Physical Sciences and Mathematics Commons^™

Impacts And Uncertainties Of Climate Change On The Chesapeake Bay, Kyle E. Hinson

Dissertations, Theses, and Masters Projects

Climate change impacts in the Chesapeake Bay will limit the efficacy of nutrient reduction efforts and decrease dissolved oxygen, but uncertainties associated with the magnitude of these effects remain. An understanding of underlying mechanisms that have driven recent warming trends will narrow uncertainties for future pathways of temperature change. Additionally, future simulations of climate impacts in the estuary are dependent on multiple different sources of uncertainty, many of which have not yet been fully evaluated. This dissertation used a three- dimensional coupled hydrodynamic-biogeochemical model to investigate recent warming trends as well as underlying uncertainties likely to influence regional projections of …

Experimental Studies Of Neutral Particles And The Isotope Effect In The Edge Of Tokamak Plasmas, Ryan Chaban

Dissertations, Theses, and Masters Projects

The H-mode plasma edge is a region of steep gradients in density and temperature known as the “pedestal” which greatly increases energy confinement. The complex links between neutral-plasma interactions and both diffusive and convective transport in the pedestal must be understood to model, predict, and achieve the high performance required for a fusion power plant. This dissertation explores the effects of different hydrogenic isotope neutral particles and plasma transport from the edge pedestal region into the Scrape-Off Layer. Current experiments typically use deuterium (H with amu=2 or D), however future fusion power plants may startup with hydrogen (H), and eventually …

Matfusion: A Generative Diffusion Model For Svbrdf Capture, Samuel Lee Sartor

Dissertations, Theses, and Masters Projects

We formulate SVBRDF estimation from photographs as a diffusion task. To model the distribution of spatially varying materials, we first train a novel unconditional SVBRDF diffusion backbone model on a large set of 312,165 synthetic spatially varying material exemplars. This SVBRDF diffusion backbone model, named MatFusion, can then serve as a basis for refining a conditional diffusion model to estimate the material properties from a photograph under controlled or uncontrolled lighting. Our backbone MatFusion model is trained using only a loss on the reflectance properties, and therefore refinement can be paired with more expensive rendering methods without the need for …

Intelligent Software Tooling For Improving Software Development, Nathan Allen Cooper

Dissertations, Theses, and Masters Projects

Software has eaten the world with many of the necessities and quality of life services people use requiring software. Therefore, tools that improve the software development experience can have a significant impact on the world such as generating code and test cases, detecting bugs, question and answering, etc. The success of Deep Learning (DL) over the past decade has shown huge advancements in automation across many domains, including Software Development processes. One of the main reasons behind this success is the availability of large datasets such as open-source code available through GitHub or image datasets of mobile Graphical User Interfaces …

Gas-Phase Proton Affinities Of Proline- And Pipecolic Acid-Containing Dipeptides, Trinh Ton

Dissertations, Theses, and Masters Projects

Mass spectrometry (MS) is one of the most used techniques in proteomics because it allows for both high-throughput and quantitative analyses. Bottom-up MS-based proteomics involves breaking down proteins into smaller chains of amino acids called peptides, ionizing and fragmenting the peptides, and identifying the fragments using sequencing databases. These databases depend on the random fragmentation at the backbone peptide bond of the peptides, as predicted by the mobile proton model. Research has shown that peptides containing proline or pipecolic acid have selective fragmentations that could lead to incorrect identification in the sequencing algorithms. These selective cleavages are called “the proline …

Efficient Parallelization Of Irregular Applications On Gpu Architectures, Qihan Wang

Dissertations, Theses, and Masters Projects

With the enlarging computation capacity of general Graphics Processing Units (GPUs), leveraging GPUs to accelerate parallel applications has become a critical topic in academia and industry. However, a wide range of irregular applications with a computation-/memory-intensive nature cannot easily achieve high GPU utilization. The challenges mainly involve the following aspects: first, data dependence leads to a coarse-grained kernel; second, heavy GPU memory usage may cause frequent memory evictions and extra overhead of I/O; third, specific computation patterns produce memory redundancies; last, workload balance and data reusability conjunctly benefit the overall performance, but there may exist a dynamic trade-off between them. …

Characterizing Molecular Environments In Acrylic Paint Via Single-Sided Nmr, Lyndi Kiple

Dissertations, Theses, and Masters Projects

Acrylic paint is a modern artistic material made of colored pigment and polymeric binder. Acrylic binder requires fundamental study at the molecular level to understand its physical properties for purposes of art conservation and general polymer chemistry. The research presented in this thesis uses single-sided nuclear magnetic resonance (NMR) as a non-invasive and non-destructive way to measure relaxation and self-diffusion, which provide insight to molecular mobility and physical properties of proton-containing samples. Specifically, this study relies on T2 relaxation to gain insight to regions within acrylic paint with different molecular mobilities. In both dry and wet paint, relaxometry data revealed …

A Comprehensive Study Of Bills Of Materials For Software Systems, Trevor Stalnaker

Dissertations, Theses, and Masters Projects

Software Bills of Materials (SBOMs) have emerged as tools to facilitate the management of software dependencies, vulnerabilities, licenses, and the supply chain. Significant effort has been devoted to increasing SBOM awareness and developing SBOM formats and tools. Despite this effort, recent studies have shown that SBOMs are still an early technology not adequately adopted in practice yet, mainly due to limited SBOM tooling and lack of industry consensus on SBOM content, tool usage, and practical benefits. Expanding on previous research, this paper reports a comprehensive study that first investigates the current challenges stakeholders encounter when creating and using SBOMs. The …

Recoverable Memory Bank For Class-Incremental Learning, Jiangtao Kong

Dissertations, Theses, and Masters Projects

Incremental learning aims to enable machine learning systems to sequentially learn new tasks without forgetting the old ones. While some existing methods, such as data replay-based and parameter isolation-based approaches, achieve remarkable results in incremental learning, they often suffer from memory limits, privacy issues, or generation instability. To address these problems, we propose Recoverable Memory Bank (RMB), a novel non-exemplar-based approach for class incremental learning (CIL). Specifically, we design a dynamic memory bank that stores only one aggregated memory representing each class of the old tasks. Next, we propose a novel method that combines a high-dimensional space rotation matrix and …

Mattanock Town Restoration Plan, Katlin Mccarter Grigsby

Dissertations, Theses, and Masters Projects

Mattanock Town's Restoration Plan is a science-based restoration process that evaluates the site's history, the tribal history, and the most current research to maximize native habitats, enhance coastal resilience, and reconnect the Nansemond people to the local river. Restoration priorities include increasing native plant species, incorporating oyster habitat, and addressing erosion. This plan details how synthesizing existing and new physical, biological, and cultural information can help the Nansemond Indian Nation prioritize projects that benefit their community and the surrounding environment.

Achieving Equitable Offshore Wind Development: Lessons From European Stakeholders, Kacey Hirshfeld

Dissertations, Theses, and Masters Projects

The Biden Administration has set aggressive offshore wind energy goals, aiming to have 30 gigawatts of offshore energy in place by 2030. This amount of energy has the potential to power 10 million homes (White House, 2022), helping the administration to reach larger clean energy goals. In Virginia, Dominion Energy aims to have 2.6 gigawatts of offshore wind energy by 2026, enough to power up to 660,000 homes (Dominion Energy).

While the upcoming offshore wind energy development will create clean energy and green jobs, the ocean is no longer an open field for development and already supports a complex matrix …

Development Of 3d And 360 Gis-Based Models To Visualize Projected Sea Level Rise In Coastal Virginia, Candice M. Vinson

Dissertations, Theses, and Masters Projects

Science communication is a skill that can be strengthened with practice. Like any skill, it helps to know what you need to practice in order to get better at it. When presenting information to an audience, the skill of science communication comes into play as early as the first draft of a presentation. As you think about how you will tell your story to the audience, you likely consider including text on slides, images, graphs, maps, or even videos. However, it is crucial to remember that accessibility barriers are something we must often consciously work to rid our presentations of, …

Voting Rules And Properties, Zhuorong Mao

Undergraduate Honors Theses

This thesis composes of two chapters. Chapter one considers the higher order of Borda Rules (Bp) and the Perron Rule (P) as extensions of the classic Borda Rule. We study the properties of those vector-valued voting rules and compare them with Simple Majority Voting (SMV). Using simulation, we found that SMV can yield different results from B1, B2, and P even when it is transitive. We also give a new condition that forces SMV to be transitive, and then quantify the frequency of transitivity when it fails.

In chapter two, we study the `protocol paradox' of approval voting. In approval …

Quantum Federated Learning: Training Hybrid Neural Networks Collaboratively, Anneliese Brei

Undergraduate Honors Theses

This thesis explores basic concepts of machine learning, neural networks, federated learning, and quantum computing in an effort to better understand Quantum Machine Learning, an emerging field of research. We propose Quantum Federated Learning (QFL), a schema for collaborative distributed learning that maintains privacy and low communication costs. We demonstrate the QFL framework and local and global update algorithms with implementations that utilize TensorFlow Quantum libraries. Our experiments test the effectiveness of frameworks of different sizes. We also test the effect of changing the number of training cycles and changing distribution of training data. This thesis serves as a synoptic …

Modern Theory Of Copositive Matrices, Yuqiao Li

Undergraduate Honors Theses

Copositivity is a generalization of positive semidefiniteness. It has applications in theoretical economics, operations research, and statistics. An $n$-by-$n$ real, symmetric matrix $A$ is copositive (CoP) if $x^T Ax \ge 0$ for any nonnegative vector $x \ge 0.$ The set of all CoP matrices forms a convex cone. A CoP matrix is ordinary if it can be written as the sum of a positive semidefinite (PSD) matrix and a symmetric nonnegative (sN) matrix. When $n < 5,$ all CoP matrices are ordinary. However, recognizing whether a given CoP matrix is ordinary and determining an ordinary decomposition (PSD + sN) is still an unsolved problem. Here, we give an overview on modern theory of CoP matrices, talk about our progress on the ordinary recognition and decomposition problem, and emphasis the graph theory aspect of ordinary CoP matrices.

Development Of A Vector Magnetometer Based On Electromagnetically Induced Transparency In 87rb Atomic Vapor, Alexander Toyryla

Undergraduate Honors Theses

We present progress towards the development of an atomic magnetometer capable of accurate scalar and vector magnetic field measurements with high sensitivity and no need for external calibration. The proposed device will use the interaction between a bi-chromatic laser field and rubidium vapor to derive magnetic field magnitude and direction from measured amplitudes of Electromagnetically Induced Transparency (EIT) resonances. Since the proposed method requires precision control of light polarization, we observe the performance capabilities of a liquid crystal device to dynamically rotate the polarization of the laser field. Another goal in this project is to establish a polarization locking mechanism …

Alkali Linewidths Under High Temperatures And Pressures Of 3he, Michael Parker

Undergraduate Honors Theses

Current research at Thomas Jefferson National Accelerator Facility is being conducted to study the spin structure of the neutron through collisions with polarized 3He nuclei. The helium is contained in high pressure glass vessels (called cells) along with nitrogen, rubidium, and potassium. To deduce the spin structure from collisions, we need to know the precise number density of 3He in the cell. The process of polarizing 3He through spin-exchange optical pumping requires nitrogen and alkali metal. We can use the absorption linewidths of rubidium and potassium to more accurately determine the density of helium. Throughout my research, I collected absorption …

Enumerating Switching Isomorphism Classes Of Signed Graphs, Nathaniel Healy

Undergraduate Honors Theses

Let Γ be a simple connected graph, and let {+,−}^E(Γ) be the set of signatures of Γ. For σ a signature of Γ, we call the pair Σ = (Γ,σ) a signed graph of Γ. We may define switching functions ζ_X ∈ {+, −}^V (Γ) that negate the sign of every edge {u, v} incident with exactly one vertex in the fiber X = ζ^{−1}(−). The group Sw(Γ) of switching functions acts X on the set of signed graphs of Γ and induces an equivalence relation of switching classes in its orbits; there are 2^{|E(Γ)|−|V (Γ)|+1} such classes. More interestingly, …

Climate Change And Conservation Of Milkweed: Evidence Of Extensive Admixture Between Common Milkweed And Poke Milkweed, Elizabeth Davies

Undergraduate Honors Theses

Global climate change can drive many changes in species interactions. One primary way it affects species is by changing climates, causing species to expand their ranges and allowing them to interact with species from whom they were previously isolated. In plants, new species interactions can result in hybridization – the creation of hybrid offspring between two separate species. This hybridization can increase gene flow between the species and lead to introgression, the transfer of genetic material from one species to another through hybrid backcrossing with the parent species. My thesis investigates hybridization in the model system Asclepias (milkweed) by analyzing …

Chemical Analysis Of Organic Compounds In Dew Water, Monica Dibley

Undergraduate Honors Theses

Water films on outdoor surfaces, such as dew, can act as a reservoir for organic molecules deposited from the atmosphere and they present a potential reactive medium for chemical transformations. To better understand the flux of volatile organic compounds from evaporating films, the composition and reactivity of the complex mixture of dissolved organic material (DOM) found in these films need to be characterized. Previous studies have measured the salts and the small organic molecules in dew collected on clean Teflon surfaces or condensers. Here, we expand on this by probing the organic chemicals found on natural outdoor surfaces covered in …

Using Deep Learning With Satellite Imagery To Estimate Deforestation Rates, Maeve Naughton-Rockwell

Undergraduate Honors Theses

Previous studies have used Convolutional Neural Networks for regional detection of deforestation breaks. However, there is limited research into the capability of deep neural networks to identify sudden shifts in global forest cover from satellite imagery. Additionally, many deforestation detection models are trained on region specific data and need manual input thresholds. In this work, we develop a deep learning model to predict the percent of deforestation in a region between two points in time, trained on globally sourced data. Using the before and after satellite images of a deforestation event as inputs, we implemented a two input Convolutional Neural …

Differential Protein Expression In Bacteriophages Crimd And Larva, Daria Moody

Undergraduate Honors Theses

Proteomics studies allow us to answer questions about differential protein expression across different systems. Mass spectrometry is a powerful tool in these studies due to the distinct masses of the amino acids that compose proteins. In our experiment, we used a bottom-up approach and focused on two bacteriophages found on the William & Mary campus, CrimD and Larva. The infection of Mycobacterium smegmatis, a nonpathogenic model for tuberculosis, by these two bacteriophages was frozen at five different timepoints, and our goal was to compare the differential protein expression across the samples in order to gain a greater understanding of …

Using A Machine Learning Model To Predict Plant Inflorescences Based Upon Its Soil Microbiome, Luke Denoncourt

Undergraduate Honors Theses

The UN estimates that the global population could reach 9.7 billion by 2050 (United Nations). As a result, the amount of food required to feed humanity is thought to double by 2050 (Ray et al., 2012). Humanity must find a way to increase crop production without increasing fertilizer usage and eutrophication, which can be done using the soil microbiome. Using potted plants with soils inoculated with Pseudomonas alcaligenes, Pseudomonas denitrificans, Bacillus polymyxa, and Mycobacterium phlei, both the shoot and root growth of pea and cotton plants was significantly increased (Egamberdieva & Höflich, 2004). In this study, utilizing a random forest …

An Atomic Magnetometer Based On Nonlinear Magneto-Optical Polarization Rotation, Jiahui Li

Undergraduate Honors Theses

Magnetometers with high precision and accuracy have wide applications across various areas. We are developing an atomic magnetometer based on nonlinear magneto-optical rotation (NMOR). The magnetometer measures the polarization rotation of a light field, which is proportional to the magnetic field strength. However, such a magnetometer usually has a limited operation range and stops working for fields stronger than the Earth's magnetic field. To overcome this shortage, we implement frequency and amplitude modulation that induces side frequencies in the Fourier space which allows us to measure strong magnetic fields, up to 200 mG. We have achieved 60 pT sensitivity for …

Bayesian Spatial Model Development Of Soil Core Organic Matter As A Proxy For Blue Carbon Stocks Within The Chesapeake Bay, Christian Longo

Undergraduate Honors Theses

Blue carbon is carbon captured and stored within bodies of water and their ecosystems. Blue carbon stocks are very important due to their ability to store carbon away from the atmosphere. The destruction of these stocks can accelerate climate change. In particular, we wish to assess blue carbon stock within the Chesapeake Bay. Previous studies have only used geographical features to predict blue carbon stock levels. The big picture question this thesis was meant to answer is: What is the best approach for building a statistical model that factors in both spatial parameters and geographical features to predict blue carbon …

Co-Planar Waveguides For Microwave Atom Chips, Morgan Logsdon

Undergraduate Honors Theses

This thesis describes research to develop co-planar waveguides (CPW) for coupling microwaves from mm-scale coaxial cables into 50 μm-scale microstrip transmission lines of a microwave atom chip. This new atom chip confines and manipulates atoms using spin-specific microwave AC Zeeman potentials and is particularly well suited for trapped atom interferometry. The coaxial-to-microstrip coupler scheme uses a focused CPW (FCPW) that shrinks the microwave field mode while maintaining a constant 50 Ω impedance for optimal power coupling. The FCPW development includes the simulation, design, fabrication, and testing of multiple CPW and microstrip prototypes using aluminum nitride substrates. Notably, the FCPW approach …

Investigation Of Tertiary Impact Cratering And Relation To Impact Physics Theory, Mikayla Huffman

Undergraduate Honors Theses

Extraterrestrial impact crater formation is important in many subfields of planetary science, including geochronology, planetary formation, and dynamic fragmentation theory. Current dynamic fragmentation theory lacks scale dependence and relies heavily on terrestrial data. Exploring a range of impact and ejecta velocities as is produced by cratering events on the Moon may bridge the gap between heavily terrestrial-based theory and planetary data. The secondary craters of secondary craters deemed “tertiary craters,” have been theorized, but planetary images have not been of sufficient resolution to effectively search for them until recently. Tertiary craters are formed by relatively low-velocity fragments ejected by nearby …

Gas-Phase Proton Affinities For Twenty Of The Proline-Containing Dipeptides, Henry Cardwell

Undergraduate Honors Theses

Peptide fragmentation plays a crucial role in the analysis of proteins through mass spectrometry-based proteomics. Most proteomics experiments take place in the low-energy regime and are governed by the mobile proton model which predicts random cleavages along the peptide backbone; however, there sometimes arise circumstances where the mobile proton model fails causing sequencing algorithms to misidentify peptides. One such example is noted in the “proline effect” wherein proline-containing peptides preferentially fragment N-terminal. While it has been established that the “proline effect” is due to the rigidity and basicity of the proline N-terminus, a further understanding of the factors influencing the …

Period Doubling Cascades From Data, Alexander Berliner

Undergraduate Honors Theses

Orbit diagrams of period doubling cascades represent systems going from periodicity to chaos. Here, we investigate whether a Gaussian process regression can be used to approximate a system from data and recover asymptotic dynamics in the orbit diagrams for period doubling cascades. To compare the orbits of a system to the approximation, we compute the Wasserstein metric between the point clouds of their obits for varying bifurcation parameter values. Visually comparing the period doubling cascades, we note that the exact bifurcation values may shift, which is confirmed in the plots of the Wasserstein distance. This has implications for studying dynamics …

Machine Learning In Healthcare: Improving The Diagnosis Of Pulmonary Embolism In Covid-19 Patients, Soheb Osmani

Undergraduate Honors Theses

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has created new challenges for clinicians diagnosing pulmonary embolism (PE). Clinicians currently rely on D-Dimer levels in conjunction with clinical prediction scores to rule out and diagnose PE. However, patients with COVID-19 (the disease caused by SARS-CoV-2) often present with elevated D-Dimer levels. D-Dimer levels in COVID-19 patients have been found to be positively correlated with the severity of disease. Symptoms of COVID-19 also often align with symptoms of PE. Therefore, it becomes more difficult for clinicians to identify which COVID-19 positive patients should undergo further testing for PE. This study evaluates …