NASA, IBM and Germany-based research institution Forschungszentrum Julich have launched an expanded version of a geospatial artificial intelligence foundational model that could help track changes in land use, predict crop yields and monitor disasters worldwide.
What Is the Prithvi Geospatial AI Foundational Model?
NASA said Wednesday the open-source Prithvi Geospatial AI foundation model is pre-trained on the space agency’s Harmonized Landsat and Sentinel-2, or HLS, dataset and can be accessed through the data science platform Hugging Face.
“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, chief science data officer at NASA.
“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,” added Murphy.
Improving Earth Science Applications
With the inclusion of global HLS data, the Prithvi Geospatial AI foundation model has been tested in several applications, such as burn scar mapping, landslide detection, post-disaster flood mapping, above ground biomass estimation and estimation of gross primary productivity.
According to NASA, the model can identify crop patterns by using the Department of Agriculture’s crop data and analyzing time-sequenced data.
NASA and IBM first launched the AI model in August 2023.
