol

30
Calorimeter). Pre-weighed (pre-burn weight, approximately
one gram of dried material) fecal samples were placed in a
microbomb. The bomb was placed in 2 liters of water at 30-
31C. The sample was ignited and burned in the presence of
oxygen. After ignition, increases in water temperature were
registered to calculate energy within the sample, expressed
as gross heat (cal.g'1).
An algal bioassay was done using thirty-six 300 ml
flasks containing FWH+Si (a standard medium for maintenance
culture of freshwater blue-green algae). Eighteen flasks
were prepared without KN03 (medium -N); and eighteen flasks
prepared without K2HP04 (medium -P). These two groups of
eighteen flasks were each subdivided into three different
subgroups of six; one subgroup containing 3 ml of shad
feces; another containing 3 ml of tilapia feces; and the
third subgroup without feces, to function as a control. Each
group of six flasks was subdivided further into two
subgroups: three flasks inoculated with a strain of blue-
green algae (Microcystis aeruginosa), and three flasks
inoculated with a strain of green algae (Selenastrum
capricornutunU These sets of flasks were incubated for
seven days in a room under constant light and temperature
conditions.
Samples from all experimental flasks were analyzed for
chlorophyll a using extractive and fluorimetric methods
(APHA, 1989). Cell counting of algae from samples collected

at the beginning and end of the experiment from each flask
was performed, using a 2 ml chamber on an inverted Leitz
microscope.
31
Results and Discussion
Bulk density calculated for tilapia feces had a mean
value of 0.2576 g.cm'3, with a minimum value of 0.2326 g.cm"
3, and maximum value of 0.2815 g.cm'3 (Table 3-1). For shad
feces, the mean value was 0.1401 g.cm"3, with a minimum
value of 0.1258 g.cm'3 and maximum value of 0.1493 g.cm"3
(Table 3-1).
Percent dry matter for tilapia feces had a mean value
of 13.0 %, with a minimum value of 12.6 %, and a maximum
value of 13.6 % (Table 3-1). For shad, the mean value was
8.0%, with a minimum value of 7.5 % and a maximum value of
8.6 % (Table 3-1). Percent organic matter calculated from
tilapia feces had a mean value of 49.6%, with a minimum
value of 48.6 %, and maximum of 51.6 % (Table 3-1). Shad had
a mean value of 68.0 %, with a minimum of 67.0 %, and a
maximum of 68.9 % (Table 3-1).
Percent protein was calculated for tilapia and had a
mean value of 21.6 %, minimum of 21.0 %, and maximum of
23.2 % (Table 3-1). For shad, the mean value was 32.4 %,
with a minimum of 30.25 %, and maximum of 33.75 %
(Table 3-1).

Table 3-1. Physico and chemical characteristics of shad and
tilapia excretions expressed as ml and g.
PARAMETER
SHAD
TILAPIA
SHAD
TILAPIA
SHAD
TILAPIA
Mean
(ml) (g)
Mean
(ml) (g)
St. Deviat.
(ml) (g)
St. Deviat.
(ml) (g)
Range(*)
(ml) (g)
Range(*)
(ml) (g)
Bulk
Density
(g. cm-5)
1.0/ 0.1401
1.0/ 0.2576
0.72/ 0.010
0.08/ 0.022
(0.90-1.06)
(0.12-0.15)
(0.9-1.09)
(0.23-0.28)
% Dry
Matter
N/A /8.0366
N/A /12.988
N/A /0.5748
N/A /0.4856
(7.49-8.64)
(12.6-13.66)
% Org.
Matter
N/A / 67.84
N/A / 49.66
N/A /0.7954
N/A /1.3929
(67.06-68.90)
(48.6-51.63)
% Protein
N/A /32.427
N/A /21.655
N/A /1.3467
N/A /1.0568
(30.25-33.75)
(21.1-23.19)
Calories
(cal. g"1)
7516.35 ml
4041.05 g
4531.18 ml
2604.13 g
253.49
136.28
278.53
160.08
(7147-7852)
(3842-4222)
(3805-5091)
(2187-2926)
TN
N/A / 52.98
N/A / 34.65
N/A /0.9307
N/A /1.7055
(48.38-54.00)
(33.4-37.14)
TP
N/A / 6.51
N/A / 6.63
N/A /0.3636
N/A /0.4708
(5.98 7.04)
(6.05- 7.13)
(*) Numbers expressed in the 1st line are in ml.
Numbers expressed in the 2nd line are in g.
When only one line presented, it is expressed in g.
u>
to

33
Caloric content, expressed as gross heat, for tilapia
feces had a mean value of 2604.13 cal.g"1, with a minimum
value of 2186.88 cal.g'1, and maximum of 2926.08 cal.g"1
(Table 3-1). Shad had a mean value of 4041.05 cal.g" 1, and
minimum and maximum values of 3842.40 cal.g'1 and 4212.81
cal.g'1, respectively (Table 3-1). Pierce ( 1977 ), from data
obtained from 27 samples of shad fecal material, calculated
its mean caloric content as 1953 cal.g"1.
Total nitrogen for tilapia feces displayed a mean value
of 34.6 pg.l"1, with a minimum of 33.42 pg.l"1, and a
maximum of 37.14 pg.l'1 (Table 3-1). For shad, the mean was
52.98 pg.l"1, with the minimum and maximum values being
48.38 pg.l"1, and 54.0 pg.l"1, respectively (Table 3-1).
Mean total phosphorus (TP) for tilapia was 6.63 pg.l'1, with
a minimum of 6.05 pg.l"1, and a maximum of 7.13 pg.l"1
(Table 3-1). Shad had a mean of 6.51 pg.l"1, with a minimum
of 5.98 jug. I"1, and a maximum of 7.04 jug.l"1 (Table 3-1).
Absolute values for total nitrogen, percent of
nitrogen, crude protein, total phosphorus and percent
phosphorus composition for gizzard shad and blue tilapia
fecal material are provided in Tables 3-2 and 3-3.
There were significant differences in the amount of
feces hand-stripped from both fishes. Tilapia had the
greatest amount of feces, and it was not possible to isolate
any particular factor responsible for such an occurrence. It
may be related to the length of the intestinal tract (which

34
Table 3-2. Total nitrogen, total phosphorus and protein
content for gizzard shad feces.
pg N
% N
Crude Protein
pg P
% P
Group 1
A
48.38
4.84
30.25
5.98
.60
B
48.95
4.89
30.56
6.84
.68
Group 2
A
54.00
5.40
33.75
6.41
.64
B
53.65
5.36
33.50
6.19
.62
Group 3
A
51.43
5.14
32.12
7.04
.70
B
53.36
5.34
33.37
6.79
.68
Group 4
A
52.44
5.24
32.75
6.55
. 65
B
53.04
5.30
33.12
6.26
. 63
Table 3-3. Total nitrogen, total phosphorus and protein
content for blue tilapia feces.
pg N
%N
Crude Protein
pg P
%P
Group 1
A
33.68
3.37
21.06
7.13
.71
B
33.42
3.34
20.87
6.86
.69
Group 2
A
37.14
3.71
23.19
6.47
.65
B
34.38
3.44
21.50
6.05
.60
Group 3
A
22.61
2.26
14.12
5.50
.55
B
22.44
2.24
14.00
5.30
.53
Group 4
A
27.40
2.74
17.12
12.66
1.27
B
28.48
2.85
17.81
11.54
1.15

35
in tilapia is longer) or the time of the year (i.e.,
spawning time). In gizzard shad, the length of the
intestinal tract increases with increasing standard length.
Schmitz and Baker (1969) found that the intestine-to-body-
length ratio for gizzard shad was about 2.8:1, while in
threadfin shad the ratio averaged 1.8:1. Although the length
of the intestine in a fish may be partially determined by
its diet, Schmitz and Baker (1969) believe that the length
and form of the intestine are probably the result of a
complex of intrinsic and extrinsic factors that can not be
explained only by diet.
Tilapia produces five times more feces than bighead
carp, whereas the growth rate of both species was comparable
(Opuszynski and Shireman, unpubl.). In this study, I found
that the average feces produced by the shad was 2.08 0.34g
fresh feces per lOOg fish body weight and 4.5 0.21g fresh
feces per lOOg fish body weight for the tilapia. The fact
that blue tilapia feces' bulk density is twice as high as
shad feces (Table 3-1), indicates that tilapia produces four
times more feces as dry matter than shad. Therefore, blue
tilapia seems to be more suitable than bighead carp
(Opuszynski and Shireman, unpubl.) and shad for use in
biological schemes for water guality improvement.
There are obvious macroscopic differences in the feces
produced by these two fish taxa. Tilapia feces are released

36
as mucilaginous pellets, while shad feces are released into
the environment as a flocculent material.
The bulk of both fish feces consisted of small green
and blue-green algae species, dominating the phytoplankton
in the assemblage at the time the fishes were collected.
Many of the algal cells did not show any signs of digestion,
or they were slightly digested but easily identifiable under
a microscope. It is known that many algal taxa may be viable
after passage through fish digestive tracts (Fish, 1951,
1955; Lowe, 1959; Smith, 1963). Fish (1951, 1955) reported
that, of the algae eaten by tilapia, blue-green and green
algae generally pass through the fish undigested and that
diatoms form the main usable food (Cailteux, 1988). In this
study, I found green algae of the genus Closterium, the
blue-green Microcystis, and the diatoms Melosira,
Fraqilaria, Tabellara. and Cyclotella in tilapia feces.
Shad feces contained the green algal genus Ankistrodesmus,
the diatom Melosira. the dinoflagellate Peridinium, and the
blue-green Microcystis.
Mean values for chlorophyll a (final initial
measurements) in shad feces were [1] for the inoculum made
in the medium deficient of nitrogen(-N) and the blue-green
algae 541.57 mg.m"3 36.85; and 859.18 mg.m"3 17.93 with
green algae; [2] for the inoculum made in the medium
deficient of phosphorus (-P) and the blue-green algae 432.21
mg.m"3 22.02; and 1173.76 mg.m'3
13.96 with green algae.

37
Mean values in tilapia feces were [1] for the inoculum made
in the medium deficient of nitrogen (-N) and the blue-green
algae -3 64.05 mg.m'3 38.06; and 329.33 mg.nf3 31.38 with
green algae; [2] for the inoculum made in the medium
deficient of phosphorus (-P) and the blue-green algae 683.14
mg.m'3 38.62; and -335.58 mg.m"3 17.98 for the inoculum
with green algae.
The experiment was harvested on the seventh day-
following daily evaluation of fluorescence. As a major
trend, it can be said that fluorescence values increased
dramatically (in the shad samples) during the first four
days, after which they leveled off and did not change very
much. In the medium containing blue-green algae, the
combination shad feces -N as well as shad feces -P displayed
a steady increase in values, which became more noticeable
after the fourth day (Figure 3-1)). For tilapia feces, the
medium deficient of nitrogen (-N) had the same value from
the beginning and declined after the fourth day; whereas the
medium deficient of phosphorus (-P), remained steady and
began to increase slightly after the fourth day (Figure
3-1) .
In the medium containing green algae, the mixture of
shad feces -N increased progressively daily and reached its
highest values after the fourth day, which was a ten-fold
increase. The combination -P displayed the same trend,
however, with higher values increasing thirteen times after

Microcystis aeruginosa
F
I
u
o
r
e
s
c
e
n
c
e
0.05
0.04
0.03
0.02
0.01
4-
' ^ I I I I ^ I I I! ^ I I, I ^
Ctrl -N
Ctrl -P
i i 1 1 1 1 *
-P
1 1* 1 1* 1 1* 2 2* 2 2* 2 2* 3 3* 3 3* 3 3- 4 4* 4 4* 4 4* 6 5* 6 5* 5 5* 6 6- 6 6* 6 6*
Time (days)
Shad
+
Tilapia
* Control
Figure 3-1. Fluorimetric activity from shad and tilapia
feces in medium with Microcystis aeruginosa.
OJ
00
Z .

39
the fourth day (Figure 3-2). For tilapia feces, the medium
deficient of nitrogen (-N), as well as the medium deficient
of phosphorus (-P), showed a daily decrease in fluorescent
values and on the fourth day of the experiment they
stabilized in a value 50% lower than the initial number.
(Figure 3-2). Shad feces displayed a five fold increase in
fluorometric values compared with the measured in the
medium with tilapia feces.
Phytoplankton biomass in the fecal material was
calculated from chlorophyll a concentrations assuming 1 mg
chlorophyll a = 67 mg dry weight (APHA, 1982). Blue tilapia
fecal material suppressed phytoplankton biomass (mg.dry
weight) in 75% of the samples of blue green and green algal
prepared in -N and -P media. Mean suppression in these
samples was approximately 50% of the initial values (Table
3-4). Gizzard shad fecal material, on the other hand,
increased phytoplankton biomass in all of the samples of
blue green and green algae. The mean concentration increase
was greater than tenfold of the initial values (Table 3-4).
Conclusions
Adult fishes that employ filter-feeding as their
primary feeding method are typically visual particulate
feeding planktivores when juveniles. They switch from a diet

Selenastrum capricornutum
F
I
u
o
r
e
s
c
e
n
c
e
Time (days)
' Shad + Tilapia Control
Figure 3-2. Fluorimetric activity from shad and tilapia
feces in medium with Selenastrum capricornutum.
4*.
o

Table 3-4. Phytoplankton biomass (mg.dry weight-1) in fish excretions
calculated from chlorophyll a values.
FISH + ALGA MEDIUM
INITIAL VALUE
mg/dry weight
FINAL VALUE
mg/dry weight
FIN-INIT (X)
mg/dry weight
(X)-CTRL.
mg/dry weight
Til (-N)+ Bluegreen
68054
43663
-24391
-23939
Til (-N)+ Green
62232
84297
22065
23269
Til (-P)+ Bluegreen
79421
33651
-45770
-51642
Til (-P) + Green
68756
46273
-22483
-43562
Shad (-N)+ Bluegreen
2710
38995
36285
36737
Shad (-N)+ Green
4818
62383
57565
58769
Shad (-P) + Bluegreen
3061
32019
28958
23086
Shad (-P)+ Green
2810
81454
78644
57565
Ctrl (-N)+ Bluegreen
903
451
-452
Ctrl (-N)+ Green
1882
678
-1204
Ctrl (-P)+ Bluegreen
502
6374
5872
Ctrl (-P) + Green
2308
23387
21079

42
composed of large zooplankton as juveniles to increasing
dependence on phytoplankton and/or smaller and non-evasive
zooplankton as adults (Kutkuhn, 1957; Cramer and Marzolf,
1970; Janssen, 1978; Durbin, 1979; Drenner et al., 1982b;
Lazzaro, 1987; and Yowell and Vinyard, 1993). Although the
feeding habits of filter-feeding fish have been extensively
studied, information about the quantitative feeding under
natural conditions is scarce.
The theoretical basis of fish bioenergetics has
received considerable attention in the fisheries literature.
Sophistication in describing basic equations and the
addition of computer analysis have enabled investigators to
partition more accurately the energy requirements of a
population. The goal of this study was to provide additional
information on the feces produced by gizzard shad and blue
tilapia including fecal composition, and nutrient and
caloric content. Experiments were conducted to compare the
potential impact of the fecal material of each respective
fish species on the primary production in aquatic systems.
Feces composed the greatest constituent of the energy
balance estimated from an experiment with bighead carp
employing energy budget equations to calculate food
consumption and food assimilation, i.e., C=P+R+F+U (1) and
A=C-F+U (2), where C=consumption, A=assimilation,
P=production, R=respiration, F=feces, and U=urea (Opuszynski
and Shireman, unpubl.).

43
I found that bulk density for tilapia fecal material
was twice the value for shad feces. Averaged shad feces
analysis had a composition of 37% more organic matter, 50%
more protein and 50% less dry matter than tilapia feces
which makes shad feces more organic, more protein-rich and
more easily dissolved in the environment. Caloric content of
shad feces was 60% higher than tilapia feces.
In general, assimilation efficiencies, which are
calculated using the balance of total calories of ingestion
and egestion material in fish, have been assumed to
approximate 80 percent of ingested energy (Winberg, 1956;
Ricker, 1968; Mann, 1969; Moriarty and Moriarty, 1973).
Although little is known about the energetics of omnivorous
fish such as shad and tilapia, an assimilation efficiency of
42 percent is reported for shad by Pierce (1977), and of 70-
80 percent for tilapia (Moriarty and Moriarty, 1973).
In this study, I found caloric content in shad fecal
material to have a mean of 4041.05 cal.g"1 (Table 3-5),
whereas tilapia had a mean of 2604.13 cal.g-1 (Table 3-6).
Considering both fishes relative to a natural diet, tilapia
is egesting material that has 60% less caloric content than
shad. This is, consequently, in accordance with what is
stated in the literature about the assimilation efficiency
for both, where tilapia has roughly two times greater
assimilation efficiency than shad, being expected to egest
material with a lower caloric content.

Table 3-5. Caloric content of gizzard shad
feces.
SHAD
CALORIC CONTENT
(Cal/g)
Group 1 (A)
4115.56
(B)
4134.18
(C)
4150.35
Group 2 (A)
4066.21
(B)
4221.81
(C)
4196.37
Group 3 (A)
4026.14
(B)
3935.90
Group 4 (A)
3894.76
(B)
3867.86
(C)
3842.40
Table 3-6. Caloric content of blue tilapia
feces.
TILAPIA
CALORIC CONTENT
(Cal/g)
Group 1 (A)
2532.20
(B)
2490.92
Group 2 (A)
2722.98
(B)
2648.89
(C)
2926.08
Group 3 (A)
2533.76
(B)
2560.45
(C)
2417.74
Group 4 (A)
2312.26
(B)
2186.88

45
The diet of the gizzard shad is one of the most
controversial aspects in the ecology of this species. It is
generally agreed that very young shad (<35 mm) feed almost
exclusively on microcrustaceans (Kutkuhn, 1958; Cramer and
Marzolf, 1970), and fish larger than 35 mm have been
reported to consume a wide variety of food materials.
Tilapia are primarily vegetative feeders, feeding on algae
or aquatic plants, though occasionally zooplankters and
insect larvae picked up with bottom debris are eaten (Lowe,
1959). However, despite the similarity in feeding habits,
differing fecal composition and assimilation rates suggest
that tilapia and shad play significantly different roles in
the nutrient dynamics of aquatic systems.
Chlorophyll a values for both shad and tilapia fecal
material were very different between the beginning and end
of the bioassays. In both bioassays for blue-green and green
algal groups, tilapia initial values were 10 to 15 times
higher than for shad (tilapia mean value of 915 mg.nf3 and
shad mean value of 67 mg.nf3); whereas at the end of
bioassays, the situation was completely inverted, with shad
displaying a tenfold increase in chlorophyll a (shad mean
value of 685 mg.nf3) regardless of the algal group tested.
(Figures 3-3 to 3-6) Tilapia chlorophyll a values at the end
of bioassay varied with the algal group tested. In the
Microcystis experiment, all values were negative indicating
suppressed chlorophyll a values (mean of -691 mg.nf3)

mg/nT3
1200
1000
800
600
400
200
0
Figure 3-3. Bioassay initial (I) and final (F) values of chi. a (mg.nf3)
with M.aeruginosa in medium -N.
I
Control Tilapia Shad
Time

mg/rrT3
Time
Figure 3-4. Bioassay initial (I) and final (F) values of chi. a (mg.nf3)
with M.aeruginosa in medium -P.

1400
1200
1000
800
600
400
200
0
Figur
.rrT3
F
Control Tilapia Shad
Time
3-5. Bioassay initial (I) and final (F) values of chi. a (mg.itf3)
with S.capricornutum in medium -N.

1400
1200
1000
800
600
400
200
0
.m~3
Control Tilapia Shad
Time
3-6. Bioassay initial (I) and final (F) values of chi. a (mg.m'3)
with S. capricornutum in medium -P. 5

50
(Figures 3-3 and 3-4), while in the Selenastrum tests, about
66% chlorophyll a had negative values indicating
suppression, with the remaining values being positive with a
mean of 600 mg.nf3 (Figures 3-5 and 3-6).
McDonald (1985) found that blue tilapia enhanced the
growth of Ankistrodesmus cells. Ware et al.(1975) concluded
that blue tilapia had displayed little potential for algal
control in Florida waters, and in fact, had spread rapidly
and become a nuisance. Phytoplankton biomass in the fish
fecal material calculated from chlorophyll a values at the
beginning and end of the current bioassay showed that blue
tilapia suppressed phytoplankton biomass in 75% of the
samples of blue green and green algae to a level exceeding
50% of the initial value. Gizzard shad increased
phytoplankton biomass concentration in all samples of blue
green and green algae more than tenfold from the initial
values.
It has been reported (Crisman and Kennedy, 1982) that
gizzard shad are not suitable for use as a biocontrol agent
for phytoplankton because they have no impact on chlorophyll
a values, productivity or phytoplankton densities, and they
can promote lake eutrophication through elevation of
orthophosphate concentrations and differential digestion of
some algae groups, especially blue-greens and greens.
On the other hand, results of extensive work with
tilapia are still inconclusive regarding the possible

51
importance of this species in biomanipulation schemes. It is
known that adult blue tilapia employ filter-feeding as their
primary feeding method, while they feed as visually-
oriented, particulate-feeding zooplanktivores as juveniles
(Yowell and Vinyard, 1993). It is assumed that minimization
of feeding costs while maximizing net energy return has been
the basis for the filter-feeding strategy. The remaining
guestion is what causes these fishes to switch their feeding
behavior.
Facultative planktivores that use both feeding modes
interchangeably (Leon and OConnel, 1969; Crowder and
Binkowski, 1983; and Helfman, 1990) usually filter-feed on
small, abundant zooplankton and particulate-feed on large,
less abundant forms. Yowell and Vinyard (1993), suggest that
this is a stratagem which maximizes net energy return in
response to a changing prey situation. The reduction in
zooplankton density may then allow an increase in non-grazed
phytoplankton (Drenner et al., 1987; Vinyard et al., 1988).
The results of this study indicate that blue tilapia
fecal material, either by its composition and/or the absence
of some essential nutrient, completely suppressed
Microcystis aeruginosa chlorophyll a production, and appears
to have suppressed Selenastrum capricornutum chlorophyll a
production in > 60% of the samples. Gizzard shad feces
increased chlorophyll a production values in both blue-
greens and greens (tenfold increase). This study suggests

52
that, in Florida systems with conditions similar to those of
the lakes studied, biomanipulation techniques can mitigate
cultural eutrophication if gizzard shad are replaced by blue
tilapia.

CHAPTER 4
SEDIMENTS AND HISTORICAL ECOLOGY OF TWO CENTRAL
FLORIDA URBAN LAKES
Introduction
In the absence of historical data, past lake conditions
can be reconstructed based on interpretations of the
sedimentary record. Assuming that lake sediments accumulate
in an orderly fashion, paleolimnology may provide
information about how the lake ecosystem has changed over
time. Reconstructing the reason for past changes in the lake
may be used to predict the lake's response to future
management strategies (Smeltzer and Swain, 1985; Brenner et
al., 1993).
Lake sediments originate from numerous sources. The
main sources are both the biochemical substances produced by
organisms, or resulting from their degeneration, and
morphological pieces of specific organisms. The sediment
constitution is influenced primarily by the geomorphology of
the lake basin and the drainage basin (Wetzel, 1983).
Paleolimnology has as a principal objective a formulation of
general principles about the way lakes change with time
(Livingstone, 1981; Cohen and Nielson, 1986; Johnson et al.,
1991) .
53

54
Florida has about 7800 lakes varying in size from 0.4
ha to over 180,000 ha. (Canfield and Hoyer, 1988). Lakes in
Florida serve many agricultural, domestic, industrial and
recreational purposes (Canfield and Hoyer, 1992). Outside
the glaciated regions in North America, Florida lakes
constitute the largest group of natural lakes (Hutchinson,
1957) on the continent and the most important group of
solution basins (Hutchinson, 1957; Crisman, 1992)
Limnological characteristics and productivity from
these lakes vary widely and range from oligotrophic to
hypereutrophic (Canfield and Hoyer, 1988; Brenner et al.
1990; Beaver et al. 1981). Although Florida's aquatic
systems are large in number and have a significant economic
impact, water quality data have been collected for few lakes
(>10%) (Brenner et al., 1990), and routine data acquisition
began fairly recently, in the 1960s and 1970s (Huber et al.,
1982).
Considering Florida's aquatic system dimensions and
economic importance, little information has been reported on
the sediments of Florida lakes, i.e. Flannery et al. (1982);
Brenner and Binford (1988); Brenner et al. (1990); Stoermer
et al. (1992); Gottgens (1992); Brenner et al. (1993). Here
I provide information on the relationship between
accumulation rates of sediment variables such as organic
matter, water content, total nitrogen and total phosphorus

55
and estimated trophic status based on sedimentation rates
(g.cm'2.yr) for the last century.
Sediment cores collected from Lake Bonny and Lake
Gibson (Polk County, Florida) were dated isotopically with
210Pb and 137Cs to estimate material accumulation rates in
the sediment profile using radioactive decay of fallout
("unsupported") 210Pb. These techniques have been widely
used to detect changes in sediment accumulation rates due to
urban development in a watershed (Smeltzer and Swain, 1985),
clear cutting of vegetation (Oldfield et al., 1980) and
major climatic events (Robins et al., 1978).
In this study, these techniques were applied to
identify lake sediment profile physicochemical
characteristics and evaluate the potential impact of gizzard
shad feces when it is the dominant filter feeder fish on
lake sediment composition and net accumulation rates of
organic matter, total nitrogen and total phosphorus for
Lakes Bonny and Gibson. Whenever possible, this information
will be correlated with the results from sediment cores for
Lake Parker and Lake Hollingsworth reported by Schelske et
al. (1992). For this study, both lakes were considered to be
dominated by blue tilapia population as the main filter
feeder fish.
According to the classification of Forsberg and Ryding
(1980), Lake Bonny is hypereutrophic and Lake Gibson is an
eutrophic lake. Mean values for all physicochemical and

56
biological parameters are provided in Chapter 5 of this
study. Both are small urban lakes subject to anthropogenic
pressures in the form of industrial, residential and
commercial development in the area and recreational uses
such as boating, fishing, and skiing. In addition, there are
agricultural activities at Lake Gibson watershed.
Study Sites
Lake Bonny is a small (144 ha of surface area), shallow
(zmax= 2.5 m), urban lake bordered by the city of Lakeland,
(Polk County, Florida). It has no known discharges or
withdrawals (Polk County Water Resources Division, 1990).
The population of Lakeland has increased from less than
500 inhabitants in 1880 to nearly 75,000 people in 1993. As
a result, all of the lake watershed was developed for
commercial and residential uses. Lake Bonny has consistently
had poor water quality (Polk County Water Resources
Division, 1990) with an average 1992 TSI of 73.5. Urban
runoff from commercial and residential development in the
watershed and low lake levels resulting from seepage have
been a major problem. Extremely poor water quality was
recorded in 1986 at a time of extreme low lake levels (Polk
County Water Resources Division, 1990). Based on current
data, the TSI was 98; total nitrogen 9.0 mg.l'1, total
phosphorus 0.7 mg.l-1, Secchi 0.2 m and chlorophyll a was
220 mg.nf3.

57
Using the classification system of Forsberg and Ryding
(1980), Lake Bonny is classified as hypereutrophic. Average
total phosphorus concentration is 59 pg.l'1 and average
total nitrogen concentration is 1858 pg.l"1. Total
chlorophyll a concentrations average 40 pg.l"1, the water
clarity as measured by a Secchi disc averages 0.6 m, and
average pH is 7.8. Table 5-10 exhibits water quality data
for Lake Bonny.
Before 1985, the plant community of Lake Bonny was
comprised mainly of Hvdrilla verticillata. Due to the
macrophyte chemical control program employed by the City of
Lakeland, it has been replaced by Typha sp. as the dominant
macrophyte. Typha sp. is responsible for a percent lake area
coverage (PAC) of 10% (cf. Canfield and Hoyer, 1992).
The fish population in this lake has changed
dramatically. The species that survived the drought of 1983
are currently repopulating the lake. The most abundant open-
water species collected in experimental gillnets are gizzard
shad and Florida gar with 17.0 and 11.3 fish/net/24 hr,
respectively (Canfield and Hoyer, 1992).
Lake Gibson is a small (192 ha of surface area),
shallow lake (zniax= 6.1 m) located on the outskirts of the
City of Lakeland in a rapidly urbanizing area where citrus
groves and pastures are giving way to large-scale commercial
and moderate-density residential development. A domestic
wastewater treatment plant from one local elementary school

58
discharges to a wetland north of Lake Gibson, which
overflows into the lake. There are no known withdrawals
(Polk County Water Resources Division, 1990).
Lake Gibson is classified as eutrophic according to
Forsberg and Ryding's (1980) classification system. The 1992
Florida TSI for Lake Gibson is 55 (Polk County Water
Resources Division, 1990). The water is slightly tannic
(color is 43 cpu) which lowers the Secchi disc value and may
bias the TSI value. The average total phosphorus
concentration is 94 pg.l-1, and the average total nitrogen
concentration is 1570 pg.l'1. Total chlorophyll a
concentration averages 23 pg.l'1 and the water clarity as
measured by use of a Secchi disc averages 1.3 m. Lake's
annual primary productivity (mean = 241.18 mgC.m"3.h
229.1) displaying noticeable seasonal variability and the
presence of blooms of cyanobacteria every few years suggests
an eutrophic status for Lake Gibson. Water guality data for
Lake Gibson are shown in Table 5-8.
The dominant macrophytes of the lake are water
primrose, alligatorweed, and pennywort, which are
responsible for 5% lake area coverage (Canfield and Hoyer,
1992).
No recent fish population evaluation was obtainable for
this lake. From the 1979 evaluation, ten species of fish
were reported, with channel catfish being dominant.

59
Methods
An 89 cm mud-water interface core was collected from
the middle of Lake Bonny using a piston corer with a 12 cm
diameter, 1.83 m long cellulose acetate butyrate (CAB) core
barrel. Externally, the core was characterized by a light
color sediment for the first 10 cm and a typical black mud
color below in the rest of the core.
A 60 cm core was taken from the central portion of Lake
Gibson using the same piston corer described above. The
presence of sand mixed with mud was noticed beginning at the
50 cm depth. Sand content increased deeper in the core. Due
to the soupiness of the sediments, in both cores a spoon was
used to collect the samples down to the 10-cm layer. From
that point on, collection was possible using a spatula. The
sediment from Lake Gibson was more consolidated on the
bottom than the sediment from Lake Bonny.
Concentrations of 210Pb and 137Cs were measured by
direct y-assay using a P-type, intrinsic-germanium detector
(Princeton Gamma Tech). The counting system used for
spectral analysis is located in the University of Florida's
Department of Environmental Engineering Sciences Low
Background Counting Room. The electronics for the system
include a preamplifier (RG11B/C, Princeton Gamma Tech,Inc.),
amplifier (TC 242, Tennelec), bias supply (5 kV, TC 950,
Tennelec), power supply (TC 909, Tennelec), and transformer
(Sola).

60
Samples for isotope analysis were dried at 95C for 24
hours, pulverized by mortar and pestle, weighed, and placed
in small plastic vials. Core sections were combined (up to 4
cm) to obtain an adequate sample weight (> 1 g). Vials were
sealed with plastic cement and left for 14 days to
equilibrate radon (222Rn) with radium (226Ra) Counting times
were never less than 23 hours. Standards were counted to
track efficiency (counts per y) and to calculate a 226Ra
conversion factor (PCi per count per sec.). Blanks were
counted to determine background radiation.
Bulk density and water content of sediments were
measured by weighing samples before and after drying at
95C. Percent organic matter was evaluated by loss on
ignition (LOI) at 550C for one hour (Hkanson and Jansson,
1983). Nitrogen was measured as total Kjeldahl nitrogen
(TKN) using a Technicon-II semi-automated manifold after
digestion following Bremner and Mulvaney (1982) but modified
to exclude selenium as a catalyst. The digestate was also
used for total phosphorus (TP) determinations. Liberated
orthophosphate was determined with the ascorbic acid method
(APHA, 1985).
Results and Discussion
The Lake Bonny Record
The 89 cm Lake Bonny sediment core had a mean bulk
density of 0.094 g.cm-3 0.042, mean dry matter of 9.16%

61
3.93, mean organic matter of 56.00% 12.69, mean total
nitrogen of 22.32 mg.g'1 2.68 and mean total phosphorus of
4.10 mg.g'1 1.53 (Table 4-1).
Lake Bonny had high organic content in surface
deposits, but this declined somewhat with depth and age down
to 19 cm of the core which coincides with the 1980s, and
then increased again to the bottom where it reached highest
values (dated at the beginning of this century, 1909)
(Figure 4-1). Total nitrogen and total phosphorus plotted
against time were constant throughout the whole core (Figure
4-2). Bulk density increased in the top 40 cm of the core
which coincides with a 210Pb determination of sediment age
of approximately 1950, i.e., about the time when much
agriculture in the watershed was being converted to urban
development (USGS Topographic Maps, Lakeland Quadrangle,
1944, 1975).
Total P reconstruction near the base of the core
suggests that the lake has always been mesotrophic to
eutrophic. However, the lake has experienced periods of much
higher nutrient enrichment, and values for total P in excess
of 5.0 mg.g'1 were computed for several sections that post
date the 137Cs peak of 1960s and 1970s. No change in total P
concentration has been noticed since then, suggesting that
the lake trophic state has remained the same for the last
thirty years.

Table 4-1. Sediment core physical and chemical characteristics
from two central Florida urban lakes.
PARAMETER
LAKE
GIBSON
LAKE BONNY
Mean
St. Dev.
Range
Mean
St. Dev.
Range
Bulk Density
0.279
0.319
0.094
0.094
0.042
0.021
(g. cm-3)
1.644
0.218
Dry Matter
21.61
13.93
9.47
9.16
3.93
2.24
(%)
76.49
19.81
Organic Matter
38.17
14.85
1.73
56.0
12.69
32.72
(%)
59.56
76.71
Nitrogen
9.33
2.99
1.46
22.32
2.68
18.07
(mg. g"1)
13.46
28.00
Phosphorus
1.86
1.49
0
4.10
1.53
1.51
(mg. g"1)
4.57
6.38

Percent organic matter
Time (year)
Figure 4-1. Lake Bonny percent organic matter sediment core
content versus time (year).

mg/g
Time (year)
Figure 4-2. Lake Bonny TN (mg.g-1) and TP (mg.g"1) sediment core
content versus time (year).

65
Sedimentation rates expressed in Figure 4-3 for Lake
Bonny are in agreement with the lake history. The first
peak of 0.1 g.cm'2.yr coincides with the 1955 development
boom in the area, and the second peak of 0.13 g.cm'2.yr in
1983 coincides with the time when the lake was almost
completely dry.
Detectable 137Cs, which is an human-produced
radionuclide that was injected into the atmosphere as a
consequence of nuclear weapons testing in the early 1950s,
matched well with the determined 210Pb chronology (Table 4-
2). This agreement is remarkable, considering the
shallowness of the lake and the flocculence of the bottom
substrate which makes these upper sediments vulnerable to
physical disturbance. Eutrophic systems such as Lake Bonny,
however, accumulate sediments rapidly, so these physical
disturbances likely affect short time-intervals only.
The Lake Gibson Record
The sediment core from Lake Gibson had a mean bulk
density of 0.279 g.cm'3 0.319, mean dry matter of 21.61%
13.93, mean organic matter of 38.17% 14.85, mean total
nitrogen of 9.33 mg.g'1 2.99, and mean total phosphorus of
1.86 mg.g'1 1.49. (Table 4-1).
The organic content of Lake Gibson sediments was
somewhat low in surface deposits (30% organic matter),
increased with depth to about the midpoint of the core,
matching with 1955, and declined to its lowest values of

g/crrT2/yr
Time (year)
Figure 4-3. Lake Bonny sedimentation rate (g.cirf2.yr)
versus time (year).
a\
CTi

Table 4-2. Lake Bonny sediment analysis.
Core
Section
(cm)
Bulk
Density
g. cm'3
2iopb
pCi. g'1
137Cs
pCi. g'1
Deposition
Period Rate
year g.cr2.yr
TN
mg. g'1
TP
mg. g'1
0-4
0.029
18.72
1.5
1993
0.05
28.00
5.16
4-7
0.050
12.30
3.0
1991
0.08
23.11
5.29
7-9
0.061
12.50
4.13
1989
0.07
23.62
5.35
9-11
0.066
11.22
4.01
1987
0.07
27.15
6.03
11 13
0.074
8.46
3.65
1985
0.09
20.02
4.62
13 15
0.078
8.14
3.20
1983
0.09
20.45
4.81
15 17
0.079
5.27
4.02
1982
0.13
22.26
5.21
17 19
0.090
6.99
4.56
1980
0.10
21.58
3.53
19 22
0.092
8.92
4.49
1978
0.07
19.15
4.93
22 26
0.092
6.51
3.28
1974
0.08
20.12
5.86
26 30
0.090
7.00
2.67
1970
0.07
21.72
6.38
30 34
0.089
6.00
1.07
1963
0.07
23.63
5.07
34 38
0.091
3.80
0.85
1958
0.09
22.85
3.64
38 42
0.102
3.43
0.92
1953
0.08
24.67
3.68
42 46
0.164
3.48
0.87
1948
0.07
19.9
3.03
46 50
0.174
3.17
0.78
1937
0.06
18.07
2.42
50 54
0.142
1.91
0.91
1921
0.06
20.59
2.49
54 58
0.137
1.94
0.59
1909
0.04
20.36
2.22

68
1.4% organic matter at the bottom of the core (approximately
1892) (Figure 4-4). Total nitrogen and total phosphorus
displayed high variability. Total nitrogen peaked about
1942, declined after that and peaked again in 1990. Total
phosphorus started with low values (1.0 mg.g'1) about 1892,
increased steadily to peak after 1966, and, from that point
on, kept increasing to a 1990 value of 5.0 mg.g-1 (Figure
4-5) .
Total P reconstruction at the base of the core suggests
that this lake, on the basis of phosphorus concentrations,
has been oligotrophic in the past. Nevertheless, values for
total P in excess of 3.0 mg.g-1 and greater were recorded
for all levels corresponding to the last 25 years (since
1966), reflecting changes in watershed use.
Bulk density increased below the 24 cm core, which
matched with a 210Pb determined sediment age of
approximately 1950. Detectable 137Cs coincided with the
determined 210Pb chronology (Table 4-3). The sedimentation
rate for Lake Gibson corresponded with the lake's history. A
significative peak of 0.12 g.cm_2.yr coincided with 1955,
the beginning of the development boom for the area (Figure
4-6) .
The organic matter content of dry bulk sediment in
Florida lacustrine surface mud averaged 39.7% from 97
surveyed lakes (Brenner and Binford, 1988). Lake Gibson had
a percent organic matter mean value of 38.2%, while Lake

70
60
50
40
30
20
10
0
Percent organic matter
92 1914 1923 1935 1942 1946 1951 1955 1966 1973 1979 1986 1990 1993
Time (year)
Figure 4-4
Lake Gibson percent organic matter sediment core
content versus time (year).
o>
3

mg/g
Time (year)
Figure 4-5. Lake Gibson TN (mg.g-1) and TP (mg.g~1) sediment core
content versus time (year).

Table 4-3. Lake Gibson sediment analysis
Core
Section
(cm)
Bulk
Density
g. cm'3
210pb
pCi. g'1
137Cs
pCi. g'1
Deposition
Period Rate
year g.cr2.yr
TN
mg. g-1
TP
mg. g'1
0-2
0.113
11.86
4.41
1993
0.07
10.50
4.45
2-4
0.167
9.48
5.02
1990
0.08
10.56
4.57
4-6
0.221
9.31
4.18
1986
0.08
9.11
4.10
6-8
0.233
6.76
3.63
1979
0.08
8.46
3.77
8-10
0.271
5.72
2.96
1973
0.08
7.83
3.28
10 12
0.284
6.07
2.96
1966
0.06
7.49
3.03
12 14
0.229
2.33
1.32
1955
0.12
10.38
2.07
14 16
0.178
2.63
1.28
1951
0.09
11.00
1.62
16 18
0.168
2.61
0.84
1946
0.08
11.70
1.70
18 20
0.160
3.34
0.75
1942
0.05
13.46
1.55
20 24
0.161
2.29
0.71
1935
0.06
13.21
1.50
24 28
0.185
1.11
0.75
1923
0.09
12.22
1.28
28 32
0.208
1.42
0.57
1914
0.05
7.26
0.70
32 36
0.183
2.00
0.30
1892
0.02
9.98
0.89

g/crrT2/yr
1892 1914 1923 1935 1942 1946 1951 1955 1966 1973 1979 1986 1990 1993
Time (year)
Figure 4-6. Lake Gibson sedimentation rate (g.cm'2.yr)
versus time (year).
-j
N>

73
Bonny had 56.0%. Total nitrogen for the same array of lakes
displayed high variability, ranging from 0.6 to 42.4 mg.g"1
(Brenner and Binford, 1988). Nitrogen concentration for
Lake Gibson was 9.33 mg.g"1 (Table 4-4), and for Lake Bonny
was 22.32 mg.g"1 (Table 4-5). Total P concentrations
recorded by Brenner and Binford (1988) varied from 0.07 to
8.09 mg.g"1. Values for Lake Gibson were 1.8 mg.g'1 (Table
4-6) and for Lake Bonny was 4.1 mg.g'1 (Table 4-7).
Considering that these lakes overlie phosphatic
limestone deposits and are located in urban settings, each
one of those isolated factors or a combination of both could
be responsible for the high total P values. The total P
value registered for Lake Bonny of 4.1 mg.g"1 is higher than
what was found in 93 of the 97 systems studied for Brenner
and Binford (1988).
The Lake Parker Record
The Lake Parker sediment core displayed usually
elevated deposition rates for bulk density, organic matter
and phosphorus (Brenner et al., 1993). Bulk sediment
accumulation rates increased from 11 mg.cm"2.yr"1 in 1922 to
a maximum of 131 mg.cm"2.yr"1 in the late 1950s (Brenner et
al., 1993). Phosphorus net accumulation rates increased
more than ten times during the last century, reaching the
highest value of 600 pg.cm"2.yr"1 in the late 1970s (Brenner
et al., 1993). Appendices D and E show physical and chemical
properties of Lake Parker sediment core.

74
Table 4-4. Total nitrogen (mg.g-1)
Lake Gibson sediment core.
| Interval (cm)
TN (mg/g)
0-2
10.4995
2-4
10.5649
4-6
9.1080
6-8
8.4608
8-10
7.8340
10-12
7.4951
12-14
10.3830
14-16
11.0008
16-18
11.6984
18-20
13.4574
20-24
13.2149
24-28
12.2212
28-32
7.2597
32-36
9.9815
36-40
10.4658
40-44
10.6539
44-48
7.2012
48-52
3.3171
52-56
1.4618
1 56-60
10.4134
| Mean
9.33
for

Table 4-5. Total nitrogen (mg.g-1
Lake Bonny sediment core.
| Interval (cm)
TN (mg/g)
I 0-2
28.0057
I 2-4
26.5107
4-7
23.1092
7-9
23.6192
9-11
27.1484
11-13
20.0185
1 13-15
20.4545
15-17
22.2599
17-19
21.5829
19-22
19.1497
22-26
20.1250
26-30
21.7168
30-34
23.6266
34-38
22.8466
38-42
24.6686
42-46
19.9064
I 46-50
18.0703
| 50-54
20.5948
54-58
20.3690
58-64
24.4444
64-72
19.5555
72-84
23.2875
| Mean
22.32
for

76
Table 4-6. Total phosphorus (mg.g-1) for
Lake Gibson sediment core.
Interval (cm)
TP (mg/g) |
0-2
4.4495
2-4
4.5700
4-6
4.1031
6-8
3.7696
8-10
3.2824
10-12
3.0312
12-14
2.0766
| 14-16
1.6263
I 16-18
1.6984
18-20
1.5514
20-24
1.5047
24-28
1.2811
28-32
0.7041
32-36
0.8948
36-40
0.9506
40-44
0.9809
44-48
0.6574
48-52
0.1171
52-56
0.0097
56-60
0.0000
Mean 1.8

Table 4-7. Total phosphorus (mg.g"1) for
Lake Bonny sediment core.
77
I Interval (cm)
TP (mg/g)
I 0-2
5.1567
I 2-4
5.2729
4-7
5.2941
7-9
5.3534
9-11
6.0308
I 11-13
4.6168
I 13-15
4.8145
I 15-17
5.2071
I 17-19
3.5261
I 19-22
4.9261
1 22-26
5.8654
26-30
6.3821
30-34
5.0733
34-38
3.6455
38-42
3.6842
42-46
3.0309
46-50
2.4239
50-54
2.4907
j 54-58
2.2233
I 58-64
2.0648
[ 64-72
1.5072
I 72-84
1.5842
H Mean 4.1
= a BBS J1

78
The Lake Hollingsworth Record
Inferences from paleolimnological core studies indicate
that Lake Hollingsworth has had high nutrient and
chlorophyll a concentrations since the 19th century;
however, the highest levels occurred between the 1950s and
the 1970s, when much of the agriculture in the watershed was
converted to urban development (Schelske et al., 1992). Lake
sediments' topmost 10 cm is poorly consolidated (densities
of <0.032 g.dry.cm-3.wet) (Schelske et al., 1992), although
density generally increases with depth and is highest in
sand-rich deposits.
Organic matter in the core generally ranges between 40
and 60% of dry weight but decreases in sandy deposits
(Schelske et al., 1992). Total phosphorus concentrations
have remained relatively constant during the last ten years,
ranging from 59 to 71 pg.L'1 (Schelske et al., 1992).
Appendix F show physical and chemical properties of Lake
Hollingsworth sediment core.
Conclusions
The data presented provide support for the conclusion
of Flannery et al.(1982) that Florida lakes of higher
trophic state have greater proportions of organic matter in
their surface sediments e.g. Lake Bonny (TSI of 73.46) with
57% organic matter and Lake Gibson (TSI of 55.18) with 35%
organic matter. Organic matter in the Lake Parker

79
superficial layers was >70.0%, and in Lake Hollingsworth was
>51.0% (Schelske et al., 1992). Lake Bonny organic matter is
positively correlated with depth and shows no correlation
with any other parameter. Bulk density was positively
correlated with depth and negatively correlated with total
nitrogen and total phosphorus at the 5% level confidence
(SAS, 1989) (Table 4-8).
Organic matter in Lake Gibson was negatively correlated
to bulk density and showed a positive correlation to total
nitrogen. Bulk density was positively correlated with depth,
and depth was negatively correlated to total phosphorus at
the 5% level confidence (SAS, 1989) (Table 4-9).
Using the reconstruction of historical limnological
conditions, paleolimnology may help address issues of lake
management (Smeltzer and Swain, 1985) and restoration.
Sedimentary records may reflect past trophic status of a
lake and, thereby, assist in setting a goal for restoration.
Models may then be employed to identify primary nutrient
sources and to foresee whether mitigation actions can reduce
loading sufficiently to improve water quality or if other
management techniques such as biomanipulation may be needed.
When reduction of nutrient loading to a lake is
feasible, a cost/benefit evaluation should be completed to
ascertain which restoration techniques are financially
practical. Then, paleolimnological records can help in the
decision making process.

80
Table 4-8. Correlation matrix for sediment variables
for Lake Bonny.
SEDIMENT VARIABLES
CORRELATION
P VALUE
| Depth/ Bulk Density
0.7848
0.0000
| Depth/ Dry Matter
0.8255
0.0000
Depth/ Organic Matter
0.5482
0.0123
Depth/ Total Phosphorus
-0.8135
0.0000
| Bulk Density/T Nitrogen
-0.5724
0.0084
| Bulk Density/T Phosphorus
-0.6747
0.0011
Table 4-9. Correlation matrix for sediment variable
for Lake Gibson.
SEDIMENT VARIABLES
CORRELATION
P VALUE
| Depth/ Bulk Density
0.5862
0.0066
Depth/ Total Phosphorus
-0.9054
0.0000
Bulk Density/ Org. Matter
-0.7335
0.0002
Org. Matter/T Nitrogen
0.6123
0.0041

81
Historical water-column total P data collected for the
last four years at Lake Bonny showed little variation, with
the exception of Spring 1991, when the values were double
the mean (mean = 0.112 mg.I'1). An increase in chlorophyll a
and total bacteria were also reported for the same period.
This information is in disagreement with Maceina and Soballe
(1990), who suggested that resuspension during wind events
is probably responsible for the highly variable total P and
chlorophyll a values measured in many shallow Florida lakes.
Lake Gibson historical records for total P showed no
significant variation for the last four years, with a mean
of 0.210 mg.l-1. Chlorophyll a, TSI and total bacteria also
displayed no notable variations. Anthropogenic alterations
in the watershed, especially increased urbanization, may be
responsible for the presence of only one genus of blue-green
algae in the lake, as well as for the occasional algal
bloom.
The Lake Parker historical record for net phosphorus
accumulation rates greatly exceed dangerous P loadings to a
basin of <5 m mean depth (13 pg. cm'2.yr'1) (Brenner et al.,
1993). Total phosphorus measured in the water column during
this study had a mean of 0.22 rag.L"1 (0.07 to 0.602
mg. L'1) .
Lake Hollingsworth historical total phosphorus
concentration show an overall increase upwards in the
sediment core (Schelske et al.,
1992). Total P values

82
changed from 2.03 mg.g'1 at the base of the core to 6.60
mg.g'1 at the top (Schelske et al., 1992). Total phosphorus
in the water column measured during this study had a mean of
0.27 rng.L"1 (0.081 to 0.61 mg.L"1)
Due to the geology in the area, Lake Bonny has been an
eutrophic lake for much of its recent history, even with
conditions of minimal human occupation in the watershed
(Figure 4-7). This should be the foundation for any
management plan for this lake. Any plans for water quality
improvement in the lake must consider that the lake is
edaphically phosphorus-rich, and even a great effort to
reduce phosphorus concentrations in the water may provide
little or unnoticeable improvement in water clarity.
Lake Gibson, meanwhile, had a different
paleolimnological record that displayed alterations during
the last 25 years (Figure 4-8) in which the lake changed
from mesotrophic to eutrophic, reflecting the switch in past
largely agricultural watershed uses to large scale
commercial and moderate density residential development. Any
attempt to reduce nutrient concentration in this lake must
consider control of point source external loading of
nutrients as well as use of biomanipulation techniques as a
means to achieve water quality improvement.
Figure 4-9 show the historic variation of organic
matter content, nitrogen and phosphorus in the Lake
Hollingsworth sediment core cf. Schelske et al. 1992. It is

Time (year)
+~ % Dry matter % Organic matter
Nitrogen (mg/g) ^ Phosphorus (mg/g)
Figure 4-7. Changes in some physicochemical characteristics in
Lake Bonny through time.
00
u>

Time (year)
+- % Dry matter % Organic matter
Nitrogen (mg/g) Phosphorus (mg/g)
Figure 4-8. Changes in some physicochemical characteristics in
Lake Gibson through time.

Organic matter (%) l Nitrogen (%) Phosphorus (mg/g)
Figure 4-9. Changes in some physicochemical characteristics in
Lake Hollingsworth through time (after Shelske et al.,1992).
00
in

86
noted that then was an increase in phosphorus concentration
after the 1950s, matching with the period of more intense
occupation in the watershed for the whole area. Water
quality measurements collected since 1966 demonstrate that
the lake has been hypereutrophic for at least 25 years
(Schelske et al., 1992). Although the lake is naturally
productive, anthropogenic impacts have accelerated the rate
of eutrophication. Eutrophication problems, especially from
cultural enrichment, should be resolved in a number of ways,
ranging from eutrophication prevention and lake
rehabilitation, to learning to live with the problem.
From the sediment core analysis of Lakes Parker and
Hollingsworth, lakes considered as having a dominant blue
tilapia population in this study, and Lakes Bonny and
Gibson, considered as having a dominant gizzard shad
population, it is not clear that there are any significant
differences in sediment organic matter content, nitrogen,
phosphorus or any other analyzed parameter. All four lakes
seem to have high organic matter content in the topmost
layers, and all of them had a record in the sediments of the
main environmental changes for the area e.g. the 1950's
switch from totally agricultural practices in the area to
watershed conversion to urban development, as well as the
efforts been made to recover these systems since the mid
1980's.

87
Sediment cores from Lakes Bonny and Gibson collected
during this study showed no noticeable changes relative to
the improvement of any water quality parameter. However,
Lakes Parker and Hollingsworth had what perhaps could be
called a modification. According to Schelske et al. (1992),
total phosphorus reconstructions in Lake Parker's topmost 15
cm (after 1960s) of sediment suggested a progressive decline
in the water-column that may signal some reversal of
cultural eutrophication. The Lake Hollingsworth core
revealed that, for at least in the last ten years, total
phosphorus inferences have remained relatively constant
(Schelske et al., 1992).
Taking in consideration that blue tilapia came into the
systems after 1961, it is reasonable to assume, first of
all, a lag time for the establishment of the fish population
in the new environment, and, secondly, time for the
manifestation of any modification that those fishes would
bring to the systems. Since more than thirty years have
passed since fish introduction, it is possible to
hypothesize that this could be responsible for changes
recorded in Lakes Parker and Hollingsworth water quality,
even though this might still provide a weak indication of
water quality improvement. However, considering that all
lakes included in this study are subject basically to the
same kind of impacts, as well as the lake recovery program
carried out by the City of Lakeland, the explanation for

88
this slight noticeable improvement and/or stabilization
process in the water measured parameters in those lakes may
be due to the either an altered biological community and/or
the modifications resulting from the introduced species.

CHAPTER 5
LIMNOLOGICAL ASSESSMENT
Introduction
Florida can be divided into three major physiographic
zones: the Northern, Central and Southern zones (Puri and
Vernon, 1964; White, 1970). All lakes reported in this study
are located in Polk County, Florida. Polk County lies within
the Central Highlands physiographic province. The vast
majority of the county lies within the Polk and Lake
Uplands. The Polk Uplands is an area of continuous high
ground located between the Gulf Coastal Lowlands and the
western edge of the Lake Wales Ridge (White, 1970). Along
the eastern ridge, deposits of the Fort Preston Formation
occur, but most of the region is underlain by deposits of
the phosphatic Hawthorn and Bone Valley Formations (Puri and
Vernon, 1964). In the eastern half of the Polk Uplands, most
lakes are in association with sand ridges. Polk County has
550 lakes covering 37.9 x 103 ha.
A majority of phosphatic sand deposits and clays of the
Miocene Bone Valley and Hawthorne Formations are located in
the Polk Uplands and Lake Uplands physiographic regions
(Puri and Vernon, 1964). Florida has led the nation in
89

90
phosphate production for over 90 years; the bulk of this
production is from Polk County (Boyle and Hendry, 1985). The
trophic status for lakes of this region is mostly
mesotrophic or eutrophic, and waters are moderately
organically stained (Canfield, 1981).
Water quality on the Polk Upland is highly variable.
Based on data from Florida Game and Fresh Water Fish
Commission, Canfield (1981), Polk County Water Resources
Division and this study, mean pH ranged from 5.4 to 9.4 and
total alkalinity concentrations averaged between 1 and 76
mg.l'1 as CaC03. Total hardness concentration averaged
between 14 and 195 mg.l'1 as CaC03, and mean specific
conductance ranged from 82 to 281 pmhos.cm'1. Calcium was
the dominant cation and bicarbonate the dominant anion in
Lake Parker, whereas in most of the other lakes in the area,
sodium and chloride were the dominant anions. Total nitrogen
concentration was between 306 and 1566 mg.m'3, mean total
phosphorus concentration from 2.4 to 144 mg.m'3, and Secchi
disk depth 0.2 to 2.7 m. The wide range of water chemistry
may be directly related to different regional geology
(Canfield, 1981; Crisman, 1993).
The following agencies have conducted studies or
developed information on the lakes in Polk County: United
States Geological Survey (USGS); Environmental Protection
Agency (EPA); Polk County Water Resources Division (PCWR);
City of Lakeland (CL); Florida Game and Freshwater Fish

91
Commission (FGFWFC); Southwest Florida Water Management
District (SWFWMD); Central Florida Regional Planning Council
(CF208).
Methods
I selected six lakes in Polk County, Florida: Lakes
Parker, Hollingsworth, Hancock, Gibson, Bonny and Hunter.
For this study all lakes were sampled quarterly during the
years of 1992 and 1993 following approved U.S. EPA
methodology. Additional data for the years of 1989 to 1991
were obtained from the above listed agencies.
Three water quality sampling stations were established
for Lakes Parker, Hollingsworth and Hancock, and two
stations were established for Lakes Gibson, Bonny and
Hunter. Water was collected from just bellow the surface
(0.5 m) in acid-cleaned Nalgene bottles. Samples were placed
on ice and returned to the laboratory for analysis. Water
temperature (C), dissolved oxygen (mg/L), pH and
conductance (pS/cm @ 25C) were measured by using a Hydrolab
Station. Secchi depth (m) was measured at each station where
water was collected.
At the laboratory, total alkalinity (mg/L as CaC03) was
determined by titration with 0.02 N H2S04 (APHA, 1985).
Total phosphorus concentrations (mg/L) were determined using
the methods of Murphy and Riley (1962) after persulfate
oxidation (Menzel and Corwin, 1965). Total nitrogen

92
concentrations (mg/L) were determined by a modified Kjeldahl
technique (Nelson and Sommers, 1975). Total suspended solids
(mg/L), organic suspended solids (mg/L) and inorganic
suspended solids (mg/L) were determined according to
standard methods (APHA, 1985). Water samples were analyzed
for color (Pt-Co units) using the platinum-cobalt method and
matched Nessler tubes (APHA, 1985). (The results of these
laboratory analyses were provided by the aforementioned
agencies).
Total chlorophyll a concentrations (pg/L) were
determined by filtering a measured portion of lake water
through a Gelman type A-E glass filter. Chlorophyll a was
determined by using the method of Yentsch and Menzel (1963)
and the equations of Parson and Strickland (1963). Primary
productivity (mgC.m'3.h) was measured using the light and
dark bottle method according to Wetzel and Likens (1991).
Phytoplankton samples for qualitative analysis were
collected quarterly at one station in each lake during the
1992/93 survey. Samples were collected at mid-Secchi depth
using a Van Dorn water bottle, and then dispensed into dark,
polypropylene bottles containing Lugol's solution (APHA,
1985). At the laboratory, samples were refrigerated in the
dark until examined. Zooplankton were sampled quarterly at
one station in each lake during the 1992/93 survey. Samples
were collected with an 8.0 L Kemmerer bottle and filtered
with a 28 pm net, and were preserved with 10 percent sugar-

93
formalin (Haney and Hall, 1973) containing rose bengal
stain. Obligue tows were started as close to the bottom as
possible, and the sampler was raised to the surface at a
slow, constant rate; this procedure was repeated four times,
filtering up to 32 L of water. Fish populations were
estimated from data of Canfield and Hoyer (1992) and Florida
Game and Freshwater Fish Commission (FGFWFC).
Description of Study Area
Lake Parker
Lake Parker is located in Polk County, Florida
(Appendix A). The basin lies in the Polk Uplands
physiographic region and is situated in phosphatic deposits
from the Hawthorn and Bone Valley Formations. Due to its
location just east of the Lakeland Ridge, the lake receives
inputs of groundwater which have been in contact with
limestone (Stewart, 1966). Lake Parker is a large urban lake
with a surface area of 924 ha, shoreline length of 19.8 km
and mean depth of 2.2 m (Table 5-1).

94
Table 5-1. Morphometry
of Lake Parkerci)
Surface Area
924 ha
Maximum Depth
3.0 m
Mean Depth
2.2 m
Development of Shoreline 70 %
Drainage Basin Area
6.18 x 10J ha
Shoreline Length
19.8 km
Macrophyte Cover
5.0 %
(1) Data from PCWR (1992).
The drainage area north of the lake receives water from
Lake Gibson and a sinkhole basin on the west side. Overflow
from Lake Mirror also enters Lake Parker. The lake also
receives groundwater inputs. FMC Corporation discharges
effluent to the lake by a drainage ditch and the City of
Lakeland Power Plant withdraws and discharges water to the
lake. Lake Parker has an outflow to Saddle Creek through a
small canal and a water level control structure at the
outlet.
Land-use in the basin is dominated by a large
commercial and residential area. The entire northern section
of the Lake Parker watershed has been impacted by phosphate
mining. Dominant macrophytic plant species (defined as
greater than or egual to 5% macrophyte cover cf. PCWR, 1992)
include cattail, hydrilla, and water naiad. Macrophyte
chemical control is employed for water hyacinths, water
lettuce, and hydrilla.

95
Using the criteria of Forsberg and Ryding (1980), Lake
Parker was hypereutrophic during this study. Canfield (1981)
also classified Lake Parker as hypereutrophic. It is
suggested that the trophic status of this lake has remained
relatively stable for at least the last decade. Table 5-2
shows water quality data for Lake Parker.
Phytoplankton primary productivity was measured in
October, December, April and July at surface, 50cm and 100
cm depths each time, using the light and dark bottle method.
The results are expressed as gross photosynthesis
(mgC.m'3.h) cf. Wetzel and Likens (1991) (Table 5-3). Values
for net photosynthesis are presented in table 5-4, community
respiration in table 5-5, and table 5-6 shows the average
annual productivity values for all study lakes. Lake Parker
does not display great seasonality in primary productivity.
The highest values were always found at the 50 cm depth,
with a mean for that depth of 154.87 mgC.m"3.h 42.51
throughout the year. A minimum of 103.86 mgC.'3.h for April
and a maximum of 203.12 mgC.m'3.h for December were recorded
at the 50 cm depth. The lowest value recorded for the lake
was 58.93 mgC.nf3.h at surface during July, while the
maximum recorded for the lake was 203.12 mgC.nf3.h at 50 cm
during December.
The phytoplankton community in Lake Parker was
represented by blue-green algae including Anabaena sp.,
Microcystis sp., Merismopaedia sp., and Aphanizomenon;

96
Table 5-2. Summary of water quality data for Lake Parker.
Numbers shown are mean values with ranges in parentheses.
Parameter
Unit
Canfield
(1981)
Several
Agencies *
(1989/91)
This study
(1992/93)
Temperature
C
N/A
24.95
(16.4-31.8)
24.81
(17.7-30.2)
D.O.
mg/L
N/A
8.5
(4.7-13.2)
9.1
(8.1-10.3)
PH
8.9
(8.8-9.1)
8.9
(5.4-10.9)
8.5
(7.5-9.3)
Alkalinity
rag/L CaC03
68
(63-76)
55
(40-64)
43.5
(34-50)
Conduct.
pmhos/cm
215
(200-260)
387
(207-470)
247
(207-271)
Nitrogen
mg/L
1.56
(0.98-1.97)
3.15
(0.63-11.4)
3.13
(2.19-3.69)
nh3-n
mg/L
N/A
0.06
(0.01-0.19)
0.05
(0.02-0.07)
TKN
mg/L
N/A
2.9
(0.6-11.0)
2.8
(1.3-3.2)
Phosphorus
mg/L
0.08
(0.06-0.11)
0.15
(0.02-.34)
0.22
(0.07-0.602)
Chi. a
mg/m3
33.7
(9.3-49.3)
75.7
(18.6-222.8)
53.8
(43.0-76.6)
Color
Pt units
22
(15-30)
24.2
(6-57)
42.6
(29-53)
Secchi
m
0.5
(0.4-0.6)
0.4
(0.25-1.5)
0.35
(0.25-0.4)
Bacteria
x 106
N/A
761.7
(1-7800)
18.7
(1-60)
TSI
N/A
79.6
(62.7-94.0)
82.7
(76.6-86.9)
(*) AGENCY CODES:
USGS; EPA; PCWR; CL; UFC; FGFWFC; CF208; and SWFWMD

97
Table 5-3.
Gross
primary productivity for all study lakes
expressed as (mgC.m .h).
I LAKE
DEPTH
08/92
12/92
04/93
07/93
MEAN
S.DEV
0 cm
64.06
78.12
64.53
58.93
66.41
8.21
y Parker
50 cm
140.62
203.12
103.86
171.87
154.87
42.51
100 cm
70.31
62.51
69.43
78.12
70.09
6.39
0 cm
178.90
437.5
356.78
506.34
369.88
141.2
1 Holl a
50 cm
46.87
31.25
304.65
406.25
197.25
187.4
100 cm
46.87
30.98
146.35
187.5
102.92
76.06
0 cm
356.45
468.75
187.5
625.0
409.42
184.4
A Hanc b
50 cm
187.67
0
0
218.75
101.60
118.0
100 cm
N/A
N/A
N/A
N/A
N/A
N/A
0 cm
346.89
15.62
25.46
468.75
214.18
229.1
Gibson
50 cm
278.26
93.75
31.25
265.6
167.21
123.7
100 cm
156.78
0
43.75
178.46
94.74
86.5
0 cm
478.34
453.12
345.12
515.6
448.04
73.2
| Bonny
50 cm
156.89
187.5
131.25
234.37
177.50
44.34
100 cm
98.14
93.75
46.87
101.56
85.08
25.67
0 cm
303.85
109.37
289.56
215.47
229.56
89.0
| Hunter
50 cm
423.74
115.62
484.37
203.12
306.71
175.6
100 cm
205.63
98.45
281.25
218.75
201.02
75.92
(a) Hollingsworth.
(b) Hancock.

98
Table 5-4. Net primary productivity for all study lakes
expressed as (mgC.m'3.h).
LAKE
DEPTH
08/92
12/92
04/93
07/93
MEAN
S.DEV
0 cm
19.22
23.44
19.36
17.68
19.92
2.46
Parker
50 cm
42.18
60.94
31.16
51.56
46.46
12.75
100 cm
21.09
18.75
20.83
23.44
21.03
1.92
0 cm
53.67
131.25
107.03
151.90
110.96
42.37
Holl a
50 cm
14.06
9.37
91.39
121.87
59.17
56.23
100 cm
14.06
9.29
43.90
56.25
30.87
22.82
0 cm
106.93
140.62
56.25
187.5
122.82
55.33
[ Hanc b
50 cm
56.30
0
0
65.62
30.48
35.40
100 cm
N/A
N/A
N/A
N/A
N/A
N/A
0 cm
124.41
5.46
8.91
164.06
75.71
80.78
1 Gibson
50 cm
97.39
32.81
10.94
92.96
58.52
43.29
100 cm
54.87
0
15.31
62.46
33.16
30.26
0 cm
167.42
158.59
120.79
180.46
156.81
25.64
II Bonny
50 cm
54.91
65.62
45.94
82.03
62.12
15.52
100 cm
34.35
32.81
16.40
35.55
29.78
8.99
0 cm
106.35
38.28
101.35
75.41
80.35
31.15
| Hunter
50 cm
148.31
40.47
169.53
71.09
107.35
61.46
100 cm
71.97
34.45
98.44
76.56
70.35
26.58
(a) Hollingsworth.
(b) Hancock.

99
Table 5-5. Community respiration for all study lakes
expressed as (mgC.m'3.h).
LAKE
DEPTH
08/92
12/92
04/93
07/93
MEAN
S.DEV
0 cm
44.84
54.68
45.17
41.25
46.48
5.74
1 Parker
50 cm
98.44
142.18
72.7
120.31
108.41
29.76
n~
100 cm
49.22
43.76
48.6
54.68
49.06
4.47
0 cm
125.23
306.25
249.75
354.44
258.92
98.86
Holl a
50 cm
32.81
21.88
213.26
284.38
138.08
131.2
100 cm
32.81
21.69
102.45
131.25
72.05
53.24
0 cm
249.52
328.13
131.25
437.5
286.6
129.1
Hanc b
50 cm
131.37
0
0
153.13
71.12
82.61
100 cm
N/A
N/A
N/A
N/A
N/A
N/A
0 cm
225.48
10.16
16.55
304.69
139.22
148.9
| Gibson
50 cm
180.87
60.94
20.31
172.64
108.69
80.40
100 cm
101.91
0
28.44
116.0
61.59
56.21
0 cm
310.92
294.53
224.33
335.14
291.23
47.62
Q Bonny
50 cm
101.98
121.88
85.31
152.34
115.38
28.82
100 cm
63.79
60.94
30.47
66.01
55.30
16.68
0 cm
197.5
71.09
188.21
140.06
149.21
57.85
Hunter
50 cm
275.43
75.15
314.84
132.03
199.36
114.1
100 cm
133.66
64.00
182.81
142.19
130.66
49.35
(a) Hollingsworth.
(b) Hancock.

Table 5-6. Average annual productivity values (ragC.m'3.h) (+S.Dev.)
at surface (1); 50 cm (2); and 100 cm (3) depths
for all study lakes.
Parameter
Parker
Hollings.
Hancock
Gibson
Bonny
Hunter
(1)
66.41
(8.21)
369.88
(141.2)
409.42
(184.4)
214.18
(229.1)
448.04
(73.2)
229.56
(89.0)
Gross prim, product. (2)
154.87
(42.51)
197.25
(187.4)
101.6
(118.0)
167.21
(123.7)
177.5
(44.34)
306.71
(175.6)
(3)
70.09
(6.39)
102.92
(76.06)
N/A
94.74
(86.5)
85.08
(25.67)
201.02
(75.92)
(1)
19.92
(2.46)
110.96
(42.37)
122.82
(55.33)
75.71
(80.78)
156.81
(25.64)
80.35
(31.15)
Net primary product. (2)
46.46
(12.75)
59.17
(56.23)
30.48
(35.40)
58.52
(43.29)
62.12
(15.52)
107.35
(61.46)
(3)
21.03
(1.92)
30.87
(22.82)
N/A
33.16
(30.26)
29.78
(8.99)
70.35
(26.58)
(1)
46.48
(5.74)
258.92
(98.86)
286.6
(129.10)
139.22
(148.91)
291.23
(47.62)
149.21
(57.85)
Respiration (2)
108.41
(29.76)
138.08
(131.20)
71.12
(82.61)
108.69
(80.40)
115.38
(28.82)
199.36
(114.14)
(3)
49.06
(4.47)
72.05
(53.24)
N/A
61.59
(56.21)
55.30
(16.68)
130.66
(49.35
100

101
greens such as Scenedesmus sp., Closterium sp., Pediastrum
sp., Cosmarium sp. and Ankistrodesmus; the dinoflagellate
Peridinium sp.; and diatoms of the genus Navicula sp.
A total of 41 zooplankton taxa were identified for Lake
Parker (Kolasa, 1993). Rotifers were the most abundant taxa
(27 out of 41), and comprised almost twice the number of
taxa as all other zooplankton combined. Rotifers comprised
94.7 percent of the zooplankton on an annual basis (Kolasa,
1993). Brachionus sp., Keratella sp., Collurella sp., and
Trichocerca sp., were the most numerically abundant
rotifers. The most abundant adult crustacean zooplankton was
Bosmina lonqirostris, a cladoceran (Kolasa, 1993).
Lake Parker's fish species richness of 21 species
(Canfield and Hoyer, 1992) is expected for a lake of this
size and is similar to that found in hypereutrophic Florida
lakes. Canfield and Hoyer (1992) reported data from nine
0.08 ha blocknets (six placed in littoral habitats and three
in open-water locations), and found that total fish biomass
averaged 71.1 kg.ha'1, with the average of harvestable
sportfish of 31.7 kg.ha"1. The greatest total fish biomass
was collected in littoral blocknets set in hydrilla and
tapegrass, whereas the lowest biomass was harvested in open-
water blocknets. The dominant fish species collected in
littoral nets were largemouth bass and bluegill, and the
dominant fish species collected in open-water nets were
gizzard shad and threadfin shad (Canfield and Hoyer, 1992).

102
ANOVAs were performed using the General Linear Model
(GLM) (SAS, 1989) in order to identify significant seasonal
limnological factors (p < 0.05). The following parameters
were analyzed: Secchi, temperature, pH, conductivity,
dissolved oxygen, total suspended solids, turbidity,
chlorophyll a, color, alkalinity, NH3, TKN, total nitrogen,
total phosphorus, orthophosphate, and total bacteria.
The data were collected between 1989 and 1993 and
grouped in four seasons: winter (December to February);
spring (March to May); summer (June to August); and fall
(September to November). The analyses of the data were
according to the year of collection, season and the
combination of year and season (Table 5-7).
Temperature, conductivity, dissolved oxygen, turbidity,
and total bacteria are the parameters that exhibited
significant seasonal variation at the 95% confidence level.
The other parameters' variability displayed either
interannual variations or no variation at all. When year and
seasonality were combined, Secchi, conductivity, turbidity,
NH3 and total bacteria are the parameters that displayed
significant variability.
Lake Hollingsworth
Lake Hollingsworth is located in Polk County, Florida
(Appendix A). The basin lies in the Bartow Embayment
division of the Central Lakes District (Canfield, 1981),

Table 5-7. Correlation analysis of water quality data for Lake Parker
using the General Linear Model procedure (GLM). Data were combined by
year, season and year*season.
PARAMETER
YEAR
SEASON
YEAR*SEASON
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
Secchi
X
X
X
Temperature
X
X
X
PH
X
X
X
Conductivity
X
X
X
D. 0.
X
X
X
TSS
X
X
X
Turbidity
X
X
X
Chlorophyll a
X
X
X
Color
X
X
X
Alkalinity
X
X
X
nh3
X
X
X
TKN
X
X
X
TN
X
X
X
TP
X
X
X
ORT
X
X
X
BACTERIA
X
X
X
103

104
where the bedrock is dominated by phosphatic deposits from
the Hawthorn and Bone Valley Formation. Lake Hollingsworth
is an urban lake situated within Lakeland, Florida, with a
surface area of 145 ha, shoreline length of 4.34 km and mean
depth of 1.5 m (Table 5-8).
able 5-8. Morphometry of Lake Hollingswortha:
Surface Area
145 ha
Maximum Depth
2.44
m
Mean Depth
1.13
m
Development of Shoreline
100 %
Drainage Basin Area
495 ha
Shoreline length
4.34
km
Macrophyte Cover
3 %
(1) Data from PCWR (1992).
No known discharges or withdrawals were reported for
Lake Hollingsworth. Its drainage basin is totally developed,
and land-use within the basin includes residential and
commercial development and Florida Southern College.
The dominant macrophytic vegetation (defined as greater than
or egual to 3% macrophyte cover cf. PCWR, 1992) includes
American lotus, elephant ear, and cattails. Chemical control
measures are employed for water hyacinth, water lettuce, and
hydrilla.
Using the Forsberg and Ryding (1980) classification,
Lake Hollingsworth was classified as hypereutrophic during
this study. The lake was also classified as hypereutrophic

105
by Canfield and Hoyer (1992). Past limnological data
indicate that this lake has been hypereutrophic for at least
the last 25 years (PCWR, 1992). Table 5-9 shows water
quality data for Lake Hollingsworth.
Lake Hollingsworth displayed seasonality for its gross
primary productivity; the highest values were recorded in
late fall and mid-summer. Overall, the surface layer had the
highest values, with an annual mean of 369.88 mgC.'3.h
141.2; the minimum recorded value was 178.9 mgC.-3.h, and
the maximum was 506.34 mgC.m~3.h (Table 5-3). Net
photosynthesis is reported in table 5-4, community
respiration in table 5-5, and table 5-6 shows the average
annual productivity values for all study lakes.
Historical phytoplankton data for Lake Hollingsworth
from 1968 and 1971 (USGS, 1968, 1971) show the Cyanophyta as
the dominant group, especially species of Oscillatoria sp.,
Lynqbya sp., and Raphidiopsis sp., followed by Chlorophyta
and Diatoms. During this study, the phytoplankton community
was almost completely dominated by the green algae
Ankistrodesmus sp. and Cosmarium sp. and the blue green alga
Spirulina sp. The latter was replaced sometimes by
Aphanizomenon sp. and Anabaena sp. The diatoms were
represented by the genus Navicula.
Rotifers were the dominant zooplankton group throughout
the year. Keratella sp., Brachionus sp., and Monostvla sp.,
were the dominant taxa. The only adult copepod identified

106
Table 5-9. Summary of water quality data for L. Hollingsworth.
Numbers shown are mean values with ranges in parentheses.
| Parameter
Unit
Canfield
(1981)
Several
Agencies *
(1989/91)
This study
(1992/93)
| Temperature
C
N/A
25.8
(15.5-33.3)
25.3
(18.5-30.9)
D.O.
mg/L
N/A
9.7
(4.8-12.8)
9.4
(6.7-11.0)
PH
9.4
(8.8-10.1)
9.4
(7.8-11.3)
8.7
(7.8-10.43)
Alkalinity
mg/L CaC03
47
(32-56)
52.4
(38-66)
47.2
(40-58)
1 Conduct.
pmhos/cm
183
(165-205)
189.8
(142-230)
181.5
(158-209)
Nitrogen
mg/L
N/A
4.04
(0.75-12.2)
5.41
(2.8-8.1)
1 nh3-n
mg/L
N/A
0.07
(0.006-0.3)
0.167
(0.036-0.7)
I TKN
mg/L
N/A
3.92
(0.75-11.6)
5.15
(2.35-7.9)
| Phosphorus
mg/L
0.774
(0.32-1.2)
0.254
(0.015-2.2)
0.268
(0.081-0.61)
| Chi. a
mg/m3
54.8
(26.7-104)
182.59
(30.1-368.1)
186.32
(64.5-285.9)
| Color
Pt units
5
(5-5)
34.7
(6-89)
58.5
(37-94)
I Secchi
m
1.6
(0.8-2.4)
0.3
(0.2-0.4)
0.22
(0.15-0.4)
1 Bacteria
x 106
N/A
2417.15
(1-17200)
800
(40-3700)
1 TSI
N/A
89.46
(70.3-103.1)
98.27
(88.2-105.6)
(*) AGENCY CODES:
USGS; EPA; PCWR; CL; UFC; FGFWFC; CF208; and SWFWMD.

107
was of the genus Diaptomus, but it did not represent a large
component of the zooplankton population. Rotifers had an
estimated density of 286,000 individuals .nf3, and
cladocerans amounted to 68,700 individuals .m-3 (Canfield and
Hoyer, 1992).
Sixteen species of fish were collected from Lake
Hollingsworth (Canfield and Hoyer, 1992). The fish
population from 1965 to 1969 for this lake was described as
having an overabundance of forage fish such as gizzard shad,
threadfin shad, and stunted bluegill (Ware et al., 1971),
and total fish biomass estimated with littoral blocknets
ranged from 70 to 257 kg.ha-1 (Ware et al., 1971). The
littoral nets in the Canfield and Hoyer report of 1992
averaged 1050 kg.ha-1. The most abundant species collected
by open-water experimental gillnets were gizzard shad and
black crappie (Canfield and Hoyer, 1992).
ANOVAs were performed using the General Linear Model
(GLM) (SAS, 1989) in order to identify significant seasonal
limnological factors (p < 0.05). The following parameters
were analyzed: Secchi, temperature, pH, conductivity,
dissolved oxygen, total suspended solids, turbidity,
chlorophyll a, color, alkalinity, NH3, TKN, total nitrogen,
total phosphorus, orthophosphate, and total bacteria.
Secchi, temperature, conductivity, turbidity,
chlorophyll a, color, total phosphorus and total bacteria
are the parameters that exhibited significant seasonal

108
variation at the 95% confidence level (Table 5-10). Other
parameters displayed either interannual variations or no
variation at all. When year and seasonality were combined,
conductivity, turbidity, color, alkalinity and total
bacteria were the parameters displaying significant
variability.
Lake Hancock
Lake Hancock is located in Polk County, Florida
(Appendix A). The basin lies in the Polk and Lake Uplands
physiographic region, and it is situated in phosphatic
deposits from the Hawthorne and Bone Valley Formations. Lake
Hancock is a large lake with a surface area of 1839 ha,
shoreline length of 16.7 km, and mean depth of 0.85 m (Table
5-11).
Table 5-11. Morphometry of
Lake Hancockd)
Surface Area
1839 ha
Maximum Depth
1.5 m
Mean Depth
0.85 m
Development of Shoreline
5 %
Drainage Basin Area
34.1 x 103 ha
Shoreline Length
16.7 km
Macrophyte Cover
1 %
(1) Data from PCWR (1992).
The lake has historically received wastewater
discharges via Banana Lake from the City of Lakeland, and
via Lake Lena Run from the City of Auburndale. Both have

Table 5-10 Correlation analyses of water quality data for Lake Hollingsworth
using the General Linear Model procedure (GLM). Data were combined by
year, season and year*season.
PARAMETER
YEAR
SEASON
YEAR*SEASON
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
Secchi
X
X
X
Temperature
X
X
X
PH
X
X
X
Conductivity
X
X
X
D.O.
X
X
X
TSS
X
X
X
Turbidity
X
X
X
Chlorophyll a
X
X
X
Color
X
X
X
Alkalinity
X
X
X
NH-,
X
X
X
TKN
X
X
X
TN
X
X
X
TP
X
X
X
ORT
X
X
X
BACTERIA
X
X
X
109

110
been eliminated. Known discharges are coming from a Coca-
Cola Citrus processing plant, Adams Citrus processing plant
and Florida Distillers via Lake Lena Run. Agrico Chemical
discharges directly into the lake. There are no known
withdrawals.
Lake Hancock's drainage basin is extremely large and
includes portions of the City of Lakeland (75,000
inhabitants) and Auburndale (18,000 inhabitants). Non-urban
land-use areas is split between mined areas and agricultural
development. The dominant vegetation (defined as greater
than or equal to 1% of macrophyte cover cf. PCWR, 1992)
includes cattails, pickerelweed, and pennywort. Macrophyte
chemical control is employed for water hyacinth, water
lettuce, and hydrilla.
Lake Hancock was reported as an alkaline, eutrophic,
hard water lake by Canfield (1981). Using the criteria of
Forsberg and Ryding (1980), Lake Hancock was classified as
hypereutrophic during this study. Lake water quality has
been improving for at least the last 10 years, although the
lake is still the most eutrophic lake in Polk County (Polk
County Water Resources Division, 1990). Table 5-12 shows
water quality data for Lake Hancock.
Lake Hancock did not display much seasonality for gross
primary productivity. The highest values were always
measured at the surface with an annual mean of 409.42 mgC.m'
3.h + 184.4, minimum value of 187.5 mgC.m'3.h in April and

Ill
Table 5-12. Summary of water quality data for L. Hancock.
Numbers shown are mean values with ranges in parentheses.
Parameter
Unit
Canfield
(1981)
Several
Agencies *
(1989/91)
This study
(1992/93)
Temperature
C
N/A
25.71
(23.1-28.6)
23.65
(23.1-24.2)
1 D'*
mg/L
N/A
7.91
(2.48-10.9)
6.54
(2.48-10.6)
I PH
9.4
(8.8-10.0)
9.06
(7.84-9.66)
9.31
(9.03-9.59)
1 Alkalinity
mg/L CaC03
76
(36-106)
N/A
N/A
1 Conduct.
^/mhos/cm
281
(240-340)
302.8
(232-394)
263
(232-294)
I Nitrogen
mg/L
N/A
7.54
(4.07-12.55)
7.64
(4.07-11.22)
1 nh3-n
mg/L
N/A
0.029
(0.018-0.057)
0.024
(0.018-0.03)
TKN
mg/L
N/A
7.54
(4.06-12.55)
7.64
(4.06-11.22)
I Phosphorus
mg/L
2.44
(1.4-3.2)
0.56
(0.366-0.944)
0.65
(0.37-0.94)
I Chi. a
mg/m3
144
(40.1-217)
184.1
(66-349.5)
221.15
(131.8-310)
B Color
Pt units
28
(25-30)
51.25
(37-75)
60
(45-75)
Secchi
m
0.8
(0.5-1.3)
0.2
(0.1-0.3)
0.25
(0.2-0.3)
(*) AGENCY CODES:
USGS; EPA; PCWR; CL; UFC; FGFWFC; CF208; and SWFWMD.

112
maximum of 625.0 mgC.m'3.h in July (Table 5-3). Values for
net photosynthesis are shown in table 5-4, community
respiration in table 5-5, and average annual productivity in
table 5-6.
During this study, blue-green algae including Spirulina
sp. and Anabaena sp. were the most abundant algae, followed
by Microcystis sp., and Merismopedia sp. A small number of
green algae such as Closterium sp. were also recorded.
The zooplankton community was dominated by rotifers,
including Brachionus sp. and Keratella sp., and Diaptomus
was the only copepod genus reported (Kolasa, 1993).
Fourteen species of fish were reported for Lake
Hancock. Field collections from fifteen-minute
electrofishing events in July 1991 and February 1992,
indicated that blue tilapia was responsible for 71.4 and
52.0 % of total fish weight, respectively, while values for
gizzard shad were 4.4 and 32.0 % by weight for the two
sampling periods. The dominant fish species collected were
blue tilapia, followed by black crappie in the 1991 sample
and gizzard shad in 1992.
ANOVAs were performed using the General Linear Model
(GLM) (SAS, 1989) in order to identify significant seasonal
limnological factors (p < 0.05). The following parameters
were analyzed: Secchi, temperature, pH, conductivity,
dissolved oxygen, total suspended solids, turbidity,
chlorophyll a, color, alkalinity, NH3, TKN, total nitrogen,

113
total phosphorus, orthophosphate, and total bacteria.
Secchi, temperature, conductivity, turbidity, chlorophyll a,
color and total bacteria exhibited significant seasonal
variation at the 95% confidence level (Table 5-13).
Interannual variation or no variation at all was reported
for the other parameters. When year and seasonality were
combined, Secchi, conductivity, turbidity, color and total
bacteria were the parameters displaying significant
variability.
Lake Gibson
Lake Gibson is located in Polk County, Florida
(Appendix A). The basin lies in the Polk Uplands
physiographic region and is situated in sandy deposits of
the Hawthorne and Bone Valley Formations. Lake Gibson is on
the outskirts of the City of Lakeland, with a surface area
of 192 ha, a shoreline length of 6.9 km, and a mean depth of
3.5 m (Table 5-14).
Table 5-14. Morphometry of Lake Gibsoncu
Surface Area 192 ha
Maximum Depth 6.1 ra
Mean Depth 3.5 m
Development of Shoreline 65 %
Drainage Basin Area 1.1 x 103 ha
Shoreline Length 6.9 km
Macrophyte Cover 5 %
(1) Data from PCWR (1992).

Table 5-13. Correlation analyses of water quality data for Lake Hancock
using the General Linear Model procedure (GLM). Data were combined by
year, season, and year*season.
PARAMETER
YEAR
SEASON
YEAR*SEASON
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
Secchi
X
X
X
Temperature
X
X
X
pH
X
X
X
Conductivity
X
X
X
o

o
X
X
X
TSS
X
X
X
Turbidity
X
X
X
Chlorophyll a
X
X
X
Color
X
X
X
Alkalinity
X
X
X
NH^
X
X
X
TKN
X
X
X
TN
X
X
X
TP
X
X
X
ORT
X
X
X
BACTERIA
X
X
X
114

115
Discharge from Padgett Elementary School passes through
a domestic wastewater treatment plant then goes into a
wetland that overflows north of Lake Gibson. No known
outflows were reported. The lake is located in a rapidly
urbanizing area. Citrus groves and pastures are giving way
to large-scale commercial and moderate-density residential
development. Dominant macrophytic vegetation (defined as
greater or equal to 5% macrophyte cover cf. PCWR, 1992)
includes water primrose, alligator weed, and pennywort.
Chemical controls are employed for water hyacinth, water
lettuce, and hydrilla.
During this study, Lake Gibson was classified as a
eutrophic lake according to the classification of Forsberg
and Ryding (1980). Canfield (1981) classified this lake as a
slightly acid, mesotrophic, soft-water lake. The Florida TSI
for this lake is 56 (the lowest for all lakes included in
this study), and it appears that water quality has remained
stable over the past decade. The water is slightly tannic
(color is 43 cpu), which lowers the Secchi disk values and
may bias the TSI value. Table 5-15 shows the water quality
data for Lake Gibson.
Lake Gibson displayed noticeable seasonality in its
phytoplankton gross primary productivity. As has been
reported by other investigators elsewhere in Florida
(Nordlie 1976; McDiffet 1980; Beaver and Crisman 1989),
maximum algal productivity occurred during summer. The

Table 5-15. Summary of water quality data for L. Gibson.
Numbers shown are mean values with ranges in parentheses
116
Parameter
Unit
Canfield
(1981)
Several
Agencies *
(1989/91)
This study
(1992/93)
Temperature
C
N/A
25.03
(16.4-30.7)
24.43
(17.6-29.7)
D. 0.
mg/L
N/A
7.09
(4.02-9.8)
7.69
(6.57-9.0)
PH
7.2
(6.8-7.6)
7.36
(5.9-10.84)
7.15
(6.4-8.12)
Alkalinity
mg/L CaC03
17
(9-21)
19.01
(4-44)
27.33
(22-32)
Conduct.
pmhos/cm
138
(120-160)
162.78
(109-205)
172.66
(148-200)
H Nitrogen
mg/L
N/A
1.206
(0.1-3.71)
1.183
(0.88-1.69)
I nh3-n
mg/L
N/A
0.076
(0.001-0.3)
0.057
(0.016-0.09)
I TKN
mg/L
N/A
1.087
(0.56-3.43)
0.926
(0.35-1.29)
1 Phosphorus
mg/L
0.650
(0.450-1.0)
0.259
(0.063-0.35)
0.183
(0.16-0.206)
1 Chi. a
mg/m3
15.4
(3.2-28.9)
15.37
(2.2-93.1)
8.83
(5.8-12.8)
| Color
Pt units
N/A
66.6
(40-133)
60.83
(37-79)
U Secchi
m
2.2
(1.4-3.1)
0.87
(0.5-1.4)
1.2
(1.1-1.3)
Bacteria
x 106
N/A
287.97
(1-3600)
23.4
(1-60)
I TSI
N/A
58.5
(46.7-79.4)
55.18
(53.06-57.3)
(*) AGENCY CODES:
USGS; EPA; PCWR; CL; UFC; FGFWFC; CF208; and SWFWMD

117
highest values recorded were at the surface; the annual mean
was 214.18 mgC.ra'3.h + 229.1, with the minimum value of
15.62 mgC.m~3.h during winter and 4 68.7 5 mgC.nf3.h during
summer (Table 5-3). Values for net photosynthesis are shown
in table 5-4, community respiration in table 5-5, and
average annual productivity in table 5-6.
The phytoplankton community in Lake Gibson was
dominated by the blue green algal genus Spirulina.
Throughout the year, other blue green genera were
represented including Merismopedia sp. and Lvnqbva sp., as
well as the dinoflagellate Peridinium sp.
The zooplankton assemblage was comprised mainly of
rotifers. Brachionus sp., Keratella sp., and Collurella sp.,
were the most numerically abundant rotifers. The only
copepod was Diaptomus sp.
No recent fish population survey was available for this
lake. Ten species of fish were reported from the last
assessment done in 1979. The dominant species at that time
was channel catfish, followed by blue tilapia, with 63.9 %
and 26.0 % of total fish weight, respectively.
ANOVAs were performed using the General Linear Model
(GLM) (SAS, 1989) in order to identify significant seasonal
limnological factors (p < 0.05). The following parameters
were analyzed: Secchi, temperature, pH, conductivity,
dissolved oxygen, total suspended solids, turbidity,

118
chlorophyll a, color, alkalinity, NH3, TKN, total nitrogen,
total phosphorus, orthophosphate, and total bacteria.
Temperature, conductivity, dissolved oxygen and chlorophyll
a, exhibited significant seasonal variation at the 95%
confidence level (Table 5-16). Other parameters displayed
either significant interannual variations or no variation at
all. When year and season were combined, Secchi,
conductivity and chlorophyll a displayed significant
variability.
Lake Bonny
Lake Bonny is located in Polk County, Florida (Appendix
A). The lake lies in the Bartow Embayment division of the
Central Lake District (Canfield, 1981). Lake Bonny is an
urban lake with a surface area of 144 ha, a shoreline length
of 8.2 km and a mean depth of 2.0 m (Table 5-17).
Table 5-17. Morphometry
of Lake Bonnyd)
Surface Area
144 ha
Maximum Depth
2.5 m
Mean Depth
2.0 m
Development of Shoreline 50 %
Drainage Basin Area
729 ha
Shoreline Length
8.2 km
Macrophyte Cover
10 %
(1) Data from PCWR (1992).
There are no known outflows from Lake Bonny. Land-use
in the basin is a combination of commercial and moderate

Table 5-16. Correlation analyses of water quality data for Lake Gibson
using the General Linear Model procedure (GLM). Data were combined by
year, season and year*season.
PARAMETER
YEAR
SEASON
YEAR*SEASON
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
Secchi
X
X
X
Temperature
X
X
X
PH
X
X
X
Conductivity
X
X
X
D.O.
X
X
X
TSS
X
X
X
Turbidity
X
X
X
Chlorophyll a
X
X
X
Color
X
X
X
Alkalinity
X
X
X
NH,
X
X
X
TKN
X
X
X
TN
X
X
X
TP
X
X
X
ORT
X
X
X
BACTERIA
X
X
X
119

120
density residential development. The most commonly
encountered plants are Typha sp., Ludwiqia repens, and
Pontederia cordata (Canfield and Hoyer, 1992). Macrophyte
chemical control measures are employed for water hyacinth,
water lettuce, and hydrilla.
Lake Bonny was hypereutrophic during this study
according to Forsberg and Ryding (1980). Historical data
indicate that water quality has been improving since
approximately 1986, a time when the lake almost dried
completely. Table 5-18 shows water quality data for Lake
Bonny.
Lake Bonny did not display any seasonality for gross
primary productivity. The only noticeable trend was that
phytoplankton productivity was always greater at the
surface, with mean annual value of 448.04 mgC.m_3.h + 73.2;
the minimum was 345.12 mgC.m'3.h. at April, and the maximum
was 515.6 mgC.m"3.h at July (Table 5-3). Net photosynthesis
is reported in table 5-4, community respiration in table
5-5, and average annual productivity in table 5-6.
Species of blue green algae including Anabaena sp.,
Spirulina sp., Merismopedia sp., and Lynqbya sp. dominated
the phytoplankton assemblage of Lake Bonny. Green algae
(Closterium sp., Cosmarium sp., and Scenedesmus sp.) were
present, as well as the dinoflagellate Peridinium sp., and
the diatom Stauroneis sp.

121
Table 5-18. Summary of water quality data for Lake Bonny.
Numbers shown are mean values with ranges in parentheses.
| Parameter
Unit
Several Agencies*
(1989/91)
This study
(1992/93)
| Temperature
C
24.87 (16.08-30.55)
24.26 (17.48-29.58)
D.O.
mg/L
7.24 (3.85-11.34)
7.47 (6.24-8.63)
PH
8.14 (5.9-9.79)
7.26 (6.28-8.34)
Alkalinity
mg/L CaC03
57.88 (48-84)
47.5 (37-56)
Conduct.
pmhos/cm
2 1 9.7 (170-257)
194.5 (158-219)
Nitrogen
mg/L
2.94 (0.74-7.2)
2.48 (1.5-4.06)
I NH,-N
mg/L
0.0 6 6 (0.001-0.20)
0.079 (0.031-0.13)
I TKN
mg/L
2.83 (0.73-6.2)
2.29 (1.1-3.75)
1 Phosphorus
mg/L
0.14 (0.019-0.36)
0.11 (0.073-0.169)
8 Chi. a
mg/m3
62.9 (18.1-182.5)
44.35 (23.3-94.7)
Color
Pt units
37.5 (13-100)
53.66 (39-67)
1 Secchi
m
0.5 (0.2-1.1)
0.56 (0.3-0.75)
| Bacteria
x 106
1035.9 (1-13400)
165.83 (10-540)
I TSI
7 6.72 (61.84-96.61)
7 3.46 (68.06-86.28)
(*) AGENCY CODES:
USGS; EPA; PCWR; CL; UFC; FGFWFC; CF208; and SWFWMD.

122
The Lake Bonny zooplankton community was dominated by
rotifers and copepods with 274,000 and 85,200
individuals .m"3, respectively (Canfield and Hoyer, 1992).
The dominant rotifers were Brachionus sp. and Keratella sp.
Diaptomus was the most abundant copepod.
Sixteen species of fish were collected in Lake Bonny
(Canfield and Hoyer, 1992). Gizzard shad was the most
abundant fish collected in open-water using experimental
gillnets, followed by Florida gar, with 17.0 and 11.3
fish.net'1.24hr, respectively (Canfield and Hoyer, 1992 ).
ANOVAs were performed using the General Linear Model
(GLM) (SAS, 1989) in order to identify significant seasonal
limnological factors (p < 0.05). The following parameters
were analyzed: Secchi, temperature, pH, conductivity,
dissolved oxygen, total suspended solids, turbidity,
chlorophyll a, color, alkalinity, NH3, TKN, total nitrogen,
total phosphorus, orthophosphate, and total bacteria.
Temperature, conductivity, dissolved oxygen, turbidity,
chlorophyll a, color, TKN, total nitrogen and total
phosphorus exhibited significant seasonal variation at the
95% confidence level (Table 5-19). Interannual variation or
no variation was reported for the other parameters. When
year and season were combined, Secchi, pH, conductivity,
TSS, turbidity, chlorophyll a, color, and total phosphorus
displayed significant variability.

Table 5-19. Correlation analyses of water quality data for Lake Bonny
using the General Linear Model procedure (GLM). Data were combined by
year, season and year*season.
PARAMETER
YEAR
SEASON
YEAR*SEASON
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
Secchi
X
X
X
Temperature
X
X
X
PH
X
X
X
Conductivity
X
X
X
O

Q
X
X
X
TSS
X
X
X
Turbidity
X
X
X
Chlorophyll a
X
X
X
Color
X
X
X
Alkalinity
X
X
X
NH,
X
X
X
TKN
X
X
X
TN
X
X
X
TP
X
X
X
ORT
X
X
X
BACTERIA
X
X
X
123

124
Lake Hunter
Lake Hunter is located in Polk County, Florida
(Appendix A). The lake lies in the Bartow Embayment division
of the Central Lakes District (Canfield, 1981). Lake Hunter
is a small urban lake located within the city limits of
Lakeland, with a surface area of 41 ha, a shoreline length
of 2.6 km, and a mean depth of 1.5 m (Table 5-20).
Table 5-20. Morphometry
of Lake Hunterd)
Surface Area
41 ha
Maximum Depth
2.7 m
Mean Depth
1.5 m
Development of Shoreline 80 %
Drainage Basin Area
208 ha
Shoreline Length
2.6 km
Macrophyte Cover
0.2 %
(1) Data front PCWR (1992).
Development in the watershed has caused Lake Hunter to
be used as a stormwater retention area. Twelve stormwater
discharge inlets enter the lake, six of which can be
considered as a major source of pollution. The lake
watershed is almost entirely developed and consists of
commercial and moderate to large residential developments.
Most typical of urban systems, Lake Hunter experiences
excessive sedimentation, poor water guality, and rapid
growth of undesirable aquatic vegetation. The lake's
macrophyte dominant vegetation (defined as greater than or
equal to 5% macrophyte cover) includes elephant ear, water

125
primrose, and pennywort. There is no aquatic weed control
program for Lake Hunter.
According to Forsberg and Ryding (1980), Lake Hunter
was hypereutrophic during this study. In 1983, Lake Hunter
was completely drained, and water quality has continued to
decline since then. Table 5-21 show water quality data for
Lake Hunter.
Lake Hunter does not have much seasonality in its gross
primary productivity (Table 5-3). The most productive layer
is at 50 cm water depth. The mean for the 50 cm depth was
306.71 mgC.m"3.h 175.6 throughout the year, with a minimum
value of 115.62 mgC.m'3.h in December and a maximum of
484.37 mgC.m_3.h in April. Values for net photosynthesis are
shown in table 5-4, community respiration in table 5-5, and
average annual productivity in table 5-6.
The phytoplankton community in Lake Hunter had
Cyanophytes as dominants with Chlorophytes being co
dominant. Species of blue-green algae such as Lvnabva sp.,
Spirulina sp., Microcystis sp..Merismopedia sp., and
Anabaena sp. were present as well as green algae such as
Closterium sp., and Scenedesmus sp.; the dinoflagellate
Peridinium sp. were abundant. No diatoms were reported for
this lake.
Rotifers and copepods comprised the dominant
zooplankton community for Lake Hunter. Dominant rotifers
were Keratella sp., Brachionus sp., and Monostvla sp.

126
Table 5-21. Summary of water quality data for Lake Hunter.
Numbers shown are mean values with ranges in parentheses.
I Parameter
Unit
Several Agencies*
(1989/91)
This study
(1992/93)
| Temperature
C
25.50 (16.50-31.51)
24.98 (17.87-30.09)
1 D.O.
mg/L
9.60 (4.72-13.90)
10.46 (9.88-11.8)
PH
9.48 (7.88-11.35)
8.74 (8.33-9.45)
| Alkalinity
mg/L CaC03
65.25 (46-99.9)
66.6 (62-71)
Conduct.
pmhos/cm
199.6 (139-253)
202.8 (160-276)
Nitrogen
mg/L
2.24 (0.51-4.92)
2.50 (1.58-3.56)
NH,-N
mg/L
0.0 6 7 (0.01-0.292)
0.0 4 2 (0.032-0.056)
TKN
mg/L
2.14 (0.51-4.81)
2.21 (1.42-3.05)
| Phosphorus
mg/L
0.150 (0.019-0.26)
0.18 (0.158-0.239)
B Chi. a
mg/m3
70.69 (25.6-152.8)
69.75 (43.3-112.3)
H Color
Pt units
28.6 (15-55)
38.66 (25-60)
| Secchi
m
0.38 (0.2-0.6)
0.36 6 (0.2-0.6)
I Bacteria
x 106
1236.3 d-20000)
342.5 (140-600)
] TSI
79.18 (66.36-94.52)
81.58 (71.94-88.88)
(*) AGENCY CODES:
USGS; EPA; PCWR; CL; UFC; FGFWFC; CF208; and SWFWMD.

127
Diaptomus was the only copepod identified. Data from
Canfield and Hoyer (1992) show a rotifer density of 31,200
individuals .m'3, and a copepod density of 14,000
individuals .m-3.
Eighteen species of fish were collected in Lake Hunter
(Canfield and Hoyer, 1992). Historically, Lake Hunter was a
hypereutrophic lake with an overabundance of forage fish,
principally gizzard shad and threadfin shad (Huish, 1955;
Ware and Horel, 1971). After the lake restoration done in
1983 and 1984, 10,000 largemouth bass and 1,000 sunshine
bass were stocked to establish a predator population (Moxley
et al. 1984). After 8 years, the most abundant open-water
species collected in experimental gillnets were gizzard shad
and sunshine bass, with 163 and 8 fish.net-1.24hr,
respectively (Canfield and Hoyer, 1992).
ANOVAs were performed using the General Linear Model
(GLM) (SAS, 1989) in order to identify significant seasonal
limnological factors (p < 0.05). The following parameters
were analyzed: Secchi, temperature, pH, conductivity,
dissolved oxygen, total suspended solids, turbidity,
chlorophyll a, color, alkalinity, NH3, TKN, total nitrogen,
total phosphorus, orthophosphate, and total bacteria.
The data were collected between 1989 and 1993, and
grouped into four seasons: winter (December to February);
spring (March to May); summer (June to August); and fall
(September to November). The results in the Table 5-22 are

Table 5-22. Correlation analyses of water quality data for Lake Hunter
using the General Linear Model procedure (GLM). Data were combined by
year, season and year*season.
PARAMETER
YEAR
SEASON
YEAR*SEASON
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
SIGNIF
N/SIGNF
Secchi
X
X
X
Temperature
X
X
X
PH
X
X
X
Conductivity
X
X
X
o

Q
X
X
X
TSS
X
X
X
Turbidity
X
X
X
Chlorophyll a
X
X
X
Color
X
X
X
Alkalinity
X
X
X
NH-,
X
X
X
TKN
X
X
X
TN
X
X
X
TP
X
X
X
ORT
X
X
X
BACTERIA
X
X
X
128

129
reported as significant at p < 0.05. Temperature,
conductivity, dissolved oxygen, TSS, chlorophyll a, and
orthophosphate exhibited significant seasonal variation.
Other parameters displayed either interannual or no
variation. When year and season were combined, Secchi, pH,
dissolved oxygen and turbidity displayed significant
variability.
Summary
For the scope of this study, the lakes were divided
into two groups: [1] lakes considered as having blue tilapia
as the major component of its filter-feeder fish population
(Lakes Parker, Hollingsworth, and Hancock), and [2] lakes
were gizzard shad is the dominant rough fish (Lakes Gibson,
Bonny and Hunter). Evaluating the limnology of the study
lakes, there seems to exist differences among them. Figure
5-1 has information on gross primary productivity (mgC.m'
3.h), community respiration (mgC.nf3.h), phytoplankton
(NU/ml) and zooplankton (individuals.m-3) for all lakes.
Lakes Parker and Hollingsworth have a significantly
larger (ANOVA, p <0.05) zooplankton population than Lakes
Bonny and Hunter, which could be indicative of the different
proportionality of filter feeder fish in the population.
Lakes where blue tilapia have proportionally high population
tend to have higher number of zooplankton, especially Lake

i Prod (mgC/m3/h)
Phyto (NU/ml x 1CT3)
Resp (mgC/m3/h)
7nnn v 10*9^
Figure 5-1. Phytoplankton and zooplankton population for all study lakes.
Values for primary productivity and community respiration are shown.
130

131
Parker, where the zooplankton population was two times
larger than the other lakes, data that are in agreement with
Crisman and Beaver (1988).
Lakes considered as having more blue tilapia also have
proportionally lower gross primary productivity (Figure 5-1)
than the group where gizzard shad is considered dominant,
which is consistent with the results of Crisman and Beaver
(1988).
Regarding to the fish population, the two extremes were
in the group where blue tilapia is prevalent. Lake Parker
has the smallest fish population of the six studied lakes
(total fish biomass = 71.1 kg.ha"1) (Canfield and Hoyer,
1991), and Lake Hollingsworth the largest fish population
(total fish biomass = 1050 kg.ha"1) (Canfield and Hoyer,
1992). The small fish population in Lake Parker could also
help to explain the presence of the large zooplankton
population in that lake. There were no noticeable
differences in the size of the fish populations for Lakes
Bonny and Hunter, both belonging to the group where gizzard
shad was dominant.

Prod (mgC/m3/h)
Zoop (#/m3 x 10*2)
Re8p (mgC/m3/h)
lilil Fish (kg/ha)
] Phyto (NU/ml x 10*3)
Figure 5-2. Fish, phytoplankton and zooplankton population for all study lakes.
Values for primary productivity and community respiration are shown.
132

CHAPTER 6
SUMMARY
Fish community impact on lake ecosystems was addressed
from the point of view of the effects of replacement of a
native filter feeder, gizzard shad, by the exotic blue
tilapia which has been present in central Florida eutrophic
lakes for the last thirty years. Evaluation of the impact of
gizzard shad and blue tilapia was through the
characterization and estimation of the fecal material
produced by each respective fish species. Analyses were
performed in order to determine how the different kinds of
feces could affect bioavailability of nutrients in the
systems.
A mesocosm experiment was performed to evaluate past
and present lake conditions. Paleolimnological data were
coupled with ambient water quality data in order to create
the necessary framework to permit further investigation of
system alterations due to modifications in the biological
community.
Several issues addressed in this study provided
information that led to a mosaic of interpretations.
Historical limnological data were analyzed for the six lakes
studied. Physicochemical and limnological data were
133

134
collected for more than twelve months in all the lakes.
Phytoplankton primary productivity and community respiration
were measured. Phytoplankton, zooplankton and fish
communities were identified. Sediment cores were taken from
two lakes, and sediment core information was gathered for
two of the others. Finally, the fecal composition of the two
major filter feeder fish in the systems was determined and
relative impact estimated.
All lakes except one were classified as hypereutrophic,
according to the classification of Forsberg and Ryding
(1980). Lake Gibson was classified as eutrophic. For
analytical purposes, I grouped the lakes into two
categories: lakes where the prevalent filter feeder fish
(defined as >50% of the filter-feeder fish population) was
gizzard shad and lakes where blue tilapia was the prevalent
filter feeder fish. Lakes Gibson, Bonny, and Hunter were
dominated by shad during this study, while Lakes Parker,
Hollingsworth and Hancock were tilapia dominated systems.
Fish feces experiment
There were significant differences in the amount of
feces hand-stripped from both species of fishes. Blue
tilapia releases its fecal material in pellet form enveloped
in mucilage, whereas gizzard shad releases its feces as a
flocculent material. Blue tilapia produced the greatest
amount of feces when compared with gizzard shad. The bulk

135
density of blue tilapia feces was two times the bulk density
for gizzard shad feces. On average, shad feces was composed
of 37% more organic matter, 50% more protein, and had a 60%
higher caloric content than did blue tilapia feces.
Furthermore, blue tilapia feces consisted of 50% more dry
matter than gizzard shad feces.
I carried out two bioassays to evaluate fecal nutrient
composition and algal content. I measured the change in
fluorescence for each treatment group. The three treatment
groups were: [1] feces plus Microcystis aeruginosa; [2]
feces plus Selenastrum capricornutum; and [3] control (no
feces). Each treatment group was inoculated with algae in
both a medium deprived of nitrogen and a medium deprived of
phosphorus.
Gizzard shad fecal samples always yielded greater
fluorescence values regardless of the algal strain or
chemical composition of the growth medium (-N or -P) used.
In treatment group feces plus Selenastrum capricornutum, a
significant increase in fluorescence occurred in the first
day of experiment; whereas in treatment group feces plus
Microcystis aeruginosa, a significant increase occurred on
the fourth day of the bioassay (Figures 3-3 and 3-4).
Results were different in the flasks inoculated with
blue tilapia fecal material. Despite high initial
fluorescence values, in the presence of treatment group
feces plus Microcystis aeruginosa, fluorescence values

136
remained constant and a decrease was noticed after the
fourth day of experimentation. When the treatment group
feces plus Selenastrum capricornutum was evaluated, high
initial fluorescence values decreased after the first day of
the bioassay, reaching minimum values on the fourth day of
the experiment. (Figures 3-3 and 3-4). Control flasks in the
medium deprived of phosphorus and in the medium deprived of
nitrogen exhibited algal fluorescence readings of zero or
near zero.
Chlorophyll a readings for blue tilapia feces were very
high in the beginning of the experiment (ten-fold greater
than the values measured for gizzard shad feces). The
results showed that gizzard shad fecal material started the
bioassay at low chlorophyll a value in the flasks containing
blue green algae both in the nitrogen and in the phosphorus
deprived medium. At the end of the experiment, chlorophyll a
values increased ten-fold in the nitrogen deprived medium
and 13.5 times in the phosphorus deprived medium. Blue
tilapia feces displayed, on the other hand, under the same
conditions, increased chlorophyll a only at the end of the
experiment and only 5 to 6 times the initial value, an
increase which is only half of that produced by gizzard shad
feces.
Phytoplankton biomass in gizzard shad and blue tilapia
feces was indicative of the different impact these fishes
can have in aquatic systems. While phytoplankton biomass was

137
suppressed (zero growth) in 73% of all samples containing
blue tilapia feces, all the samples containing gizzard shad
feces displayed a ten-fold increase.
Paleolimnolooical analysis
Analyses of sediment cores from Lakes Bonny and Gibson,
both defined in this study as having a prevalent filter-
feeder population of gizzard shad, were done to assess the
distribution of organic deposits throughout the lake basin
and for sediment age determination. Sediment
characterization was based on one core collected from the
deepest part of each lake.
Lake Bonny showed high organic matter content in the
superficial layers (58% of organic matter). TP and TN were
constant throughout the whole core. Bulk density and dry
matter began to decline after 1950, which coincides with
urban development of the watershed. Total phosphorus
reconstructions near the base of the core suggest that the
lake has always been mesotrophic to eutrophic or
hypereutrophic. However, the lake has experienced periods of
much higher nutrient enrichment, and values for total
phosphorus in excess of 5.0 mg.g'1 were common for layers
that post-dated the 137Cs peak after 1960. Detectable 137Cs,
injected into the atmosphere as a consequence of nuclear
weapons testing in the early 1950s, matched well with the
determined 210Pb chronology!

138
Lake Gibson had less organic matter in its superficial
layers than did Lake Bonny. However, values of up to 40%
organic content were registered throughout the core, which
coincides with the average organic matter content of dry
bulk sediment in Florida lakes of 39.7% (Brenner and
Binford, 1988). Total nitrogen and total phosphorus declined
at depths > 24 cm in the core. Below 48 cm in the core,
total phosphorus was undetectable.
Total phosphorus reconstructions at the base of the
core suggest that the lake was oligotrophic for at least the
last century. Nevertheless, values for TP exceeding 3.0
mg.g'1 were registered for the sediment corresponding to
the last 25 years, reflecting changes in the watershed use.
Using 210Pb detrminations, it appears that the lake had
much less organic sediment before the 1950s. Detectable
137Cs coincided with the determined 210Pb chronology.
Lakes Parker and Hollingsworth, both defined in this
study as having a blue tilapia dominant fish population,
were the subject of a sediment survey done by Schelske et
al. (1992). They stated that Lake Parker has always been
mesotrophic to eutrophic. Analyses of diatom assemblages
indicated the presence of alkaline conditions. Determination
of total phosphorus in the topmost 15 cm of sediment
suggested eutrophic conditions. However, these analyses
further indicated a progressive decline in water column
total phosphorus that may signal some reversal of cultural

139
enrichment (Schelske et al., 1992). Diatom reconstructions
of historical water column total phosphorus concentrations
indicate that Lake Hollingsworth has been eutrophic to
hypereutrophic for the past 100 years (Schelske et al.,
1992). During the last ten years, however, total phosphorus
inferences (cf. Schelske et al., 1992) suggest that the rate
of eutrophication remained relatively constant.
Since both of these lakes were classified as having a
blue tilapia dominant fish population in this study, it is
possible that some of the trophic reversal can be credited
to a filter feeder fish replacement. Additional support for
this hypothesis can be drawn indirectly from the absence of
an improvement trend in the sediment nutrient concentration
in Lakes Bonny and Gibson, both gizzard shad dominant lakes.
However, additional study is required in order to confirm
this hypothesis.
Lakes assessment
The lakes were analyzed for the years 1989 to 1993 and
separated by season within each of these years. The seasons
were defined as winter (December, January and February),
spring (March, April and May), summer (June, July and
August), and fall (September, October and November). ANOVAs
were performed using General Linear Model (GLM) (SAS, 1989)
in order to identify significant seasonal limnological
factors (p < 0.05).

140
Cluster analysis (Pearson Correlation Coefficient)
(SAS, 1989) divided the six lakes into 4 aggregations. Lakes
Hunter and Bonny, both with a prevalent population of
gizzard shad, were the most statistically similar lakes in
this study. Lake Parker, a blue tilapia dominant lake,
displayed limnological characteristics that differed
significantly from the mean values reported for the other
study lakes.
Cluster analyses of lakes by fish species yielded
similar relationships. Lake Parker again did not show any
correlation with the other study lakes. Seasonally, summer
and fall appeared to be the most stable seasons of the year
for the study lakes, and winter is the one which exhibited
the greatest statistical variability. This observation is
certainly a consequence of temperature, with minimum annual
temperatures occurring during the winter months.
Primary productivity determinations showed high gross
photosynthesis for most of the lakes, with a mean of 334.22
mgC.m'3.h + 106.35. The only deviation occurred in Lake
Parker, where a mean value of 66.41 mgC.m'3.h was recorded.
Conclusion
In view of the evidence presented in this study,
coupled with the findings of mesocosm experiments performed
by Crisman and Kennedy (1982), Drenner et al. (1986), and
Crisman and Beaver (1988), it seems possible that

141
biomanipulation techniques employing fish as a major element
could lead to a significant improvement in water clarity in
subtropical and tropical lakes. This study certainly
provides evidence that gizzard shad and blue tilapia feces
have a different composition and may stimulate different
plankton community dynamics.
The results of this investigation suggest intriguing
possibilities for the potential role of filter feeding fish
in the control of eutrophication in Florida lacustrine
systems. Gizzard shad feces clearly stimulated algal growth,
while blue tilapia feces appeared to suppress algal growth.
However, additional research is necessary in the use of
filter feeding fish in biomanipulation schemes.

LITERATURE CITED
AMERICAN PUBLIC HEALTH ASSOCIATION. 1985. Standard methods
for the examination of water and wastewater. 16th
edition. APHA, Washington D.C.
ANDERSSON, G., H. Berggren, G. Cronberg and C. Gelin. 1978.
Effects of planktivorous and benthivorous fish on
organisms and water chemistry in eutrophic lakes.
Hydrobiologia 59 (1):9-15.
APPLEBY, P.G., P.J. Nolan, D.W. Gifford, M.J. Godfrey, F.
Oldfield, N.J. Anderson and R.W. Battarbee. 1986. 210Pb
dating by low background gamma counting. Hydrobiologia
143:21-27.
, and F. Oldfield. 1983. The assessment of 210Pb
data from sites with varying sediment accumulation
rates. Hydrobiologia 103:29-35.
BAGENAL, T. (ed.) 1978. Methods for assessment of fish
production in fresh waters. IBP Handbook no. 3.
Blackwell Scientific Publications, Oxford. 365 pp.
BAKER, C.D. and E.H. Schmitz. 1971. Food habits of adult
gizzard and threadfin shad in two Ozark reservoirs, p.
3-10. In G.E. Hall (ed) Reservoir Fisheries and
Limnology. Special Publ. #8, Amer. Fish. Soci. 511 p.
BAYS, J.S. and T.L. Crisman. 1983. Zooplankton and trophic
state relationships in Florida lakes. Can. J. Fish.
Aquat. Sci. 40:1813-1819.
BEAVER, J.R. 1990. Importance of organic color,
bacterioplankton and planktivore grazing in structuring
ciliated protozoan communities in subtropical lakes.
Ph.D. Dissertation. University of Florida. 227 pp.
r and T.L. Crisman. 1989. Replacement of an
endemic pump-filter feeding fish by an exotic omnivore:
Implications for water quality management in sub
tropical lakes.
142

143
, T.L. Crisman and J.S. Bays. 1981. Thermal
regimes of Florida lakes. A comparison with biotic and
climatic transitions. Hydrobiologia. 83:267-273.
BENNDORF, J. 1988. Objectives and unsolved problems in
ecotechnology and biomanipulation:A preface.
Limnologica (Berlin) 19:5-8.
, H. Kneschke, K. Kossath and E. Penz. 1984.
Manipulation of the pelagic food web by stocking with
predacious fishes. Int. Revue ges. Hydrobiologia
69:407-428.
, H. Schultz, A. Benndorf, R. Unger, E. Penz, H.
Kneschke, K. Kossatz, R. Dunke, U. Hornig, R. Kruspe
and S. Reichel, 1988. Food-web manipulation by
enhancement of piscivorous fish stocks:Long-term
effects in the hypertrophic Bautzen Reservoir.
Limnologica 19:97-110.
BERRY, F.H. 1958. Age and growth of the gizzard shad (D.
lecepedi) (LeSueur) in Lake Newnan, Florida. Proc.
Eleventh Amer. Conf., Southeastern Assoc. Game and Fish
Commissioners. 318-331.
BJORK, S. 1985. Lake restoration technigues. In:Proceedings
International Congress Lakes Pollution and Recovery",
Rome:281-292.
BODOLA, A. 1966. Life history of the gizzard shad, Dorosoma
cepedianum (LeSueur), in western Lake Erie. Fishery
Bull. 65:391-425.
BOYLE, J.R. and C.W. Hendry, Jr. 1985. The mineral industry
of Florida, 1983. In Minerals Yearbook, 1983. U.S.
Bureau of Mines, Washington, v.2, pp. 137-148.
BREMNER, J.M. and C.S. Mulvaney. 1982. Nitrogen-total. Pages
599-609 in A.L. Page, R.H. Miller, and D.R. Keeney
(eds.), Methods of Soil Analysis, Part II, Chemical and
Microbiological Properties. Amer. Soc. Agro. & Soil
Sci. Soc. Amer., Madison, Wise. 599-609.
BRENNER, M. and M.W. Binford, 1988. Relationships between
concentrations of sedimentary variables and trophic
state in Florida lakes. Canadian J. Fish. Aquat. Sci.
45:294-300.
/ M.W. Binford and E.S. Deevey. 1990. Lakes. Pages
364-391 in R.L. Myers and J.J. Ewel (eds.), Ecosystems
of Florida. Univ. Central Florida Press, Orlando.

144
, T.J. Whitmore, M.S. Flannery, and M.W. Binford.
1993. Paleolimnological methods for defining target
conditions in lake restoration: Florida case studies.
Lake and Reserv. Manage. 7(2) :209-217.
BROOKS, J.L. and S.I. Dodson, 1965. Predation, body size and
composition of plankton. Science 150:28-35.
CAILTEUX, R.L. 1988. Food resource utilization by two
population of blue tilapia, Tilapia aurea, in north
central Florida. M.Sc. thesis, University of Florida,
Gainesville.
CAIRD, J.M. 1945. Algae growth greatly reduced after
stocking pond with fish. Water Works Engineering
98:240.
CAMPBELL, K.M. 1986. Geology of Polk County, Florida.
Florida Geological Survey. Open File Report no. 13.
CANFIELD, Jr., D.E. 1981. Chemical and trophic state
characteristics of Florida lakes in relation to
regional geology. Final report to Coop. Fish and
Wildlife Res. Unit., University of Florida,
Gainesville.
, and M.V. Hoyer. 1988. Regional geology and the
chemical and trophic state characteristics of Florida
lakes. Lake and Reserv. Manage. 4:21-31.
, and M.V. Hoyer. 1991. A characterization of the
fish population in lake Parker (Polk County), Florida.
Final report to SWFWMD. Brooksville.
, and M.V. Hoyer. 1992. Aguatic macrophytes and their
relation to the limnology of Florida lakes. Final
report to Bureau of Aguatic Plant Management. FDNR,
Tallahassee.
CARPENTER, S.R. and J.F. Kitchell. 1988. Consumer control of
lake productivity. Bioscience 38:764-769.
, J.F. Kitchell and F.R. Hodgson. 1985.
Cascading trophic interactions and lake productivity.
Bioscience. 35:634-639.
, J.F. Kitchell, J.R. Hodgson, P.A. Cochran, J.J.
Elser, M.M. Elser, D.M. Lodge, D. Kretchmer & X. He,
1987. Regulation of lake primary productivity by food
web structure. Ecology 68: 1863-1876.

145
COHEN, A.S. and C. Nielson. 1986. Ostracodes as indicators
of paleohydrochemistry in lakes: a late Quaternary
example from Lake Elementeita, Kenya. Palaios 1:601-
609.
COOKE, G.D., E.B. Welch, S.A. Peterson, and P.R. Newroth.
1993. Restoration and management of lakes and
reservoirs. Second edition. Lewis Publishers, Boca
Raton, Fl.
COURTENAY, W.R.Jr. and C.R. Robins. 1973. Exotic aquatic
organisms in Florida with emphasis on fishes: a review
and recommendations. Trans. Am. Fish. Soc. 102: 1-12.
, and J.R. Stauffer, Jr. 1984. Distribution, Biology,
and Management of Exotic Fishes. The Johns Hopkins
University Press, Baltimore. 430 pp.
COVENEY, M.F., G. Cronberg, M. Enell, K. Larson, and L.
Olofsson. 1977. Phytoplankton, zooplankton and bacteria
standing crop and production relationships in a
eutrophic lake. Oikos 29:5-21.
COWELL, B.C., C.W. Dye and R.C. Adams. 1975. A synoptic
study of the limnology of Lake Thonotosassa, Florida.
Part I: Effects of primary treated sewage and citrus
wastes. Hydrobiologia 46(2-3):301-345.
, and D.S. Vodopich. 1981. Distribution and
seasonal abundance of benthic macroinvertebrates in a
subtropical Florida lake. Hydrobiologia 78:97-105.
CRAMER, J.D. and G.R. Marzolf. 1970. Selective predation on
zooplankton by gizzard shad. Trans. Amer. Fish. Soc.
99:320-332.
CRISMAN, T.L. 1980. Chydorid Cladoceran assemblages from
subtropical Florida. Pages 599-608 in W.C. Kerfoot
(ed), Evolution and Ecology of Zooplankton
Communities. Univ.Press of New England, Hanover, N.H.
. 1992. Natural lakes of the southeastern
United States: Origin, structure, and function. Pages
387-405 in W.H.Martin (ed), Biotic Communities of the
Southeastern United States. Wiley Press, New York.
, and J.R. Beaver. 1988. Lake Apopka trophic
structure manipulation Phase I Final Project Report.
St. Johns River Water Management District,
Jacksonville, FL. 127 pp.

146
, and J.R. Beaver. 1990. Applicability of
planktonic biomanipulation for managing eutrophication
in the subtropics. Hydrobiologia 1(11):177-186.
, and H.M. Kennedy, 1982. The role of gizzard
shad (Dgrgsgma cepedianum) in eutrophic Florida lakes.
Publ. 64. Water Resources Research Center, University
of Florida, Gainesville, 83 pp.
CROWDER, L.B. 1985. Optimal foraging and feeding mode shifts
in fishes. Environ. Biol. Fishes 12(l):57-62.
CROWDER, C.B., R.W. Drenner, W.C. Kerfoot, D.J. McQueen,
E.L. Mills, V. Sommer, C.N. Spencer, and J.J. Vanni.
1988. Food web interactions in lakes. Pages 314-397 in
S.R. Carpenter (ed), in Complex Interactions in Lake
Communities. Springer Verlag, New York.
DALQUEST, W.N. and L.J. Peters. 1966. A life history study
of four problematic fish in Lake Diversion, Archer and
Baylor counties, Texas. Texas Park and Wildlife Dept.
I.F. Series, no. 6:87 p.
DICKMAN, M. and H. Nanne. 1987. Impact of tilapia grazing on
plankton composition in artificial ponds in Guanacaste
Province, Costa Rica. J. Freshwat. Ecol. 4: 93-100.
DRENNER, R.W. 1977. The feeding mechanism of the gizzard
shad (Dgrgsgma cepedianum). Ph.D. Thesis Univ. Kansas
Lawrence.
, F. de Noyelles Jr. and D. Kettle. 1982a. Selective
impact of filter feeding gizzard shad on zooplankton
community structure. Limnol. Oceanogr. 27:965-968.
, K.D. Hambright, G.L. Vinyard, M. Gophen and V.
Pollingher. 1987. Experimental study of size-selective
phytoplankton grazing by a filter-feeding cichlid and
the cichlid's effects on plankton community structure.
Limnol. Oceanogr. 32:1140-1146.
, J.R. Mummert, F. de Noyelles Jr. and D. Kettle.
1984a. Selective particle ingestion by a filter-feeding
fish and its impact on phytoplankton community
structure. Limnol. Oceanogr. 29(5):941-948.
, W.J. O'Brien and J.R. Mummert. 1982b. Filter
feeding rates of gizzard shad. Trans. Am. Fish. Soc.
111:210-215.
, J.R. Strickler and W.J. O'Brien. 1978. Capture
probability. The role of zooplankton escape in the

147
selective feeding of planktivorous fish. J. Fish. Res.
Bd. Can. 34:1370-1373.
, S.B. Taylor, X. Lazzaro and D. Kettle. 1984 b.
Particle grazing and plankton community impact of an
omnivorous cichlid. Trans. Am. Fish. Soc. 113:397-402.
, S.T. Threlkeld and M.D. McCracken. 1986.
Experimental analysis of the direct and indirect
effects of an omnivorous filter-feeding clupeid on
plankton community structure. Can. J. Fish. Aquat.
Sci. 43:1935-1945.
EDMONDSON, W.T. and S.E.B. Abella. 1988. Unplanned
biomanipulation in Lake Washington. Limnologica
(Berlin) 19:73-79.
ELLIOTT, J.M. 1976. Energy losses in the waste products of
brown trout (Salmo trutta.L). J. Anim. Ecol. 45:561-
580.
FOOTE, K.J. 1977. Annual performance report. Blue tilapia
investigations. Florida Game Fresh Water Fish Comm.,
Tallahassee, Florida.
FLANNERY, M.S., R.D. Snodgrass and T.J. Whitmore. 1982.
Deepwater sediments and trophic conditions in Florida
lakes. Hydrobiologia 92:597-602.
FLORIDA BOARD OF CONSERVATION. 1969. Florida lakes. Part 3.
Gazeteer. Div. Water Resour., Tallahassee.
FORSBERG, C. and S.O. Ryding. 1980. Eutrophication
parameters and trophic state indices in 30 Swedish
waste-receiving lakes. Arch, fur Hydrobiol. 88:189-207.
FROST, T.M., D.L. DeAngelis, S.M. Bartell, D.J. Hall and
S.H. Hulbert. 1988. Scale in the design and
interpretation of aquatic community research. Pages
229-261 in S.R. Carpenter (ed), Complex interactions in
lake communities. Springer- Verlag.
GARLAND, C.R. 1972. A comparative study of the trophic
relationships of the gizzard shad (Dqrqsgma cepedianuim
in Acton Lake and Four-Mile Creek. M.S. thesis, Miami
University, Oxford, Ohio. 51p.
GOTTGENS, J.F. 1992. Quantitative impacts of lake-level
stabilization on sediment and nutrient dynamics:
coupling limnology with modeling. Ph.D. dissertation,
University of Florida, Gainesville.

148
GOPHEN, M. 1990. Biomanipulation: retrospective and future
development. Hydrobiologia 200/201:1-11.
, R.W. Drenner and G.L. Vinyard. 1983. Cichlid
stocking and the decline of the Galilee Saint Peters
Fish (Sarotherodon galilaeus) in Lake Kinneret, Israel.
Can. J. Fish. Aquat. Sci. 40: 983-986.
HAITH, D.A., W. Rast, K.H. Reckhow, L. Somlydy and G. van
Straten. 1989. The use of models. Pages 85-113 in S.-O.
Ryding and W. Rast (eds.), The Control of
Eutrophication of Lakes and Reservoirs, Parthenon Publ.
Group Inc., Park Ridge, N.J. 314 pp.
HAKANSON, L. and M. Jansson. 1983. Principles of Lake
Sedimentology. Springer-Verlag, New York.
HALL, D.J., W.E. Cooper & E.E. Werner. 1970. An experimental
approach to the production dynamics and structures of
freshwater animal communities. Limnol. Oceanogr.
15:839-928.
HALVER, J.E. 1989. Fish Nutrition. 2nd edition. San Diego
Academic Press, New York. 798 pp.
HANAZATO, T., H. Hayashy, T. Ichikawa and Y. Watanabe. 1989.
Dynamics of zooplankton community in enclosures of
different types in a shallow eutrophic lake. Jpn. J.
Limnol. 50(l):25-37.
HANEY, J.F. and D.J. Hall. 1973. Sugar-coded Daphnia A
preservation technique for cladocera. Limnol. Oceanogr.
18:331-333.
HEAT, R.C. and C.S. Conover. 1981. Hydrologic Almanac of
Florida. U.S. Geological Survey, Tallahassee, Florida.
HENDRICKS, M.K. and R.L. Noble. 1980. Feeding interactions
of three planktivorous fishes in Trinidad Lake, Texas.
Proc. Annu. Conf. SE Ass. Fish. Wildl. Agencies 33:
324-330.
HENRIKSON, L., H.G. Nyman, H.G. Oscarson and J.A. Stenson.
1980. Trophic changes without changes in the external
nutrient loading. Hydrobiologia 68:257-263.
HOBBIE, J.E. and R.J. Daley. 1977. Use of nucleopore filters
for counting bacteria by epifluorescent microscopy.
App. Environ. Micro. 33:1225-1228.

149
HOLANOV, S.H. and I.C. Tash. 1978. Particulate and filter
feeding in threadfin shad, (Dorosoma petenense). at
different light intensities. J. Fish. Biol. 13:619-625.
HOSPER, S.H. and M.-L. Meijer. 1986. Control of phosphorus
loading and flushing as restoration methods for Lake
Veluwe, The Netherlands. Hydrobiol. Bull. 20:183-194.
HRBACEK, J., M. Dvorakova, V. Korinek and L. Prochazkova.
1961. Demonstration of the effect of the fish stock on
the species composition of zooplankton and the
intensity of metabolism of the whole plankton
association. Verh. Int. Ver. Limnol. 14:192-195.
HUBER, W.C., P.L. Brezonik, J.P. Heaney, R.E. Dickinson,
S.D. Preston, D.S. Dwornik and M.A. DeMaio. 1982. A
classification of Florida lakes. Rep. ENV-05-82-1 to
Florida Dep. Environ. Reg., Tallahassee.
HURLBERT, S.M. 1984. Pseudoreplication and the design of
ecological field experiments. Ecol. Monogr. 54:187-211.
HUTCHINSON, G.E. 1957. A Treatise on Limnology.I Geography,
Physics, and Chemistry. John Wiley and Sons, Inc, New
York, 1015 pp.
JANUSZKO, M. 1974. The effect of three species of
phytophagous fish on algae development. Pol. Arch.
Hydrobiol. 21:431-454.
. 1978. The influence of silver carp
(Hypophthalmichthvs molitrix) on eutrophication of the
environment of carp ponds. III. Phytoplankton. Rocz.
Nauk Roln. ser. H 99:55-79.
JESTER, D.B. and B.L. Jensen. 1972. Life history and ecology
of the gizzard shad, Dorosoma cepedianum (Le Sueur),
with reference to Elephant Butte Lake, New Mexico State
University, Agricultural Experiment Station Research
Report 218:1-56.
JOHNSON, T.C., J.D. Halfman and W.J. Showers. 1991.
Paleoclimate of the past 4000 years at Lake Turkana,
Kenya, based on the isotopic composition of authigenic
calcite. Paleogeography, Paleoclimatology,
Paleoecology 85:19-198.
KAJAK, Z., J. Zawiska and A. Hillbricht-Ilkowska. 1976.
Effect of experimentally increased fish stock on
biocenosis and recovery processes of a pond type lake.
Limnologia, Berlin 10(2):595-601.

150
KOLASA, K.V. 1993. Lake Parker diagnostic feasibility study.
Final report. Southwest Florida Water Management
District, Brooksville, FL.
KUSHLAN, J.A. 1974. Effects of a natural fish kill on the
water quality, plankton, and fish population of a pond
in the Big Cypress Swamp, Fl. Trans. Amer. Fish. Soc.
2:235-243.
KUTKUHN, J.H. 1958. Utilization of plankton by juvenile
gizzard shad in a shallow prairie lake. Trans. Amer.
Fish. Soc. 87:80-103.
LAMARRA, V.A.,Jr. 1975. Digestive activities of carp as a
major contributor to the nutrient loading of lakes.
Verh. Int. Ver. Limnol. 19:2461-2468.
LANGELAND, A., J.I. Kolsvik, Y. Olsen and H. Reinertsen.
1987. Limnocorral experiments in a eutrophic lake -
Effects of fish on the planktonic and chemical
conditions. Pol. Arch. Hydrobiol. 34(l):51-65.
LAZZARO, X. 1987. A review of planktivorous fishes: their
evolution, feeding behaviors, selectivities, and
impacts. Hydrobiologia 146:97-167.
LEVENTER, H. 1981. Biological control of reservoirs by fish.
Bamidgeh 33(l):3-23.
LEWIS, W.M. 1953. Analysis of the gizzard shad population of
Crab Orchard Lake, Illinois. Ill. Acad. Sci. Trans.
46:231-234.
LIVINGSTONE, D.A. 1981. Paleolimnology. The ecology and
utilization of African inland waters. Pages 176-182 in
J.J. Symoens, M. Burgis, J.J. Gaudet (eds.), United
Nations Environmental Programme. Nairobi, Kenya.
LUND, J.W.G. & J.F. Tailing. 1957. Botanical limnological
methods with special reference to algae. Botanical
Review 23(8-9):489-583.
LYNCH, M., 1979. Predation, competition and zooplankton
community structure: an experimental study. Limnol.
Oceanogr. 24: 253-272.
McBAY, L.G., 1961. The biology of Tilapia nilotica Linneaus.
Proceedings of the Fifteenth Annual Conference,
Southeastern Association of Game and Fish
Commissioners, pp. 208-218.

151
MACEINA, M.J. and D.M. Soballe. 1990. Wind-related
limnological variation in Lake Okeechobee, Florida.
Lake and Reserv. Manage. 6:93-100.
MACKERETH, F.J.H., J. Heron and J.F. Tailing. 1978. Water
analysis: some revised methods for limnologists.
Freshwater Biological Association Scientific
Publications no. 36, Ambleside. 120 pp.
MALLIN, M.A., 1986. The feeding ecology of the blue tilapia
(T. aurea) in a North Carolina reservoir. Pages 323-326
in Lake and Reservoir Management. Vol. 2.
MANN, K. 1969. The dynamics of aquatic ecosystems. Pages 1-
81 in J. Cragg (ed.), Advances in Ecological Research
Academic Press, London. Vol. 6 236 pp.
MENZEL, D.W. and N.Corwin. 1965. The measurement of total
phosphorus in seawater based on the liberation of
organically bound fractions by persulfate oxidation.
Limnol. Oceanogr. 10:280-282.
MILLER, R.R. 1960. Systematics and biology of the gizzard
shad (Dorosoma cepedianum) and related fishes. Fishery
Bull., Fish and Wildlife Ser. 60:371-392.
MIURA, T. 1989. Effects of fish plankton feeders on the
plankton community in a small eutrophic lake. Programme
and Abstracts of International Conference
Biomanipulation Tool for Water Management, Amsterdam,
The Netherlands. Aug. B-ll. 116 pp.
MOXLEY, D., V. Williams and C. Harris. 1984. Resource
restoration section 1983-84 annual report. Florida Game
and Fresh Water Fish Commission, Annual Report,
Tallahassee, Florida.
MURPHY, J. and J.P. Riley. 1962. A modified single solution
method for the determination of phosphate in natural
waters. Anal. Chim. Acta 27:31-36.
NAKAMOTO, N. and T. Okino. 1972. Activity of phytoplankton
excreted by fish. The Bulletin of Plankton Society of
Japan 19(1):1-4.
NAUWERCK, A., 1963. Die Beziehungen zwischen Zooplankton und
Phytoplankton in See Erken. Symb. Bot. Upsal. 17(5):163
PP-
NELSON, D.W. and L.E. Sommers. 1975. Determination of total
nitrogen in natural waters. J. Environ. Qual. 4:465-
468.

152
NOBLE, R.L., R.D. Germany and C.R. Hall. 1975. Interactions
of blue tilapia and largemouth bass in a power plant
cooling reservoir. Proc. Annu. Conf. SE Game Fish.
Comm. 29: 247-251.
O'NEIL, R.V., D.L. DeAngelis, J.B. Waide and T.F.H. Allen.
1986. A Hierarchical Concept of Ecosystems. Princeton
University Press, Princeton, N.J.
OPUSZYNSKI, K. 1978. The influence of the silver carp
Hypophthalmichthvs molitrix, on eutrophicaton of the
environment of carp ponds. Part VII. Recapitulation.
Roczn. Nauk Roln. ser. H. 99:127-150.
. (undated). Fish manipulation as a tool for
keeping the lakes clean. The theory of
ichthyoeutrophication. Inland Fisheries Institute,
Varsaw, Poland.
, and J.V. Shireman. (unpubl). A new approach to
quantitative evaluation of grazing by filter feeding
fish and to their use for water purification and
aqua/agriculture production enhancement.
, J.V. Shireman, and W. Opuszynski. (unpubl). Feces
collection of filter-feeding fish as a method for
wastewater treatment and enhancement of
aquaculture/agriculture production.
PARSONS, T.R. and J.D. Strickland. 1963. Discussion of
spectrophotometric determination of marine-plant
pigments, with revised equations of ascertaining
chlorophylls and carotenoids. J. Mar. Res. 21:155-163.
PIERCE, R.J. 1977. Life history and ecological energetics of
the gizzard shad (Dqrqsqma cepedianum) in Acton Lake,
Ohio. Ph.D. dissertation, Miami University, Oxford,
Ohio. 201 p.
POLK COUNTY WATER RESOURCES DIVISION. 1992. Lake water
quality report. Polk County, Florida.
POLLMAN, C.D. 1982. Internal loading in shallow lakes. Ph.D.
dissertation, University of Florida, Gainesville, Fl.
POPMA, T.J. 1982. Digestibility of selected feedstuffs and
naturally occurring algae by Tilapia. Ph.D.
dissertation, Auburn University, Auburn, Ala.
POST, J.R. and D.J. McQueen. 1987. The impact of
planktivorous fish on the structure of a plankton
community. Freshwater Biology 17:77-89.

153
PULLIN, R.S.V. and R.W. Lowe-McConnell. 1984. The biology
and culture of Tilapia. Pages 34-57 in ICLARM
Conference Proceedings 7, International Center for
Living Aquatic Resources.
PURI, H.S. and R.O. Vernon. 1964. Summary of the geology of
Florida and a guidebook to the classic exposures.
Florida Geological Survey Special Publication n. 5.
REINERTSEN, H. and Y.Olson. 1984. Effects of fish
elimination on the phytoplankton community of a
eutrophic lake. Verh. Internat. Verein. Limnol. 22:649-
657.
ROUND, F.E. 1977. The Biology of the Algae. Edward Arnold,
London. 278 pp.
SAS Institute, Inc. 1989. SAS user's guide: statistics. SAS
Institute, Inc.
SCHWOERBEL, J. 1975. Mtodos de Hidrobiologia. H. Blume
Ediciones, Madrid. 262 pp.
SCOTT, W.B. and E.J. Crossman. 1973. Freshwater fishes of
Canada. Fish. Res. Bd. Can. Bull 184R. 966 p.
SHAFLAND, P.L., 1978. A status report on tilapia in Florida.
Non-native Fish Research Laboratory, Contribution no.
16. 5 pp.
SHAPIRO, J., 1978. The need for more biology in lake
restoration. Pages 161-167 in Lake Restoration. EPA-
440/5-79-001, Washington, D.C.
, 1990. Biomanipulation: the next phase making
it stable. Hydrobiologia 200/201:13-27.
, V. Lamarra and M. Lynch, 1975. Biomanipulation:
An ecosystem approach to lake restoration. Pages 85-86
in P.L. Brezonik and J.L. Fox (eds.), Water Quality
Management Through Biological Control. Dept. Environ.
Engineering Sciences, University of Florida,
Gainesville, 164 pp.
, and D.I. Wright. 1984. Lake restoration by
biomanipulation: Round Lake, Minnesota, the first two
years. Freshwat. Biol. 14:371-383.
SMELTZER, E. and E.B. Swain. 1985. Answering lake management
questions with paleolimnology. Pages 268-274 in Lake
Reserv. Manage.- Practical Applications. Proc. 4th
Annu. Conf. and Symp. (NALMS) 390pp.

154
SMITH, A.D. 1971. Some aspects of trophic relations of
gizzard shad, Dorosoma cepedianum. Ph.D. thesis.
Virginia Polytechnic Institute and State University,
Blacksburg, 86 pp.
SMITH, W.L., 1963. Viable algal cells from the gut of the
gizzard shad Dorosoma cepedianum (LeSeur). Proc. Okla.
Acad. Sci. 43:148-149.
SOLOMON, D.J. and A.E. Brafield. 1972. The energetics of
feeding, metabolism and growth of perch
(Perca fluviatilis, L.) J. Anim. Ecol. 41:699-718.
SPATARU, P. and M. Zorn. 1978. Food and feeding habits of
Tilapia aurea (Steindachner) (Cichlidae) in Lake
Kinneret (Israel). Aquaculture 13: 67-79.
STEEMAN-NIELSEN, E. 1965. On the terminology concerning
productions in aquatic ecology, with a note about
excess production. Arch. Hydrob. 62(02):184-189.
STEWART, H.G. 1966. Ground-water resources of Polk County,
Florida. Fla. Geol. Survey Rept. Invest. No.44.
STOERMER, E.F., N.A. Andersen and C.L. Schelske. 1992.
Diatom succession in the recent sediments of Lake
Okeechobee, Florida, U.S.A. Diatom Research 7(2):367-
386.
THRELKELD, S.T. and E.M. Choinski. 1987. Rotifers,
cladocerans and planktivorous fish: what are the major
interactions? Hydrobiologia 147:239-243.
, and R.W. Drenner. 1987. An experimental mesocosms
study of residual and contemporary effects of an
omnivorous, filter-feeding, clupeid fish on plankton
community structure. Limnol. Oceanogr. 32(6):1331-1341.
TUNDISI, J. and T.M. Tundisi. 1976. Produco orgnica em
ecossistemas aquticos. Ciencia e Cultura
28(8) : 864-887.
U.S. Environmental Protection Agency. 1979. Handbook for
analytical quality control in water and wastewater
laboratories. EPA 600/4-79-019, Cincinnati, Ohio.
Van DONK, E., M.P. Grimm, R.D. Gulati and J.P.G. Klein
Breteler. 1990. Whole-lake food-web manipulation as a
means to study community interactions in a small
ecosystem. Hydrobiologia 200/201:275-289.

155
, R.D. Gulati and M.P. Grimm. 1989. Food web
manipulation in Lake Zwemlust: positive and negative
effects during the first two years. Hydrobiol. Bull.
23:19-34.
Van LIERE, E. 1986. Loosdrecht lakes, origin,
eutrophication, restoration and research program.
Hydrobiol. Bull. 20:9-15.
VELASQUEZ, G.T. 1939. On the viability of alga obtained from
the digestive tract of the gizzard shad, Dorosoma
cepedianum (LeSeur). Am. Midi. Nat. 22:376-412.
VINYARD, G.L., R.W. Drenner, M. Gophen, U. Pollingher, D.L.
Winkelman and K.D. Hambright. 1988. An experimental
study of the plankton community impacts of two
omnivorous filter-feeding cichlids, Tilapia qalilaea
and Tilapia aurea. Can.J. Fish. Aquat. Sci. 45:685-690.
VOLLENWEIDER, R.A. 1974. A Manual on Methods for Measuring
Primary Production in Aquatic Environments. IBP
Handbook no. 12. Blackwell Scientific Publications,
Oxford. 225 pp.
WATKINS, C.E.,II J. V. Shireman and W.T. Haller. 1983. The
influence of aquatic vegetation upon zooplankton and
benthic macroinvertebrates in Orange Lake, Florida. J.
Aquat. Plant Manage. 21:78-83.
WARE, F.J. 1973. Status and impact of Tilapia aurea after
twelve years in Florida. Unpubl. Rep. Florida Game
Fresh Water Fish Comm., Tallahassee, Florida.
, and G. Horel. 1971. 1970-71 annual progress report
for research and development project. Florida Game and
Fresh Water Fish Commission, Federal Aid in Fish
Restoration Dingell-Johnson Project F-12-12 Lake
Management, Research and Development, Tallahassee,
Florida.
, R.D. Gasaway, R.A. Martz and T.F. Drda. 1975.
Investigations of herbivorous fishes in Florida, pp.
79-84 In Water Quality Management Through Biological
Control. Report n ENV-07-75-1. Department of
Environmental Engineering Sciences, University of
Florida, Gainesville, FI.
WETZEL, R.G. 1983. Limnology. 2nd edition. Ed. Saunders
Coll., Philadelphia.
/ and G.E. Likens. 1991. Limnological Analyses. 2nd
edition. Springer-Verlag, New York.

156
WHITE, W.H. 1970. The geomorphology of the Florida
peninsula. Florida Department of Natural Resources
Geological Bulletin n. 51.
WILLIAMS, V.P., D.E. Canfield, Jr., M.M. Hale, W.E. Johnson,
R.S. Kautz, J.T. Krummrich, F.H. Langford, K. Langland,
S.P. McKinney, D.M. Powell, and P.L. Shafland. 1985.
Lake habitat and fishery resources of Florida. Pages
54-69 in Seaman, Jr., (ed.), Florida Aquatic Habitat
and Fishery Resources. American Fisheries Society, New
York, 543 pp.
WYNGAARD, G.A., J.L. Elmore and B.C. Cowell. 1982. Dynamics
of a subtropical plankton community, with emphasis on
the copepod Mesocvclops edax. Hydrobiologia 89:39-48.
YENTSCH, C.S. and D.W.Menzel. 1963. A method for the
determination of phytoplankton chlorophyll and
phaeophytin by fluorescence. Deep Sea Res. 10:221-231.
ZALE, A.V. 1984. Applied aspects of the thermal biology,
ecology and life history of the blue tilapia, Tilapia
aurea (Pisces: Cichlidae). Technical Report no. 12,
Florida Cooperative Fish & Wildlife Research Unit,
Gainesvile, FL. 196 pp.

157
APPENDIX A
Map of the study area

158
APPENDIX B
Lake Bonny sedimentation rate (g.cm"2.yr)
by year.
LAKE BONNY
Core
Sect
(cm)
Bulk
Density
(g/cm3)
Unsupp.
Pb-210
(pCi/g)
Unsupp.
Pb-210
(pCi/c m2
Cum.Res.
Uns.Pb-21
(pCi/cm2)
Age
(yrs BP)
(1993)
Sed.
Rate
(g/cm2.yr)
Cs-137
(PCi/g)
0
4
0.029
18.72
2.17
31.84
0.00
0.03
1.3
4
7
0.030
12.30
1.84
29.67
2.27
0.08
3
7
9
0.061
12.30
1.33
27.83
4.32
0.07
4.13
9
11
0.066
11.22
1.48
26.31
6.13
0.07
4.01
11
13
0.074
8.46
1.23
24.83
7.99
0.09
3.63
13
13
0.078
8.14
1.27
23.37
9.65
0.09
3.20
13
17
0.079
3.27
0.83
22.30
11.43
0.13
4.02
17
19
0.090
6.99
1.26
21.47
12.63
0.10
4.36
19
22
0.092
8.92
2.46
20.21
14.39
0.07
4.49
22
26
0.092
6.31
2.39
17.75
18.76
0.08
3.28
26
30
0.090
7.00
2.32
13.36
23.40
0.07
2.67
30
34
0.089
6.00
2.14
12.84
29.16
0.07
1.07
34
38
0.091
3.80
1.38
10.70
33.01
0.09
0.83
38
42
0.102
3.43
1.40
9.32
39.44
0.08
0.92
42
46
0.164
3.48
2.28
7.92
44.67
0.07
0.87
46
30
0.174
3.17
2.21
3.64
33.56
0.06
0.78
30
34
0.142
1.91
1.08
3.43
71.32
0.06
0.91
34
38
0.137
1.94
1.06
2.33
83.64
0.04
0.39
38
64
0.124
0.00
-0.10
1.29
102.86
ERR
0.14
64
72
0.123
0.13
1.39
100.56
ERR
0.33
72
84
0.126
1.24
1.24
104.23
ERR
0.37

159
APPENDIX C
Lake Gibson sedimentation rate (g.cm"2.yr)
by year.
LAKE GIBSON
Core
Sect
(cm)
Bulk
Density
(g/cm3)
Unsupp.
Pb-210
(pCi/g)
Unsupp.
Pb-210
(pCi/cm2
Cum.Res.
Uns.Pb-21
(pCi/cm2)
Age
(yrs BP)
(1993)
Sed.
Rate
(g/cm2.yr)
Cs-137
(pCi/g)
0
2
0.113
11.86
2.68
28.29
0.00
0.07
4.41
2
4
0.167
9.48
3.16
23.61
3.20
0.08
3.02
4
6
0.221
9.31
4.11
22.43
7.43
0.08
4.18
6
8
0.233
6.76
3.13
18.34
13.92
0.08
3.63
8
10
0.271
3.72
3.10
13.19
19.97
0.08
2.96
10
12
0.284
6.07
3.43
12.09
27.30
0.06
2.96
12
14
0.229
2.33
1.07
8.64
38.09
0.12
1.32
14
16
0.178
2.63
0.94
7.37
42.32
0.09
1.28
16
18
0.168
2.61
0.88
6.64
46.36
0.08
0.84
18
20
0.160
3.34
1.07
5.76
51.11
0.05
0.75
20
24
0.161
2.29
1.47
4.69
37.70
0.06
0.71
24
28
0.183
1.11
0.82
3.22
69.78
0.09
0.73
28
32
0.208
1.42
1.18
2.40
79.22
0.03
0.37
32
36
0.183
2.00
1.46
1.22
100.93
0.02
0.30
36
40
0.161
1.42
0.91
-0.24
ERR
-0.01
0.25
40
44
0.177
0.12
0.08
-1.13
ERR
-0.30
0.37
44
48
0.269
1.41
1.32
-1.23
ERR
-0.03
0.19
48
32
0.317
0.39
0.81
-2.73
ERR
0.22
0.03
32
36
1.167
0.00
-0.93
-3.36
ERR
ERR
0.04
36
60
1.644
0.00
-2.63
-2.63
ERR
ERR
-0.05

APPENDIX D
Physical and chemical properties of Lake Parker
sediment core (cf. Schelske et al., 1992)
Interval
Mid-depth
Rho
Organic
Ctot
Ntot ,
^tot .
(mg. g'1)
(cm)
(cm)
(g. dry. cm'3)
Matter (%)
(%)
(mg. g"1)
0-2
1
0.01820
71.7
34.9
63.7
4.9
2-4
3
0.02575
71.4
35.3
N/A
N/A
4-6
5
0.02702
70.1
34.3
71.8
5.5
6-8
7
0.02895
63.4
31.9
N/A
N/A
8-10
9
0.03554
59.1
28.6
23.3
4.5
10-12
11
0.03667
54.0
26.3
N/A
N/A
12.-14
13
0.03652
52.1
27.2
23.5
4.5
14-16
15
0.04000
52.0
26.6
N/A
N/A
16-18
17
0.03919
52.3
27.7
25.4
4.9
18-20
19
0.04368
50.1
26.8
22.7
5.1
22-24
23
0.04846
48.6
24.9
24.8
5.5
26-28
27
0.04934
47.7
24.6
25.6
5.9
30-32
31
0.05047
48.8
26.3
22.7
6.7
34-36
35
0.05258
48.4
26.6
23.8
6.5
38-40
39
0.05625
41.5
20.7
22.3
6.0
42-44
43
0.05751
37.1
20.4
19.0
6.5
46-48
47
0.05741
37.4
21.7
19.6
6.8
o

APPENDIX E
Physical and chemical properties of Lake Parker
sediment core (cf. Schelske et al., 1992) Cont.
Interval
(cm)
Mid-depth
(cm)
Rho
(g.dry. cm*3)
Organic
Matter (%)
Ctot
(%)
N.nt
tot 1
(mg.g*1)
- -
^tot .
(mg.g*1)
50-52
51
0.06073
40.4
23.4
18.8
5.4
54-56
55
0.05879
41.0
21.6
21.4
5.4
58-60
59
0.06078
49.1
28.8
26.8
5.7
60-62
63
0.06016
44.3
23.6
25.0
5.7
66-68
67
0.07233
49.6
28.3
25.6
5.5
70-72
71
0.07785
54.6
31.3
28.0
4.5
74-76
75
0.07166
54.3
30.9
29.2
4.1
78-80
79
0.06960
55.0
31.0
28.6
4.2
82-84
83
0.07834
56.6
32.4
28.3
4.1
86-88
87
0.07616
57.5
32.6
23.2
4.7
90-92
91
0.08900
55.8
31.3
25.1
3.4

162
APPENDIX F
Physical and chemical properties of Lake
Hollingsworth sediment core (cf. Schelske et al., 1992).
Depth
(cm)
Density
(g. dry. cm'3. wet)
Organic
Matter
(%LOI)
Ctot
(%>
Ntot
(%)
^tot .
(mg. g'1)
0-2
0.01772
54.3
27.6
3.04
6.99
2-4
0.02082
51.8
27.4
3.35
7.24
4-6
0.02487
54.5
27.4
2.89
6.42
6-8
0.02693
52.8
26.0
3.02
7.32
8-10
0.02846
52.8
26.7
N/A
N/A
I 10-12
0.03139
52.2
26.1
2.92
7.03
12-14
0.03583
48.7
25.5
2.76
6.47
I 14-16
0.03796
48.8
25.4
2.71
6.58
16-18
0.04688
40.7
20.8
2.36
5.43
18-20
0.06550
30.5
14.9
2.17
5.13
20-22
0.09093
24.2
13.9
1.34
5.18
22-24
0.08989
25.8
13.8
1.66
6.47
24-26
0.09053
28.9
15.0
1.67
8.87
26-28
0.08030
34.3
17.3
1.97
9.78
28-30
0.08537
30.6
16.3
1.53
9.64
30-32
0.10669
25.7
14.8
1.35
7.52
40-42
0.07557
43.4
24.3
2.43
6.22
50-52
0.07180
49.8
27.8
2.70
4.30
| 60-62
0.07493
55.5
34.3
3.04
3.57
70-72
0.07412
56.7
33.8
3.16
3.96
80-82
0.06584
62.0
35.3
3.53
3.20
90-92
0.07100
55.5
34.4
2.99
2.29
98-100
0.08207
48.4
28.4
2.81
2.48

APPENDIX G
Sediment core dating for Lake Hollingsworth
(cf. Schelske et al., 1992).
Depth
(cm)
Total 210Pb
(dpm.g-1)
214Bi
(dpm. g 1)
Excess 210Pb
(dpm.g-1)
137Cs
(dpm. g'1)
Age
(years)
Date
(AC)
Sed.rate
(g. cm-2. yr"1)
0-4
31.41
7.90
23.62
2.79
1.5
1990
0.051
4-8
32.98
8.25
24.84
3.06
3.8
1988
0.046
8-12
29.01
8.32
20.78
3.04
6.2
1986
0.051
12-16
28.63
7.46
21.26
3.28
9.4
1983
0.045
16-18
24.99
5.93
19.17
3.06
11.4
1981
0.046
18-20
18.24
6.54
11.77
2.84
13.3
1979
0.071
20-22
19.40
7.53
11.93
3.69
16.1
1976
0.065
22-24
20.12
9.35
10.82
2.89
18.8
1973
0.066
24-26
23.04
11.99
11.11
2.29
21.9
1970
0.059
26-28
26.71
13.99
12.80
3.17
25.4
1967
0.046
28-30
21.73
10.54
11.26
2.58
29.0
1963
0.047
30-32
18.20
9.14
9.11
1.57
33.2
1959
0.051
32-40
17.05
8.70
8.39
1.20
51.6
1940
0.040
40-42
15.90
8.26
7.67
0.82
56.7
1935
0.030
42-50
11.51
6.46
5.07
0.70
75.5
1916
0.031
a\
u>

APPENDIX H
Sediment core dating for Lake Hollingsworth
(cf. Schelske et al., 1992). Cont.
Depth
(cm)
Total 210Pb
(dpm. g~1)
214Bi
(dpm. g"1)
Excess 210Pb
(dpm.g-1)
137Cs
(dpm.g'1)
Age
(years)
Date
(AC)
Sed. rate
(g.cm"2.yr"1)
50-52
7.13
4.66
2.48
0.59
78.7
1913
0.045
52-60
5.83
3.68
2.16
0.47
93.7
1898
0.039
60-62
4.53
2.70
1.84
0.35
98.2
1894
0.034
62-70
5.22
2.85
2.38
0.33
144.8
1847
0.013
70-72
5.91
3.00
2.92
0.31
N/A
N/A
N/A
72-80
4.68
3.06
0.00
0.21
N/A
N/A
N/A
80-82
3.44
3.11
N/A
0.11
N/A
N/A
N/A
82-90
3.56
3.08
N/A
0.12
N/A
N/A
N/A
90-92
3.68
3.06
N/A
0.12
N/A
N/A
N/A
92-98
4.11
2.78
N/A
0.17
N/A
N/A
N/A
o>

BIOGRAPHICAL SKETCH
Carlos A. Fernandes was born in Mossor, State of Rio
Grande do Norte, Brazil, on April 11, 1950. He received a
Bachelor of Science degree in molecular biology from
University of Brasilia, Brasilia, Brazil, in 1976. He began
working as a professor at the University of Brasilia, in
1977. He received his Master of Science degree in ecology,
with area of specialization in limnology from University of
Brasilia, in 1979. Later that year he started working as a
biologist for the Water and Wastewater Company of Brasilia
(CAESB). From 1984 to 1989 he worked as a director for
Pollution Control for the Secretariat of Environment,
Science and Technology (SEMATEC), in Brasilia. His desire
for more knowledge brought him to the University of Florida
in 1990. He earned a Doctor of Philosophy degree from the
University of Florida, Department of Environmental
Engineering Sciences in 1994. He is a mostly proud father of
two boys, Carlos Filho and Diego.
165

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Thomas L. Crisman, Chair
Professor of Environmental
Engineering Sciences
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
G. Ronnie Best
Scientist of Environmental
Engineering Sciences
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Edward J.
Associate Professor of
Forest Resources and Conservation
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Frank G. Nordlie
Professor of Zoology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Pjjilosophy.
Horst 0. Schwassmann
Professor Emeritus of Zoology

This dissertation was submitted to the Graduate Faculty
of the College of Engineering and to the Graduate School and
was accepted as partial fulfillment of requirements for
the degree of Doctor of Philosophy.
May 1995 /
yvwinfred M. Phillips
/ Dean,College of Engineering
Karen A. Holbrook
Dean, Graduate School



56
biological parameters are provided in Chapter 5 of this
study. Both are small urban lakes subject to anthropogenic
pressures in the form of industrial, residential and
commercial development in the area and recreational uses
such as boating, fishing, and skiing. In addition, there are
agricultural activities at Lake Gibson watershed.
Study Sites
Lake Bonny is a small (144 ha of surface area), shallow
(zmax= 2.5 m), urban lake bordered by the city of Lakeland,
(Polk County, Florida). It has no known discharges or
withdrawals (Polk County Water Resources Division, 1990).
The population of Lakeland has increased from less than
500 inhabitants in 1880 to nearly 75,000 people in 1993. As
a result, all of the lake watershed was developed for
commercial and residential uses. Lake Bonny has consistently
had poor water quality (Polk County Water Resources
Division, 1990) with an average 1992 TSI of 73.5. Urban
runoff from commercial and residential development in the
watershed and low lake levels resulting from seepage have
been a major problem. Extremely poor water quality was
recorded in 1986 at a time of extreme low lake levels (Polk
County Water Resources Division, 1990). Based on current
data, the TSI was 98; total nitrogen 9.0 mg.l'1, total
phosphorus 0.7 mg.l-1, Secchi 0.2 m and chlorophyll a was
220 mg.nf3.


Table 3-4. Phytoplankton biomass (mg.dry weight-1) in fish excretions
calculated from chlorophyll a values.
FISH + ALGA MEDIUM
INITIAL VALUE
mg/dry weight
FINAL VALUE
mg/dry weight
FIN-INIT (X)
mg/dry weight
(X)-CTRL.
mg/dry weight
Til (-N)+ Bluegreen
68054
43663
-24391
-23939
Til (-N)+ Green
62232
84297
22065
23269
Til (-P)+ Bluegreen
79421
33651
-45770
-51642
Til (-P) + Green
68756
46273
-22483
-43562
Shad (-N)+ Bluegreen
2710
38995
36285
36737
Shad (-N)+ Green
4818
62383
57565
58769
Shad (-P) + Bluegreen
3061
32019
28958
23086
Shad (-P)+ Green
2810
81454
78644
57565
Ctrl (-N)+ Bluegreen
903
451
-452
Ctrl (-N)+ Green
1882
678
-1204
Ctrl (-P)+ Bluegreen
502
6374
5872
Ctrl (-P) + Green
2308
23387
21079


144
, T.J. Whitmore, M.S. Flannery, and M.W. Binford.
1993. Paleolimnological methods for defining target
conditions in lake restoration: Florida case studies.
Lake and Reserv. Manage. 7(2) :209-217.
BROOKS, J.L. and S.I. Dodson, 1965. Predation, body size and
composition of plankton. Science 150:28-35.
CAILTEUX, R.L. 1988. Food resource utilization by two
population of blue tilapia, Tilapia aurea, in north
central Florida. M.Sc. thesis, University of Florida,
Gainesville.
CAIRD, J.M. 1945. Algae growth greatly reduced after
stocking pond with fish. Water Works Engineering
98:240.
CAMPBELL, K.M. 1986. Geology of Polk County, Florida.
Florida Geological Survey. Open File Report no. 13.
CANFIELD, Jr., D.E. 1981. Chemical and trophic state
characteristics of Florida lakes in relation to
regional geology. Final report to Coop. Fish and
Wildlife Res. Unit., University of Florida,
Gainesville.
, and M.V. Hoyer. 1988. Regional geology and the
chemical and trophic state characteristics of Florida
lakes. Lake and Reserv. Manage. 4:21-31.
, and M.V. Hoyer. 1991. A characterization of the
fish population in lake Parker (Polk County), Florida.
Final report to SWFWMD. Brooksville.
, and M.V. Hoyer. 1992. Aguatic macrophytes and their
relation to the limnology of Florida lakes. Final
report to Bureau of Aguatic Plant Management. FDNR,
Tallahassee.
CARPENTER, S.R. and J.F. Kitchell. 1988. Consumer control of
lake productivity. Bioscience 38:764-769.
, J.F. Kitchell and F.R. Hodgson. 1985.
Cascading trophic interactions and lake productivity.
Bioscience. 35:634-639.
, J.F. Kitchell, J.R. Hodgson, P.A. Cochran, J.J.
Elser, M.M. Elser, D.M. Lodge, D. Kretchmer & X. He,
1987. Regulation of lake primary productivity by food
web structure. Ecology 68: 1863-1876.


Table 3-5. Caloric content of gizzard shad
feces.
SHAD
CALORIC CONTENT
(Cal/g)
Group 1 (A)
4115.56
(B)
4134.18
(C)
4150.35
Group 2 (A)
4066.21
(B)
4221.81
(C)
4196.37
Group 3 (A)
4026.14
(B)
3935.90
Group 4 (A)
3894.76
(B)
3867.86
(C)
3842.40
Table 3-6. Caloric content of blue tilapia
feces.
TILAPIA
CALORIC CONTENT
(Cal/g)
Group 1 (A)
2532.20
(B)
2490.92
Group 2 (A)
2722.98
(B)
2648.89
(C)
2926.08
Group 3 (A)
2533.76
(B)
2560.45
(C)
2417.74
Group 4 (A)
2312.26
(B)
2186.88


BIOGRAPHICAL SKETCH
Carlos A. Fernandes was born in Mossor, State of Rio
Grande do Norte, Brazil, on April 11, 1950. He received a
Bachelor of Science degree in molecular biology from
University of Brasilia, Brasilia, Brazil, in 1976. He began
working as a professor at the University of Brasilia, in
1977. He received his Master of Science degree in ecology,
with area of specialization in limnology from University of
Brasilia, in 1979. Later that year he started working as a
biologist for the Water and Wastewater Company of Brasilia
(CAESB). From 1984 to 1989 he worked as a director for
Pollution Control for the Secretariat of Environment,
Science and Technology (SEMATEC), in Brasilia. His desire
for more knowledge brought him to the University of Florida
in 1990. He earned a Doctor of Philosophy degree from the
University of Florida, Department of Environmental
Engineering Sciences in 1994. He is a mostly proud father of
two boys, Carlos Filho and Diego.
165


106
Table 5-9. Summary of water quality data for L. Hollingsworth.
Numbers shown are mean values with ranges in parentheses.
| Parameter
Unit
Canfield
(1981)
Several
Agencies *
(1989/91)
This study
(1992/93)
| Temperature
C
N/A
25.8
(15.5-33.3)
25.3
(18.5-30.9)
D.O.
mg/L
N/A
9.7
(4.8-12.8)
9.4
(6.7-11.0)
PH
9.4
(8.8-10.1)
9.4
(7.8-11.3)
8.7
(7.8-10.43)
Alkalinity
mg/L CaC03
47
(32-56)
52.4
(38-66)
47.2
(40-58)
1 Conduct.
pmhos/cm
183
(165-205)
189.8
(142-230)
181.5
(158-209)
Nitrogen
mg/L
N/A
4.04
(0.75-12.2)
5.41
(2.8-8.1)
1 nh3-n
mg/L
N/A
0.07
(0.006-0.3)
0.167
(0.036-0.7)
I TKN
mg/L
N/A
3.92
(0.75-11.6)
5.15
(2.35-7.9)
| Phosphorus
mg/L
0.774
(0.32-1.2)
0.254
(0.015-2.2)
0.268
(0.081-0.61)
| Chi. a
mg/m3
54.8
(26.7-104)
182.59
(30.1-368.1)
186.32
(64.5-285.9)
| Color
Pt units
5
(5-5)
34.7
(6-89)
58.5
(37-94)
I Secchi
m
1.6
(0.8-2.4)
0.3
(0.2-0.4)
0.22
(0.15-0.4)
1 Bacteria
x 106
N/A
2417.15
(1-17200)
800
(40-3700)
1 TSI
N/A
89.46
(70.3-103.1)
98.27
(88.2-105.6)
(*) AGENCY CODES:
USGS; EPA; PCWR; CL; UFC; FGFWFC; CF208; and SWFWMD.


107
was of the genus Diaptomus, but it did not represent a large
component of the zooplankton population. Rotifers had an
estimated density of 286,000 individuals .nf3, and
cladocerans amounted to 68,700 individuals .m-3 (Canfield and
Hoyer, 1992).
Sixteen species of fish were collected from Lake
Hollingsworth (Canfield and Hoyer, 1992). The fish
population from 1965 to 1969 for this lake was described as
having an overabundance of forage fish such as gizzard shad,
threadfin shad, and stunted bluegill (Ware et al., 1971),
and total fish biomass estimated with littoral blocknets
ranged from 70 to 257 kg.ha-1 (Ware et al., 1971). The
littoral nets in the Canfield and Hoyer report of 1992
averaged 1050 kg.ha-1. The most abundant species collected
by open-water experimental gillnets were gizzard shad and
black crappie (Canfield and Hoyer, 1992).
ANOVAs were performed using the General Linear Model
(GLM) (SAS, 1989) in order to identify significant seasonal
limnological factors (p < 0.05). The following parameters
were analyzed: Secchi, temperature, pH, conductivity,
dissolved oxygen, total suspended solids, turbidity,
chlorophyll a, color, alkalinity, NH3, TKN, total nitrogen,
total phosphorus, orthophosphate, and total bacteria.
Secchi, temperature, conductivity, turbidity,
chlorophyll a, color, total phosphorus and total bacteria
are the parameters that exhibited significant seasonal