Environment and Climate Change Canada announced on Thursday its initiative to enhance the accuracy of weather forecasts by integrating artificial intelligence technology. The department is set to introduce a new hybrid model in the upcoming spring season, combining AI capabilities with traditional forecasting methods to improve the precision of weather predictions.
According to a news release, the hybrid model leverages AI for advanced forecasting of weather patterns, while integrating the conventional physics-based model to consider local factors like wind patterns, temperature variations, and precipitation levels. By analyzing extensive historical data spanning across an entire continent within minutes, AI models establish correlations between temperature, wind speed, and atmospheric pressure to forecast future weather conditions, particularly for significant events like heatwaves and hurricanes.
The hybrid model excels in predicting extreme weather phenomena such as high winds and heatwaves by retaining intricate details that AI algorithms may overlook. Environment Canada stated that with this new model, the accuracy of its six-day forecast will match that of its five-day forecast, marking a notable advancement that typically required years of research and development.
Moreover, the department highlighted that the hybrid system will expedite the prediction of major weather systems like winter storms, heatwaves, and atmospheric rivers. Throughout the past year, Environment Canada has conducted extensive testing on the hybrid model, running it concurrently with the traditional model to assess its performance in forecasting Canadian weather conditions.
Despite the technological advancements, Environment Canada emphasized the indispensable role of its meteorologists in interpreting forecast results and communicating them effectively to the public. Cindy Day, a seasoned meteorologist based in Halifax with over 40 years of experience, expressed enthusiasm about the rapid analysis of climate data and its potential impact on improving storm warnings for public safety.
However, Day raised concerns about the efficacy of historical data analysis in light of climate change’s rapid effects on weather patterns. She questioned the relevance of analyzing vast historical datasets in producing accurate forecasts amidst the rapidly changing climate conditions.