Our Students and Their Research

The Arctic is warming two to three times more rapidly than the global average, causing benthic communities in the region to undergo rapid and pronounced ecological changes. Researchers have compiled time series of benthic biological, biogeochemical, and physical data in the Bering and Chukchi Seas which document northward shifts among benthic species. However, Arctic data collection is expensive. Our work has two objectives: (a) to minimize costs of future research by identifying redundant information reported among research stations; (b) to augment existing knowledge of how Arctic benthic communities are responding to rapidly changing conditions. Conventional Euclidean distance metrics for comparing time series can only provide simultaneous comparisons; however, Arctic research stations may report similar information at a time lag due to changing conditions. In this study, we applied dynamic time warping (DTW) to these time series of Arctic water column and benthic community data, realigning their time indices to highlight similarities in their shapes. A cluster analysis was then performed using the distance values output by DTW for each time series comparison. For each variable under study, we produced two clusterings using one globular and one density-based clustering algorithm. The time series of water column variables were largely clustered by region, highlighting the existence of similar spatial dynamics within, but not across, regions. For each of the sedimentary and benthic community variables, the stations were grouped into one large cluster with the exception of one or two extreme observations. Removing these outliers could improve the clustering results, rendering them more informative. Further research which integrates a holistic approach (e.g. multidimensional DTW) with a region-specific analysis (e.g. formal tests for trend synchrony) will be necessary to conclusively identify sources of redundancy among the stations of study.  

Visual estimation is a technique used to measure vegetation cover in marshes and is crucial to perform effectively to monitor marsh health and the progression of restoration projects. However, there are two vastly different existing techniques that are both ‘visual estimation,’ and that is the ungridded and gridded quadrat. Both techniques require the user to place the quadrat over a plot and give total and species-specific cover. The ungridded quadrat covers a larger area (1.0 m² vs. 0.25 m²), but the gridded quadrat has physical guidelines to help the user (25 cells of equal size). There is a lack of comparative studies of these two methods which can cause inconsistencies in data. In this study, we aimed to compare the two methods for estimating total cover and diversity across marsh types and observers. Multiple observers went to marshes of various zones (low/high) and ages (natural, 8 to 20 y/o, and >1 y/o) and gave their ungridded and gridded quadrat estimates. We also investigated observer biases between different observers and within the same observer by analyzing the same sites and on two different occasions. A regression analysis of the composite data produced an equation of conversion between the two methods. Our results revealed no significant difference between the ungridded and gridded quadrat total cover and diversity estimates between different marsh zones and ages. Significant observer bias was detected but within the same observer there were not significant differences between observations in the same locations. Further investigation is needed to confirm these results and to fine-tune the conversion factor. It is also strongly suggested to train observers prior to the field to avoid biases.  

Seasonal eutrophication occurs throughout Chesapeake Bay, leaving the bottom waters to be anoxic. Eutrophic waters lead to higher production of methane in the sediments, which is a powerful greenhouse gas that can contribute to global warming if released into the atmosphere. Methane oxidation, a process done both aerobically and anaerobically, transforms methane into carbon dioxide, which is a less powerful greenhouse gas. If the rising temperatures associated with global warming continue on a similar trend, it is possible methane oxidation rates will increase with it. To test this, anoxic and oxic sediment slurry incubations using ¹³C-enriched methane were conducted under varying temperatures using surficial sediment from the lower Chesapeake Bay. While there was an increase in carbon dioxide associated with increasing temperatures both aerobically and anaerobically, the production of methane measured and the lack of ¹³C-isotope incorporation into the CO₂ pool suggest that methane oxidation was not observed under these experimental conditions. As such, rising temperatures will increase carbon dioxide levels in the sediments of Chesapeake Bay, but it cannot be concluded that these increased temperatures will have a direct effect on methane oxidation.  

Dissolved Organic Matter (DOM) is a dynamic carbon pool with largely unknown molecular composition. We are focused on the subset of DOM that fluoresces (FDOM) and can be linked to overall DOM molecular signatures derived from ultrahigh resolution mass spectrometry (HRMS). Complex HRMS and fluorescence data were collected from samples at Bermuda Atlantic Time-series Station (BATS) and Hawaii Ocean Time-series (station ALOHA). Using sparse orthogonal factor regression (SOFAR), a method previously used in genetic analysis, we aim to learn underlying association networks between the HRMS and fluorescence datasets (Uematsu et al. 2019). Increasing sparsity in a large dataset minimizes noise and promotes selection of the most significant latent factors. On the other hand, orthogonality ensures independence.  SOFAR applies these two concepts to select smaller sets of predictor m/z ions while increasing independence of the fluorescence response variable. This project is one example of a growing collaboration between big data and environmental sciences that showcases the powerful role of machine learning techniques across disciplines, from genetics to analytical chemistry.  

Aquatic and marine viruses are present in all known aquatic environments. A proportion of these viruses infect cyanobacteria including the invasive filamentous cyanobacteria Lyngbya. In the Chesapeake Bay and other ecosystems, Lyngbya can produce toxins that are hazardous to the organisms in their surrounding environment. However, viruses have been shown to be associated with bloom decline or termination and with toxin production. Our study aimed to determine if Lyngbya in the Susquehanna Flats are producing viruses and to estimate their production rates. Lyngbya from one location in the Susquehanna Flats appeared to produce viruses, while Lyngbya from another location did not. As the location with detectable virus production was also the one with higher Lyngbya abundance, we hypothesize that viral production may only be possible when the density of Lyngbya is sufficiently high in the environment. We recommend further studies to observe viral production by Lyngbya later in the summer when there are denser mats. We also advocate determining the taxonomic identity of the viruses infecting Lyngbya, as well as exploring their effects on the blooms, as they could alter toxin production and regulate bloom termination.  

Solomon’s Harbor, MD, a shallow water estuary connected to the Chesapeake Bay, is suffering from a lack of dissolved oxygen. The two primary drivers of this oxygen depletion are the introduction of excess nutrients and climate change, which are known to increase temperatures and decrease dissolved oxygen (DO) levels. To better understand the factors that influence DO, we performed a univariate analysis of a 36-year-long historical dataset, deployed temperature sensors at seven sites for a month, and conducted dark box incubations on water samples collected on-site. Temperature was determined to be the strongest factor that explains DO, especially in bottom waters. Hydrodynamics were also impactful on DO and temperature because they determine water mixing, residence times, and variability at individual sites. Temperature was also strongly impacted by distance to the mouth of Solomon’s Harbor, with sites further away having higher temperatures than sites close to the mouth. The sites furthest from open water have less mixing, higher residence times, more variability, are more heavily impacted by sunlight and weather, and have higher average temperatures, so these sites contain the most vulnerable habitat and should be closely monitored and made a conservation priority.