In the midst of a week filled with groundbreakings, partnership announcements, and the SMD Symposium, NASA and IBM stepped up to give an important update on their current project.

Back in February, the two companies started work on an AI foundation model for NASA’s earth science satellite data that would allow for NASA’s Earth-observing satellite to analyze geospatial satellite data more quickly and more efficiently in an effort to combat climate change.

In May, IBM unveiled watsonx, their AI and data platform that will allow NASA and other enterprises to scale and accelerate impact of the most advanced AI with trusted data. Watsonx includes watsonx.ai: an enterprise studio that includes a model library consisting of three model categories: fm.geospatial — which will include the IBM-NASA geospatial model later this year via the IBM Environmental Intelligence Suite — along with fm.nlp and fm.code.

The latest update provided from IBM – United States revealed that IBM’s watsonx.ai geospatial foundation model will now be openly available as the first open-source AI with built in collaboration from NASA. This will be possible with Hugging Face, an American company that develops tools for building applications using machine learning. It is most notable for its transformers library built for natural language processing applications and its platform that allows users to share machine learning models and datasets.

Even though NASA suggests that by 2024 scientists will have 250,000 terabytes of data from new missions, scientists and researchers still face obstacles in analyzing these large datasets. With the newly available open-source AI, these obstacles will prove easier to overcome.

“We believe that foundation models have the potential to change the way observational data is analyzed and help us to better understand our planet,” said Kevin Murphy, Chief Science Data Officer, NASA. “And by open sourcing such models and making them available to the world, we hope to multiply their impact.”

The model has been trained jointly by IBM and NASA over the last year across the continental United States and has been fine-tuned on data for flood and burn scar mapping. According to IBM, it has also demonstrated to date a 15 percent improvement over state-of-the-art techniques using half as much labeled data.

With additional fine tuning, the base model can also be used for tasks like tracking deforestation, predicting crop yields, or detecting and monitoring greenhouse gasses. Researchers from IBM and NASA are also working with Clark University to adapt the model for applications such as time-series segmentation and similarity research.

IBM and open-source AI platform Hugging Face also announced that IBM’s watsonx.ai geospatial foundation model – built from NASA’s satellite data – will now be openly available on Hugging Face. It will be the largest geospatial foundation model on Hugging Face and the first-ever open-source AI foundation model built in collaboration with NASA.

Access to the latest data remains a significant challenge in climate science where environmental conditions change almost daily. And, despite growing amounts of data — estimates from NASA suggest that by 2024, scientists will have 250,000 terabytes of data from new missions — scientists and researchers still face obstacles in analyzing these large datasets. As part of a Space Act Agreement with NASA, IBM set out earlier this year to build an AI foundation model for geospatial data. And now, by making a geospatial foundation model available via Hugging Face — a recognized leader in open-source and a well-known repository for all transformer models — efforts can advance to democratize access and application of AI to generate new innovations in climate and Earth science.

“The essential role of open-source technologies to accelerate critical areas of discovery such as climate change has never been clearer,” said Sriram Raghavan, Vice President, IBM Research AI. “By combining IBM’s foundation model efforts aimed at creating flexible, reusable AI systems with NASA’s repository of Earth-satellite data, and making it available on the leading open-source AI platform, Hugging Face, we can leverage the power of collaboration to implement faster and more impactful solutions that will improve our planet.”

The model – trained jointly by IBM and NASA on Harmonized Landsat Sentinel-2 satellite data (HLS) over one year across the continental United States and fine-tuned on labeled data for flood and burn scar mapping — has demonstrated to date a 15 percent improvement over state-of-the-art techniques using half as much labeled data.

With additional fine tuning, the base model can be redeployed for tasks like tracking deforestation, predicting crop yields, or detecting and monitoring greenhouse gasses. IBM and NASA researchers are also working with Clark University to adapt the model for applications such as time-series segmentation and similarity research.

For more information about this collaboration, visit the IBM Research Blog.