In wetland sediments, organic matter provides a substrate for microbial activity. During metabolism, microbes release extracellular polymeric substances, which accumulate to bind soil particles. A similar concept can be implemented on a large scale to reduce wetland loss in Louisiana. Hypothetically, hydraulically dredged sediment can be amended with polymer and deposited on subsiding marshes as a restoration method where the polymer increases sediment stabilization until plants become established. This lab study focused on investigating the influence of natural polymer additions on particle aggregation to increase sediment stability and the effects on microbial activity. Sediments from three sites (i.e. freshwater, intermediate, and marine) were used, which varied in moisture content, organic matter content, salinity, and texture. The soil amendments were xanthan gum, a microbially produced polymer, and guar gum, a plant polysaccharide. Following polymer application, sediment-polymer mixtures were incubated for 1, 8, 16, or 26 weeks before analysis. Response variables included moisture content, redox potential, pH, dewatering, consolidation, aggregate size, microbial biomass, and basal respiration. Polymer addition increased microbial activity in the first week. Lower redox potentials indicate that more carbon substrates were available to serve as electron donors for microbial use. High respiration rates suggest a microbial response to polymer addition with increased activity and growth, followed by rapid turnover of the biomass. At the 0.5% polymer concentration level, microbes assimilated carbon as indicated by respiration similar to control samples. At the 1% polymer concentration level, increased respiration indicates a transition to an increasing biomass pool. Microbial response to added polymer carbon indicates that microbial communities degraded the polymers within one week of application. No evidence of increased aggregation was found, supported by no polymer effects on dewatering and consolidation. Natural polymer additions may not achieve the goal of increasing sediment stability, due to their water-solubility and simple structure, which contributed to rapid degradation by microbes. High moisture content of wetland sediments may require the use of synthetic polymers for aggregation. A material that maintains structure in water and resists microbial activity may be more successful in stabilizing wetland sediments.

Hurricanes account for a significant portion of damages, injuries, and fatalities in the Caribbean. The Dominican Republic has had 3 major hurricane strikes in this century resulting in loss of lives for thousands of people and billions of US $ in economic loss. Major hurricanes had been relatively infrequent in north-eastern Nicaraguas modern history until Hurricane Felix, a Category 5 storm, made landfall in the north-east in 2007. It caused 130 confirmed deaths and brought devastation to many villages. These events highlight a need for re-evaluation of hurricane risk based on a more comprehensive and long-term hurricane history. Using paleotempestological methods, paleo- hurricane history of this region, going back 7,455 years in Nicaragua and 3,640 years in the Dominican Republic, was reconstructed. To achieve this, 27 cores of sediment from 4 coastal sites in the north-eastern part of Nicaragua and 5 sites on the coast of the Dominican Republic were collected. Presence of sand layers in a mainly lacustrine environment, were used as proxies to identify past storm events. Radiocarbon dating, 137Cs, and 210Pb analysis was used to establish chronology. Data from Nicaragua indicate a shift in the hazard regime occurring on a millennial scale - a high activity period in the last 950 years preceded by a low activity period between 950 - 3,350 BP. Using the average return period of hurricane landfall derived from prehistoric and historic data, Poisson probability of strikes was computed for all coastal provinces and municipalities in the study areas. Results reveal a redistributed likelihood of strike for the coastal areas. Vulnerability of coastal areas was determined using indicators and spatially represented. A hazard analysis was conducted by creating storm surge and wind speed models in ArcGIS. Ultimately, the probability information was combined with vulnerability and hazards data resulting in a comprehensive, spatial risk model for the case studies. Jurisdictions containing the seats of government for the Dominican Republic and RAAN face the highest risk.

Eutrophication and coastal wetland loss are the major environmental problems affecting estuaries around the world. In Louisiana, controlled diversions of the Mississippi River water back into coastal wetlands are thought to be an important engineering solution that could reverse coastal land loss. There are concerns, however, that freshwater diversions may increase nutrient inputs and create severe eutrophication problems in estuaries and wetlands adjacent to the diversion sites. My dissertation research concerns modeling the effects of the observed and hypothetical freshwater diversion discharges on the hydrodynamics, salinity and water quality in the Barataria estuary, a deltaic estuary in south Louisiana. This estuary receives freshwater and nutrient discharges from the Davis Pond diversion, the worlds largest freshwater diversion project. I have implemented two Barataria Bay simulation models of differing complexity, a simple 6-box mass-balance model and a high resolution two-dimensional (2-D) coupled hydrology-hydrodynamic- water quality model. Model results have shown that the Barataria estuary imports nitrogen and exports carbon to the coastal ocean. Compared to the lower Mississippi River, the Barataria estuary appears to be a very small source of total organic carbon for the northern Gulf of Mexico and is unlikely to have a significant influence on the development of the Gulfs hypoxia. Model simulations pointed out that the effects of different diversion discharges on salinity are most apparent in the middle and lower sections of the Barataria estuary. Further, tracer simulation experiments have shown that residence times differ markedly at different locations within the same water body due to differences in small scale hydrodynamics. Model simulations clearly demonstrated the importance of residence times for the overall functioning of the estuary. Model simulations also pointed out the differences in spatial patterns in phytoplankton response to distributed freshwater and nutrient inflows, reflecting the near-field control of nutrients and far-field control of residence times on phytoplankton standing stock. The models reiterate the fact that there are significant tradeoffs in using freshwater diversions in coastal restoration efforts, namely tradeoffs between hydrologic restoration and water quality effects.

Louisianas estuaries and coastal waters play an important role in providing habitat for several estuarine-dependent species and also serve as a nursery for the juveniles of several other species. The current rate of land loss in Louisiana and the decline of estuarine-dependent species force us to improve our understanding of the recruitment of these species into Louisianas estuaries. Habitat suitabilities were developed for several fish species of concern (SOC) and other estuarine-dependent species across three environmental gradients, abundance patterns were developed for seasons, five-year intervals, and coastal study areas (CSAs). Chi-square analyses and ANOVAs were then used to test for differences in the suitability patterns and abundance patterns, respectively. The pipefish species, chain pipefish, Syngnathus louisianae (SOC) and gulf pipefish, Syngnathus scovelli, had significantly different suitability patterns for all three environmental variables. Chain pipefish suitability patterns were significantly different for all three variables, whereas, gulf pipefish only had strong patterns for salinity and Secchi depth. The pipefishes also had different temporal and spatial abundance patterns. The two goby species, violet goby, Gobioides broussonnetii (SOC) and sharptail goby, Gobionellus hastatus, had similar suitability patterns for all three environmental variables, but for sharptail goby did not have a strong salinity suitability pattern. They also had similar seasonal abundance patterns with peaks during the spring months. The three SOC fundulids, diamond killifish, Adinia xenica, bayou killifish, Fundulus pulvereus, and saltmarsh topminnow, Fundulus jenkinsi, shared similar, strong patterns of suitability for temperature and salinity. For Secchi depth, only diamond killifish and bayou killifish had strong suitability patterns. All three species also shared seasonal abundance patterns with peaks in the winter, and spatial abundance patterns with peaks in the central CSAs. The two candidate SOCs, Atlantic threadfin, Polydactylus octonemus and southern puffer, Sphoeroides nephelus, were chosen for further analyses based on their declining annual abundances. They did not have any similar temporal or spatial abundance patterns. The two species had strong suitability patterns for all three environmental variables; however, they did not share any similar patterns. The strong suitability patterns suggest that some of the environmental variables effect habitat selection for several of these species.

The presence or absence of microphytobenthos on the seafloor provides clues about whether benthic oxygen evolution contributes significantly to the oxygen budget of the hypoxic area in the northern Gulf of Mexico. Hypoxia (oxygen < 2 mg l-1) creates inadequate concentrations of dissolved oxygen to support most organisms, such as fish, shrimp and crabs, and occurs over large areas of the Louisiana continental shelf from spring through summer in most years. Oxygen production by benthic autotrophs may offset a decline in oxygen concentrations if there is a functioning community and sufficient light. I sampled three stations (14, 20 and 23 m depths) ~ 100 km west of the Mississippi River over three hypoxic annual cycles (2006 2008), and 11 stations along a 14 - 20 m contour on the shelf in late-July in 2006, 2007 and 2008. I used microscopy and high-performance liquid chromatography to estimate the biomass and composition of phytoplankton and microphytobenthos. The potential seasonal oxygen production was estimated in 2007 and 2008 by incubating coupled light/dark sediment cores and bottom water from two stations. The sediment community (cells > 3 um) differed from those in the water column and were frequently benthic pennate diatoms and filamentous cyanobacteria (58-88% seasonally and 1-99% in mid-summer). The concentration of microphytobenthic biomass was usually < 2.0 ug g dry sed-1, and various biotic parameters were influenced by light at the seafloor. Declines in dissolved oxygen over a seasonal cycle in 2007 and 2008 were affected more by the initial dissolved oxygen concentration than by the presence of microphytobenthos that could generate oxygen. The sediment (1.2 - 27.3 mmol O2 m-2 d-1, n = 97) and bottom-water (1.1 - 17.5 mmol m-2 d-1, n = 23) oxygen consumption rates were within the range of the few previously-reported data. This work adds to these data and also provides the only sediment oxygen consumption rates at fixed sites over seasonal time scales. These results provide critical input to three-dimensional, physical-biological models of oxygen dynamics for this hypoxic area.

The northwestern Gulf of Mexico currently experiences a large hypoxic area (dead zone) during the summer. While the local effects of hypoxia on organisms have been documented, the population-level effects are largely unknown. I developed a spatially-explicit, individual-based model to analyze how hypoxia effects on Atlantic croaker reproduction, growth, and mortality in the northwestern Gulf of Mexico could lead to population-level responses. The model follows the hourly growth, mortality, reproduction, and movement of individuals on a 300 x 800 spatial grid of 1 km2 cells for 100 years. Chlorophyll-a concentration, water temperature, and dissolved oxygen were specified daily for each grid cell for an average year, which was repeated every year. A bioenergetics model was used to represent growth, mortality was assumed age- and size-dependent, and the movement behavior of juveniles and adults was modeled as an unconditioned response to external cues (kinesis) coupled with a neighborhood search algorithm that emulated hypoxia avoidance. Hypoxia effects were imposed using vitality-repair submodels that convert time-varying exposures to reduced hourly growth, increased hourly mortality, and reduced annual fecundity. Results showed that 80 years of either mild or intermediate hypoxia produced small reductions in population abundance, while severe hypoxia caused a 31% reduction in long-term population abundance. The response to severe hypoxia was added age-1 and age-2 daily mortality (8-9%) and a 5% average reduction in eggs per individual. Relatively few individuals (5%) were exposed each hour but many individuals (20-40%) experienced at least one hour of low DO each year. The effects slowly built up in the model over years; population abundance slowly declined for 40 years before the 31% reduction was realized. Under more realistic hypoxia conditions of mild, intermediate, and severe hypoxia years occurring in proportion to their historical frequency, the model predicted an 18-29% decrease in the long-term population abundance. Sensitivity analysis showed hypoxia effects via reduced growth were small, and aspects of avoidance behavior were important in determining the population response. I discuss the strengths and weaknesses of the modeling, and future plans for refining the analysis based on data from ongoing field and laboratory studies.

The Caernarvon Diversion meters Mississippi River water into coastal marshes of Breton Sound. Elevated levels of nitrogen in river water have sparked concerns that nutrient loading may affect marsh resilience and belowground biomass, as evidence from several marsh fertilization studies. These concerns resulted from observation that fresh and brackish Breton Sound marshes suffered extensive damage from Hurricane Katrina. The goal of this study is to determine the fate of nitrate (the dominant inorganic nitrogen form in the Mississippi River) in Breton Sound Estuary marshes. We hypothesized that most nitrate would be removed by denitrification and that nitrate loading would not affect belowground biomass. To test this hypothesis, a mass balance study was conducted using 15N-labeled nitrate. Twelve plant-sediment cores were collected from a brackish marsh located proximal to Delacroix, Louisiana. Six cores received dionized water (control), while another six (treatment) received 2 mg L-1 of 15N-labeled potassium nitrate twice a week for three months. A set of three control and treatment cores were destructively sampled after three months and analyzed for 15N in the above and below ground biomass, as well as the soil. The remaining three treatment cores received 20 mg L-1 of 15N-labeled potassium nitrate twice a week for one month, and a similar mass balance was determined to distinguish N removal, including denitrification, surface algae and microbial uptake and incorporation into aboveground and belowground biomass. Twelve hrs after the addition of 2 mg N L-1 water for each flooding event, nitrate levels were below detection (0.014 mg NO3- L-1. In comparison, after 24 hrs, 20 N mg L-1 water column nitrate levels were approximately zero. The 15N analyses determined 68, 65, and 74% of added labeled nitrate as unaccounted for, which represents gaseous losses. The remaining 15N was incorporated in plant and soil compartments. Labeled N data from the 2 mg N L-1 treatment and 20 mg N L-1 treatment suggests denitrification as the major removal pathway for nitrate in Caernarvon Diversion. Comparison of nitrate loss in bayou sediment and marsh soil suggests nitrate removal would be enhanced if diverted Mississippi River water flooded the marshes.

The Gulf of Mexico (GOM) red snapper stock has been exploited since the mid 1800s; yet it is still one of the most economically important fisheries in the GOM. Red snapper have been managed as a unit stock and are currently overfished, but perhaps no longer undergoing overfishing. Habitat varies greatly throughout the GOM and while numerous studies have aged red snapper, none have simultaneously compared the age and size structure and growth rates among standing and toppled oil and gas platforms with natural habitats. The objectives of this study were to examine the size and age structure and growth rates of red snapper among three different habitats (shelf-edge banks, standing platforms, toppled platforms) and six recreational fishing regions of the GOM (South Texas, North Texas, Louisiana, Alabama, Northwest Florida, Central Florida). Across all of the habitats and regions, red snapper were small (mean TL = 526.84 mm, mean TW = 0.97 kg) and from younger age classes (mean age = 4.44 yr), representing the strong recruitments of 2004, 2005 and 2006, with few fish older than seven years (1.5%). Total length, weight, and age frequencies, and growth models differed significantly among the habitats. Red snapper from the banks were significantly smaller at age and slower growing than red snapper from the artificial habitats. Also, shelf-edge banks appear to support a higher predominance of older red snapper compared to the artificial habitats. Demographic differences in red snapper size and age frequencies and growth parameters exist across the GOM. Small, fast-growing individuals dominated the recreational catches of South Texas, Northwest Florida, and Central Florida, whereas larger, slower growing red snapper constituted the majority of the Alabama and Louisiana catches. Also, both of the Florida regions catches were comprised of significantly younger red snapper than catches in the north-central and western regions. To prevent habitat- and region-specific overfishing and promote stock recovery, these differences should be weighed when evaluating future stock assessments and management decisions. It is also important for fisheries managers to note the absence of old red snapper in this study and its implications for the stocks recovery status.

Red snapper, Lutjanus campechanus, has been an economically important reef fish in the Gulf of Mexico (Gulf) for over 150 years and is currently overfished. Catch statistics and demographic differences have lead to the population being categorized into eastern and western substocks divided by the Mississippi River, but data is recombined to set a Gulf-wide annual catch limit. The two objectives of this study were to apply otolith nursery chemical signatures to estimate red snapper mixing dynamics in the western Gulf, and to determine if signatures based upon trace metals associated with oil and gas platforms could discriminate between region and habitat of origin to further examine population connectivity. Nursery otolith signatures were developed from age-0 red snapper belonging to the 2005 - 2007 year classes and collected from six nursery regions in the Gulf (Florida, Alabama, Louisiana, Texas, Veracruz, and Campeche Banks). Year class-specific quadratic discriminant function analyses (QDFAs) distinguished nursery regions with 71 84% accuracy. Maximum likelihood analyses identified sources of sub-adult and adult red snapper sampled during the summer of 2006 - 2008 from the western Gulf and Mexico regions based on year class-specific otolith core chemical concentrations. Locally derived and Louisiana recruits were apparent among red snapper collected off Texas, but data were inconclusive to estimate connectivity between the western Gulf and Mexico regions. Otoliths of red snapper collected from platforms and other habitats off Alabama, Louisiana and Texas during the summer of 2007 and 2008 were analyzed to determine if platforms impart detectable signatures based on seventeen trace metals. Mean jackknifed classification accuracies from QDFAs indicated higher success for discriminating among regions (86%) than habitats (79%). Maximum likelihood analyses estimated region and habitat of origin of red snapper collected from natural habitats off Florida, Louisiana and Texas during the summer of 2009. Platform signatures were evident in otoliths from red snapper collected off Florida, a region devoid of platforms, possibly reflecting a western Gulf contribution to the eastern substock. The microchemical otolith signatures of western Gulf red snapper in this study demonstrated discrete regional populations with some interpopulation mixing, further supporting a metapopulation structure.

Greenhouse gases were sampled during spring and summer months in 2010 and 2011 at two sites in the Mississippi River Basin: a riparian forested wetland under restored flood pulsing near Memphis, TN and a coastal estuarine marsh complex impacted by a river diversion at Caernarvon, LA. Mean methane fluxes at Caernarvon (-0.09 g CH4-C m-2 d-1) were lower than those at Memphis (0.53 g CH4-C m-2 d-1), and did not differ between seasons. Nitrous oxide fluxes were generally below detection limits. Fluxes were uncorrelated with water depth, soil temperature, and ambient air temperature, though sampling occurred over a narrow range of these variables due to limitations in site accessibility. A water quality dataset spanning 8 years of monthly sampling was analyzed to determine long-term spatial and temporal patterns in nutrient concentrations at Caernarvon. The estuary served as a source of ammonium, and chlorophyll a and a sink for nitrate, total nitrogen and total suspended sediments. Classification of the estuary as a source or sink for phosphate, total phosphorus and silicate changed with the seasons. Nitrate removal efficiency varied seasonally, with efficiency highest in the fall (98%), summer (92%) and spring (87%) and lowest in the winter (74%). Removal efficiency was inversely related to loading rate. Water quality samples were collected at Memphis in a secondary channel of the Mississippi River under restored hydrology during spring and summer 2010-2011. Seasonal variations existed in concentrations of nitrate, total nitrogen, total phosphorus and total suspended solids, but not ammonium, phosphate, silicate, dissolved oxygen and chlorophyll a. No parameters varied longitudinally or laterally within the channel. Nitrate loading rate was high and removal efficiency was low, especially in comparison with Caernarvon and other study sites within the Mississippi River Basin, due to the limited size of the receiving wetland.