Agricultural development in the Mississippi River Basin has contributed to an 3-fold increase in NO3 loading of the river. Increased NO3 loading is a primary cause of eutrophication in the northern Gulf of Mexico. Identification of best management practices (BMPs) to reduce NO3 loss and wetlands restoration to remove NO3 through denitrification are critically needed. The objectives of this research were to determine factors controlling denitrification potential of different landscape units in an agricultural watershed and quantify the effects of BMPs and organic C amendments on denitrification rates of cultivated lands and restored forested wetlands. N2O, CH4 and CO2 emissions from restored and natural forested wetlands were measured to determine if restoration for NO3 removal will increase greenhouse gas emissions, thereby contributing to global warming and compromising the water quality benefits of restoration. Low-elevation, wetland clay soils exhibited 6.3 and 2.5 times greater denitrification potential than the high-elevation silt-loam and low-elevation clay soils under cultivation, respectively. Denitrification potentials of vegetated ditches were 1.3 to 4.2 times greater than the unvegetated ditches and cultivated soils, respectively. Soil cores collected from forested wetlands displayed 2.0 - 6.6 times greater denitrification rates than cultivated soils when incubated at 70 to 100% water-filled pore space (WFPS). Significantly lower N2O:N2 emission ratios were observed from wetlands than from cultivated soils. Denitrification rates in the cultivated and restored forested wetland soils increased 200% and 42%, respectively, when amended with cotton gin trash (CGT). BMPs increased denitrification rates of restored wetlands. Nitrate addition to forested wetlands led to a 48% increase in N2O emissions. Forested wetlands exhibited net CH4 sink of 438-1050 g CH4 ha-1 y-1. CO2 emission decreased as WFPS increased from 40 to100% in forested wetlands. Low-elevation clay soils in agricultural watersheds are the best candidates for wetland restoration for water quality improvement. CGT amendment of cultivated and restored soils in conjunction with BMPs can help reduce on-site NO3 loss. Wetland restoration in the Lower Mississippi valley will not significantly affect the global greenhouse gas emissions budget; however, increased N2O emissions due to NO3 additions merit consideration when establishing CO2 storage credits on restored wetlands.

As biological invasions have become a common phenomenon throughout the world, ecologists have intensified efforts to understand why natural communities are susceptible to invasion. Invading species can cause shifts in community structure that result in irreversible changes to ecosystem function. Phragmites australis has rapidly spread in North American coastal wetlands during the past 50 years and has become a dominant feature in Northern Gulf of Mexico brackish marshes. The rate at which Phragmites is spreading or the mechanisms controlling its establishment in these marshes is unknown. My research objectives were to: (1) determine the spatial and temporal patterns of Phragmites invasion and expansion; (2) evaluate how disturbance and nutrient enrichment controls brackish marsh invasibility and Phragmites establishment, and (3) identify the ecosystem impacts occurring within a brackish marsh during Phragmites invasion. I found substantial increases in the abundance and size of clones of Phragmites during the past 75 years. Annual increases of 11-23% occurred in area covered by clones, which had intrinsic rates of increase in size of 0.07 - 0.23 yr-1. To test marsh invasibility, I manipulated both nutrient levels and disturbance regimes in conjunction with purposeful introductions of Phragmites seed and rhizome material. Phragmites demonstrated the potential for active growth and spread when rhizomes were introduced into brackish marsh. To examine the ecosystem impacts of Phragmites invasion, I located three isolated Phragmites invasions and identified four distinct community types along a transect from the center of each invasion to adjacent un-invaded marsh. My results demonstrate for the first time that Phragmites increases marsh surface elevation relative to un-invaded marsh. Phragmites invasion resulted greater aboveground biomass, increased organic matter accumulation and peat development and lower cellulose decomposition rates relative to un-invaded marsh. The numbers and sizes of Phragmites invasions are increasing without apparent restriction in this Louisiana brackish marsh. These communities remain vulnerable to future Phragmites invasions if rhizomes are transported to new locations. Furthermore, Phragmites has an obvious affect as an ecosystem engineer and may allow invaded marshes to better tolerate increasing water levels due to sea-level rise/land subsidence than native short-stature graminoids.

Little is known about the morphometrics and the feeding ecology of paddlefish in Louisiana. I examined the morphometrics and feeding ecology of paddlefish from the Mermentau River, Louisiana from March 2002 to February 2003. Paddlefish were examined morphometrically by taking measurements of eye-fork length (EFL), body cavity length (BCL), girth (G), vent girth (VG), mouth vertical (MTB), mouth horizontal (MSS), and mouth gullet bottom (MGB) to determine which parameters would give a better estimate of weight (W). Regression analysis resulted in a morphometric model of the form: logW = 10.10 + 1.52 (logEFL) + 1.21 (logG) + 0.24 (logVG) (r2 = 0.97), though I recommended a more inclusive conservation model of the form: logW = -9.82 + 1.15 (logG) + 1.39 (logEFL) + 0.25 (logVG) + 0.15 (logBCL) (r2 = 0.97). Reanalysis of range wide data revealed reservoir/lake paddlefish were generally heavier than river paddlefish at larger eye-fork lengths. This may reflect denser patches of zooplankton in reservoir/lake systems, which may allow for greater paddlefish foraging success than in rivers. Copepods and cladocerans occurred in 90% and 82% of the stomachs, respectively. Feeding seasonality peaked in summer and winter. Diet appeared to switch from copepods in the summer to cladocerans in the winter. Diatoms dominated most of the full May stomachs. Paddlefish diets were similar between seasons except for the summer and winter seasons. Copepods and cladocerans contributing to most of the similarities and dissimilarities. Paddlefish diets were similar between all stages of sexual maturity with copepods and cladocerans contributing to most of these similarities and dissimilarities. An evaluation of size selectivity using Chessons Alpha indicated non-selective feeding and thus paddlefish may be described as indiscriminate planktonic feeders. Calculations of apparent ingestion time indirectly suggest that paddlefish were able to locate dense patches of zooplankton. This study followed the Mississippi Interstate Cooperative Resource Associations (MICRA) paddlefish studies.

The paucity of research into the environmental requirements, stock membership, abundance and residency patterns of bottlenose dolphins (Tursiops truncatus) in coastal Louisiana creates difficulty in understanding how local ecosystems and threats (such as fishery interactions, habitat degradation and pollution) affect populations. This study combined fine-scale environmental measurements and photo-identification techniques to describe patterns of habitat usage and abundance of bottlenose dolphins in lower Barataria Basin from June 1999 to May 2002. In addition I investigated the validity and limitations of using mark-recapture models to estimate abundance from cetacean photo-identification data. Bottlenose dolphins were present year-round in a wide range of water temperatures (10.9 33.9 ºC), dissolved oxygen levels (3.7 16.6 mg/L), salinities (11.7 31.5 psu), turbidity levels (1.4 34.0 NTU), distances from shore (3 800 m), and water depths (0.4 - 12.5 m). However, feeding activity was concentrated in a narrower range of conditions, 20 24 ºC water temperature, 6 9 mg/L of dissolved oxygen, turbidity values between 20 28 NTU, 200 500 m from shore, and depths of 4 6 m. Spatial mapping showed differences in the seasonal distribution of individuals and a tendency for feeding activity and larger group sizes to be concentrated in passes. Using distinctive natural markings present on dorsal fins, I identified 133 individual dolphins. Closed-population models were improved by inclusion of temporal and individual heterogeneity as sources of sighting variability and produced estimates of between 138 and 238 (95% CL range = 128 297) bottlenose dolphins for the study area. Analysis of Jolly-Seber model assumptions demonstrated the importance of ensuring cetacean surveys accurately represent temporal, geographic and demographic properties of a study population. In addition such factors as non-preferential image acquisition, group size, gender, behavior, stability and distinctiveness of natural markings, weather conditions and boat traffic must be considered. Evidence of a relatively closed Barataria Basin population agrees with current assumptions that bay bottlenose dolphin stocks are distinct from those found in deeper, offshore waters. Furthermore, the characterization of environmental usage patterns for this bay population strengthens adequate description and management of this relatively discrete Gulf of Mexico bottlenose dolphin stock.

An extensive search for data on life-history and ecological variables was conducted for a representative number of the species within the family Lutjanidae. After creating different databases and standardizing all data, these variables were examined in statistical, correlation and graphic analyses. Additionally, a series of Principal Component Analyses were used to examine patterns among variables. Life-history variables included age at length zero, asymptotic length, maximum length, longevity, asymptotic weight, length at maturity, age at maturity, reproductive life span, growth rate and mortality rate. Ecological variables included latitudinal and vertical distribution, habitat selection (represented by substrate type), and spawning seasons. The feeding habits of snappers were examined after determining the most common items reported in the literature and summarizing them in ten categories. Results indicated that life-history variables related to the size of the species were highly correlated, supporting the findings of other studies that length at maturity is dependent on asymptotic length. Longevity and reproductive life span were also highly correlated, while longevity and age at maturity had a low correlation, indicating that maturity among snappers is independent of life span and with relatively low variation among species. A review of the geographic and vertical distribution of snappers suggests that pre-settlement larvae in several species may exhibit strong nursery habitat selection, and that adult snappers have distinctive vertical distributions. Finally, based on the variability of prey items found in snapper diets, two basic feeding strategies were evident, species in the subfamily Lutjaninae feed mainly on benthic organisms, mainly crustaceans; while species in the subfamily Etelinae feed in the water column on a variety of pelagic fishes and invertebrates.

The diets of a transitory fish (spotted sea trout, Cynoscion nebulosus) and a fish with presumed greater site-fidelity (pinfish, Lagodon rhomboides) were analyzed with respect to habitat. Sampling occurred in Barataria Bay, Louisiana between May 2003 and May 2004. Spotted sea trout were caught using gillnets, and pinfish were caught using baited fish traps. Each gear was used on three habitats: mud bottom, oyster shell, and marsh edge. In addition, sampling for spotted sea trout was repeated at three locations along a salinity gradient. A total of 175 spotted sea trout stomachs and 137 pinfish stomachs were examined. Seventeen prey species were identified in the guts of spotted sea trout, and thirteen species were identified for pinfish. Fifty-three percent of spotted sea trout stomachs were empty, and 37% of pinfish stomachs were empty. Unidentifiable fish and brown shrimp(Farfantepenaeus aztecus) were consistently the most important prey items for spotted sea trout in weight, frequency, and caloric value. Plant material, detritus, and small amounts of fish and shrimp, were consistently most important for pinfish by weight, frequency, and caloric value. The diet of spotted sea trout caught adjacent to marsh edge had the least dietary overlap with fish from other habitats, especially during winter and spring. The diet of pinfish caught over shell had the least dietary overlap with conspecifics from other habitats. Dietary breadth for spotted sea trout was similar across habitats and lowest in winter. Dietary breadth for pinfish was lowest on the shell habitat and in fall. Neither habitat, season, water quality (i.e., dissolved oxygen, temperature, salinity), nor any other effect could be shown to be statistically significant in distinguishing the diet of either fish using Analysis of Variance (ANOVA) and Analysis of Similarities (ANOSIM) statistical techniques. Diets of these two fish were not notably diverse, but the dietary breadth for both fish species was high, suggesting that of those prey items consumed, there was no strong preference among the prey. This may be attributable to the transitory nature of the fish and the ubiquity of some of the prey types across habitat types.

Elevated soluble aluminum concentrations can adversely affect plant growth. During a drought, wetland soils may experience higher than normal soluble aluminum due to the oxidation of metal sulfides and resulting decreases in pH, which mobilizes metallic cations. Louisiana coastal salt marshes were subject to a record-setting drought in the winter and spring of 2000 which was coincident with the die-off of large expanses of salt marsh, termed " brown marsh ". Spartina alterniflora was the primary plant species affected. However, because some individuals within large areas of die-off survived the brown marsh event, they may have been the more resistant genotypes. To determine if genotypic resistance to aluminum existed, six genotypes of the common salt-marsh cord-grass Spartina alterniflora, five surviving genotypes, and a commercial variety (Vermillion), were dosed with aluminum chloride (AlCl₃) at concentrations ranging from 0.2 mM to 10.8 mM. No death was observed in any of the genotypes at aluminum concentrations as high as 10.8 mM, although growth rates decreased to near zero. The results of this study indicate that, as a species, the resistance of Spartina alterniflora to aluminum may surpass the threshold of any plant species studied to date. All genotypes in the experiment were found to tolerate extremely high concentrations of aluminum, although declines in stem elongation rate and cumulative stem height were evident in all Al treatments. I estimated the differential aluminum tolerance by using the first significant decrease in growth rate when the genotype x concentration effect was significant. The first significant decrease approach had the best resolution for determining genotype variability when used with the stem elongation data. Although insufficient evidence exists to determine if aluminum toxicity caused the brown marsh event in Louisiana, based on the results of this thesis, the aluminum concentrations would have had to reach extremely high levels to have been the sole cause of the brown marsh dieback.

Copepods are important components of marine ecosystems. Understanding copepod population dynamics can help interpret variations in both primary producers and higher trophic levels. Egg production, stage duration, and stage-specific mortality rates are key parameters describing copepod population dynamics. Estimation of stage-specific mortality is complicated due to a complex life history, patchiness, and sampling biases. This study was undertaken to quantify the population dynamics of the copepod <i>Clausocalanus furcatus</i> in the northern Gulf of Mexico and to assess the utility of available mortality estimation methods in a highly advective environment. Zooplankton samples were taken every 12h from March 18-April 6, and May 15-June 9, 2003 from an offshore petroleum platform using a 153μm net and a 30L Niskin bottle to characterize the mesozooplankton assemblage. Incubation experiments were conducted during June-July 2002, March-April 2003, and May-June 2003 to measure egg production rates and stage durations. Stage-specific mortality rates were estimated using the horizontal life table (HTL), vertical life table (VTL), quadratic programming method (QPM) and inverse matrix method (IMM). Mesozooplankton communities in the study area were influenced by the Mississippi River plume. Field estimates of the mean egg production rate of <i>C. furcatus</i> were lower than measurements from lab incubation experiments. Egg production rates did not appear to be limited by food availability. A complete generation time ranged from ~13-20d. Early naupliar stages had shorter durations than late copepodite stages. Comparisons among HTL, VTL, QPM and IMM showed that the HTL and VTL had the disadvantage of producing negative mortality estimates, while the QPM likely overestimated mortality rates. Simulation experiments indicated that variability in stage abundances was a key factor affecting estimates of copepod mortality by the QPM and IMM techniques. Neither the QPM nor the IMM performed well when stage abundance variability was high. IMM estimates of instantaneous egg mortality rates were 1.30d^-1 in March-April and 1.60d^-1 in May-June. While instantaneous mortality rates for NI-CCIV stages ranged from 0.02 to 0.18d^-1. Simulated populations using the mortality rates estimated from the IMM technique were consistent with observed field population trajectories.

Rice fields are a major source of the greenhouse gases methane (CH₄) and nitrous oxide (N₂O) and contribute to nitrate (NO₃^-) pollution in waters. Ferric iron (Fe³⁺) and manganic manganese (Mn⁴⁺) are two intermediate alternative electron acceptors (AEAs) capable of regeneration in freshwater soils. In this investigation, the influences of iron-centered intermediate redox processes on NO₃^- reduction and CH₄ formation in rice soils were studied using soil slurries, soil columns, and potted rice. Reduction of Fe³⁺-centered intermediate AEAs was mainly mediated by obligate anaerobes relying on fermentation products. Ferric iron reducers are bioelectrochemically active, supporting bioelectricity generation through a fuel cell process from the flooded soil coupled to the reduction of O₂ or NO₃^- in the overlying water. As a major electron accepting process in anaerobic carbon decomposition, Fe³⁺ reduction stimulated N₂O production but had little influence on overall NO₃^- reduction in the homogenized soil slurries under near-neutral pH conditions. In the flooded soil column and pot experiments, intensification of iron-centered intermediate redox processes under amendments of iron and/or manganese oxides changed the fate of NO₃^- in the overlying water, decreasing heterotrophic denitrification and increasing NO₃^- percolation and N₂O emission. Ferric iron reduction competitively suppressed methanogenic activity in the homogenized soil slurries. The diffusion of the stronger oxidants O₂ and NO₃^- controlled temporal and vertical variations of iron-centered intermediate redox processes, which subsequently controlled temporal and vertical variations of methanogenic activity in the flooded soil columns. In the pot experiment, Fe³⁺ reduction had small effect on CH₄ emission in the early season when CH₄ emission was low but effectively reduced CH₄ emission after midseason drainage intervals through Fe³⁺ regeneration. The roles of iron-centered intermediate redox processes need to be considered in the evaluation and predication of NO₃^- reduction and CH₄ formation in rice fields.

Sidescan Sonar, chirp sonar sub-bottom profiles, and grab samples were collected on the north-central Gulf of Mexico continental shelf as part of an interdisciplinary study mapping juvenile red snapper habitat. Demarcation of essential fish habitat for juvenile red snapper (Lutjanus campechansis) in the Gulf of Mexico is considered critical for effective management of this valuable species. The first goal of this study was to map and describe the geology of this region. The second goal was to attempt to relate variations in geology to juvenile red snapper abundance and distribution. Sidescan mosaics were created for ten polygons, ranging in size from 2 to 20 km² on the inner to middle shelf south of Mississippi-Alabama, in water depths of 17-40 m. Geological observations delineated three contrasting seabed types: (1) linear to patchy shell regions on the inner-middle shelf, (2) muddy sand sheets on the middle shelf, and (3) prodeltaic muds in the southwest of the study area, marking the eastern extent of recent shelf deposits from the modern Mississippi delta. The shell ridges stand 1-3 m above the surrounding seabed, and may extend 200 m across. They are composed of > 50% CaCO₃, including shell fragments from both estuarine and marine taxa, and contrast sharply with adjacent muddy sands containing minor shell. Radiocarbon dating of shell material, along with the geological characteristics of the ridges suggests that they are remnants of Holocene coastal environments. This region was previously described as either an extension of the MAFLA sand sheet or a transitional zone between the MAFLA sands and prodelta muds (Ludwick, 1964). In the present study, we identified a range of geological features of estuarine, shoreface, and wholly marine origin. The diversity of deposits described records a wide range of geological processes active from early-middle Holocene to recent time. The integration of geological observations with coordinated biological observations reveals that geologic structures and sediment composition on the northern Gulf of Mexico continental shelf are major controls on the distribution of juvenile red snapper (Patterson et al, in press), and record both coastal depositional histories and open-shelf processes active during Holocene transgression.