Nvidia announces Earth-2 digital twin to forecast planet’s climate change

Nvidia unveiled Earth-2 Cloud Platform for predicting climate change across the entire planet, using simulation by AI supercomputers.

In a bid to combat the increasing economic losses due to extreme weather patterns caused by climate change, Nvidia has introduced Earth-2, a revolutionary climate digital twin cloud platform. Revealed at GTC, Earth-2 offers groundbreaking APIs designed to simulate and visualize weather and climate at an unprecedented scale, paving the way for more accurate forecasts and timely warnings.

The company announced Earth-2 during the keynote by Jensen Huang, CEO of Nvidia, at the Nvidia GTC 2024 event in San Jose, California. Earth-2 was first announced in 2021, and now it’s ready.

“Climate disasters are now normal — historic droughts, catastrophic hurricanes and generational floods appear in the news with alarming frequency,” said Huang. “Earth-2 cloud APIs strive to help us better prepare for — and inspire us to act to moderate — extreme weather.”

Earth-2 itself is software that can run across a variety of supercomputers, with its detailed forecasting dependent on just how much computer power various partners throw at the problem. The long-term aim is to forecast climate change for decades to come.

The Earth-2 platform, part of the Nvidia CUDA-X microservices software, leverages advanced AI models and the CorrDiff generative AI model to produce high-resolution simulations 1,000 times faster and 3,000 times more energy-efficient than current numerical models.

On stage, Huang said most of the damage from disasters is concentrated in certain parts of the world.

“In scientific computing, climate research remains a grand challenge,” said Rev Lebaredian, vice president of simulation at Nvidia, in a press briefing. “We are announcing our Earth 2 platform for interactive, high resolution climate and weather prediction. Extreme weather events cost $140 billion in economic losses every year. Affordable climate simulations are too high level.”

He added, “We need kilometer-scale simulations to really make an impact. Longer scale simulation is costly, requiring a million times the compute. [It’s] a cloud platform for government agencies, weather forecasters, and prediction agencies to accelerate weather employment prediction.”

To elaborate on Earth 2, Nvidia is making software for it that can run on different supercomputers, and the customers or government agencies of the world will determine what kind of supercomputer that will be. Nvidia is showing off new services that can calculate detailed models of the Earth’s climate that are 1,000 times faster and 2,000 times more energy efficient than today’s numerical weather prediction processes, Lebaredian said.

Nvidia is providing accelerated simulation as a service, with better performance, energy efficiency, lower cost and better visualization powered by Omniverse Cloud APIs.

How it works

The Earth-2 APIs empower users to create interactive simulations ranging from global atmospheric conditions to local weather phenomena like typhoons and turbulence, enabling them to deliver real-time forecasts and warnings in seconds.

Taiwan’s Central Weather Administration is among the first to adopt Earth-2’s diffusion models to enhance typhoon forecasts, aiming to minimize casualties through early evacuations. By utilizing Nvidia’s advanced generative AI technology, Earth-2 promises to revolutionize disaster preparedness and response strategies, potentially saving lives in regions prone to extreme weather events.

Additionally, Earth-2 integrates with Nvidia Omniverse, enabling organizations like The Weather Company to develop 3D workflows that incorporate real weather data, enhancing their digital twin environments for better analysis and simulation of weather impacts. The Weather Company plans to leverage Earth-2 APIs to create higher-resolution simulations and develop new weather modeling products, ultimately improving weather intelligence services for enterprise clients.

Early adopters of Earth-2 APIs include weather analytics platforms like Spire and Meteomatics, as well as startups exploring climate tech solutions. Powered by Nvidia DGX Cloud, Earth-2 provides full-stack acceleration for climate and weather solutions, offering optimal AI pipelines and GPU acceleration for numerical weather prediction models.

With Earth-2, Nvidia aims to empower organizations worldwide to better understand, analyze, and mitigate the impacts of weather and climate change, marking a significant step forward in the fight against environmental challenges. It visualizes and simulates weather at a two-kilometer scale.

When combined with proprietary data owned by companies in the $20 billion climate tech industry, the Earth-2 application programming interfaces help users deliver warnings and updated forecasts in seconds compared to the minutes or hours in traditional CPU-driven modeling.

“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, in a statement.

With more than 136 typhoons striking the island since 2000, using Earth-2 to mitigate these impacts is key to improving the quality and resolution of disaster informatics, NCDR said.

The Weather Company, a global leader in weather data forecasting and insights, plans to integrate its meteorological data and Weatherverse tools with Omniverse, enabling 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.

“To help effectively address current and future weather- and climate-related challenges, it’s critical now more than ever to incorporate reliable, globally scaled real weather data and insights into digital twin environments to better analyze, plan and simulate the impacts of weather,” said Sheri Bachstein, CEO of The Weather Company, in a statement. “We’ve worked with Nvidia for years on GPU acceleration of GRAF, our proprietary weather modeling systems, and we plan to adopt Earth-2 APIs to create
higher resolution, energy-efficient simulations at a lower cost.”

Earth-2 APIs use DGX Cloud to provide full-stack acceleration for climate and weather solutions. This includes optimal AI pipelines for models such as FourCastNet, GraphCast and Deep Learning Weather Prediction. It also includes GPU acceleration of numerical weather prediction models like Icon on the latest Grace Hopper systems.

Running on Nvidia DGX GH200, HGX H100 and OVX supercomputers, Earth-2 may provide a path to simulate and visualize the global climate simulations at unprecedented speed and scale.