Welcome to GeeFetch¶

GeeFetch is a Python library and command-line tool for downloading large-scale satellite data from Google Earth Engine directly to your computer. Built as a higher-level wrapper around geedim, GeeFetch focuses on reproducibility, scalability, and ease of use, especially for those who want to use satellite data without being remote sensing experts.
Why GeeFetch?¶
- Reproducible data acquisition through a declarative configuration approach
- Scalable downloads for national or international coverage
- Robust execution resume your download after your internet went down
- Pre-configured defaults for popular satellite sources
- Simple CLI for straightforward data access without programming expertise
GeeFetch doesn't aim to be a general-purpose tool for crafting custom algorithms on Earth Engine data. Instead, it provides a streamlined way to get standardized satellite data that you can then use for your own applications, such as machine learning, deep learning, or other analytical workflows.
Supported Satellites¶
- Sentinel-1 (SAR imagery)
- Sentinel-2 (multispectral imagery)
- GEDI (forest structure)
- Dynamic World (land use/land cover)
- Landsat-8 (multispectral imagery)
- Palsar-2 (SAR imagery)
- NASADEM (elevation data)
Basic Example¶
# Install geefetch
pip install geefetch
# Download Sentinel-2 imagery using a configuration file
geefetch s2 -c my_config.yaml
Configuration example (my_config.yaml), targetting an area around Les Landes, France:
data_dir: ~/satellite_data
satellite_default:
aoi:
spatial:
left: -0.7
right: -0.2
bottom: 43.8
top: 44.2
epsg: 4326
temporal:
start_date: "2023-06-01"
end_date: "2023-08-31"
gee:
ee_project_ids: ["my-ee-project"] # can be more to increase throughput
max_tile_size: 10
tile_size: 1000
resolution: 10
s2:
cloudless_portion: 40
cloud_prb_threshold: 40
Note
GeeFetch is currently under active development. New satellite datasets and features may be added in the near future.