First protein foldingNow weather forecasting: London-based AI firm DeepMind continues to be powered by applying deep learning to challenging science problems. Working with the Met Office, the UK’s national weather service, DeepMind has developed a deep-learning tool called DGMR that can accurately predict the probability of precipitation in the next 90 minutes – one of the toughest weather forecasting challenges.
In a blind comparison with existing tools, dozens of experts judged the DGMR forecast to be the best across a range of factors — including its predictions of location, range, movement, and rain intensity — 89% of the time. The results were published in nature paper Today.
New DeepMind Tool nothing AlphaFold, who solved a major problem in biology Which scientists have been struggling with for decades. However, the slight improvement in the prediction is significant.
Forecasting rain, especially heavy rain, is critical to many industries, from outdoor events to aviation to emergency services. But doing it well is difficult. Knowing how much water is in the sky, and when and where it will fall, depends on a number of weather processes, such as changes in temperature, cloud formation, and wind. All of these factors are complex enough by themselves, but they are even more complicated when taken together.
The best current forecasting techniques use massive computer simulations of atmospheric physics. These work well for long-term forecasting but are less good at predicting what will happen in the next hour or so, known as nowcasting. Previous deep learning techniques have been developed, but they usually work well at one thing, such as predicting location, at the expense of another, such as predicting density.
“Precipitation nowcasting remains a huge challenge for meteorologists,” says Greg Karpin, head of forecast operations at the NOAA Weather Prediction Center in the United States, who was not involved in the work.
The DeepMind team trained the AI on radar data. Many countries release frequent snapshots throughout the day of radar measurements that track cloud formation and movement. In the UK, for example, a new reading is released every five minutes. Putting these footage together provides an updated stop-motion video that shows how rain patterns are moving across the country, similar to the forecast images you see on TV.