{"id":422641,"date":"2026-05-08T00:23:35","date_gmt":"2026-05-07T14:23:35","guid":{"rendered":"https:\/\/science.nasa.gov\/science-research\/ai-foundation-model-in-orbit\/"},"modified":"2026-05-08T00:23:35","modified_gmt":"2026-05-07T14:23:35","slug":"nasas-prithvi-becomes-first-ai-geospatial-foundation-model-in-orbit","status":"publish","type":"post","link":"https:\/\/www.vibewire.com.au\/?p=422641","title":{"rendered":"NASA\u2019s Prithvi Becomes First AI Geospatial Foundation Model In Orbit"},"content":{"rendered":"
\n
\n
\n

4 min read<\/p>\n

NASA\u2019s Prithvi Becomes First AI Geospatial Foundation Model In Orbit<\/h1>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
\n
\"Florida<\/a><\/figure>
\n
Florida as seen from the International Space Station. A NASA geospatial AI foundation model was deployed to a platform aboard the space station for the first time, unlocking new opportunities for Earth observation.<\/div>\n
NASA<\/div>\n<\/figcaption><\/div>\n<\/div>\n<\/div>\n

A team of researchers from Adelaide University and the SmartSat Cooperative Research Center in South Australia has successfully uploaded and demonstrated NASA and IBM\u2019s open-source Prithvi Geospatial artificial intelligence (AI) foundation model aboard two in-orbit platforms, making it the first geospatial foundation model to be deployed in orbit. Trained on 13 years\u2019 worth of data, Prithvi can facilitate a wide variety of Earth observation tasks.<\/p>\n

By uploading a compressed version of Prithvi to the South Australian government\u2019s Kanyini satellite and to the Thales Alenia Space IMAGIN-e (ISS Mounted Accessible Global Imaging Nod-e) payload aboard the International Space Station, the researchers tested the model\u2019s flood and cloud detection performance across two different orbiting platforms and computing environments.<\/p>\n

\n
\n
\n
\"The<\/a><\/figure>
\n
Prithvi\u2019s demo prediction of burn scars from the Gifford Fire, which occurred northwest of Los Angeles on August 17, 2025. When deployed aboard an Earth-observing satellite, foundation models can perform advanced analyses before the data even reaches the ground.<\/div>\n
NASA<\/div>\n<\/figcaption><\/div>\n<\/div>\n<\/div>\n

The team chose Prithvi for their research because of its strong generalization across Earth observation tasks, and because of its availability as an open-source model.<\/p>\n

\u201cIf Prithvi weren\u2019t open source, I would have to train my own foundation model,\u201d said Dr. Andrew Du, the project\u2019s lead researcher, who is a postdoctoral researcher at Adelaide University and an AI engineer at the SmartSat Cooperative Research Center. \u201cHaving that model openly available saved a lot of time and effort.\u201d<\/p>\n

A foundation model is an AI model trained on an enormous amount of unlabeled data, which allows the model to begin detecting patterns in the data that humans wouldn\u2019t notice on their own. The model can then be fine-tuned for specific applications using much smaller amounts of labeled data.<\/p>\n

\n
\n
\n
\"Flooding<\/a><\/figure>
\n
Flooding around Lake Norman in North Carolina caused by Hurricane Helene on October 7, 2024. The blue areas of the image are the Prithvi foundation model demo\u2019s prediction of the extent of the flooding.<\/div>\n
NASA<\/div>\n<\/figcaption><\/div>\n<\/div>\n<\/div>\n

\u201cPrithvi is the first model of its kind to be deployed in orbit, and that demonstrates exactly why we make our AI models open source,\u201d said Kevin Murphy, chief science data officer<\/a> at NASA Headquarters in Washington, whose office led the collaboration that created Prithvi. \u201cBy sharing these tools with anyone who wants to use them, we accelerate scientific and technological development into the future.\u201d<\/p>\n

Developed by a team of data scientists from IBM and NASA\u2019s IMPACT team within the Office of Data Science and Informatics at NASA\u2019s Marshall Space Flight Center in Huntsville, Alabama, the Prithvi Geospatial model was trained on the Harmonized Landsat and Sentinel-2 dataset. This dataset compiles over a decade of global geospatial data from NASA\u2019s Landsat and ESA (European Space Agency) Sentinel-2 satellites. Prithvi can be adapted for tasks such as mapping flood plains, monitoring disasters, and predicting crop yields.<\/p>\n

\n
\n
\n
\n\t\t\t\t <\/path><\/path><\/g><\/svg>\n\t\t\t<\/div>\n
\n
\n

By sharing these tools with anyone who wants to use them, we accelerate scientific and technological development into the future.<\/span><\/h2>\n<\/p>\n<\/div>\n
\n
\n

Kevin Murphy<\/p>\n

NASA Chief Science Data Officer and Acting Chief Data Officer\/Chief AI Officer<\/p>\n<\/p>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n

Earth-observing satellites collect enormous amounts of data about our planet. Processing and analyzing the data in orbit before the satellite sends it back to Earth can help researchers gain insights more quickly. However, active satellites often can\u2019t accept large software updates because of bandwidth limits, so the AI models they carry for data analysis tend to be lightweight and highly specialized.<\/p>\n

Researchers can use the flexibility of a foundation model to facilitate a wide range of Earth observation tasks in one software architecture. If they want the model to take on a new task once the satellite is in orbit, they only need to upload a small extra decoder package \u2013 using far less bandwidth than uploading a whole new model to the satellite.<\/p>\n

\n
\n
\n
\"On<\/a><\/figure>
\n
On June 22, 2013, the Operational Land Imager (OLI) on Landsat 8 captured this false-color image of the East Peak fire burning in southern Colorado near Trinidad. Burned areas appear dark red, while actively burning areas look orange. Dark green areas are forests; light green areas are grasslands. Data from Landsat 8 were used to train the Prithvi foundation model, which can help detect burn scars.<\/div>\n
NASA Earth Observatory<\/div>\n<\/figcaption><\/div>\n<\/div>\n<\/div>\n

Sending Prithvi to orbit is an early demonstration of how foundation models could transform Earth observation. In addition to data analysis, foundation models could eventually help scientists interact with the instruments collecting the data.<\/p>\n

\u201cA large language model is also a type of foundation model,\u201d Du said. \u201cIn the future, this could allow operators to interact with satellites in natural language, asking questions about onboard data or system status and receiving responses in a conversational way.\u201d<\/p>\n

The NASA team behind Prithvi continues to work on open-source foundation models trained on NASA data. A heliophysics model, Surya, was released in 2025, and the team intends to create foundation models for planetary science, astrophysics, and biological and physical sciences as well.<\/p>\n

The Prithvi Geospatial foundation model is funded by the Office of the Chief Science Data Officer<\/a> within NASA\u2019s Science Mission Directorate at NASA Headquarters in Washington. The Office of the Chief Science Data Officer advances scientific discovery through innovative applications and partnerships in data science, advanced analytics, and artificial intelligence. To learn more about NASA\u2019s AI foundation models and other AI tools for science, visit:<\/p>\n

https:\/\/science.nasa.gov\/artificial-intelligence-science<\/a><\/p>\n

By Lauren Leese
Web Content Strategist for the Office of the Chief Science Data Officer<\/em><\/strong><\/p>\n

\n
\n
\n
\n
\n
\n

Share<\/h2>\n<\/p>\n<\/div>\n
\n