1 EFFECTS OF POND AERATION METHODS AND POST HARVEST HOL DING TREATMENTS ON ZEBRA DANIO GROWTH AND SURVIVAL By NATALIE STECKLER A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF T HE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2012
2 2012 Natalie Steckler
3 To my family
4 ACKNOWLEDGMENTS Funding for this project and my graduate assistantship was provided b y the United States Department of Agriculture (USDA) and the University of Florida Thank you also to the University of Florida College of Agricultural and Life Sciences for providing my graduate assistantship in the Program of Fisheries and Aquatic Scien ces. I am grateful to my major professor, Dr. Roy Yanong, whose wealth of knowledge on ornamental fish medicine has taught me much within the field. I would like to thank my advisory committee members, Dr. Jeff Hill and Dr. Kathleen Hartman, not only for providing insight in their respective fields but for their helpful advice in matters of life and career. I am grateful to the faculty, staff, and students at the University of Florida Tropical Aquaculture Laboratory who contributed time and effort to var ious aspects of this project, including Kevin Barden, Yuri Barreda, Jeremy Butts, Eric Cassiano, Daniel Conner, Christine Creamer, Dr. Claire Erlacher Reid, Dr. Mark Flint, Dr. Trevor Gerlach, Scott Graves, Larry Lawson, Carlos Martinez, Deborah Pouder, So nya Sampson, Craig Watson, Po Ting Wong, and Amy Wood. I want to thank Dr. Chuck Cichra for his patient and enthusiastic help with statistical analyses and for being so supportive of all scientific endeavors that advance the fields of fisheries and aquacu lture. Finally, I want to acknowledge those who donated supplies and equipment for use in this study. Thank you to Claude Brown, Mark Hoyer, and the Florida LAKEWATCH laboratory for their donation of chlorophyll testing supplies, and for kindly offering t heir time and services to analyze samples. I am also grateful to Ty Simmons and Steve Simmons Aquatics, Inc. for their extremely generous donation of over 1 million zebra danio fry, and to Kasco and Colorite Wat erworks for their contributions of aerators and aerator materials.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 A BS TRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ ..... 12 2 LITERATURE REVIEW ................................ ................................ ........................... 15 Pond Production in Florida ................................ ................................ ...................... 15 Aeration in Aquaculture ................................ ................................ ........................... 19 Zebra Danio Species Profile ................................ ................................ ................... 22 3 PRELIMINARY INVESTIGATION OF AERATION METHODS IN SMALL AQUACULTURE PONDS ................................ ................................ ....................... 28 Overview ................................ ................................ ................................ ................. 28 Methods ................................ ................................ ................................ .................. 28 Pond Preparation ................................ ................................ ............................. 28 Aeration Methods ................................ ................................ ............................. 29 Dissolved Oxygen and Temperature ................................ ................................ 30 Water Chemistry and Quality ................................ ................................ ............ 31 Statistical Analysis ................................ ................................ ............................ 31 Results ................................ ................................ ................................ .................... 32 Discussion ................................ ................................ ................................ .............. 33 4 AERATION STUDY IN SMALL OUTDOOR PONDS STOCKED WITH ZEBRA DANIOS ................................ ................................ ................................ .................. 43 Overview ................................ ................................ ................................ ................. 43 Methods ................................ ................................ ................................ .................. 44 Pond Preparation ................................ ................................ ............................. 44 Aeration Methods ................................ ................................ ............................. 45 Zebra Danio Stocking ................................ ................................ ....................... 46 Pond Maintenance ................................ ................................ ........................... 47 Dissolved Oxygen and Temperature ................................ ................................ 48 Water Chemistry and Quality ................................ ................................ ............ 49 Chlorophyll a ................................ ................................ ................................ .... 50 Zooplankton Densities ................................ ................................ ...................... 50
6 Danio Growth ................................ ................................ ................................ ... 51 Danio Health Assessments ................................ ................................ .............. 52 Pond Harvest ................................ ................................ ................................ .... 53 S tatistical Analysis ................................ ................................ ............................ 54 Results ................................ ................................ ................................ .................... 55 Zebra Danio Growth, Survival, and Health ................................ ....................... 55 Dissolved Oxy gen and Temperature ................................ ................................ 57 Pond Productivity ................................ ................................ ............................. 58 Discussion ................................ ................................ ................................ .............. 60 5 POST HARVEST MEDICATION STUDY USING ZEBRA DANIOS ........................ 84 Overv iew ................................ ................................ ................................ ................. 84 Methods ................................ ................................ ................................ .................. 86 Pond Harvest ................................ ................................ ................................ .... 86 Medication Trial ................................ ................................ ................................ 87 Examined Medications ................................ ................................ ..................... 89 Biosecurity and Tank Maintenance ................................ ................................ .. 91 Statistical Analysis ................................ ................................ ............................ 92 Results ................................ ................................ ................................ .................... 92 Discussion ................................ ................................ ................................ .............. 93 6 CONCLUSIONS ................................ ................................ ................................ .... 100 LIST OF REFERENCES ................................ ................................ ............................. 104 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 113
7 LIST OF TABLES Table page 3 1 Mean dissolved oxygen and other water quality and chemistry values in small fishless pond s aerated using one of three methods ................................ 41 3 2 Nitrogen components in small fishless ponds following fertilization with cottonseed meal ................................ ................................ ................................ 42 4 1 Harvest data (mean SE) of zebra danios following a 10 week growout period in small outdoor ponds using various aeration methods ......................... 80 4 2 External biopsy findings on zebra danios examined three days prior to harvest from small ponds equipped with various aeration methods ................... 81 4 3 Mean dissolved oxygen and other water quality and chemistry values in s mall ponds stocked with zebra danio fry and aerated using various methods 82 4 4 Nitrogen components in small ponds stocked with zebra danio fry following fertilization with cottonseed m eal ................................ ................................ ....... 83 5 1 Ranked categorical scoring system used by judges for evaluation of zebra danio appearance and behavior (Kilgore et al. 2009) ................................ ...... 98 5 2 External biopsy findings on zebra danios examined before and after a 14 day holdin g period with one of three treatments ................................ ........................ 99
8 LIST OF FIGURES Figure page 2 1 Photograph of a typical earthen pond used in ornamental fish production ........ 27 3 1 Aerial photograph of the University of Florida Tropical Aquacult ure Laboratory ................................ ................................ ................................ ......... 37 3 2 Photograph of a diffuser box aerator designed for the fishless pond aeration study ................................ ................................ ................................ ................... 38 3 3 Schema tic indicating aerator location and water flow in small fishless ponds .... 39 3 4 Dissolved oxygen and oxygen saturation over a 6 week period in small fishless ponds aerated using three meth ods. ................................ .................... 40 4 1 Schematic indicating aerator location and water flow in small aquaculture p onds stocked with zebra danios ................................ ................................ ........ 68 4 2 Photograph of a diffuser box aerator designed for zebra danio pond aeration study ................................ ................................ ................................ ................... 69 4 3 Growth curves over a 10 week period for zebra danios raised in small pond s aerated using va rious methods ................................ ................................ ........... 70 4 4 Photograph illustrating size difference among zebra danio fry raised in small ponds using various aeration methods ................................ .............................. 71 4 5 Scatterplot of zebra danio average individual fish weight (g) vs. harvest yield (# of fish/pond) ................................ ................................ ................................ ... 72 4 6 Length frequency histograms of zebra danios after 10 wee ks of growth in small outdoor ponds aerated using various methods ................................ ........ 73 4 7 Comparison photographs of a normal zebra danio compared to a severely kyphotic fish ................................ ................................ ................................ ....... 74 4 8 Dissolved oxygen and oxygen saturation measurements over a 10 week period in small ponds stocked with zebra danio fry and aerated using three methods ................................ ................................ ................................ .............. 75 4 9 Diurnal oxygen curves from a 24 hour s tudy in small ponds stocked with zebra danio fry and aerated using various methods ................................ .......... 76 4 10 Oxygen profiles constructed from measurem ents in small ponds aerated using various methods ................................ ................................ ....................... 77
9 4 11 Total ammonia nitrogen and nitrite concentrations following cottonseed meal application in small ponds stocked with zebra danio fry ................................ ..... 78 4 12 Weekly z oopl ankton densities d uring a 10 week study for ponds stocked with zebra danio fry and aerated using various methods ................................ .......... 79 5 1 Zebra danio appearance scores before and after a 14 day holding period in aquaria with one of four treatments ................................ ................................ .... 96 5 2 Zebra danio behavior scores before and after a 1 4 day holding period in aquaria with one of four treatments ................................ ................................ .... 97
10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of M aster of Science EFFECTS OF POND AERATION METHODS AND POST HARVEST HOL DING TREATMENTS ON ZEBRA DANIO GROWTH AND SURVIVAL By Natalie Steckler December 2012 Chair: Roy P.E. Yanong Major: Fisheries and Aquatic Sciences Low dissolved oxygen (DO) increas es risk of disease and mortality in cultured fish; however, ae ration methods have not been extensively researched for pond production of ornamental fish. This study compared three pond aeration methods and four post harvest holding medications for their e ffect s on zebra danio Danio rerio growth and survival. A 6 week study was performed in fishless ponds using three aeration types: air stones (AS); vertical pump (VP) ; and fine bubble diffuser box (DB) These aerators were then compared in ponds stocked w ith zebra danio fry. Fry were raised for 10 weeks and harvested to evaluate fish size and yield Dissolved oxygen, water temperature, ammonia, nitrite, pH, Secchi depth, alkalinity, hardness, c hlorophyll a and zooplankton densities were evaluated. Pr e dawn mean DO was highest in VP ponds, followed by DB then AS. Nearly all other water chemistry and quality parameters were simila r among aerator types, although high ammonia and nitrite took significantly longer to decrease in AS ponds Also, oxygen st ratification was observed only in AS ponds. Danio growth rate was
11 much faster in VP ponds versus DB and AS ponds ; however, survival was lowest in VP ponds. In contrast, AS ponds had the highest survival rates and the smallest fish. DB p onds had intermed iate fish size and yield The inverse relationship between size and yield suggests density dependence. Overall fish health was good for all aeration methods All danios in the post harvest holding treatment study survived, with no difference in saleabil ity across treatments (bronopol, diquat, hydrogen peroxide, control).
12 CHAPTER 1 INTRODUCTION Since its beginning in the 1930s, ornamental fish production in Florida has grown e industry. Ninety five percent of freshwater tropical ornamental fish raised in the United States are currently produced in Florida. Farmgate income to Florida producers reached an all time industry high of $57.2 million in 1997 (FASS/FDACS 2001, 2002), and t he latest survey reported $33.2 million in farmgate value for ornamental fish produced in 2005 (FASS/FDACS 2005) Florida raised ornamental species are sold and distributed within the United States and exported to many other countries. There is a shortage of published research studying ornamental species, especially when compared to larger industries such as food fish culture; thus, there is great potential to improve current production practices in the ornamental industry. Many traditional orname ntal husbandry practices are based either on empirical (or qualitative) observations or on accepted methods for non ornamental species such as food fish. Because food fish are larger and, in many cases, more robust than small bodied ornamentals, husbandry methods derived from food fish research may be inappropriate. There fore, there is a need for research directed specifically towards improvement of ornamental fish husbandry practices. In particular, there is relatively little scientific research concerni ng the use of aeration in ornamental fish production. Dissolved oxygen (DO) is a critical component for fish health Low oxygen events account for more fish mortalities than any other water quality parameter (Boyd 1990). Despite the importance of DO for fish health and
13 survival, aeration standards have not been scientifically established for ornamental fish production. Another potential cause of disease and mortality occurs when fish are harvested for sale. Handling associated with harvest can cause phy sical trauma, and crowded conditions during holding and transport can quickly lower DO and compromise water quality, exposing fish to toxic levels of ammonia. Stress associated with handling and transport may increase susceptibility to pathogens, thereby decreasing yield and qual ity of fish for sale (Noga 2010; Pasnik et al. 2010). The goal of this research was to study two important factors in ornamental fish production: 1) the method of pond aeration and 2) medical treatment of fish after harvest. Spec ifically, three pond aeration methods (the most commonly used method in ornamental production and two newer methods) were studied for their influence on zebra danio Danio reri o growth and survival. Pond physical, biological, and chemical parameters were a lso examined for differences in water quality and productivity as a result of aeration method. After harvest, danios were treated with one of three broad spectrum medications (bronopol, hydrogen peroxide, and diquat). Fish survival, appearance and behav ior were compared to determine relative saleability before and after the application of holding treatments. Because relatively little published literature exists in ornamental fish production, this research may potentially help producers improve current production practices. To date, supplemental aeration is not universally implemented in ornamental pond production, and low dissolved oxygen is an ongoing health concern, particularly in the warm Florida climate in which the majority of domestically farmed freshwater
14 ornamental species are produced. Also, stress induced disease outbreaks occasionally lead to significant losses after harvest. Improvements in pond production and harvest techniques can increase survival and quality of farmed ornamental fish and lead to a stronger aquaculture industry for the state.
15 CHAPTER 2 LITERATURE REVIEW Several topics relevant to the studies described in Chapters 3 5 will be discussed in this literature review. In particular, the subjects of ornament al fish production, husbandry practices, and commonly cultured species will be addressed. Pond Production in Florida Ornamental fish culture in Florida began in the 1930s, and it has since expanded into a multi million dollar industry for the state (Wats on 1999). Florida is an ideal location for outdoor pond culture of tropical fish. The year round warm climate falls within the optimal temperature ranges for thousands of freshwater ornamental species (Shireman and Lindberg 1985). Also, the combination of flat topography, high aquifer (water table), and high soil permeability that exists in the central and southern portions of the state is ideal for the use of excavated earthen ponds ( Haman et al. 1991; Watson and Shireman 1996). The highest concentrati on of ornamental facilities in the state is located in Hillsborough and Polk counties. A large international airport located in Tampa is conveniently close, making transportation of Florida farmed fish easy and fast (Miller Morgan 2010). Outdoor ornamenta l production ponds in central and south Florida are usually kept small, at a surface area of approximately 8 x 25 m and a maximum depth of 2 m (Watson and Shireman 1996; Figure 2 1). allows ponds to retain water wit hout the use of a liner, and water is supplied from the shallow aquifer Ground water in central and south Florida is typically hard and alkaline, with total alkalinity and total hardness each exceeding 100 mg/L CaCO 3 and pond pH ranges from 8 to 9 in mo st locales (Watson and Shireman 1996). The wide variety of
16 ornamental fish raised in small outdoor ponds is remarkable. Thousands of fish species from several families, including minnows (Cyprinidae), cichlids (Cichlidae), tetras (Characidae), rainbowfish es (Melanotaeniidae and Pseudomugilidae) and catfishes (Loricariidae and Callichthyidae) are currently produced in Florida (Hill and Yanong 2002). Ponds are typically prepared for stocking by draining, then pressure washing the banks to remove the loose upper layer of detritus and organic material. Once ponds are cleaned, hydrated lime (calcium hydrox ide [CaOH]) is applied evenly to the substr ate to quickly raise pH above 11 and target remaining organisms in the pond, including potential predators, pests and pathogens (Arce and Boyd 1975). Hydrated lime also acts as a temporary source of hardness and alkalinity for ponds. Alkalinity is the collective term for all titratable bases in the water, including carbonate, bicarbonate, and hydroxide, that buffe r against pH fluctuations. Alkalinity levels in ponds are primarily influenced by the soil type. Most central and southern Florida ponds have naturally high alkalinity and do not require a supplemental alkalinity source; however, if alkalinity is lower than 50 mg/L, a long acting source of lime, such as agricultural limestone (CaCO 3 ), is also recommended to increase alkalinity (Boyd 1990). Limestone also acts as a long term source of water hardness (i.e., the total concentration of divalent cations, pr imarily calcium and magnesium). Water hardness is important for egg structure and musculoskeletal development in fry, as well as metabolism and acid/base balance in all life stages of fish (Lall and Lewis McCrea 2007). After the addition of lime, a fert ilizer source is also added to the pond. The most commonly used fertilizer in Florida ornamental production ponds is cottonseed meal (C.
17 Martinez, University of Florida, personal communication). Fertilizer provides the pond with vital nutrients, primaril y nitrogen, phosphorus, and potassium, which are needed to support macrophyte and phytoplankton populations. Nutrients initially stimulate a bloom in phytoplankton, which in turn serves as the base of the food web for pond organisms (Boyd and Tucker 1992, Kurten et al. 1999). The smallest zooplankton, rotifers, increase in number first and are followed by increasingly larger plankton (copepods and cladocerans). This process, known as succession, can take up to several weeks to complete after fertilizatio n, and timing is dependent on several factors including temperature, w eather, and fertilizer dose (Ludwig et al. 1998). Phytoplankton and zooplankton both serve as vital food sources for small fish in a pond system. The size of plankton consumed depends on fish size and age. Rotifers are an ideal first food for fry because their size (0.04 2.5 mm body length) is smaller than the gape of many fish species (Hoff and Snell 1993). Daphnia commonly known as water fleas, are members of a larger class of zoo plankton known as cladocerans. Cladoceran body length depends on life stage and may reach 3 mm making them an ideal food source for larger fry (Ludwig 1999). Copepods, the third major type of zooplankton, also reach 2 3 mm as adults and are a common foo d source for larger fry (Ludwig 1999). It is important t o properly time stocking of fish after pond fertilization so that appropriately sized plankton are available for consumption (Boyd 1990). Fertilizer, as a source of nitrogen, temporarily causes inc reases in total ammonia nitrogen (TAN) and nitrite in the pond. TAN includes both ionized ammonium (NH 4 + ) and unionized ammonia (NH 3 ). The unionized form is much more toxic to fish, with as little as 0.06 mg/L adversely affecting some fish species (Noga 2010). The ratio of
18 unionized and ionized ammonia in water depends on several factors, primarily temperature and pH. As pH and temperature increase, the portion of unionized (i.e., more toxic) ammonia increases. Water with a pH below 7.0 has almost no u nionized ammonia. Besides fertilizer, the main source of ammonia in ponds is excretion from fish; however, other sources include rotting organic matter and animal waste. Nitrite is a result of bacterial oxidation of ammonia. Like ammonia, nitrite can b e toxic to fish at very low concentrations. Nitrite sensitivity varies among fish species (Lewis and Morris 1986) and some groups of fish (e.g., centrarchids) are even resistant to nitrite toxicity. Bacterial species also convert nitrite into nitrate, a relatively nontoxic form of nitrogen. After a new pond is cleaned, limed, and fertilized, it takes several days to weeks for nitrifying bacteria populations to establish. Until this occurs, TAN and nitrite may rise above safe levels for fish; therefore, w ater quality parameters should be monitored prior to stocking. Recommended stocking practices for ornamental fish species depend on whether they are egg laying or livebearing. Livebearing species are usually stocked as adults and allowed to spawn in t he pond. Fry are then periodically harvested based on size and grade. Egg laying species, on the other hand, are usually spawned in a hatchery setting and t hen stocked into ponds as fry for a finite growout period, after which the entire pond is harveste d for sale. Recommended stocking density in ponds is variable, depending on the fish species, pond design, and whether any life support is supplemented to the pond. Like any agricultural commodity, the goal of ornamental aquaculture is to produce large y ields of a high quality product, as quickly and economically as possible. While increased stocking rates may potentially grow more
19 fish and generate more profit, there are drawbacks to heavy stocking. Crowded ponds are at a higher risk for issues that ca n hinder growth and survival. For example, heavy feeding and dense stocking can lead to poor water quality. In particular, higher numbers of fish excrete more ammonia into the water. Also, crowded ponds have higher oxygen demand and are at risk for low DO issues. Finally, there is an increased risk of disease issues in densely stocked habitats (Noga 2010). Aeration in Aquaculture Dissolved oxygen is arguably the most important water quality parameter in aquaculture (Boyd 1990). Oxygen is a critical pa rameter for many reasons. First, example, the atmosphere at sea level contains 21% oxygen, whereas water at 28C is saturated with oxygen at only 7.8 mg/L (or 0.078%). Anoth er important factor regarding oxygen is that DO levels can change rapidly, fluctuating from satisfactory to critically low within minutes to hours. Therefore, a low oxygen event may not be noticed without close observation, and intervention may be too lat e Oxygen is supplied to water via three basic mechanisms: 1) diffusion across the water surface from the atmosphere; 2) wind and wave action; 3) photosynthesis by algae and plants. Photosynthesis is the primary source of oxygen in outdoor ponds. Phytop lankton produce oxygen during daylight hours but, along with other organisms in the pond, they also consume oxygen via respiration. During the overnight hours, the absence of photosynthetic activity in the water causes an overall decrease in oxygen conten t in the water. Ponds typically have a diurnal rise and fall in DO levels, with peak levels usually reached in the afternoon before dusk and lowest levels just before dawn.
20 where suboptimal DO levels cause morbidity or mortality in the fish population (Francis Floyd 1992; Torrans 2005). While oxygen depletion events can occur at any time, in Florida they are most common during summer months. This is due to a number of fact ors. As ambient temperatures increase, water temperatures rise and DO concentration in the water decreases. Because fish are poikilothermic, metabolic rate goes up with increasing temperature and causes oxygen demand to increase. Thus, any available oxy gen is consumed more quickly by fish in warmer water. Other factors leading to decreased DO in the summer are also related to environmental conditions Sunlight and warm temperature s can cause stratification in the water, leading to a warm, oxygen rich s urface layer and a cooler, oxygen depleted bottom layer (Chang and Ouyang 1988). Once disturbed via wind or rain, mixing of these two layers occurs and the total DO concentration i n the pond may quickly decrease in what is known as a nt Pr olonged periods of cloudy weather can also decrease photosynthetic activity in the pond steadily decreasing DO levels until an oxygen depletion even t occurs (Boyd and Tucker 1979; Francis Floyd 1992). Supplemental aeration is the primary method for med iating low DO events in pond systems (McGee and Cichra 1991). Besides raising oxygen levels for fish, there are multiple benefits to providing aeration. Increased oxygen levels have been shown to support heavier populations of nitrifying bacteria than st atic ponds, allowing for more efficient bacterial decomposition of w aste and faster mediation of poor water quality (Fernandes et al 2010). Also, many aerator designs minimize oxygen and temperature stratification by increasing water circulation (Rogers 1 989 ) resulting in better DO levels
21 throughout the pond and an increased habitable volume for fish (Rogers and Fast 1988). A minimum DO concentration of 5 mg/L is recommended for optimal freshwater f ish health (McKee and Wolf 1963; Swingle 1969), and mor talities may occur if DO concentration is prolonged at less than 2 mg/L (Noga 2010). Critical DO levels are species specific; however, DO levels at or near saturation (7.8 mg/L at 28C) are recommended for most ornamental fish species (Lawrence 2007). Lo w DO concentrations have been linked to decreased hatching and growth rates, lower feed consumption, and even mas s mortality events in extreme cases (Brungs 1971; Buentello et al. 2000; Das et al. 2012). Chronically low DO levels have been associated wit h exacerbation of a number of fish pathogens, including Mycobacterium (Noga 2010), Streptococcus spp. (Evans et al. 2003; Russo et al. 2006), and Flavobacterium columnare (Noga 2010). The majority of pond aeration studies conducted thus far have been perfo rmed on cultured food fish species, including catfish and carp (Boyd 1990 ; Pawar et al. 2009 ). Channel catfish ( Ictalurus punctatus ) production is typically intensive, with heavy stocking and feeding rates Food fish species are large bodied and have a h igh oxygen demand; therefore, DO is often consistently monitored in food fish operations, and emergency aeration is implemented as needed to prevent nighttime DO from falling to stressful levels (Boyd and Tucker 1979). The most common aerator types in fo od fish culture include paddle wheels and pumps that are electrically operated and powerful enough to treat large 3 20 hectare ponds. Aeration is most commonly used on an as needed basis, since research suggests there is no added benefit in channel catfis h
22 growth rate and survival when aeration is provided continuous ly versus nightly (Boyd 1990). While aeration has been studied extensively in food fish production (Boyd and Tucker 1979 ; Boyd 1998 ), aeration standards have not been established for ornament al fish pond production. So far, most ornamental husbandry practices have been based either on other areas of aquaculture or on empirical observation within the ornamental fish industry Information based on non ornamental species (e.g., food fish) canno t be considered universally applicable or appropriate for all areas of aquaculture, particularly ornamental production. M ost ornamental fishes are much smaller than food fish species and some popular ornamentals (e.g., fancy goldfish and freshwater angel fish) are more delicate. Powerful aerators that are ideal for large catfish ponds may not perform similarly in ornamental fish ponds that are 100 times smaller; therefore, food fish practices may not be appropriate for small bodied ornamental species. C urrently, the most popular aeration method for ornamental production ponds consists of air stone s attached to a floating line that is aerated using a low energy air blower. This design was developed in th e early 1980s and is known as Aeration (He ideman 1983; Figure 2 1). Most ornamental p roducers in central Florida either use air stones or no aeration at all. O ther aeration methods, such as diffusers, airlift systems, and electrically powered pumps (Boyd 1998) are not widely used in ornamental p ond production Zebra Danio Species Profile A wide variety of freshwater ornamental fishes are currently raised in Florida, including cyprinids, tetras, cichlids, livebearers, and catfishes (Hill and Yanong 2002). T he zebra danio was chosen as the res earch model for the aeration and pond harvest
23 studies described in Chapters 4 and 5. Also known as the zebrafish, t he small size and active, hardy nature make it one of the most popular aquarium fish species. Several natural and manmade aqu arium vari ants exist, including longfin morphs and blue, gold, and albino colorations. In 2003, a transgenic variety of ze was developed and introduced for sale to the public (Gong et al. 2003). Four vividly fluorescent colors of have been trademarked so far as a genetic model has also recently increased m aking it one of the most common species for vertebrate research The zebra danio belongs to the Cyprinidae family, which is one of the largest families of freshwater fishes. There are 44 recognized danioin (danio and similarly related) species throughout the world (Fang 2001). Adults grow up to 4 cm in standard length and both sexes have a characteristic striped appearance on the body and fins (Laale 1977). The zebra danio is native to the Indian subcontinent where it is found in a wide variety of habit ats, including floodplains, manmade lakes, ponds, irrigation channels, rivers and streams (Daniels 2002; Spence et al. 2007b). Natural habitats typically contain aquatic vegetation and have a silty substrate. Danios travel in small shoals and occupy the entire water column in both open and planted areas (Spence et al. 2007b). Zebra danios are commonly perceived as a hardy fish in captivity; however, water quality is still important for providing optimal husbandry conditions. Wild zebra danios are found in slow moving water that is slightly alkaline with a pH around 8.0, and recommended pH range for domesticated zebra danios is 7.0 8.0 (Lawrence 2007). The annual temperature range is approximately 6 38C (Spence et al. 2007b), and controlled laboratory e xperiments show that zebra danios have a temperature tolerance
24 range of 6.7 41.7 C ( Cortemeglia and Beitinger 2005). Zebra danios also prefer hard water; the recommended hardness for systems is greater than 100 mg/L CaCO 3 (Lawrence 2007). The zebra dani unknown; however, it is recommended to maintain DO levels as close to saturation as possible (7.8 mg/L at 28.0C) to prevent stress and decreased production (Lawrence 2007). The mean lifespan of domesticated zebrafish is 42 months (Spence et al. 2007a), though individuals have been documented at over 5 years of age in captivity (Lawrence 2007). Reproductive maturity is relative to size rather than age. Captive zebra danios raised at 25.5C reach maturity ar ound 75 days at a size of approximately 23 24 mm SL (Eaten and Farley 1974). Wild danios also reach sexual maturity at this size, although growth rate is slower (Spence et al. 2007a). Zebra danios spawn in groups and lay eggs directly onto the substrate with no parental care (Spence et al. 2007a). In aquaculture, it is standard to breed pairs or small groups in small tanks or boxes with a mesh or grill bottom, which allows eggs to fall through the mesh and avoid cannibalism by broodstock (Lawrence 2007) Females are capable of spawning several hundred eggs as often as every 2 3 days (Kimmel et al. 1995). Eggs hatch approximately 48 72 hours after they are laid, and larvae then attach to hard surfaces where, after swim bladder inflation at 2 3 days old, they become free swimmin g (Kimmel et al. 1995). Larval feeding, which begins 4 6 days after hatching coincides with absorption of th e yolk sac (Goolish et al. 1999; Lawrence 2007).
25 Food selection and water quality are both very important factors when raising zebra danio fry. Fry are gape limited, and the best feeds for youngest fry include a mixture of ciliates, rotifers, Artemia and commercially formulated feeds (Brown 1997). In a hatchery setting, fry should be fed several times a day or continuo usly (Spence et al. 2007a). Although individual tanks with recirculating systems can be used, in Florida zebra danio larvae are typically stocked into outdoor earthen ponds for growout. Growth rate of zebra danios is fastest in the first three months of life (Spence et al. 2007b). In Florida, growout to market size is typically 10 16 weeks during the summer, depending on stocking and feeding rates (J. Diaz, 5 D Tropical; C. Martinez, University of Florida; T.Simmons, Steve Simmons Aquatics Inc., personal communications). Zebra danios are generalists that feed on a wide variety of food items. Gut content analysis performed in wild danios revealed phytoplankton, zooplankton, insects, arachnids, filamentous algae and plants (McClure et al. 2006; Spence at al. 2007a). Feeding rate is typically based on percent body weight, with larval fish in indoor systems being fed 50 300% BW per day and adults 1 10% BW (Lawrence 2007). Because zebra danios are cyprinids and lack a true stomach, frequent meals with an a ppropriate diet should be provided for maximum growth. There are several reasons why the zebra danio was chosen as the research model for the studies described in Chapter 4 5. First, cyprinids make up one of the largest and most popular fish families in t he ornamental in dustry; therefore, the zebra danio is a r epresentative model for a substantial portion of ornamental fish p roduced in Florida. Also, the aeration and harvest studi es outlined in Chapter 4 and 5 required a combined stocking rate of over 1 m illion danio fry ; therefore, a readily available
26 ornamental species was n eeded The zebra danio is currently the second most common freshwater ornamental species in the United States by volume (M. Meyers, PIJAC, personal communication) and high numbers of fry were available from local producers The zebra danio species also made it suitable for the 3 month growth study designed in Chapter 4.
27 Figure 2 1. Photograph of a typical earthen po nd used in ornamental fish production, measuring 18 x 7.5 m. This pond is equipped with a FAT CAT style aerator constructed of poly tubing and multiple air stones, the most common aeration method for ornamental production ponds in Florida.
28 CHAPTER 3 PRELIMINARY INVESTIGATION OF AERATION METHODS IN SMALL AQUACULTURE PONDS Overview Th e use of supplemental aeration has been studied in various areas of aquaculture, including baitfish and food fish culture (Boyd 1998 ; Pawar et al. 2009 ). Aeration provides benefits to intensive pond culture, not only by directly correcting low oxygen prob lems but also by improving water quality via oxygenation and water mixing. One area of aquaculture that remains relatively unstudied with regards to aeration is small pon d culture of ornamental fish. Freshwater o rnamental fish production in the United Sta tes primarily occurs in the state of Florida. Warm temperatures in the region potentially contribute to low oxygen events, due to the decreased oxygen carrying capacity of water at warmer temperatures. To date, the most commonly used aeration method in o rnamental pond culture is air stones, and alternative methods have not been scientifically researched (C. Martinez, University of Florida, personal communication) This study examined the effect of three aeration methods (air stones; fine bubble diffuser box; vertical pump) on dissolved oxygen and other water quality and chemistry parameters in small fishless ponds. Methods The aeration study took place at the University of Florida Tropical Aquaculture Laboratory (UF TAL) located in Ruskin, Florida (Fig ure 3 1). The trial occurred during the period of June to August, 2011. Pond Preparation Existing outdoor earthen ponds at UF TAL were used for the aeration trial. Ponds measured approximately 18 x 7.5 m, with an average water depth of 1.4 m.
29 Each po nd was already equipped with adjustable flow PVC pipes supplying aerated well Prior to the trial, ponds were drained and banks were pressure washed to remove any loose detritus and organic material. Cottonseed me al fertilizer and hydrated lime (CaOH) were applied to the banks at a rate of 11.3 kg/pond and 22.6 kg/pond, respectively. Dosing rates for lime and fertilizer were representative of common ornamental industry practices. Following cleaning, ponds filled with ground water within approximately 72 hours. Once ponds were full, aerators were installed and the study began, lasting for 42 days. Pond maintenance throughout the study included observation three times daily to ensure aerators were operating correc tly. Any aerator malfunctions, such as lost air stones or clogged motors on vertical pump aerators, were repaired within 24 hours. Aeration Methods Three types of aerators were used in this study: 1) air stones; 2) diffuser box; 3) vertical pump aerato r. Aerators were randomly assigned to ponds, and four replicate ponds were used for each aeration method. Air stone aerators were constructed at UF (Heideman 1983; Figure 3 2). UF TAL design consisted of a single 18 m length of 13 mm diameter rigid poly pipe. Five 0.75 m drop lines made of airline tubing were installed into the poly pipe, positioned in 1.5 m intervals and each fitted with single 50 mm silica air stone. The pol y pipe was then anchored at each end of the pond, suspending it at the water surface along the middle of the pond (Figure 3 3a). The f i ne bubble airlift pump aerators used in this study were created by Carlos Martinez, UF TAL extension faculty, and adapt ed from a three sided b ox aerator design
30 that generate s directional water flow (J.A. Chappell, Auburn University, personal communication). Airlift pump a erators consisted of multiple rows of high efficiency diffuser tubing fitted to a PVC frame, which was suspended in the water using a weighted PVC frame and pontoons (Figure 3 2). White poly liner covered three of four lateral sides of the cube, leaving on e side open for directional flow of diffused air to exit the aerator (Figure 3 3b). An OEM style ro tameter (Omega Engineering, Inc., Stamford, Connecticut) was used weekly to ensure consistent airflow among all diffuser box aerators. The third aeration method was a vertical pump pond aerator ( Model 2400AF ; Kasco Marine, Inc., Prescott, Wisconsin). Th e aerator consisted of a submersible electric HP, 120 V, single phase motor with an impeller attached to the shaft. The motor was suspended by a circular thermoplastic float, and water propelled directly upwards from the impeller into the air The aera tor was positioned in a central area of the pond (Figure 3 3c), and an insulated cable supplied power from a pond side electrical outlet. Dissolved Oxygen and Temperature Dissolved oxygen concentration, oxygen saturation, and water temperature were meas ured three times a day for 42 consecutive days. Readings were taken 30 minutes prior to sunrise (0545 0700 hours, depending on sunrise time), midday (1130 1300 hours) and in late afternoon (1600 1800 hours). Lightning prohibited pond work a total of eigh t times, resulting in the loss of one morning data collecti on, four midday collections, and three late afternoon collections. A handheld dissolved oxygen meter Oxygen and temperature measurements were obtained from the pond bank at a dept h
31 of 0.25 m. For consistency, three recordings were taken at the same locations in each pond and averaged to obtain mean values for each pond. Water Chemistry and Quality Water samples were collected from each pond once daily for water chemistry analysis. Samples were obtained using a 3 m PVC water column sampler with a swing check valve on one end. The sampler was submerged approximately 1.0 m beneath the water surface from three points on the pond bank and emptied into a bucket. From the col lected volume, a 500 mL subsample was saved for testing. Samples were analyzed within 1 hour of collection. Total ammonia nitrogen (TAN), nitrite, and pH were analyzed daily. Total alkalinity and total hardness were tested weekly. All parameters wer e analyzed using a Hach Freshwater Fish Farmer's Kit (Model FF 1A; Hach Co., Loveland, Colorado). Secchi disk measurements were obtained once weekly in each pond. The disk was submerged until it first vanished from sight. Measurements greater than the le ngth of the pipe (1.5 m) were marked Statistical Analysis Daily DO, temperature, and water quality data were analyzed using one way, repeated measures analysis of variance (ANOVA) (MIXED procedure). Water quality dynamics following pond fertil ization were analyzed using one way ANOVA with the general linear models (GLM) procedure. Both procedures were followed with post hoc performed in SAS version 9.3 (SAS Institute Inc., Cary, North Carolina). All numerical data are represented as the mean standard error (SE). Statistical differences were considered significant at P < 0.05.
32 Results Pre dawn DO concentrations (mg/L) and oxygen saturation (%) were high est in ponds with vertical pump aerators, while ponds with air stones had the lowest values (Figure 3 4, Table 3 1). Ponds with diffuser boxes had intermediate levels (DO concentration: F 80,359 = 3.39, P < 0.0001; oxygen saturation: F 80,359 = 3.11, P < 0. 0001). Pre dawn DO levels in air stone ponds fell below the minimum desirable concentration of 5 mg/L on 13 out of 41 sampling days, while ponds with the other two aeration methods maintained mean values above 5 mg/L throughout the study. Midday DO con centrations and oxygen saturation percentages followed similar patterns to the morning values, with vertical pump aerators having the highest mean values, diffuser boxes having intermediate values, and air stone ponds having the lowest values (DO concentra tion: F 74,330 = 1.71, P = 0.0009; oxygen saturation: F 74,330 = 2.31, P < 0.0001; Figure 3 4; Table 3 1). In the late afternoon, all ponds became supersaturated; however, ponds with air stones and diffuser boxes reached significantly higher DO concentratio ns and percent saturation levels than ponds with vertical pump aerators (DO concentration: F 76,342 = 1.84, P < 0.0001; oxygen saturation: F 76,342 = 1.89, P < 0.0001; Figure 3 4; Table 3 1). Ultimately, diurnal fluctuation was greatest in air stone ponds, followed by ponds with diffuser boxes. Ponds with vertical pump aerators had minimal diurnal oxygen fluctuation. Throughout the experiment, water temperature in ponds with vertical pump aerators was, on average, 1.0 C cooler than other ponds (Table 3 1). This finding was statistically significant for morning and midday readings (morning: F 80,359 = 2.36, P < 0.0001; midday: F 74,330 = 2.94, P < 0.0001), but not for afternoon readings (F 76,342 = 1.96, P = 0.1866).
33 Mean daily values for the majority of tested water qualit y and chemist r y parameters including TAN, pH, alkalinity, hardness, and Secchi depth, were not significantly different among aeration methods. One exception was nitrite, which was higher throughout the study in diffuser box ponds than in ponds with other aeration methods (F 82,369 = 1.58, P = 0.0026). Nitrogen components (TAN and nitrite) followed similar trends among aeration methods after pond s were fertilized with cottonseed meal, and no differences were observed in peak TAN, peak n itrite, and nitrogen cycle length (Table 3 2). Discussion The air stone aerator, which is Aeration design is the most common aeration method currently used in ornamental fish production ponds (C. Martinez and C. Watson, University of Flor ida, personal communications); h owever, this study showed that fishless ponds with air stones consistently had the lowest early morning DO levels of three tested aeration methods. Diffuser boxes and vertical pump aerator s, two aeration methods used more commonly in pond culture of non ornamental fish species (Boyd 1998) each maintained acceptable morning DO levels throughout the study. Results from this preliminary investigation suggest that alternative methods besides air stones may be more suita ble for aeration of small ornamental ponds. Of the three aeration methods tested, ponds with vertical pump aerators had the least amount of diurnal oxygen fluctuation, with morning and afternoon saturation levels staying closer to 100% compared to the ot her two aeration methods (Table 3 1). Differences in diurnal fluctuation may have been due to several factors, including varying phytoplankton populations in ponds and different mixing ability of ea ch aerator type. Phytoplankton contribute to diurnal cha nges in ponds, because the daily rise and
34 fall of oxygen is a direct result of photosynthesis and respiration by phytoplankton. Although phytoplankton content was not directly measured in this study, Secchi depths were recorded on a weekly basis. As a me asurement of water clarity, Secchi is also considered an indirect parameter for estimating biological productivity in a water body (Florida LAKEWATCH 2000). Secchi measurements in ponds with vertical pump aerators were not statistically different than oth er ponds; however, phytoplankton is only one of three factors contributing to Secchi readings the other two being color and suspended sediments. A more accurate indicator of phytoplankton content would have been chlorophyll a ; however, chlorophyll was no t analyzed in this study. A s the base of the aquatic food web, phytoplankton is essential for establishing populations of zooplankton for small fish to eat. Phytoplankton and zooplankton are both desirable in ornamental production ponds as they serve as critical sources of food for small bodied fish. Zooplankton populations also were not measured in this study and would have been useful for comparing biological productivity among ponds. Differences in diurnal oxygen levels in this study may also have b een caused by the differing m ixing abilities of each aerator. Aerators with powerful mixing a ction such as the vertical pump, vigorously agitate the water to increase gas diffusion across the water surface. Ultimately, mixing brings gas levels (e.g., ox ygen and carbon dioxide) closer to equilibrium with atmospheric concentrations (Boyd 1990 ; Rogers 1990 ). Of the three aeration methods tested in this study, the vertical pump was observed to have the strongest mixing action. The cooler temperature observ ed throughout the study in vertical pump aerator ponds also implies better mixing of all thermal layers in those ponds.
35 Although vertical pump aerators had the highest morning oxygen levels and least diurnal fluctuation compared to other aerators additio nal factors are also important for producers choosing an aeration method. One key cons ideration is the cost of purchasing, operating, and maintaining the aerator. Among the three aeration methods tested in this study, air stones were the cheapest aerator to build ($25 USD per unit), and they were simple to construct with minimal maintenance after installation. Diffuser boxes were more expensive at $92 USD each for materials, yet they required no maintenance during the study other than routine rotameter r eadings to check for consistent air flow. When purchased directly from the manufacturer, a HP vertical pump aerator similar to the model used in this study retails for $795 USD making it the most expensive of the three tested aerators Also, the verti cal pump was the only aerator requiring pond side electrical supply. Electricity installation costs for a small ornamental fish farm can exceed $10,000 (C. Watson, University of Florida, personal communication), and monthly energy costs may exceed those n eeded to run a single air blower. The vertical pump aerators in this study also required the most maintenance of all tested methods, because they frequently became clogged with aquatic plants. Clogging compromised aerator efficiency, and unclogging was n ecessary to prevent motors from malfunction ing abilities. In conclusion, b aseline information gained from this pre liminary s tudy shows that aeration and water circulation were the l east pronounced in ponds with air stones which are currently the most common method used in ornamental production ponds. Data
36 collected from fishless ponds suggests that a lternative aerat ion methods may provide additional benefits along with improved aeration, such as better water quality through increased oxygenation and circulation throughout ponds Future studies using ponds stocked with fish in a production setting will ultimately det ermine whether alternative aeration methods lead to improved production of ornamental fish species.
37 Figure 3 1. Aerial photograph of the University of Florida Tropical Aquaculture Laboratory s design, with multiple rows of small excavated ponds, is representative of ornamental fish production facilities in central and south Florida.
38 Figure 3 2. Example of a fine bubble diffuser box aerator constructed for the study. This a erator was designed by Carlos Martinez, UF TAL extension faculty. The aerator frame (A) provided buoyancy for the aeration mechanism, which consisted of multiple rows of diffuser tubing supplied by a flexible PVC airline (B).
39 Figure 3 3. Schematic indicating aerator location within ponds for a) air stones; b) fine bubble diffuser box; and c) vertical pump pond aerator. Arrows indicate directional water flow generated by each aeration method.
40 Figu re 3 4. Dissolved oxygen and oxyge n saturation for research ponds containing one of three aeratio n methods Readings were taken three times daily for 6 weeks Data points are expressed as mean values for each aeration method.
41 Table 3 1. D issolved oxygen (DO) and other water quality and chemistry values in ponds aerated with three aeration methods. Data were collected over a 6 week period and are represented as mean SE with ranges in parentheses. Significant differences ( P < 0.05) are noted by different letters Variable Air stones D iffuser box Vertical pump Pre dawn DO (mg/L) 5.2 0.10 (2.3 7.9) x 6.3 0.06 (4.3 7.1) y 7.4 0.03 (6.9 7.9 ) z Midday DO (mg/L) 7.0 0.12 (4.0 11.1) y 7.6 7.1 (5.3 10.8) y 8 .4 0.09 (7.0 11.4 ) z Afternoon DO (mg/L) 9.0 0.26 (4.8 16.5) z 8.6 0.17 (5.9 13.7) z 8.4 0.08 (7.3 11.1) y Pre dawn oxygen saturation (% ) 68.5 1.6 (29.9 98.7) x 82.8 1.6 (55.9 9 1.3) y 94.9 1.6 (81.7 98.1) z Midday oxygen saturation (%) 92.4 1.7 (52.1 143.8) x 102.1 2.7 (68.9 280.4) y 109.5 2.6 (91.0 144.0 ) z Afternoon oxygen saturation (%) 119.7 3.8 (64.4 219.6) z 114.6 2.6 (79.4 183.3) z 109.9 3.4 (91.2 144.3) y Morning temperature (C) 29.5 0.17 (26.8 31.6) z 29.7 0.17 (26.9 31.8) z 28.5 0.19 (26.0 30.4) y Midday temperature (C) 30.3 0.19 (27.6 32.1) z 30.2 0.18 (25.5 32.3) z 29.3 0.17 (27.2 31.2) y Afternoon temperature (C) 31.5 0.20 (22.6 33.4) 31.6 0.20 (23.5 33.7) 30.5 0.18 (22.4 32.3) pH 8.4 0.04 (7.8 9.5) 8.4 0.03 (7.8 9.0) 8.6 0.02 ( 8.0 9.5) Total ammonia nitrogen (mg/L) 0.6 0.02 (0.3 1.5) 0.6 0.03 (0.3 1.5) 0.5 0.02 (0.2 1.4) Nitrite (mg/L) 0.17 0.03 (0 1.52) y 0.26 0.06 (0 1.99) z 0.18 0.04 (0 1.55) y Hardness (mg/L CaCO 3 ) 342.5 8.8 (307.8 444.6) 359.1 14.8 (290.7 444.6) 356.9 11.9 (290.7 442.5) Alkalinity (mg/L CaCO 3 ) 149.6 1.9 (119.7 171.0) 144.6 3.2 (136.8 171.0) 148.2 3.8 (136.8 171.0) Secchi (m) 0.72 0.02 0.90 0.02 0.92 0.02 (0.45 1.15) (0.70 1.20) (0.60 1.30)
42 Table 3 2. Nitrogen components following fertilization with cottonseed meal in small ponds aerated using three methods. Values are represented as mean ( SE). Nitrogen cycle length equals the number of days after cottonse ed meal application until nitrite concentration fell and stabilized below 0.1 mg/L. None of the differences were statistically significant using analysis with one way ANOVA ( P < 0.05). Aeration method Peak TAN (mg/L) Peak TAN (days after pond fertiliz ation) Peak nitrite (mg/L) Peak nitrite (days after pond fertilization ) Length of nitrogen cycle (days) Air stones 1.18 (0.18) 13.0 (2.0) 1.00 (0.35) 21.5 (1.0) 29.0 (5.3 ) Diffuser box 1.20 (0.06) 17.5 (1.9) 1.56 (0.17) 24.0 (0.9) 32.5 (3.0) Vertical pump 1.15 (0.18) 13.8 (1.7) 1.18 (0.26) 21.0 (0.7) 27.3 (2.4)
43 CHAPTER 4 AERATION S T UDY IN SMALL OUTDOOR PONDS STOCKED WITH ZEBRA DANIOS Overview Small ponds are used worldwide for growing fish in the food, bait, and ornamental fish in dustries. Some challenge s in pond culture include poor water quality and its associated negative effects on fish. In particular, dissolved oxygen is a critical parameter for fish survival. Dissolved oxygen (DO) can change rapidly in pond systems and is capable of causing large scale fish mortalities. There are known benefits to providing aeration in aquaculture, including improved water quality, increased stocking densities, and improved fish health. Studies have shown that fish growth and feeding rat es improve when dissolved oxygen is maintained at acceptable levels ( Brungs 1971 ; Buentello et al. 2000; Pichavant et al. 2005; Torrans 2005). A preliminary study performed at the University of Florida Tropical Aquaculture Laboratory (UF TAL) tested the e ffects of three aeration methods on small fishless ponds. Results revealed that electrically powered vertical pump aerators had the highest pre dawn DO levels as well as improved pond mixing compared to the other two tested methods (diffuser box and air s tones). In this study, the same three aeration methods were used in small ponds stocked with zebra danio fry. Pond physical, biological, and chemical parameters were examined, and d anio growth rate, health, and survival were evaluated The zebra dani o was chosen for this study based on its popularity in the ornamental industry. Native to the Indian subcontinent, the zebra danio is a small cyprinid reaching adult s ize of approximately 4 cm standard length (Laale 1977) By volume, the zebra danio is c urrently the second most commonly produced freshwater ornamental fish in the United
44 States (M. Meyers, PIJAC, personal comm unication). The normal (i.e., wild type ) morph was used for this study. Methods Pond Preparation This study took place at UF TAL i n Ruskin, Florida, during the months o f June to August 2012, using existing outdoor ponds. Each pond measured approximately 18 x 7.5 m, with an average depth of 1.4 m, and was equipped with PVC lines supplying s blower. Five weeks prior to cleaning, ponds were each treated with 1 L of 19.8% copper sulfate pentahydrate (Aqua tec; Cheltec, Inc., Sarasota, Florida) to reduce periphyton (filamentous algae) content. Ponds were later retreated with copper sulfate on day 58 of the 70 day trial in preparation for harvest. Heavy periphyton is considered u ndesirable in ornamental ponds because it can clog pond equipment and complicate harvest procedures Ponds were prepared for stocking by draining and pressure w ashi ng the banks to remove the upper layer of substrate containing detritus and organic material. Cottonseed meal fertilizer was applied in a thin layer to the substrate at a rate of 11.3 kg/pond, followed by 22.6 kg of hydrated lime (CaOH). Dosing rates of lime and fertilizer were representative of common ornamental industry practices (C. Martinez, University of Florida, personal communication) Following cleaning, it took approximately 5 days for ponds to fill with groundwater, and well water was supplemen ted to ponds to expedite the filling process due to a drought. Once ponds were full, aerators were installed and started.
45 Aeration Methods Three types of aerators were used in this study: 1) air stones; 2) diffus er box; 3) vertical pump Aerators were r andomly assigned to ponds, and four replicate ponds were used for each aeration method. Air stone aerators were constructed at UF TAL 1983). UF TAL design consisted of a s ingle 18 m length of 13 mm diameter rigid poly pipe. Five 0.75 m drop lines made of airline tubing were installed into the poly pipe in 1.5 m intervals and each fitted with single 50 mm silica air stone. The poly pipe was then anchored at each end of the pond, suspending it at the water surface along the middle of the pond (Figure 4 1a). The f i ne bubble airlift pump aerators used in this study were created by Carlos Martinez, UF TAL extension faculty, and adapted from a three sided box aerator design th at generates directional water flow (J.A. Chappell, Auburn University, personal communication). Aerators consisted of multiple rows of high efficiency diffuser tubing fitted to a PVC frame, which was suspended in the water using a weighted PVC frame and p ontoons (Figure 4 2). White poly liner covered three of four lateral sides of the cube, leaving one side open for directional air flow of diffused air to exit the aerator (Figure 4 1b). An OEM style rotameter (Omega Engineering, Inc., Stamford, Connecti cut) was used weekly to ensure consistent airflow among all diffuser box aerators. The third aeration method was a vertical pump pond aerator ( Model 2400AF ; Kasco Marine, Inc., Prescott, Wisconsin). The aerator consisted of a submersible electric HP, 120 V, single phase motor with an impeller attached to the shaft. The motor was suspended by a circular thermoplastic float, and water was propelled directly
46 upwards by the impeller into the air for aeration. The aerator was positioned in a central area of the pond (Figure 4 1c). An insulated cable supplied power from a pond side electrical outlet. After aerator installation, pH, total ammonia nitrogen (TAN), and nitrite were monitored every other day until stocking. Zebra Danio Stocking Zebra danio f ry were obtained for this study from a local ornamental fish broodstock in 20 L spawning containers with fine mesh partitioning to prevent cannibalism of eggs. Broodstock w ere removed after spawning, and fry remained in the spawning containers until they became free swimming (8 days after broodstock introduction and 5 days post hatch), at which point they were collected for stocking into ponds. Prior to s tocking, number of fry per spawning container was determined at UF TAL by counting fry in a subsample of ten containers. Fry were euthanized in a 2.5 g/L solution of tricaine methanesulfonate (MS 222; Tricaine S; Western Chemical, Ferndale, Washington) buffered with sodium bicarbonate. After euthanasia, batches of fry were placed with in a small volume of water in a Petri dish, numbered and photographed. Fry in each photograph were then hand counted to determine average number of fry per spawning container. This average n umber (7,230 fry per container) was used to verify pond stocking densities. Typical stocking densities in ornamental fish production are difficult to determine and may be highly variable among producers, because fry are usually not individually counted. Zebra danio s tocking rates are approximate and range anywhere from 30,000 to 210,000 per pond (J. Diaz, 5 D Tropical, and T. Simmons, Steve Simmons Aquatics
47 Inc., personal communications). A stocking rate of 79,500 fry per pond was chosen for this study as representative of industry practice. Fry were transported from the TAL in oxygenated plastic bags. At UF TAL, bags of fry were arbitrarily distributed to ponds and floated in the water for 10 minutes for temperature acclimati on. Fry were then released into ponds. Total transport and acclimation time was less than 1 hour. The aeration trial began on the day of stocking and lasted for 70 days. Upon stocking, fry were offered a commercially prepared diet at a daily rate of 15% mean body weight (BW) divided into two feedings per day. Feeding weights were based on average pooled wet weights obtained from fish subsamples during the pre trial fry counts. A 45% protein tropical fish meal (Cargill Animal Nutrition, Inc., Minneapoli s, Minnesota) was fed throughout the study. After the first week of the trial, subsequent feeding rate was adjusted weekly according to weight measurements taken at 7 day intervals. Each week, feeding rate for all ponds was increased to 15% BW of the hea viest subsample of fish. In week 5, daily feeding rate was reduced from 15% to 5% BW based on ornamental production recommendations (Francis Floyd 2002; C. Martinez, University of Florida, personal communication). Daily feed rates throughout the study ranged from 0.15 kg/pond (week 1 ) to 1.08 kg/pond (week 5) and a final feed ing rate of 2.7 kg/pond (week 10). Pond Maintenance Pond maintenance included twice daily observation to ensure aerators were operating correctly. All aerator malfunctions wer e repaired within 12 hours, excluding the unforeseen failure of three out of four vertical pump aerators throughout the study.
48 In each case of aerator failure, a replacement motor had to be ordered and shipped from the manufacturer, a process which took s ix days to occur after each malfunction. Biosecurity measures were used throughout the study to minimize risk of any pathogen spread among fish populations. Strict biosecurity was impossible due to the outdoor, open environment and the inability to comp letely prevent disease vectors such as aquatic turtles, snakes, and birds from accessing the ponds; however, complete biosecurity is difficult in outdoor pond studies. Disinfectants, including hydrogen peroxide and potassium peroxymonosulfate sodium chlor ide (Virkon Aquatic; Western Chemical, Inc., Ferndale, Washington), were used to clean pond equipment after use, and bird netting and pyrotechnics were used to deter predators. Dissolved Oxygen and Temperature Dissolved oxygen concentration, oxygen satu ration, and water temperature were measured twice daily for 70 consecutive days. Readings were taken 30 minutes prior to sunrise (0545 0700 hours, depending on sunrise time) and in the late afternoon (1600 1800 hours). A handheld dissolved oxygen meter ( ings, Ohio) was used to obtain measurements Oxygen and temperature were measured from the pond bank at a depth of 0.25 m. For consistency, three recordings were taken at the same locations in each pond and averaged to obtain mean values for each pond. In week 5 of the aeration trial, a diurnal oxygen study was performed. DO, oxygen saturation, and temperature were recorded in each pond every 2 hours for a 24 hour period, beginning and ending just before dawn. In week 10 o f the aeration trial, an oxygen profile study was performed. Readings were taken at three locat ions in each pond: 1) the central /deepest area; 2) an area adjacent the pond bank; 3) a shallow point halfway between locations 1 and 2. To determine the degre e of oxygen and
49 temperature stratification in ponds, DO and temperature measurements at each location were recorded in 0.5 m depth intervals, starting at the surface (0 m) and descending until the probe touched the substrate. The study occurred at 1400 ho urs. Lightning prohibited pond work a total of 14 times, resulting in the loss of five morning and nine late afternoon data collections. During the fourth week of the study, Tropical Storm Debby resulted in significant flooding to research ponds. Floodi ng did not rise above ground level, and therefore fish were not able to escape; however, flooding did coincide with the electrical failure of two of four vertical pump aerators on day 30 of the study. A third vertical pump aerator malfunctioned earlier on the study on day 8. In each case of failure, aerators were pulled from the ponds and motors were determined to be flooded with pond water. On site repair attempts were not successful; therefore, replacement motors had to be ordered, a process which too k six days after every case of failure. During the wait, supplemental aeration was not provided to the affected ponds. DO and temperature were still recorded dur ing periods of motor failure, a lthough those values were not incl uded in statistical analyses for this report. Water Chemistry and Quality Water samples were collected from each pond once daily for water chemistry analysis. Samples were obtained using a 3 m PVC water column sampler with a swing check valve on one end. The sampler was submerged approximately 1.0 m beneath the water surface from three points on the pond bank and em ptied into a bucket. From the collected volume, a 500 mL subsample was saved for testing. Samples were analyzed within one hour of collection. Tota l ammonia nitrogen, nitrite, and pH were analyzed daily until all three parameters stabilized (during week 5), after which measurements were taken every
50 other day. Total alkalinity and total hardness were tested weekly. All parameters were analyzed using a Hach Freshwater Fish Farmer's Kit (Model FF 1A; Hach Co., Loveland, Colorado). Secchi disk measurements were obtained once weekly in each pond. The disk was submerged until it first vanished from sight. Measurements greater than the length of the pip e (1.5 m) were marked Chlorophyll a Chlorophyll a analysis was performed once weekly using samples collected from each pond. Water samples were collected by submerging a 1 gallon ( 3.8 L ) plastic jug upside down to elbow depth in the water, the n turning the container upright to fill it. Water samples were processed within 1 hour of collection. Phytoplankton was concentrated by pouring water through a membrane filter. The filter was then stored in silica dessicant at 0C until submission to th e Florida LAKEW ATCH laboratory in Gainesville for analysis. The method used for chlorophyll a measurement was pigment extraction with ethanol, followed by spectro photometric methods (Sartory and Grobbelaar 1984). Zooplankton Densities Weekly water sampl es were collected from each pond for zooplankton density analysis. Results for the first two weeks of the study were di scarded due to partial loss of zooplankton samples. A new method of sample preservation was designed in week 3 and successfully used fo r the remainder of the experimental period. Two liter (L) water samples were collected from each pond using a 3.0 m PVC water column sampler with a swing check valve on one end. The sampler was submerged approximately 0.75 m below the water surface, ta king care to avoid touching the bottom of the pond. Samples were analyzed within 2 hours of collection. Each 2 L
51 water sample was filtered through a 35 m mesh sieve to retain all zooplankton while letting smaller debris (e.g., phytoplankton, fine detri tus) filter through. Retained zooplankton were washed from the sieve into a glass vial using 20 mL of filtered pond water. After immobilizing plankton with acetic acid, samples were immediately processed by placing a 1 mL subsample on a Sedgewick Rafter slide. Zooplankton were counted according to type (rotifers, copepods, and cladocerans) using light microscopy at 100x magnification. Counts were then used to determine densities of each type of zooplankton per L of pond water. Danio Growth Fish were collected for weekly mean total length (TL) and weight analyses. A fine mesh net was used to catch 30 fish from each replicate pond in weekly intervals. As fish increased in size during the study, they became increasingly difficult to catch using the net so the capture method was switched to wire minnow traps with 3 mm mesh in week 4. Traps were baited with frozen shrimp and collected within three hours Fish were euthanized in a 2.5 g/L solution of tricaine methanesulfonate (MS 222; Tricaine S; West ern Chemical, Ferndale, Washington) buffered with sodium bicarbonate. Total length of each fish was measured in 0.01 mm increments using electronic calipers. A pooled weight for 30 fish was recorded to the nearest 0.1 mg using a digital laboratory scale. Pooled weight was used to calculate average individual fish weight in each pond. The ornamental producer who initially donated zebra danio fry also donated a TL and weight were obtaine d on these fish to determine a representative saleable size for zebra danios.
52 These values were used as comparisons for research fish sizes, ultimately to determine the growout period for each aeration method. Danio Health Assessments Three days prior t o harvest, a subsample of 20 fish was collected from each pond to examine physical appearance and external pathogen load. Fish were captured from ponds using either a fine mesh net or a baited wire minnow trap. Live f ish were held in plastic bags contai ning pond water until examination, which was completed within two hours of capture. Any gross pathological signs (such as anatomic deformities or lesions) or abnormal behavior found on physical examination were noted. Skin, fin, and gill biopsies were the n collected from each fish, based on methodology described by Roberts et al. (2010). The skin biopsy was collected from the left lateral body wall starting just behind the operculum and extending to the caudal fin. Approximately 50% of the caudal fin was removed for examination. A gill biopsy was collected from one gill arch on the right side. Wet mounts of all tissues were prepared using fresh water. Immediately following biopsy collection, fish were euthanized in 2.5 g/L tricaine methanesulfonate (MS 222; Tricaine S; Western Chemical, Ferndale, Washington) buffered with sodium bicarbonate. External tissue wet mounts were examined within 5 minutes of biopsy collection. Wet mounts were observed using light microscopy at 40x, 100x, and 200x magnifica tions. Pathological findings were quantified using three gradations (light, moderate, and heavy), and a ll samples were read by the author in order to minimize interpretation variation. surrounding the gill fi lament measured less than 25% the width of the widest section of 50% the width of the filament, and
53 or more of the secondary lamellae affected). Digene infections in the gill were 3 metacercariae 6 metacercari ae or more metacercariae per 40x field). Pond Harvest Ponds were completely harvested at the end of the 70 day aeration trial. Due to limited capacity of the faci consecutive days to complete. Forty eight hours prior to harvest of each pond, commercial feed was discontinued to reduce oxygen consumption and ammonia production by fish during holding and transport ( Wedemeyer 1996). Immediately prior to harvest, ponds were drained halfway to concentrate fish into a smaller volume for easier collection. Fish were collected using a seine net (18.3 m x 2.4 m, with 1.6 mm mesh) operated by two people. The first collect in order to avoid overloading the seine net with fish. Three pulls spanning the entire pond area followed. Each pond was pumped nearly empty prior to the final pull. Fish were dip netted into insulated coolers con taining a 1:1 mixture of pond water and aerated well water (i.e., holding system water), then transported to UF building for acclimation (Crosby et al. 2005b). Fish were weighed using a digital scale with a sensitivity of 2 g. (Prior to harvest, scale accuracy was confirmed by comparing multiple fish weights with a more sensitive, calibrated digital scale.) Gravid females and fry (estimated to be less than 3 weeks old, based on size) were observed in all vertical pump aerator ponds and in three of four
54 diffuser box ponds; however, fry were excluded from weight measurements and total yield as they were not considered to be marketable size. To weigh harvested fish, plastic containers partially filled with well water were tared to zero. Using dip nets, fish were separated from pond debris and pests and transferred to the containers after allowi ng excess water to drain from the net Seven to ten subsamples per pond, totaling approximately 1,000 fish, were also hand counted after weighing to calculate average individual fish weight Average fish weight was then used to determine total number of fish per pond via the following equation: Estimated fish (#/pond) = Pond yield (g) / Mean fish weight (g) A subsample of 30 fish per pond was eut hanized for TL and weight analysis, using similar methodology for size analyses during the aeration trial. Statistical Analysis Daily DO, temperature, and water quality data were analyzed using one way, repeated measures analysis of variance (ANOVA) (MIX ED procedure). Nitrogen cycle and harvest data were analyzed using one way ANOVA with the general linear models (GLM) procedure. Both procedures were followed with post multiple comparison procedure. Health assessment data wer e analyzed using the Kruskal comparison procedure. All statistical analyses were performed in SAS version 9.3 (SAS Institute Inc., Cary, North Carolina). Numerical data are represen te d as mean standard error (SE) Statistical differences were considered significant at P < 0. 05, which was significant at P < 0.15.
55 Results This study revealed that aeration method affected zebra danio growth rate and surviv al. Pond dissolved oxygen and temperature values were different among aeration methods, and oxygen values followed trends similar to those observed during a preliminary study examining the same aeration methods in fishless ponds. The majority of other wa ter quality and chemistry values, as well as estimates of pond productivity, were statistically similar regardless of aeration method. Zebra Danio Growth, Survival, and Health Zebra danio growth was significantly faster for fish raised in vertical pump pon ds than in ponds with air stones and diffuser boxes ( TL : F 18,81 = 6.40, P < 0.0001; average weight: F 18,81 = 8.77, P < 0.0001; Figure 4 3 and 4 4). In fact, maximum daily growth rate (which was observed during week 2 for all fish) was nearly twice as fas t for fish in vertical pump aerator ponds (1.09 0.18 mm/day) compared to fish from ponds with diffuser boxes (0.68 0.1 mm/day) and air stones (0.60 0.04 mm/day). At harvest, total mass of fish collected per pond was statistically similar among aer ation methods (F = 2.22, df 2, P = 0.1647; Table 4 1); however, average individual fish weight and estimated number of fish per pond varied significantly depending on aeration method (individual fish weight: F=17.74, df 2, P = 0.0008; estimated number of f ish: F = 8.82, df 2, P = 0.0076; Table 4 1). There was an inversely proportional relationship between size and number of fish at harvest (Figure 4 5). On average, ponds with air stones had the smallest size of individual fish but the largest yield in num bers. In contrast, vertical pump aerator ponds yielded a smaller number of fish, yet individual fish were 2.5 times heavier than those from ponds with air stones. Number and size of fish were intermediate for ponds with diffuser box aerators.
56 L ength fre quency histogram s constructed using TL of danio subsamples measured at harvest illustrates the difference in size and uniformity among ponds according to aeration method (Figure 4 6). Total length was normally distributed for air stone ponds and vertical pump aerator ponds; however, the size distribution of fish from diffuser box ponds was bimodal with one replicate pond having notably larger fish than the other three ponds The most uniform fish population indicated by standard deviation (SD) of TL at harvest, occurred in danios raised in ponds with air stones (SD 2.66). The only fish to uniformly reach saleable size within the 10 week growout period were those harvested from ponds with vertical pump aerators (SD 3.03) A small portion of fish raised in ponds with diffuser box aerators also reached saleable size after 10 weeks ; however, the majority of fish were too small to be considered saleable and fish size was the least uniform of the three tested aeration methods (SD 4.65) External biopsies p erformed just prior to harvest revealed t hat fish were generally healthy regardless of aeration method (Table 4 2). The only pathogenic organism s detected were digene metacercariae ( the larval stage of a parasitic trematode belonging to the subclass Digen ea of the phylum Platyhelminthes). Digenes were present in the gills of one third of the fish examined from ponds with air stones. Only 16% of sampled fish from diffuser box ponds had digenes, and digenes were absent in all fish sampled from vertical pum 2 = 32.1, df 2, P < 0.0001). Digene levels were light in all positive fish (3 or less metacercariae per 40x field). Excess mucus and telangiectasis in the gills, two indicators of possible irritation and/or trauma, were present in a sig nificantly higher portion of fish from ponds with vertical pump aerators compared to the other two aeration 2 = 31.0099, df 2, P < 0.0001;
57 telangiectasis: 2 = 33.7573, df 2, P < 0.0001 ). Excess mucus on skin biopsies was seen only in fis 2 = 6.8961, df 2, P = 0.0318 ), while occurrence of excess mucus on the fins was statistically similar 2 = 4.9950, df 2, P = 0.0823 ). During the 10 week tri al, musculoskeletal deformities were observed in varying degrees in subsamples of fish from nearly every pond. Occurrence was typically no more than 1 in 30 fish (3%), except for one diffuser box pond which had a notably higher occurrence. In weeks 8 and 10, nine out of 30 (30%) fish in that pond had obvious signs of kyphosis and/or scoliosis (Figure 4 7). Dissolved Oxygen and Temperature Pre dawn DO concentrations (mg/L) and oxygen saturation (%) were highest in ponds with vertical pump aerators, whi le ponds with air stones had the lowest values (DO concentration: F 126,546 = 3.47, P < 0.0001; oxygen saturation: F 126, 546 = 3.25, P < 0.0001; Figure 4 8, Table 4 3). Ponds with diffuser boxes had intermediate morning oxygen levels. Vertical pump aerato rs maintained DO levels above the minimum acceptable level of 5 mg/L throughout the study. The lowest observed morning DO in a pond aerated by a diffuser box was 4.9 mg/L, only slight ly under the minimum recommended concentration for ornamental fish produ ction Mean DO levels in air stone ponds fell below 5 mg/L on 18 of 65 mornings. In the late afternoon, all ponds became supersaturated; however, ponds with air stones and diffuser boxes reached significantly higher DO concentrations and supersaturation levels than ponds with vertical pump aerators (DO concentration: F 120,523 = 1.83, P < 0.0001; oxygen saturation: F 120, 523 = 1.92; P < 0.0001). Diurnal fluctuation was the most pronounced in air stone ponds, followed by ponds with diffuser
58 boxes. Ponds with vertical pump aerators had the least diurnal fluctuation. Similar diurnal pattern s were seen during the 24 hour diurnal oxygen study, revealing greater diurnal fluctuation in DO for diffuser box and air stone ponds than for vertical pump aerator pond s (Figure 4 9). All ponds reached similar peak DO levels in the late afternoon ; however, morning DO levels were more representative of concentrations observed throughout the entire trial. Several days of cloudy weather preceded the diurnal study which, along with an afternoon thunderstorm on the day of the study, may have reduced photosynthetic activity in the ponds, resulting in less oxygen supersaturation compared to sunny periods of the aeration trial. Oxygen profiles assembled for all ponds reveal ed minimal oxygen stratification in ponds with vertical pump and diffuser box aerators (Figure 4 10) In the deepest area of the pond, DO decreased an a verage o f only 0.1 mg/L (vertical pump) and 0.8 mg/L (diffuser box) between the water surface and botto m Ponds with air stones had a greater average decrease in DO of 2.8 mg/L between surface and bottom in the deepest area Water temperature decreased no more than 0.1C from pond surface to bottom in all ponds regardless of aeration method. DO and tempe rature r eadings taken at the pond bank and shallow areas were similar among aeration methods and remained constant from the water surface to pond bottom. Throughout the 10 week aeration trial, pre dawn and late afternoon temperatures were, on average, 1.1 C cooler in ponds with vertical pump aerators than other aeration methods (morning, F 126,546 = 3.36, P < 0.0001; afternoon, F 120,523 = 2.38, P < 0.0001; Table 4 3). Pond Productivity No significant differences among aeration methods were detected f or the majority of measured water chemistry and quality parameters (Table 4 3) including TAN (F 86,387
59 = 1.09, P = 0.2968), pH (F 86,386 = 1.02, P = 0.4444), and total alkalinity (F 18,81 = 0.66, P = 0.8361). Although water hardness was significantly diff erent a mong aeration methods on a weekly basis (F 18,81 = 2.23, P = 0.0077), average water hardness for the entire study was similar ( P = 0.9417 ). Mean nitrite levels were significantly different among all three aeration methods (F 86,387 = 2.79, P < 0.000 1). Ponds with air stones had the lowest levels, ponds with vertical pump aerator had the highest nitrite, and ponds with diffuser boxes had intermediate nitrite levels (Table 4 3). Although TAN and nitrite reached similar peak concentrations in all pond s following fertilization with cottonseed meal (TAN: F value = 0.05, df 2, P = 0.9559; nitrite: F value = 1.67, df 2, P = 0.2411 ), the nitrogen cycle took approximately 12 days longer to complete in ponds with air stones compared to other aeration methods (F value = 75.21, df 2, P < 0.0001; Figure 4 11; Table 4 4 ). Mean Secchi readings were slightly lowe r in vertical pump ponds (Table 4 3 ), although the difference was not significant (F 18,81 = 1.27, P = 0.2298) and no obvious trends occurred during the st udy among aeration methods. Although chlorophyll a measurements were statistically similar among aeration methods (F 18,81 = 0.97, P = 0.5030), mean values were marginally higher for vertical pump ponds th an other aeration methods (Table 4 3 ). Because chl orophyll a is an indirect measurement for phytoplankton (Dust and Shindala 1970) this finding implies slightly heavier phytoplankton populations in ponds with vertical pump aerator s. Zooplankton levels were low in all ponds until the fifth week of the study (Figure 4 12) Rotifer, copepod, and cladoceran d ensities were statistically similar among al l ponds (rotifers: F 14,63 = 1.58, P =0.1113; copepods: F 14,63 = 1.17, P =0.3168;
60 cladocerans: F 14,63 = 1.67, P =0.0840). When plotted, results show increas ing trends in populations of all zooplankton types occurring at different time points for each aeration method. Increases in densities of all three zooplankton types were seen after week 4 in ponds with vertical pump aerators. Increases in zooplankton populations occurred later in the study for other ponds. Diffuser box ponds showed an increase in rotifer populations after week 5, cladocerans after week 6, and copepods after week 7. Ponds with air stones showed increasing rotifer and copepod densities after week 6 and cladocerans after week 8. Discussion y is determined not only by its oxygenation and water mixing abilities, but also by its effect on the health, gro wth, and survival of the fish maintained in the system. Mater ial and maintenance costs are also important factors when choosing an aerator for a production setting. This study found differences for all of the above factors among the three tested aeration methods. Notable findings included the universally h igh zebr a danio mortality (i.e., 75% or greater ) between stoc king and harvest, as well as significant differences in danio growth rate among aeration methods. The cause and timing of fish mortality in pond production can be difficult to determine The majority of mortalities in this study may have occurred around the time of stocking, immediately after fry became free swimming. F ry across species are generally the most delicate during the yolk sac stage (Howe ll et al. 1998), which in danios lasts until approxim ately 5 days post hatc h. However, even free swimming zebra danio fry are still very small (< 0.5 cm TL) and slow moving, making them vulnerable to aquatic predators such as larval insects and water beetles (Spence et al.
61 2007a ; Kumar et al. 2012 ). Young fry are also more susceptible to transport injury than older juveniles (Kurten et al. 1999). Previous s tudies with zebra danios recommend that water flow should be kept as low as possible for l arvae because high water flow may exhaust larvae and decrease survival (Lawrence 2007). T he universally high mortality rate in this study may have resulted from the yo ung a ge at which fry were transported and stocked, exposing them not only to changes in water quality but also to predation. Data suggest that morta lity occurred early in the study. Growth curves (Figure 4 3) show that danio fry growth and weight were similar for all three aeration methods until the third week of the study. This may be due to one of several reasons. First, fish size may have indeed varied early in the study, but measurement methods were not capable of detecting differences due to the small (3 mm TL) fish size. Another possibility is that significant mortality occurred early in the study (week 1 2), and subsequently growth rates cha nged as a result of differences in food and other factors important for growth. Studies examining fry populations have found that significant mortalities occur particularly in the first few days after stocking (Turner 1998), suggesting that early mortalit y may be due to stress, transportation, and water quality changes associated with moving fish for stocking. Other potenti al cause s for significant mortality in this study include starvation disease, and cannibalism Starvation is a concern in larval cu lture of both freshwater and marine species (Rao 2003 ), and improved survival can be seen by increasing feeding rates and observing fry closely to ensure they successfully feed Intracohort cannibalism has been documented in first feeding larvae raised in culture settings (Smith and Reay 1991; Baras and Jobling 2002), such as walleye Sander vitreus
62 (Johnston and Mathias 1993; Peterson et al. 1997). Disease is a concern in cultured fish particularly in intensive culture where fish are densely populated H ealth assessments performed at harvest did not detect widespread disease in any of the ponds other than an atypically high occurrence of musculoskeletal deformities in a single diffuser box pond. Kyphosis and scoliosis are known issues with zebra danios (Gerhard et al. 2002) and have typically been associated with senescence in adult wild and cultured zebra danios (Spence et al. 2007). Possible causes for musculoskeletal deformities in fish i nclude nutritional deficiencies genetics, poor water quality, and overcrowding (Halver et al. 1969; Lovell 1991; Yanong 1996; Francis Floyd 2002; Gerhard et al. 2002). The exact cause for musculoskeletal deformity w as not determined in this study. Although disease may have been a contributing factor to mortality in this study, it is not evident that musculoskeletal deformities were responsible for the universally low survival in all ponds. Ultimately, the data collected in this study is insufficient for determining the exact cause for mortality, or at what point th e majority of mortalities occurred; however, future studies involving inventory of pond populations may help determine this. In addition to the university high mortality rate, s urvival and growth rate were both significantly different among the three teste d aeration methods. Variations in dissolved oxygen levels, water quality, and pond productivity may have contributed to differences in fish growth and yield. fish growth rate has been studied Decreased oxygen saturation ha s been associated with reduced food consumption in multiple species, resulting in slower growth rate in turbot Scophthalmus maximus and sea bass Dicentrarchus labrax (Pichavant et al. 2001), as well as Atlantic
63 cod Gadus morhua ( Chabot and Claireaux 2008). One study using zebra danios found that expression of the appetite suppressing hormone leptin is markedly higher in oxygen deprived fish after just 4 days of exposure to 1.0 mg/L DO, suggesting a cause for anorexia and slow growth rate in hypoxic conditi ons (Chu et al. 2010). Low oxygen environments have also been shown to exacerbate stress r elated illnesses, including infection with the freshwater bacteria Streptococcus spp. (Evans et al. 2003 ). In the current study, early morning oxygen saturation p ercentages were consistently highest in ponds with vertical pump aerators, indicating that the vertical pump was the most effective at correcting nighttime decreases in DO. Also, p re dawn oxygen concentrations were directly proportional with danio growth rate in all ponds. Variations in w ater chemistry and quality parameters among aeration methods may also have contributed to varying danio growth rate s Although all ponds reached similar peak TAN and nitrite levels following pond fertiliza tion (Table 4 4 ) TAN and nitrite stabilized most quickly in vertical pump ponds, followed by diffuser box ponds, then air stone ponds (Figure 4 11 ). Nitrifying bacteria, being aerobic, require oxygen for metabolism; therefore, ponds with higher oxygen content likely sup ported heavier populations of nitrifying bacteria and were more effective at oxidizing toxic ammonia and nitrite (Fernandes et al. 2010) T he three aeration methods tested in this study also had varying water circulation abilities Although water flow w as not quantitatively studied, observation showed that vertical pump aerators caused vigorous agitation of the water surface Diffuser box aerators likewise had improved pond wide water circulation compared to air stone aerators, which primarily caused mi nimal surface agitation in a l ocalized area around
64 each air stone. Oxygen profiles showed differences in water circulation by comparing the presence and extent of oxygen stratification in each pond Minimal stratification (i.e., more mixing) was recorded in vertical pump and diffuser box ponds compared to air stone ponds. Efficient pond wide circulation benefits fish populations in several ways. First, mixing allows decaying organic matter to be processed by aerobic bacteria rather than settling into a n anoxic layer on the pond bottom. Mixing also increases the overall oxygen content in a pond, providing fish and other organisms with a greater habitable volume of water ( Rogers 1989 ) Further evidence of differences in mixing among aeration method s occ urred after the high winds and heavy rainfall associated with Tropical Storm Debby in week 4. Temporary spikes in TAN and nitrite were observed in all ponds in the immediate days following the storm (as well as after other period s of heavy rainfall during the study), indicating that all pond s experienced so me degree of turnover regardless of aeration method However, increases were less pronounced in vertical pump aerator and diffuser box ponds compared to air stone ponds (Figure 4 11) This suggests tha t the increased pond mixing abilities of vertical pump and diffuser box aerators resulted in less turnover action after strong wind and rain and, ultimately, better water quality. Varying dissolved oxygen levels caused by aeration type also may affect biol ogical productivity within a pond. Oxygen is a key requirement for phytoplankton and zooplankton m etabolism, and in turn plankton populations serve as important food sources for la rval fish including zebra danios (Lawrence 2007). Based on Secchi and chlo rophyll a analyse s performed throughout the aeration trial, all experimental ponds in
65 this trial were eutrophic (Florida LAKEWATCH 2000). Although differences in plankton populations were not statistically significant, vertical pump aerator ponds did hav e slightly higher phytoplankton and zooplankton counts and lower Secchi readings than other ponds Although chlorophyll a sampling only approximates phytoplankton content and does not assess periphyton, benthic algae, and macrophyte content (all of which zooplankton, and Secchi values a ll s uggest that ponds with vertical pump aerators had marginally higher biological productivity and, ultimately, more available food for dan io fry compared to the other ponds. Illustrated on a scatterplot (Figure 4 5), t he differences in fish size and survival among aeration methods show an inversely proportio nal relationship between the two variables This relationship supports a concept k where mortality within a population affects subsequent growth rate of remaining survivors (Shepherd and Cushing 1980) In high density populations, individual fish growth rate tends to be slower than that of fish in low densi ty populations Compensatory growth following mortality is desirable in fishery settings because rapid growth and recruitment helps to offset the loss of individuals (Rose et al. 2001) In this study, fish size and yield at harvest suggested some level o f density dependence (Figure 4 5). Ponds with vertical pump aerators yielded relatively low numbers of larger fish and had the highest mortality rate of all aeration methods ( 91 1% ), while diffuser box and air stone ponds had lower mortality (76 1% an d 85 1%, respectively) and individual fish size was smaller.
66 From a production standpoint, the vertical pump aeration method was the only tested method to result in uniformly saleable populations within the 10 week experimental period. However, the n eed for pond side electricity and increased energy costs associated with electrical power may be cost prohibitive for some ornamental producers. While the diffuser box aerator is powered by an air blower and can therefore be ope rated more cheaply, this me thod did not result in significant increases in fish size or yield at harvest compared to the current most commonly used method, air stones. One notable benefit observed with diffuser box aerators was increased pond wide water circulation, which may poten tially improve water quality. Economic factors are important when choosing an aeration method for a production setting A ir stones were the cheapest of the three tested aerator s to build ($25 USD per unit), and this design was simple to construct with m inimal maintenance after installation. Diffuser boxes were more expensive at $92 USD each for materials, yet they require minimal maintenance other than infrequent cleaning (i.e., 1 2 times per year) The vertical pump aerator was the most expensive of t he three tested aerators at $ 795 USD per unit The vertical pump was also the only aerator requiring pond side electrical supply. Electricity installation costs vary depending on facility size and may total more than $10,000 for a small ornamental fish f arm (C. Watson, University of Flori da, personal communication); also, monthly energy costs may exceed those needed to run a single air blower. The vertical pump aerators in this study also required the most maintenance of all tested methods, because they frequently became clogged with aquatic plants. Clogging compromised aerator efficiency, and unclogging was necessary to prevent motors from malfunctioning due to overheating Ultimately, a
67 p of rela tive cost and Future studies sho uld address potential reasons for the universally high mortality rates observed among all aeration methods in this study. Studies utilizing periodic inv entories of fish may help determine the timing and rate of mortality at various life stages in ornamental pond production. If mortality is indeed at or shortly after stocking, then perhaps husba ndry methods can be modified to increase survival of larval f ish. Future analyses of larval rearing methods may pinpoint major causes of mortality in first feeding fish and help increase s urvival rate s eit her by r aising fry in a n indoor hatchery until they outgrow the delicate larval stage or by using a gentle aer ation method with minimal water flow until fry grow into more robust juveniles This research provides helpful information that can be applied in future pond aeration studies using other ornamental species. Survival and growth data may differ in other s pecies based on their environmental and water quality requirements For example, larger and /or more robust ornamental species (e.g., cichlids, goldfish) as well as species that are typically stocked into ponds as adults (e.g., livebearers), may prove mor e tolerant of the increased water movement associated with some of the aeration methods u sed in this study. Future studies examining air stones, diffuser box aerators, and vertical pumps in these species will provide additional information to determine th eir potential uses in the ornamental fish industry.
68 Figure 4 1 Schematic indicating aerator location within ponds for a) air stone s; b) fine bubble diffuser box; and c) vertical pump pond aerator. Arrows indicate directional water flow generated by each aeration method.
69 Figure 4 2 Example of a fine bubble diffuser box aerator constructed for the study. This aerator was desig ned by Carlos Martinez, UF TAL e xtension f aculty. The aerator frame (A) provided buoyancy for the aerat ion mechanism, which consisted of multiple rows of diffuser tubing supplied by a flexible PVC airline (B).
70 Figure 4 3 Zebra danio growth curves during a 10 week growout period in ponds aerated using three methods. Fry were stocked at 5 d ays post hatch. Total length (in mm) is based on subsamples (n=120 per aeration method) measured in weekly intervals. Individual fish weight (g) was calculated from pooled weights measured for subsamples (n=120 per aeration method) in weekly intervals. L ines represent mean SE for each aeration method.
71 Figure 4 4. Photograph illustrating size difference among Danio rerio fry raised in ponds with different aeration methods. These f ish were stocked on day 1 at 5 days post hatch and sampled on day 3 0 of the 70 day aeration trial. Difference in growth rate was statistically significant among aeration methods in weeks 3 10 of the study.
72 Figure 4 5. Scatterplot illustrating the inversely proportional relationship between number of fi sh at harvest and average individual fish weight. This relationship is suggestive of compensatory density dependence.
73 Fig. 4 6. Total length frequency histograms of zebra danio fry after 10 weeks of growth in outdoor production ponds aerated us ing three different methods. Fish were stocked at 5 days post hatch. Histogram data is based on representative fish s ubsamples measured at harvest. Mean s aleable size is indicated by the dotted line
74 Figure 4 7. A clinically normal zebra danio ( A ) compared to a fish with severe kyphosis /scoliosis ( B ). Both fish were collected from the same pond during week 8 of the aeration trial. Occurrence of skeletal deformities ranged from 3 to 30% in experimental ponds.
75 Figur e 4 8 Dissolved oxygen and oxygen saturation measurements for research ponds containing one of three aeration methods (vertical pump aerator; fine bubble diffuser box; air sto nes). Ponds were stocked with zebra danio fry on day 1, and readings were taken t wice daily for 10 weeks Data points are expressed as mean values for each aeration method.
76 Figure 4 9. Diurnal curves constructed from dissolved oxygen measurements performed in two hour intervals over a 24 hour period (starting and ending b ef ore dawn) during week 5 of the zebra danio aeration study. Lines represent mean values for each aeration method.
77 Figure 4 10. Dissolved oxygen profiles constructed from oxygen measurements taken in the deepest area of experimental ponds aer ated using three methods. Readings were taken in 0.5 m increments beginning at the water surface (0 m). Lines indicate mean values for each aeration method.
78 Figure 4 11. Total ammonia nitrogen (TAN) and nitrite concentrations following application of cottonseed meal in small ornamental ponds aerated using three methods. Curves illustrate an initial increase in TAN followed by high nitrite levels. Parameters stabilized by week 4; however, Tropical Storm Debby (indicated by arrow s ) correlated with tem porary increases in TAN and nitrite.
79 Figure 4 12. Zooplankton densities measured during a 10 week trial using ponds stocked with zebra danio fry and using three aeration methods. Densities are illustrated for weeks 3 10.
80 Table 4 1. Means ( SE) for zebra danios harvested following a 10 week growout period in small outdoor ponds aerated using three methods. Initial stocking dens ity was 79,500 fry per pond. Significant differences ( P < 0.05) are noted by different letters. Aeration method Tot al mass (kg) Avg. fish weight (g) Estimated # fish/pond (x1000) Diffuser box 3.46 (0.54) 0.33 (0.06) y 11.76 (2.93) y z Air stones 5.23 (0.36) 0.27 (0.02) y 19.36 (1.82) z Vertical pump 4.44 (0.80) 0.65 (0.06) z 6.89 ( 1.25) y
81 Table 4 2. External biopsy findings on zebra danios examined three days prior to harvest after a 10 week growout period in ponds aerated with three methods. Twenty fish were examined per replicate, with four repli cates per aeration method. Intensity levels for findings were recorded as Statistically significant pathologic findings among aeration methods are noted by different letters. Tissue Pathologic finding Aerat ion method Occurrence (%) Mean intensity Intensity range Skin Excess mucus Air stones 0 y n/a n/a Diffuser box 6.3 yz L L Vertical pump 8.8 z L L Fin Excess mucus Air stones 0 n/a n/a Diff user box 5.0 L L Vertical pump 6.3 L L Gill Excess mucus Air stones 23.8 y L L H Diffuser box 31.3 y L L M Vertical pump 63.8 z L L H Telangiectasis Air stones 23.8 y L L H Diffuser box 31.3 y L L M Vertical pump 63.8 z L/M L H Digene larvae Air stones 33.8 z L L Diffuser box 16.3 yz L L Vertical pump 0 y n/a n/a
82 Table 4 3. Mean SE (r ange in parentheses) of dissolved oxygen (DO), water quality and chemistry variables in ponds stocked with zebra danio fry and aerated using three methods. Means were calculated from data collected over a 10 week period. Significant differences ( P < 0.05 ) are noted by different letters. Variable Air stones Diffuser box Vertical pump Pre dawn DO (mg/L) 5.1 0.05 (3.8 6.6) x 6.4 0.03 (4.9 7.6) y 7.2 0.02 (6.4 7.7) z A fternoon DO (mg/L) 8.8 0.14 (5.5 14.1) z 8.5 0.09 (6.9 11.1) z 8.1 0.05 (6.8 9.5) y Pre dawn oxygen saturation (%) 66.5 0.5 (49.0 82.9) x 83.7 0.3 (63.0 93.6) y 91.0 0.2 (83.0 95.1) z A fternoon oxygen saturation (%) 117.7 2.1 (68.0 184.7) z 114.0 1.5 (90.1 154.2) z 107.1 0.8 (91.8 124.1 ) y Pre dawn temperature (C) 29.0 0.16 (25.3 31.2) z 29.1 0.09 (25.3 31.2) z 27 .9 0.15 (25.0 29.9) y A fternoon temperature (C) 31.1 0.21 (26.3 33.4) z 31.1 0.2 1 (27.0 33.4) z 30.0 0.19 (26.4 32.3) y pH 8.04 0.04 (7.5 9.5) 8.03 0.03 (7.0 8.5) 8.23 0.05 (7.0 9.0) Total ammo nia nitrogen (mg/L) 0.74 0.11 (0.3 2.3) 0.67 0.08 (0.4 2.3) 0.65 0.08 (0.4 1.8) Nitrite (mg/L) 0 .12 0.05 (0 2.31) y 0.17 0.09 (0 1.98) yz 0.19 0.1 (0 2.97) z Hardness (mg/L CaCO 3 ) 309.1 10.4 (25 6.5 393.3) 306.1 13.6 (239.4 410.4) 304.4 10.6 (222.3 376.2) Alkalinity (mg/L CaCO 3 ) 134.7 5.2 (102.6 171.0) 128.7 5.0 (102.6 171.0) 131.2 5.1 (102.6 171.0) Chlorophyll a 32.1 3.8 (6.6 151.8) 31.7 5.0 (7.8 97.0) 43. 5 4.7 (6.2 112.6) Secchi 1.18 0.06 (0.5 >1.5) 1.19 0.06 (0.6 >1.5) 1.02 0.06 (0.5 >1.5)
83 Table 4 4. Nitrogen components in ponds following fertilization with cottonseed meal. Values are given as mean ( SE) for each aer ation method. Nitrogen cycle length equals the number of days after cottonseed meal application until nitrite concentration fell below stabilized below 0.1 mg/L. Significant differences ( P < 0.05) are noted by different letters. Aeration method Pea k TAN (mg/L) Peak TA N (days after pond fert ilization ) Peak nitrite (mg/L) Peak nitrit e (days after pond fer tilization ) Length of nitrogen cycle (days) Diffuser box 1.65 (0.30) 13.3 (0.5) z 1.74 (0.19) 20.8 (1.1) yz 27.5 (0.9) y Air stones 1.65 (0 .34) 12.0 (1.8) yz 1.02 (0.45) 26.8 (2.5) z 40.3 (1.0) z Vertical pump 1.55 (0.12) 7.8 (0.9) y 1.87 (0.38) 20.0 (1.1) y 28.3 (0.6) y
84 CHAPTER 5 POST HARVEST MEDICATION STUDY USING ZEBRA DANIOS Overview In pond production of ornamental fish harvest technique typically involves using seine and dip nets as the primary methods of collection (Crosby et al. 2005a). Harvest is a known cause of trauma and stress related disease outbreaks in cultured fish (Minchew et al. 2007). Ne tting fish potentially causes mechanical trauma, such as torn fins and loss of scales and protective mucus. Mechanical trauma is not only natural protective barriers, cr eating routes for pathogen introduction. Traditional handling also requires temporary exposure to air, which is a known stressor in fish (White et al. 2008) Crowding in nets and holding containers may potentially compromise water quality, resulting part icularly in increased ammonia and carbon dioxide, and low dissolved oxygen (Carmichael et al. 1984). Low DO is especially a concern in hot weather, which is often unavoidable in central Florida where the majority of domestic freshwater ornamental species are raised. Ultimately, harvest presents many potential stressors for f ish, and catastrophic losses, although not common, are possible. Stress has been shown to negatively affect the health of many verte brate species, including fish, through associated im munocompromise (Pasnik et al. 2010). The physiological effects of stress have been studied in many teleost species (Barton and Iwama 1991). Disease outbreaks after stressful events such as harvest may potentially lower fish survival and quality Therefo re, some ornamental producers hold fish for a certain time period after harvest for acclimation and disease surveillance
85 (Crosby et al. 2005a). Common pathogens that may present after harvest of ornamental species include bacterial ( Aeromonas spp., Flavob acterium columnare and Streptococcus spp.), parasitic ( Trichodina spp., Ichthyophthirius multifiliis ), fungal ( Saprolegnia spp.), and viral diseases (spring viremia of carp) (Noga 2010, Pasnik et al. 2010). As a preventative measure, some farmers use th e post harvest holding period as an opportunity to institute one or more prophylactic medications to minimize risk of disease outbreaks. Columnaris disease in fish is an infection with the bacterium Flavobacterium columnare (formerly known as Flexibact er columnaris ), which is typically a secondary pathogen associated with stress and trauma (Noga 2010). Although c olumnaris is commonly considered an external pathogen, serious infections may become systemic (Hawke and Thorne 1992). Clinical signs include listlessness and a characteristic gray to white discoloration of the skin, fin, and gills, indicating tissue necrosis and infiltration of bacteria (Plumb 1999). Most coldwater and warmwater freshwater fish species are susceptible to c olumnaris (Plumb 199 9). It is considered one of the most costliest diseases in channel catfish Ictalurus punctatus production because it has the potential to cause rapid, high mortality rates and result in devastating economic losses (Wagner et al. 2002). c olumnaris is also considered an economically important disease within the trop ical ornamental fish industry, and it can affect all life stages of fish within multiple families (Starliper and Schill 2011). The use of proph ylactic medications in ornamental fish for pathoge ns such as Flavobacterium columnare may vary widely depending on the fish species, season and locale drug availability and producer One of the most common additive s
86 applied to freshwater ornamental species is salt (NaCl). Salt is an effec tive tr eatment for many ectoparasites, and its effectiveness for management of diseases caused by Flavobacterium columnare and water mold s ( Saprolegnia ) has also been studied (Noga 2010). Salt is also commonly used for osmoregulatory support during times of stress ( Crosby et al. 2005b; Pasnik et al. 2010). Other therapeutic agents have also been used; however, the comparative success of commonly used drugs for broad spectrum prophylactic use is currently unknown. This study examined the effects of three broad spectrum medications in zebra danios after harvest from small ponds. Following harvest, f ish were held for 14 days in indoor systems and treated with one of three medications: 1) bronopol; 2) diquat; 3) hydrogen peroxide. The major goal of this st udy was to examine the relative effect of these three agents on post harvest appearance, behavior, and survival of ornamental fish. Methods Pond Harvest This study took place at UF TAL in Ruskin, Florida, in July 2012. An outdoor earthen pond at UF TAL was used to raise zebra danio fry for this trial. One day prior to harvest, a subsample of 60 fish was collected from the pond for health assessments, which were performed within 3 hours of capture. Assessments consisted of physical examination and exter nal tissue biopsies to estimate external pathogen load and overall health in the population. Diagnostic procedures were performed on all 60 fish using methods described by Roberts et al. (2010) Any gross pathological signs (such as anatomic deformities or lesions) were noted. Skin, fin, and gill biopsies were then collected from each fish. The skin biopsy was collected from the left lateral body wall
87 starting just behind the operculum and extending to the caudal fin. Approximately 50% of the caudal fin was removed for examination. A gill biopsy was collected from a right gill arch. Wet mounts of each tissue were prepared using fresh water. Immediately following biopsy collection, fish were euthanized in 2.5 g/L tricaine methanesulfonate (MS 222; Tricaine S; Western Chemical, Ferndale, Washington) buffered with sodium bicarbonate. External tissue wet mounts were examined within 5 minutes of biopsy collection. Wet mounts were examined using light microscopy at 40x, 100x, and 200x magnifications. Pathological findings were quantified using three gradations (light, moderate, and heavy). filament measured less than 25% the width of the widest section of the filament. mucus extended 25 mucus measured 50% or greater than the gill filament width. Telangiectasis was noted 40% of secondary la secondary lamellae affected). The following day, 1 200 fish were harvested from the pond using a seine net (18.3 m x 2.4 m, 1.6 mm mesh) and transported using coolers to the indoor resea rch facility Fish were arbitrarily stocked into 32 L aquaria at a density of 50 fish per tank. A total of 24 tanks were used for the experiment. Total transport and acclimation time was approximately 1 hour. Medication Trial The experimental system was located in an indoor facility with a constant temperature of 26 27 C. Each aquarium was equipped with a 5 c m silica air stone for aeration and flow through supply of aerated well water. Water flow to each
88 aquarium was adjusted to 32 L per hour (one aquarium volume per hour) and discontinued during the administration of any treatments. A 12:12 (light:dark) photoperiod was maintained throughout the trial. Fish were fed once daily to satiation with 2% body weight of tropical fish meal (Cargill Animal Nutrition, Inc., Minneapolis, Minnesota). Waste and uneaten food were siphoned from each aquarium once daily. Aquaria were observe d four times daily (0900, 1200, 1500, and 1800 hours) throughout the 14 day trial for mortalities. Two hours after initia l stocking, fish appearance and behavior in each aquarium were assessed based on a ranked, categorical scale for evaluation of ornamental fish ( Kilgore et al. 2009; Table 5 1). This scoring system evaluates appearance on a scale of 1 5 and behavior on a 1 4 scale. Higher scores indicate superior appearance and behavior, and scores of 3 or above in each category indicate fish populations of saleable quality. Examples of appearance criteria include scale loss, external ulceration or hemorrhage, torn fins, excess mucus, emaciation, and anatomical deformities such as kyphosis. Behavioral criteria include abnormal position or orientation in the water column, spinning, flashing, piping or increased respiratory effort, listlessness, and clamped fins. Six judge s were chosen for this trial based on their backgrounds in aquaculture and fis h health. Jud ges were trained on the scoring criteria prior to judging and were shown example aquaria to evaluate and become familiarized with using the scoring system. All jud ges were blinded to the experimental trea tments throughout the trial. Following judging, each aquarium was randomly assigned to one of four treatments The examined medications ( bronopol, diquat, and hydrogen peroxide ) were chosen for this study based on their potential use against common post harvest
89 pathogens, namely Flavobacterium columnare. A control (no treatment) was also examined. Six replicate tanks per treatment were used. Medicated aquaria received the first treatment on day 1 (i.e., stocking day) immediately after the pre trial judging concluded. Water flow was discontinued to individual tanks receiving treatment and was restarted with conclusion of each treatment period. Water flow to control tanks remained constant throughout the entire h olding period. Examined Medications Bronopol ( 2 bromo 2 nitro 1,3 propanediol ) is a thiol containing dehydrogenase enzyme inhibitor with antifungal and antibacterial properties. This biocide has traditionally been used in the cosmetic and pharmaceutical industries and only recently gained popularity as a therapeutic agent in fish. Marketed as Pyceze (Novartis Animal effectiveness is similar to malachite green, a commonly used medication in cultured f ish species (Cawley 1999). Bronopol has been advertised as a safer alternative to malac hite green, which has known carc inogenic and teratogenic risks to h umans (Meyer and Jorgensen 1983; Sudova et Saprolegnia a com mon water mold infection in fish, has been validated (Cawley 1999). Pyceze is currently approved in the European Union only for treatment of water mold infections in salmonids and salmonid eggs and is available with veterinary prescription (Noga 2010). In this study, bronopol (Acros Organics, Geel, Belgium) was dosed as a 5 mg/L bath immersion for 6 hours on days 1, 3, 5, and 7. Di quat [ 6,7 dihydrodipyrido (1,2 a:2 ',1' c) pyrazinediium dibromide] is a U.S. EPA approved herbicide that is used for contr mechanism results in damage and death to the cellular structure of the target species.
90 Diquat also has uses in aquaculture as a therapeutic treatment for external bacterial diseases in fish, namely Flavobacterium columnare (Chen et al. 1982; Darwish and Mitchell 2009 ). Experimentally, diquat dosed at a 5.4 mg/L concentration has resulted in 100% survival in channel catfish challenged with Flavobacterium columnare compared to 100% mortality in challenged untreate d fish (Thomas Jinu and Goodwin 2004). Other studies show that a minimum dose of 5.0 mg/L is effective for treatment of c olumnaris disease (Darwish and Mitchell 2009). In this study, d iquat dibr omide at a rate of 4.9 mg/L for 6 hours on days 1, 3, 5, and 7. Hydrogen peroxide (H 2 O 2 ) is a strong oxidizing agent used for a wide range of industrial and medical purpo spectrum properties are known be effective against numerous bacterial, fungal, and parasitic pathogens in aquatic species (Russo et al. 2007). In recent years, hydrogen peroxide has increased in popularity as a potential th erapeutic agent for cultured fish, and in 2007 a 35% formulation known as PEROX AID (Eka Chemicals, Marietta, GA) was approved by the FDA for treatment of Saprolegnia, Flavobacterium branchiophilum, and Flavobacterium columnare in eggs and fish of select food fish species. Other documented target pathogens include numerous external parasites such as Ambiphyra and Gyrodactylus (Rach et al. 2000) and Amyloodinium (Montgomery Brock et al. 2001). In this study, h ydrogen peroxide (Walgreen Co., Deerfield, Ill inois) was dosed as a 5 mg/L immersion bath for 6 hours on days 1 and 3. At the end of each 6 hour treatment, hydrogen peroxide concentrations were measured in treated aquaria using a hydrogen peroxide titration kit (Model HYP 1; Hach Co., Loveland, CO). Residual
91 concentrations, which measured 4 5 mg/L in all aquaria, confirmed t he medication had persisted at therapeutic concentrations for the duration of treatment. Biosecurity and Tank Maintenance Biosecurity measures were implemented throughout the tria l to avoid contamination among systems and the rest of the facility. Prior to stocking, the tank system was disinfected with 10% sodium hypochlorite solution. Each aquarium was supplied with separate water and air lines and was fitted with a glass lid. Maintenance equipment was disinfected with Virkon Aquatic between aquariums and after use. Foot baths containing Virkon Aquatic were placed at each entrance to the building, and after daily siphoning the floor was disinfected to prevent spread of any path ogens. Water parameters dissolved oxygen, water temperatu re, total ammonia nitrogen (TAN), nitrite, pH, total al kalinity, and total hardness were measured during the experimental period to confirm similar water chemistry among treatments. DO and temperat Inc., Yellow Springs, Ohio). TAN, nitrite, pH, total hardness, and total alkalinity were analyzed using a Hach Freshwater Fish Farmer's Kit (Model FF 1A; Hach Co., Loveland, Colorado) Water parameters were indeed statistically similar among treatments and mean values ( standard error, when applicable) were as follows: DO = 8.3 mg/L; temperature = 26.7 C; pH = 8.2; TAN = 0.5 mg/L; nitrite = 0 mg/L; total alkalinity = 180 1.1 mg/L ; total hardness= 348.7 2.4 mg/L. At the end of the 14 day experimental period, behavior and appearance scoring was repeated with the same six judges used prior to the trial. After judging, health assessments were performed on a 10% representative sam ple (i.e., five fish per
92 replicate ) Health assessment methodology was similar to that described for pre trial assessments. Statistical Analysis Water quality parameters were analyzed using one way, repeated measures analysis of variance (ANOVA) (MIXED pr ocedure). Health assessment findings and judging scores were analyzed using the Kruskal Wallis (NPA1WAY) procedure, followed performed in SAS version 9.3 (SAS Institu te Inc., Cary, North Carolina). Numerical data are represented as mean SE Statistical differences were considered significant at P < 0.05 with the exception of which was significant at P < 0.15. Results Survival was 100% for all trea tment groups; therefore, no statistical analysis was required to assess survivability. Pre trial appearance and behavior scores were similar among treatment types (appearance: F value = 0.2428, df 3, P = 0.8426; behavior: F value = 5.4993, df 3, P = 0.518 3). After the 14 day experimental trial, mean appearance score for bronopol tanks was significantly lower than other treatments ( F value = 5.4993, df 3, P = 0.0013 ); yet, mean score still ranked fish from all treatments as saleable. Behavior scores afte r treatment were similar among treatments ( F value = 1.2460, df 3, P = 0.2935). Fish from all categories improved in mean appearance and behavior scores after treatment (Figure 5 1 and 5 2), with the exception of those treated with bronopol. Average beha vior scores improved for all fish after treatment. Sixty fish examined prior to the drug trial showed no occurrence of pathogens on skin, fin, and gill biopsies. Excess mucus was observed on the skin (1.7% occurrence), fin (5%), and gill (35%) tissue. Telangiectasis was observed in 71.7% of examined fish,
93 and 5% of fish also had gill congestion. A fter the 14 day experimental period, most health assessment findings were not significantly different in occurrence or severity among treatment gro ups These 2 = 2.8287, df = 3, P 2 = 2.2655, df = 3, P = 0.1534), which each occurred only in light levels in 13% or fewer fish and had similar occurrence compared to the pre trial population (Table 5 2). G ill telangiectasis also occurred in statistically similar levels among treatment groups 2 = 5.5361, df = 3, P = 0.1365) but was decreased in all groups compared to pre trial fish The severity of excess mucus on the gills was significantly different among 2 = 9.8512, df = 3, P = 0.0199). Fish treated with diquat had no decrease in gill mucus compared to pre treatment fish, while those treated with bronopol or hydrogen peroxide had significantly decreased occurrences of excess gill mucus versus pre treatment fish. Excess gill mucus was only slightly decreased compared to pre treatment fish. Discussion This study was designed to experimentally test three potential treatments for Flavobacterium columnare in zebra danios; however, we were unable to induce disease in the experimental population from harvest and handling performed immediately prior to the trial. Indeed, all fish used in this study were genera lly healthy throughout the study Twenty two out of 24 replicate samples were rank ed as saleable bot h before and after treatment, and t he remaining two tanks only received one unsaleable score by a single judge. Scores were similar for control samples as well as fish that were treated with one of three tested medications C orrespondin gly, health assessments revealed few physical abnormalities and no pathogenic organisms
94 The overall good health of animals used in this study is perhaps a testament to ely hardy nature. However, reproducing disease in a control led environment particularly for a single trial, is a complex and difficult task. Unfortunately the results of this study should not be considered universally conclusive for the use of prophylaxis in ornamental aquaculture. Based on the results of this study, it is tempting to assume that apparently healthy f ish do not requ ire prophylactic treatment after harvest. T his study showed that prophylaxi s did not significantly improve health or survival in treated fish versus untreated controls It is import ant however, to r ecognize the potentially serious risk of disease outbreaks in production settings M any common ornamental fish pathogens (e.g., Flavobacterium columnare and Ichythyophthirius multifiliis ) may not manifest for days to weeks following a st ressf ul event like harvest and can result in high mortality rates In outbreaks of disease, treatment is recommended to m inimize losses. In addition to instituting prophylaxis when necessary production facilities should construct a disease management pla n to minimize disease risks and prepare for potential outbreak s Treatment protocols should consider the many variables that contribute to disease epidemiology such as presence of predisposing environmental factors, the fish species being cultured and it s relative susceptibility to various pathogens For example, poeciliids (e.g., platies, mollies) are highly susceptible to infection with Flavobacterium columnare and prompt intervention is recommended to minimize the risk of devastating losses from an o utbreak. Other factors, such as environment (e.g., season, temperature, and locale) and husbandry conditions (e.g., system design and biosecurity measures) are equally important considerations when
95 designing a fish health management plan All of the abo ve factors should be included when constructing a disease prevention protocol for a production facility. Practicing responsible husbandry measures such as biosecurity and quarantine, are also important, an d facilities should consult with a fish health pr ofessional when designing a health management plan or whenever disease is a concern.
96 Figure 5 1. Zebra danio appearance scores before ( A ) and after ( B ) a 14 day holding period in tanks with four treatment types: control, bronopol, diaquat, and hydrogen peroxide. Scores of 3 5 indicate saleable fish populations as determined by six judges for 6 replicate aquaria per treatment, each stocked with 50 fish.
97 Figure 5 2. Zebra danio behavior scores scores before ( A ) and after ( B ) a 14 day holding period in tanks with four treatment types: control, bronopol, diaquat, and hydrogen peroxide. Scores of 3 4 indicate saleable fish populations as determined by six judges for 6 replicate aquaria per treatment, each stocked with 50 fish.
98 Table 5 1 Ranked categorical scoring system u sed by judges for evaluation of zebra danio appearance and behavior at the beginning and end of a 14 day holding period. Score Appearance Score Behavior UNSALEABLE U NSALEABLE 1 > 50% abnormal with mild t o moderate damage or 25 50% with severe damage 1 >50% abnormal 2 25 50% with mild to moderate damage or 10 25% with severe damage 2 25 50% abnormal S ALEABLE SALEABLE 3 3 4 Overall normal, few ab normalities 4 Overall normal, all fish alert 5 Overall excellent, no abnormalities Kilgore, K.H., J.E. Hill, J.F.F. Powell, C.A. Watson, and R.P.E. Yanong. 2009. Investigational use of metomidate hydrochloride as a shipping additive for t wo ornamental fishes. Journal of Aquatic Animal Health 21: 133 139.
99 Table 5 2. External biopsy findings on zebra danios examined before and after a 14 day holdin g period with one of four treatments (TX) Intensity levels for findings were different pathogenic findings among aeration methods are noted by different letters Tissue Pathologic finding Pre /Post treatment T reatment Occurrence (%) Mean intensity Range Sk in Excess mucus Pre n/a 1.7 L L Post Bronopol 10.0 L L Diquat 6.7 L/M L M H 2 O 2 0 n/a n/a Control 6.7 L L Fin Excess mucus Pre n /a 5.0 L L Post Bronopol 3.3 L L Diquat 13.3 L L H 2 O 2 0 n/a n/a Control 6.7 L L Gill Excess mucus Pre n /a 35.0 z L L M Post Bronopol 6.7 y L L Diquat 33.3 z L L H H 2 O 2 10.0 y L L Control 13.3 yz L L Telangiectasis Pre n /a 71.7 z L L H Post Bronopol 40.0 y L L M Diquat 13.3 y L L M H 2 O 2 26.7 y L L Control 23.3 y M L H
100 CHAPTER 6 CONCLUSION S Recent i mprovement s of aquaculture production p ractices are due in part to extensive research aimed at evaluating and bettering husbandry and health of cultured fish populations The majority of aquacultur e research conducted thus far has focused on non ornamental fish species. Contrastingly basic h usbandry standards are still being scientifically established for ornamental species, and even less is known regarding the emerging area of marine ornamental species culture. Most freshwater ornamental fish raised in Florida are grown in outdoor ponds. Some standards for ornamental species have been adapted from pond production of food fish. Other husbandry practices are based on empirical observation but have not been formally studied. The research in this thesis was designed specifically to address t he lack of knowledge regarding freshwater ornamental fish production. In particular, little is known regarding the use of supplemental aeration in earthen pond culture. Aeration is accepted as an important component of outdoor food fish production, parti cularly due to extensive catfish research conducted in the southeastern United States. By contrast, the importance of aeration in ornamental pond production has not yet been validated. The aeration trials performed in this study compared the current st andard ornamental pond aeration method (small air stone diffusers) with two less commonly used aerators in the industry (fine bubble diffuser box, vertical pump aerator). The latter two methods are more commonly used in other areas of aquaculture ; however their use in ornamental systems has not been formally evaluated until now. This research
101 examined the influence of pond aeration methods not only on fish productivity but also on biological, physical, and chemical pond parameters. Air stones (the stan dard method of ornamental pond aeration) resulted in the lowest morning dissolved oxygen (DO) levels in small outdoor ornamental ponds, and DO frequently fell below the minimum recommended concentration of 5 mg/L. Ponds with diffuser boxes had higher morn ing DO levels, and vertical pump vertical pond aerators maintained morning oxygen concentrations close to saturation. Zebra danio ( Danio rerio ) growth rate differed significantly among aeration methods. The maximum growth rate for fry raised in ponds with vertical pump aerators was more than twice as rapid as fish growth rate in ponds aerated with the other two methods. After a 10 week growth period, danios from ponds with vertical pump aerators were visibly larger and more robust than those harveste d in other ponds. By contrast, fish raised in ponds with air stones were much smaller at harvest. This remarkable increase in growth rate may be beneficial for producers seeking a faster turnaround time in pond production. Fish survival was also notab ly different among aeration methods. Survival rates were universally poor, with the best survival occurring in ponds with air stones (24%). Ponds with diffuser boxes had 15% survival, and only 9% survival occurred in ponds with vertical pump aerators. T he reason for such high mortality in all ponds could not be determined; however, heavy losses may have occurred from stocking fry at a young age (5 days post hatch). Higher mortality rates in ponds with vertical pump aerators may have been due to the vigo rous water agitation caused by those aerators. Although the cause for mortality was not determined, the inversely proportional relationship between
102 survival and fish size was suggestive of density dependence, a concept that has been described in both wild and cultured fish populations. Pond productivity, which was assessed through the measurement of several physical, biological, and chemical parameters, was generally similar among ponds with different aeration methods. Plankton analysis, performed by me asuring chlorophyll a and zooplankton densities, revealed only minor differences among aeration methods. Water clarity, measured via Secchi disk readings, was also statistically similar among aeration types. The majority of measured water quality paramet ers, such as pH, hardness, and alkalinity, were also similar. Interestingly, poor water quality issues (specifically high ammonia and nitrite) took significantly longer to correct in ponds with air stones. This was seen both after the application of cot tonseed meal fertilizer and after a tropical storm that occurred in week 3 of the aeration trial and caused some degree of turnover in all ponds. Poor water quality likely took longer to correct in ponds with air stones due to a combination of factors. N amely, dissolved oxygen levels were lower, and the air stones had less mixing ability than the other two pond aeration designs. A drug study performed after harvest yielded high survival and good overall health both in zebra danios that received medication treatments (bronopol, diquat, hydrogen peroxide) and those that were left untreated. This single trial was inconclusive for establishing relative efficacies of the three medications as prophylactic treatments in ornamental fish species. In conclusion, lo w oxygen levels are associated with poor growth rate, decreased food consumption, and increased occurrence of disease. Indeed, this study showed
103 that morning oxygen levels correlated positively with growth rate in zebra danios. Fish that were raised in p onds with oxygen levels at or just under saturation grew twice as quickly as those exposed to lower overnight DO levels. On the other hand, DO did not have a positive impact on zebra danio survival. In fact, at harvest the highest yields of fish came fro m the ponds with the lowest morning DO levels. Dissolved oxygen and water quality continue to be important parameter s for fish health. Low DO is also one of the most preventable of poor water quality issues, and oxygen issues can be corrected through t he use of supplemental aeration techniques. It is my hope that this research illustrates the influence that aeration has not only on cultured fish but also on pond parameters such as water quality and productivity.
104 LIST OF REFERENCES Arce, R.G., and C.E. Boyd. 1975. Effects of agricultural limestone on water chemistry, phytoplankton production, and fish production in soft water ponds. Transactions of the American Fisheries Society 2: 308 312. Baras, E., and M. Jobling. 2002. Dynamics of intr acohort cannibalism in cultured fish. Aquaculture Research 33: 461 479. Barton, B.A., and G.K. Iwama. 1991. Physiological changes in fish from stress in aquaculture with emphasis on the response and effect of corticosteroids. Annual Review of Fish Dise ases 1: 3 26. Boyd, C.E. 1990. Water quality in ponds for a quaculture. Birmingham Pu blishing Company, Birmingham, Alabama Boyd, C.E. 1998. Pond water aeration systems. Aquacultural Engineering 18(1): 9 40. Boyd, C.E. and C.S. Tucker. 1979. Emerg ency aeration of fish ponds. Transactions of the American Fisheries Society 108: 299 306. Boyd, C.E., and C.S. Tucker. 1992. Water quality and pond soil analyses for a quaculture. Auburn University Agricultural Experiment Station, Auburn University, A labama. Brown, D.D. 1997. The role of thyroid hormone in zebrafish and axolotl development. Proceedings of the National Academy of Sciences of the USA 94 : 13011 13016. Brungs, W.A. 1971. Chronic effects of low dissolved oxygen concentrations on the fa thead minnow ( Pimephales promelas ) Journal of the Fisheries Research Board of Canada 2 8(8) : 1119 1123. Buentello, J.A., D.M. Gatlin III, and W.H. Neill. 2000. Effects of water temperature and dissolved oxygen on daily feed consumption, feed utilization and growth of channel catfish ( Ictalurus punctatus ). Aquaculture 1 82(3 4) : 339 352. Carmichael, G.J., J.R. Tomasso, B.A. Simco, and K.B. Davis. 1984. Confinement and water quality induced stress in largemouth bass. Transactions of the American Fisheri es Society 113: 767 777. Carvalho, A.P., L. Arajo, and M.M. Santos. 2006. Rearing zebrafish ( Danio rerio ) larvae without live food: evaluation of a commercial, a practical and a purified starter diet on larval performance. Aquaculture Research 3 7(11) : 1107 1111. Cawley, G.D. 1999. Bronopol: an alternative fungicide to malachite green? Fish Veterinary Journal 3: 79 81.
105 Chabot, D. and G. Claireaux. 2008. Environmental hypoxia as a metabolic constraint on fish: the case of Atlantic cod, Gadus morhua Marine Pollution Bulletin 5 7(6) : 287 294. Chang, W.Y.B. and H.Ouyang. 1988. Dynamics of dissolved oxygen and vertical circulation in fish ponds. Aquaculture 74: 263 276. Chen, C. R.L., Y.Y. Chung, and G.H. Kuo. 1982. Studies on the pathogenicity of Flexibacter columnaris 1. Effect of dissolved oxygen and ammonia on the pathogenicity of Flexibacter columnaris to eel ( Anguilla japonica ). CAPD Fisheries series No. 8, Reports on Fish Disease Research 4: 57 61. Chu, D.L.H., V.W.T. Li, and R.M.K. Yu. 2010. Leptin: clue to poor appetite in oxygen starved fish. Molecular and Cellular Endocrinology 319: 143 146. Corte meglia, C., and T.L. Beitinger. 2005. Temperature tolerances of wild type and red transgenic zebra danios. Transactions of the America n Fisheries Society 134: 1431 1437. Crosby, T.C., J.E. Hill, C.V. Martinez, C.A. Watson, D.B. Pouder, and R.P.E. Yanong. 2005. Harvesting ornamental fish from ponds. Fact sheet FA117, Department of Fisheries and Aquatic Sciences, Florida Cooperative Ext ension Service, Institute of Food and Agricultural Science, University of Florida. Available: http://edis.ifas.ufl.edu/pdffiles/FA/FA11700.pdf (October 2012). Crosby, T.C., J.E. Hill, C.V. Martinez, C.A. Watson, D.B. Pouder, and R.P.E. Yanong. 2005. On farm transport of ornamental fish. Fact sheet FA119 Department of Fisheries and Aquatic Sciences, Florida Cooperative Extension Service, Institute of Food and Agricultural Science, University of Florida. Available: http://edis.ifas.ufl.edu/pdffiles/FA /FA11900.pdf (October 2012). Daniels, R.J.R. 2002. Freshwater fishes of peninsular India Universities Press, Hyderabad. Darwish, A.M., and A.J. Mitchell. 2009. Evaluation of diquat against an acute experimental infection of Flavobacterium columnare i n channel catfish Ictalurus punctatus (Rafinesque). Journal of Fish Diseases 32: 401 408. Das, P.C., J. Jena, B. Mishra, and B.K. Pati. 2012. Impact of aeration on the growth performance of silver barb, Puntius gonionotus during fingerling rearing. Journal of the World Aquaculture Society 43(1) : 128 134. Dust, J.V. and A. Shindala. 1970. Relationship of chlorophyll a to algal count and classification in oxidation ponds. Journal of Water Po llution Control Federation 42:1362 1 369. Eaton, R.C., and R .D. Farley. 1974. Spawning cycle and egg production of zebrafish, Brachydanio rerio in the laboratory. Copeia 1974 : 195 204.
106 Evans, J.J., C.A. Shoemaker, and P.H. Klesius. 2003. Effects of sublethal dissolved oxygen stress on blood glucose and suscep tibility to Streptococcus agalactiae in Nile tilapia Oreochromic niloticus Journal of Aquatic Animal Health 15: 202 208. Fang, F. 2001. Phylogeny and species diversity of the South and Southeast Asian cyprinid genus Danio hamilton (Teleostei, Cyprinida e). PhD Thesis, Stockholm University, Stockholm, Sweden. FASS/FDACS. 2001, 2002. Florida Agriculture Statistics Service/ Florida Department of Agriculture and C onsumer Services online article. Available: http://www.nass.usda.gov/Statistics_by_State/Flor ida/Publications/Aquaculture/0 2aqua4.pdf. (September 2012). FASS/FDACS. 2005. Florida Agriculture Statistics Service/Florida Department of Agriculture and Consumer Services online article: http://www.nass.usda.gov/Statistics_by_State/Florida/Publications/A quaculture/0 6aqua4.pdf. (September 2012). Fernandez, S.O., S.S. Kulkarni, R.R. Shirodkar, S.V. Karekar, R.P. Kumar, R.A. Sreepada, C. Vogelsang, and P.A.L. Bharathi. 2010. Water quality and bacteriology in an aquaculture facility equipped with a new aera tion system. Environmental Monitoring and Assessment 164(1 4): 81 92. Information Circular No. 101. Department of Fisheries and Aquatic Sciences and Institute of Food and Agr icultural Sciences, University of Florida, Gainesville, FL. Available: http://edis.ifas.ufl.edu/pdffiles/FA/FA07800.pdf (September 2012). Francis Floyd, R. 1992. Dissolved oxygen for fish production. Fact Sheet FA 27, Department of Fisheries and Aquat ic Sciences, Florida Cooperative Extension Service, Institute of Food and Agricultural Science, University of Florida. Available: http://edis.ifas.ufl.edu/pdffiles/FA/FA00200.pdf (August 2012). Francis Floyd, R. 2002. Fish nutrition. VM114, Department of Fisheries and Aquatic Sciences, Florida Cooperative Extension Service, Institute of Food and Agricultural Science, University of Florida. Available: http://edis.ifas.ufl.edu/pdffiles/FA/FA09600.pdf (October 2012). Gerhard, G.S., E.J. Kauffman, X. Wan g, R. Stewart, J.L. Moore, C.J. Kasales, E. Demidenko, and K.C. Cheng. 2002. Life spans and senescent phenotypes of ebrafish ( Danio rerio ). Experimental Gerontology 37: 1055 1068. Gong, Z., H. Wan, T.L. Tay, H. Wang, M. Chen, and T. Yan. 2003. Develop ment of transgenic fish for ornamental and bioreactor by strong expression of fluorescent proteins in the skeletal muscle. Biochemical and Biophysical Research Com munications 308(1) : 58 63.
107 Goolish, E.M., K. Okutake, and S. Lesure. 1999. Growth and surv ivorship of larval zebrafish Danio rerio on processed diets. North America Journal of Aquaculture 61:189 198. Halver, J.E., L.M. Ashley, and R.R. Smith. 1969. Ascorbic acid requirements of coho salmon and rainbow trout. Transactions of the American Fish eries Society 98: 762 771. Haman, D.Z., G.A. Clark, and D.J. Pitts. 1991. Excavated pond construction in Florida. Circular 939, Department of Fisheries and Aquatic Sciences, Florida Cooperative Extension Service, Institute of Food and Agricultural Scien ce, University of Florida. Available: http://edis.ifas.ufl.edu/pdffiles/AE/AE01300.pdf (August 2012). Hawke J.P., and R.L. Thune. 1992. Systemic isolation and antimicrobial susceptibility of Cytophaga columnaris from commercially reared channel catfish. Journal of Aquatic Animal Health 4: 109 113. Heideman, R.C. 1983. Aeration/circulation method and apparatus utilizing low pressure air. U.S. P atent 4,488,508. December 18, 1984. Hill, J.E. and R.P.E. Yanong. 2002. Freshwater ornamental fish commonly cultured in Florida. Circular 54, Department of Fisheries and Aquatic Sciences, Florida Cooperative Extension Service, Institute of Food and Agricultural Science, University of Florida. Available: http://edis.ifas.ufl.edu/pdffiles/FA/FA05400.pdf (Octob er 2012). Hoff, F.H., an d T.W. Snell. 1993. Plankton culture manual, 3rd edition. Florida Aquafarms, Inc., Dade City, Florida. Howell, B.R., O.J. Day, and T. Ellis. 1998. Early life stages of farmed fish. Pages 27 66 in K.D. Black and A.D.Pickerin g, e ditor s Biology of farmed f ish. Sheffield Academic Press, Boca Raton, Florida. Johnston, T.A., and J.A. Mathias. 1993. Mortality of first feeding postlarval walleye ( Stizostedion vitreum ) in culture ponds. Canadian Journal of Fisheries and Aquatic Sciences 50(9): 1835 1843. Kilgore, K. H. J.E. Hill, J.F.F. Powell, C.A. Watson, and R.P.E. Yanong. 2009. Investigational use of metomidate hydrochloride as a shipping additive for two ornamental fishes. Journal of Aquatic Animal Health 21: 133 139. Kimmel, C.B., W.W. Ballard, S.R. Kimmel, B. Ullmann, and T.F. Schilling. 1995. Stages of embryonic development of the zebrafish. Developmental Dynamics 203: 253 310.
108 Kumar, R., S. Souissi, and J.S. Hwang. 2012. Vulnerability of carp larvae to copepod predation as a function of larval age and body length. Aquaculture 338 341: 274 283. Kurten, G., L. Hall, and N. Thompson. 1999. Evaluation of cottonseed meal supplementation of inorganically fertilized Florida largemouth bass spawning ponds. North American Journal of Aquaculture 61: 115 125. Laale, H.W. 1977. The biology and use of zebrafish, Brachydanio rerio in fisheries research. A literature review. Journal of Fish Biology 10 : 121 173. Lall, S.P. and L.M. Lewis McCrea. 2007. Role of nutrien ts in skeletal metabolism and pathology in fish an overview. Aquaculture 267: 3 19. Lawrence, C. 2007. The husbandry of zebrafish (Danio rerio) : a review. Aquaculture 269: 1 20. Lewis, W.M., and D.P. Morris. 1986. Toxicity of nitrite to fish: a r eview. Transactions of the American Fisheries Society 115(2): 183 195. Lovell, R.T. 1991. Nutrition of aquaculture species. Journal of Animal Science 69: 4193 4200. Ludwig G.M 1999. Zooplankton succession and larval fish culture in freshwater ponds. Publication 700 Southern Regional Aquaculture Center, Cooperative States Research, Education and Extension Service, United States Department of Agriculture. Available: https://srac.tamu.edu/index.cfm/event/getFactSheet/whichfactsheet/140/ (November 2012 ). Ludwig, G.M., Stone, N.M., and C. Collins. 1998. Fertilization of fish fry ponds. Publication 469, Southern Regional Aquaculture Center, Cooperative States Research, Education and Extension Service, United States Department of Agriculture. Available: h ttps://srac.tamu.edu/index.cfm/event/getFactSheet/whichfactsheet/116/ (November 2012). McClure, M.M., P.B. McIntyre, and A.R. McCune. 2006. Notes on the natural diet and habitat of eight danionin fishes, including the zebrafish Danio rerio. Journal of Fish Biology 68: 1 18. McGee, M., and C. Cichra. 1991. The role of aeration in pond management. Fact sheet FA6, Department of Fisheries and Aquatic Sciences, Florida Cooperative Extension Service, Institute of Food and Agricultural Science, University of Florida. Available: http://edis.ifas.ufl.edu/pdffiles/FA/FA02100.pdf (October 2012).
109 McKee, J.E. and H.W. Wolf, editors. 1963. Water quality c riteria, 2 nd ed ition State of California, State Water Quality Control Board, Publ. No. 3 A, Sacramento, California Meyer F.P., and T.A. Jorgensen. 1983. Teratological and other effects of malachite green on the development of rainbow trout and rabbits. Transactions of the American Fisheries Society 112: 818 824. Miller Morgan, T. 2010. A brief overv iew of the ornamental fish industry and hobby Pages 25 32 in H.E. Roberts, editor. Fundamentals of ornamental fish h ealth. Wiley Blackwell, Ames, Iowa Montgomery Brock, D., V.T. Sato, J.A. Brock, and C.S. Tamaru. 2001. The application of hydrogen peroxide as a treatment for the ectoparasite Amyloodinium ocellatum (Brown 1931) on the Pacific threadfin Polydactylus sexfilis. Journal of the World Aquaculture Society 32: 250 254. Noga, E.J. 2010. Fish disease: d iagnosis and t reatment, 2 nd ed ition W iley Blackwell, Ames, Iowa. Pasnik, D.J., J.J. Evans, and P .H. Klesius. 2010. Stress in f ish. Pages 33 38 in H.E. Roberts, editor. Fundamentals of ornamental fish h ealth. Wiley Blackwell, Ames, Iowa Pawar, N.A., J.K. Jena, P.C. Das, and D.D. Bhatna gar. 2009. Influence of duration of aeration on growth and survival of carp fingerlings during high density seed rearing. Aquaculture 290(3 4): 263 268. Peterson, D.L., R.F. Carline, T.A. Wilson, and M.L. Hendricks. 1997. Production scale methods for intensive culture of walleye fry. The Progressive Fish Culturist 59(1): 14 19. Pichavant, K., J. Person Le Ruyet, N. Le Bayon, A. Severe, A. Le Roux, and G. Boeuf. 2005. Comparative effects of long term hypoxia on growth, feeding and oxygen consumption in juvenile turbot and European sea bass. Journal of Fish Biology 59(4): 875 883. Plumb J.A. 1999. Health m ai ntenance and principal microbial diseases of cultured f ishes. Iowa State University Press, Ames, Iowa Rach, J.J., S.M. Schleis, M.Gaikowski, an d A. Johnson. 2003. Efficacy of hydrogen peroxide in controlling mortality associated with external columnaris on walleye and channel catfish fingerlings. North American Journal of Aquaculture 65: 300 305. Rao, T.R. 2003. Ecological and ethological p erspectives in larval fish rearing. Journal of Applied Aquaculture 13: 145 178.
110 Roberts, H.E., E.S. Weber III, and S.A. Smith. 2010. Nonlethal diagnostic techniques. Pages 172 184 in H.E. Roberts, editor. Fundamentals of ornamental fish h ealth. Wiley Blackwell, Ames, Iowa Rogers, G.L. 1989. Aeration and circulation for effective aquaculture pond management. Aquacultural Engineering 8: 349 355. Rogers, G.L., and A.W. Fast. 1988. Potential benefits of low energy water circulation in Hawaiian pra wn ponds. Aquacultural Engineering 7: 155 165. Rose, K.A., J.H. Cowan Jr., K.O. Winemiller, R.A. Myers, and R. Hilborn. 2001. Compensatory density dependence in fish populations: importance, controversy, understanding and prognosis. Fish and Fisheries 2 : 293 327. Russo, R., H. Mitchell, and R.P.E. Yanong. 2006. Characterization of Streptococcus iniae isolated from ornamental cyprinid fishes and development of challenge models. Aquaculture 256: 105 110. Russo, R., E.W. Curtis, and R.P.E. Yanong. 2007. Preliminary investigations of hydrogen peroxide treatment of selected ornamental fishes and efficacy against external bacteria and parasites in green swordtails. Journal of Aquatic Animal Health 19: 121 1 27. Sartory, D.P., and Grobbelaar, J.U. 1984. Extraction of chlorophyll a from freshwater phytoplankton for spectrophotometric analysis. Hydrobiologia 114 : 117 187. Shepherd, J.G., and D.H. Cushing. 1980. A mechanism for density dependent survival of lar val fish as the basis of a stock recruitment relationship Conseil International pour l'xploration de la Mer 39(2): 160 167. Shireman, J.V. and W.J. Lindberg. 1985. A first look at Florida aquaculture. Circular 702, Department of Fisheries and Aquatic Sciences, Florida Cooperative Extension Service, Institute of Food and Agricultural Science, University of Florida. Available: http://ufdc.ufl.edu/UF00072584/00001/ (September 2012). Smith, C., and P. Reay. 1991. Cannibalism in teleost fishes. Reviews in Fish Biology and Fisheries 1: 41 64. Spence, R., M.K. Fatema, S. Ellis, Z.F. Ahmed, and C. Smith. 2007. Diet, growth and recrui tm ent of wild zebrafish in Bangladesh. Journal of Fish Biology 71 : 304 309. Spence, R., G. Gerlach, C. Lawrence, and C. Sm ith. 2007. The behaviour and ecology of the zebrafish, Danio rerio Biological Reviews 83: 13 34. Starliper, C.E., and W.B. Schill. 2011. Flavobacterial diseases: columnaris disease, coldwater disease, and bacterial gill disease. Pages 606 631 in P. T.K. Woo and D.W. Bruno, editors. Fish diseases and disorders, volume 3: viral, bacterial, and fungal infections, 2 nd edition CAB International, London, UK.
111 Sudova, E., J. Machova, Z. Svobodova, and T. Vesely. 2007. Negative effects of malachite green and possibilities of its replacement in the treatment of fish eggs and fish: a review. Veterinarni Medicina 52: 527 539. Sw ingle, H.S. 1969. Methods of analysis for waters, organic matter, and pond bottom soils used in fisheries r esearch. Auburn University, Auburn, Alabama. Thomas Jinu, S. and A.E. Goodwin. 2004. Acute columnaris infection in channel catfish, Ictalurus punctatus (Rafinesque): efficacy of practical treatments for warmwater aquaculture ponds. Journal of Fish Diseases 27: 23 28 Torrans, E.L. 2005. Effect of oxygen management on culture performance of channel catfish in earthen ponds. North American Journal of Aquaculture 67: 275 288. Turner, C.J. 1988. Detection of stocking mortality in striped bass and hybrid striped bas s culture ponds. The Progressive Fish Culturist 50(2): 124 126. Wagner B.A., D.J. Wise, L.H. Khoo and J.S. Terhune. 2002. The epidemiology of bacterial diseases in food size channel catfish. Journal of Aquatic Animal Health 14: 263 272. Watson, C.A. 1999. The ornamental fish industry. Pages 87 93 in B.C. Paust and A.A. Rice, editors. Marketing and shipping live aquatic products, proceedings of the second international conference and exhibition. Fairbanks: Alaska Sea Grant College Program. Watso n, C.A. and J.V. Shireman. 1996. Production of ornamental aquarium fish. Fact sheet FA 35, Department of Fisheries and Aquatic Sciences, Florida Cooperative Extension Service, Institute of Food and Agricultural Science, University of Florida. Available : http://edis.ifas.ufl.edu/pdffiles/FA/FA03900.pdf (September 2012). Wedemeyer, G.A. 1996. Interactions with water quality condition s. Pages 60 110 in Physiology of fish in intensive culture s ystems. Chapman and Hall, New York, NY. White, A.J., J.F Schreer, and S.J. Cooke. 2008. Behavioral and physiological responses of the congeneric largemouth (Micropterus salmoides) and smallmouth bass (M. dolomieu) to various exercise and air exposure durations. Fisheries Research 89: 9 16. Yanong, R.P.E. 1 996. Reproductive management of freshwater ornamental fish. Seminars in Avian and Exotic Pet Medicine 5(4): 222 235.
112 Yanong, R.P.E. 2008. Use of hydrogen peroxide in finfish aquaculture. Fact Sheet 157, Department of Fisheries and Aquatic Sciences, Florida Cooperative Extension Service, Institute of Food and Agricultural Science, University of Florida. Available: http://edis.ifas.ufl.edu/pdffiles/FA/FA15700.pdf (August 2012).
113 BIOGRAPHICAL SKETCH Natalie Steckler was born and raised in Lexington, KY. Natali e always had a love for animals helped spark a particular interest in aquatic animals which quickly developed into a passion. After attending Bellarmine University where she graduated with a Bachelor of Science n moved to Florida to pursue a m es and Aquatic Sciences at the University of Florida. Natalie hopes to devote her career to the exciting field of fish health.