Google DeepMind has unveiled its new artificial intelligence system, AlphaEarth Foundations, releasing its “Satellite Embedding dataset” through Google Earth Engine, marking a significant advancement in Earth surface analysis.
The AlphaEarth Foundations system is part of Google’s broader “Google Earth AI” initiative, designed to analyze the Earth’s surface with high accuracy by integrating data from satellites, radar, and other sources. The technology’s core innovation lies in its ability to create highly compressed digital summaries for 10×10 meter areas of the planet. According to Google’s research, this method achieves a notable reduction in error rates by approximately 23.9% and requires 16 times less storage space compared to other AI systems. A key capability of AlphaEarth Foundations is its support for “continuous time,” which enables the system to generate accurate maps for any specific date range by interpolating data to fill in gaps caused by factors like cloud cover.
The system has been tested by over 50 organizations in the past year and is already being utilized for practical applications. For instance, the organization MapBiomas is leveraging AlphaEarth Foundations in Brazil to monitor deforestation in the Amazon, while the Global Ecosystems Atlas initiative employs it to map the world’s ecosystems for conservation purposes. These applications demonstrate the system’s potential in environmental monitoring and conservation efforts.
The dataset now available through Google Earth Engine includes annual snapshots from 2017 through 2024, containing over 1.4 trillion data footprints per year. Google has emphasized that the system’s 10-meter resolution is designed for large-scale environmental monitoring and is not intended to capture information that could identify individual people or objects, thereby addressing privacy concerns.
