short: Aardvark Weather, a system based on artificial intelligence, is a significant enhancement of weather forecasting by providing predictions of dozens of times faster while using thousands of times the power of computing is less than current methods. This system was developed by researchers at the University of Cambridge, with the support of the Alan Torring Institute, Microsoft Research, and the European Medium -range weather forecast.
The speed and efficiency of modern prediction systems is vital, as traditional methods depend on strong giant computers and wide teams of experts, and often require several hours to produce predictions.
Modern innovations of technology giants such as Huawei, Google and Microsoft have proven that artificial intelligence can significantly improve specific aspects of the prediction process, including numerical solutions, which are decisive in weather prediction while simulating how weather conditions develop over time. These companies have made faster and more accurate predictions by integrating artificial intelligence into these solutions.
For example, Google develops Amnesty International weather forecasts and is currently marketing Two models of its cloud customers for the institution. The models were developed by Google DeepMind, historical weather data is used to predict future conditions 10 to 15 days.
Aardvark is a great progress by replacing traditional predictions with one simplified educational model. Using a standard desktop computer, it can process data from different sources, including satellites and weather stations, to generate global and local expectations in minutes.
“Aardvark is re -depicting the current weather, providing the ability to make weather forecast faster, cheaper, more flexible and more accurate than ever,” Make up Professor Richard Turner from the Ministry of Engineering at Cambridge, who led the research. “Aardvark is thousands of times faster than all previous weather.”
Despite working with only a small part of the data used by the current systems, Aardvark exceeds the American national prediction system in many major standards and is still able to compete with expectations from national weather service, which usually includes multiple models and expert analysis.
“these results It is just the beginning of what Aardvark can achieve. The first author Anna Allen of the Cambridge Computer and Technology Department. She said that a comprehensive learning approach can be easily applied to other weather forecast problems, such as air quality and marine ice dynamics.
One of the most interesting aspects of Aardvark is its flexibility and simple design. Since he learns directly from the data, it can be quickly adapted to produce detailed predictions for specific industries or sites, whether it is a prediction of temperatures to support African agriculture or wind conditions for European renewable energy companies. This contradicts sharply with traditional systems, which require years of work by major specialization teams.
This ability has the ability to convert weather forecasts in developing countries, where access to experience and mathematical resources is limited. “By converting weather forecasts from giant computers to desktop computers, we can weaken prediction, making these strong technologies available to developing countries and regions that combine data around the world,” said Dr. Scott Hawking of the Alan Torring Institute.
Aardvark is expected to play an important role in expanding weather forecasting. Turner stated that the model can eventually predict accurately with eight -day expectations, exceeding the capabilities of current models for three days. This progress, along with the ability to adapt and Aardvark efficiency, places it as a transformative meteorological power.
The next steps for Aardvark include the development of a new team at the ANAN TURING Institute, which will explore the deployment of technology in the global south and integrate it into broader environmental prediction initiatives.