Our Students and Their Research

Anne Arundel County, Maryland, operates a destratification system used to mitigate the effects of eutrophication-driven oxygen depletion in a small tidal tributary called Rock Creek. The system provided the opportunity for ecosystem-scale experimentation of manipulated oxygen levels by turning the system on or off, effectively inducing hypoxia and anoxia. Eutrophication can lead to overall declines in dissolved oxygen concentrations via the elevated consumption of oxygen during the decomposition of organic matter. There are known impacts of these oxygen reductions on phosphorus cycling by increasing the flux of phosphate from sediment into the water column. To determine the severity or duration of hypoxia needed to induce phosphate accumulation in the water column, both model simulations and field observations were used to quantify changes in phosphate fluxes from sediments and associated water column increases. Model predictions indicate that phosphate flux is sensitive to exposure to low oxygen, primarily severe hypoxia (0.5 mg/L). Model results paired with observed field data suggest that within 3–7 days of hypoxic conditions, a large signal of phosphate increase (500–900%) is experienced, likely due to elevated sediment-water flux. Future studies should include a focus on system recovery to increased phosphate flux resulting from hypoxia.

Non-competitive and competitive substrates are two mechanisms in which methane can be produced in sediment. The methane production mechanism is due to oxidation-reductions reactions that are occurring within the sediment. Methane is a powerful greenhouse gas, and estuarine systems are one of the understudied provider of methane. The objective of this study is to understand how methane is being produced in sulfate rich sediments of an eutrophic estuary, the Chesapeake Bay. The study consisted of a top and bottom sediment incubation experiment which contained multiple non-competitive and competitive substrates, the non-competitive substrates included trimethylamine, dimethyl sulfide, and methanol and the competitive substrate was bicarbonate. The sediments were studied for a total of 42 days with 5-time points. The original hypotheses included non-competitive substrates having the highest methane production rates while the competitive substrate did not produce methane and the top sediment will have higher methane production rates in comparison to the bottom sediment. While the data showed evidence of the top sediment incubation producing higher methane production rates and the highest methane production rates being produced from competitive substrates, more questions arose about the experiment; therefore, a future study must be conducted to understand the methane production rates. 

The leatherback sea turtle population (Dermochelys coriacea) has suffered a major decline in recent years. This drop has many causes, including coastal egg poaching and pelagic fisheries bycatch; the latter is believed to be the leading cause of pelagic adult leatherback mortality. Then, in order for the population to recover, fisheries bycatch must be reduced. In this study, we created a spatiotemporal Species Distribution Model (SDM) that incorporates remote sensing data of the seascape into the dynamic management of the Eastern Pacific (EP) leatherback population in the Southeast Pacific. We obtained leatherback observation data from multiple fisheries that have operated in the southeastern Pacific over the past several decades. We applied a dynamic Poisson Process Model to generate pseudo-absence data and predictions of leatherback intensity as a function of dynamic environmental covariates. Model comparisons were performed based on likelihood ratio tests and prior expectations for leatherback intensity. The chosen model revealed a somewhat circular movement pattern in the southeast pacific, as well as greater leatherback presence in the southern reaches of the study range in December, as opposed to June. It is our hope that this model can be used to reduce future leatherback bycatch. 

Three different sources of oysters, Crassostrea virginica, were settled in three water condition treatments: 5 µm, 100 µm, and unfiltered ambient sea water to determine if larval source or water treatment impacted larval settlement success and/or spat growth. Treatments were chosen based on commonly used hatchery conditions, creating a trade-off between food availability and predators. The larvae were placed in the downwelling system for 8 days to settle in various water conditions. Larvae that set, metamorphosed, and grew were transferred to the upwelling system for 21 days, using ambient water conditions. Comparing 100 µm treatments, there was no statistically significant difference (p>0.05) in settlement or growth between sources. There were significant differences in larval settlement and spat growth between the 5 and 100 µm filtered water treatments. The 5 µm treatment had lower settlement rates than the 100 µm treatment. From Day 8 to 15, the 5 µm treatment grew slower than the 100 µm, indicating a carry-over effect, highlighting that larval rearing conditions determine subsequent spat growth. From days 15 to 31, the 5 µm treatment grew quicker than the 100 µm treatment. This indicates spat grown in a food-limited environment experience a carry-over effect followed by compensatory growth. 

Although species-rich and valuable ecosystems, which offer a vast amount of ecosystem services, tidal freshwater marshes have been increasingly threatened by relative sea-level rise and reduced sediment availability in the past couple of decades. The goal of this study, conducted at a tidal freshwater marsh in the upper Patuxent River (Jug Bay) is to determine how the vegetation community and sediment accretion rates in the marsh have responded historically to these threats. Seeds from Nuphar lutea, Pontederia cordata, Polygonum arifolium, Peltandra virginica, Rosa palustris, Leersia oryzoides, and Hibiscus moscheutos species were recovered from cores; however, evaluating changes in vegetation with time were limited by the number of seeds recovered, especially with depth in cores. Robust correlations between sediment accretion rates and annual mean sea level indicate that flooding patterns drive the delivery of sediment to the marsh surface. For example, greater variability in accretion rates were observed at infrequently flooded high-marsh sites, while lower variability was observed at low-marsh sites, which are more frequently flooded. This study represents a positive approach in better understanding potential future impacts of increased flooding associated with relative sea-level rise in tidal freshwater marshes, while highlighting challenges in establishing seed geochronologies.

Existing ocean color phytoplankton models are not designed to handle the high levels of backscattering and absorbance characteristic to optically complex coastal and estuarine systems, and so are useless in measuring phytoplankton biomass in these environments. An understanding of the behavior of these inherent optical properties is vital in the development of accurate regionally-specific phytoplankton remote sensing models in “Case II” systems like the Chesapeake Bay estuary. This study examined how absorbance and backscattering varied along an “estuarine gradient” in the Chesapeake Bay and one of its freshwater tributaries. We found that both backscattering and absorption decreased from the mouth of the Bay to the headwaters of the freshwater tributary, primarily as a function of depth. In addition, we also made some basic changes to the Quasi-Analytic Algorithm for Optically Deep Waters, an ocean color phytoplankton model designed for use in the open ocean, to improve its performance in the optically complex waters of the Bay. 

Salt marsh morphologies are commonly affected by strong tidal processes that regulate sediment accretion rates. Here, we investigate how the interplay between tides and wind affect sediment dynamics in a low energy salt marsh at the Paul S. Sarbanes Ecosystem Restoration Project at Poplar Island, an active restoration site where fine-grained material dredged from the upper Chesapeake Bay is being used to restore tidal marsh habitat. Tidal currents were measured over multiple tidal cycles in the inlet and tidal creeks of one marsh at Poplar Island, using Acoustic Doppler Current Profilers (ADCP) to estimate water flows and marsh connectivity. Sediment fluxes were determined by estimating suspended sediment concentrations (SSC) with ADCPs and validated against total suspended solids measurements taken on site. Channel morphology influences flood-ebb dominance in marshes, where deep, narrow channels promote high tidal velocities and incision, increasing SSC and reducing resilience. Multiple regression analysis identified a significant relationship between water flows, depth, and SSC at the culvert. Concerns about erosion and sediment export have been highlighted by our study, showing the highest fluxes in SSC coinciding with peak ebb tides. Understanding the drivers of salt marsh morphodynamics is vital for informing restoration practices and designs to improve resilience.

Neomysis americana are an abundant species of key importance in coastal and estuarine communities, linking primary and secondary productivity as an omnivore in the plankton community and as an important food source for many fish species. Yet little is known about their preferential diets within the plankton community. This study utilizes flow cytometry to assess the grazing habits of N. americana at two sites in the Patuxent River, Maryland and at both the surface (both sites) and bottom (one site) of the water column. Several taxonomic groups of phytoplankton consistently occurred in both sites and in the upper and lower water column. Individuals collected from an automated sort of each group indicated these included cyanobacteria, dinoflagellate, diatom, and green algae groups. An overall increase in cell concentrations of cyanobacteria was seen in the majority of the mysid grazing trials from both sites. A decreasing trend in dinoflagellates and diatoms at both sites suggests a preference for these functional groups as forage. Across all three treatment levels, the largest loss of cells was observed for green algae. This study developed a methodology for pairing flow cytometry with live animal grazing experiments, yielding techniques that refine and build on more traditional methods for studying zooplankton grazing. This work also informs future studies on potential avenues for research in mysid grazing, which will ultimately lead to a greater understanding of their role in Chesapeake Bay food webs.

The frequency and severity of low oxygen events—anoxia and hypoxia—have been variable in past decades and have become a particular problem in Chesapeake Bay. Due to anoxia, there can be profound and deleterious effect on the biology and chemistry of the Chesapeake Bay. There are a number of known factors that are correlated with the development of anoxia including the nitrogen loading and freshwater flow into the bay, many of which are controlled anthropogenically. These variables are used to build linear regression models to forecast anoxic volumes. This project improves the forecasting performance of these models through the utilization of a residual bootstrap. In a cross-validation testing, the residual bootstrap approach outperforms the traditional parametric approach by yielding better coverage probabilities and smaller widths of the prediction intervals. Additionally, we detect a change in the relationships between variables, by examining model coefficients in a moving time window. This can provide a better understanding of the factors resulting in anoxia as well as inform a future direction for improvements in forecasting anoxic volume. Furthermore, these more accurate predictions can be used to better inform the public, legislators, and policy-makers concerned about the health of the estuary. 

Tidal marshes are among the most productive ecosystems supporting a wide range of biodiversity, but they are being threatened by anthropogenic and natural causes. Restoration projects must be monitored to understand biogeochemical processes and their evolution towards achieving natural functionality. In Poplar Island, where dredge material with a high N pool was used for its construction, N₂fixation have not been determined and it will serve as an indicator of marsh chronosequence trajectory toward resembling a natural marsh. The acetylene reduction assay was implemented to measure N₂fixation rates in the marsh chronosequence, and to assess the rates in two planting configurations, clumped and dispersed, on two substrates, sand and fine grained dredged material. Preliminary determination of diazotroph organisms was made by adding DCMU to inhibit oxygenic photosynthesis. A positive correlation between N₂fixation rates and marsh chronosequence was found. In the planting design study higher fixation was found in sandy sediments and inside plant clumps during the day, demonstrating the substrate characteristics and planting design importance in the development of these projects to achieve proper morphodynamics and biota interactions leading to efficient functionality. N₂ fixation after oxygenic inhibitor was added increased relatively rapidly indicating a heterotrophic microbial community presence.

Golden tilefish are a demersal, stenothermic species that support a sizable fishery off the Southeast coast of the United States. The current stock assessment model heavily relies upon fishery-dependent data over fishery-independent data, and does not incorporate environmental effects on catchability. We analyzed longline survey data from the fishery-independent MARMAP (Marine Resources Monitoring Assessment and Prediction) program, catch data from the South Atlantic fishery, and publicly available climate data to determine the effects of a variety of environmental variables on catchability and abundance indices. Our results suggest the presence of sex-varying catchability and time-varying catchability, neither of which is accounted for in the current assessment. We also detected delayed declines in abundance indices following anomalously cold years, and identified a potential predictor of abundance in climate data describing the North Atlantic Oscillation. These tentative findings require more data to verify, but suggest that survival of juvenile tilefish is affected by low temperatures and that existing climate data could be used to predict future stock abundance. We recommend that future stock assessments for this species increase the weight of fishery-independent data relative to fishery-dependent data, and begin collecting sex-specific data to support stock assessments.

Despite the increasing importance of aquaculture for oyster restoration, relatively little is known about the genetic implications of using aquaculture-grown oysters for restoration. Nutrient-rich environments, such as those seen in aquaculture, could significantly reduce larval resilience to starvation. To examine the potential for adaptation of larval feeding traits to the hatchery environment, we exposed larvae from both aquaculture and wild oyster populations to starvation and ad libitum (control) conditions and monitored growth, survivorship, and respiration. Both wild and aquaculture groups grew at a rate of 10.849 µm/day and 8.5977 µm/day, respectively. Starved larvae stopped growing during starvation at 114 microns, but resumed growing when food was reintroduced. Survival was reduced for starved aquaculture oysters (8-10 times less than wild larvae by day 25) after starvation, indicating that wild larvae were more tolerant of starvation. Preliminary respiration data, correlated with metabolism, indicated that starved wild larvae exhibited reduced respiratory responses, which eventually surpassed that of the fed treatments, possibly demonstrating compensatory performance. Overall, our findings suggest that wild larvae possess more tolerance to starvation than aquaculture oysters. Future work will examine the underlying physiology of starvation tolerance in wild oysters, and shed light on effects of hatchery propagation on shellfish restoration. 

Chromophoric dissolved organic matter (CDOM) is an important component of the marine dissolved organic matter pool. It absorbs solar radiation and thus effects light attenuation in the ocean, as well as the biogeochemical cycling of elements such as carbon and nitrogen. The marine picocyanobacteria Synechococcus and Prochlorococcus have recently been found to be a contributor to the deep-sea fluorescent dissolved organic matter (FDOM) pool. However, little is known about the composition or structure of these picocyanobacteria-derived fluorophores. Phycocyanobilin (PCB) is a water-soluble tetrapyrrole that functions as the chromophore of the light-harvesting structures in the photosynthetic apparatus of Synechococcus and is a plausible precursor to picocyanobacteria-derived FDOM. In this study we make preliminary measurements and observations of both a standard PCB solution and real Synechococcus culture DOM that will help with structure elucidation. We used photo-degradation experiments to produce the PCB-derived fluorescent photo-products. We implemented excitation-emission matrix spectroscopy to measure fluorescence, high-performance liquid chromatography to separate the fluorophores, and high-resolution mass spectrometry to begin to obtain compositional and structural information about the fluorophores. We found that both the PCB standard and the Syn. culture show similar changes in fluorescence and mass after photo-degradation, confirming that PCB may be a precursor to deep-sea picocyanobacteria-derived FDOM.

Biogeochemical measurements from 1978-2003 were analyzed to develop a carbon and nutrient budget for the Delaware Bay estuary using a box-model. Two different box-models were constructed to calculate nutrient transport and production throughout the estuary. The first consisted of a single box to capture estuary wide net production and transport, the second was comprised of four boxes in the salinity gradient. The models indicated massive amounts of nutrient production, specifically NO3 and NH4 in the estuary. Overall heterotrophy was prominent in the upper regions and shifted to autotrophic tendencies in the lower estuary. Box-model calculations were then compared to geochemistry data in the upper river and box-models of other similar estuaries, including Chesapeake Bay and Roskilde Fjord. With further analysis, the upper regions of the estuary exhibited uncharacteristically higher production of both NO3 and NH4 (almost twenty times higher). Competition between nitrification and remineralization led us to expect production of one nutrient, not both, suggesting the box-model is not accounting for some external inputs. In this case we hypothesized sewage input to be the main contributor. The box-model interpreted these inputs as apparent production in the upper boxes and stoichiometric conversions of net nutrient production to net carbon allowed us to derive an estimate of 17 mmol N m2*day-1 from sewage input. Overall, three major biogeochemical regions were identified. The first region we labeled as the urban river, characterized by ammonium and nitrate net production coinciding with sewage inputs. The second, is the turbidity maximum splitting the first and third region. Finally, the third region was labeled the lower estuary experiencing minimal external input and high productivity.

Because amphibian populations are already being threatened globally for a multitude of reasons, it is necessary to understand how increasing temperature resulting from climate change will impact these ectothermic species. Two anuran species of frog, Cope’s gray tree frogs (Hyla chrysoscelis, Family Hylidae) and southern leopard frogs (Lithobates sphenocephalus, Family Ranidae) were selected to determine the effects of chronically elevated temperatures and quality (protein content) of their diet on their metabolic rate, growth, and development. Metabolic rates of H. chrysoscelis were significantly impacted by diet quality at low temperatures (ANCOVA, p = p = 0.009), but were not significantly affected at higher temperatures. While not significant, in H. chrysoscelis anurans, those fed the high protein diet at low temperatures on average experienced a larger proportional change in weight than those fed the same treatment at 30°C. In H. chrysoscelis, there was no significant difference in developmental rates between individuals fed the high protein diet and those fed the low protein diet for a given temperature. Metabolic rates in L. sphenocephalus were significantly impacted by diet quality at low temperatures (ANCOVA, p = 0.032), but were not significantly affected at higher temperatures. Similar to the results of H. chrysoscelis, at both temperatures the frogs fed the high protein diet were insignificantly younger at the time of forelimb emergence (e.g. developed faster) that those fed a low protein diet. Lithobates sphenocephalus fed the high protein diet at low temperatures experienced an insignificantly larger proportional change in weight than those fed the high protein diet at the higher temperature.

Farmers have increased farm yields to help sustain the growing human population by adding nitrogen (N) and phosphorus (P) fertilizers and ditches to their fields. However, these ditches allow excess water containing N and P to leave farm fields quickly and enter nearby rivers, lakes, and estuaries and cause eutrophication. Drainage control structures is a management practice used to decrease the amount of N and P leaving the fields. My project focuses on how effectively drainage control structures decrease the outflow of water from the field by testing the amount of water flowing around the drainage control structure through groundwater. The groundwater around a free flowing ditch (control) and controlled drainage ditch (treatment) was sampled and measured for water level over the summer to compare the discharge and amount of nitrate (NO₃^-) around each ditch. There was no significant difference in discharge between the two ditches (P< 0.001) However, there was a significant difference in NO₃^- flux between the two ditches (P= 0.019).

Sulfide oxidizing bacteria hold ecological importance in their ability to convert naturally formed, yet toxic, hydrogen sulfide into the inert sulfate in marine sediment. Two sulfide oxidizing bacteria, Beggiatoa and cable bacteria, share common territory in marine sediments; however, not much is known about their interactions and natural distribution. The factors controlling the sulfide oxidizing bacterial distribution are not known, but it is hypothesized that salinity may be one of the determinants. To test this hypothesis, we sampled across a salinity gradient in the Choptank River and into the Chesapeake Bay, Maryland, collecting sediments from sites at conductivities from < 1 mS/cm up to 18 mS/cm. We found that Beggiatoa were at low abundance at all sites except at station on the west side of Chesapeake Bay, indicating that salinity alone does not define its geochemical niche. Enumeration of cable bacteria is still on-going, with preliminary results identifying their presence at the mid-Bay station in Chesapeake Bay. 

The pelagic macroalgae Sargassum is a keystone species and is economically important. Previous models estimating its distribution and growth suggest that the Gulf of Mexico may be a seed region for Sargassum, where enough biomass is retained yearly to maintain a self-sustaining population in the Atlantic. We examined how adding inertia to a particle model could influence the retention rate and potential growth of modeled Sargassum within the Gulf of Mexico to determine its effectiveness as a seed region. We determined that adding inertia affected particles near regions of high divergence or high convergence more than the particles in more quiescent areas. We also concluded that there was a westward trend of particles when inertia was added which led to significantly fewer particles leaving the Gulf. Finally, we found that the potential for Sargassum growth is likely strongest near regions of modest convergence and divergence.

In an effort to reverse the effect of eutrophication, average nutrient inputs are being reduced in highly dynamic coastal shallow ecosystems; however, little is known about the response trajectories following these reductions. Long-term data (tide, nutrients, community respiration and hydrographic data) from the Chesapeake Biological Laboratory’s (CBL) monitoring program and a ~daily 10-day sampling, of the same measurements, were analyzed to understand the dynamics of the CBL pier ecosystem. A Water Column Model (WCM) was then constructed and coupled to an existing Sediment Flux Model (SFM) to simulate our ecosystem’s response to nutrient load reductions. A strong seasonal correlation (R² = 0.7149) between community respiration and particulate nitrogen was observed as well as a dominance of nitrate in the early spring and a dominance of particulate nitrogen in the summer (as NO₃ decreased). Daily variability for particulate nitrogen and algal N were generally low while high variability in community respiration seemed correlated with mean sea level (tide). Preliminary model runs showed significant differences in annual data and model dynamics. These results suggest that seasonal shift in the dominance of different nitrogen species reflect nitrogen inputs from the watersheds in winter-spring and accumulated algal material in the summer. Seasonal community respiration seems to be driven by the amount of organic material in the ecosystem while ~daily variability seems to be driven by the lability of the organic matter available as the tides changes. Finally, future improvements to the Shallow-Water Ecosystem Model are necessary for better simulations dynamics. 

Organic matter in highly eutrophic estuaries goes through several degradation steps before being buried in the sediments. These steps are carried out by microbes that live in both the water column and the sediments. Understanding the breakdown of this organic matter is important, as it produces carbon dioxide and ultimately, methane, which are powerful greenhouse gases. The objective of this study was to characterize the organic matter that is available to the sulfate reducing and methane producing bacteria. Doing this would allow us to see how it changes in relation to the activity of sulfate reducers and methane producers and if there are differences in how it is transformed through different biogeochemical zones. This was done by first determining biogeochemical zonation in the Chesapeake Bay through down core changes in concentrations of sulfate, sulfide, methane, chloride, and dissolved organic carbon. Pore water was then analyzed for fluorescent dissolved organic matter components using excitation emission matrix analysis and statistical PARAFAC analysis. Three statistically different dissolved organic matter components were confirmed to be in the pore water. Changes in fluorescence of each component through the biogeochemical zones was observed, and thus it is possibility that it is utilized differently within these zones.