In a joint venture, NASA and IBM Research have reportedly developed a new Artificial Intelligence model to support various weather and climate applications. The new Prithvi-weather-climate foundational model leverages AI to significantly increase the resolution we receive, paving the way for more accurate regional and local weather and climate models.  

Understanding the weather and climate dynamics 

Large-scale base models, known as foundational models, can be optimized for a range of uses. They are trained in extensive unlabeled datasets. NASA’s Modern-Era Retrospective Analysis for Research and Applications (MERRA-2) data is used to train the Prithvi weather-climate model. Subsequently, the model leverages AI to apply patterns extracted from the original data to various additional scenarios. 

According to Karen St. Germain, director of the Earth Science Division of NASA’s Science Mission Directorate, “Advancing NASA’s Earth science for the benefit of humanity means delivering actionable science in ways that are useful to people, organizations, and communities. The rapid changes we’re witnessing on our home planet demand this strategy to meet the urgency of the moment”. “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,” Karen added. 

Unlocking a wide array of climatic applications 

With the Prithvi-weather-climate model, researchers can unlock a multitude of climatic applications that will boost weather and climate forecasting.

These applications include:

• The identification and forecasting of extreme weather patterns and natural disasters
• The development of concentrated forecasts based on localized observations
• The enhancement of global climate simulations’ spatial resolution down to the regional level
• The improvement of the depiction of the inclusion of physical processes into weather and climate models.

This model holds the potential to reshape our understanding of weather and climate dynamics. 

“These transformative AI models are reshaping data accessibility by significantly lowering the barrier of entry to using NASA’s scientific data,” said Kevin Murphy, NASA’s chief science data officer, Science Mission Directorate at NASA Headquarters. “Our open approach to sharing these models invites the global community to explore and harness the capabilities we’ve cultivated, ensuring that NASA’s investment enriches and benefits all.”  

The rise of Prithvi-weather-climate 

An open partnership between IBM Research, Oak Ridge National Laboratory, and NASA—including the agency’s Interagency Implementation and Advanced Concepts Team (IMPACT) at the Marshall Space Flight Center in Huntsville, Alabama—led to the development of Prithvi-weather-climate.  

Because of the architecture’s flexibility, Prithvi-weather-climate can depict the intricate dynamics of atmospheric physics even without data. Without sacrificing detail, this fundamental weather and climate model can be scaled to global and regional locations.  

“This model is part of our overall strategy to develop a family of AI foundation models to support NASA’s science mission goals,” said Rahul Ramachandran, who leads IMPACT at Marshall. “These models will augment our capabilities to draw insights from our vast archives of Earth observations.”   

A giant step toward democratisation of NASA’s science and observation mission 

Prithvi-weather-climate is a part of the Prithvi family of models, including models trained using Sentinel-2 and Harmonized LandSat data from NASA. In keeping with NASA’s open scientific ideals, the most recent model functions as an open collaboration to ensure that all data is available and helpful to communities worldwide. Hugging Face, a platform for data science and machine learning that assists users in creating, deploying, and training machine learning models, will host its release later this year.  

Tsendgar Lee, program manager for NASA’s Research and Analysis Weather Focus Area, High-End Computing Program, and Data for Operation and Assessment, stated that “the development of the NASA foundation model for weather and climate is an important step towards the democratization of NASA’s science and observation mission.” “We will continue developing new technology for climate scenario analysis and decision making.”  

The Office of the Chief Science Data Officer at NASA, the Global Modeling and Assimilation Office at Goddard Space Flight Center, Oak Ridge National Laboratory, Stanford University, University of Alabama in Huntsville, and Colorado State University also joined the IMPACT and IBM Research that made significant contributions to the development of Prithvi-weather-climate. 

• Source: NASA
• Image: Unsplash