IBM and NASA Unveil AI Model for Weather and Climate Crisis

As unpredictable weather patterns increase across the globe, so does the looming threat of climate change and consequentially, the need for advanced tools to understand and predict atmospheric phenomena.

AI has emerged as a powerful tool in this fight, offering unprecedented capabilities to analyse vast datasets, identify trends and predict future climate scenarios.

By leveraging AI, scientists and policymakers can make more informed decisions to mitigate the impacts of climate change and build a more sustainable future.

One significant development in this area is the collaboration between IBM and NASA to create a new AI foundation model specifically designed to address weather and climate-related challenges.

This model represents a step forward in the field of climate science and technology, offering a versatile and adaptable tool for a range of applications.

IMB and NASA’s tool for climate analysis

The AI foundation model, developed in partnership with Oak Ridge National Laboratory, is a highly flexible tool capable of handling a wide range of tasks related to short-term weather forecasting and long-term climate projection.

Karen St. Germain, director of the Earth Science Division of NASA's Science Mission Directorate, says: "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." 
 

The AI models technical specifications and availability

The AI foundation model was pre-trained on 40 years of Earth observation data from NASA's MERRA-2 dataset, providing it with a solid foundation of knowledge about the Earth's climate system.

Its architecture allows for fine-tuning at various scales, making it suitable for both global and regional applications.

The model is now also available for download on Hugging Face, a popular platform for machine learning models.

Additionally, two fine-tuned versions of the model have been developed to address specific scientific and industry-relevant applications:

Climate and weather data downscaling: This version can downscale climate and weather data to higher resolutions, enabling more localised forecasts and climate projections.
​​​​​​​Gravity wave parameterisation: This version aims to improve the representation of gravity waves in numerical models, which can enhance the accuracy of weather and climate predictions.

The impact of IBM and NASA’s collaboration 

IBM and NASA’s collaboration dates back to the 1960s when IBM computers were used in NASA's Apollo missions.

However, in recent years, the two organisations have focused on leveraging AI to address climate change.

One key outcome of this partnership is the IBM watsonx.ai geospatial foundation model, built using NASA's satellite data.

This model is designed to analyse global weather patterns, track land use changes and predict crop yields.

The collaboration has also produced an AI model trained on nearly 300,000 Earth science journal articles to organise and make scientific literature more accessible.

This project uses IBM's PrimeQA, an open-source multilingual question-answering system.

Juan Bernabe-Moreno, Director of IBM Research Europe (UK and Ireland) and IBM's Accelerated Discovery Lead for Climate and Sustainability concludes: "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.”

******

Make sure you check out the latest edition of Technology Magazine and also sign up to our global conference series - Tech & AI LIVE 2024

******

Technology Magazine is a BizClik brand