Jump to content

Expanded AI Model with Global Data Enhances Earth Science Applications 


Recommended Posts

  • Publishers
Posted

4 min read

Expanded AI Model with Global Data Enhances Earth Science Applications 

A false-color image of the East Peak fire burning in southern Colorado near Trinidad, captured by the Operational Land Imager on the Landsat 8 satellite. Dark green forests and light green grasslands cover most of the image, but a red patch in the middle represents a burn scar, and some orange spots around it represent actively burning areas.
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 artificial intelligence model, which can help detect burn scars.
NASA Earth Observatory

NASA, IBM, and Forschungszentrum Jülich have released an expanded version of the open-source Prithvi Geospatial artificial intelligence (AI) foundation model to support a broader range of geographical applications. Now, with the inclusion of global data, the foundation model can support tracking changes in land use, monitoring disasters, and predicting crop yields worldwide. 

The Prithvi Geospatial foundation model, first released in August 2023 by NASA and IBM, is pre-trained on NASA’s Harmonized Landsat and Sentinel-2 (HLS) dataset and learns by filling in masked information. The model is available on Hugging Face, a data science platform where machine learning developers openly build, train, deploy, and share models. Because NASA releases data, products, and research in the open, businesses and commercial entities can take these models and transform them into marketable products and services that generate economic value. 

“We’re excited about the downstream applications that are made possible with the addition of global HLS data to the Prithvi Geospatial foundation model. We’ve embedded NASA’s scientific expertise directly into these foundation models, enabling them to quickly translate petabytes of data into actionable insights,” said Kevin Murphy, NASA chief science data officer. “It’s like having a powerful assistant that leverages NASA’s knowledge to help make faster, more informed decisions, leading to economic and societal benefits.”

AI foundation models are pre-trained on large datasets with self-supervised learning techniques, providing flexible base models that can be fine-tuned for domain-specific downstream tasks.

Three images show the process of crop classification with NASA and IBM’s open-source Prithvi Geospatial artificial intelligence model. The first shows a true color composite image of a cropland area. The second shows a Ground Truth Mask of the types of land cover in the image — natural vegetation is colored red, forest is orange, corn is yellow, soybeans are light green, wetlands are mid-green, developed or barren land is darker green, open water is light blue, winter wheat is mid-blue, alfalfa is dark blue, fallow/idle cropland is dark purple, cotton is pink, and sorghum is dark pink. The third image shows a Predicted Mask that closely matches the Ground Truth Mask.
Crop classification prediction generated by NASA and IBM’s open-source Prithvi Geospatial artificial intelligence model.

Focusing on diverse land use and ecosystems, researchers selected HLS satellite images that represented various landscapes while avoiding lower-quality data caused by clouds or gaps. Urban areas were emphasized to ensure better coverage, and strict quality controls were applied to create a large, well-balanced dataset. The final dataset is significantly larger than previous versions, offering improved global representation and reliability for environmental analysis. These methods created a robust and representative dataset, ideal for reliable model training and analysis. 

The Prithvi Geospatial foundation model has already proven valuable in several applications, including post-disaster flood mapping and detecting burn scars caused by fires.

One application, the Multi-Temporal Cloud Gap Imputation, leverages the foundation model to reconstruct the gaps in satellite imagery caused by cloud cover, enabling a clearer view of Earth’s surface over time. This approach supports a variety of applications, including environmental monitoring and agricultural planning.  

Another application, Multi-Temporal Crop Segmentation, uses satellite imagery to classify and map different crop types and land cover across the United States. By analyzing time-sequenced data and layering U.S. Department of Agriculture’s Crop Data, Prithvi Geospatial can accurately identify crop patterns, which in turn could improve agricultural monitoring and resource management on a large scale. 

The flood mapping dataset can classify flood water and permanent water across diverse biomes and ecosystems, supporting flood management by training models to detect surface water. 

Wildfire scar mapping combines satellite imagery with wildfire data to capture detailed views of wildfire scars shortly after fires occurred. This approach provides valuable data for training models to map fire-affected areas, aiding in wildfire management and recovery efforts.

Three images show the process of burn scar mapping with NASA and IBM’s open-source Prithvi Geospatial artificial intelligence model.The first shows a true color composite satellite image, which contains shades of green and purple. The second shows a Ground Truth Mask, which shows the true extent of a burn scar on the land by blacking out the area around the burn scar. The third shows a Predicted Mask, which almost exactly matches the Ground Truth Mask.
Burn scar mapping generated by NASA and IBM’s open-source Prithvi Geospatial artificial intelligence model.

This model has also been tested with additional downstream applications including estimation of gross primary productivity, above ground biomass estimation, landslide detection, and burn intensity estimations. 

“The updates to this Prithvi Geospatial model have been driven by valuable feedback from users of the initial version,” said Rahul Ramachandran, AI foundation model for science lead and senior data science strategist at NASA’s Marshall Space Flight Center in Huntsville, Alabama. “This enhanced model has also undergone rigorous testing across a broader range of downstream use cases, ensuring improved versatility and performance, resulting in a version of the model that will empower diverse environmental monitoring applications, delivering significant societal benefits.”

The Prithvi Geospatial Foundation Model was developed as part of an initiative of NASA’s Office of the Chief Science Data Officer to unlock the value of NASA’s vast collection of science data using AI. NASA’s Interagency Implementation and Advanced Concepts Team (IMPACT), based at Marshall, IBM Research, and the Jülich Supercomputing Centre, Forschungszentrum, Jülich, designed the foundation model on the supercomputer Jülich Wizard for European Leadership Science (JUWELS), operated by Jülich Supercomputing Centre. This collaboration was facilitated by IEEE Geoscience and Remote Sensing Society.  

For more information about NASA’s strategy of developing foundation models for science, visit https://science.nasa.gov/artificial-intelligence-science.

Share

Details

Last Updated
Dec 04, 2024

Related Terms

View the full article

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Similar Topics

    • By European Space Agency
      Image: A close-up view of Vienna, Austria’s capital city, is featured in this image from April 2025. View the full article
    • By USH
      Since November 2024, strange blinking lights have been reported worldwide, an unexplained phenomenon that’s left many puzzled. MrMBB333 believes he may have found a connection. 

      Also known as electrical pollution, dirty electricity refers to high-frequency voltage spikes that ride along standard power lines. These rogue signals, forms of electromagnetic interference (EMI), can spread through our infrastructure, causing devices to glitch or behave unpredictably. 
      If this interference is appearing globally, the source might be something massive, possibly deep within Earth’s core. Rogue frequencies from the core could travel up and interact with power grids, solar systems, and transmission lines, triggering widespread anomalies. 
      Supporting this idea is a discovery from NASA’s ANITA project in Antarctica. While searching for cosmic neutrinos, scientists instead detected impossible radio signals rising from deep within Earth, signals that defy current physics.  
      According to current science, these waves should have been absorbed by the Earth’s crust long before reaching the detectors. But they weren’t. 
      When researchers checked their findings against other experiments, nothing lined up. This means they didn’t detect neutrinos, but something entirely unknown. Could this be a new kind of particle? A glitch in reality? Or something even stranger? 
      Although it is not known whether the strange radio signals detected deep beneath the Antarctic ice are related to the rogue signals believed to originate from Earth's core, MrMBB333 suggests there could be a connection. He proposes that similar forms of electromagnetic interference (EMI) might be disrupting global electronics and even contributing to the mysterious blinking light phenomenon. 
      Another possible factor at play is that the magnetic field is weakening as well as Solar Cycle 25 — the current 11-year cycle of solar activity marked by the Sun’s magnetic field reversal and increasing sunspot activity. This cycle began in December 2019 and is expected to reach its peak in 2025. 
      Therefore, could this solar phenomenon be interfering with the rogue electromagnetic signals from the Earth’s core are behind the strange blinking lights observed around the world? 
      If that’s the case, although I don’t recall the blinking light phenomenon ever appearing this intensely before, then the strange lights may begin to fade as Solar Cycle 25 winds down. Still, that doesn’t explain the origin of the mysterious radio signals rising from deep beneath Antarctica’s ice.
        View the full article
    • By European Space Agency
      Image: Copernicus Sentinel-1 captured this image over part of eastern Borneo, a tropical island in Southeast Asia. View the full article
    • By NASA
      Explore This Section Earth Earth Observer Editor’s Corner Feature Articles Meeting Summaries News Science in the News Calendars In Memoriam Announcements More Archives Conference Schedules Style Guide 9 min read
      The Earth Observer Editor’s Corner: April–June 2025
      NASA’s Earth science missions have continued to demonstrate remarkable adaptability and innovation, balancing the legacy of long-standing satellites with the momentum of cutting-edge new technologies. The Terra platform, the first of three Earth Observing System flagship missions, has been in orbit since December 1999. Over a quarter-century later, four of its five instruments continue to deliver valuable data, despite recent power challenges. As of this writing, Terra’s Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) – Visible–Near Infrared (VNIR) and Thermal Infrared (TIR) bands, Multi-angle Imaging SpectroRadiometer (MISR), Moderate Resolution Imaging Spectroradiometer (MODIS), and one of the two Clouds and the Earth’s Radiant Energy Systems (CERES) instruments onboard, are all still producing science data. For reasons explained below, only the Measurement of Pollution in the Troposphere (MOPITT) instrument has been shut down completely, after 25 years of successful operations. The longevity of the Terra instruments is credited to Terra’s instrument team members, who have skillfully adjusted operations to compensate for the reduction in power and extend Terra’s scientific contributions for as long as possible.
      Terra has been experiencing power-based limitations caused by platform orbital changes and solar array impacts. On November 28, 2024, one of Terra’s power-transmitting shunt units failed. A response team reviewed Terra’s status, and discussed potential impacts and options. Consequently, the team changed the battery charge rate and reduced spacecraft power demands by placing the ASTER instrument into safe mode.
      In order to maintain power margins, the Terra team also moved the MOPITT instrument from science mode into safe mode on February 4, 2025, ceasing data collection. On April 9, 2025, the Terra project determined that additional power was needed for the platform and MOPITT was moved from safe mode and fully turned off, ending the instrument’s carbon monoxide data record of near-global coverage every three days.
      MOPITT was the Canadian Space Agency’s (CSA) contribution to the Earth Observing System. Launched as part of Terra’s payload in 1999, it became the longest-running air quality monitor in space, and the longest continuously operating Canadian space mission in history. MOPITT’s specific focus was on the distribution, transport, sources, and sinks of carbon monoxide (CO) in the troposphere – see Figure. The spectrometer’s marquee Earthdata products have included MOPITT Near Real-Time Datasets and offerings from the MOPITT Science Investigator-led Processing System (MOPITT SIPS). From tracking pollution from wildfires to providing data that informs international climate agreements, MOPITT served as a powerful tool for gathering data about pollution in the lowest portion of Earth’s atmosphere, informing research, policies, and even helping to advance forecasting models used by scientists worldwide. Congratulations to the MOPITT team for more than 25 years of groundbreaking science and international collaboration!
      To view this video please enable JavaScript, and consider upgrading to a web browser that
      supports HTML5 video
      Figure. This data visualization of total column carbon monoxide was created using MOPITT data from 2000-2019. In these maps, yellow areas have little or no carbon monoxide, while progressively higher concentrations are shown in orange, red, and dark red. Figure Credit: NASA’s Goddard Space Flight Center/SVS As chance would have it, the MOPITT Team had planned a 25th anniversary celebration in April, 10–11, 2025, at CSA headquarters in Longueuil, Quebec and online – which began one day after the instrument was shut down. The celebration was a fitting closeout to the MOPITT mission and a celebration of its accomplishments. Over the two days, more than 45 speakers shared memories and presented findings from MOPITT’s quarter-century record of atmospheric carbon monoxide monitoring. Its data showed a global decline in carbon monoxide emissions over two decades and could also track the atmospheric transport of the gas from fires and industry from individual regions. MOPITT is a testament to remarkable international collaboration and achievement. As it is officially decommissioned, its data record will continue to drive research for years to come.
      The Director General of the Canadian Space Agency—a key MOPITT partner—delivered remarks, and both Ken Jucks [NASA HQ— Program Manager for the Upper Atmosphere Research Program (UARP)] and Helen Worden [National Center for Atmospheric Research— MOPITT U.S. Principal Investigator] attended representing the U.S.
      More information is available in a recently-released Terra blog post and on the Canadian Space Agency MOPITT website.
      After continued investigation and monitoring of platform battery status, the Terra Flight Operations Team (FOT) determined there was sufficient power to resume imaging with ASTER’s VNIR bands, and as a result, ASTER once again began collecting VNIR data on January 17, 2025. Subsequently, ASTER resumed acquisitions for the TIR bands on April 15, 2025. (The ASTER Shortwave Infrared (SWIR) bands have been shut down since 2008).
      As one long-serving mission sunsets its operations, new missions are stepping in to carry forward the legacy of Earth system science with fresh capabilities and approaches. Launched on May 25, 2023, the NASA Time-Resolved Observations of Precipitation structure and storm Intensity with a Constellation of Smallsats (TROPICS) mission provides a groundbreaking approach to studying tropical cyclones using a passive microwave sounder CubeSat constellation. TROPICS uses multiple small satellites flying in a carefully engineered formation to measure precipitation structure as well as temperature and humidity profiles both within and outside of storms.
      Unlike traditional polar-orbiting satellites, TROPICS’ low-inclination orbits allow for hourly revisits over tropical regions, enabling scientists to better monitor storm structure, intensity changes, and key processes like upper-level warm core formation and convective bursts.
      The mission has already significantly contributed to operational forecasting and scientific research. With over 10 billion observations to date, TROPICS data have been used to validate storm models, support early-warning systems, and improve forecasts for events like Hurricane Franklin and Typhoon Kong-rey. Collaborations with agencies like the National Hurricane Center and the Joint Typhoon Warning Center have shown the value of TROPICS channels, particularly the 204.8 GHz channel, in identifying storm structure and intensity. The data are publicly available through the Goddard Earth Sciences Data and Information Services Center (GES DISC), and TROPICS continues to set the stage for the next generation of rapid-revisit Earth observation missions. To read more about the last two years of successful science operations with TROPICS, see NASA’s TROPICS Mission: Offering Detailed Images and Analysis of Tropical Cyclones.
      While some missions focus on monitoring atmospheric processes, others are expanding the frontiers of Earth observation in entirely different domains—ranging from seafloor mapping to land surface monitoring and beyond. NASA’s Ice, Clouds, and land Elevation Satellite–2 (ICESat-2) mission continues to provide critical data on Earth’s changing ice sheets, glaciers, and other environmental features. In March 2025, the satellite achieved a significant milestone by firing its two trillionth laser pulse, measuring clouds off the coast of East Antarctica. Despite challenges, such as a solar storm in May 2024 that temporarily disrupted operations, the mission has resumed full functionality, providing high-resolution data that has enabled scientists to map over 16 years of ice sheet changes. The mission’s advanced laser altimeter system, ATLAS, continues to deliver unprecedented detail in monitoring Earth’s changing ice sheets, glaciers, forests, and ocean floor.
      The ICESat-2 Satellite-Derived Bathymetry (SDB) workshop, held on March 17, 2025, in conjunction with the US-Hydro meeting, brought together experts and stakeholders from government, academia, and industry to explore the current capabilities and future potential of satellite-based seafloor mapping. With over 2000 journal articles referencing ICESat-2 in the context of bathymetry, the workshop underscored the growing importance of this technology in coastal management, navigation, habitat monitoring, and disaster response. For more details, see the ICESat-2 Applications Team Hosts Satellite Bathymetry Workshop report.
      As satellite technologies continue to evolve, so do the scientific communities that rely on them, bringing researchers together to share insights, refine data products, and explore new applications across a range of Earth and atmospheric science disciplines. As of early 2025, NASA’s Stratospheric Aerosol and Gas Experiment III (SAGE III) aboard the International Space Station (ISS) continues to provide critical insights into Earth’s atmospheric composition. In addition to scientific advancements, SAGE III/ISS has enhanced public accessibility to its data. In February 2025, the mission launched updates to its Quicklook and Expedited data portal, introducing a new ‘Highlights’ tab to showcase major stratospheric events and a ‘Comparisons’ tab for validating measurements with ground-based stations. These enhancements aim to make SAGE III/ISS data more accessible and increase its utilization for atmospheric research.
      The most recent SAGE III/ISS Science Team Meeting took place in October 2024 at NASA Langley Research Center and was held in hybrid format. Around 50 scientists gathered to discuss recent advancements, mission updates, and future directions in upper troposphere–stratosphere (UTS) research. The SAGE III/ISS team celebrated eight years of continuous data collection aboard the ISS and presented Version 6.0 of SAGE III/ISS data products during the meeting, which addresses previous data biases and enhances aerosol profile recovery. Presentations also covered aerosol and cloud studies, lunar-based aerosol retrievals, and collaborative projects using data from multiple satellite platforms and instruments. To learn more, see the full Summary of the 2024 SAGE III/ISS Meeting.
      Moving on to personnel announcements, I wish to extend my condolences to the friends and family of Dr. Stanley Sander, who passed away in March 2025. Sander devoted over 50 years to atmospheric science at NASA’s Jet Propulsion Laboratory, making groundbreaking contributions to stratospheric ozone research, air pollution, and climate science. His precise laboratory work on reaction kinetics and spectroscopy became foundational for atmospheric modeling and environmental policy, including the Montreal Protocol. Sander also played a key role in satellite calibration, mentored dozens of young scientists, and held several leadership positions at JPL. Remembered for his brilliance, humility, and kindness, his legacy endures through both his scientific achievements and the many lives he influenced. See In Memoriam: Dr. Stanley Sander.
      On a happier, though bittersweet, note, my congratulations to Jack Kaye [NASA HQ—Associate Director for Research, Earth Science Division (ESD)] who retired from NASA on April 30, 2025, after 42 years of distinguished service. With a background in chemistry and atmospheric science, he played a leading role in NASA’s efforts to understand Earth’s atmosphere and climate using satellite data and modeling. Throughout his career, Kaye has held various key leadership positions, managed major missions, e.g., the series of Shuttle-based Atmospheric Laboratory of Applications and Science (ATLAS) experiments, and supported the development of early-career scientists. He also represented NASA in national and international science collaborations and advisory roles. Kaye received numerous awards, published extensively, and was widely recognized for his contributions to Earth science and global climate research. I extend my sincere thanks to Jack for his many years of vital leadership and lasting contributions to the global Earth science community!
      Barry Lefer [NASA HQ—Tropospheric Composition Program Manager] has taken over as Acting Associate Director for Research in ESD. Reflecting on Kaye’s impact, Lefer said, “Jack has been a wonderful friend and mentor. The one thing about Jack that has had the biggest impact on me (besides his incredible memory) is his kindness. He has an enormous heart. He will be missed, but his impact on Earth Science will endure for a very long time!” See the full announcement, Jack Kaye Retires After a Storied Career at NASA.
      Steve Platnick
      EOS Senior Project Scientist
      Share








      Details
      Last Updated Jun 11, 2025 Related Terms
      Earth Science View the full article
    • By NASA
      A group of students huddle around two of their classmates using virtual reality headsets to get an up-close view of a rocket during Education Day with the Lake Erie Crushers on Thursday, May 15, 2025. Credit: NASA/Chris Hartenstine NASA’s Glenn Research Center headed to the ballpark for Education Day with the Lake Erie Crushers on May 15. NASA Glenn staff showcased the science of NASA using portable wind tunnel demonstrations, virtual reality simulations, and other interactives inspired by NASA’s Artemis missions.  
      NASA Glenn Research Center engineers Heath Reising, far left, and Dave Saunders, far right, provide a wind tunnel demonstration to a group of aspiring STEM professionals during Education Day with the Lake Erie Crushers on Thursday, May 15, 2025.Credit: NASA/Chris Hartenstine Guests snapped photos at an “out-of-this-world” selfie station and learned how to take the first step toward a career in the aerospace or space industry through NASA’s internship programs. The mid-day game welcomed 3,575 fans, many who came from local schools on field trips for the special day. 
      Return to Newsletter View the full article
  • Check out these Videos

×
×
  • Create New...