Our Students and Their Research

Weighted flow networks are structures that arise naturally when analyzing complex systems. The countable properties of unweighted networks are not easily generalized to weighted networks. One candidate measure of complexity is the number of roles, or specialized functions in a network. It is easy to identify the number of roles in a linear or cyclic unweighted network. There is only one logically consistent way to generalize the measures of nodes, flows, connectivity, and roles into weighted networks, and these generalizations are equivalent to indices derived from information theory and used by ecologists since the late seventies. Data from ecosystem networks suggests that ecosystems inhabit a narrow window of the parameter space defined by these measures.

To fully understand interannual variations in the sizes of fish populations, the relationship between abiotic (density-independent) and biotic (density-dependent) factors must be analyzed. Moreover, particular attention must be given to the influences of physical factors in the field, and how it influences fish behavior, and ultimately larval recruitment. Small-scale turbulence in the field has been found to increase prey encounter with planktonic predators. Feeding success, however, is limited to a certain threshold of turbulence for visual predators, where the probability of successful seizure of prey eventually decreases with increasing turbulence. The effects of turbulence on non-visual predators is not investigated as fully as the effects of turbulence on visual predators, but is a worthy topic to completely understand microscale turbulence effects on an entire food web. The purpose of this investigation is to analyze the effects of varying amounts of small-scale turbulence on the feeding success of the non-visual ctenophore predator, Mnemiopsis leidyi. Small ctenophores were exposed to varying amounts of turbulence in a novel turbulence chamber apparatus while being fed an identical Artemia salinia nauplii concentration for each turbulence condition. Ingestion rates were analyzed as the mean naupllii/gut, and standard error bars were used to compare ingestion rates among all tested turbulences (0 mm/s- 26.2 mm/s). It was observed that no significant effect of turbulence on ingestion rate existed, as a non-linear relationship was discovered. This counters expectations as non-visual predators were proposed to encounter continually increasing feeding success with increasing turbulence due to ingestion based on contact, not visual pursuit. Further turbulence investigations concerning other non-visual predators should be undertaken to confirm these findings.

Bacteria incubations carried out using samples collected from chlorophyll max., oxic/anoxic interface, and marsh sites in the Delaware and Chesapeake Bays were run to examine breakdown and consumption of the protein BSA, as well as total DOC consumption. Experimental and control incubations were run simultaneously, and 20.0mg of BSA was added to experimental incubations. Protein measurements were made using NanoOrange Protein Assay (Molecular Probes; Eugene, Oregon). Station 1 incubations (chlorophyll max.) were found to consume the most protein at the fastest rate. Station 2 (oxic/anoxic) incubations showed lower rates and total consumption, and station 3 (marsh) was observed to have the slowest rate of degradation. Bacteria populations in experimental incubations were higher than in controls. Overall, rates of BSA degradation were similar for all three sites (~ 4% h^-1), and may indicate excessive run times. Station 3 (marsh) samples were found to have high DOC content which could represent an abundance of terrigenous compounds such as humic substances, and may have affected results. The use of NanoOrange as a seawater protein assay reinforced expected trends found in this study. However as a relatively new protein assay, its abilities to accurately measure small amounts of protein using the method we employed is still uncertain.

The rate of predation and feeding behavior of the blue crab Callinectes sapidus on the soft-shelled clam Mya arenaria, in relation to the presence of artificial submerged aquatic vegetation (SAV) was examined in 1 m² raceway tanks at the Academy of Natural Sciences Estuarine Research Center in St. Leonard, Maryland. Two types of experiments were conducted. The first experiment tested whether individual crabs would exhibit different rates of predation in single habitat tanks, where the habitat is defined by zero, low, medium, and high SAV density. The second experiment gave each individual crab a choice of three habitats: zero, medium, and high density SAV. Crabs participating in this experiment were video monitored to observe their feeding efficiency by quantifying the proportion of time each crab spent in each habitat compared to the amount of clams consumed in each habitat. The results of the first experiment showed that when not given a choice of habitat, crabs do not display significantly different predation rates between habitats, indicating that clams were not benefiting from the potential refuge created by higher densities of SAV. However, when given a choice of habitat, the crabs showed preference for feeding where SAV was absent over medium and high SAV habitats, but did not display significant differences of predation between the medium and high-density habitats. This suggests that the clams are getting protection from the presence of SAV, but may not get enhanced protection as the density of SAV increases. The results of the video monitoring showed highly variable feeding efficiency and behavior between individuals.

Epiphytes are considered detrimental to seagrasses as they reduce the amount of light, i.e. photosynthetically available radiation (PAR) that reaches the plant surface. This study evaluated the possibility that epiphytes can also be beneficial to seagrasses by reducing the amount of UV-B that reaches seagrass leaves. As UV-B may not only negatively affect seagrasses but also their epiphytes, possibly eliminating a positive filtering effect, artificial seagrass leaves made out of a UV-B transparent material were deployed at Horn Point Cove (Ruppia maritime and high nutrients) and Chincoteague Bay (Zostera marina and low nutrients) under a UV-B screen as well as under natural conditions (unscreened). At days 1, 3, 5, 7 and 12 after deployment, the leaves were collected and transmittance of UV and PAR quantified. No significant difference was found in the epiphytic load of UV-B-screened and unscreened leaves, suggesting that the present UV-B levels are not detrimental to epiphytic growth on seagrasses. In contrast, epiphytes transmitted a significantly lower amount of radiation in the UV-B range than in the PAR range. Therefore, epiphytic layers are effective UV-B filters on seagrass leaves. This benefit of epiphytes is lost when PAR transmission is reduced to levels below the compensation point.

During the course of two months, a device was created and field-tested which continually samples atmospheric air for concentrations of ammonia and ammonium. Though still being refined, this sensor reliably measured nitrogen species during a variety of conditions: urban to rural settings, calm days to thunderstorms, dangerously high levels of ammonia/ammonium to next to nothing, and even while offshore in the Chesapeake Bay and Baltimore's Back River. Continual testing ranged in time from several hours to just under a week. All field-testing was carried out within the confines of the state of Maryland because it provided no death of infrastructural settings. Within the next several months this real-time sensor will be used as the primary data collection tool in a major study to model the fate and transport of ammonia/ammonium as a result of the changing agricultural trends on Maryland's eastern shore.

Sediment toxicity is a major area of current research in the Chesapeake Bay region. Much research has focused on the lethal effects of contaminants on benthic organisms. Sub-lethal effects of such contaminants, however, also need to be taken into consideration. To begin to understand the effects of these contaminants on a population of organisms, a long-term, multi-generational model named BAMM, Benthic Amphipod Multi-generational Model, was employed to focus on the sub-lethal effects of sediment contaminants on the amphipod Leptocheirus plumulosus.

The dinoflagellate, Pfiesteria piscicida, was exposed to varying levels of shear comparable to those found in its natural environment. The shear rates tested, produced by Couette devices, were 0, 1, 3, and 10 s^-1. P. piscicida and its in-culture prey, Storeatula major, were put in Couette devices at known concentrations and, at 24 hour intervals over 4 days, samples were withdrawn to measure the effect of shear on P. piscicida's growth and ability to graze on S. major. Additional experiments were conducted to determine how shear affected S. major, and how shear would affect a starved culture of P. piscicida. It was found that below 1 s^-1, shear was not significant on Pfiesteria growth. However, at 3 and 10 s^-1, shear significantly affected Pfiesteria growth. In the supplemental experiment testing the feeding rates of Pfiesteria after an hour of exposure to shear, the proportion of Pfiesteria with orange fluorescent inclusions (feeding cells) was dependent on the level of shear the specific treatment received.

The ingestion rates, clearance rates and selectivity of Neomysis americana on copepods, rotifers and phytoplankton were of key investigation. Since there was no significant difference in mysid length between all experiments mysid predation was not a result of various sizes but rather various prey. Mysids did prey on Thalassiosira weissflogii but it was so insignificant further consideration was extraneous. The functional response of N. americana on copepods was linear suggesting no saturation, whereas the function response of mysids on rotifers became logarithmic at concentrations exceeding 300 rotifers/liter. Ingestion rates significantly increased when copepods were offered in the presence of rotifers when compared to copepods offered alone; the same trend was seen for clearance rates. Rotifers offered in the presence of copepods showed no significant differences between rotifers offered alone. Clearance rates did, however, show a significant decrease suggesting that N. americana does prefer copepods to rotifers when presented with a choice. The presence of rotifers may be interfering with copepod escape and defense tactics resulting in greater mysid predation. Copepods may also be of greater nutritional value than rotifers to mysids. This investigation will act as a precursor for future studies applying these results in a non-turbid environment to the turbid waters of the Chesapeake Bay Estuary.

In recent decades, the populations of SAV in the Chesapeake Bay have decreased in number. A large factor in this decline has been the amount of nutrients deposited in the Bay by human sources, which in turn cause a buildup of algae that blocks light to the plants. In an attempt to determine how much epiphytic algae a plant can tolerate, artificial substrates were used to mimic actual plants. The first experiments were arranged to find whether an increase in nutrients resulted in higher levels of epiphytic growth. We found a correlation in the spring, but not in the summer. The second experiment compared epiphytic growth on the artificial substrate to the grassbed density it was placed in. We found that as bed density increased, epiphytic growth decreased. The ultimate goal is to determine at what level of epiphytic growth plants can survive. The reasoning behind the difference between spring and summer is a combination of factors, including differences in grazing, precipitation, temperature, and the source of nutrients.

It was hypothesized that desiccation will cause mortality in Crassostrea virginica (Eastern oyster) spat; small oyster spat are more susceptible to desiccation than larger spat; misting with ambient Choptank river water will reduce spat mortality; covering spat (thus reducing exposure to solar radiation) will decrease mortality; and decreased temperature/increased humidity will decrease mortality. These hypotheses were tested by subjecting spat of varying sizes and known ages to desiccation under various stressors. These included: direct exposure to atmospheric ambient weather conditions; exposure to the atmosphere while covered by a tarp thus preventing exposure to solar radiation; and exposure to ambient conditions with lessoned desiccation by providing a constant mist of Choptank River water flowing over the spat. The exposure times for each stressor ranged from 2-48 hrs. During the entire exposure time data loggers recorded the temperature, humidity, and solar radiation of the ambient conditions the spat were being exposed to. The results showed that desiccation did cause mortality in spat, and that the larger the individual spat, the longer it could to survive under any stressor. Positive correlations between decreased solar radiation, increased humidity, and decreased temperature were found. The spat positioned under the tarp faired slightly better than those under the mist of river water and both faired significantly better than those exposed to ambient conditions.
It was hypothesized that the mortality rate of Crassostrea virginica spat exposed to ambient conditions will decrease as their size increases and as their exposure time decreases was tested by subjecting spat of varying sizes and known ages to desiccation under various stressors. These stressors included direct exposure to ambient weather (sunny) conditions; exposure to the atmosphere while covered by a tarp thus preventing exposure to solar radiation; and exposure to ambient conditions with lessoned desiccation by providing a constant mist of Choptank River water flowing over the spat. The exposure times for each stressor ranged from 2-48 hrs. These stressors determined if various other factors (solar radiation, time of day, humidity, temperature, and a constant mist of sea water as compared to complete desiccation) would have any effect on the survival rate of the spat once removed from their natural environment. The results showed that the larger the individual spat, the longer it could to survive under any stressor. Positive correlations between decreased solar radiation, increased humidity, and decreased temperature were found. It was also determined that time of day did play a significant role in spat survival. The spat positioned under the tarp faired slightly better than those under the mist of river water and both faired significantly better than those exposed to ambient conditions.

The soft bottom coastal sediments of the Chesapeake Bay are inhabited by several species of macroinvertebrates. The geochemistry of the sediment is affected by the activities of the fauna within the sediment. It is important to understand how organisms affect the habitat in which they live inorder to understand how other species will be affected. The feeding behavior of a tentilid clam, Macoma balthica, in varying population densities was studied in laboratory conditions. The effects on nutrient flux, pore water, and primary production were analyzed in relation to population density.

The bioaccumulation of Hg, as methylmercury (MMHg), can reach toxic levels in top predatory fish that can be unsafe for human consumption. Over 40 states have fishing advisories due to the health risks associated with Hg. This research was completed to determine the HgT concentration in top predatory fish of MD reservoirs and investigate the relationships between Hg concentration and size. Large mouth bass and black crappie and/or bluegill species of fish were collected at five different reservoirs, and were analyzed for HgT. Large mouth bass was the only species that showed increased HgT concentrations with size class. Weight proved to be a better predictor of HgT concentrations, as four out of the five reservoirs showed a significant correlation between HgT and weight in large mouth bass. Only one reservoir was significantly different in terms of the amount of HgT in large mouth bass. In this study, we found that there were several large mouth bass from MD reservoirs that exceeded the common Hg advisory level (0.5 mg Hg/ kg fish tissue) and a few that approached the FDA action level (1.0 mg Hg/kg fish).

The fine-scale vertical distribution of zooplankton in an estuarine system was analyzed by addressing the following hypotheses: 1) Physical parameters such as low dissolved oxygen (DO), salinity, and temperature have an effect on the fine-scale vertical distribution of zooplankton. 2) Given that the copepod, Acartia tonsa is relatively sensitive to hypoxia, there will be a strong relationship between fine-scale vertical distribution of mesozooplankton and DO concentrations. 3) Given that the ctenophore, Mnemiopsis leidyi is relatively tolerant of hypoxia, there will not be a strong relationship between fine-scale vertical distribution gelatinous zooplankton and DO levels. 4) The greatest M. leidyi density will be found in the same fine-scale vertical layer as their prey, mesozooplankton. To test these hypotheses sampling was completed in the Patuxent River in July 1999 and in the mainstem of Chesapeake Bay in July 2000. DO, temperature and salinity were measured at 1m intervals using a YSI meter. Zooplankton samples and underwater video footage were attained at 2 meter intervals. Laboratory studies included identification and counting of the zooplankton samples and analysis of video footage in order to determine the ctenophore abundance in each depth interval. Zooplankton distribution data relative to DO concentrations supports the first hypothesis. There was a strong oxycline in Chesapeake Bay where DO concentrations where nearly anoxic below 10m. Zooplankton were not found below 10m in these severe oxygen conditions. However, when bottom DO does not reach stressful levels (that is, in the Patuxent River), zooplankton are distributed to lower depths. The second and third hypotheses were neither supported nor refuted. This is because in order to address these hypotheses DO levels had to be stressful (1-2 mg/L), but not lethal for ctenophores and zooplankton in a large part of the water column. These physical conditions never occurred in either the Bay or the River during sampling times. Distribution of zooplankton and ctenophores supports the fourth hypothesis. Both the Bay and River stations demonstrate an overlap in zooplankton and ctenophore peaks. The results received in this project can be used to further the study of the effects that low DO levels has upon the trophic interactions taking place in an estuarine system.

The effects of sediment disturbance on methylmercury production were studied at Mackall Cove, off the Patuxent River estuary, near St. Leonard, MD. The study consisted of six mesocosms and three treatments. The sediment was stirred twice over three weeks in the disturbed treatments. Porewater and sediment samples were taken to determine the concentrations of methylmercury, sulfate and sulfide. Deeply disturbed sediment (10 cm) was shown to have a significant increase in the production of methylmercury due to an increase in microbial activity.

Labyrinthilomycota spp. (or "Laby") is an unnamed species of the same phylum as the clam parasite QPX (quahog parasite unknown). QPX disease has been shown to cause mass mortalities in the clam (Mercenaria mercenaria). Research was conducted to determine if the cell-mediated immune system of M. mercenaria responds to this potential pathogen via chemotaxis, defined as directional migration of blood cells (hemocytes) along a concentration gradient of substances secreted by Laby. Test substances (i.e., live Laby cells, cell-free culture media, dead or lysed Laby cells, concentrated Laby proteins from culture media) were placed in the bottom of a blind-well chamber and clam hemocytes were placed in the top well, with a 5 micron pore size polycarbonate membrane separating the two wells. After an hour incubation, membranes were fixed, stained, and mounted, and cells were counted under 400x for the top and bottom sides of the membrane to determine the number of cells migrating through the membrane towards the putative chemoattractant. This information was used to calculate a chemotactic index, to compare control and experimental data. Preliminary results indicate that the presence of Laby cells does increase hemocytic chemotaxis, and that this response is dose-dependent. Results also indicate that hemocytes are responding more to a secreted chemoattractant, possibly protein, than to the Laby cell surface components. It is suggested that Laby cells secrete a mix of proteins, some that are attractive and some that are inhibitory to the clam chemotactic response. Since Laby is closely related to QPX, conclusions drawn from these experiments may be applicable to QPX as well.

Through the addition of sediment pore water extracted from various sites around the Delmarva Peninsula, the hypothesis was tested which stated that high-nutrient filtrate would stimulate growth of a local isolate of Prorocentrum minimum to a greater extent than would nutrient-poor filtrate. Growth rates of the dinoflagellate were analyzed over a period of 10-18 days amongst tubes containing pore water of varying nutrient compositions and concentrations from 18 sample sites. The sample sites were as well representative of a wide array of particle sizes and were situated in very diverse environments. Enhancement of the growth rate higher than the control was observed in 100% of experimental cultures, and some significance was found between growth rate enhancement and the existence of particular nutrients within pore water samples. Positive correlations between nutrient concentrations and physical characteristics of sediment demonstrate the possibility of classifying sample sites based on a variety of parameters. Results show the actual stimulation of growth from pore water examples and suggest that the nutrients within the benthos may serve as a source for pelagic algal growth.