This spring, Environment and Climate Change Canada will introduce a new hybrid weather forecasting model that combines artificial intelligence with traditional physics-based forecasting systems. The goal is to improve the accuracy, speed, and reliability of weather predictions while strengthening public safety and emergency preparedness across the country.
The new system uses AI to analyze vast amounts of historical and atmospheric data to identify weather patterns, while the existing physics-based model continues to account for local factors such as wind, temperature, and precipitation. By integrating both approaches, forecasters will be able to deliver more precise weather predictions and more timely alerts for Canadians.
The hybrid model is expected to significantly enhance forecasting performance at all time ranges, from short-term to long-range predictions. For example, a six-day forecast using the new system is expected to match the accuracy of a current five-day forecast. It will also provide earlier warnings for major weather events such as storms, heat waves, and atmospheric rivers—sometimes improving lead time by more than 24 hours.
In addition, the model will increase confidence in predicting the timing and movement of severe weather systems, helping communities prepare more effectively. These improvements are particularly important as Canada faces more frequent and intense climate-related events, including wildfires, floods, and extreme heat.
Over the past year, scientists at Environment and Climate Change Canada have tested the hybrid model alongside existing forecasting tools and used it to reanalyze past storms. Early results show improved detection of severe weather, which can strengthen early warning systems and enhance national preparedness.
Despite the integration of advanced AI, the department will continue to rely on meteorologists to interpret model outputs and communicate forecasts to the public, ensuring that human expertise remains central to weather prediction.
