Unified Access to Geospatial Data and AI Pipelines — an open-source platform unifying scattered geospatial labels from the Global South into AI-ready pipelines for climate resilience, food security, and weather forecasting.
Climate science in the Global South is severely hampered by label fragmentation. Geospatial labels critical for AI — crop types, land cover, building footprints, field polygons — exist across dozens of incompatible repositories, formats, and metadata standards. This makes it nearly impossible to train reliable AI models for the regions that need them most.
Three components working together to make geospatial data AI-ready.
Aggregating and standardizing multi-source geospatial labels using Croissant/GeoCroissant metadata standards.
Serving data in formats like GeoParquet, Cloud-Optimized GeoTIFF (COG), and Zarr for efficient access.
Ready-to-use pipelines for training resource-efficient models and fine-tuning geospatial foundation models on Global South observations.
Targeting the most impactful domains for the Global South.
Building tools and models to help communities adapt to climate change impacts.
Crop type mapping, yield prediction, and agricultural monitoring at scale.
Improving weather forecasting and climate modeling for underserved regions.
Everything we build is open and freely available to the community.
uGAP targets five pilot countries across Africa to validate and refine the platform.
A collaborative effort bringing together leading organizations in geospatial AI and climate science.