82837 research outputs found
Sort by
Dynamics of the initiation and the evolution of Cordilleran orogeny
No full textOrogenesis is a complex tectonic process, involving interactions between tectonic plates and internal deformation of the continental crust and mantle lithosphere over millions of years. Cordilleran orogeny, where a mountain belt forms above active subduction zones, has created mountain belts spanning more than 15,000 km along the western margins of North and South America. This thesis investigates the dynamic evolution of these Cordilleran orogens, focusing on processes occurring during the pre-orogenic, syn-orogenic, and post-orogenic phases. Specifically, three geodynamic problems are addressed: (1) dynamics of lower crustal flow and its influence on Moho geometry, (2) initiation of the Andean orogeny, and (3) subduction erosion dynamics. These issues are explored using two-dimensional thermo-mechanical models developed with the finite-element code SOPALE.
Firstly, a series of numerical models investigate the relationship between lower crustal flow and the evolution of Moho geometry during and after tectonic perturbations caused by lithospheric extension. Results demonstrate that localized thinning from extension generates lateral pressure gradients that trigger viscous flow in the lower crust. This flow acts to smooth Moho deflections both during active extension and for 5–10 Myr following the cessation of extension. The key factors controlling this behavior are crustal temperature and rheology. Effective viscous flow requires minimum Moho temperatures around 750°C in wet, quartz-rich lower crust, limiting Moho deflections (<12 km) and generating low lateral gradients (<0.1 km/km). Conversely, cooler or more mafic crust inhibits flow, resulting in a more rugged Moho. These findings imply that observed flat Moho geometries in regions such as the western North American Cordillera, western Europe, Tibet, and Archean cratons reflect hot, weak lower crust conditions favorable to viscous redistribution, whereas rugged Moho geometries suggest cooler, stronger crust with limited flow.
The second part explores the initiation of the Andean orogeny and the contrasting deformation magnitudes between the central and southern Andes (>300 km vs. <100 km shortening, respectively). Through systematic modelling, it is demonstrated that regional factors, particularly slab folding within a high-viscosity lower mantle and rapid trenchward continental motion, are critical in triggering continental compression. Both factors reduce slab dip slightly at depths shallower than 200 km, enhancing compressive stress transfer from the subduction interface into the continent. The onset of continental deformation depends strongly on continental strength relative to plate boundary stresses. Deformation is enhanced with rapid trenchward continental movement and local conditions such as higher subduction interface friction and weaker continental lithosphere rheology. This combination of regional and local controls explains the substantial shortening and uplift observed in the central Andes. Conversely, limited deformation in the southern Andes likely results from reduced interface friction due to abundant trench sediments or stronger continental lithospheric rheology.
The final part investigates mechanisms controlling subduction erosion and the fate of eroded crust. Subduction erosion occurs when crustal material from the overriding continent is mechanically entrained by the subducting plate into the mantle. Numerical modeling indicates that subduction erosion depends on continental mantle lithosphere rheology (assuming weak continental crust) and strength of the subduction interface (controlled by oceanic crust frictional strength in the models). Subduction erosion is favored by a relatively strong plate interface and moderately strong continental mantle lithosphere (comparable to wet or dry olivine); significantly weaker or stronger conditions inhibit erosion. Additionally, models reveal that most subduction-eroded crust is not carried into the deep mantle but detaches from slab at depths of ~120–200 km due to buoyancy. It may then either: (1) underplate the continental mantle lithosphere, (2) intrude within the continental mantle lithosphere, or (3) underplate the continental crust. The fate of eroded material depends on its buoyancy and the continental mantle lithosphere viscosity. The eroded crust is generally limited to depths below where the mantle lithosphere viscosity is above ~1020 Pa s, and therefore it either underplates or intrudes the mantle lithosphere, depending on the rheology. Furthermore, increased buoyancy from greater erosion amounts favors crustal underplating, especially in weak, hot mantle lithosphere environments. These results highlight subduction erosion’s potential role in crustal thickening, explaining anomalously thick crust in areas such as the northern Puna plateau of the Andes, where minimal crustal shortening occurs. Collectively, this thesis emphasizes the roles of crust and lithosphere rheology, alongside local lithospheric dynamics, in shaping the large-scale evolution of Cordilleran orogens
Zoe Vernham - Abstract 18 - Innovate Conference 2025
No full textThis analysis uses Driscoll’s model of reflection to investigate high employee turnover in a Canadian home care practice setting. Using the Driscoll’s, what, so what, and now what questions; the author explores employee satisfaction, retention and specific challenges faced by home care workers and recipients
Value-Oblivious Secretaries with Advice
No full textIn the value-oblivious secretary problem, a sequence of candidates arrives in uniformly random order, and the decision-maker has access only to pairwise comparisons between candidates, rather than their exact values. The goal is to maximize the probability of selecting the candidate with the highest value. This work examines the value-oblivious secretary problem through the lens of algorithms with predictions (also known as learning-augmented algorithms), by incorporating potentially erroneous advice indicating the position of the best candidate in the arrival sequence. The objective is to bridge the gap between worst-case guarantees and empirical performance by designing online algorithms that can effectively leverage accurate predictions while maintaining resilience to arbitrary prediction errors.
We develop both deterministic and randomized algorithms that build upon the classical wait-and-accept strategy, using a novel optimization-based frame-work. This framework captures the trade-off between consistency (performance under accurate predictions) and robustness (performance under adversarially wrong predictions) by formulating them as optimization objectives and constraints. Algorithmic performance is then optimized by solving the resulting problem. We further demonstrate the versatility of our approach by extending it to variations of the secretary problem, including the multiple-choice setting and variants with rehiring options and time fairness constraints
Exploring Restorative Practices as Alternatives to Police in Schools for K-12 Education
No full textRestorative justice originated in the 1960s as an alternative to traditional rehabilitation and crime prevention approaches. In education, suspensions were commonly used to address student misbehaviour. However, this undercuts students' abilities to solve problems, increases chances of incarceration and dropping out for Black students, and disproportionately disciplines minority students and correlates to a higher risk of involvement in the juvenile justice system. This research poster explores the possibilities for and results of a restorative justice system model in Alberta's K-12 classrooms as a alternative to exclusionary punishment models
Mathematical Analysis of Go-or-Grow Type Systems in Cancer Modelling
No full textGo-or-grow approaches represent a specific class of mathematical models used to describe populations where individuals either migrate or reproduce, but not simultaneously. The reaction-diffusion ODE formulation of these models has a wide range of applications in biology and medicine, chiefly among those used to model the spread of brain cancer. The analysis of go-or-grow models has inspired new mathematics, particularly in connection to the Fisher–Kolmogorov–Petrovsky–Piskounov (FKPP) equation and the areas of travelling waves and steady-state solutions. In this dissertation, we focus on two special types of solutions of the go-or-grow models: travelling wave solutions and steady-state solutions in bounded domains with hostile boundaries. For travelling waves, we provide an existence and non-existence result for a general class of cooperative go-or-grow models and a particular non-cooperative glioma model. For the cooperative go-or-grow models, we show formal convergence of travelling wave solutions of the go-or-grow models to travelling wave solutions of an FKPP-type equation with nonlinear diffusion.
For steady-state solutions in bounded domains, we extend the constant rate critical domain results by Hadeler-Lewis's by changing from constant diffusion to a uniformly elliptic operator and constant transition rates to nonlinear functions of the two populations. The degenerate nature of the equation of the sedentary compartment requires us to use Young-measure valued weak solutions, which we call alright solutions. We show, under certain conditions, that the domain size determines whether a nontrivial, alright steady-state solution exists or not
Tonya Roy - Abstract 5 - Innovate Conference 2025
No full textIn 2023, Frankadua, Ghana, experienced a high incidence of malaria, with 80% of schoolchildren testing positive. The high malaria rate was primarily attributed to stagnant and contaminated water sources, which served as breeding grounds for mosquitoes (Serengbe et al., 2015). This global health initiative aimed to reduce malaria transmission by providing access to clean water through boreholes, water receptacles, and pumps that are sustainably maintained by communities. Educational programs are implemented in schools promoting hygiene practices. Building trust and relationships with community members and empowering the community to participate in malaria prevention are at the core of this nurse-led project, emphasizing reducing the burden of this communicable disease
Ensemble Methods for Unsupervised Syntactic Parsing
No full textUnsupervised parsing aims to uncover the underlying structure of sentences without relying on annotated data. Existing models often exhibit instability and capture different aspects of parsing structures.
In this thesis, we systematically investigate ensemble methods for unsupervised parsing, including constituency and dependency parsing, to enhance performance and robustness. We first introduce a novel tree averaging technique that effectively merges multiple constituency parse structures, leading to improved constituency parsing results. For discontinuous constituency parsing, we analyze the computational complexity of tree averaging under different conditions and develop an efficient exact algorithm. Furthermore, in unsupervised dependency parsing, we identify error accumulation as a key challenge in ensemble methods and propose an ensemble-selection approach that mitigates this issue by leveraging error diversity.
In our experiments across multiple datasets and settings, our proposed ensemble framework and ensemble-selection strategies consistently outperform previous approaches, demonstrating its superior effectiveness and robustness under various conditions, including domain shifts
Transmit Radiofrequency Field (B1+) Map Prediction Using Machine Learning
No full textQuantitative magnetic resonance imaging (qMRI) enables measurement of tissue parameters such as longitudinal T1 and transverse T2 relaxation times, which can reveal microstructural changes relevant to neurological disease. Accurate T1 and T2 mapping requires modelling of the signal and knowledge of the actual flip angle distribution, typically obtained from transmit radiofrequency field (B1⁺) mapping. However, B1⁺ acquisitions are often excluded from clinical and large-scale research protocols, limiting the reliability of downstream quantitative analyses.
This thesis investigates the use of deep learning to predict B1⁺ maps in the brain from routinely acquired anatomical MR images. The Alberta 300 dataset was used including 267 healthy adult subjects (ages 19–90, 151 females) acquired at 3T at the Edmonton site. Available anatomical images included: volumetric T1-weighted magnetization-prepared rapid gradient echo (MPRAGE), and dual-echo proton density (PD) and T2-weighted turbo spin echo. In addition, a B1+ mapping sequence was included in each study, enabling a gold standard for model development. A 3D generative adversarial network (GAN) was trained to synthesize subject-specific B1⁺ distributions from the corresponding anatomical MR images (240 subjects for training, 27 for inference). Multiple input combinations (up to two channels) were tested to identify the best-performing configuration. Both whole-cohort (n = 267) and age-classified models (three groups of 89 subjects each) were evaluated using structural similarity (SSIM), mean absolute percentage difference (APD), and regional analyses across whole brain and subcortical regions. To further increase the number of inference subjects and assess model robustness, a four-fold randomized cross-validation was conducted, expanding the test set from 27 to 108 subjects.
The GAN-predicted B1+ maps showed strong agreement with measured B1⁺ values. Among the input combinations, the single-channel T1-weighted input yielded the best whole-brain accuracy (APD = 3.17%, SSIM = 96.0%, averaged for all of 27 inference subjects in 3D space). After age separation, performance improved further (APD = 2.60%, SSIM = 97.0%, averaged for the same 27 inference subjects in 3D space across the three age groups). The four-fold randomized cross-validation confirmed stable performance (APD = 2.49%, SSIM = 96.2%, averaged for all of 108 inference subjects in 3D space across the three age groups).
Regional analysis of B1⁺ maps across seven regions of interest showed errors typically below 3%, with the lowest error in gray matter (APD = 2.28%, averaged for all of 108 inference subjects in 3D space across the three age groups). When integrated into a T2 mapping pipeline that required dual echo PD and T2-weighted images and a B1+ map for accurate modelling, the predicted B1⁺ maps produced quantitative T2 values within 1.16% of those obtained using measured B1⁺ across the whole brain. Regional T2 errors were generally below 1%, with the lowest discrepancy in the putamen (APD = 0.28%, averaged for all of 108 inference subjects in 3D space across the three age groups).
These findings demonstrate that accurate B1⁺ estimation can be achieved directly from standard MR contrasts, enabling retrospective correction of existing datasets and reducing reliance on direct transmit field mapping. This approach has the potential to make quantitative MRI more accessible in both research and clinical settings
It’s Not What I Thought It Would Be: Ever-Becoming Teacher
No full textFor those who teach, education is a place of dizzying personal successes, bitterly demoralizing defeats, and an acceptance of the benignly ordinary. Looking at my 23 years of classroom teaching, I reflect on what teaching means to me and how I identify as a teacher. My teacher identity is a puzzle——a puzzle both of pieces and of understanding. In an effort to better understand my teacher identity puzzle, I engaged in a narrative inquiry (Clandinin & Connelly, 2000) through initially engaging in interviews with two teachers in the first 2 years of their continuous teaching contracts. I discerned 4 resonant threads as I looked across their experiences and then used these threads to engage in an autobiographical narrative inquiry into my own experiences. This thesis is an inquiry into two beginning teachers’ identities and my own identity. Deborah Britzman (2015) has noted that classrooms are sites of conflict resulting in a chaos of learning, while Maxine Greene (1995) suggests that imagination may awaken and sustain a career requiring an internal voice that reconciles the dichotomy found between teacher and student and teacher as learner. This inquiry into my teaching journey involved looking backwards and forward, inwards and outwards as I asked myself - Why do I teach? Who am I as a teacher? What makes me a teacher? In the final chapter, I weave various depictions of teachers from film and education researchers against my inquiry into who I am as a becoming teacher. Rarely in this profession do things go as planned (Aoki, 2004). Knowing one’s teacher identity provides the ability to live in the moment, and through my inquiry, I now understand how I live, and continue to live, in the moment