AI weather models can now beat the best traditional forecasts

By Vassili Kitsios, Senior Research Scientist, Climate Forecasting, CSIRO

A new machine-learning weather prediction model called GenCast can outperform the best traditional forecasting systems in at least some situations, according to a paper by Google DeepMind researchers published today in Nature.

Using a diffusion model approach similar to artificial intelligence (AI) image generators, the system generates multiple forecasts to capture the complex behaviour of the atmosphere. It does so with a fraction of the time and computing resources required for traditional approaches.
…

Read complete article

~	Australia has a new autism strategy but questions remain for those who don’t get NDIS support

~	Fruit and veg, exercise, frequent bloodletting and more tips on staying healthy from medieval travellers

~	The world’s largest gathering: how India plans to keep 400 million pilgrims safe at the Maha Kumbh Mela festival

~	First Person: Syrian migrant shipwreck survivor vows to reconstruct shattered country

~	Cyberattack affecting school boards spotlights the need for better EdTech regulation in Ontario and beyond

~	LA fires: Why fast wildfires and those started by human activities are more destructive and harder to contain

~	Fertile land for growing vegetables is at risk — but a scientific discovery could turn the tide

~	Ecology of Fear: Mike Davis’ history of LA and natural disaster is re-read whenever fire rages in California

~	Surveillance tech is changing our behaviour – and our brains

~	Long-acting contraceptives seem to be as safe as the pill when it comes to cancer risk: new study