IBM and NASA have released an open-source AI model on Hugging Face, a platform for machine learning models, to aid scientists, developers, and businesses in better understanding and analyzing weather and climate data. This new AI foundation model, developed by IBM and NASA with contributions from Oak Ridge National Laboratory, offers a flexible and scalable way to address various challenges related to short-term weather forecasting and long-term climate projection.
The model can tackle applications beyond existing weather AI models, including creating targeted forecasts based on local observations, detecting and predicting severe weather patterns, and improving the spatial resolution of global climate simulations. Key individuals involved in this collaboration include Karen St. Germain, director of NASA’s Earth Science Division, Juan Bernabe-Moreno, Director of IBM Research Europe, and Arjun Shankar, director of the National Center for Computational Sciences at Oak Ridge National Laboratory.
Open-Source AI Model for Weather and Climate Applications
The recent collaboration between IBM and NASA has led to the development of an open-source AI foundation model for various weather and climate use cases. This model is designed to be flexible and scalable, allowing it to tackle a wide range of challenges related to short-term weather forecasting as well as long-term climate projection. The model’s unique architecture enables it to be fine-tuned to global, regional, and local scales, making it suitable for a variety of weather studies.
The AI foundation model was pre-trained on 40 years of Earth observation data from NASA’s Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2). This extensive training dataset allows the model to accurately reconstruct global surface temperatures from a random sample of only five percent original data. The model’s flexibility makes it an ideal tool for scientists, developers, and businesses seeking to better understand and analyze weather and climate data.
Advantages Over Existing Weather AI Models
The IBM-NASA AI foundation model offers several advantages over existing weather AI models. Its unique design and training regime enable it to tackle a broader range of applications beyond forecasting. The model can be tuned to various inputs and uses, allowing it to be applied in both global and local contexts. This flexibility makes it an ideal tool for understanding meteorological phenomena such as hurricanes or atmospheric rivers, reasoning about future potential climate risks by increasing the resolution of climate models, and informing our understanding of imminent severe weather events.
Applications in Weather Forecasting and Climate Modeling
The AI foundation model has already been tested with Environment and Climate Change Canada (ECCC) to explore its flexibility in additional weather forecasting use cases. The model is being used to test very short-term precipitation forecasts using a technique called precipitation nowcasting, which ingests real-time radar data as input. Additionally, the team is testing the downscaling approach from global model forecasts at 15 km to km-scale resolution.
Integration with Other AI Foundation Models
The weather and climate AI foundation model is part of a larger collaboration between IBM Research and NASA to use AI technology to explore our planet. It joins the Prithvi family of AI foundation models, which includes the Prithvi geospatial AI foundation model. This geospatial foundation model has been used by governments, companies, and public institutions to examine changes in disaster patterns, biodiversity, land use, and other geophysical processes.
Accessibility and Future Development
The weather and climate AI foundation model, as well as the gravity wave parameterization model, can be accessed through the NASA-IBM Hugging Face page. The downscaling model can be accessed through the IBM Granite Hugging Face page. This open-source approach enables researchers and developers to build upon the model’s capabilities, driving further innovation in weather and climate applications.
Collaboration and Impact
The collaboration between IBM, NASA, and Oak Ridge National Laboratory demonstrates the power of interdisciplinary research in advancing our understanding of complex systems like weather and climate. The development of this AI foundation model has the potential to significantly impact various industries, from agriculture to urban planning, by providing actionable insights into weather patterns and climate risks. As stated by Karen St. Germain, director of the Earth Science Division of NASA’s Science Mission Directorate, “The rapid changes we’re witnessing on our home planet demand this strategy to meet the urgency of the moment.”
External Link: Click Here For More
