Water hazards pose a significant risk to railway infrastructure, leading to service disruptions, costly repairs, and safety concerns. Recognizing these challenges, Canadian Pacific Kansas City (CPKC) sought an innovative, proactive solution to enhance water hazard detection and mitigation across its network. In response, Tetra Tech developed an AI-driven water hazard monitoring platform that integrates advanced satellite data, artificial intelligence, field observations, and historical environmental information into a collaborative geospatial environment. This platform enables railway operators and stakeholders to identify, predict, and communicate water-related threats in near real-time, fostering a shared, interactive environment for decision-making, risk assessment, and infrastructure resilience. The project began in September 2023 as a 2,000-mile pilot study, monitoring high-risk track segments using Synthetic Aperture Radar (SAR) and multispectral imagery. Based on the success of this initial phase, CPKC expanded the system to monitor 12,600 miles of track, covering all legacy Canadian Pacific assets. Today, the platform processes over 3,000 satellite images weekly, analyzing 120,000 GB of data to detect more than 1.2 million water bodies and identify hazards such as beaver dams, washouts, and flooding risks. Results are processed within four hours of satellite capture, enabling rapid hazard identification and response. The system now provides continuous coverage across 75,500 square miles, having generated 500+ alerts to date, including 50 urgent hazard warnings that supported timely interventions and proactive maintenance. This project represents a significant shift in railway risk management, moving from reactive inspections to AI-powered predictive maintenance. By integrating remote sensing and AI, CPKC has set a new industry benchmark for infrastructure resilience, demonstrating the power of data-driven decision-making in mitigating environmental risks.

The CPKC Water Hazard Monitoring Program leverages an innovative approach that integrates Artificial Intelligence (AI), Synthetic Aperture Radar (SAR), and multispectral satellite imagery to proactively monitor and mitigate water-related threats.

The system processes over 3,000 satellite images (up to 16 cm resolution) and analyzes 120,000 GB of data weekly to detect water hazards. SAR enables all-weather, day-and-night monitoring, while multispectral imaging provides detailed insights into water bodies and vegetation. AI models analyze time-series data and spatial patterns to detect water level changes, blockage hazards, and potential washouts.

Data is processed within four hours of capture, ensuring near real-time alerts. AI-powered risk assessments classify hazards based on severity, triggering automated alerts to CPKC operators. The system continuously refines its models through machine learning, field validation, and feedback loops, improving accuracy over time. This approach transforms railway risk management, shifting from reactive inspections to a predictive,

AI-driven monitoring system that enhances safety, efficiency, and resilience.
Originally launched in September 2023 as a 2,000-mile pilot, the program’s success led to a full-scale expansion to 12,600 miles, covering 75,500 square miles of railway corridors.

The program is transforming how CPKC monitors and responds to water hazards, shifting from reactive inspections to AI-driven predictive maintenance. Near real-time alerts—processed within four hours—are reducing response times and service disruptions. The system has issued 500+ alerts, including 50 urgent warnings, enhancing safety, increasing operational efficiency, and optimizing maintenance costs across 12,600 miles of track.