Google DeepMind and Google Research have introduced WeatherNext 2, a new artificial intelligence model for weather forecasting. It is the successor to the GenCast model that was announced last year. The company said that WeatherNext 2 is eight times faster and offers higher resolution for one hour, which allows more detailed monitoring of weather conditions. This is also the first time Google is offering access to the model outside its research labs.
In a blog post, the company shared details about the new model and confirmed its wider availability. WeatherNext 2 data can now be accessed through Earth Engine and BigQuery. It is also being offered on Google Cloud’s Vertex AI platform through an early access programme. Google already uses this technology in Search, Gemini, Pixel Weather and the Weather application programming interface of Google Maps Platform.
WeatherNext 2 can generate hundreds of possible weather scenarios from a single input in less than a minute, using only one Tensor Processing Unit. This is Google’s own chip designed for artificial intelligence processing. This is important because traditional weather forecasting models depend on physics based simulations which can take hours to run on supercomputers.
The company said the new model shows measurable improvements. WeatherNext 2 is reported to outperform GenCast on 99.9 percent of key variables such as temperature, humidity and wind. It performs strongly across lead times of up to fifteen days. It also offers higher temporal resolution, which provides more detailed forecasts on an hour by hour basis.
The technology behind the model uses an architecture called a Functional Generative Network. Instead of producing one single forecast, this network introduces structured “noise” into the model’s parameters. This allows it to create multiple realistic weather outcomes.
The model learns meteorological elements known as marginals and joints. Marginals refer to individual variables like temperature or wind speed in a specific place. Joints show how different variables interact across a region. Although WeatherNext 2 is trained only on marginal data, Google DeepMind says it performs strongly when predicting joints as well. This is important for complex weather events such as heatwaves or storms.
Also read: Viksit Workforce for a Viksit Bharat
Do Follow: The Mainstream formerly known as CIO News LinkedIn Account | The Mainstream formerly known as CIO News Facebook | The Mainstream formerly known as CIO News Youtube | The Mainstream formerly known as CIO News Twitter
About us:
The Mainstream formerly known as CIO News is a premier platform dedicated to delivering latest news, updates, and insights from the tech industry. With its strong foundation of intellectual property and thought leadership, the platform is well-positioned to stay ahead of the curve and lead conversations about how technology shapes our world. From its early days as CIO News to its rebranding as The Mainstream on November 28, 2024, it has been expanding its global reach, targeting key markets in the Middle East & Africa, ASEAN, the USA, and the UK. The Mainstream is a vision to put technology at the center of every conversation, inspiring professionals and organizations to embrace the future of tech.
