Analysis of the Effects of Severe Weather with the COAWST- Model Comparing Coupled and Decoupled Simulations, as well as Using Machine Learning Algorithms to Predict the Impact on Populations

Identificador:  TDX:4348

Handle: https://hdl.handle.net/20.500.11797/TDX4348

Autores:  Iglesias Deutú, Jordi

Resumen:
This doctoral thesis is structured into three interconnected chapters rooted in the field of fluid mechanics. The first two chapters delve into the analysis of extreme weather effects, with the first chapter specifically comparing coupled and uncoupled numerical simulations of the environment. These simulations are critical for climate policies and socio-economic decisions, and the study assesses the performance of coupled models in simulating a high-energy storm (Gloria, January 2020) in the northwestern Mediterranean region. The second chapter shifts focus to the application of Machine Learning techniques for predicting incidents caused by adverse weather, aiding emergency management. Using various municipality predictors, the model outperforms existing meteorological warning systems, helping to assess the impact of urbanization on the severity of such events. The third chapter addresses the decline in engineering program enrollments and introduces an aerodynamics workshop designed to engage and motivate students to pursue engineering studies. The workshop combines theoretical learning with hands-on experimentation in a wind tunnel, and surveys indicate positive student reception and their potential influence on career decisions. Overall, the thesis explores crucial topics in fluid mechanics, extreme weather analysis, machine learning for emergency management, and student engagement in engineering, contributing to scientific knowledge and practical applications in these domains.