They didn’t see that coming! Google DeepMind AI can accurately predict the weather up to 10 days in advance, leaving forecasters and even top supercomputers out in the cold

• Google DeepMind AI can predict storms like Debi quicker than normal methods 

An AI developed by Google can accurately predict the weather as many as ten days in advance, outperforming traditional forecasting methods and even top supercomputers.

The new Google DeepMind model, GraphCast, was trained using 40 years of meteorological data from weather stations, satellite images and radar records and can create forecasts in less than a minute drawing from its machine learning.

The tool was used to predict the path of Hurricane Lee in September three days before traditional methods were able to, as it spotted it would make landfall in Nova Scotia nine days before it hit.

The forecasts made by GraphCast could mean that storms including Debi, which has battered the south coast this week, could be spotted much sooner.

An AI developed by Google can accurately predict the weather as many as ten days in advance, outperforming traditional forecasting methods and even top supercomputers

Pushmeet Kohli, Google DeepMind’s vice president of research said: ‘Weather prediction is one of the most challenging problems that humanity has been working on for a long, long time.

‘GraphCast is really very accurate compared to traditional systems and it’s incredibly fast. It can make predictions in under a minute. It is very exciting.’

Developers said the technology ‘marks a turning point in weather forecasting’.

A scientific paper published in the journal Science on Tuesday concluded that Google’s AI was more accurate than the most advanced forecasting system, the European Medium Range Weather Forecasting Model.

It was able to outperform the EMRWFM on 90 per cent of the 1,380 metrics that were tested, including temperature, air pressure, wind speed, humidity and atmosphere levels.

The forecasts made by GraphCast could mean that storms including Debi, which has battered the south coast this week, could be spotted much sooner (waves crashing in Folkestone, Kent)

Machine-learning coordinator at ECMWF Matthew Chantry said that weather forecasting systems were progressing ‘far sooner and more impressively than we expected even two years ago’.

Google DeepMind’s Remy Lam from said that AI could not yet replace the currently used supercomputers and that the new technology will complement the older system.

‘We are standing on the shoulders of giants to build those models’, he said, according to the BBC.

‘AI models are trained from data and that data is generated by traditional approaches, so we still need the traditional approach to gather data to train the model,’ Mr Lam said.

Source: Read Full Article