The new system based on artificial intelligence predicts the weather quickly and accurately

Aardvark Weather's artificial intelligence system can predict the weather faster and more accurately than traditional methods, according to a study published in The Guardian's Nature journal.

Currently, meteorologists use complex algorithms to forecast the weather. This process takes hours on supercomputers and is managed by a team of specialists. However, scientists from the University of Cambridge, in collaboration with the Alan Turing Institute, Microsoft Research, and the European Centre for Medium-Range Weather Forecasts, have developed a new, AI-based approach.

The researchers trained a neural network to analyze data from various sources—weather stations, satellites, aerostats, ships, and airplanes. This allows for accurate forecasts without using traditional modeling methods.

According to Richard Turner, a professor of artificial intelligence at the University of Cambridge, the Aardvark Weather system provides tailored forecasts for various sectors, from agriculture to wind energy. Additionally, this technology enables a specialist to create accurate weather forecasts even on a simple computer.

Moreover, Aardvark Weather can provide accurate forecasts up to eight days in advance. This is significantly longer than the five-day forecasts offered by modern methods. The developers emphasize that the new AI system can also provide precise information for specific regions.

Photo: Murdo MacLeod/The Guardian