The behaviors of oyster larvae are difficult to monitor or experimentally manipulate, especially in field conditions. As a result, little is known of the fate of oysters in the larval portion of their life cycle, prior to recruitment. At the transition from pelagic larvae to benthic adults, larvae are likely to come into contact with many invertebrates resident on oyster reefs. Of these, fouling epifauna are generally believed to reduce the settlement of interspecific larvae through competitive exclusion and predation. Studies of these interactions, however, often utilize artificial settlement panels, which can exhibit different recruitment patterns to those observed on natural substrates. I therefore investigated the interactions between reef-associated fauna and settling oyster larvae on natural shell substrates. Over a series of laboratory microcosm studies, native (Crassostrea virginica) and non-native (Crassostrea ariakensis) larvae were exposed to reef-collected shells, each supporting a single species of reef-associated fauna. The presence of adult bryozoans (Membranipora tenuis) had little effect on either larval settlement rate or mortality. The boring sponge (Cliona sp.) significantly decreased oyster larval settlement, and generally increased oyster mortality. Barnacles (Balanus improvisus) typically facilitated settlement. Barnacle molds and empty barnacle tests, intended to mimic the surface area and rugosity of live barnacles, did not significantly affect settlement. However, in some trials, adult barnacle bathwater enhanced settlement of both oyster species, implicating the role of waterborne cues. Such bathwaters were found to cause oyster larval mortality, as were bathwaters created by adult clamworms or even adult oysters. Predation by clamworms (Neanthes succinea), which were found at very high densities on field-collected oyster shells, caused significant oyster larval mortality in these experiments. The combined roles of both positive and negative interactions between oyster larvae and reef fauna require enumeration under field conditions. The results from this study highlight the need for clarification of these roles in order to optimize shell supplementation restoration efforts, and to more thoroughly understand the settlement behaviors and mortality sources of recruiting oyster larvae

The phytoplankton of the Ross Sea have been intensively studied during the last decade, as opposed to the Amundsen Sea, where virtually nothing is known about phytoplankton taxonomy and distribution. Blooms in the Ross Sea are usually composed of diatoms and the prymnesiophyte Phaeocystis antarctica; diatoms are often dominant in strongly stratified waters during the summer, whereas P. antarctica usually dominates in less stratified waters in the south-central polynya during spring. This study focused on understanding the environmental variables that influence the spatial patterns of phytoplankton assemblages during late summer and early fall, 2007, and late spring and early summer, 2008 in the Amundsen and Ross Seas. Large differences in the distribution of phytoplankton were noted in space and time. In general, P. antarctica dominated the blooms in southwestern part of the Ross Sea. The silicoflagellate Dictyocha speculum was present in relatively high abundance and co-dominated with P. antarctica and diatoms in the eastern part of the Ross Sea during February. In the Amundsen Sea P. antarctica co-dominated with diatoms, and diatom blooms were more common than in the Ross Sea. Shallow mixed layer depths supported the growth of diatoms in the Ross and Amundsen Seas, but it was not the only factor required for diatom bloom development. Blooms dominated or codominated by diatoms were also more frequent in relatively fresher waters of the Amundsen Sea than in the Ross Sea in late summer and early fall. Modified Circumpolar Deep Water (MCDW), a water mass that can potentially be a source of Fe to phytoplankton, intruded the upper waters (from 80-120 m) near the Ross Ice Shelf. I hypothesized that this water mass intrusion may have favored P. antarctica blooms, releasing them from Fe limitation because they occurred in waters where the MCDW was shallow. However, pigment and quantum yield data show that P. antarctica blooms were approaching Fe stress in waters where the MCDW was observed, suggesting that intrusions of MCDW strengthen stratification and restrict Fe inputs into the surface. Because the Ross and the Amundsen Sea have a wide range of environmental and climatic conditions, understanding the factors that influence phytoplankton distribution in these areas will provide information of how phytoplankton respond to a changing climate.

The land-­‐ocean margin is characterized by a shallow water column, which allows light to reach the benthos and supports a diverse community of benthic autotrophs. One group of benthic autotrophs, consisting of benthic diatoms, cyanobacteria and other photosynthetic microorganisms living near the sediment surface (i.e., benthic microalgae) accounts for a substantial amount of this primary production. Benthic microalgae are also tightly coupled to carbon and nutrient cycling processes carried out by the sediment bacterial community. Benthic microalgae exude complex biopolymers, called extracellular polymeric substances (EPS), which consist mainly of carbohydrates, but can contain proteins and nucleic acids. EPS serves multiple functions for the benthic microalgae including: sediment attachment and stabilization and provides a labile substrate that may facilitate the tight coupling between benthic microalgae and sediment bacteria. A novel experimental apparatus, called the perfusionator, was used to examine carbon and nitrogen cycling through the benthic microalgal and sediment bacterial communities during a 48-­‐day field experiment. Dual stable isotopic tracers (H13CO3-­‐ and 15NH4+) were added to the porewater within the perfusionators to trace carbon and nitrogen cycling through inorganic and organic pools in order to assess the role of benthic microalgae in sediment nutrient cycling and sequestration. Physical and biological processes were characterized by: (1) installing covers over half of the perfusionators on day 14 to dampen physical mixing and sediment reworking, (2) using glass beads to track sediment mixing and reworking within the perfusionators, (3) continuous measurements of current speeds using an acoustic Doppler velocimeter and (4) monitoring a suite of environmental variables. The dominant feature of the experimental period was a harmful algal bloom (chl a concentrations peaked on day 14 at 107 μg L-­‐1) dominated by Cochlodinium polykrikoides, which reduced biomass of benthic microalgae. Sediment total organic carbon and nitrogen (TOC and TN), benthic microalgae isolated through Ludox extractions, and EPS were enriched in 13C and 15N during the labeling period. Two fractions of EPS were extracted from the sediments, a hot water (HW-­‐EPS) or intracellular fraction and a hot bicarbonate (HB-­‐EPS) or extracellular fraction. Concentrations of HB-­‐EPS were higher than HW-­‐ EPS, but the HW-­‐EPS showed enrichment in 13C before the HB-­‐EPS, consistent with its association with photosynthetic pathways inside the cell. The carbohydrate content of the EPS was higher later in the experiment when sediment chlorophyll concentrations were lower. The carbohydrate composition of the EPS indicated that a higher proportion of material was probably derived from sources other than benthic microalgae including bacterial sources. Increased concentrations of phaeopigments in surface sediments corroborated increased contributions from decaying or degraded material, likely derived from the harmful algal bloom. Overall, this field experiment traced carbon and nitrogen through the benthic microalgal community in a shallow coastal system, and captured the response of the system to a harmful algal bloom event. This experiment provided new insights about carbon and nitrogen dynamics within shallow systems and benthic microalgae community responses to key environmental events.

The relative strength of bottom-up and top-down processes operating within food webs is a fundamental determinant of community structure and function. In marine systems, inconspicuous but often highly abundant invertebrate herbivores (mesograzers) are implicated as strong consumers of primary production and important prey for higher-order consumers. Because of their small size, however, mesograzer abundance is not easily manipulated in the field, which limits our ability to adequately assess their grazing impacts. Seagrass systems present a pressing need for the study of food web dynamics because anthropogenic nutrient and sediment inputs decrease the amount of light reaching seagrass leaves, which limits the depth distribution of seagrasses via reduced photosynthesis to respiration ratios. Mesograzers benefit seagrass through their consumption of epiphytic algae and thus may mitigate the loss of seagrass beds due to nutrient enrichment. I test the relative impacts of nutrient enrichment and crustacean mesograzer abundance on epiphytes in a natural seagrass bed without using cages. My work presents the first cage-free tests of crustacean mesograzing impacts in natural seagrass communities. I successfully decreased crustacean abundance for extended periods of time in multiple experiments using a degradable chemical deterrent. Crustacean mesograzer reduction led to concomitant increases in epiphytic algal biomass, while nutrients increased epiphytes only in the absence of mesograzers. My results validate early work from mesocosm and field cage studies designed to test grazing impacts of mesograzers and support the hypothesis that mesograzers indirectly benefit seagrass through a positive indirect interaction.

The Mid-Atlantic Ridge (MAR) is the largest topographic feature in the Atlantic Ocean, yet little is known about the food web structure and trophic ecology of fishes inhabiting mesopelagic waters along the MAR. To better understand the food web structure and to compare the feeding ecology of abundant MAR fishes to those in offridge areas, MAR-ECO, a Mid-Atlantic Ridge ecosystem field project of the Census of Marine Life, sampled the pelagic fauna of the northern MAR (Iceland to the Azores) during June-July, 2004. Samples were collected with two midwater trawls outfitted with multiple opening and closing cod ends to sample discrete depths from the surface to >2300 m at predefined stations within four ridge sections. Fishes of the family Myctophidae were the most abundant fishes collected, with the myctophid Benthosema glaciale the numerically dominant species. The feeding ecology of three myctophid species, B. glaciale, Protomyctophum arcticum, and Hygophum hygomii, were quantified through dietary analysis. Each species was predominantly zooplanktivorous, consuming copepods, euphausiids, ostracods, and other crustacean zooplankton, with each myctophid species exhibiting unique dietary patterns. Different abiotic parameters influenced variability in the diet of B. glaciale and P. arcticum as determined by canonical correspondence analysis. Of the measured parameters, ridge section and time of day were significant explanatory variables in the diet of B. glaciale, while only depth was significant in the diet of P. arcticum. Daily consumption by B. glaciale and P. arcticum was highest at the Reykjanes Ridge and Charlie-Gibbs Fracture Zone, respectively; consumption by H. hygomii was only measured at the Azorean Zone. Daily consumption of all three species was less than 1% of dry body weight. Active transport of carbon out of the euphotic zone by diel vertically migrating MAR myctophids through respiration of CO2, excretion of dissolved organic carbon, and egestion of particulate organic carbon (POC) was also estimated, and compared to passive sinking of POC in the North Atlantic. Active carbon flux by the 0-200 m integrated migrant myctophid biomass, uncorrected for trawl efficiency, ranged from 0.01-0.4% of sinking POC flux at 150 m and 0.02-0.95% at 300 m. If myctophid biomass was corrected for low gear efficiency, flux increased to 0.1-1% and 0.3-3.5%, respectively, of sinking POC. Lower MAR myctophid biomass resulted in lower active carbon transport compared to myctophids in the Pacific and zooplankton in the North Atlantic estimated in previous studies, but myctophid active transport should still be considered in models of MAR carbon cycling. Due to the role of myctophids as both predators of numerous zooplankton taxa and as prey of higher taxa, information on the feeding ecology and daily consumption of these fishes is necessary to accurately assess their role in the MAR food web and the overall trophic structure of this hydrodynamically and topographically unique ecosystem.

The northern Mid-Atlantic Ridge, from Iceland to the Azores (MAR), ranges in depth from 800 – 4500 m and extends over an area of 3.7 million km2. Recent evidence from MAR-ECO, a Census of Marine Life field project, reported increased abundance and biomass of deep-pelagic fishes below 1000 m on the MAR. Among the fishes sampled, Bathylagus euryops was found to be the biomass-dominant species and ranked third in total abundance. In this thesis, we characterize the distribution and feeding ecology of B. euryops as a function of physical, biological, and life history parameters along a mid-ocean ridge system. Multiple biologically plausible general linear models were fitted to B. euryops catch-per-unit-effort (CPUE) data to investigate the role of various combinations of explanatory variables on the distribution of this species. Results indicated that a model containing categorical depth and geographic location variables provided the most parsimonious description of B. euryops CPUE data. Vertical migration analyses were also conducted to investigate the vertical distribution of B. euryops along the MAR and results indicated that time of day had little influence, whereas ontogeny likely influenced the vertical distribution of B. euryops. To describe the feeding ecology of B. euryops, a general diet composition was determined. Multivariate analyses, including a cluster analysis and a canonical correspondence analysis, were utilized to investigate factors that cause variability within the diet of B. euryops. Results revealed that fish size and geographic location significantly influenced the diet of B. euryops. Furthermore, daily rations were estimated to better understand the role of B. euryops in the food webs of the North Atlantic and results were consistently estimated to be less than 1% of the average wet weight along the MAR. The general trend observed was a southward increase in daily ration estimates along the MAR.

Eight species of Prionotus, collected from 5-100 m in the Gulf of Mexico along a cross-shelf transect off Freeport, Tx during October 1977-August 1981, were studied to determine life history patterns and how they partition resources. Only four species were abundant: P. tribulus, P. paralatus, P. longispinosus, and P. stearnsi. Prionotus spp. mature at 80 mm (P. stearnsi and P. rubio), 85 mm (P. tribulus and P. paralatus), 100 mm (P. ophryas), 105 mm (P. roseus), and 120 mm (P. longispinosus). Sizes at Age 1 were 99-140 mm (P. tribulus), 99-138 mm (P. longispinosus), 98-122 mm (P. paralatus), 75-125 mm (P. stearnsi), and 95-129 mm (P. ophryas). Most fish were less than 200 mm. Typical maximum sizes were 155-175 mm (P. tribulus), 145-160 mm (P. longispinosus), 160-165 mm (P. paralatus), 180-195 mm (P. rubio), 125 mm (P. stearnsi), and 155-160 mm (P. ophryas). Fish were typically 1-3 years old at these typical maximum sizes, and most were Age 1 and less. Apparent total annual mortality rates were 80-100%. The eight species fit into three categories based on their bathymetric distributions and community memberships: (1) members of the inshore white shrimp community which occurs from 5 m to about 16 m (P. tribulus); (2) members of the offshore brown shrimp community which occurs from about 36 to at least 100 m (P. paralatus, P. stearnsi, and P. roseus); and (3) members of a transition fauna which occurs between the outer edge of the white shrimp community and the inshore edge of the brown shrimp community from about 18-27 m (P. longispinosus, P. rubio, P. ophryas, and P. scitulus). Comparisons among species indicate a common pattern of population dynamics that favors r-strategy (short life spans, high mortality rates, and rapid turnover of biomass), but with temporally segregated spawning and spatially segregated distributions. Spawning grounds seem to lie toward the northcentral Gulf with current transport mechanisms carrying young toward the northwestern Gulf. Prionotus spp. from the northwestern Gulf show quite different life history attributes than their congeners from mid Atlantic and New England regions.

The disease caused by the protozoan Perkinsus marinus has been a major source of mortality in the eastern oyster, Crassostrea virginica. Variations in susceptibility to P. marinus infection among eastern oysters collected from the Chesapeake Bay and Gulf of Mexico, as well as between eastern and Pacific (Crassostrea gigas) oysters were determined. Since oyster host defense may play a role in determining susceptibility to pathogen infection, cellular and humoral defense activities of the oyster and their interactions with P. marinus were investigated. Procedures also had to be established to isolate, purify, and propagate in vitro, P. marinus. Eastern oysters from all sites were found to be highly susceptible to the pathogen. Cellular and humoral activities were significantly affected by heavy intensity of P. marinus infection. Prevalence and intensity of P. marinus infection were lower in Pacific oysters than in eastern oysters. Pacific oysters may offer a less favorable environment for the development of P. marinus compared to eastern oysters for at least two possible reasons: the elevated cellular and humoral activities may degrade the parasite more effectively, and lower plasma protein levels may limit parasite growth. Incubation of merozoites with hemocytes of eastern and Pacific oysters in vitro suggested that limited intracellular killing of P. marinus occurred but that killing was not mediated by oxygen metabolites. Perkinsus marinus was successfully propagated in vitro in a culture medium containing most of the known constituents of cell-free oyster hemolymph. Cultures of the parasite were initiated from heart fragments of infected oysters. The cultured protozoan was similar in morphology to P. marinus, enlarged in fluid thioglycollate medium, reacted with polyclonal antibodies raised against hypnospores and was infective. Continuous cultures of P. marinus could also be initiated from hypnospores. Two types of division, progressive cleavage and successive bipartition of the mother cell protoplast, were observed.

Hypoxia, low dissolved oxygen, is an important environmental stressor in estuarine systems. In this dissertation, I examine the effects of hypoxia on the macrobenthic communities of the York and Rappahannock Rivers, Chesapeake Bay, USA, and in particular its effects on the infaunal clam Macoma balthica. A survey of the macrobenthic community was performed using box-coring before and after hypoxia in 2003 and 2004 in both rivers. Hypoxia was associated with a change in the macrobenthic community towards smaller, shorter-lived, opportunistic species; and a substantial decrease in biomass. M. balthica recruited into all areas of the river but suffered local extinction in hypoxic areas, demonstrating that these areas represent sink habitats. I developed an enzyme-linked immunosorbent assay to quantify fecundity in M. balthica and used it determine the effect of hypoxia on clam fecundity. In laboratory experiments performed in 2005 and 2006, M. balthica migrated toward the sediment surface and decreased egg production in response to hypoxia. In field caging experiments, performed during the summers of 2005 and 2006, episodic hypoxia caused a three-fold increase in the rate of predation on M. balthica , suggesting that the behavioral responses of M. balthica to hypoxia make it more vulnerable to predation. These results further suggested that hypoxia may change the functional response of epibenthic predators to M. balthica from a stabilizing type III to a destabilizing type I or II. Using the results of the previous experiments, I constructed a density-independent model of the M. balthica population, which predicted that increasing the spatial extent and duration of hypoxia could cause the population to decline toward extinction. A second model, which incorporated density-dependent predation, predicted that, under mild hypoxic conditions, trophic transfer of biomass from M. balthica to predators could be enhanced, but that increasing the severity of hypoxia would decrease trophic transfer. The model further predicted that increasing hypoxia would decrease the resilience of the M. balthica population to disturbance, making functional extinction of the population more likely. This body of work underscores the negative effects of hypoxia on the levels of the individual, the population, and the ecosystem.

Continental shelves are generally believed to play a critical role in ocean biogeochemical cycling, however this has raised the question as to the relative importance of various nitrogen flux terms such as denitrification, burial, net community production and advective fluxes. Quantifying these fluxes on an annual area-integrated basis using traditional observational means is often difficult, due to the fact that these fluxes rapidly change on relatively small spatial scales, making inadequate data resolution a significant problem. Satellite remote sensing data and numerical modeling provide alternative ways to fill the data gaps, and hence have the potential to generate quantitative estimates of these various biogeochemical fluxes. However, they both suffer from distinct shortcomings, e.g., satellite data are only limited to the surface whereas numerical modeling can be pointless without rigorous skill assessment. Thus caution is warranted when using these tools to generate quantitative estimates of biogeochemical fluxes. The two were combined in this dissertation project by assimilating the satellite-derived data into the models, selecting the optimal ecosystem model, as well as evaluating the model before using the model simulations to explore the nitrogen fluxes on the Mid-Atlantic Bight (MAE). First, multiple satellite-derived data products were assimilated into a one-dimensional assimilative model framework to determine the relative advantages of assimilating different satellite data types. The variational adjoint method, a parameter optimization method, was applied to a series of experiments assimilating synthetic and actual satellite-derived data, including total chlorophyll, size-fractionated chlorophyll and particulate organic carbon (POC). The experiments revealed the importance of assimilating data from multiple sites simultaneously as the optimal parameter sets produced by assimilating data at individual sites were often unrealistically over-tuned and deteriorated model skill at times and depths when data were not available for assimilation. The model-data misfits from the experiments also demonstrated that optimal results were obtained when satellite-derived size-differentiated chlorophyll and POC were both assimilated simultaneously. These two types of satellite data were then assimilated simultaneously to rigorously evaluate how food web model complexity affects the ability of a lower trophic level model to reproduce observed patterns in satellite-derived data. This was again implemented in the one-dimensional model framework to minimize the computational costs. Five ecosystem model variants with various levels of complexity in the phytoplankton (P) and zooplankton (Z) structures were examined by assimilating satellite-derived size-differentiated chlorophyll and POC data at four MAE continental shelf sites, and testing the optimal parameter values at five independent sites in a cross-validation experiment. Although all five models showed improvements in model skill after the assimilation, the moderately complex 2P2Z model best reproduced the surface fields throughout the MAE. Additional experiments were conducted in which random noise was added to the satellite data prior to assimilation. Whereas the most complex model was sensitive to the random noise added to the data, the simpler models successfully reproduced nearly identical optimal parameters regardless of whether or not noise was added to the assimilated data, highlighting that random noise inherent in data into these simple models. The moderately complex 2P2Z ecosystem model was thus coupled with a three-dimensional circulation model and forced by a dynamic land ecosystem/watershed model to simulate the biogeochemical cycling on the MAB shelf and to quantitatively assess key components of the annual area-integrated nitrogen budget from 2004-2007. The simulation indicated that over these four years similar amounts of nitrogen were removed by denitrification and burial (∼0.1 Tg N y-1). Net community production was larger and varied more between the four years (∼0.2 to 0.3 Tg N y-1), but overall was positive, indicating that the MAB was net autotrophic. The advective fluxes of nitrogen into and out of the MAB were dramatically different between the four years investigated (by about ∼.26 Tg N y-1), presumably as a result of changes in the positions of the Gulf Stream and Labrador Sea waters. The accumulative effects of these fluxes resulted in a near zero net rate of change in total nitrogen, indicating the MAB remained unchanged in the amount of total nitrogen in the water column over these the four years. Sensitivity tests varying the initial conditions and simplifying the modeled plankton structure showed distinct impacts on these nitrogen fluxes: the former strongly affected the advective fluxes, but had little impact on denitrification, burial or NCP, whereas the latter significantly reduced denitrification, burial, and NCP but did not significantly impact the advective fluxes. Overall the strong seasonality and interannual variability in the nitrogen fluxes highlight the importance of data coverage throughout all seasons and multiple years in order to accurately resolve the current status and future changes of the MAB nitrogen budget.