This repository contains demonstration notebooks showcasing the Earthmover Platform for accessing and analyzing large-scale climate and ice sheet datasets. The demos illustrate how to use Arraylake and Icechunk to work with petabyte-scale geoscience data directly from cloud storage.
The Earthmover Platform provides seamless access to large-scale Earth observation datasets through:
- Arraylake: A cloud-native data platform for scientific datasets
- Icechunk: A version-controlled, cloud-optimized data format built on Zarr
- Direct cloud access: Query datasets without downloading, using familiar Python tools like Xarray
Demonstrates working with ERA5 weather reanalysis data:
- Discover and access datasets via Arraylake Client
- Use Icechunk and Xarray for efficient cloud-based data access
- Explore Icechunk's version history and branching capabilities
- Perform spatial and temporal analysis on the ERA5 dataset
- Create custom data subsets and write to new Icechunk repositories
Combines ICESat-2 ATL15 ice sheet data with ERA5 climate data:
- Access ATL15 dataset through Arraylake's hierarchical data structure
- Work with different spatial resolutions using Zarr groups
- Reproject climate data to match ice sheet grids
- Analyze correlations between temperature and ice sheet height changes
- Demonstrate multi-dataset analysis workflows
pip install arraylake icechunkVisit https://app.earthmover.io to explore available datasets
git clone https://github.com/earth-mover/icesat2-hackweek-demo.gitRun the notebooks and authenticate when prompted using your hackweek email address or Google account. The authentication uses OAuth2 and will open a browser window for login.
This project is released under the MIT License.