Assembling a National Forecast from an Ensemble of Global Weather Models
The Royal Society of Tasmania invites you to its April 2025 Public Lecture by guest speaker Dr Robert Johnson. All RST members, their guests, and the public are welcome.
Where: Geology Lecture Theatre, UTAS, Sandy Bay Campus on Sunday 6 April, 2024.
Time: 3.30pm for pre-lecture drinks, 4pm for the lecture.
A reminder that this will be the first Sunday following the time change to mark the end of Daylight Savings. I know it threw a few attendees last year.
Can artificial intelligence outsmart Mother Nature? From supercomputers crunching vast climate data to neural networks refining forecasts in real time, AI is transforming the way we predict the weather. But how does it work, and can it really make forecasts more reliable? Join us for a fascinating dive into the future of meteorology, where machine learning meets the atmosphere to shape tomorrow’s forecasts.
Every day, weather forecasts shape our decisions – from what we wear to how we prepare for extreme weather events. But where do these forecasts come from? This talk will take you behind the scenes of modern weather prediction, showing how we combine multiple global weather models into a single, more accurate national forecast for Australia. We’ll explore how AI and machine learning are quietly improving the forecasts you rely on – especially for rainfall predictions – and why uncertainty is an essential part of predicting the future.
Dr. Rob Johnson is a senior research scientist at the Bureau of Meteorology, where he leads the Forecast Improvement Research Team. His work focuses on improving weather forecast accuracy through model post-processing, nowcasting, and integrating advanced AI and machine learning techniques into operational meteorology. He recently led the implementation of Australia’s first AI-driven public weather forecasting system, replacing traditional statistical methods and enhancing the way forecasts are produced and communicated to millions of Australians.
