NVIDIA's CEO Jensen Huang has introduced an Earth Climate Digital Twin, a platform designed to simulate and visualize weather and climate at an unprecedented scale. This new offering aims to accelerate efforts in combating the staggering $140 billion in economic losses caused by extreme weather events driven by climate change.
Earth-2 cloud APIs, part of the newly announced NVIDIA CUDA-X microservices enable users to create AI-powered emulations. These emulations can deliver interactive, high-resolution simulations of various weather and climate phenomena, ranging from the global atmosphere and local cloud cover to typhoons and turbulence, at a remarkable speed. When combined with proprietary data from companies in the $20 billion climate tech industry, Earth-2's APIs can generate warnings and updated forecasts in mere seconds, a significant improvement from the minutes or hours required by traditional CPU-driven modeling.
One of the key components of Earth-2 is CorrDiff, a revolutionary NVIDIA generative AI model that employs state-of-the-art diffusion modeling. CorrDiff generates images at a resolution 12.5 times higher than current numerical models, while operating 1,000 times faster and 3,000 times more energy-efficiently. This first-of-its-kind generative AI model can deliver super-resolution, synthesize new metrics of interest to stakeholders, and learn the physics of fine-scale local weather from high-resolution datasets.
The Central Weather Administration (CWA) of Taiwan plans to leverage these diffusion models to predict more precise locations of typhoon landfall. "Taiwan is a critical component of the global supply chain, and flooding risk analysis and evacuation preparedness are core to our mandate," said Chia-Ping Cheng, administrator of CWA. With over 136 typhoons striking the island since 2000, using Earth-2 to mitigate these impacts is crucial for improving the quality and resolution of disaster informatics, according to Taiwan's National Science and Technology Center for Disaster Reduction (NCDR).
Another essential component of Earth-2 cloud APIs is NVIDIA Omniverse, a computing platform that allows individuals and teams to develop Universal Scene Description (OpenUSD)-based 3D workflows and applications. The Weather Company, a global leader in weather data forecasting and insights, intends to integrate its meteorological data and Weatherverse tools with Omniverse. This integration will enable customers building digital twins to better understand and visualize the impact of actual weather conditions for the first time. The Weather Company also plans to explore the use of NVIDIA score-based generative AI for its Weatherverse services, Weather Engine solution for enterprise-level weather intelligence, and new high-resolution weather modeling products.
Other early adopters of Earth-2 APIs include weather analytics platform companies like Spire and Meteomatics, as well as startups such as Tomorrow.io, north.io, and ClimaSens, which are exploring new solutions for climate tech applications.
Earth-2 APIs utilize DGX Cloud to provide full-stack acceleration for climate and weather solutions, including optimal AI pipelines for models like FourCastNet, GraphCast, and Deep Learning Weather Prediction, as well as GPU acceleration of numerical weather prediction models like ICON on the latest NVIDIA Grace Hopper systems. Running on NVIDIA DGX GH200, HGX H100, and OVX supercomputers, Earth-2 may provide a path to simulate and visualize global climate simulations at unprecedented speed and scale.
As climate disasters become increasingly frequent and severe, platforms like NVIDIA's Earth-2 offer a glimmer of hope in our ability to better prepare for and potentially moderate the impacts of extreme weather events.