Masters Theses

Permanent URI for this collection

https://hdl.handle.net/10919/9291

Browse

Now showing 1 - 20 of 23766

Effects of Surface Wettability on Cavitation Inception and Cloud Cavitation
Chandrasekaran, Swathiga Devi (Virginia Tech, 2026-05-26)
Cavitation, the phenomenon involving the formation and collapse of bubbles due to pressure variations in liquid flows. This causes noise, vibrations, surface damage, and reduced efficiency in hydraulic systems. Cavitation is strongly dependent on the surface of the systems involved. However, the effect of surface wettability on controlled flow conditions is largely unexplored. The experiment in this study investigates the impact of cavitation formation and development over Venturi geometry. The Venturi models were engineered using three techniques: spray-coating, chemical etching, and initiated chemical vapor deposition, to induce different surface wetting characteristics. The models were characterized through contact angle and roughness measurements. High-speed imaging, with synchronized pressure data, was obtained to study cavitation: inception bubble statistics and cloud cavitation dynamics. Space-time diagrams, power spectral density, spatial occupancy maps, and RAFT-based optical flow velocimetry were performed to quantify the vapor and liquid-phase dynamics. The results revealed that the surface with increased hydrophobicity was incepted earlier and demonstrated localized near-wall inception at the Venturi throat. For example, the carbon black spray-coated surface, with a contact angle of 140 deg, promoted inception at 45 L/min, whereas the clear surface (Contact angle of 84 deg) incepted at 50 L/min. In addition, the hydrophobic surface led to reduced modal bubble size, nearly half the size as seen over the clear surface. The hydrophobicity also influenced the peak inception event density, with inception events 10 times higher than in the clear case. In the cloud cavitation analysis, hydrophobic and superhydrophobic surfaces consistently produced thinner and shorter cavities with higher shedding frequencies. At the same Reynolds numbers, the carbon black spray-coated surface reduced the cavity length by about 4.9 mm in comparison to the cavities over the clear surface. The etched Aluminum, with increased surface roughness and hydrophobicity, showcased suppression of cavitation cycles. In contrast, the iCVD superhydrophilic surface produced larger, longer cavities, reaching 16.78 mm at Re = 238697, compared to cloud lengths of 15.61 mm and 12.9 mm for the clear and iCVD hydrophobic surfaces, respectively. The RAFT-based velocimetry illustrated more attached shear layers, with reduced upstream circulation for hydrophobic surfaces. On the other hand, the surfaces at the other end of the wettability spectrum, clear and superhydrophilic surfaces, led to broader and detached vapor structures. As a whole, the findings indicate that the surface wettability acts as an important parameter for altering the cavitation flow dynamics, both at initial and advanced stages of cavitation. The experimental results guide engineering strategies focused on delaying, suppressing, or controlling cavitation in related engineering systems.
La Cuna Del Volcan
Manzanares, Rafael Nicolas (Virginia Tech, 2026-05-26)
This thesis explores the concept of care in architecture through horizontal and vertical movement, emphasizing the journey of healing through space. Situated on a site overlooking a volcano, the project reinterprets ancient architectural principles as a foundation for healing environments. While volcanoes are often associated with danger, this proposal reframes them as places of refuge "The Cradle of the Volcano" symbolizing safety, renewal, and transformation. The design investigates how natural light, green space, material texture, and views can shape well-being and personal growth. Through its spatial organization and connection to context, the thesis proposes architecture as a reflection and belonging.
Parameter Identification and Validation of a Control-Oriented Vehicle Dynamics Model for an Autonomous Chevrolet Bolt EUV
Kang, Hyunbin (Virginia Tech, 2026-05-26)
Accurate and computationally tractable vehicle dynamics models are essential for autonomous vehicle development, supporting offline software validation, controller design, and simulation- based testing. This thesis presents a systematic methodology for identifying and validating the parameters of a control-oriented dynamics model of a Chevrolet Bolt EUV used by the Virginia Tech AutoDrive team. The platform introduces modeling challenges not adequately addressed by off-the-shelf tools, including a steer-by-wire steering system with speed-dependent nonlinear gain, and a Controller Area Network (CAN) interface that im- poses implementation-specific command scaling on the brake channel. A baseline Simulink model is constructed from a longitudinal force balance with assumed second-order actuator dynamics and a kinematic bicycle model for lateral motion. Parameters are identified using MATLAB's patternsearch derivative-free solver. Longitudinal identification is treated as a physics-based estimation of vehicle mass, linear viscous rolling resistance, brake command gain, and the torque actuator's dynamics. Lateral identification is treated as a data-driven functional correction, in which a two-dimensional lookup table parameterized by vehicle speed and steering command magnitude captures the nonlinear steer-by-wire response. System-level validation across four combined longitudinal-lateral maneuvers from AutoDrive competition tasks demonstrates sufficient fidelity to support con- troller development using only the GNSS, IMU, and CAN signals available on the vehicle.
Invisible Barriers
Verdura, Francesca Natalia (Virginia Tech, 2026-05-26)
This research will explore the intersection of accessibility and nature within the built environment. It is essential to advocate for thoughtful architectural design that is inclusive of all abilities while being deeply connected to natural elements. Drawing from my own personal experiences with disabilities in my own family, I argue that accessibility is often treated as an afterthought in design. It is limited by rigid compliance to standardized building codes that primarily prioritize wheelchair users while overlooking other forms of disabilities. This thesis project emphasizes the potential of biophilic and sensory-based design to create inclusive outdoor environments that activate all five senses for people of all ages and abilities. I plan to research this topic by analyzing natural forms and movements, critically engaging with ADA standards, and designing spaces that accommodate a diverse range of needs. This work aims to re-imagine accessibility as an essential and expressive part of architecture, not a constraint.
Biomechanical Analysis of Equipment Modifications to Reduce Concussion Risk due to Projectile Impacts in Women's Lacrosse
Metrey, Katherine Lee (Virginia Tech, 2026-05-26)
Participation in women's lacrosse has nearly doubled since 2003, despite the non-contact sport boasting concussion incidences as high as some contact sports. Most head injuries result from accidental equipment impacts, whether during passing, shooting, or ground-ball pursuits. Though headgear is not mandated, ASTM F3137 specifies test methods to certify soft-shell headgear for optional in-game use. Additionally, lacrosse manufacturers have begun proposing equipment modifications to reduce head injury rates, such as a lightweight ball. Previous research suggests that headgear can reduce overall concussion rate, though the effects during projectile impacts are not understood biomechanically. For short-impulse impacts, not only is rigid-body motion (RBM) of interest, but also the head's vibrational response, where skull resonant frequencies can be excited. Quantifying biomechanical responses to risk reduction interventions in women's lacrosse can increase player safety by influencing rule changes and informing sport manufacturers. Through the recreation of projectile impacts observed in women's lacrosse, this study aimed to analyze the effects of three risk reduction interventions on biomechanical response: (1) Headgear effects (impacts with standard balls on headgear and bare headform conditions), (2) ball effects (impacts with lightweight and standard balls on a bare headform), and (3) combined effects (impacts with a lightweight ball on headgear against standard ball, bare headform impacts). Both interventions, headgear and the lightweight ball, were individually effective in reducing biomechanical response during projectile impacts. We estimate headgear can reduce RBM by 17.2 – 24.1 % and vibrational response by 20.3%, while the lightweight ball can reduce RBM by 12.6 – 22.4% and vibrational response by 8.5%. Variations between headgears may result from different padding configurations, while ball types may exhibit varying coefficients of restitution. The greatest reductions in injury response were observed when interventions were combined and compared against control impacts involving only a bare headform and standard balls. We estimate the combined treatment can reduce RBM by 35.6 – 40.7% and vibrational response by 33.4%, suggesting the implementation and modification of equipment could play a critical role in athlete safety by mitigating high concussion incidences in women's lacrosse.
Economic Evaluation of Agricultural and Related Research and Extension in Virginia
Rahman, Raisa (Virginia Tech, 2026-05-26)
Public investment in agricultural research and extension (RandE) significantly drives productivity and rural development, yet state-level data can be outdated. This thesis revisits Virginia's public RandE returns from 1949 to 2022 using a reconstructed multi-factor productivity (MFP) index and a distributed lag model. Findings show that Virginia's agricultural productivity rose by 1.49% annually, while knowledge stocks increased by 4.03% annually. The estimated elasticity of 0.37 indicates a modified internal rate of return of 6.2%, an internal rate of return of 12.18%, and a benefit-cost ratio of 4.53 for each $1,000 investment. While these figures are lower than previous Virginia results, they align with national trends of a post-1970s productivity slowdown, demonstrating that public RandE still provides substantial economic returns. Importantly, the analysis did not reveal a statistically stable long-run relationship between knowledge stocks and MFP, suggesting these return metrics are growth-accounting indicators. The thesis also outlines data sources and methods for replicability and emphasizes the need for a regular evaluation system to ensure timely, relevant evidence for Virginia's agricultural policy and investment decisions.
The High Life Girl
Raab, Anne Elizabeth (Virginia Tech, 2021-05-26)
The High Life Girl is a novel written in a close first perspective that explores the cost of grief and loss through the eyes of a teenaged girl living with new instability in Milwaukee, Wisconsin.
Brain-Inspired Drone Navigation Using the Conflict Architecture
Chauhan, Vikhyat (Virginia Tech, 2026-05-21)
Novel forms of brain-inspired computer architectures are required both to understand the structural basis of intelligence and to overcome fundamental limits in computer parallelism and real-time autonomous decision-making. The brain does not compute a single answer and commit; it runs multiple, heterogeneous processing pathways in parallel on the same input, preserves conflicting candidate responses, and defers commitment until an external event or physical constraint forces selection. This organizational strategy is not arbitrary: as Rent's Rule [1-3] establishes for any large-scale system, the brain cannot sustain dense global synchronization without prohibitive communication costs, and so distributes computation across isolated, specialized modules — preserving local conflict rather than resolving it globally — until an external event forces commitment. Prior work introduced the Conflict Architecture (CA) to demonstrate that this principle can be instantiated in an engineered system, showing through maze-solving experiments that maintaining multiple conflicting algorithmic interpretations in parallel — and resolving conflict through time-driven arbitration rather than early convergence — can outperform any single strategy under unpredictable deadline constraints. That work established the CA as a proof of concept but left open the question of which classes of real-world problems benefit from this approach and whether the principle holds under a richer set of simultaneous physical constraints. This thesis addresses both questions. First, it systematically investigates the structural properties of problems for which the CA provides a principled advantage: critical decisions under unpredictable deadlines, multiple algorithmically distinct solution strategies, feasibility determined by physical constraints rather than purely on theoretical optimality, and continuous interaction with a dynamic and partially observable environment. This investigation motivates autonomous aerial navigation as a representative real-world domain, where unmanned aerial vehicles must simultaneously satisfy timing, energy, and safety constraints and where no single navigation strategy is optimal across all operating conditions. Second, this thesis introduces the CA-based Navigation Engine as a direct instantiation of the CA applied to UAV navigation in dynamic, obstacle-rich corridor environments. It runs three Algorithm Processing Elements (APEs) in parallel — a hard-failsafe reflex (APE1), a moderate planning depth reactive planner (APE2), and a high planning depth corridor navigator (APE3) — on a shared, time-stamped state snapshot, and commits whichever proposal is both complete and highest in planning depth before each externally imposed deadline expires. This organization instantiates Flynn's MISD[4] category — Multiple Instruction, Single Data — historically dismissed as either nonsensical because it produces conflicting answers or confined to fault- tolerant systems that treat those conflicts as errors to be caught. The CA reframes both: conflicting answers are not a defect but the design, and MISD becomes not merely coherent but advantageous precisely when the goal is to sustain multiple competing answers simultaneously until a physical deadline forces selection. The architecture is evaluated in a Gazebo–ROS 2 simulation of procedurally generated Manhattan-grid corridor worlds under a stochastic, event- driven deadline model that prevents any fixed strategy from reliably completing before action is required. Experiments across 1,000 simulation runs per strategy confirm and extend the core finding of the prior CA work. APE2 and APE3, deployed as standalone strategies, had event violations in over 80% and 86% of runs respectively — at mean violation rates of 28.02% and 40.39% respectively — rendering both non-feasible for deployment under real-time constraints despite their apparent speed advantages in uninterrupted conditions. Only APE1 and the CA achieved zero deadline event violations across all 1,000 runs. Between the two feasible strategies, the CA achieved a 34.55% reduction in mission elapsed time relative to APE1 (38.98 s versus 59.56 s; p < 0.0001) and a 34.52% reduction in algorithmic computation energy (97.51 J versus 148.92 J; p < 0.0001), with no statistically significant difference in no-fly zone violations or total physical flight energy. These results demonstrate that CA-style arbitration is Pareto- improving over the only other feasible standalone strategy. Significantly, the CA simultaneously reduces both completion time and computational energy cost, while conventional power reduction comes at the expense of reduced performance. This work extends the CA evaluation framework to include energy as an explicit metric alongside time and lays the foundation for further research in conflict-aware architectures applied to multi-constraint, real-time autonomous systems as well as a broader class of problems.
Extracellular Vesicle‑Driven CNS Niche Formation in Triple Negative Breast Cancer Metastasis
Oyediran, Khadijat Olorunsefunmi (Virginia Tech, 2026-05-21)
Triple negative breast cancer (TNBC) frequently metastasizes the central nervous system, where outcomes remain poor, yet how tumor-derived signals condition before tumor arrival is still not well understood. Tumor-derived extracellular vesicles (EVs) are known mediators of long-range communication that can alter distant tissues and contribute to pre-metastatic niche formation, but their impact on the meningeal lymphatic system and glial cells, two systems that regulate fluid balance, immune signaling, and CNS homeostasis, remains unclear. To address this, we examined how TNBC-derived EVs influence lymphatic endothelial cells (LECs), meningeal stromal cells (HMCs), astrocytes, and microglia across early and extended time points. In the meningeal lymphatic model, EV exposure did not change overall cell coverage, but LECs showed clear changes in junctional organization, with increased spacing at early time points followed by reorganization, alongside a consistent increase in EV secretion, indicating an active cellular response rather than passive disruption. HMCs maintained structural stability while showing a more modest increase in EV secretion, suggesting a supportive role within the system. In parallel, glial cells did not exhibit strong morphological or proliferative changes, but both astrocytes and microglia increased EV secretion, and cytokine profiling revealed elevated inflammatory signaling, with astrocytes showing increased IL-6 expression linked to aquaporin-4 regulation and fluid movement within the CNS. These findings suggest that TNBC-derived EVs drive early changes by modulating cellular communication and signaling rather than inducing immediate structural damage, shifting the CNS microenvironment into a more responsive and potentially permissive state. This work provides insight into how early tumor-independent signaling may influence CNS function and highlights EV-mediated interactions as a key mechanism in shaping disease progression before metastasis occurs.
Sensory Properties, Consumer Acceptability, Nutritional Composition, and Flavor Profiles of Dark Chocolates Produced from Wild Fine-Flavor Amazonian Cacao
Adornetto, Julia Christine (Virginia Tech, 2026-05-21)
Consumer interest in fine flavored cocoa products has continued to grow annually alongside a global decline in cocoa production due to climate change and crop epidemics. Wild cacao from the Brazilian Amazonian regions have been shown to produce high quality cocoa and chocolates. There is a noted lack of literature pertaining to the sensory profiles of these wild cocoa beans and their products. This project set out to characterize the nutritional, chemical, and sensory properties of 70% and 81% cocoa dark chocolate bars manufactured from wild cacao beans harvested along the Tocantins, Cassiporé, Acará, Juruá, and Purus rivers in the Brazilian Amazon. In the free sorting study, panelists (n = 31) were asked to group ten chocolate samples (5 origins, 70 and 81% cocoa) according to similarity (k = n-1) and describe them using open-ended responses. DISTATIS and textural analysis yielded 26 descriptors. In the second study, panelists (n=90) evaluated the five 70% cocoa chocolates (9-pt hedonic; overall-liking, appearance, aroma, flavor, and mouthfeel liking; "Check-All-That-Apply"), and answered a short questionnaire. There were significant differences in consumer acceptance of the chocolates (p<0.05) and Tocantins was most liked overall (6.77 ±1.30). Penalty analysis revealed "creamy," "melting" as positive drivers of acceptability, while "leather" and "hay" contributed to lower hedonic scores. Proximate analysis showed 81% cocoa chocolates had higher ash, moisture, lipid, and protein content than 70% cocoa for all regions. Purus had the highest protein content for 70% and 81%, and lipid for 81%. Tocantins had the highest ash and lowest moisture for 70%, and the lowest lipid but highest protein content for 81% samples. Volatile compounds related to nutty, floral, and fruity were identified. Future research is recommended to optimize fermentation and roasting processes based on cacao origin to preserve desired flavors associated with higher consumer acceptability.
The Patch
Wysocki, Megan (Virginia Tech, 2026-05-21)
This thesis explores the impact of the embodied experience on cognitive development in early learning environments through the intersection of neuroscience and architecture. We as human beings spend 90% of our lives within architecture. Architecture should be designed to improve human quality of life and well-being. To do that, we must look at neuroscience to better understand the population we are designing for. Neuroscience has established, when you enter into a space, in milliseconds your brain has perceived the space, shifting your mood and attention/awareness. Children are our future, to bring about a positive change, this exploration investigated children, ages 3-6. During this period, children are developing executive functioning skills, while brain plasticity and curiosity is at its highest, They use their senses to absorb information about their environments. Which presents this period as an ideal time to gather if learning environments could be improved to enhance cognitive development? The architectural experience is multimodal, resulting in your senses intersecting or stitching together. The design approach contextualizes the stitch into a tangible entity. The Patch is a network of programmatic patches that are stitched together through contrast and sensory intersections. The stitches draw attention to architectural affordance and foster cognitive development through awareness of one's self and the world around them. Utilizing architecture as a catalyst for teaching students about a their environment, themselves, or their culture/community. Stitching takes place at every scale of the project, from location, learning philosophies, and architectural elements. The Patch Preschool becomes a quilt of memorable moments in architecture that teaches the next generations to come.
Enabling Small Language Models as Efficient and Capable Agents
Srivastava, Gaurav (Virginia Tech, 2026-05-21)
Most agentic systems today are built around large language models (LLMs) accessed through proprietary APIs (for example, GPT, Claude, and Gemini), which raises concerns about cost, latency, and privacy. This thesis argues that small language models (SLMs), typically under 30 billion parameters, can serve as efficient and capable agents when paired with the right system design choices. The work proceeds as a sequence of five studies that together support this case. We begin with ThinkSLM, a study of 72 small models across 17 reasoning tasks, which shows that training methodology and data quality drive reasoning more than parameter count. This motivates Debate, Train, Evolve (DTE), a self-evolution framework that turns multi-agent debate traces into reinforcement learning signals, improving small-model reasoning without ground-truth supervision and matching or surpassing the multi-agent system at single-model inference cost. The limits we observe in DTE prompt a closer look at how models allocate compute, leading to our overthinking analysis (LLMThinkBench), which shows that reasoning-trained models often produce around 18 times more tokens on basic math while achieving lower accuracy. To investigate memorization further, we develop BeyondBench, a contamination-resistant evaluation framework that algorithmically generates problems from combinatorial spaces of more than 10^15 instances across 44 tasks and 117 variations. Evaluating 101 models shows that hard-suite language-only performance remains low for many strong models, such as Gemini-2.5-Pro at 56.21%, while tool-augmented GPT-5 reaches 71.68% on the same suite, suggesting that agentic capabilities are a useful complement to raw scale. We synthesize these insights into EffGen, an open-source agentic framework built from the ground up for SLMs. EffGen contributes prompt optimization that compresses context by 70--80%, complexity-based routing, task decomposition into parallel and sequential subgraphs, a unified three-tier memory system, and the first unified implementation of the MCP, A2A, and ACP protocols. Across 13 benchmarks, EffGen consistently outperforms LangChain, AutoGen, and Smolagents in success rate, latency, and memory use. Together, these results show that with the right system design, small models combined with tools, memory, and intelligent orchestration can perform competitively with much larger models on a meaningful range of tasks. The contribution of this thesis is to identify the regimes in which SLMs are effective, to characterize where they break, and to provide an open framework that lets practitioners deploy them responsibly.
Optimization of a UAV for Wildfire Management
Hargan, Nathaniel Steele (Virginia Tech, 2026-05-20)
Wildfires pose a significant environmental risk, cause substantial economic damage, and are a danger to human life. Uncrewed aerial vehicles (UAVs) have the potential to support hotshot crews to combat wildfires more effectively. UAVs are valuable due to their low cost and high maneuverability. UAVs with the ability to carry large payloads can effectively transport fire retardant or propellant to create controlled burns remotely, mitigating the risk of uncontrolled wildfires. A UAV designed for this application should be able to carry a large payload, have enough battery capacity to remain in flight for the extent of a mission, and be rigid enough to resist vibration. The UAV must be large enough for the propellers and electrical components needed to lift the payload. The goal of this project is to design a conceptual model of an octocopter UAV and to examine the design space to find an optimal solution. The UAV has a 2-meter wheelbase and is designed to carry a 45kg payload. The UAV model is an 1/8-symmetric sector model of the full UAV, represented as a finite element model, and is used to estimate the deflection, stress, fatigue life, frequency response, and damage. The UAV is modeled as a 3D Timoshenko beam finite element model. The mass and mass moment of the UAV are minimized using non-linear programming.
Reinforcement Learning Benchmarking for Sustainable Energy Systems: Perturbation Robustness, Safety Constraints, and Multi-Agent Coordination
Koruturk, Mehmet (Virginia Tech, 2026-05-20)
Reinforcement learning (RL) has emerged as a promising approach for sequential decision making in sustainable energy systems, yet the lack of systematic benchmarks hinders principled algorithm comparison and deployment. This thesis presents SustainRL-Bench, a comprehensive benchmarking study that evaluates RL algorithms across three sustainable energy environments—electric vehicle (EV) charging coordination, building HVAC control, and cogeneration plant dispatch—along three experimental axes: perturbation robustness, safety constraints, and multi-agent coordination. Our evaluation spans 116 unique configurations totaling over 310 training runs. Several findings challenge conventional expectations. Perturbation sensitivity varies by over an order of magnitude across environments: EV Charging PPO degrades less than 11% under state perturbation, while Building policies collapse by over 100% at the same relative scale, and in Cogeneration, on-policy PPO remains far more robust than off-policy methods (<24% degradation across all channels vs. order-of-magnitude losses for SAC). Constrained MDP-based safe RL achieves precise constraint tracking only when the cost signal is sufficiently orthogonal to the reward; partial overlap permits tracking at tight limits, while full alignment renders constraints inactive, and off-policy Lagrangian methods fail entirely across all environments. In multi-agent settings, environment coupling structure rather than agent count determines algorithm feasibility, with the single-agent algorithm ranking reversing in tightly coupled domains. All code and configurations are publicly available.
The Impacts of Prescribed Burning on Vectors of Laurel Wilt in Post Outbreak Forests in South Carolina, USA
Allan, Shane (Virginia Tech, 2026-05-20)
Laurel wilt (LW) is a nonnative, invasive disease complex caused by the fungal pathogen Harringtonia lauricola and primarily vectored by the redbay ambrosia beetle (Xyleborus glabratus). This disease was first reported in the United States in 2002 and has caused the mortality of hundreds of millions of trees in the Lauraceae family. Currently, there are no successful management strategies for controlling LW at the forest scale, but prescribed burning presents a promising option for managing LW in fire prone ecosystems. To assess the efficacy of prescribed burning to control LW and its ambrosia beetle vectors, we conducted a study in a post-LW forest in South Carolina. We established multiple plots to determine the colonization and emergence rates of H. lauricola vectors and the population of ambrosia beetles (Coleoptera: Curculionidae) attracted to Lauraceae volatiles. Additionally, tissue samples were collected from redbay (Tamala borbonia) trees within plots to test for the presence of H. lauricola and track the incidence of LW before and after prescribed burns. Our results indicate that prescribed burns do not affect population dynamics of H. lauricola vectors or incidence of LW in forests where prescribed burns are regularly conducted. However, prescribed burning may influence host physiology, which impedes LW vectors host detection and reproduction.
Mapping and Modeling the Aerial Extent of Ipomoea Hildebrandtii Across Rangelands in Northern Tanzania
Babchak, Timothy Glen (Virginia Tech, 2026-05-19)
This study examines the spatial distribution of Ipomoea hildebrandtii Vatke (purple morning glory), a native invader in the semi-arid rangelands of northern Tanzania whose expansion appears to share important ecological and social dynamics with invasive species worldwide. Using high-resolution PlanetScope imagery (3 m) and a Random Forest classification approach, the study maps species distribution in Simanjiro District and evaluates its association with environmental and anthropogenic factors. Field observations collected in June 2025 supported training data development and image interpretation. The classification model achieved high performance (accuracy = 0.985, AUC = 0.994) and produced a continuous probability surface used to estimate spatial extent and patch structure. Results indicate that I. hildebrandtii occurs in highly fragmented patches across the landscape. Distance to roads emerged as the strongest predictor of occurrence. This relationship suggests that transportation infrastructure is associated with species distribution and may reflect the influence of disturbancerelated processes. Elevation and NDVI showed weaker positive relationships, while other variables were not statistically significant. These findings point to the importance of disturbance in shaping vegetation patterns and suggest that human and livestock movement along road networks may contribute to the expansion of I. hildebrandtii.
Understanding and Meeting Urban Energy Demand with Small Modular Reactors
Oredipe, Albert Oluwadamilare (Virginia Tech, 2026-05-19)
Rising expansion of data centers, largely driven by the information age as well as the increase in utilization of generative artificial intelligence has led to the development of challenges related to the growing energy demands of urban energy systems. Cities as they stand are required to not only accommodate for the energy demands associated with the increase in population and commercial needs, but now also must address the demand associated with large scale data center development. This thesis looks to evaluate the potential of small modular reactors (SMRs) as an alternative energy source that is capable of addresses these rising challenges. By using Loudoun County, Virginia as a case study due to it being a noticeable global epicenter of new data center development, this thesis looks to utilize a mixed-methodological framework which combines life cycle assessment, geospatial suitability analysis, and energy economics modeling to compare SMRs with conventional and renewable energy sources. Quantitative methods include assessing greenhouse gas emissions, land use, water consumption, levelized cost of energy, and geospatial constraints using GIS-based multi criteria decision analysis, while qualitative policy analysis examines the political and regulatory landscape currently governing SMR development. Findings from this thesis show that SMRs stand advantageous in respect to land efficiency, water consumption, long-term emissions, and can be economically competitive in high demand regions under synergetic political frameworks. Zoning and Conservation easement however present significant constraints on siting feasibility. The results from this thesis suggest that SMRs can play a significant role in meeting the future urban energy demands especially when co-located with datacenters if available provided that the supporting regulatory and policies evolve to accommodate growing and advancing nuclear technologies.
Fuels of the Southern Appalachian Fire and Fire Surrogate Study following Hurricane Helene and the Black Cove Wildfire Complex
Harris, Christopher Alan (Virginia Tech, 2026-05-19)
Wildland fuel loading estimates are used to approximate potential fire behavior when planning and implementing wildland fire management. Managers consider the relationships of fuel volume and structure to wildland fire behavior and resulting ecological responses. As active fire and fuels management becomes more widespread after decades of active fire suppression and limited or infrequent forest management, managers increasingly seek to incorporate prescribed fire for a range of stewardship goals and outcomes. As development in rural areas expands the wildland urban interface, managers may consider alternative fuel treatments in areas where prescribed fire may not be feasible or practical. Initiated in the early 2000s, the National Fire and Fire Surrogate Study (FFSS) compared the effects of prescribed fire to those of fire surrogate treatments on a variety of ecological responses, including fuels and potential fire behavior, at multiple locations across the country. For the southern Appalachian Mountains, the treatments implemented in the study included prescribed fire, mechanical felling (shrubs and midstory), a combination of prescribed fire and mechanical felling, and untreated controls. Over a period of 24 years, sites were repeatedly treated and monitored to determine the efficacy of each treatment in producing desired outcomes. Following 24 years of management and study, the FFSS site at the Green River Game Land, in Henderson and Polk counties, North Carolina, was impacted by two major disturbances over a six-month period – Hurricane Helene in September 2024 followed by a large wildfire in March 2025. From June to July 2025, fuel loading and iii structure at the site were measured and assessed to determine how changes in fuels may have differed between the treated areas. Our results suggest that, for the period between 2014 and 2025, decreases in ground fuel (O Horizon) mass differed between the treatments. The decrease was greatest in the control treatment, which did not differ from the mechanical-only treatment, but did differ from the burn-only and mechanical+burn treatment. Fine woody fuel mass decreased in all treatments and did not differ between the treatments. Coarse woody debris only decreased in the burn-only and mechanical+burn treatments, but differences between the treatments were not statistically significant. Currently, the mean total fuel mass range for the control, mechanical-only, and mechanical+burn treatments is 45.32-50.60 Mg ha-1. The burnonly mean total fuel mass is 26.93 Mg ha-1, nearly half of what is present in the other treated or control areas. Understanding the long-term fuel dynamics at this location will provide critical information for the southern Appalachian Mountain region as scientists and managers consider fuel reduction treatments to alter fuel complexes, with and without storm impacts from Hurricane Helene.
From Shock to Strategy: How Planners and Practitioners Conceptualize Food System Resilience Across Place
Goodman, Zenobia Makini (Virginia Tech, 2026-05-19)
Food systems planning has become a significant part of the field of urban and regional planning. However, as the world continues to change and shocks occur due to significant stressors, planners and practitioners must be creative with how they support the local food system and respond to these shocks. Through the employment of semi-structured interviews with various planners and practitioners affiliated with the American Planning Association Food Division (APA FOOD), this study employed qualitative methods to identify various epidemiological, , climate and politico-economic shocks to the food system. This is done while investigating the shocks through the lens of geographical diversity, institutional capacity, and community-based collaboration. This study found that food system resilience is conceptualized and operationalized through place-based, collaborative strategies shaped by regional conditions, institutional capacity, and the interconnected nature of social, economic, and environmental shocks. This research underscores that food systems are not merely technical or logistical challenges, but social and relational systems that reflect broader questions of equity, power, and belonging.
Hydrogel-Integrated MIM-SERS Platforms for In-Situ, Spatiotemporal Profiling of E. coli Biofilm Responses to Chemo-Physical Perturbation
Zong, Ze (Virginia Tech, 2026-04-07)
Monitoring biofilm responses to antimicrobial and physical interventions requires in situ, minimally perturbative sensing in tissue-like environments. Here, we present a hydrogel-integrated surface-enhanced Raman spectroscopy (SERS) platform in which Escherichia coli is cultured beneath an LB–agarose layer on a nanolaminated metal–insulator–metal (MIM) substrate, enabling label-free, spatiotemporal readouts under chemo-physical perturbations. Time-resolved spectra (785 nm) were acquired at 1, 11, 24, 24*h (immediately post-intervention), and 36 h over 4 mm² maps (~400 spectra per map), with electronic Raman scattering (ERS) used as an internal standard to reduce instrumental drift. Interventions introduced at 24 h included ampicillin (AMP), femtosecond (fs) laser-induced nanobubbles (950 nm), or their combination. Spectra consistently exhibited a band near 445 cm⁻¹, tentatively associated with carbohydrate-rich EPS components and a feature near 732 cm⁻¹ that is adenine-associated with a substantial hydrogel contribution. Principal component analysis (PCA) and linear discriminant analysis (LDA) discriminated timepoints and treatments with high apparent accuracy (leave-one-out cross-validation, LOOCV >95%), while 2D maps visualized treatment-dependent heterogeneity. Immediate global intensity increases at 24*h were consistent with transient hotspot re-exposure of plasmonic hotspots after fs excitation. Overall, this study provides a method-focused validation of hydrogel-integrated MIM-SERS for real-time biofilm profiling and motivates replicate-level validation, orthogonal biological readouts, and translation toward flexible device formats.

Browse

Recent Submissions