AlphaEarth Foundations Geospatial AI

2 posts · Jul 2025 – Nov 2025

What this trail covers

AlphaEarth Foundations integrates petabytes of Earth observation data (optical satellite images, radar, 3D laser mapping, climate simulations) into a unified digital representation (embeddings) for global mapping and monitoring. It analyzes land and coastal waters in 10x10 meter squares with high precision and tracks changes over time. The model achieves a 24% lower error rate compared to traditional and other AI mapping systems, excelling in scenarios with scarce label data. The Satellite Embedding dataset is released in Google Earth Engine.

All DeepMind trails

Nov 2025Three ways Google scientists use AI to better understand natureIntroduced a new AI-powered approach for producing species range maps at unprecedented scale by combining open databases of field observations with satellite embeddings from AlphaEarth Foundations and species trait information using a Graph Neural Net (GNN) model. This allows for inference of geographical distributions for many species simultaneously and refinement with additional local data. Released 23 species maps for Australian mammals via UN Biodiversity Lab and Earth Engine.
Jul 2025AlphaEarth Foundations helps map our planet in unprecedented detailThis post details the release of the Satellite Embedding dataset, powered by AlphaEarth Foundations, in Google Earth Engine. It highlights the model's ability to integrate diverse Earth observation data into a unified digital representation (embeddings) for global mapping and monitoring. The model analyzes land and coastal waters at 10x10 meter resolution, tracks changes over time with high precision, and requires 16x less storage than other tested AI systems. It achieved a 24% lower error rate than competing systems, particularly in low-label data scenarios. The post also showcases partner use cases, including ecosystem classification and agricultural/environmental change monitoring.
Advertisement