izvonkov's picture
Update README.md
e0f76c5 verified
metadata
license: cc-by-4.0

Dataset Card for Togo Presto Embeddings

Geospatial embeddings for Togo generated using the Presto geospatial foundation model.

Dataset Details

Presto geospatial embeddings provide a compressed representation of Earth Observation data, enabling more efficient mapping and analysis. Embeddings are generated by using the Presto encoder to compress location information, optical imagery (Sentinel-2), radar imagery (Sentinel-1), climatology data (ERA5), and elevation data (SRTM) over the course of a year (March 2019 - March 2020). Each embedding contains 128 features representing a single 10m2 pixel on Earth. Embeddings can be used in place of raw Earth Observation data for various machine-learning tasks, such as classification, clustering, and anomaly detection.

  • Curated by: Ivan Zvonkov, Gabriel Tseng, Inbal Becker-Reshef, Hannah Kerner
  • License: cc-by-4.0

Dataset Sources

Uses

Geospatial embeddings offer a novel, efficient, and accessible way to map landscape features (such as cropland).

Dataset Structure

The embeddings are represented in a series of geotiff files covering Togo. They are also available as a Google Earth Engine asset: https://code.earthengine.google.com/?asset=users/izvonkov/Togo/Presto_embeddings_v2025_06_19

Dataset Creation

Our embeddings contain modified Copernicus Sentinel-1 and Sentinel-2 data (© ESA), ERA5 data (© ECMWF/Copernicus Climate Change Service), and SRTM DEM data (NASA/USGS).

Source Data