Environment and Climate Change Canada announced on Thursday its adoption of artificial intelligence to enhance the accuracy of weather forecasts. The department is set to introduce a new hybrid model this spring, combining AI technology with traditional forecasting methods to improve prediction precision.
According to a news release, the hybrid model leverages AI capabilities to enhance future weather predictions while integrating traditional physics-based models to account for local variables like wind, temperature, and precipitation. By analyzing vast historical data spanning decades across entire continents within minutes, AI models can establish patterns between temperature, wind, and pressure to estimate forthcoming atmospheric conditions, particularly for significant weather events like heat waves and hurricanes.
The department highlighted that the hybrid model excels in predicting extreme weather phenomena such as strong winds and heat waves, as it retains intricate details that AI models might overlook. Environment Canada stated that the new system will elevate its six-day forecast accuracy to match its five-day forecast, a notable advancement typically achieved only after years of research and development.
Furthermore, the hybrid system is anticipated to expedite the prediction of major weather systems like winter storms, heat waves, and atmospheric rivers. Extensive testing of the hybrid model over the past year has been ongoing to evaluate its performance in forecasting Canadian weather conditions in parallel with the traditional model.
While emphasizing the critical role of meteorologists in interpreting and communicating forecast results to the public, Environment Canada underlined the significance of their judgment in the forecasting process. Veteran meteorologist Cindy Day from Halifax expressed enthusiasm over the rapid analysis of vast climate data and its integration into practical forecasting tools. She acknowledged the potential public safety benefits of early system identification, yet raised concerns about the efficacy of historical data analysis amidst the rapidly changing climate landscape.