Electronic Theses and Dissertations (ETD)

Permanent URI for this collection

https://dspace.library.uvic.ca/handle/1828/12

For information on how to submit your thesis to this collection, please go to our ETD website on the UVic Libraries Website.

Access to the full text of some theses may be restricted at the request of the author.

All theses from 2011 to the present are in this collection, as well as some from 2010 and earlier years.

Browse

Now showing 1 - 20 of 11719

Creating representative skeletal cohorts through statistical modelling and synthetic CT image generation
(2025) Beagley, Aren; Giles, Joshua W.
Finite Element (FE) analysis is an important tool for orthopaedic research that allows studying the effects of orthopaedic devices, such as joint replacements, in ways that would be difficult or impossible to investigate experimentally. High fidelity FE studies are typically performed using a cohort of subject-specific FE bone models created from Computed Tomography (CT) images to ensure realistic bone shapes and material properties. However, this limits sample sizes because acquiring CT images exposes subjects to harmful radiation and such exposure should be avoided unless medically necessary. Unfortunately, small sample sizes create significant limitations on the applicability and generalizability of the insights provided by most FE studies. Ideally, FE studies should use sufficiently large and diverse cohorts to ensure that the resulting insights generalize to the broader population. If the population-level distribution of shape, size, and stiffness for a bone was known prior to conducting the FE study, subjects could be systematically chosen to create a representative cohort; however, even the process of choosing these subjects can be a challenge. Statistical Shape and Intensity Models (SSIMs) are an established tool for characterizing the population-level variance of size, shape, and material properties, and these models can be created from pre-existing, medically necessary, CT images. Furthermore, SSIMs can generate new instances that are representative of the population. Unfortunately, previous methods of developing SSIMs do not capture sufficient detail and produce models incapable of generating new instances suitable for use in FE studies. This thesis describes the development and validation of a method for creating high resolution SSIMs capable of generating new instances that contain data comparable to that of CT images. An associated method for converting SSIM-generated instances into SSIM-derived synthetic images, that are comparable to CT images, was also developed and validated. In combination, these two methods result in a model capable of systematically sampling a population-level distribution to create representative cohorts suitable for use in FE studies. To determine how representative the combined method is, generalization of both SSIM-generated instances and SSIM-derived synthetic images were assessed by comparing shape and material properties against real subjects. SSIM-generated instances were assessed to have Mean Generalization Root-Mean-Square (MGRMS) errors of 2.15 mm and 228.1 Hounsfield Units (HU) for shape and material properties respectively, and an average surface distance (ASD) of 1.145 mm. After converting SSIM-generated instances to synthetic images, the MGRMS error for shape could not be assessed, but for material properties it increased to 286.2 HU (+58.1) and the ASD showed improvement by decreasing to 1.028 mm (-0.117). These results demonstrate that the high resolution SSIM presented in this work has similar generalizability as previously published SSIMs while capturing significantly greater variance and that the conversion to synthetic images introduces minimal additional error. As such, the combined high resolution SSIM and conversion algorithm developed for this thesis represent a viable method of improving generalization for FE studies by overcoming the current barriers to systematically producing representative cohorts. This work set the stage for future work investigating the impacts of conducting FE studies with SSIM-derived synthetic images and addressing a range of clinically relevant biomechanical questions.
Local convergence of grounded lipschitz functions on d-ary trees
(2025) Butler, Nathaniel; Ray, Gourab
We consider the uniform sampling of grounded M-Lipschitz functions on the d-ary tree with n levels, with special interest as n → ∞. In the case M = 1, it was shown in [2] that this sampling converges weakly (in the infinite d-ary tree) iff 2 ≤ d ≤ 7. We continue this work by putting the computations into a form that a computer can handle, and we use this to confirm convergence for several other values of M and d. As in [2], the main idea is use the recursive structure of the d-ary tree to reduce the problem to studying the fixed points of a certain function on ℓ∞(N). In [2], the authors also showed an even-odd phenomenon for M-Lipschitz functions on any infinite bipartite graph with ‘rapid expansion’ (i.e. with sufficiently large Cheeger constant). Specialized to our original problem of grounded M-Lipschitz functions on Tn d, this shows that the samplings for n even and n odd both converge, but to separate limits when d ≫ M logM. We reproduce this proof here.
Machine-learning framework to identify and validate biochemical regime clusters in the global blue carbon ecosystem
(2025) Singh, Bhan; Popli, Navneet; Sima, Mihai
The Earth’s climate system is undergoing profound transformation, driven by changes in natural and anthropogenic stressors that disrupt environmental balance across land, air, and sea. Among these domains, the ocean stands as both a stabilizer and a sentinel, absorbing excess heat and carbon while revealing the earliest signs of ecological stress. Yet, the ocean itself is changing, shaped by interacting forces such as temperature, salinity, oxygen depletion, depth stratification, and biological productivity. Understanding how these stressors combine to reshape marine ecosystems requires not just observation but intelligent pattern recognition. This thesis approaches the problem as one of learning structure within complexity. Rather than relying on political boundaries or fixed geographic regions, it asks: can we allow the data itself to define the ocean’s natural divisions? Using in-situ observations from the World Ocean Database (WOD), a machine-learning framework was developed to uncover underlying biogeochemical regimes, clusters of ocean states defined by their physical and chemical signatures. Through careful preprocessing and hierarchical spatialtemporal imputation, the dataset was refined to reflect true environmental variability rather than sampling noise. The analysis employed multiple clustering algorithms to let ocean data “self-organize,” followed by classification models that validated and explained the separability of the discovered regimes. This hybrid approach revealed five coherent and interpretable patterns corresponding to familiar yet dynamically interconnected oceanic systems: productive coastal upwellings, oligotrophic gyres, polar waters, oxygen-minimum zones, and transitional open-ocean regimes. Together, these patterns tell a story of a living ocean, one organized not by political maps, but by the natural language of its own chemistry and biology. By combining unsupervised discovery with supervised validation, the research demonstrates how global ocean observations can be transformed into quantitative, interpretable indicators of ocean health. The resulting framework contributes to the emerging vision of a digital twin ocean, a system where data, models, and machine learning work together to monitor, predict, and ultimately safeguard the resilience of the planet’s largest ecosystem.
The definition of head and the syntactic structure of verbs in the composition of Yorùbá serial verb constructions
(2025) Ariyo, Oluwabukola Oluwaseun; McGinnis, Martha
I investigate serial verb constructions (SVCs) in Yorùbá within the Minimalist framework, addressing two fundamental questions: (i) which verb serves as the syntactic head of the extended projection associated with the SVC in Yorùbá, and (ii) what hierarchical relation exists between verb phrases in these constructions. Focusing specifically on SVCs in which both verbs are transitive and select distinct internal arguments, this dissertation employs multiple empirical and syntactic diagnostics to establish the head of these complex predicate formations. The analysis shows that the first verb (V1) functions as the head of the extended projection in Yorùbá SVCs. Three primary lines of evidence support this determination: (i) verb nominalization and clefting operations, which systematically target V1, while excluding the second verb (V2); (ii) adverbial modification patterns, wherein manner adverbs scope exclusively over V1, despite the ability to modify any verb in simple Yorùbá clauses; (iii) the distribution of functional categories including aspectual markers, negation, and modals, which appear only before V1. Drawing on Chomsky's bare phrase structure theory (1995, 2000) and Stepanov's late adjunction hypothesis (2001, 2007), I establish the structural properties of the verbs in the SVC. I show that VP2 is an adjunct to the VP1, rather than a complement. This conclusion is substantiated through the examination of extraction asymmetries, where wh-movement and focus movement proceed freely from V1. and its complements (including DP, PP, infinitival CP, and finite CP complements), and extraction from VP2 is categorically blocked regardless of the syntactic category or structural position of the displaced element. This extraction behavior suggests that VP2 is in an adjoined position, making it inaccessible to syntactic movement. Additional evidence from adjunct placement possibilities and reflexive binding across V1 and V2 shows that the object DP of V1 cannot be an antecedent to the object DP of V2, corroborating this structural analysis. This work contributes to the cross-linguistic understanding of SVCs by demonstrating that the mono-clausal properties characteristic of SVCs, including unified event structure, shared external arguments, and single tense, aspect, and mood specifications, can derive from adjunction configurations rather than being exclusively derived from the complement structure. This research advances both the descriptive understanding of Yorùbá syntax and theoretical discussions concerning headedness, complement/adjunct distinction, and complex predicate formation.
Sustaining student motivation and well-being: Academic and non-academic pressures, supports, and coping
(2025) Garavellos, Victoria; Cunningham, Bart
The purpose of this report is to gain a better understanding of how academic and non-academic pressures affect student well-being and how these pressures influence their academic performance. A literature review was conducted to explore the meanings of stress and student well-being, the effects of stress and stress management, and university wellness supports available to students. From the literature review, a conceptual framework was created to capture academic and non-academic pressures, motivations and demotivations, and faculty and peer support. This framework was applied to the interview guide. Semi-structured interviews were conducted with current and former university students to gather insight into their academic and non-academic needs. Interview data were analyzed to identify key themes from participant responses. Participants described several challenges and supports they encountered during their time in university, including heavy workloads, financial pressures, and the value of peer and personal relationships. The study highlights the importance of understanding the diverse stressors students face and offers recommendations to improve institutional support and promote student well-being.
Tracking uncertainty in knowledge graphs using Kalman filtering
(2025) Tkachenko, Alina; Thomo, Alex
Knowledge graphs (KGs) represent structured knowledge as networks of entities and relations, forming a foundation for reasoning in artificial intelligence. To make these symbolic structures usable by machine learning systems, knowledge graph embeddings (KGEs) map entities and relations into continuous vector spaces. However, traditional KGE models are typically static and deterministic. They treat all facts as equally certain and require full retraining when new data arrive, making them unsuitable for evolving, uncertain knowledge. This thesis introduces a new framework that reframes knowledge graph embedding as an online state estimation problem. By integrating the Kalman filter, a recursive algorithm that updates beliefs under uncertainty, into KGE training, the proposed approach enables continuous and uncertainty aware learning of entity and relation representations. The framework treats each embedding as a probabilistic latent state, updated incrementally as new triples arrive, blending prior knowledge with new, possibly noisy, observations. Two models instantiate this framework. KalmanKG2E extends the probabilistic Gaussian embedding model KG2E with Kalman-based online updates of means and covariances. KalmanComplEx adapts the non-probabilistic, complex-valued ComplEx model to a dynamic, uncertainty-tracking setting. Together, these demonstrate the frameworks generality across fundamentally different embedding architectures. Extensive experiments on six benchmark datasets show consistent improvements over static baselines. The Kalman-based models converge faster, achieve higher predictive accuracy, and exhibit greater robustness in sparse, evolving graphs. These results validate Kalman filtering as a principled and efficient mechanism for online knowledge graph learning. Overall, this work bridges classical state estimation and modern representation learning, advancing knowledge graph embeddings from static snapshots to dynamic, continuously adaptive systems that better reflect the evolving nature of real-world knowledge.
BBAE: Bit-to-byte alignment with entropy analysis for binary protocol field identification
(2025) Zhang, Leijie; Wu, Kui
Protocol Reverse Engineering (PRE) is crucial for analyzing undocumented or proprietary network protocols, particularly in the fields of network security and the Internet of Things (IoT). To conserve network bandwidth, many protocols adopt a compact binary format that maximizes bit utilization. However, this compactness introduces significant challenges for PRE, because (1) the number of potential field boundaries grows exponentially, and (2) byte-oriented PRE tools become ineffective for these scenarios. To address these challenges, we propose Bit-to-Byte Alignment with Entropy (BBAE) analysis, an innovative approach designed to enhance boundary detection in bit-oriented protocols. BBAE leverages entropy analysis and bit-congruence calculations across multiple window sizes to identify field boundaries more effectively. In addition, it enables systematic verification of detected boundaries. We conducted extensive evaluations of BBAE’s performance in identifying field boundaries of binary protocols and compared its effectiveness with existing tools, including byte-oriented semantic inference tools like BinaryInferno and bit-oriented tools such as Auto-ETLV. Our experimental results disclose that BBAE achieves outstanding performance in reverse engineering binary network protocols.
An interpretive analysis of the effectiveness of non-traditional or ‘Structured Discovery’ blindness rehabilitation in Canada from the perspective of blind service recipients and teachers
(2025) Lalonde, Elizabeth; Wiebe, Sarah Marie
Blindness rehabilitation in Canada has traditionally emphasized maximizing residual vision and functional skills through prescriptive, vision-centered methods. While these approaches provide supports, they often reinforce dependency and limit adaptability for people facing progressive vision loss. Structured Discovery, an alternative model that originated in the United States, reframes blindness as a characteristic rather than a deficit and emphasizes non-visual skill development, problem-solving, and empowerment through the mentorship of blind instructors and peers. Despite its influence in the United States, Structured Discovery is largely absent from both practice and scholarship in the Canadian context. This thesis explores how blind Canadians experience Structured Discovery training as participants and teachers. Using Interpretative Phenomenological Analysis (IPA) and grounded in Critical Disability Theory, the study examines how individuals make sense of their training experiences, the skills and perspectives they gained, and how they compare Structured Discovery to more traditional rehabilitation services. The researcher conducted ten qualitative interviews with Canadians directly engaged in Structured Discovery programs, including those delivered through the Pacific Training Centre for the Blind (PTCB), one of the few Canadian organizations applying this model. This study makes several contributions. It represents the first academic examination of Structured Discovery in Canada and addresses a gap in both disability studies and rehabilitation research. It provides practical insights for rehabilitation practitioners by showing how empowerment-based, non-visual training can better prepare blind people for independence and social participation.
Reinforcement learning based resource allocation in fog computing
(2025) Mokhtari, Masoud; Ganti, Sudhakar
The Internet of Things (IoT) has revolutionized connectivity by enabling seamless data exchange among diverse devices, fostering intelligent services and informed decisionmaking. However, the rapid surge in data traffic has exposed the limitations of traditional cloud-based solutions, particularly in meeting the quality-of-service (QoS) demands of latency-sensitive applications. Fog computing has emerged as a transformative paradigm, extending computational resources closer to end-users and bridging the gap between centralized cloud systems and edge devices. This approach addresses QoS challenges by providing critical services and resources at the network’s edge. Despite its advantages, fog computing faces resource limitations at the node level, necessitating efficient resource allocation to optimize performance and meet application-specific QoS requirements. Deciding whether to process data at the fog or cloud level involves navigating complex trade-offs dictated by resource availability, offloading criteria, and diverse application scenarios. This thesis addresses these challenges through a comprehensive approach to resource allocation in fog and cloud computing environments. First, a reinforcement learning-based method is introduced to optimize resource allocation for a single fog node. By formulating the problem as a Markov Decision Process (MDP), the approach maximizes fog resource utilization while considering the number of resource blocks and delay tolerance for each request. Experimental evaluations demonstrate the superiority of the E-SARSA algorithm in terms of speed, utilization, and adaptability compared to Q-learning, SARSA, and a Fixed-Threshold approach. The study then extends to multi-fog/cloud systems, introducing a two-phase process. In the first phase, the optimal fog node for resource allocation is identified. In the second phase, reinforcement learning is applied to determine whether tasks should be processed locally or offloaded to the cloud. This method ensures efficient resource utilization, with experimental results highlighting the superior performance of the Selection-2 approach compared to Genetic Algorithms (GA), Round Robin (RR), and Random strategies, particularly in speed, utilization, and load balancing. Finally, the framework is further enhanced with a hybrid approach combining Genetic Algorithms and Reinforcement Learning (GA/RL) for dynamic resource allocation in integer-based multi-fog/cloud systems. This method applies the two-phase process, achieving significant improvements in speed, utilization, and load balancing compared to existing methods. By dynamically allocating fog resources and optimizing offloading strategies, this work addresses the limitations of traditional cloud computing systems and ensures seamless performance for latency-sensitive IoT applications. The proposed approaches advance resource allocation strategies in fog and cloud computing, offering scalable, efficient, and adaptive solutions for future IoT ecosystems.
Advancing cell-type annotation and deconvolution in human bronchoalveolar lavage through single-cell transcriptomics and benchmarking protocols
(2025) Hu, Yushan; Zhang, Xuekui; Shao, Xiaojian
Bronchoalveolar lavage (BAL) provides a unique view to analyze immunological aspects of the human lung. Single-cell RNA sequencing data (scRNA-seq) of BAL offers great potential for immunotherapy of lung diseases. Despite promising challenges, persist in identifying disease-relevant high-resolution sub-cell types, standardizing annotations across studies, and accurately interpreting bulk RNA-seq data. This dissertation is approached from three perspectives. Chapter one serves as the introduction, while chapter two provides the background. Chapter three presents scRNA-seq data utilized to characterize macrophage and monocyte populations in chronic obstructive pulmonary disease (COPD). The analysis identified dysfunctional alveolar macrophages and hyperinflammatory monocytes, indicating potential therapeutic targets and emphasizing the modulatory effects of inhaled corticosteroids. The fourth chapter presents a standardized atlas of human BAL cells through the synthesis of multiple scRNA-seq datasets, with ensemble auto-annotation tools and reliable cross-study markers. This atlas deals with discrepancies in previous studies and provides a foundation for BAL research. Chapter five introduces the first true-paired benchmarking study of cellular deconvolution in BAL. Including 30 human BAL samples, each divided into bulk RNA-seq and matched single-cell libraries. After systematically evaluating 15 popular algorithms across multiple references and cell-type resolutions, this study demonstrates a modestly designed pairing strategy substantially improves both benchmark realism and practical accuracy. Our true-paired data, comparative analyses, and three-step protocol provide a blueprint for future deconvolution studies. Together, these chapters deliver disease insights, community resources, and methodological frameworks that advance the study of lung immunity through both single-cell and bulk transcriptomics.
Understanding the role of insulin signaling in female reproductive ageing
(2025) Athar, Faria; Templeman, Nicole M.
Preserving female reproductive health is crucial for maintaining the survival and sustenance of a species as well as for overall health and well-being. Reproductive health is mutable and has a strong life-history basis; two of its important regulators are chronological age and nutrition. Insulin signaling is an evolutionarily conserved mechanism for interpreting nutrition levels. To better understand the role of this pathway, I leverage its conservation to study impacts on reproductive function in a cross-species approach using human, mouse and Caenorhabditis elegans data. Analysis of longitudinal data from the Study of Women's Health Across the Nation (SWAN) revealed that women with higher fasting insulin levels at mid-life experienced an earlier onset and longer duration of vasomotor symptoms, independent of body mass index. To test the causal role of insulin in female reproductive ageing, I conducted longitudinal analysis of a mouse model and found that reducing endogenous levels of insulin production protects against high-fat, high-sucrose-induced reproductive dysfunction. Insulin-reduced dams maintained higher pregnancy rates and ovarian reserve compared to hyperinsulinemic littermates, despite similar levels of glycemia. In C. elegans we showed that glucose enrichment accelerates reproductive ageing by compromising oocyte quality and altering mitochondrial dynamics. However, reducing insulin signaling through mutation of the daf-2 insulin-like receptor protected against reproductive decline, even though it did not mitigate somatic ageing. I also found that insulin signaling in the intestine and body wall mediates reproductive ageing through non-autonomous mechanisms. Together, these studies provide evidence of insulin signaling being an evolutionarily conserved regulator of reproductive ageing.
Equity, decolonization, and the urban forest: Exploring Indigenous-led urban forest planning practices in the Capital Regional District
(2025) Stoltz, Sydney; Wiebe, Sarah Marie
Equitable access to urban green spaces is vital for citizen health, climate change mitigation, and reconciliation. However, the urban forest planning processes in British Columbia’s Capital Regional District (CRD) currently do not adequately support Indigenous inclusion, knowledge, and self-determination. This deficiency in planning impedes efforts to achieve urban forest equity, decolonization, and reconciliation. Addressing this issue is essential to ensuring that urban forest management is inclusive, equitable, and respectful of Indigenous perspectives. This Master’s thesis examines potential barriers, best practices, and approaches to collaborative urban forest policy within the CRD in order to advance greenspace equity, decolonization, and reconciliation for all residents. Promoting collaborative urban forest planning policy is supported under B.C.'s Declaration on the Rights of Indigenous Peoples Act Action Plan, which outlines a framework for the province and municipalities to fulfill the goals of the United Nations Declaration. While the CRD facilitates regional decision-making and positive relationships with local Indigenous communities, it currently lacks specific policies for Indigenous participation in greenspace policy and planning. Using interpretive policy analysis, thematic analysis, and a critical policy lens, this thesis reviews findings from jurisdictional scans, a literature review, and eight interpretative interviews with Indigenous and non-Indigenous community members to determine potential pathways towards collaborative urban forest planning. The collective findings suggest that there are several approaches that the CRD (or the municipalities within the region) could adopt in order to increase Indigenous inclusion in local urban forest planning. Participants emphasized the need for shared priorities, engaging early and often, relationship-building, and clear communication. Key barriers included considerations around working within ongoing colonial systems, such as honoring Indigenous cultures and traditions in ways that are non-extractive or appropriative, ensuring continuity in work and relationship-building, and working within potential funding constraints. Preferred approaches emphasized proper engagement (such as through establishing protocols in the early stages), ensuring that all voices and concerns are heard equally, and an emphasis in bringing knowledge together in a relational way rather than an extractive one. Through analysing these findings, this thesis presents several short, medium, and long-term recommendations to increase education and capacity-building within government, continue to build relationships with local First Nations, and create ongoing spaces for co-governance in urban forest planning at the regional level in order to foster improved collaboration and equity. This thesis advances local regional efforts towards reconciliation, sustainability, and environmental equity by identifying existing barriers and proposing potential ways forward through collaboration built on trust and partnership.
How a hot topic affects governance in British Columbia: An examination of extreme heat exposure emergency management planning
(2025) Emenike, Jumai; Wiebe, Sarah Marie
This study explores how British Columbia (BC) is addressing extreme heat under the new Emergency and Disaster Management Act (EDMA, 2023), with a focus on equity and planetary health. The research analyzes provincial policy documents and draws on interviews with actors from health, emergency management, housing, and climate adaptation at both provincial and municipal levels. Using interpretive policy analysis, the research explores how EDMA provisions are understood across the four phases of emergency management (preparation, response, mitigation, and recovery) and how existing extreme heat measures address heat vulnerability and adaptation. Findings reveal notable advances in preparedness and public health response, but significant gaps in mitigation, recovery, and cross-ministerial coordination. While policy documents officially acknowledge equity through tools such as heat vulnerability maps and the Population Environmental Risk Characteristic (PERC) file, implementation is limited by built environment constraints and institutional fragmentation. Planetary health considerations, including but not limited to ecological impacts and nature-based solutions, are recognized but remain marginal. The study concludes that EDMA provides policy opportunities, but effective heat governance requires a coordinated, multi-sectoral approach.
Memories from the land of amnesty: Historical narratives of the armed right in Brazil
(2025) Santana Bertho, Ana Paula; Milton, Cynthia E.
In 2018, Brazilians elected the far-right candidate Jair Bolsonaro, a retired military captain, as President. His open praise of the military dictatorship (1964–1985) and the support it received among civilians called into question the hegemony of victim-centered memories about that period. This thesis dialogues with this context and aims to investigate the role of the armed right memories in contemporary Brazilian democracy, focusing on how these narratives have shaped public discourse and national identity. Drawing on Ksenija Bilbija and Leigh A.Payne’s concept of “memory market,” the study analyzes two memory products: the commemorations of March 31st (chosen by the military as the date of the military coup d'état of 1964) between 2014 and 2022 and the Army’s Historical Museum and Fort Copacabana in Rio de Janeiro. Based on the examination of newspaper coverage, government documents, interviews, and exhibitions, this study argues that the military has sought to refashion its role in the new democracy, reaffirming its authoritarian saviour role while attempting to engage with the era of human rights speech. The rise of “uncivil groups” after 2013 empowered the authoritarian nostalgia in the public sphere, mobilizing symbols of the past to justify authoritarian projects in the present. Bolsonaro emerged as a spokesperson for these actors. His government represented the institutionalization of the right armed narratives, paradoxically generating tensions with some military sectors while instrumentalizing these memories as fuel for the storming of Brasília in January 2023.
Constructing disability in policy: The discursive construction of disability in the Accessibility for Ontarians with Disabilities Act and the Accessible British Columbia Act
(2025) Fortin, Ashley; Wiebe, Sarah Marie
This research examines how cultural and political biases are embedded within Canadian accessibility legislation, focusing on the Accessible British Columbia Act (ABCA) and the Accessibility for Ontarians with Disabilities Act (AODA). It is guided by two research questions: 1. What biases are present in the AODA and ABCA, and how do they impact disabled individuals in these provinces? 2. What specific language and terminology in the AODA and ABCA express these biases? The study employs a qualitative, interpretive design that combines Critical Discourse Analysis (CDA) with a comparative case study of British Columbia and Ontario. Using Fairclough’s CDA framework, the research reveals the powers and ideologies embedded in accessibility legislation and examines how subsequent regulations and amendments may influence accessibility for disabled individuals. Five key biases were identified within the AODA: (1) structural and authoritative bias, (2) ambiguity and neoliberal governance, (3) symbolic inclusion, (4) economic framing, and (5) enforcement and compliance. Four primary biases emerged within the ABCA: (1) equality over equity, (2) tokenism and volunteerism, (3) technocracy, temporality, and optimism, and (4) ambiguity and authority. Each theme is illustrated with examples from the legislation and analyzed through a critical discourse lens. For practice, the findings underline the need for policymakers to shift from reactive, compliance-driven approaches to proactive, rights-based frameworks that anticipate and prevent barriers before they arise. Current systems often place the burden on disabled individuals to identify barriers and file complaints, rather than embedding accessibility as a baseline expectation. From a research perspective, the comparative analysis of the AODA and ABCA highlights the value of examining accessibility legislation across jurisdictions rather than in isolation. Expanding this approach to other provinces and territories could provide a more comprehensive understanding of how disability is discursively constructed in law across Canada.
Development of a secure underwater sensor suite for real-time environmental monitoring of blue carbon ecosystems
(2025) Singh, Rudra Pratap; Popli, Navneet; Dong, Xiaodai
The health of Canada’s blue-carbon ecosystems—kelp forests, seagrass meadows, and salt marshes—plays a vital role in marine biodiversity and long-term carbon sequestration. Yet these ecosystems are increasingly vulnerable to anthropogenic and natural stressors such as temperature variation, pH fluctuations, heavy-metal pollution, and hydrocarbon extraction. Traditional monitoring methods, relying on sporadic field sampling and manual analysis, fail to capture the temporal and spatial complexity of these changes. This thesis, Development of Machine Learning-Based Techniques for Monitoring and Analyzing the Effects of Natural and Manmade Stressors on Canada’s Blue Carbon Ecosystem Using a Secure Underwater Communication Suite, presents a comprehensive hardware-driven approach to address these gaps. The research involves the design, fabrication, and laboratory validation of a modular underwater sensor suite deployed via a Blue Robotics ROV platform to collect high-resolution oceanographic data. The integrated system measures temperature, salinity, dissolved oxygen, pH, turbidity, and chlorophyll concentrations through a network of calibrated probes, ensuring precise and repeatable environmental sensing. To support continuous operation, a secure underwater communication and data-handling framework was developed using a hybrid Ethernet-acoustic link and lightweight encryption protocols to preserve data integrity and mitigate cyber vulnerabilities within the Internet of Underwater Things (IoUT). Extensive laboratory testing in controlled aquatic environments demonstrated stable sensor calibration, minimal noise drift (< 0.05% FS), and consistent data throughput at depths up to 1 m. Complementary studies explored intrusion detection and federated-learning frameworks for distributed underwater nodes, strengthening the resilience of the proposed communication network. The system enables near-real-time environmental monitoring and data synchronization between underwater nodes and surface control units. By combining reliable hardware sensing with secure data transport, the work advances Canada’s capacity for sustained observation of blue-carbon habitats. The results contribute both an open hardware architecture for scalable underwater sensing and a validated communication protocol for secure marine data acquisition foundations that can inform future autonomous monitoring networks and adaptive management strategies for coastal ecosystems.
Persistent changes in coral-microalgal symbioses with climate change-amplified heat stress
(2025) Buzzoni, Daisy; Baum, Julia Kathleen
Climate change-amplified marine heatwaves pose the greatest threat globally to the future of coral reefs. The algal symbionts (family Symbiodiniaceae) hosted by corals are intimately linked to host physiology and as such the composition of corals’ symbiont assemblages can have dramatic consequences for their survival under stress and resilience through bleaching events. This dissertation addresses persistent changes in symbiont assemblages using empirical datasets from study sites characterising exposure to globally relevant stressors. The island of Kiritimati (central Pacific) harbours reefs across a gradient of local human disturbance and was the epicentre for the 2015-2016 tropical Pacific marine heatwave, whilst the reefs of Palau (Micronesia) have experienced consistent local differences in historical marine heatwave exposure. Prior to analysing datasets from these study sites, I review evidence of the effects of climate change on the interactions between eukaryotic hosts and their microbial symbionts from a range of ecosystems to better inform hypotheses involving coral-Symbiodiniaceae responses: I summarise ongoing shifts in diversity, flexibility, and degree of parasitism in terrestrial and aquatic symbioses and reveal a pervasive lack of longitudinal data tracking long-term symbiotic changes or stability. Then, using ITS2 DNA metabarcoding to characterise Symbiodiniaceae in coral tissue samples from Kiritimati collected from 2014 to 2019 and spanning the 2015-16 marine heatwave, I reveal greater temporal and spatial flexibility in the symbiont taxa hosted by coral species that inherit their symbionts horizontally (from their environment) compared to those that inherit symbionts vertically (from their parents). The heatwave-induced changes detected over 5 years within corals’ symbiont assemblages were smaller with increasing human disturbance in corals with horizontal symbiont inheritance and greater in those with vertical symbiont inheritance, exposing contrasting effects of multiple stressor exposure on symbioses for these two coral life histories. I then report findings from a decade-long time series which revealed factors shaping the recovery trajectories of heatwave-transformed coral-Symbiodiniaceae associations for one coral species on Kiritimati. The extreme intensity of acute heat stress exposure disrupted reversion to pre-heatwave symbiont assemblages, promoting the dominance of stress-tolerant generalist symbionts, with differential local disturbance exposure shaping the succession of recovered assemblages through intermediary symbiont community compositions. Finally, in a coral species with comparatively more specific symbiont associations in Palau, I uncover increased symbiont diversity at fine taxonomic scales associated with increased historical heat stress exposure spanning more than 20 years. Experimental heat exposure revealed that the symbiont taxa driving this increased diversity were linked to low heatwave tolerance in corals from historically more heat-stressed reefs, but not in corals from historically less heat-stressed reefs, reflecting a putative interaction between symbiont function and chronic heat stress. This dissertation addresses uncertainty surrounding the fate of climate change-transformed symbioses by assessing the long-term changes in other symbiotic study systems, the spatial and temporal flexibility in coral-symbiont associations over multiple years, the persistence of heatwave-transformed symbiont assemblages, and the links between chronic historical heat stress, Symbiodiniaceae, and heatwave tolerance. This research has applicability to coral reef conservation through improved understanding of the mechanisms driving natural variation in bleaching resilience and addresses foundational questions in symbiosis ecology under climate change and human disturbance, with relevance to other symbiotic systems.
Modelling flood disruptions to urban transport: A spatio-temporal lens using coupled hydrodynamic–traffic models
(2025) Rebally, Aditya; Valeo, Caterina
The transportation sector is an essential pillar of both economic prosperity and social well-being, and its functionality and resilience are increasingly challenged by the impacts of climate change. Transportation systems are directly and indirectly affected by extreme climatic scenarios on a range of spatial and temporal scales, with floods and heavy rainfall being the most critical hazards. In recent decades, urban regions around the globe have experienced notable increases in flood intensity and frequency. Such extreme events can significantly strain transportation networks in the short term through congestion, delays, and trip cancellations; while also producing medium- and long-term impacts associated with infrastructure damage, system recovery, and cascading disruptions that reverberate across economic and social systems. The present research reviews and advances the understanding of how flooding affects transportation networks across different timescales. Flood effects are classified according to their connection to both the type of flooding and the nature of impact whether direct, indirect, or cascading on the transportation system. Existing literature demonstrates that most studies concentrate on assessing direct and tangible impacts, typically emphasizing short- and medium-term resilience at smaller spatial scales. By contrast, there is relatively limited attention given to indirect or intangible consequences, or to longer-term temporal horizons where recovery, adaptation, and broader socio-economic feedback become more apparent. This imbalance highlights a gap in both methodological approaches and conceptual frameworks, particularly when considering how multiple stressors such as rainfall and flooding interact to magnify disruptions. To address these gaps, this dissertation applies a combined hydraulic and traffic modeling frameworks to capture the compounded effects of rainfall and flooding on transportation. The 2013 flood in the City of Calgary is selected as the case study considering its severity, dual riverine sources (the Bow and Elbow Rivers), and its well-documented impacts on both urban systems and transportation infrastructure. A hydraulic model (HEC-RAS®) is used to simulate flood dynamics, while a traffic microsimulation model (SUMO®) is employed to replicate traffic conditions under four distinct scenarios: dry/no rainfall baseline, rainfall, flooding with and without rainfall, and post-flooding conditions. Both static and dynamic routing simulations are conducted to compare traveler responses and system performance under varying levels of disruption and adaptability. A new penalty model is proposed that deals with the limitations and enhances the realism of SUMO simulations leading to better quantification of indirect impacts. Results at the overall network level demonstrate clear degradation of performance across all flood-related scenarios. Compared to the dry baseline, average delay increased by ~165%, average distance travelled grew by almost 49%, and the proportion of lost or uninserted vehicles rose by ~210%, while average speed decreased by ~25%. The rainfall-only scenarios also contributed significantly to these degradations, further exacerbating network inefficiencies by 2 - 17% depending on the performance metric. These findings underscore the importance of considering not only floodwater but also antecedent rainfall conditions when evaluating transportation resilience, as rainfall effects can serve as both precursors to flooding and independent stressors on urban networks. It also demonstrates that indirect impacts can be quantified appropriately when using traffic micro-simulation models. Beyond aggregated network performance, localized spatial analyses provide further insights. Origin-based assessments reveal zones of vulnerability where the network is less capable of inserting and processing trips, thereby identifying spatial points of systemic weakness. Destination-based assessments, by contrast, highlight the consequences of flooding for accessibility, congestion, and serviceability, demonstrating how certain areas become isolated or disproportionately burdened by disrupted flows. Together, the origin- and destination-based perspectives capture the intertwined nature of vulnerability and congestion, and illustrate how direct, indirect, and cascading impacts manifest unevenly across space and time. Overall, this dissertation delivers four major contributions: (a) a unified, replicable framework that integrates hydraulic and traffic microsimulation for multi-stage flood and rainfall assessment, (b) a novel penalty model that improves the realism of SUMO outputs for disrupted and congested conditions, (c) a spatio-temporal classification of flood-induced mobility impacts, offering conceptual clarity absent in prior studies, and (d) empirical insights demonstrating how direct, indirect, compound, and cascading disruptions evolve over time and space in a real-world urban network. Collectively, these contributions strengthen the scientific foundation for flood-resilient transportation planning, provide methodological advancements that can be adapted to other cities, and emphasize the importance of multi-hazard, multi-scale mobility analysis under a changing climate.
Adaptive authorization through transformer-based learning
(2025) Sinha, Pratik; Popli, Navneet; Neville, Stephen William
Access control is a cornerstone of information security, defining how entities interact with protected digital resources. Traditional rule-based frameworks, though effective in static environments, struggle to adapt to modern, data-intensive ecosystems where roles, attributes, and contextual conditions evolve continuously. Recent advances in machine learning have introduced new opportunities to automate access control through predictive and adaptive modeling yet progress remains constrained by the scarcity of real-world datasets, inconsistent benchmarking methodologies, and limited evaluation under controlled data conditions. This thesis presents a reproducible framework for evaluating machine-learning based access control models using synthetic, configurable datasets. The proposed data generation process emulates healthcare authorization structures, incorporating tunable role hierarchies, permission ratios, and anomaly patterns to simulate varying data noise and complexity. A suite of ML architectures, including decision-tree ensembles, feed-forward networks, residual networks, and transformer-based tabular models are systematically benchmarked using standardized preprocessing and evaluation metrics. Experimental results show that decision-tree ensembles provide strong baselines on small, structured datasets, while neural and transformer-based models generalize more reliably as data volume and complexity increase. This advantage depends on sufficiently rich, application-specific data where real datasets are typically sparse near decision boundaries, motivating targeted augmentation through synthetic generation around boundary cases. These findings validate the effectiveness of synthetic datasets for reproducible access-control research and demonstrate the impact of data scale on model elasticity and stability. Collectively, this work advances the development of reliable, ML-driven authorization systems through a transparent methodology for benchmarking and comparative analysis.
Investigating materials with nanoscale imaging and spectroscopy: Insights into perovskites and plasmonic substrates
(2025) Gartside, Hannah; Brolo, Alexandre G.
The structure of an object is integral to its function. When considering nanostructured materials such as thin films of perovskites and gold nanoparticles, subtle differences in the structure can have a significant impact on their optical properties and function. In this thesis, high resolution optical imaging techniques were applied to study the properties of nanostructured materials. Laser scanning confocal microscopy (LSCM) was used to visualize phase segregation in mixed-halide perovskite through the formation of I-rich domains. An important component of perovskite solar cells (PSCs) is hole transport layers (HTLs) which are responsible for keeping electron-hole pairs separate as they are transported to their respective electrodes. As a contribution to the study of HTLs by Ahmed et al.1, near-field scanning optical microscopy (NSOM) was used to spatially visualize the distribution of no-contact areas in high molecular weight (HMW) and low molecular weight (LMW) Poly(triarylamine) (PTAA) through their photoluminescence (PL) distribution. NSOM was also used to characterize a rough gold substrate coated in 4-MBN. During an NSOM scan, a photochemical reaction of 4-MBN was observed to have occurred. Through Raman and tip-enhanced Raman spectroscopy (TERS) experiments, it was determined that the photochemistry occurred only when the probe was within the near-field of the surface. Some products of the reaction were tentatively assigned from TERS spectra. To understand the cause of the photodegradation, NSOM experiments were performed to investigate the possible presence of dark plasmons. Dark plasmons decay non-radiatively and are therefore extremely good at producing hot electrons which could have feasibly catalyzed the photodegradation reaction. Fano resonance was observed, and the near-field spectrum was blue-shifted relative to the far-field spectrum. This confirmed the activation of dark plasmons and suggested that they are the most likely cause of the observed photodegradation.

Browse

Recent Submissions