Earth-2

Platform for developing accelerated, AI-augmented, high-resolution climate and weather solutions with interactive visualization.

Overview

Climate Digital Twin Cloud Platform

NVIDIA Earth-2 combines the power of AI, GPU acceleration, physical simulations, and computer graphics to develop applications that can simulate and visualize weather and climate predictions at a global scale with unprecedented accuracy and speed. The platform consists of development tools, microservices, and reference implementations for AI, visualization, and simulation. NVIDIA NIM™ microservices for Earth-2 allow users to leverage AI-accelerated models to optimize and simulate real-world outcomes for climate and weather.

Climate Tech Companies Adopt NVIDIA Earth-2 for High-Resolution, Energy-Efficient, More Accurate Weather Predictions and Disaster Preparedness

Reference architecture for developing AI-powered weather analysis and forecasting applications.

Read Press Release Try the Blueprint Now

The Development Platform for Climate Science

Higher Resolution and Large-Scale AI Training

The Earth-2 accelerated systems will let climate scientists produce kilometer (km)-scale climate simulations, conduct large-scale AI training and inference, and achieve low-latency interactivity. NVIDIA PhysicsNeMo integrates support for numerous neural network models for climate and weather simulation.

GPU-Optimized and Accelerated Climate Simulation

The Earth-2 development platform is optimized for GPU-accelerated numerical climate simulations at km-scale to maximize simulated days per day (SDPD).

Data Federation and Interactive Weather Visualization

NVIDIA Omniverse enables ultra-large-scale, high-fidelity, interactive visualizations that depict weather conditions across the globe. Omniverse Nucleus includes a data federation engine that offers transparent data access across external databases and live feeds.

Earth-2 Tools and Microservices

AI

With Earth-2’s AI tools and microservices, climate and weather application developers can leverage reference AI inference pipelines using pretrained models and training pipelines and NVIDIA PhysicsNeMo for fine-tuning with custom data. Earth-2 offers a portfolio of community models. These models are transformational in their ability to efficiently generate large ensembles or high-resolution predictions by downscaling.

Watch Demo Try Earth-2 NIMs

Visualization

Interactive visualization microservices enable large-scale weather and climate data to be visualized and analyzed. NVIDIA Omniverse™ Blueprint for Earth-2 weather analytics showcases how developers can use the Omniverse SDK and microservices to build NVIDIA RTX™-powered visualization pipelines for rendering geospatial and weather data. The blueprint also provides a template for partners to integrate their data platforms into AI pipelines.

Watch Demo Try the Blueprint Now

Simulation

Simulation microservices will make it possible to encapsulate, orchestrate, and accelerate numerical weather prediction (NWP) models on NVIDIA GPU platforms.

Simulation services will be coming soon—please reach out to the Earth-2 team to be part of the development process.

Watch Demo

Earth-2 AI Stack

Earth2Studio

Python-based GPU-accelerated package designed to quickly get users up and running with experimenting and prototyping with various state-of-the-art AI weather and climate models.

Get Started

CorrDiff NIM for Generative AI Downscaling

Enables generative AI downscaling by 500X faster and improvement of energy efficiency by 10,000X. This augments current applications and workflows, allowing enterprises—now available in the U.S.—to generate more datasets to get better probabilistic distributions for weather events​.​

Get Started With NIM

FourCastNet NIM for Global Weather Forecasting

Accelerates AI-based global weather forecasting, enabling enterprises to develop solutions that use large datasets by up to 20X larger to capture extreme weather events while accelerating and maintaining energy efficiency.

Get Started With NIM

PhysicsNeMo

Physics AI training framework used to train the NIM microservices at scale on petabyte-size datasets like ERA5, HRRR, etc. Developers can use the training pipelines to customize AI weather models on custom data.

Download PhysicsNeMo

Leading Adopters Across Industries

Explore Earth-2 Demos

Earth-2 Goes Down to Street Level

City-scale simulation data is now incorporated  into the Earth-2 planetary digital twin. In this video, we demonstrate the combination of high-resolution simulation data from ICON, WRF, and PALM with Google Photoreal Tiles served by Cesium using the Earth-2 visualization service to help address questions in the urban environment.

Watch Demo (02:50)

Visualize AI-Augmented Weather Simulations

Researchers are using interactive visualization of AI weather forecasts, simulation data, and archived data to analyze extreme weather.

Watch Generative AI Super-Resolution Local Prediction of Weather Extremes Demo (01:27)
Accelerating Carbon Capture and Storage with Fourier Neural Operator and NVIDIA PhysicsNeMo

Accelerating Carbon Capture and Storage with Fourier Neural Operator and NVIDIA PhysicsNeMo

By accelerating analysis 700,000X, NVIDIA Omniverse and PhysicsNeMo can help engineers with planning and operating carbon capture and storage, ensuring safe operation and long-term storage and reducing the amount of carbon dioxide released into our atmosphere.

Watch Carbon Capture Storage Demo (01:17)
Predicting Extreme Weather Events Three Weeks in Advance With FourCastNet

Predicting Extreme Weather Events Three Weeks in Advance With FourCastNet

By running FourCastNet in NVIDIA PhysicsNeMo, we were able to generate 21-day weather trajectories of 1,000 ensemble members in a tenth of the time it previously took to do a single ensemble—and with 1,000X less energy consumption.

Watch Predict Extreme Weather Events with FourCastNet Demo (01:40)
Interactive Visualization of High-Resolution, Global-Scale Climate Data in the Cloud

Interactive Visualization of High-Resolution, Global-Scale Climate Data in the Cloud

Built on NVIDIA Omniverse and the OpenUSD 3D framework, the Earth-2 platform enables aggregation and visualization of diverse, global-scale climate simulation and geospatial datasets. Made possible with cloud-native technology, visualizations can be explored by anyone around the globe.

Watch Global Climate Visualization Demo (01:31)

See Earth-2 in Action

Launch Visualization Demo Launch Forecasting Demo

Announcements

Leading Climate Tech Companies Adopt NVIDIA Omniverse Blueprint for Earth-2 Weather Analytics

Enabling high-resolution, energy-efficient, and more accurate weather prediction.

Read Press Release

Announcements

First-Ever Earth-2 NIMs CorrDiff NIM for Generative AI

New NVIDIA NIM microservices offer a leap in enabling weather technology companies to develop and deploy AI models for predictions of snow, ice and hail.

Read Blog: NVIDIA Launches Earth-2 NIM Microservices for 500x Speedup in Delivering Higher-Resolution Simulations

Announcements

Groundbreaking Generative AI for Climate Tech

The Weather Company and Central Weather Administration of Taiwan are among the first to adopt the new Earth-2 Cloud APIs, using AI to speed creation of high-resolution simulations and visualization of the global climate and weather at a groundbreaking 2-kilometer scale

Read Press Release: NVIDIA Announces Earth Climate Digital Twin

Announcements

AI, Digital Twins Key to Addressing Climate Challenge, NVIDIA CEO Says

Key technologies will help climate researchers achieve the miracles they need to achieve breakthroughs in climate research.

Read Blog: Digital Twins and Climate Research Innovation

Announcements

Who’ll Stop the Rain? Scientists Call for Climate Collaboration

Three researchers describe plans to build digital twins of Earth to help people understand and adapt to the effects of global warming.

Read Blog: Scientists Call for Climate Collaboration
Products
Technologies
Resources
Company Info