Abstract Manager

Traffic collision risk and severity are strongly influenced by environmental and roadway operating conditions, particularly in cold-climate cities where winter weather introduces substantial variability in surface friction, driver behavior, and system performance. Despite extensive research on collision occurrence, limited attention has been given to the combined and interacting effects of seasonality, roadway surface condition, and speed environment. This constrains municipalities’ ability to develop targeted, evidence-based safety and winter maintenance strategies.

This study investigates the spatiotemporal impacts of season (winter versus non-winter) and roadway surface conditions (dry, wet, snow- or ice-covered) on traffic collision frequency and severity across the City of Edmonton. Using multi-year collision records categorized by severity, from property-damage-only (PDO) to severe and fatal injuries, the analysis integrates roadway characteristics, posted speed limits, traffic volumes, neighbourhood population data, and snow and ice clearance prioritization schedules.

A comprehensive analytical framework is employed. Descriptive and exploratory analyses quantify annual and seasonal trends, cross-tabulating collision severity by season and surface condition. Collision rates are normalized by vehicle-kilometres travelled and population to account for exposure effects. Spatial risk patterns are examined using Kernel Density Estimation (KDE) and Getis-Ord Gi* statistics to identify statistically significant hot and cold spots based on normalized rates rather than raw counts. Emerging Hot Spot Analysis (EHSA) is applied to capture the temporal evolution of collision risk, distinguishing persistent, intensifying, and newly emerging hotspots. Statistical modeling using generalized count regression models evaluates collision frequency, while discrete severity modeling frameworks assess the influence of seasonal, surface, speed, and exposure-related factors on injury outcomes.

The results will provide actionable insights into how winter conditions and surface states amplify collision risk and severity, and how these patterns align with roadway speed environments and maintenance prioritization. The proposed framework aims to support data-driven safety planning, winter operations optimization, and context-sensitive speed management, thereby contributing to safer, more resilient urban transportation systems in cold-climate cities.