- Google DeepMind develops AI-based ‘now-casting’ system to predict rain
- It makes its predictions based on the last 20 minutes of high-resolution radar data
- Now-casting model predicts moderate to heavy rainfall for next 90 minutes
- Meteorologists have ranked it ahead of two existing instruments for accuracy.
We’ve all been there: running out of the house without an umbrella only to be caught in an unexpected rain shower.
But now experts at Google DeepMind have developed an artificial intelligence-based ‘now-casting’ system, which they claim is more accurate in predicting the probability of rain within the next 90 minutes than existing models.
It uses high-resolution radar data from the past 20 minutes to predict whether moderate to heavy rain is likely to fall two hours ahead.
scroll down for video
How does this work? This graphic shows how Google DeepMind’s system uses high-resolution radar data from the past 20 minutes to make accurate predictions on upcoming rainfall
Experts at Google DeepMind have developed an artificial intelligence-based ‘now-casting’ system, which they claim is more accurate in predicting the probability of rain within the next 90 minutes than existing models.
HOW GOOGLE’S AI NOW-CASTING SYSTEM WORK?
The now-casting system developed by Google’s London-based tech company DeepMind relies on high-resolution radar data.
The radar repeatedly fires a beam into the lower atmosphere which then tracks the amount of moisture in the air.
It does this by measuring the relative speed of the signal, which is slowed down by water vapor.
Google DeepMind’s AI system, which uses radar data collected every five minutes, then looks at information from the past 20 minutes and provides predictions for the next 90 minutes using modeling tools.
This is called the Deep Generative Model of Rain (DGMR).
The radar repeatedly fires a beam into the lower atmosphere to track the amount of moisture in the air, which is measured by the relative speed of the signal and how much it is slowed by water vapor.
This data is used by AI modeling tools in an effort to pinpoint the timing, location and intensity of rainfall.
It is expected that this can improve the accuracy of short-term weather forecasts, and in particular the prediction of storms and heavy rain.
That’s because current supercomputer models — used largely to forecast weather over the next day or week — don’t perform so well with a shorter time frame of two hours.
They rely heavily on Numerical Weather Prediction (NWP) systems, which use mathematical equations to predict the probability of rain and other types of weather based on the movement of fluids in the atmosphere.
Suman Ravuri, a staff research scientist, said, “These models are really amazing in terms of weather prediction, ranging from six hours to almost two weeks, but there is one area – especially from about zero to two hours – in which the model is particularly useful.” perform poorly.” DeepMind in London and co-head of the project.
The Meteorological Department’s four-day forecast is now as accurate as one day’s forecast 30 years ago, while 92 percent of its next day’s temperature forecast is accurate to within 2 degrees Celsius and the next day’s wind speed forecast is as accurate as the next day’s forecast. 91 percent are correct. Within 5 knots.
But short-term forecasts for rain are often not as reliable as some experts would like.
The DeepMind team’s tool was evaluated by more than 50 meteorologists in conjunction with two existing rain prediction systems, which ranked it first for accuracy and usefulness in 89 percent of cases.
However, DeepMind has not revealed how accurate its tool is compared to the existing tools used by Met Office.
DeepMind’s senior staff scientist said, “It’s still very early days, but this testing shows that AI can be a powerful tool, enabling forecasters to spend less time sifting through a growing pile of prediction data and instead focuses on better understanding the implications of their predictions.” Shakir Mohammed.
‘This will be integral to mitigating the adverse effects of climate change today, supporting adaptation to changing weather patterns and potentially saving lives.’
DeepMind worked with the Met Office in the UK to try to turn this tool into something that could be useful to prophets.
The AI system uses the last 20 minutes of radar data (pictured at left as the target) and then predicts rainfall using its Deep Generative Model of Precipitation (right).
The study showed that this AI-based ‘deep generative modeling’ outperformed other ab-casting methods on a wide range of measures.
The system ranked first for accuracy and usability by 89 percent of a panel of 56 Met Office meteorologists
Niall Robinson, Head of Partnerships and Product Innovation at the Met Office, said: ‘Improving the accuracy of short-term forecasts is an incredibly important effort.
‘Extreme weather has catastrophic consequences, including the loss of life, and as the effects of climate change reveal, these types of events become more common.
‘That way, better short-term weather forecasts can help people stay safe and thrive.
‘This research demonstrates the potential AI can offer as a powerful tool to improve our short-term forecasts and our understanding of how our weather patterns are evolving.’
He said the Met Office is now looking at how to use DeepMind research in its forecast.
The authors write in their paper: ‘We show that generative now-casting can provide probabilistic predictions that improve forecast value and support operational utility, and at resolution and lead time where alternative methods struggle. ‘
The research is published in the journal Nature.