The cardiovascular tissues of 17 rough toothed dolphins (Steno bredanensis) that mass stranded in the Florida panhandle during 1997, 1998 and 2001 were examined. Routine hematoxylin-eosin staining demonstrated the presence of extensive lesions in the cardiovascular systems of these stranded dolphins, such as myocarditis, arteritis, arteriosclerosis, cardiomyopathy and fibrosis. Histopathologic techniques were used to characterize these lesions, i.e. to locate the pathologic changes; to identify the cells and extracellular fibers involved; and to grade them according to their degree of severity. Masson trichrome, Movat pentachrome, Verhoeff elastic, and phosphotungstic acid hematoxylin stains were used to differentiate cells, collagen, elastin, and ground substance. Lesions were present in all animals, regardless of age or sex. The most common findings in the heart were fibrosis in 17 animals, myocarditis in 14, cardiomyopathy in 13 and fibroelastosis in nine animals. The aortas showed evidence of subintimal thickening in seven animals, inflammation in five, and thickening and degeneration of elastic fibers in three animals. Coronary arteries were affected by arteriosclerosis in seven animals, arteritis in seven, subintimal thickening in five, and muscle cell atrophy in four animals. The severity of the lesions was related to the presence of multiple pathological changes in the same animal. In females, the severity of the changes varied with age, with adult females demonstrating moderate to severe changes, while juvenile females had mild to moderate changes. In general, 25% (2/8) of the adult females had mild changes, 62.5% (5/8) had moderate changes, and 12.5% (1/8) had severe changes, while 67 % (2/3) of the juvenile females had mild changes and 33% (1/3) had moderate changes. In contrast, 50% (2/4; 1/2) of the adult and juvenile males had moderate changes and 50% (2/4; 1/2) of the adult and juvenile males had severe changes. Although the causes of these changes are unknown, they likely contributed to the cause of these mass strandings.

CHAPTER 2 The burgeoning shark-fishing industry has caused severe declines in many shark populations, prompting an urgent need for conservation and management of sharks on a species-specific basis. Many fisheries are managed on the basis of species groups due to difficulty in identifying the morphologically similar requiem sharks (Carcharhinidae) and their body parts. One diagnostic trait readily discernible in landed shark carcasses is the presence or absence of an interdorsal ridge, which allows shark carcasses to be categorized as either ridgeback or non-ridgeback sharks and narrows the identification process. Further identification to species, however, remains problematic and is a major impediment to management and conservation on a species-specific basis. To facilitate the goal of species-specific management, I have developed a rapid and cost-efficient DNA-based method for use in species identification. It employs a high-density, multiplex polymerase chain reaction (PCR) assay that combines seven species-specific primers and two universal primers to discriminate one ridgeback (Galeocerdo cuvier) and six non-ridgeback (Carcharhinus brevipinna, C. limbatus, C. acronotus, C. isodon, C. leucas, and Negaprion brevirostris) sharks common in U.S. Atlantic fisheries. The primers were designed based on species-specific nucleotide differences in the nuclear ITS2 locus. These seven primers were tested on 73 non-target species worldwide to assess their diagnostic utility on a global scale. Five of the seven primers are species-specific, with only the C. leucas and the C. limbatus primers each amplifying one other congener, C. perezi and C. tilstoni, respectively. This diagnostic assay was tested in a practical management context and successfully identified shark fins confiscated by NMFS law enforcement agents, indicating that it will be useful in management of the U.S. Atlantic shark fishery. Furthermore, the primers developed for G. cuvier, C. leucas and C. brevipinna successfully identified their respective target species from the Atlantic and Pacific Oceans, suggesting that they will also be useful for global applications such as monitoring international trade in shark products. CHAPTER 3 Accurate species delineation is essential to sound management and conservation of fishery resources. The species status of two butterfly rays, Gymnura marmorata and G. crebripunctata, occurring in the fisheries in the Gulf of California and on the Pacific Mexican coast, has long been debated. Direct sequence comparison and a phylogenetic analysis were carried out using a segment of the mitochondrial cytochrome b locus to elucidate the relationship between G. marmorata and G. crebripunctata and five other congeners. Sequence divergence between G. marmorata and G. crebripunctata was minimal compared to sequence divergence among the five other established congeners. A neighbor-joining analysis showed strong statistical (100%) support and a maximum likelihood analysis showed reasonable support (62%) for the monophyletic clade consisting. of G. marmorata and G. crebripunctata. Importantly, neither analysis supported the reciprocal monophyly of either species, as would be expected for true separate species. These data suggest that the current classification of G. marmorata and G. crebripunctata as two separate species needs to be revised to support their designation as a single species. Genetic analysis of the cyt b locus also revealed three single-nucleotide synapomorphies that partitioned animals from the two geographic sites sampled in Mexico (Sahuimaro, Sonora and Bahia Almejas, Baja California Sur), indicating the existence of possible population genetic structure between the two sites. By analyzing cyt b sequences from seven ofthe 10 described Gymnura species, I also provide the first phylogenetic hypothesis for interspecies relationships in the genus Gymnura.

The effects of a seven-day exposure to a 2°C increase above the maximum summer temperature and an 86% reduction in subsurface irradiance on the net photosynthetic output and mortality of the scleractinian coral species Montastrea cavernosa were examined in vitro as both single- and simultaneous multiple-stressor events. The effect on photosynthetic output was expressed as the percent change in the Pg:R24 ratio following the seven-day treatment period. Mortality was defined as the absence of viable coral tissue at the end of the seven-day recovery period. The level of elevated temperature (31.5°C) was based on partial datasets of in situ measurements of subsurface temperature, spanning the years 2000-2002, recorded on the crest of the outer reef tract of Ft. Lauderdale, FL at a depth of 15m. The degree of irradiance attenuation was equivalent to an increase in turbidity from the ambient value of 0.3 NTU at the average specimen collection depth of 21.3 m to a value 10 NTU, induced by an increase in total suspended solids from 7.6 to 57.0 mg L-1. The individual stressors of elevated temperature and reduced irradiance both produced significant, sub-lethal depressions of the Pg:R24 ratio of M. cavernosa colonies, but a significant interaction between the two stressors was not detected. However, simultaneous exposure to the two stressors did result in significantly greater post-recovery mortality, evidence of an exacerbative interaction between elevated temperature and reduced irradiance for the relatively deep dwelling colonies of M. cavernosa in this study. The finding of a significant exacerbative interaction between elevated temperature and reduced irradiance on coral mortality illustrates the importance of considering multiple-stressors when assessing the effects of anthropogenic and natural disturbances on the current and future health of coral reefs. Pertinent to local issues of coastal zone management, the finding of a significant effect of attenuated radiance on the Pg:R24 ratio and a potentially lethal interaction between attenuated irradiance and elevated temperature indicates the current Florida regulatory limit of 29 NTU above background may not adequately protect scleractinians occupying the deeper reef zones.

The processes structuring coral communities involve many factors including recruitment dynamics, competitive interactions, and acute and chronic disturbances. The shallow reef at Sunabe (Okinawa, Ryukyu Islands, Japan) was dominated by a dense cover of mainly Acropora spp. until the bleaching event in 1998, when most colonies died. Subsequent monitoring of the community showed that the number of juvenile colonies was high in 2000 and competition among these colonies may affect community dynamics. The aim of this study was to determine the possibility of predicting community change and recovery, using short-term observation data of size-specific mortality and growth rates of coral recruits. Predictions were made using a demographic model based on a Leslie-matrix approach with data collected on the shallow reefs of Sunabe in 2000. This existing data showed a tendency of high mortality of competing colonies, but with no significant difference among species. Over the period of 2000-2004 percent cover of Acropora sp. increased from 16% to 22% however the number of colonies decreased (from 6.5 per quadrat to 2.5 per quadrat). The prediction of size fraction derived from the matrix population model and the observed size fraction based on the 2004 data showed some agreement; however the model predicted a high number of colonies in the largest size class. There were no adult colonies in 2000. A known bleaching event in 2002, causing mortality of reproducing adult colonies (high mortality and low growth rates) is most probably the reason for the disagreement between is the matrix model-based prediction and the observed community structure. Therefore, it is suggested that one more size class is add to the model, representing reproducing colonies. It is also suggested that further improvement of the model is needed for disturbances (such as bleaching, typhoon, and mass predation events). Despite these suggested improvements, the model presented here shows that it is possible to use simple community data taken over the course of 1 year to predict short-term community size class dynamics and population growth on juvenile Acropora spp. dominated reefs. Thus, such investigations have practical value for the management of coral resources.

Coral reef degradation from ship groundings and construction activities is unfortunately becoming an increasingly common occurrence worldwide, especially in densely populated regions like southeast Florida, U.S.A., where each year there are a number of coral reef injury events that require mitigation and/or restoration. These activities generally begin with the rescue and reattachment (triage) of scleractinian corals dislodged during the injury event. Because dislodged colony mortality is typically high, colonies may die during legal enforcement action delays. Additionally, natural reef recovery in southeast Florida is typically slow; therefore, transplantation of additional (donor) scleractinian corals into an injured area has been used to restore natural species richness, percent cover, and density, which can accelerate reef recovery. Donor colonies available for transplantation have been grown in situ, grown in laboratories, and taken from uninjured reef areas. An alternative source of donor coral colonies for transplantation is “corals of opportunity,” which I define as scleractinian corals that have been detached from the reef through natural processes or unknown events. My thesis was part of a project, the Coral Nursery Project, initiated in 2001 in Broward County, Florida, U.S.A. that was developed to collect these dislodged colonies and transplant them to a coral nursery. Coral nurseries are interim locations that function as storage sites for corals of opportunity where they can be cached, stabilized, and allowed to grow until needed as donor colonies for future restoration activities. My thesis utilized a partnership between a local university, county government, and a volunteer dive group. Two hundred fifty-three corals of opportunity were located, collected, and transported to coral nurseries. Corals of opportunity were tagged, transplanted to the coral nurseries, and monitored quarterly for survivorship and growth from date of transplantation to February 2004. Survival of corals of opportunity transplanted to coral nurseries was found to be statistically similar to that of control corals naturally attached to reef, and significantly greater than that of control corals of opportunity left unattached. Growth rates of control corals naturally attached to reef were similar to that of corals of opportunity transplanted to coral nurseries. Results provide resource managers with information on the utility of using corals of opportunity as a source of transplant donor colonies, on coral species- and colony size-specific transplantation success, and on transplanted colony survival and growth. I discuss the value of using coral nurseries as cache sites for corals of opportunity to be used in future coral reef restoration activities, and provide recommendations for coral nursery characteristics for optimum coral survival, developing community-based restoration projects, and fostering participatory management. However, the ultimate goal of the Coral Nursery Project is to use these rescued corals of opportunity from this nursery as a source of donor coral transplants in future coral reef restoration activities. Stabilized and healthy transplanted corals of opportunity from this coral nursery will provide a living resource for coral transplantation during coral reef restoration projects. Corals of opportunity, coral nurseries, and integrated stakeholder involvement may become important tools in creating proactive coral reef management programs.

A beach renourishment project was initiated in May 2005 and completed in February 2006 to restore 11.1 km of shoreline in Broward County, Florida, USA. For mitigation of predicted nearshore hardbottom burial, a boulder reef totaling 3.6 ha was deployed in 2003. To examine the replacement value of the mitigation relative to fishes, this study compared fish assemblages on boulder reef to those on adjacent natural hardbottom. Twenty-five natural hardbottom sites and twenty-five boulder reef sites were surveyed six times between March 2005 and August 2007. Two non-destructive visual census methods, a transect count (30 m long x 2 m wide x 1 m high) and a 20 minute rover diver count (approximately 30 m x 30 m), were conducted at each site to assess abundance and species richness. On transect counts, 7,117 fishes of 96 species were counted on natural hardbottom, while 11,769 fishes of 119 species were counted on boulder reef. Across both survey types, a total of 271 species was recorded. Significant differences among reef fish assemblages were found in both abundance and species richness (pHaemulon spp., while boulder reef showed higher densities of early juvenile (2-5 cm TL) Haemulon spp. Boulder reef also had a higher abundance of fishes greater than 5 cm and piscivorous fishes in general. While boulder reef may provide a suitable habitat for many fishes, it does not mimic natural hardbottom-associated fish assemblages, nor does it provide a similar nursery habitat for juvenile fishes.

Coral reefs in Broward County are increasingly threatened by coastal and marine construction activities. Infrastructure installation (cables, pipelines, and sewer outfalls), beach renourishment, and dredge and fill operations have both anticipated and inadvertent impacts which may result in reef damage. Mitigation practices consisting of impact avoidance and minimization are necessary to prevent and offset reef damage. Coral transplantation is a commonly utilized impact minimization tool which aims to protect corals threatened by impending construction projects. Coral colonies are removed from their natural habitats and relocated to analogous unthreatened sites prior to construction. This study focuses on coral transplantation that occurred as impact minimization for two recently completed coastal and marine construction projects in Broward County. The first project involved relocation of colonies to natural reef away from potential damage by a natural gas pipeline geotechnical survey. The second project involved relocation of colonies to an artificial reef in relation to a beach renourishment project. Colonies from both projects were monitored and analyzed in regards to survival and growth to determine transplantation success. A combined 1100 colonies were transplanted for these projects, with approximately 25% of these chosen for 18-24 months of monitoring. The projects had high attachment rates between 92-100% and survival rates between 87-99%. A majority of colonies exhibited positive growth at an average rate of 0.03%-0.04% of initial tissue area per day. There were no apparent detrimental effects of transplantation such as reduced initial growth rates, disease, bleaching, or increased partial mortality. These results indicate that coral transplantation was highly successful from a biological standpoint. This study also demonstrates that coral transplantation is a valid and effective component of impact minimization whereby colonies that may otherwise be destroyed, can be successfully preserved. For this study, coral transplantation proved to be an economically worthwhile method of conserving an ecologically and economically valuable reef resource. However, coral transplantation is only one facet of impact minimization and mitigation. Ideally, future mitigation projects should also aim to reduce habitat loss and preserve biota in addition to stony corals.

The Dania Beach Erojacks Artificial Reef was deployed off the coast of Broward County, Florida on December 31, 1967 as a way to help combat beach erosion. Over the last forty years, the linear pile of concrete hexapods has become an important habitat, for scleractinian corals, octocorals, algae, crustaceans, mollusks, and fish. This study focuses on the density and size of the scleractinian corals found on this artificial reef and how it compares to that of the nearby natural reef. In addition, the impact of two hurricanes on the shallow portion of the artificial reef was documented. In the 2-year study, results indicate that there was higher coral cover on the natural reef (6.45%) compared to 4.27% cover on the artificial reef. Most of the colonies on the natural reef are larger than those found on the artificial reef; 65.6% vs. 29.7% were greater than 25cm2. When comparing colony numbers, there are more than three times as many on the artificial reef (3870) compared to the same area of natural reef (1133). This corresponds to colony densities of 5.0/m2 on the artificial reef, compared to 1.5/m2 on the natural reef. The passage of two hurricanes in close proximity to the artificial reef resulted in no decrease in the number and surface area of corals when pre- and post-hurricane values were compared. Surprisingly, there were significant increases in both coral abundance (GEE Analysis, p < 0.0001) and cover (GEE Analysis, p =0.0001), however these changes were attributed to improved proficiency of the researchers in finding corals rather than actual increases in these values.

Chapter 1 Effective spatial management of coral reefs including design of marine protected areas requires an understanding of inter-population genetic connectivity. We report a comparative analysis of gene flow among three commensal invertebrates occupying the same host sponge (Callyspongia vaginalis) but displaying contrasting reproductive dispersal strategies: the broadcast spawning brittle star Ophiothrix lineata and two brooding amphipods Leucothoe kensleyi and Leucothoe ashleyae. Sequence variation in the mitochondrial COI gene was used to infer connectivity along 355km of the Florida reef system and between Florida and Belize. O. lineata showed significant genetic structuring between Florida and Belize, and a pattern of isolation by distance but no significant genetic structuring along the Florida coastline. Both amphipods despite direct development also showed high gene flow along the Florida coastline. A nested clade analysis on the L. kensleyi data identified two major processes operating at different temporal scales. Inference of restricted gene flow due to isolation by distance dominated younger clades, whereas the inference for older clades was long distance dispersal from northern populations south into the Florida Keys. Multiple inferences of long distance dispersal support the hypothesis that rafting in detached sponges is the likely mechanism responsible for the high levels of overall gene flow detected by AMOVA along the Florida reef system. However, rafting appears to be less likely across deep open water, as connectivity between Florida and Belize (1072km) is highly restricted. Our results show that making assumptions about connectivity based simply on reproductive strategy may be misleading, and highlight the need for examining coral reef connectivity across diverse taxa. Chapter 2 Although over half of all known elasmobranchs are batoids, with many species exploited and several of conservation concern, little is known of their population genetic structure and intraspecific evolutionary history. Here I investigate gene flow among geographic populations of the southern stingray (Dasyatis americana), a large, recreationally important, demersal batoid that ranges from New Jersey, USA and the northern Gulf of Mexico to southern Brazil. Sequence variation in 648 bp of the mitochondrial control region was used to infer patterns of gene flow among North American and Caribbean populations. Analysis of 267 individuals from eight regions identified 67 distinct haplotypes and AMOVA results revealed high levels of population structure (overall φST = 0.50, P < 0.001) highlighting that populations need to be managed as distinct evolutionary units. The highly structured nature of D. americana populations was clearly evident in a nested clade analysis (NCA), which detected restricted gene flow at three levels of nesting. In addition, numerous range expansions, dispersals, and colonizations were detected and their path through the Caribbean identified a major phylogeographic break at the Florida Channel.

Endocommensal leucothoid amphipods are ecologically important in marine ecosystems worldwide. However, our understanding of the evolution and taxonomy of the group is confused. Although leucothoids typically occupy specialized niches, Leucothoe spinicarpa, the type species for the genus Leucothoe, has been reported from a wide variety of habitats. This study compared the taxonomy, ecology, and functional morphology of Caribbean commensal amphipods previously attributed to L. spinicarpa and incorporated new host and ecological information and new diagnostic characters. As a result 11 new leucothoid species are described from Southeast Florida, the Florida Keys, and the Western Caribbean Sea.