Nvidia has announced the launch of Earth-2, a new family of open-source AI models and tools designed to transform weather and climate forecasting. The announcement, made at the American Meteorological Society’s annual meeting, introduces a powerful suite of technologies aimed at making advanced weather prediction more accessible to scientists, startups, and government agencies across the globe.
The Earth-2 platform offers an open, accelerated software stack that significantly reduces the computational time and cost associated with traditional physics-based weather modeling, which has historically relied on powerful and expensive supercomputers.
Introducing The Earth-2 AI Models
The new platform includes several state-of-the-art models, each targeting different forecasting needs. The Earth-2 Medium Range model can generate high-accuracy global forecasts for over 70 weather variables up to 15 days in advance.
For short-term, high-impact events, the Earth-2 Nowcasting model uses generative AI to produce high-resolution predictions of local storms and hazardous weather within a zero- to six-hour window. Additionally, the Earth-2 Global Data Assimilation model rapidly processes atmospheric data from satellites, balloons, and weather stations to create the initial conditions needed to launch accurate forecasts.
Israel Meteorological Service Adopts Nvidia’s AI
The new technology is already being tested and implemented by organizations in the MENA region. The Israel Meteorological Service is using Earth-2 models to generate high-resolution forecasts, enabling faster and more effective responses to extreme weather events while cutting down on computational expenses.
“NVIDIA Earth-2 models give us a 90% reduction in compute time at 2.5-kilometer resolution compared with running a classic numerical weather prediction model without AI on a CPU cluster,” said Amir Givati, director of the Israel Meteorological Service. “After a recent rainstorm, our AI model trained with CorrDiff was the best of all our operational models for a six-hour verification of accumulated precipitation.”
This adoption highlights the significant potential for AI-driven meteorology to enhance public safety and resource management in a region frequently facing climate-related challenges.
Global Adoption and Industry Impact
Beyond government weather agencies, Earth-2 is attracting interest from various industries. Energy companies like TotalEnergies and Eni are evaluating the models to improve risk awareness and forecast energy demand. The platform’s ability to provide faster, more accurate, and localized weather data is crucial for optimizing grid operations and managing renewable energy sources.
Financial firms are also leveraging the technology for risk assessment, using the models to generate scenarios for events like hurricanes to better understand potential impacts. By making these powerful tools open source, Nvidia aims to foster a global ecosystem of innovation in weather intelligence.
About Nvidia
Nvidia is a global technology company known for its invention of the GPU (graphics processing unit). The company’s work in accelerated computing and AI has been a driving force in markets such as gaming, professional visualization, data centers, and automotive. Nvidia’s platforms and software are widely used to tackle complex computational challenges across various scientific and industrial fields.
Source: NVIDIA Blogs
