No description available.

The coastal landscape is a naturally shifting mosaic of distinct ecosystems that are rapidly migrating with climate change. While directional changes in climate, such as warming and sea level rise, are fundamentally reorganizing the coastal landscape, ecosystem function, especially carbon storage, is affected to an unknown degree. This dissertation presents four chapters that examine the role of ecosystem transitions in coastal carbon dynamics across a range of spatial scales – within individual ecosystems, between two ecosystems, and at the landscape between an array of ecosystems. Ghost forests, or the marsh-forest ecotone, serves as an ideal example of a migratory ecotone. As sea levels rise, terrestrial forests die-off from salt water intrusion and are replaced by salt-tolerant marsh species. While this transition is widely seen and studied, we present the first field study that quantifies carbon loss during this transition (Chapter 1). Significantly, we find that the loss of carbon during marsh migration can be replaced by the accumulating marsh soils, but the timescale for this replacement is at the scale of centuries. Warming, a co-occurring climate stressor, is expected to affect carbon storage to an unknown degree as it affects both antagonistic properties to soil carbon storage: production and decomposition. In a whole-ecosystem soil warming experiment, we find that moderate amounts of warming consistently maximized root growth, marsh elevation gain, and belowground carbon accumulation (Chapter 2). However, our work indicates nonpermanent benefits as global temperatures continue to rise and elevated temperatures exacerbate marsh elevation and carbon loss. At the landscape scale, we see that while climate change can drastically reduce or increase the extent of coastal habitats, compensatory mechanisms largely maintain individual ecosystem extents (Chapter 3). However, coastal squeeze in some environments still reduce extents of ecosystem critical to regional carbon storage. Blue carbon habitats that comprise the coastal zone are able to compensate this loss in less time than it takes to accrue that loss. These findings reveal unique functional compensatory mechanisms at the landscape scale that quickly absorb carbon losses and could facilitate increased regional carbon storage in the face of accelerating climate change. Finally, we concentrate on ecosystem vulnerability of salt marshes, an ecosystem with a critical role in global and local carbon dynamics. By leveraging decadal SET data, we are able to identify early warning signals of marsh collapse in the changing microtopography of the marsh surface (Chapter 4). Increasing microtopographic heterogeneity in degrading salt marshes mirrored trends in a diverse array of systems with alternative stable states – indicating that early warning signals of marsh drowning and ecosystem transition are observable at small-spatial scales prior to runaway ecosystem degradation. Congruence between traditional and novel metrics of marsh vulnerability indicate that microtopographic metrics can be easily applied to existing SET records to identify hidden vulnerability before widespread marsh degradation.

Harmful algae can illicit adverse effects on aquatic and human health through various mechanisms, including through the production of bioactive compounds called phycotoxins. In the Chesapeake Bay, the largest estuary in the United States, little information was known about the distribution of phycotoxins, even though this region is known to harbor 37+ different species of harmful algae. Due to the presence of multiple species that can produce distinct groups of phycotoxins, a multi-toxin approach was taken to study this region. Two methods for the quantification of 13 phycotoxins (microcystin-RR, -LR, YR, azaspiracid-1, -2, karlotoxin 3, goniodomin A, yessotoxin, brevetoxin-2, pectenotoxin-2, okadaic acid, dinophysistoxin-1, and -2) in a single sample were developed using novel technology: ultra-performance liquid chromatography coupled to tandem mass spectrometry equipped with a trapping dimension and at-column dilution. This instrumentation allows for high-volume (up to 1 mL) injections of extracts in 100% organic solvents, reducing time and labor normally required for sample preparation steps and resulting in low limits of detection compared to current literature values. To evaluate the distribution of these phycotoxins and their causative species in the environment, a field study was carried out at 12 sites in the lower Chesapeake Bay between May 2017 and June 2018. Solid Phase Adsorption Toxin Tracking (SPATT), a passive sampling technique for dissolved phycotoxins, was used throughout the field study. The resulting dataset allowed for a spatiotemporal comparison of these compounds across the region. Surface water samples were also enumerated by light microscopy to compare toxin amounts to the presence of harmful algal cells, and environmental parameters (e.g. temperature, salinity) were evaluated over the course of the study to determine if any correlations existed with toxin data. Before samples could be analyzed, a method suitable for the extraction of multiple toxins from SPATT resin (Diaion® HP-20) was developed. The developed multi-toxin methods for extraction and quantification were then applied to SPATTs collected during the field study, and additional analyses for domoic acid were performed using enzyme-linked immunosorbent assays. Out of 15 toxins analyzed for, 8 were detected in this region: microcystin-LR, azaspiracid-1, azaspiracid-2, goniodomin A, pectenotoxin-2, okadaic acid, dinophysistoxin-1, and domoic acid. This study marks the first report of azaspiracids in Chesapeake Bay, and is among the first to report domoic acid in this region. While multiple toxins were present in samples from all 12 sites throughout the study period, harmful algae were sparse in corresponding water samples. This finding stresses the usefulness of passive sampling, a method that provides an integrated measurement of trace amounts of toxin present in the water column. The results from this study show that multiple phycotoxins, spanning both salt- and fresh-water origin, are present throughout the lower Chesapeake Bay. This region, however, does not see recurring seafood harvest closures or human health illnesses due to phycotoxin contamination, suggesting that amounts of these compounds are currently low enough to avoid major human health implications. The potential for low-level chronic exposure remains possible, and these results highlight the importance of considering the effects of chronic exposure to environmental contaminants.

Aroclor 5432, a mixture of polychlorinated terphenyls (PCT), was detected in sediment and several biological compartments including: saltmarsh cordgrass (Spartina alterniflora), American oysters (Crassostrea virginica), red-jointed fiddler crabs (Uca minax), wharf crabs (Sesarma reticulatum) and mummichogs (Fundulus heteroclitus) collected from Tabbs Creek. This tidal creek is located in the southern Chesapeake Bay region. Species from several phyla were selected in order to examine PCT accumulation in physiologically and ecologically different organisms. In general, PCT concentrations in sediment and biota decreased with distance downstream from the PCT outfall. The organism with the highest mean concentration (18,300 &\mu&g/kg dry weight) was the native oyster, a benthic filter feeder. A second study examined the dose-response effects of PCT mixtures on levels of hepatic cytochrome P4501A (CYP1A) and associated EROD activity in the mummichog, relative to a polychlorinated biphenyl (PCB) mixture Aroclor 1254. Fish were injected intraperitoneally with PCT formulations Aroclor 5432, Aroclor 5460, or the PCB magnitude as that caused by equivalent doses of the PCB mixture Aroclor 1254. Treatment with Aroclor 5460 did not result in significant induction. This work represents the first report of hepatic CYP1A induction caused by Aroclor 5432 in teleosts and, similar to work in mammalian systems, suggests that the effects of this mixture may be mediated through Ah receptor binding. PCT mixtures contain small amounts of PCB. Therefore, PCB may have contributed to the induction observed following Aroclor 5432 injection. The planar PCB congener 3,3',4,4'-tetrachlorobiphenyl was identified in Aroclor 5432 by GC-MS operating in negative chemical ionization mode. Environmental induction was observed in mummichogs collected from Tabbs Creek. This study revealed CYP1A induction at the two most contaminated sites along the PCT gradient. Fish at the upper creek site exhibited inhibition of EROD activity. Definitive environmental induction by PCT could not be established due to the presence of other inducing pollutants in the creek.

As a way to gain better understanding of how aquatic organisms respond within their natural environment, this dissertation set up a series of controlled laboratory experiments to investigate the effects of chemical and biological stressors on the zebrafish (Danio rerio). I performed a series of studies in which animals were simultaneously exposed to the toxicant phenanthrene, and infected with the bacterium Mycobacterium marinum. Unexpectedly, we found that at a high dose of phenanthrene (284 mug L-1) results in an antagonistic relationship between stressors with dually challenged fish having significantly higher survivorship than those exposed solely to phenanthrene. Dually challenged fish also had reduced phenanthrene metabolism, suggesting higher mortality in phenanthrene only exposed fish may be due to increased toxicity from phenanthrene metabolites. We postulated metabolic reduction was due to inflammatory cytokines suppressing metabolism. However, in latter experiments using a more environmentally relevant dose of phenanthrene (157 mug L-1), dually challenged fish had significantly higher mortalities than all other treatments, and there was no disruption in phenanthrene metabolism. Our second study clearly showed an additive effect: the summation of mortalities for each stressor individually is approximate to what is observed in dually challenged fish. My third experiment was designed to better elucidate the effects of dose and exposure order on stressor interaction. I used a matrix of dose and order such that two phenanthrene doses were used (157 mug L -1 and 86 mug L-1), with a staggered order of exposure. Significantly higher mortalities in dually challenged fish for the high dose of phenanthrene versus the low dose, regardless of exposure order, demonstrated phenanthrene concentration, not exposure order is a significant factor. This study also showed activation of the cytochrome P450 pathway by phenanthrene. These studies highlight the exceptionally complex interactions between multiple stressors and how minor alterations in experimental design can produce dramatic changes in stressor interaction. I have concluded that toxicant dose plays a significant role in this interaction causing an antagonistic interaction at high concentrations; however, at lower doses an additive effect is seen. Additionally, I have shown the importance of phenanthrene metabolites in toxicity. Finally I have clearly demonstrated that phenanthrene can induce the cytochrome P450 pathway. In addition to dual-stressor experiments, I conducted a comparative study between two Mycobacterium spp.: M marinum and M. pseudoshottsii. The goal of this study was to find a bacterium to model disease recrudescence. Despite the close relationship between these two species, we observed dramatically different virulence and pathology. M. marium infected fish had <10% survivorship over a 4-week exposure; however M. pseudoshottsii infected fish had ∼98% survivorship. Additionally, M. marinum infected fish displayed a classic granulomatous inflammation with bacilli sequestered within, or in immediate proximity to, well formed granulomas. In contrast, M. pseudoshottsii infected fish displayed little granuloma formation, instead having large area of diffuse inflammation and cellular necrosis. M. pseudoshottsii were seen disseminated both extra and intra-cellularly throughout areas of inflammation, a phenomenon not seen with M. marinum. I postulate that these differences are due to a unique mycolactone toxin secreted by M. pseudoshottsii. M. ulcerans is the only other Mycobacterium spp. to produce a mycolactone toxin, and produces pathologies similar to what we observe here for M. pseudoshottsii.

The role of science and scientists in environmental policy and management is and has been an important, complex, and controversial subject for many years. The objective of this study is to determine how science and scientists interact in environmental policy formation and management and how science is or could be used in the development of policy which can ultimately be used as a basis for effective resource management plans. In the very broad sense this study attempts to evaluate the general hypothesis that "Scientists do not play a role in promoting or encouraging science as a means of changing attitudes and opinions of management and the public so as to influence public policy and ultimately environmental management.". The use of science in establishing well developed management plans for coral reef areas in Australia's Great Barrier Reef; Jamaica's - Ocho Rios Marine Park System; St. Croix's - Buck Island; Anguilla; the Netherlands Antilles - including Bonaire and Curacao Marine Parks; Puerto Rico's - La Parguera National Marine Sanctuary; and several of the Florida State reefs such as Key Largo and Looe Key Marine Sanctuaries were examined through analysis of management plans. The second component of the study involved structured interviews with a number of scientists and managers. These individuals included scientists who had been working on coral reefs as well as managers of these systems--individuals who have an interest in formulating public policy as well as those who do not have. The general hypothesis was divided into a number of statements or subhypotheses which were examined to help evaluate the hypothesis. Close-ended questions allowed determination of the reasons why scientists and/or managers feel the way they do in their responses. Appropriate statistics were used to determine if there is a difference in the way scientists perceive their role, as compared to how managers perceive the role of scientists. The null hypothesis that no significant difference exists between attitude of scientists and managers could not be rejected. The general hypothesis was accepted both by scientists and managers.

No description available.

No description available.

No description available.

No description available.