Climate change is a global challenge with profound environmental, health, andsocio-economic implications. Canada's diverse geography offers a unique lens to study localized climate trends. This thesis models and clusters climate variables, focusing on temperature trends, using Bayesian hierarchical models and clustering techniques to uncover regional patterns and health impacts. Three decades of hourly temperature data from the Meteorological Service of Canada were split into 18 annual parts to capture seasonal variations. Metrics like mean, minimum, and extreme temperatures were analyzed. Bayesian models revealed regional variability, with examples of British Columbia and the Northern regions exhibiting notable trends. Clustering identified regional dependencies and linked temperature trends with morbidity and mortality risks from air pollutants (PM2.5, O3). Summer risks stemmed from O3, while winter risks were PM2.5 driven. Findings highlight the need for region-specific strategies, offering actionable insights for policy makers addressing climate-health linkages.
Author Keywords: Bayesian models, Climate change, Clustering, Temperature Trends, Time Series