NASA and IBM Research Have Developed a New Artificial Intelligence Model

What is The News About?

A novel AI model supporting a variety of climate and weather applications has been developed in collaboration between IBM Research and NASA. Thanks to its innovative application of artificial intelligence (AI), the Prithvi-weather-climate fundamental model has the potential to significantly enhance the resolution we can achieve, paving the way for more accurate regional and local climate and weather predictions.

Week’s Top Read Insight:10 AI ML In Supply Chain Management Trends To Look Out For In 2024

Foundational models are large-scale, baseline models that may be fine-tuned for different uses after being trained on big, unlabeled datasets. Using artificial intelligence learning capabilities, the Prithvi-weather-climate model is trained on a large dataset, specifically NASA data from the Modern-Era Retrospective analysis for Research and Applications (MERRA-2), and then applied to a wide variety of additional scenarios based on the patterns derived from the initial data. Researchers will have a solid foundation for a wide variety of climate applications with the Prithvi-weather-climate model. These applications encompass a wide range of tasks, such as the detection and prediction of extreme weather events, the development of targeted forecasts using localized observations, the enhancement of global climate simulations’ spatial resolution to regional levels, and the enhancement of the representation of physical processes in weather and climate models.

Read: 10 AI ML In Data Storage Trends To Look Out For In 2024

Why Is It Important?

IBM Research, Oak Ridge National Laboratory, and NASA—specifically the agency’s Interagency Implementation and Advanced Concepts Team (IMPACT) in Huntsville, Alabama’s Marshall Space Flight Center—worked together openly to develop Prithvi-weather-climate.

Because of its adaptable design, Prithvi-weather-climate is able to capture the intricate dynamics of atmospheric physics even in the absence of data. Without sacrificing resolution, this fundamental weather and climate model can be scaled to both global and regional locations. The Prithvi family of models comprises Prithvi-weather-climate and other models trained using Sentinel-2 and NASA’s Harmonized LandSat data. In keeping with NASA’s open scientific ideals, the most recent model is an open cooperation that aims to make all data accessible and useable by communities worldwide. Later this year, it will be available on Hugging Face, a platform for data science and machine learning that facilitates model development, deployment, and training.

Not only did IMPACT and IBM Research play a significant role in developing Prithvi-weather-climate, but so did NASA’s Global Modeling and Assimilation Office at Goddard Space Flight Center, Oak Ridge National Laboratory, Stanford University, the University of Alabama in Huntsville, Colorado State University, and the Office of the Chief Science Data Officer at NASA.

Must Read: What is Experience Management (XM)?

Benefits

1. Enhanced Accuracy: The Prithvi-weather-climate model improves regional and local climate predictions’ accuracy significantly.

2. Versatile Applications: Supports diverse applications like extreme weather detection and regional climate simulations, enhancing research capabilities.

3. Open Collaboration: Developed with open scientific principles, making data and models accessible to global communities on Hugging Face.

[To share your insights with us as part of editorial or sponsored content, please write to sghosh@martechseries.com]