EEDL 0.2023.10.18

Code to export arbitrarily large areas of data from Earth Engine to local storage and optionally process data locally afterward

Earth Engine Downloader

EEDL is a Python package that makes downloading and processing of bulk data from Earth Engine feasible and simple. Current support includes individual image exports, as well as a helper class that will iterate through items in a filtered ImageCollection and export them all iteratively.

Many existing workflows exist for downloading areas small enough to fit into a single tile, but this tool uses Earth Engine's functionality to tile larger and full resolution exports, then download the pieces and reassemble them, with optional further processing the data using an arbitrary function (zonal statistics tools are included).

Earth Engine's export quotas still apply, especially for EECUs. For academic accounts, they are frequently generous - we have not tested them on a commercial account.

Installation

The package is still in development and we have not yet published to PyPI (pip) or conda, but have built infrastructure for both. Current installation is to download this repository then run python setup.py install

EEDL is tested on Python 3.8-3.11 on Windows and Linux with both standard CPython and Anaconda distributions. EEDL is pure Python, but depends on GDAL, which has numerous compiled C++ dependencies.

Windows

Windows users may want to use Anaconda, or see this writeup about installing GDAL and other spatial packages on Windows.

Linux

Linux users should follow the GDAL installation guide and 1) Ensure that the gdal-bin and gdal-dev packages are installed and 2) The gdal version they install for Python matches the gdal version of the system packages (ogrinfo --version). We don't pin a version of GDAL to allow for this workflow. Further details in the GDAL documentation

Documentation

Documentation is under development at https://eedl.readthedocs.io. API documentation is most complete, but noisy right now. We are working on additional details to enable full use of the package.

Licensing

Licensing is still in progress with the University of California, but we are aiming for a permissive license. More to come.

Authors

EEDL has been built by Nick Santos and Adam Crawford as part of the Secure Water Future project. This work is supported by Agriculture and Food Research Initiative Competitive Grant no. 2021-69012-35916 from the USDA National Institute of Food and Agriculture. EEDL was built in support of Water3D