4 min read<\/p>\n
Expanded AI Model with Global Data Enhances Earth Science Applications\u00a0<\/h1>\n<\/div>\n<\/div>\n<\/div>\n\n\n\n![\"A](\"https:\/\/images-assets.nasa.gov\/image\/GSFC_20171208_Archive_e001431\/GSFC_20171208_Archive_e001431~large.jpg?w=1920&h=1920&fit=clip&crop=faces%2Cfocalpoint\")
<\/a><\/figure>\nOn 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.<\/div>\nNASA Earth Observatory<\/div>\n<\/figcaption><\/div>\n<\/div>\n<\/div>\nNASA, IBM, and Forschungszentrum J\u00fclich 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.\u00a0<\/p>\n
The Prithvi Geospatial <\/a>foundation model, first released in August 2023 by NASA and IBM, is pre-trained on NASA\u2019s Harmonized Landsat and Sentinel-2<\/a> (HLS) dataset and learns by filling in masked information. The model is available on Hugging Face<\/a>, 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.\u00a0<\/p>\n\u201cWe\u2019re excited about the downstream applications that are made possible with the addition of global HLS data to the Prithvi Geospatial foundation model. We\u2019ve embedded NASA\u2019s scientific expertise directly into these foundation models, enabling them to quickly translate petabytes of data into actionable insights,\u201d said Kevin Murphy, NASA chief science data officer. \u201cIt\u2019s like having a powerful assistant that leverages NASA\u2019s knowledge to help make faster, more informed decisions, leading to economic and societal benefits.\u201d<\/p>\n
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.<\/p>\n
\n\n\n![\"Three](\"https:\/\/assets.science.nasa.gov\/dynamicimage\/assets\/science\/cds\/ai-for-science\/features\/prithvi-crop.png?w=2548&h=1102&fit=clip&crop=faces%2Cfocalpoint\")
<\/a><\/figure>\nCrop classification prediction generated by NASA and IBM\u2019s open-source Prithvi Geospatial artificial intelligence model.<\/div>\n<\/figcaption><\/div>\n<\/div>\n<\/div>\nFocusing 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.\u00a0<\/p>\n
The Prithvi Geospatial foundation model has already proven valuable in several applications, including post-disaster flood mapping and detecting burn scars caused by fires.<\/p>\n
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\u2019s surface over time. This approach supports a variety of applications, including environmental monitoring and agricultural planning. \u00a0<\/p>\n
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\u2019s Crop Data, Prithvi Geospatial can accurately identify crop patterns, which in turn could improve agricultural monitoring and resource management on a large scale.\u00a0<\/p>\n
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.\u00a0<\/p>\n
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.<\/p>\n
\n\n\n![\"Three](\"https:\/\/assets.science.nasa.gov\/dynamicimage\/assets\/science\/cds\/ai-for-science\/features\/prithvi-burn-scar.png?w=2534&h=884&fit=clip&crop=faces%2Cfocalpoint\")
<\/a><\/figure>\nBurn scar mapping generated by NASA and IBM\u2019s open-source Prithvi Geospatial artificial intelligence model.<\/div>\n<\/figcaption><\/div>\n<\/div>\n<\/div>\nThis 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.\u00a0<\/p>\n
\u201cThe updates to this Prithvi Geospatial model have been driven by valuable feedback from users of the initial version,\u201d said Rahul Ramachandran, AI foundation model for science lead and senior data science strategist at NASA\u2019s Marshall Space Flight Center in Huntsville, Alabama. \u201cThis 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.\u201d<\/p>\n
The Prithvi Geospatial Foundation Model was developed as part of an initiative of NASA\u2019s Office of the Chief Science Data Officer <\/a>to unlock the value of NASA\u2019s vast collection of science data using AI. NASA\u2019s Interagency Implementation and Advanced Concepts Team<\/a> (IMPACT), based at Marshall, IBM Research<\/a>, and the J\u00fclich Supercomputing Centre, Forschungszentrum, J\u00fclich, designed the foundation model on the supercomputer J\u00fclich Wizard for European Leadership Science <\/a>(JUWELS), operated by J\u00fclich Supercomputing Centre. This collaboration was facilitated by IEEE Geoscience and Remote Sensing Society<\/a>. \u00a0<\/p>\nFor more information about NASA\u2019s strategy of developing foundation models for science, visit https:\/\/science.nasa.gov\/artificial-intelligence-science<\/a>.<\/p>\n\n\n\n\n\n\nShare<\/h2>\n<\/div>\n\n
<\/a><\/figure>NASA, IBM, and Forschungszentrum J\u00fclich 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.\u00a0<\/p>\n
The Prithvi Geospatial <\/a>foundation model, first released in August 2023 by NASA and IBM, is pre-trained on NASA\u2019s Harmonized Landsat and Sentinel-2<\/a> (HLS) dataset and learns by filling in masked information. The model is available on Hugging Face<\/a>, 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.\u00a0<\/p>\n \u201cWe\u2019re excited about the downstream applications that are made possible with the addition of global HLS data to the Prithvi Geospatial foundation model. We\u2019ve embedded NASA\u2019s scientific expertise directly into these foundation models, enabling them to quickly translate petabytes of data into actionable insights,\u201d said Kevin Murphy, NASA chief science data officer. \u201cIt\u2019s like having a powerful assistant that leverages NASA\u2019s knowledge to help make faster, more informed decisions, leading to economic and societal benefits.\u201d<\/p>\n 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.<\/p>\n 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.\u00a0<\/p>\n The Prithvi Geospatial foundation model has already proven valuable in several applications, including post-disaster flood mapping and detecting burn scars caused by fires.<\/p>\n 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\u2019s surface over time. This approach supports a variety of applications, including environmental monitoring and agricultural planning. \u00a0<\/p>\n 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\u2019s Crop Data, Prithvi Geospatial can accurately identify crop patterns, which in turn could improve agricultural monitoring and resource management on a large scale.\u00a0<\/p>\n 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.\u00a0<\/p>\n 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.<\/p>\n 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.\u00a0<\/p>\n \u201cThe updates to this Prithvi Geospatial model have been driven by valuable feedback from users of the initial version,\u201d said Rahul Ramachandran, AI foundation model for science lead and senior data science strategist at NASA\u2019s Marshall Space Flight Center in Huntsville, Alabama. \u201cThis 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.\u201d<\/p>\n The Prithvi Geospatial Foundation Model was developed as part of an initiative of NASA\u2019s Office of the Chief Science Data Officer <\/a>to unlock the value of NASA\u2019s vast collection of science data using AI. NASA\u2019s Interagency Implementation and Advanced Concepts Team<\/a> (IMPACT), based at Marshall, IBM Research<\/a>, and the J\u00fclich Supercomputing Centre, Forschungszentrum, J\u00fclich, designed the foundation model on the supercomputer J\u00fclich Wizard for European Leadership Science <\/a>(JUWELS), operated by J\u00fclich Supercomputing Centre. This collaboration was facilitated by IEEE Geoscience and Remote Sensing Society<\/a>. \u00a0<\/p>\n For more information about NASA\u2019s strategy of developing foundation models for science, visit https:\/\/science.nasa.gov\/artificial-intelligence-science<\/a>.<\/p>\n<\/a><\/figure><\/a><\/figure>Share<\/h2>\n<\/div>\n