Google DeepMind, the artificial intelligence company, has announced a new AI-powered weather forecasting model that can provide 15-day forecasts with superior accuracy and speed. This tool is seen as a potential lifesaver in the context of escalating climate change.
The GenCast model forecasts more accurately than existing leading global models.
According to the London-based company, the model named GenCast has demonstrated the ability to forecast more accurately than the leading models in the world. Currently, the European Centre for Medium-Range Weather Forecasts (ECMWF) provides forecasts for 35 countries and is considered the global standard for accuracy in meteorology.
However, representatives from DeepMind stated that GenCast’s forecasts surpassed ECMWF’s accuracy in over 97% of 1,320 real-world scenarios tested since 2019. The new model was trained using 40 years of data on temperature, wind speed, and air pressure from 1979 to 2018, and it can deliver a 15-day weather forecast in just 8 minutes, significantly less than the current hourly timeframe.
DeepMind’s statement confirms that GenCast provides more accurate forecasts for both daily weather and extreme weather events compared to leading forecasting systems. With its superior forecasting capabilities, DeepMind believes that this intelligent model could help mitigate the damage to life and property caused by natural disasters.
| Scientists warn that extreme weather is becoming more common and severe due to climate change. In August 2023, a series of wildfires in Hawaii (USA) resulted in approximately 100 deaths. Local residents criticized authorities for failing to provide early warnings of the wildfire risks. This summer, an unusual heatwave in Morocco resulted in at least 21 deaths within 24 hours. Last September, Hurricane Helene claimed the lives of 237 people in Florida and other southeastern states in the USA. |
