12466 research outputs found
Sort by
COASTAL AND MARINE WAVE HAZARDS IN THE CHANGING ARCTIC
No full textThis work is embargoed by the author and will not be publicly available until May 2029.The rapid warming of the Arctic and consequential sea ice decline has resulted in growth in the regional wave climate. The increasing wave energy already has had marked impact on coastal regions, driving extreme rates of erosion and threatening coastal communities with unprecedented wave and flood hazards. Nonetheless, further growth in the regional wave climate is expected to continue well into the current century as sea ice decline proceeds and open-water area expands allowing for further generation and growth of ocean waves. In this context, this dissertation presents research investigating how future Arctic coastal and marine wave hazards will evolve in the coming decades in regions already affected by or expected to be impacted by wave hazards. To accomplish this objective, the latest generation of climate models’ projections of sea ice were assessed to gain insight towards bias inherent in the mean-state simulation of the Arctic sea ice and identify regions or seasons which models struggle to represent. Based on this analysis, a multi-model ensemble sea ice projection was utilized in conjunction with a dynamic wave model to simulate how Alaskan Arctic coastal wave exposure resulting from an extreme storm event would be modified under future sea ice conditions. Derived results predict that future coastal exposure into the fall could be extended by one month by 2050 and up to three months by 2070 for the high emissions scenario. Subsequently, an ensemble of pan-Arctic simulations was produced to investigate how future wave hazards would evolve along Arctic maritime sea routes. This investigation demonstrates that concurrent with increasing maritime accessibility resulting from sea ice retreat, there is a rapid and significant increase in extreme maritime wave hazards present in the late fall leading to unprecedented wave heights for the analyzed region. These projections and analyses advance the knowledge of future Arctic wave hazards by employing the latest modeling methodologies and utilizing the latest climate model data to derive assessments of wave hazards at scales applicable to user groups adapting to the rapidly changing Arctic.2029-05-1
ESSAYS ON THE SOCIOECONOMIC DETERMINANTS AND IMPACT OF INFORMATION AND COMMUNICATIONS TECHNOLOGY IN SUB-SAHARAN AFRICA
Get PDFDuring the last quarter century, the diffusion of information and communications technology (ICT) in Sub-Saharan Africa (SSA) has been shaped by noteworthy socioeconomic determinants and arguably resulted in significant socioeconomic impact. The first essay in this dissertation is an enquiry into the statistically significant socioeconomic determinants of Natural Language Processing (NLP) in SSA languages. Logit and probit regressions reveal that the presence of Artificial Intelligence (AI) startups, effective governance, trade volume and mean years of schooling are the positive statistically significant socioeconomic determinants of NLP in SSA. The second essay is an enquiry into the socioeconomic impacts of mobile phone technology diffusion in SSA, as measured by poverty headcounts and income inequality at the country level. Fixed effects regressions reveal that mobile phone technology diffusion has a statistically significant negative relationship with poverty headcount ratios and income inequality at the country level in SSA. The third essay links ICT diffusion in SSA and socioeconomic impacts with a theory of increased ease of capability realization by SSA residents resulting from the use they make of ICT applications. Using mixed methods, this essay also finds that ICT diffusion in SSA has a statistically significant positive relationship with increased financial account ownership by the poorest 40 percent in SSA, as well as for the women and youth in the region. The third essay concludes with a discussion of the beneficial effects of ICT-augmentation of especially economically disadvantaged people relative to the limitations they face in the capability realization process. The dissertation concludes with policy recommendations based on insights from the essays
Revealing the Global Morphology of the Magnetosphere during Substorms Using Data Mining-Driven Empirical Magnetic Field Modeling
Get PDFThe Earth's magnetosphere undergoes global dynamical reconfigurations termed magnetospheric substorms in response to changes in the solar wind. Understanding how the 3D magnetic field and associated current systems evolve in time during these events is a critical component needed to understand and predict the magnetosphere system. However, modeling their description using magnetohydrodynamic (MHD) approaches is complicated because non-ideal MHD processes, such as the formation of ion-scale thin current sheets and magnetic reconnection, are defining features of substorms. As such, several unanswered questions persist about the global morphology of the magnetosphere and its evolution during these events: (1) What is the global-scale configuration of the magnetospheric magnetic field and current systems and how do they evolve during a substorm? (2) Where does magnetic reconnection occur in the magnetotail during a substorm and what is the structure of the associated X-line? (3) Where do ion and electron isotropy boundaries (IBs) map to in the magnetotail during the substorm growth phase? The dissertation will address these questions by empirically reconstructing the global 3D magnetic field and electric currents. For a given time, a multi-decade, multi-mission archive of magnetospheric magnetic field observations is mined to find a subset of data from other times when the magnetosphere was in a similar substorm and storm state. This subset of data is used to fit an empirical model of the magnetic field that analytically describes the key magnetospheric current systems associated with substorms. This procedure is repeated for each time step during an event, revealing the global morphology and evolution of the magnetosphere during substorms. The resulting model, termed SST19, represents the first empirical magnetic field model capable of capturing substorm features. To address (1), we show that SST19 captures the primary substorm features including the formation of an embedded thin current sheet (TCS) that stretches the magnetotail throughout the substorm growth phase. Following substorm onset, the tail undergoes a rapid reconfiguration. The TCS collapses as the tail dipolarizes and magnetic flux piles up in the near-Earth region. To address (2), we use in situ observations of tail reconnection from the Magnetospheric Multiscale (MMS) Mission and compare them to their SST19-reconstructed locations. We demonstrate that the SST19 analytical structure is sufficiently flexible to resolve most tail X- and O-lines and that their modeled locations generally match the MMS observed reconnection sites to Earth radii (). The SST19 reconstructed X-lines vary in length from to , with the shorter ones tending to form inside of while the longer ones, , appear beyond . Question~(3) is addressed by inferring the equatorial location of ion and electron IBs using the SST19 model. The fidelity of these SST19 inferred IB locations is confirmed by mapping them to low altitudes and comparing them to their observed location using precipitating particle data from the Electron Losses and Fields Investigation (ELFIN) mission. Both the observed and modeled IBs move equatorward during the growth phase and diverge in latitude after substorm onset. Further, they reveal a ``checkmark" pattern in energy vs. time/latitude plots indicative of an accumulation of flux in the magnetotail during the substorm growth phase
Statistical Estimation of Agent-based Models Using Econometric and Machine Learning Techniques
Get PDFAgent-based models (ABMs) are computational models used to simulate the interactions between people, things, places, and time. Financial markets are one area ABMs can be applied to. Understanding the behavior of the market and how different agents affect it prior to making financial decisions can be very beneficial. The volatile behavior of markets can make them difficult to study, but ABMs can help explain some of the characteristics of market-related time series that are not explained by standard models. Financial ABMs were chosen for this dissertation due to the importance and complexity of the financial market. These financial ABMs can only be effective if their parameters can be accurately estimated. In this dissertation, I study how to statistically estimate ABMs in the context of a financial market model. To do this, I used several calibration methods of different types including simulated minimum distance (SMD) methods, Bayesian Density Estimation (BDE) methods, simulated annealing, and particle swarm. I applied these calibration methods to three different financial ABMs including Franke and Westerhoff, Farmer-Joshi, and Rock Around the Clock. I found that overall the Bayesian Density Estimation, specifically Platt’s method, was found to be the most effective method with the best results emerging after only 12 iterations. In terms of the other methods, Nelder Mead significantly deviated in terms of its approach and solution. It only had significant improvements in its optimization when provided with a larger number of iterations. The Genetic Algorithm outperformed Nelder Mead however, I learned that the Genetic Algorithm led to a more unusual global solution, is more sensitive to the quality of bounds and constraints, and is highly dependent on the initial guess. Further, in comparison, I found that Simulated Annealing offers better tolerance to constraints and optimization setup. Lastly, for surrogate models, I found that surrogate models with active learning can be useful when generating intuitive and less random parameters is beneficial. Overall, this dissertation provides a comparison of various calibration methods and a framework for how to best estimate different financial ABMs
A Decision-Rule and Spatial Transfer Learning-Based Approach For Automated Mapping of Local Climate Zones (LCZs) Using Multi-Source Geospatial And Remote Sensing Data
Get PDFUrbanization, industrialization, and population growth are driving rapid changes in global climate patterns, posing significant challenges to human health and environmental sustainability. Local Climate Zones (LCZs) classification offers a structured approach to understanding urban morphology and its relationship with climate, providing valuable insights for urban planning and policy-making. Leveraging remote sensing technologies, this study aims to advance LCZ mapping by addressing key limitations in current classification approaches and integrating spatial information into machine learning models. Using a combination of decision rules based on remote sensing and spatial parameters, this research automates the generation of training samples for LCZ classification on a global scale. By establishing universal decision rules, training samples are generated automatically, overcoming geographic and climatic variations. Additionally, a spatial transfer learning method is proposed to address the challenge where certain categories of training samples are scarce in one geographic location but plentiful in another. This model is designed to integrate local covariates, local spatial information, and global covariates. This integration enables the model to address spatial dependencies and transfer knowledge about scarce categories from locations where they are abundant. Consequently, this improves the precision, accuracy, and scalability of solving local classification problems. The study produces LCZ maps with the proposed method and compares them with existing products, demonstrating significant advancements in accuracy and detail. Statistical analyses confirm the promising performance of the proposed spatial transfer learning model, with overall accuracies consistently above 80%. Visual comparisons reveal discrepancies between LCZ maps generated by the proposed model and those from existing databases, highlighting the superiority of the spatial transfer learning approach. Additionally, the study identifies limitations in current classification approaches, including scale constraints, reliance on supervised methods, and inconsistencies in training data. Recommendations for future research include the refinement of decision rules, integration of more accurate building height data, and consideration of cloud cover in analysis. By addressing these limitations, LCZ mapping holds immense potential for informing urban planning, climate adaptation, and sustainable development efforts globally
Ground-based Light Curve Follow-up Validation observations of TESS object of interest TOI 5886.01
No full text“TESS's full name is The Transiting Exoplanet Survey Satellite. TESS's mission is to observe stars to find exoplanets. TESS conducts two-year observations of the solar neighborhood and monitors the periodic decrease in brightness of stars caused by transits caused by planets passing by. The goal of TESS is to identify a large number of asteroids. It will identify asteroids and measure their sizes. Through follow-up observations, we can get the masses of some planets. Select a star, use AstroImageJ (AIJ) to generate a light curve, and use NEB check to check whether the star has exoplanets. There are usually transits on the light curve, and the RMS of NEB is too large, which means that the transiting planet is not the planet of the target star we are observing.
Project 5b: Trend of Liver Transplant Recipients and Donor Cause of Death Before vs During the Covid-19 Pandemic
Get PDFThis project included one student research who wished to remain anonymous.Objective: Our study intends to investigate how liver donor and transplant recipient characteristics change pre- and post-pandemic. While also comparing transplant outcomes that used donor livers from different causes of death pre- and post-pandemic, such as drug overdose, stroke, blunt injury, cardiovascular disease, natural death, and more
Sex Differences in Concussion Incidence and Recovery in Middle School Students
No full textThis work is embargoed by the author and will not be publicly available until May 2026.Sex differences in sports-related concussion (SRC) incidence and recovery have been established in late adolescent and adult athlete populations, but little is known about early adolescent athlete populations. Early adolescence (10-14 years) is an important developmental period marked by considerable physical, cognitive, and psychosocial developmental changes associated with puberty. During this time, early adolescents attend middle school (MS) which provides low-cost sports, which are often a barrier to participation in youth sports programs. Understanding SRC concussion during this time is critical for concussion care in this population and the long-term health of MS athletes. Therefore, the aims of this dissertation were as follows: (i) to describe the incidence of SRC in sex-matched MS sports; (ii) to identify potential sex differences using three measures of clinical recovery; and (iii) to explore the utility of the application of a machine-learning algorithm to predict concussion outcome in MS athletes. Participants in these studies (N= 13,870) are participants of MS-sponsored sports over six school years (Male: n=7267, 52.4%, age: 12.2±0.9). Certified athletic trainers (AT) collected injury information and athlete exposure data during school-sponsored competitions and practices. We observed the following: (i) MS girls were over 2 times more likely to sustain an SRC than boys, regardless of event type in basketball, soccer, and softball. (ii) There were no sex differences in two of the three measures of clinical recovery, with girls having a significantly longer recovery in comparison to boys. MS athletes, regardless of sex, missed less than one full day of school after sustaining an SRC. (iii) A random forest algorithm was created using the following factors: sex, age, grade level, participation in a fall contact sport, participation in a winter contact sport, participation in a spring contact sport, number of sports seasons, identifies as Black, identifies as Hispanic, English is the primary language spoken at home, pre-existing health condition, and history of concussion. The model, with the inclusion of all factors, was used to predict concussion outcomes with 25 % specificity and 77% sensitivity. This dissertation is among the first to research SRC in an understudied early adolescent population. This search aimed to explore sex differences in different aspects of concussion within MS-aged athletes. Collectively, the primary findings from this research suggest that sex differences within the MS population are significant in concussion incidence and return-to-play, specifically for female athletes. These findings can lead to improved development of prevention strategies and patient education within the MS population. As with all research, this dissertation is not without limitations, and as such findings may not be generalizable to all early adolescent MS athlete populations. Future research should further investigate the influence of sex differences in concussion assessment and recovery within MS athletes to further refine concussion care in children.2026-05-1
Student Conceptions of Scientific Research Methods and their Impact on Evaluations of Scientific Claims
Get PDFThe field of science education has long advocated for students to leave K-12 education with an appreciation for the diversity of research methods scientists employ in order to effectively engage with research generated across many scientific fields of study. Yet, research has indicated that students typically believe research proceeds according to “the scientific method,” a linear or cyclic model of experimental research that the science education research community considers a “myth.” Very few studies have specifically explored student conceptions of the diversity of research methods and how these conceptions relate to student engagement with novel scientific information, however. Through interviews, this dissertation investigates student conceptions of the diversity of research methods in US college freshman to better understand how students conceptualize the diversity of research methods around the time of high school graduation. This study also explores how these conceptions relate to student evaluations of news articles about scientific innovations. This dissertation finds that students can appreciate that researchers use a diversity of research methods to conduct scientific research while simultaneously envisioning the scientific method as a framework that guides scientific research. This exploratory research provides foundational insights for the science education research community and concludes with potential implications for science educators and suggestions for future research on student conceptions of scientific research
Ground-based Light Curve Follow-up Validation observations of TESS object of interest TOI 5886.01
Get PDF“Context. It is important to discover new systems with exoplanets because each new system shows a different aspect of the lifecycle of a solar system. Discovering new solar systems can also act as a way to confirm our understanding of the lifecycle and interactions between planets and stars. The launch of the TESS mission helped to identify thousands of possible exoplanets that need to be manually confirmed.
Aims. The goal of this study was to attempt to define the status of TOI 5886 as an exoplanet or as a false positive or false result.The goal was also to further research into exoplanets in order to make future studies of TOI 5996.01 more successful.
Methods. The George Mason 0.8m space telescope was used to observe TOi 5996.01 on the night of 6/18/24. AstroImageJ was later used to analyze the results and create a light curve and an NEB plot.
Results. Results of the analysis were inconclusive. Although the transit occurred during the predicted time and had a depth close to the predicted value no stars passed the NEB check which indicated that the result could be a false positive.