Our Students and Their Research

Ocean City’s inlet is a conduit for nutrients and pollutants between the Atlantic Ocean and Maryland’s coastal bays. This study investigated the response of phytoplankton to increased nutrient availability and temperature as part of a larger investigation of phytoplankton dynamics, in relation to eutrophication and climate change effects. Three nutrient bioassays were conducted on phytoplankton biomass in the Ocean City Inlet, with additions of 10_μM nitrate, 10_μM ammonium, 5_μM phosphate, and 10_μM Nitrate+5_μM Phosphate. The first two bioassays were subjected to natural light and temperature while bioassay 3 was conducted in two temperature control rooms averaging 22.1C and 36.5C. Chlorophyll a and phycocyanin were measured using a handheld fluorometer; Chlorophyll a was extracted to standardize the in-situ measurements. All bioassays showed a strong positive increase in phytoplankton biomass with nitrogen treatments, supporting the hypothesis that nitrogen is the major limiting nutrient for marine phytoplankton. There was no significant response with phosphate additions alone, however, N+P additions were highest in bioassay 1, suggesting co-limitation of phosphorus when excess nitrogen is supplied. There was no significant difference between nitrate and ammonium treatments. Phytoplankton biomass did not increase with temperature. However, temperature had synergistic effects on nutrient intake. This data, as part of the larger study, will allow for improved management and preservation of coastal ecosystems.

The diamondback terrapin (Malaclemys terrapin), is an estuarine turtle found within brackish marshes of the eastern and gulf coasts of the United States and it is the official state reptile of Maryland. The diamondback terrapin unfortunately may become a victim of increasing global temperatures due to its temperature-dependent sex determination (TSD). Species that display TSD may produce more female than male hatchlings or vice versa depending on the temperature of the individual nest. This study used several individual nest temperatures along with the air temperatures at the nesting site, recorded in the summer of 2018, to select a predictive model with R software package rstanarm, to predict nest temperatures using known air temperatures. The regression spline mixed model was selected for having the lowest leave one out information criterion. Its accuracy was determined by using air temperatures recorded during the summer of 2019 to predict 2019 nest temperatures, and comparing the predictions to field recorded values. The RSMM had a predictive capability of 83.6-98.7% coverage of five actual nest temperatures recorded in 2019, with a predictive interval of 95%. The resulting model will inform a collaborative project evaluating climatic influence on individuals and populations of diamondback terrapins.

Due to global climate change, Talbot and Dorchester County on the Chesapeake Bay are experiencing some of the fastest rates of sea-level rise (SLR) in the country. Historically, communities have used grey engineering to protect their shorelines from erosion and storms, such as breakwaters, seawalls, or rip-rap. However, over time, these will lose effectiveness without major upgrades. Additionally, these structures provide poor habitat for aquatic species. This project seeks to create effective hybrid coastal infrastructure by retrofitting grey infrastructure with oysters using oyster castles. Oysters are ecosystem engineers, offering both protective and non-protective benefits to coastal communities. This study focused on filling a knowledge gap and building the foundation of a long-term BACI-designed ecosystem project, to measure the impact of installing oyster castles. We constructed model-informed oyster castle breakwaters at Horn Point Laboratory with the aim of comparing both the protective and non-protective benefits of seeded and unseeded breakwaters by monitoring wave dampening effects, shoreline response, biodiversity, and oyster growth and accretion. Focusing mainly on the ‘before’ part of the BACI design, we took the first steps necessary in understanding the use of oyster castles in Maryland, building a foundation of knowledge.

Mixotrophy is a nutrient acquisition method that combines photosynthesis and heterotrophy. Although mixotrophy is known to be prevalent throughout aquatic microbial ecosystems, mixotrophy has been poorly represented in current marine food web models. The ability to quantify carbon and nutrient acquisition associated with mixotrophy would be a major step forward in the attempt to better understand the behavior. Heterocapsa rotundata is a dinoflagellate species indigenous to the Chesapeake Bay that has been confirmed to exhibit mixotrophy and begins to graze on bacteria when light levels are low. Heterocapsa rotundata is a prominent species in Chesapeake Bay during the winter season. The objective of this study was to measure mixotrophic grazing in a culture of Heterocapsa rotundata isolated from the Choptank River, across a gradient of light levels. To measure ingestion rates, H. rotundata were allowed to feed on fluorescent microspheres for a brief incubation period. Ingestion rates were calculated as the rate of fluorescent beads engulfed by H. rotundata, as observed using fluorescence microscopy. The study demonstrated a significant negative relationship between light level and ingestion in indigenous Heterocapsa rotundata.

The common bottlenose dolphin, Tursiops truncatus, is a highly social and vocal species that uses a variety of acoustic signals to aid communication, foraging, and orientation. Bottlenose dolphins have the ability to convey identity information via signature whistles—an individually specific whistle that is unique to individual dolphins. These bottlenose dolphin signature whistles play a similar role in identity communication as unique names for individual humans. Based on a bout analysis method of signature whistle identification, we identified signature whistles in audio recordings collected via passive acoustic monitoring between July 2016 and October 2017 off Ocean City, Maryland, U.S.A., within a proposed wind energy area. Through signature whistle analysis we estimated the minimum abundance of dolphins within our detection area as 174 dolphins with the highest number in summer 2016 and the most frequent re-occurrence of individuals during winter 2017. These results can be used to inform assessments of the potential exposure of bottlenose dolphins to high levels of noise associated with offshore wind farm construction.

In the United States, oysters are induced to settle using either whole oyster shells or ground oyster shells (“cultch”). Settlement on cultch is used in aquaculture to provide single oysters for market. However, a common method used in European countries is to induce settlement using the hormone epinephrine. This study compared the settlement success of triploid Crassostrea virginica larvae of the Lola strain using cultch versus epinephrine as an inducer of metamorphosis and settlement in oysters. Specifically, the study was concerned with how epinephrine affected settlement success, as well and the short- and long-term growth of the oysters. To determine these effects, the larvae were raised for 8–13 days in a downwelling system, then the resulting spat raised for several weeks in an upwelling system. An additional treatment was also employed—the effects of using ambient, unfiltered water from the Patuxent River versus water filtered using a 5µm mesh screenduring the settlement and metamorphosis phases of the oysters’ life cycle. The effects of these treatments were examined in conjunction with the effects of the two inducers used to mimic a “good” versus a “poor” environmental condition (i.e. high and low food availability). In both treatments, exposure to epinephrine led to higher settlement success than cultch. Mortality was lower for the epinephrine-treated oysters than cultch, and growth in the first 8 days was similar for epinephrine and cultch oysters in the ambient treatment, but higher for the epinephrine oysters in the filtered treatment. Long-term growth was similar among all treatments and inducers. The filtered treatment led to lower settlement success and higher mortality than the ambient treatment in the first 8 days.

Oyster aquaculture has found success in using a variety of oyster strains which have been developed to perform in specific environmental conditions. Strains differ in areas such as disease resistance, salinity tolerance, and growth rates. Although differences in growth rates among strains has been observed, the physiological cause of the growth differences remains unknown. The goal of this study was to investigate and begin to quantify the bioenergetic budget of oysters, specifically the relationships among environmental variables, respiration rate and growth. It was predicted that the faster-growing strain would have the lowest respiration rate and therefore more energy leftover to dedicate towards growth. This study utilized microrespirometry methods using Firesting O₂ fiber-optic oxygen probes to measure changes in oxygen concentration within the vials. Trials were run at various temperature and salinity combinations chosen to reflect the variation typical of the Chesapeake Bay. Changes in oxygen concentrations were converted to respiration rates and data was analyzed using response surface methodology. ANOVA results indicated that strain effects were not significant (F=1.0122). Some aspects of the results supported the hypothesis and other aspects were in conflict with the hypothesis such that an overall conclusion is not possible and more work needs to be done in order to understand the bioenergetics of oysters.

Studies have shown that thawing permafrost has the potential to release significant amounts of greenhouse gases directly to the atmosphere. Yet, some areas of the world with continuous permafrost-containing mineral soils have not been characterized and need to be constrained to gain a global picture of the potential impact of thawing permafrost on global warming. In this study we report on recent cores drilled from terrestrial permafrost, as well as permafrost underlying shallow marine waters, in Tuktoyaktuk Island, Northwest Territories, Canada. Material was subsampled from ~1m increments and measured for concentrations of dissolved methane. Methane concentrations ranged from 0.03 to 0.13 mg CH₄per kg soil, with a peak in concentration at 2 m, just below the active layer. These concentrations are similar to other mineral soil permafrost cores. Material from the active layer and 2.8 m below the surface was incubated for ~6 weeks under aerobic and anaerobic conditions at -20°C, -5°C, and +15°C to quantify the amount of methane and carbon dioxide being released from microbial activity. These biotic treatments were compared to killed controls to determine if any abiotic processes may be contributing to methane and carbon dioxide release. Under anaerobic conditions, methane concentrations increased over time at all temperatures compared to the killed controls, suggesting microbial production. In addition, methane concentrations increased faster under warmer conditions, but we did measure appreciable amounts of methane in the -20°C treatment. This paper reports on both the aerobic and anaerobic incubation results within the broader context of the downcore characterization work. It will also discuss how the results fit into the larger framework of permafrost incubation experiments.

The Chesapeake Bay and its tidal tributaries experience significant bottom water hypoxia during the summer months. This study investigates the effects of a field of concrete reef balls in the Severn river, a tidal tributary of the Chesapeake Bay, on the overlying water column, and whether these effects have any significant impact on disruption of the summer pycnocline and reaeration of hypoxic bottom waters. Current velocities, turbulence, and several indicators of water quality were measured at the study site using an Acoustic Doppler Current Profiler and a Yellow Springs Instrument (YSI) Exo1 probe. These instruments were deployed for a 16-day period before reef balls were placed at the site. Following placement of the reef balls the instruments were deployed at the site for another 16 days. Comparison of these two data sets revealed a 5.62% decrease in time averaged velocity and a 28.26% increase in turbulence intensity following the placement of the reef balls. However, wind influence was detected to a depth of six meters. Considering only the portion of the water column not influenced by wind, there was a 7.49% decrease in time averaged velocity and 2.63% increase in turbulence intensity. Although, the increase in turbulence intensity in the lower water column was minimal, and dissolved oxygen concentrations at depth were not improved, the results have enough significance to conclude that further investigation is warranted.

Plastic degrades overtime into smaller particles of plastic known as microplastics through a process of fragmentation. The development of microplastics has been studied in a variety of contexts, such as salt marshes and the open ocean. However, little is known about the factors that influence plastic degradation and microplastic production in coastal environments, even though the majority of the plastic that ends up in the ocean must pass through coastal environments on its way from land to ocean. To understand this process, this experiment takes the experimental design from Weinstein et al. (2016) and adapts it to coastal processes such as tides, erosion and deposition, wave action, biomass development, and other coastal factors. As a result, strips of high-density polyethylene plastic and polystyrene plastic were field deployed in a variety of locations corresponding to critical variables such as erosion, deposition, and subtidal and intertidal influences. These strips were collected at four weeks, eight weeks, sixteen weeks, and thirty-two weeks from the start of the experiment and were analyzed for mass loss, biomass development, microplastic production, and fragmentation. The experiment remains in progress, but the four-week results show the development of biofilm and no significant plastic strip mass loss, as expected. The results also show large scale macroplastic fragmentation occurring in the subtidal polystyrene strips in the erosion zone. 

Coastal waterways receive fecal pollution from many sources each year in varying proportions around the world. Waters that have been polluted with storm water and urban runoff, sewage treatment plants, combined sewage overflows and sanitary sewer overflows can contain human pathogens. Methods that use fecal indicator bacteria (FIB), including E. coli and Enterococci, have traditionally been used to assess the human pathogen risk in waterways. However, these culture methods can be unreliable for several reasons: (1) FIB presence does not necessarily indicate pathogen presence, (2) they can appear from sources other than humans, and (3) sediments help FIB to persist and survive over the long term. Molecular methods are being developed that are more reliable in pointing to sources and pathogen presence. The World Harbour Project compares these two methods in twelve harbor systems around the world. Baltimore Harbor is one of the sites. We measured FIB and water quality parameters in eight sites that reflect more to less impacted in June and July and correlated them. Significantly higher FIB was present in a wet period sampling compared to a dry and there was also a gradient of FIB numbers from more to less impacted sites. This was also reflected in the relationship with salinity, with higher numbers of FIB in lower salinity (i.e. indicative of higher rainfall and also more impacted inner harbor sites). Additionally, a positive correlation of FIB numbers with total nitrogen was determined, but generally inverse correlations with all other water quality parameters. A longer time scale of sampling will be necessary to make more quantitative assessments. Samples are also being analyzed using next generation Illumina sequencing, and these molecular results will be analyzed across the World Harbour Project cities to help determine whether molecular techniques are more informative in predicting actual human pathogens rather than traditional FIB culturing methods and whether new methods could be implemented on a broader global scale to improve human health impacts associated with the fecal pollution problem worldwide. 

Sargassum is a pelagic macroalgae that accumulates in the warm Mid-Atlantic providing shelter, food, and a breeding ground to over 60 marine species. Sargassum can export about 43% of its production both as particulate organic carbon (POC) and dissolved organic carbon (DOC). It is also known to sequester heavy metals, such as mercury. Due to its vast coverage and high activity, Sargassum mats seem like an ideal place for production of methylmercury (MeHg), the only mercury to bioaccumulate in fish and humans. After analysis of the values of MeHg, T-Hg, and optical properties present inside and outside Sargassum mats in the Gulf Stream and in Puerto Rico, a connection between the quantity of dissolved organic matter (DOM) present and T-Hg being methylated can be made. DOM can act as a mercury carrier and is released the most when Sargassum degrades. Sargassum does contain mercury (3.92 ng/g wet wt) and methylmercury (0.011 ng g wet wt). We found that Sargassum beds release Hg (2.62 ng/L) and MeHg (1.03 ng/L) into the ocean water, concentrations much higher than the Hg (0.67 ng/L) and MeHg (0.02 ng/L) seen in the open ocean. The more DOM present correlates with the more MeHg produced. High methylation is often observed when the Sargassum is stressed, such as in coves, where microbial activity changes as DOM export from the Sargassum increases. In the waters of the Laguna Grande, Puerto Rico, a lagoon impacted by a Sargassum invasion, the input of fresh DOM from Sargassum resulted in low oxygen conditions and Hg and MeHg concentrations in waters outside Sargassum mats ranging from 0.02 to 0.7 and 0.02 to 0.4 ng/L, respectively. This suggests Sargassum can have significant local impacts on water quality. Laboratory simulations of these low oxygen conditions yielded concentrations of 6.43 ng/L T-Hg and 4.47 ng/L MeHg. Further research can be done to look at the type of bacteria responsible for methylation in Sargassum mats as well as where the MeHg attaches itself on the DOM.

We examined photo-ammonium production from dissolved organic matter (DOM) in three different sites. Photoproduced ammonium has been found to be an important source of nutrients in nitrogen-limited waters, but this process is not well understood. In this study, samples were taken from St. Mary’s Lake and the Patuxent River in Maryland and three different landfill leachates in Florida—active, closed reverse osmosis brine (brine), and closed landfill leachate. These samples were chosen to reflect a range of conditions to allow us to compare photoammonification across different sites. All samples were filtered through glass microfiber filters (Whatman GF/F) and solid phase extraction was done on all samples. Photo-degradation experiments were conducted on both filtered samples and extracts. Fluorescence excitation emission matrices (EEMs) were collected every twenty minutes on an Aqualog fluorometer and irradiation lasted either twenty or twenty-eight hours for all samples. Significant ammonium photo-production was observed in brine landfill leachate solid phase extract and St. Mary’s Lake filtered sample and solid phase extract. Significant ammonium loss was observed in active and closed landfill leachate, which may have been due to outgassing of ammonia during the experiment. Photoammonifcation rates were strongly correlated with total dissolved nitrogen (TDN) concentrations. Samples with higher TDN tended to have higher rates of photo-ammonium loss. Parallel factor analysis (PARAFAC) was used to analyze EEM data. This is the first time that photoammonification has been studied in any of these sites.

The Susquehanna River is the Chesapeake Bay’s largest tributary, providing half of the bay’s freshwater and 2 x 10⁶ tons of sediment on average annually. This discharge passes over the Susquehanna Flats at the mouth of the river, the most upstream part of the bay. Submersed Aquatic Vegetation (SAV) presence here was found to have an important influence on water quality as it initiates a feedback loop that decreases turbidity and eutrophication, improving the health of ecosystems in the area. Additionally, this feedback loop adds to the strength and resilience of the beds during flood events. Previous research was conducted on the larger, more permanent beds in the center of the flats; we identified smaller, periphery beds for this study. We gathered push-core samples inside and outside of the perimeters of fifteen beds before the peak of SAV growing season to study sedimentation patterns and composition. We used Beryllium-7 activity, grain size, and nutrient composition to infer sedimentation patterns of the ephemeral beds in the Flats. We found that there is preferential sedimentation to the southeast of the Flats, and that the geomorphology of the beds plays an important role in deposition patterns when SAV is not present.

Microplastic pollution has become a major concern in today’s environmentally aware society. It is now a worldwide problem that has been documented extensively in scientific journal articles and also in the mainstream media. The long-term goal of our study is to analyze different sediment samples taken from Baltimore Harbor to determine the type and amount of microplastics found. Sediments collected in the 1970s and 1980s will be used along with modern samples to determine the extent of microplastic pollution over the years. This paper focuses on developing the best method to extract microplastic particles from the sediments. The samples were first disaggregated using a Calgon solution with a rotary shaker and sonicator. Next, the larger particles were removed using 500 and 106 μm mesh sieves. Dissolved organic matter was removed by adding hydrogen peroxide. Zinc chloride solution was then used in a density separation device to separate the microplastics from the sediment through flotation and filtration. The method was applied to four samples in order to test and, if possible, optimize each individual step.

Ocean acidification disrupts the carbonate chemistry of marine ecosystems and threatens calcifying organisms, which then harms coastal communities that depend on them. There is a lack of understanding of the effects of ocean acidification in estuaries due to their complexity and the minimal spatial and temporal monitoring being conducted. Because high-quality commercial CO₂ sensors are very expensive, monitoring at greater frequencies is a costly and difficult task that is never accomplished. Acknowledging the need to understand the carbonate chemistry, the objective of this project was to design and build a deployable autonomous CO₂sensor (DACS) at a price of under a thousand dollars. The study consisted of the construction of a DACS followed by verification of the product through field testing and laboratory analysis. The laboratory analysis consisted of a collection of water samples and measurement of total alkalinity and dissolved oxygen to calculate pCO₂. The construction of the DACS was a success after several design iterations. The DACS CO₂ measurements were verified with the laboratory results. While the DACS and laboratory monitoring of Horn Point Laboratory (HLP) hatchery water data exhibited similar trends, results somewhat diverged indicating the need for further development of the DACS to improve accuracy. Despite inaccuracy, optimizing and verifying the DACS performance will continue, as development of an inexpensive and reliable CO₂ sensor is imperative for future ocean acidification research. 

The Anacostia River, “the forgotten river,” has an overall poor quality caused by several anthropogenic sources including sewage, excess nutrients from runoff, and combined sewer overflows (CSOs). To assess the impact of changing nutrient concentrations and forms on the bacterial and phytoplankton community composition and productivity, samples were collected from the Anacostia River into fifteen cubitainers with different nutrient additions (different combinations of +N, +P, +Si). Our project focused on one aspect: the influence of different nutrient additions on urease activity, an enzyme responsible for the breakdown of urea to ammonium. We hypothesized that there would be more urease activity in the +NO₃^- and +urea additions because NH₄⁺ suppresses urease activity and there will be no difference in +P treatments because urease activity does not require ATP. Urease activity increased in the NH₄⁺ treatments once NH₄⁺ was exhausted and the microbes were physiologically stressed. Urease activity and NO₃^- concentrations remained mostly steady throughout the experiment for the other treatments, with the exception of in +P treatments where urease activity varied even though PO₄ concentrations decreased on specific days. Based on these results, changing nutrient dynamics in the Anacostia River will have an influence on urease activity if ammonium levels are reduced due to the new storage tunnel that went on-line in March 2018. 

Phytoplankton are important photosynthetic microorganisms that support aquatic food webs and influence the physical and biogeochemical properties of aquatic ecosystems. This project examined how nitrogen, phosphorus, and light regulated phytoplankton growth rates during summer months in the Chesapeake Bay. Water samples from a mesohaline monitoring station and the Choptank River were collected monthly, incubated, and analyzed to assess how phytoplankton respond to experimental enrichment of nitrogen and phosphorus by using both flow cytometry and chlorophyll concentrations. Growth rates between 0 and 1.0 day^-1 in the Bay and 0.5 and 2.0 day^-1 in the Choptank River were observed. Comparing growth between treatments showed no statistically significant differences, suggesting light limitation. However, large within treatment variations revealed a need for more rigorous methodology moving forward. As growth rates are measured into the future, Chesapeake Bay models can be evaluated for accuracy, thus improving predictions of nutrient management plans seeking to improve water quality and ecosystem services.

The goal of this research was to statistically downscale and bias-correct several climate variables from the most recent Community Climate Systems Model (CCSM) to the American Mid-Atlantic and Northeastern region, encapsulating the Chesapeake Bay watershed. The downscaling and bias-correction of a global climate model (GCM) increases the spatial resolution and reduces some of its biases, making the model more useful at the regional scale. The Coupled Model Intercomparison Project (CMIP) 5 CCSM4 Representative Concentration Pathway (RCP) 8.5 GCM was downscaled and bias-corrected with the intention of enabling the projection of hypoxia in the Chesapeake Bay from 2081 to 2100. Although the main cause of hypoxia is nutrient loading (Li et. al. 2016), climate change is also a factor that influences the timing and extent of hypoxic zones in the Chesapeake Bay. Specifically, air temperature has been shown to be the most influential climate variable in its effect on hypoxia (Altieri & Gedan 2015). The results project an increase in mean temperature, air pressure, and downward longwave radiation flux (DLWRF) and a decrease in relative humidity between the historical period (1986-2005) and the future period (2081-2100). These climate variables, including their seasonal and geographical focuses, can have varying impacts on the timing and extent of hypoxia, thus deserving further study.