2024 is shaping up to be a spectacular year for digital twin of Earth.

In March, NVIDIA launched the Earth-2 climate digital twin platform, while the European Union's DestinE entered its first phase in June. The platform will continuously integrate data and aims to become a "complete" digital twin of Earth by 2030.

With the assistance of AI, the development of Earth’s digital twin is accelerating. Typhoon prediction paths are becoming faster and more accurate; scientists can simulate landslides repeatedly to find the best methods for disaster prevention; and businesses can determine the optimal locations for wind farms. The EU also plans to use this technology to protect biodiversity, assess the impact of climate change on society and the economy, and even apply it to policy adjustments.

How is Earth-2 Created? Insights from NVIDIA Experts

NVIDIA's CEO Jensen Huang stirred up a "NVIDIA whirlwind" during his visit to Taiwan in May. In June, he took the stage at National Taiwan University, showcasing Earth-2, which is touted as one of the "most ambitious projects in human history," on the global stage.

Earth-2 utilizes artificial intelligence and high-resolution interactive simulations to accelerate climate and weather forecasting. It enhances satellite image resolution by more than 12 times and reduces typhoon prediction range errors from 25 kilometers to just 2 kilometers, providing more accurate forecasts of landfall locations.

While digital twins are not a new technology—first applied in manufacturing in 2002—their use has exploded with the development of AI. A digital twin is a virtual representation of a physical product. For example, an aircraft has a virtual counterpart created with the same details and processes from the design and production phases. Flight data is recorded in this virtual twin, allowing for pre-testing and adjustments before applying changes to the physical aircraft. The same principle applies to digital twin factories.

Earth-2 emphasizes its ability to rapidly predict climate and weather. However, the weather system is extremely complex. To virtually replicate Earth's weather system, it requires not only high computing power and AI technology but also extensive training with large datasets. Dion Harris, Director of Product Marketing for NVIDIA’s Data Center, told Live Science that the training data includes hourly estimates of land, atmospheric, and oceanic climate variables. After integrating hundreds of observational results, Earth-2 can reconstruct Earth's weather and climate from the past 50 years.

NVIDIA’s team explained that Earth-2 can assist not only in predicting typhoons but also in long-term climate change predictions. Before this, Earth-2 must first rapidly analyze vast streams of data, including greenhouse gas emissions, temperature changes, global water cycles, land use changes, and ocean chemistry. NVIDIA’s supercomputing capabilities are well-suited for this task. 

The European Union’s Ambition: A "Complete" Digital Twin Earth

The "DestinE" project was launched in 2022. The European Union’s ambition is not only to analyze climate change and extreme weather events but also to use the digital twin to study their social and economic impacts, aiming to find adaptation and mitigation strategies. Besides applications in biodiversity protection, water resource management, and promoting renewable energy, it may also be used to adjust EU policy-making and implementation.

DestinE utilizes Europe’s vast data lake for simulations and predictions, which includes data from the European Space Agency, the European Organisation for the Exploitation of Meteorological Satellites, and the European Centre for Medium-Range Weather Forecasts. Additionally, it incorporates IoT data from the ground and social and economic data provided by governments.

By the end of this year, DestinE has completed its core platform, data lake, and two digital twin models, with plans to integrate data on oceans, cities, and forests. By 2030, it aims to become a "complete" digital twin of Earth.

Want to know how heatwaves will affect food security? How extensive is the flooding? Where is the best location for wind farms? The digital twin Earth will provide the answers.

Reference:

Live Science（2024/4/16），AI-powered 'digital twin' of Earth could make weather predictions at super speeds

Space.com（2024/3/5），Scientists made a 'digital twin' of Earth to simulate possible natural disasters

The Next Web（2024/6/10），EU launches AI-powered ‘digital twin’ of the Earth

European Commission，Destination Earth

Nvidia，earth-2

Source: Environmental Information Center