IBM and NASA open source an AI model for geospatial data analysis

With the model, the companies aim to make NASA's satellite data more accessible and accelerate climate-related discoveries

Earlier last week, we learned that NASA is launching its own streaming platform for space enthusiasts. Today, the organization has teamed up with IBM and HuggingFace to announce a new open-source project.

According to NASA, via Engadget, it estimates that its Earth Science missions would generate roughly 250,000 terabytes of data in 2024, and to sift through the climate and research data, the companies have built an open-source geospatial foundation model.

The model will serve as the basis for creating new kinds of climate and Earth science AIs, which could potentially monitor deforestation, forecast crop yields, and measure greenhouse gas emissions using satellite imagery.

The project is based on Watsonx.ai, IBM’s recently-released foundation model that can learn from multiple domains and tasks. The model was trained on a year’s worth of NASA’s Harmonized Landsat Sentinel-2 (HLS) data, which covers the Earth’s land surface and coastal regions in 13 spectral bands.

“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.”

IBM also fine-tuned the model to allow users to map the extent of past U.S. floods and wildfires, which can help predict future areas of risk.

HuggingFace, on the other hand, is hosting the open-source model on its platform, allowing other researchers and developers to access, share and build on the model. “AI remains a science-driven field, and science can only progress through information sharing and collaboration,” said Jeff Boudier, head of product and growth at Hugging Face. “This is why open-source AI and the open release of models and datasets are so fundamental to the continued progress of AI, and making sure the technology will benefit as many people as possible.”

For more information on the collaboration, click here.

Image credit: IBM

Source: IBM, Via: Engadget