The Sheffield Ice Velocity ExploreR (SHIVER): A unified satellite-derived ice velocity dataset for Earth's ice sheets

NOTE: This code and the associated manuscript are in preparation and may be subject to change.

For more details, check out the SHIVER readthedocs

Overview

This repository contains:

1. The scripts used to generate the dataset described in the manuscript in preparation: "The Sheffield Ice Velocity ExploreR (SHIVER): an online tool for low latency exploration, analysis and sub-setting of unified satellite-derived ice velocity data for Earth's ice sheets".
2. Everything required to produce the SHIVER readthedocs website (https://shiver-zarr.readthedocs.io/), which describes how to access cloud-hosted Zarr stores of the SHIVER data.

Note: This repository only contains the code to generate the dataset. The resulting dataset is hosted by Source Cooperative at https://source.coop/uos-shiver.

Directory Structure

• src/ : Contains the Python scripts used to generate the unified dataset.
  • citations.py : Provides formatted citation records for each contributing dataset.
  • greenland/
    • build_multisource_zarr.py : Compiles a Zarr store (*spatial_200.zarr) from all contributing datasets for Greenland. The chunk definition is 1024 x 1024 x 1 (x, y, t) and the resolution is 200 x 200 m.
    • build_multisource_omezarr.py : Compiles an OME-Zarr store (*spatial.zarr) from all contributing datasets for Greenland. The chunk definition is 1024 x 1024 x 1 (x, y, t) and the resolution is telescopes from 200 m to 3200 m.
    • rechunk_zarr_safe.py : Rechunks the *spatial_200.zarr Zarr store to either a time-series optimised (chunk def: 64 x 64 x -1) Zarr or general-purpose Zarr (chunk def: 256 x 256 x 300).
    • update/
      • scrape_promice.py : Download only new PROMICE (edition 5) files.
      • append_new_promice_to_spatial_200_zarr.py : Appends new PROMICE files to the spatial_200 Zarr store.
      • append_new_promice_to_spatial_zarr.py : Appends new PROMICE files to the spatial OME-Zarr store.
  • antartica/
    • build_multisource_zarr.py : Compiles a Zarr store (*spatial_200.zarr) from all contributing datasets for Antarctica. The chunk definition is 1024 x 1024 x 1 (x, y, t) and the resolution is 200 x 200 m.
    • build_multisource_omezarr.py : Compiles an OME-Zarr store (*spatial.zarr) from all contributing datasets for Antarctica. The chunk definition is 1024 x 1024 x 1 (x, y, t) and the resolution is telescopes from 200 m to 6400 m.
    • rechunk_zarr_safe.py : Rechunks the *spatial_200.zarr Zarr store to either a time-series optimised (chunk def: 64 x 64 x -1) Zarr or general-purpose Zarr (chunk def: 256 x 256 x 300).
• docs/ : Everything required to build the SHIVER readthedocs website.

environment.yml

This code was run using Python 3.11.14. To install the necessary dependencies, you can use conda: conda env create -f environment.yml

Usage

To reproduce the dataset creation:

1. Download the raw data from the sources given in citations.py (see docs/examples/utils).
2. Update the directory paths in each script.
3. Run the scripts. Note the build*.py scripts are designed to run on a HPC, where each contributing dataset is submitted as a separate batch job.

License

This code is licensed under the GPL-3.0 License - see the LICENSE file for details.

The data are licensed under CC-BY-4.0.

Cite

A dataset description paper for SHIVER is in preparation. The citation for that paper will be something like:

Davison, B. J. et al. (in prep). The SHeffield Ice Velocity ExploreR (SHIVER): an online tool for low latency exploration, analysis and sub-setting of unified satellite-derived ice velocity data for Earth's ice sheets. [specify journal]. https://doi.org/10.xxxx/XXXXXXX

The dataset itself should also be cited like this:

Davison, B. J. (2026). The SHeffield Ice Velocity ExploreR (SHIVER): A unified satellite-derived ice velocity dataset for Earth's ice sheets (Version v[specify version number]) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.21375859

Or if your outputs relied on the SHIVER interactive application, then you should cite the application like this:

Davison, B. J. (2026). The SHeffield Ice Velocity ExploreR (SHIVER): an online tool for low latency exploration, analysis and sub-setting of unified satellite-derived ice velocity data for Earth's ice sheets (Version v[specify version number]) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.21378057

The SHIVER dataset is a compilation of many published velocity datasets and so would not have been possible without those. Therefore, those contributing datasets should also be cited. Specifically, you should cite all contributing datasets that your specific query intersected. If you are using the SHIVER application, all exports will contain citation details for the relevant contributing datasets. If you are extracting data programmatically, please see this notebook that describes how you can generate the correct citations for your query.