Particulate levels within marine, estuarine and freshwater vegetated shallows are often lower than in nearby open water, although most of the studies quantifying this trend are from non-tidal, freshwater systems. The potential positive feedbacks between vegetation, water clarity, and zooplankton clearance and the potential negative feedback from microbially-mediated sulfide production were investigated in several eelgrass (Zostera marina) beds in the lower York River and Mobjack Bay, Chesapeake Bay, Virginia and through the construction of a computer simulation model. Paired automated chlorophyll a and turbidity sensors were deployed for eight one-week periods to compare particulate levels inside and outside of eelgrass beds. The vegetated estuarine shallows monitored appeared to behave differently than those in freshwater vegetated systems, in that they were not able to consistently maintain improved water clarity relative to adjacent, unvegetated areas. Predictive equations for particulate levels inside the eelgrass beds were developed by regressing chl a and turbidity against wind and tidal influences for use in a Zostera simulation model. Zooplankton were sampled during two summer seasons to quantify their impact on water clarity. In 2006, zooplankton densities were significantly higher in vegetated than non-vegetated areas, but not in 2007. Zooplankton densities were significantly higher at night, both inside and outside of the vegetated beds. Overall, the zooplankton densities encountered within the SAV beds had the potential to filter approximately 2-6% of the water column per day, much less than typically encountered in freshwater. Eelgrass density, sediment organic content and porewater sulfide levels were quantified in situ in several SAV beds throughout spring and summer. There was no significant difference in [S] between vegetated and unvegetated areas, [S] was not correlated with eelgrass cover or sediment organic levels, but field results demonstrated that porewater [S] above 900-1000 muM inhibited eelgrass growth within the study area. An iron enrichment experiment demonstrated some potential for iron to positively affect Z. marina growth and survival, but responses were site specific and highly variable. Finally, a computer simulation model was constructed that incorporated positive and negative effects within Z. marina beds, including tidal- and wind-induced particulate loading, resulting attenuation of light, particulate removal due to biological and physical filtration, temperature stress and sulfide toxicity. Modeled Z. marina responded to reduced light with approximately proportional reductions in year-end shoot and root/rhizome biomass. The model was less sensitive to increased sulfides; increases of 1.5, 2.0 and 2.5 times background sulfide levels resulted in incremental reductions of year-end shoot biomass by 20-25% and root/rhizome biomass by 15-20%. The model was most sensitive to temperature; a 1??C increase reduced year-end shoot and root/rhizome biomass by 41%; sulfide and temperature stress combined reduced shoot and root/rhizome biomass by 64%. With eelgrass in the Chesapeake Bay growing near its southern limits, model results indicate that either sulfide or temperature stress may limit restoration efforts and induce continued losses of eelgrass. Internal feedbacks reduce some of the stress caused by light limitation, but do not compensate for a 1??C increase in temperature or increases in sulfide levels.

In estuarine crabs, there is a gradient of larval dispersal from retention within hatching habitats to export to the continental shelf. This dissertation examined the influence of physical transport and nutritional stress on recruitment success in species with different larval dispersal. In the first chapter, I examined larval vertical migration. Larvae from different families exhibited different tidally and light driven behaviors. The role of transport in selecting behaviors is associated with constraints on larval survival within the estuary. The transport of larvae was examined using a hydrodynamic model. This study compared the dispersal of vertically migrating larvae to that of non-migrating particles. Tidal migration and light limited behaviors influenced advective dispersal. Tidal shear fronts influenced larval distribution and likely effect the strength of biotic interactions. In chapter 3, I examined the impact of starvation and food quality on survival and biochemical composition of larvae with different dispersal strategies. Larvae that develop on the continental shelf generally had greater resistance to starvation than those that develop in the estuary, and they showed more efficient utilization of essential biochemical constituents during nutritional stress. Phospholipid metabolism was important in the starvation response. The biochemical composition of larvae in the field was examined to assess variability in nutritional condition over tidal time scales. Larvae in the plankton were larger and had higher contents of energetic lipids than those in laboratory studies. In the natural habitat, nutritional stress is rare and feeding conditions promote the accumulation of energetic reserves. There was a large degree of short term variability in phospholipid composition of larvae associated with small scale physical factors driving contact between zoeae and their prey. The results from this dissertation highlight the importance of small scale processes in the recruitment success of crab zoeae. While large scale gradients play an important role in selecting for dispersal, larval vertical migration behaviors reflect the influence of both small and large scale selective pressures. Tidal fronts with short durations play an important role in driving the distribution of larvae and likely influence larval feeding. Small-scale physical processes strongly influence variability in recruitment success of crab zoeae.

Habitat fragmentation leads to small, isolated habitat patches in which ecological processes may differ substantially from those in larger, continuous habitats. Seagrass is a structurally complex but fragmented subtidal habitat that serves as a refuge from predation for juveniles of the blue crab, Callinectes sapidus Rathbun. I compared the effects of eelgrass ( Zostera marina L.) patch size and shoot density on juvenile blue crab survival both before (June) and after (September) shoot defoliation and cownose ray disturbance changed eelgrass habitat, and used artificial seagrass to determine the influence of eelgrass patch size on juvenile blue crab survival in the absence of covarying shoot density. Under natural conditions, eelgrass patch size, eelgrass shoot density and density-dependent cannibalism all influenced juvenile blue crab survival, but crab survival varied temporally. Crab survival was inversely correlated with eelgrass patch size in the absence of covarying shoot density, in contrast to patterns typically observed in fragmented terrestrial landscapes; this was likely due to low predator abundance in small patches. I tested the hypothesis that crab survival is maximized at an intermediate level of seagrass fragmentation due to the inverse relationship between crab survival and patch size by modeling the joint effects of patch size and proportional cover on juvenile blue crab survival. When I assumed predation on crabs to be independent of crab density, maximal crab survival (ca. 34%) occurred at intermediate values of seagrass fragmentation. Approximately 18% of crabs survived irrespective of the proportion of the landscape covered by seagrass when crab survival was assumed to be density-dependent. My findings indicate that (i) effects of habitat fragmentation on survival may differ between seagrass and terrestrial landscapes, (ii) seagrass habitat fragmentation has a significant but not overriding influence on faunal survival, and (iii) seagrass patch size, seagrass complexity, and blue crab density all influence juvenile blue crab survival, but their effects vary temporally. Habitat fragmentation studies should incorporate multiple scales of space and time, as well as potentially confounding environmental variables.

There currently exists a major dichotomy in the nesting population trends of leatherback turtles (Dermochelys coriacea) worldwide. Eastern Pacific populations have been declining precipitously while populations in the Atlantic and western Indian Oceans have been either stable or increasing. The populations in the western Pacific have also declined with some near extirpation. Factors attributed to the Pacific population decline are incidental fishery mortality and egg harvesting. Fishery mortality occurs throughout all basins, and with the exception of the western Pacific population, egg harvesting has not been a major factor at the major nesting complexes for almost two decades. Populations in the Atlantic and western Indian Oceans currently tolerate mortality from top-down factors, while the eastern Pacific populations continue to decline. Therefore, trophic forcing from bottom-up factors may be the major cause of this dichotomy. This study addressed the effects of bottom-up and climatic forcing on leatherback populations worldwide, with an emphasis on leatherbacks in the eastern Pacific. Nesting leatherbacks in the eastern Pacific were highly influenced by ENSO. Cool, highly productive La Nina events caused a high nesting probability whereas warm, less productive El Nino events caused a low probability. Areas of the equatorial Pacific that produced the most accurate nesting estimates, as indicated by sea surface temperature anomalies, were located around the northeast. Foraging conditions in the eastern equatorial Pacific were a function of ENSO governed primary production transitions that determined leatherback nesting numbers in Costa Rica. It appeared that resource availability in the temperate, southeastern Pacific was not sufficient on its own to support vitellogenesis and the nesting process. While coastal foragers were common among most leatherback populations worldwide, they were rare within the eastern Pacific population, possibly from high mortality associated with coastal gillnet fisheries along Central and South America. Among populations worldwide, eastern Pacific leatherbacks had the lowest reproductive output derived from climate driven, inconsistent resource availability. This study demonstrated that the highly sensitive nature of eastern Pacific leatherbacks to anthropogenic mortality derived from interannual and multidecadal climate variation in their foraging areas, thus accounting for their decline rates despite continued beach protection.

Community models predict the effects of stress on community structure and processes. I tested the Menge and Sutherland 1987 model in an estuarine epifaunal community experiencing low oxygen stress, termed hypoxia. Epifauna, animals living on the surfaces of substrates, are ecologically important in many estuaries where hypoxia occurs, yet little is known about the effects of hypoxia on these communities. Epifauna formed dense communities in the York River, a tributary of the Chesapeake Bay, USA, despite frequent hypoxia. Abundance and species composition was similar in two areas of the river, even though the downstream study area often experienced lower oxygen concentrations during hypoxic episodes than the upstream study area. Many dominant species exposed to high and low oxygen in the laboratory had a median lethal time (LT50) in hypoxia greater than the duration of typical hypoxic episodes in the York River, suggesting that hypoxia may cause little mortality for many species in this system. Predation by a variety of taxa decreased during hypoxia in the laboratory, because many mobile predators had higher mortality than sessile prey, and because predation rates decreased. Peak recruitment of dominant taxa, and of total epifauna, in the York River occurred during neap tides in the downstream study area, coinciding with the lowest oxygen concentrations. In the laboratory, low oxygen decreased recruitment of dominant taxa, but some recruitment continued in hypoxia for most taxa, indicating that larvae of dominant epifauna are tolerant of hypoxia. Larval tolerance of hypoxia may allow communities to persist even though the recruitment season of many epifaunal species coincides with the peak season of hypoxia. These findings supported some predictions of the consumer stress model, but not all. as predicted, when stress increased, the importance of disturbance for determining community structure increased, while the importance of predation decreased. Unlike predictions, stress changed recruitment rates in the laboratory. There were few effects of stress on abundance and diversity, possibly because in this system hypoxia is mild, brief, and because the species in this community can tolerate stress, colonize disturbed areas quickly, and grow quickly enough to complete life-cycles between hypoxic episodes.

Otoliths, scales, dorsal spines, and pectoral fin rays were compared to determine the best hardpart for ageing weakfish, Cynoscion regalis. Sectioned otoliths showed the clearest marks and were validated by the marginal increment method for ages 1-5. Traditionally-used scales were found to be less-precise and to underage older fish. Most weakfish from the Chesapeake Bay region were 200-600 mm TL and ages 1-4. Weakfish were not fully-recruited to commercial foodfish grades until age 2. Maximum observed age was 17 from a Delaware Bay fish collected in 1985. Current maximum observed ages were age 12 in Chesapeake Bay and age 11 in Delaware Bay. Fish older than age 6 were rare in both areas. Weakfish growth was well-described by the von Bertalanffy growth model (R&\sp2& = 0.98, N = 857). There was no evidence Delaware Bay weakfish reached a larger asymptotic length or size-at-age than Chesapeake Bay fish. However, maximum size and age from both areas has fluctuated in the past thirty years. Maximum size greatly increased from the late 1960's until roughly 1985, as did the numbers of large fish, apparently due to a series of strong year-classes, beginning in the late 1960's. Weakfish are multiple spawners with indeterminate fecundity and a spawning season from May to August, in the Chesapeake Bay region. Sex ratios were approximately 3:1, females to males, in 1990-1992. Mean length at first maturity for males and females was 164 and 170 mm TL, respectively. Most fish were mature by age one and all fish were mature by age 2. Spawning activity was not consistent throughout the spawning season or between 1991 and 1992. Batch fecundities ranged from 75,289-517,845 eggs/female and significantly increased with both TL and somatic weight. Spawning frequency was every 2-3 days in 1991 and every 12-13 days in 1992, leading to an average annual fecundity of 7,369,750 eggs/females in 1991 and 1,808,056 eggs/female in 1992. Patterns of spawning activity within and between years appeared closely associated with feeding success. Yield-per-recruit analysis indicated that, over a likely range of natural mortality rates, growth overfishing is currently occurring in the Chesapeake Bay region. Current t&\sb{lcub}\rm c{rcub}& is &\le& age 2, whereas maximum yield consistently occurred at t&\sb{lcub}\rm c{rcub}& &\ge& age 6.

Hundreds of stranded turtles wash up deceased on Virginia’s coastline each year, yet the causes of most stranding events are poorly understood. In this thesis, a carcass drift model was developed for the Chesapeake Bay, Virginia, to predict likely locations of mortality from coastal sea turtle stranding records. First, field studies were carried out to better parameterize the drift characteristics of buoyant sea turtle carcasses, namely, probable oceanic drift times and the impact of direct wind forcing on carcass drift. Based on the duration that tethered, free-floating turtle carcasses were buoyant, we determined that oceanic drift duration of turtle carcasses was highly dependent on water temperature and varied from 2-15 days during typical late spring to early fall bay water conditions. The importance of direct wind forcing for turtle carcass drift was assessed based on track divergence rates from multiple simultaneous deployments of three types of surface drifters: bucket drifters, artificial turtles and real turtle carcasses. Turtle drift along-wind leeway was found to vary from 1-4% of wind speed, representing an added drift velocity of approximately 0.03-0.1 m/s for typical bay wind conditions. The information obtained from these field studies were used to parameterize the oceanographic carcass drift model, which was applied to reported strandings during 2009-2014. Predicted origin of stranding records with probable cause of death identified as vessel strike were compared to commercial boating data. Locations of potential hazardous turtle-vessel interactions were identified in high traffic areas of the southeastern Chesapeake Bay and James River. Commercial fishing activity of various gear types with known sea turtle interactions were compared in space to predicted mortality locations for stranded turtles classified with no apparent injuries, suggesting possible fisheries-induced mortality. Probable mortality locations for these strandings were found to vary between spring peak and summer off-peak stranding periods, but two distinct hotpots were identified in the southwest and southeast portions of the lower bay. Spatial overlap was noted between potential mortality locations and gillnet, seine, pot, and pound net fisheries. These predictions provide clear space-time locations for focusing future research and prioritizing conservation efforts. Nevertheless, the lack of fine temporal and spatial resolution fishing data limited our ability to quantitatively assess most likely causes for specific stranding events. This study both highlights the importance of addressing these data gaps and provides a meaningful conservation and management tool that can be applied to stranding data of sea turtles and other marine megafauna around the globe.

The effects of water quality and sediment composition on mid-Atlantic semi-annual and perennial Zostera marina reproductive success, seed-bank viability, and seed germination were elucidated using laboratory and in situ experiments, quantitative field observations, and ecological model simulations. The sediment seed-bank was found to play a large role in the recovery of perennial Z. marina beds in the Chesapeake Bay and in the yearly re-establishment of beds in North Carolina which were determined to have a semi-annual life history. However, the resiliency provided by sediment seed-bank for both semi-annual and perennial Z. marina beds was limited as seeds remained viable for less than one year. When comparing the two life forms, semi-annual Z. marina beds produced a greater proportion of flowering shoots and more seeds than nearby perennial beds. Seed germination was significantly affected by sediment type and burial depth with maximum germination of seeds occurring in sediments containing >3% organic content and buried at depths <3 cm. Model simulations indicate that York River Z. marina beds are currently at their maximum temperature threshold and that projected increases of >1??C in water temperature within the Chesapeake Bay may result in large scale declines. While the sediment seed-bank may provide a mechanism for recovery following one year of increased temperature stress, seed-banks are depleted following large scale germination events and may not provide resiliency to multiple consecutive years of stressful conditions. Further research into the interactive effects of sediment and water column conditions and seed physiology on seed viability are required to gain a more comprehensive understanding of seed-bank dynamics in Z. marina beds. Monitoring of semi-annual and perennial Z. marina beds in North Carolina indicated that multiple life history strategies may be found within one Z. marina bed. Shoots within the semiannual Z. marina bed germinated from seeds, a portion of seedlings flowered during their first year of growth, and all shoots completed their life cycle and died within one year of germination like a semi-annual plant; however, not all shoots flowered and shoots reproduced both sexually and asexually similar to a perennial plant. Since the individual plants found within the semiannual bed did not display the all of the defining characteristics of either perennial or annual Z. marina life histories, this population cannot be completely described by either life history strategy. Research into the development of a semi-annual life history strategy for Z. marina within this site and the possibility of this form occurring at other geographic regions requires additional research. Sexual reproduction is an important component of both semi-annual and perennial Z. marina populations that should be included in ecological studies and models. Although most perennial Z. marina beds rely on asexual reproduction as a primary form of bed maintenance, the ability to reproduce sexually is maintained and, as shown here, may play a large role in the recovery, maintenance, and expansion of these populations. For these reasons and due to the use of seeds in restoration of Z. marina beds within large systems such as the Chesapeake Bay, research into the dynamics of sexual reproduction within existing beds, a better understanding of seed physiology, and additional research into environmental effects (including the sediment) on seed germination and viability are essential.

Sexual reproduction provides submerged aquatic vegetation (SAV) populations unique opportunities for dispersal, genetic mixing, and resilience in the event of catastrophic population declines. Relative to asexual reproduction, sexual reproduction is a risky resource investment and can have a lower probability of success. A wide variety of abiotic and biotic interactions common in both terrestrial and aquatic environments can lead to significant mortality of seeds and seedlings. The goal of this dissertation is to explore the diversity of biological interactions that influence seed and seedling survival in SAV that drive the population dynamics and restoration success of SAV species. A combination of survey and experimental methods were used to test if three biological interactions, disturbance, herbivory, and competition, compromised seedling recruitment for three different SAV species growing in three different coastal environments. Chapter One explored the influence of sediment bioturbators on seedling establishment for the seagrass Posidonia australis in a marine environment. Field surveys demonstrated that dispersed seeds of P. australis overlap with a suite of sediment bioturbators that disturb the sediment in the coastal lagoons of Western Australia. The movement of sediment bioturbators found in these areas (sand dollars, sea stars, and heart urchins) dislodged and moved recently settled P. australis seeds. The overlap in habitat suitability between these animals and P. australis seeds suggests high densities of these animals consistently disturb non-dormant P. australis seeds and may act as a bottleneck to seedling recruitment. Chapter Two explored the role of grazers on seedling recruitment of a freshwater angiosperm. Field surveys recorded high grazing levels of isolated Vallisneria americana propagules in oligohaline areas of the James and Chickahominy Rivers, VA. Camera surveys identified the blue crab, Callinectes sapidus, as a likely herbivore. Subsequent surveys and experiments suggested C. sapidus in this system consumes SAV as part of their diet. The emergence of non-native SAV in the system, but not V. americana, suggests grazing prevents the recruitment of some SAV, but not others. Chapter Three evaluated interactions between adult plants and seedlings and how they impacted seedling establishment of an estuarine seagrass species. Field surveys consistently recorded seedlings establishing among existing Zostera marina shoots in a meso/polyhaline region of Chesapeake Bay, VA. Concurrent experiments indicated that seed supply influenced seedling establishment rates in some areas. Further surveys and experiments showed that negative interactions between seedlings and adult shoots influenced the subsequent survival of these seedlings within existing Z. marina meadows. Results from the research studies in this dissertation, conducted across three diverse coastal habitats, demonstrate that interactions between SAV seeds or seedlings and other biota can be very important in ultimately determining seed or seedling survival. The diverse mechanisms through which biota compromise seedling recruitment and sexual reproduction for SAV observed here suggest there may be many additional, unexplored biological interactions affecting successful sexual recruitment for many SAV species. Because sexual reproduction provides substantial benefits to SAV populations, incorporating risks associated with seedling recruitment into population models and restoration strategies may help better predict SAV population health, resiliency and expansion as well as help optimize SAV restoration efforts.

Anthropogenically-induced climate change has resulted in increases in water temperature and the frequency and severity of hypoxic events in coastal areas worldwide. Temperature and hypoxia affect fishes' energetics which can, in turn, be reflected in changes in reproductive success and shifts in spatial distributions. in an effort to quantify these changes in Atlantic croaker (Micropogonias undulatus) and spot (Leiostomus xanthurus) in Chesapeake Bay. I: (1) estimated standard and maximum metabolic rates and hypoxia tolerances at five temperatures (10, 15, 20, 25 and 30°C) using intermittent-flow respirometry, (2) examined the effects of hypoxia exposure on metrics of reproductive potential and, (3) developed an individual-based, dynamic-seascape model of Atlantic croaker and spot based on data from the respirometry trials, The first set of experiments showed that metabolic scope (i.e., the difference between standard and maximum metabolic rates, and within which all aerobic metabolic processes must operate) increased with increasing temperature in both species between 10 and 20°C, but plateaued above 25°C in Atlantic croaker and above 20°C in spot. Except at 10°C, the metabolic scope of Atlantic croaker was less than that of spot at all temperatures. in contrast to previous studies with Atlantic croaker from the Gulf of Mexico, the relative expression of hypoxia-inducible factors and metrics of reproduction (gonadosomatic index, most-advanced oocyte stage, and proportion of atretic oocytes) did not differ between Atlantic croaker captured under normoxic and hypoxic conditions in Chesapeake Bay. Simulations of the movements and distribution of Atlantic croaker and spot using individual-based models suggested that these species would occupy areas with warmer and better-oxygenated water than indicated by trawl survey observations from 1988-2014. Additionally, simulations indicated that a greater proportion of Atlantic croaker and spot in the Virginia waters of Chesapeake Bay would occupy the lower portion of Chesapeake Bay than indicated by capture rates from the trawl survey. My research suggests Atlantic croaker and spot are well-adapted to the environmental conditions of Chesapeake Bay during summer and are likely not affected by the frequent hypoxic episodes occurring in the subestuaries of the lower Chesapeake Bay. The apparent larger effect of elevated temperature on the metabolic scope of spot may provide them a greater capacity for movement, growth, and reproduction in warmer conditions and thus, a competitive advantage over Atlantic croaker as water temperatures continue to rise due to anthropogenically-induced climate change. My results indicate that intermittent exposure to hypoxic conditions is unlikely to negatively affect the reproductive potential of Atlantic croaker. Additional research, however, is necessary to better understand how this intermittent hypoxia exposure affects the endocrine pathways controlling reproduction. Finally, although climate-change science frequently focuses on the effects of rising coastal water temperature, and fisheries science and management on the effects on fish distributions, the results of my individual-based models suggest that predicting the effects of anthropogenically-induced climate change should not focus on temperature alone, as this may not be the most important driver of changes in fish distribution. More specifically, other factors such as time-area specific hypoxic events, prey availability, and predator avoidance likely contribute to the spatial distributions of these species in Chesapeake Bay.