The AI-based Sea-Ice Presence Forecasting Model is an artificial intelligence system developed by the National Research Council Canada in collaboration with the University of Waterloo to provide accurate and timely forecasts of sea ice conditions in the Canadian Arctic and other ice-infested waters. The system combines sea-ice model outputs with observational data using advanced statistical and machine learning methods to generate daily seasonal forecasts that support safe shipping operations.
This system affects maritime operators, shipping companies, and Arctic researchers who rely on accurate ice forecasts for tactical and strategic planning. The forecasts are integrated into the Canadian Arctic Shipping Risk Assessment System (CASRAS) and made available to both NRC employees and external stakeholders, including shipping companies and government agencies. The system does not process personal information and is designed specifically to improve maritime safety and operational decision-making in Arctic waters.
As a production system in active use, the model generates daily forecast products that help reduce risks associated with Arctic shipping by providing state-of-the-art predictions of harbor ice conditions and sea ice extent across different seasons.
