Electronics Help How a Single AI World Model Simulates the Earth System

Related Vendors

FAULHABER Drive Systems

Taiwan Machine Tool & Accessory Builders' Association (TMBA)

Wuxi InfiMotion Technology Co., Ltd.

Researchers at KIT are developing an AI world model that combines global climate, weather, and environmental models to predict such events worldwide more accurately, quickly, and energy-efficiently than traditional physics-based simulations on supercomputers.

Global climate change is gradually altering weather and the environment—both globally and locally. The consequences range from more frequent extreme events to profound impacts such as the collapse of entire ecosystems. "Numerical models of climate, weather, and the environment now allow many aspects of these developments to be represented relatively consistently across different time and spatial scales," says Tenure-Track Professor Peer Nowack from the Institute of Theoretical Informatics at KIT and coordinator of the project.

"Modern AI methods, in turn, offer the potential to either emulate such models cost-effectively—that is, mimic them—or even directly learn relationships from observational data. This often leads to better results, for example, in weather forecasting. This presents great opportunities for overall environmental modeling."

In the project "WOW—a World Model of Our World," Nowack and seven other KIT researchers are taking it a step further: they are investigating how multiple AI models for various processes in the Earth's system can be coupled into a single AI system across spatial and temporal scales. To achieve this, they plan to use an approach from computer science that enables the linking of different models—into an AI world model.

How the AI World Model is Created

With WOW, the scientists are developing methods to connect various AI models. These include, among others, emulators for global climate models, AI-based weather models, and models that represent local phenomena such as wildfires or floods. The goal is to link these previously separate systems into a continuous process chain to represent the local impacts of global changes more consistently. To enable these advancements, new developments in AI methodologies are particularly required.

By coupling the different models, the researchers also aim to better understand the interactions between the atmosphere, water, and land surface. "We are interested in how changes in one part of the Earth's system affect other areas—such as how droughts or altered cloud formation trigger feedbacks in the climate," says Professor Almut Arneth from the Institute of Meteorology and Climate Research—Atmospheric Environmental Research, at KIT's Alpine Campus in Garmisch-Partenkirchen, who is also involved in the research project. "It is possible that we might uncover previously hidden connections of climate change this way."

Also Relevant for Other Fields of Knowledge

In the medium term, the new AI world model could help to better assess risks—and make informed decisions for climate protection and adaptation. "Our methods could also be applied in other natural sciences where complex systems are modeled," explains Dr. Markus Götz from the Scientific Computing Center at KIT, another scientist involved in the project. "If we learn to efficiently couple AI models, we can understand complex relationships faster and more accurately. Additionally, we can integrate observational data much more directly into the modeling process. This overall offers tremendous opportunities for science." The Carl Zeiss Foundation is funding the WOW project for five years with a total of $6.5 million USD. (mk)

Secure and Compliant Authentication in Laboratories

Secure & Contactless User Authentication at GMP/GLP