IBM Releases Open-Source AI Model On Hugging Face For Weather, Climate Applications

IBM Corp. has announced a new artificial intelligence foundation model for weather and climate use cases, available in open-source to the scientific, developer and business communities. Developed by IBM and NASA, the model aims to address a variety of challenges related to short-term weather as well as long-term climate projection.

According to IBM, because of its design and training regime, the foundation model can tackle more applications than existing weather AI models, as outlined in a paper recently published on arXiv, "Prithvi WxC: Foundation Model for Weather and Climate." Potential applications include creating targeted forecasts based on local observations, detecting and predicting severe weather patterns, improving the spatial resolution of global climate simulations, and improving how physical processes are represented in numerical weather and climate models.

The model was pre-trained on 40 years of Earth observation data from NASA's Modern-Era Retrospective analysis for Research and Applications, Version 2, IBM said. As a foundation model, it has an architecture that allows it to be fine-tuned to global, regional and local scales. It is available for download on Hugging Face, along with two fine-tuned versions that tackle specific scientific and industry-relevant applications:

• Climate And Weather Data Downscaling: A common meteorological practice is downscaling—inferring high-resolution outputs from low-resolution variables. Typical data inputs include temperature, precipitation and surface winds, all of which can have varied resolutions. The model can depict both weather and climate data at up to 12 times resolution, generating localised forecasts and climate projections. The downscaling model is available on the IBM Granite page on Hugging Face.

• Gravity Wave Parameterisation: Gravity waves are ubiquitous throughout the atmosphere and can affect many atmospheric processes related to climate and weather, such as cloud formation and aircraft turbulence. The model can help scientists better estimate gravity wave generation to improve the accuracy of numerical weather and climate models and constrain uncertainty when simulating future weather and climate events. This model is being released as part of the NASA-IBM Prithvi family of models on Hugging Face.

"Advancing NASA's Earth science for the benefit of humanity means delivering actionable science in ways that are useful to people, organisations and communities," said Karen St. Germain, director of the Earth Science Division of NASA's Science Mission Directorate. "The NASA foundation model will help us produce a tool that people can use: weather, seasonal and climate projections to help inform decisions on how to prepare, respond and mitigate." 

"This space has seen the emergence of large AI models that focus on a fixed dataset and single use case—primarily forecasting. We have designed our weather and climate foundation model to go beyond such limitations so that it can be tuned to a variety of inputs and uses. For example, the model can run both on the entire earth as well as in a local context," added Juan Bernabe-Moreno, director of IBM's Accelerated Discovery Lead for Climate and Sustainability.