cryoswath

cryoswath is a python package containing processing pipelines, a tool library, and some pre-assembled data to retrieve and study CryoSat-2 data.

Adaptability lies at its core. The user can access many options simply by passing arguments to functions; everything else can be customized changing the concerned function or adding a new one.

🌱 state

Currently, it is in the beta phase. main contains those parts that I believe to work if used as intended and that tested to some extent. Other branches are for development.

✨ features

• find all CryoSat-2 tracks passing over your region of interest
• download L1b data from ESA
• retrieve swath elevation estimates
• aggregate point data to gridded data
• fill data gaps using tested methods
• calculate change rates

🚀 getting started

To use cryoswath, pull this repo and download ArcticDEM and the RGI glacier and complex shape files into the data/auxiliary/DEM and -RGI directories. Then, either use the provided docker container or set up an environment and install the software dependencies. If you choose the latter, you will have to either limit xarray to <2024.3 or to apply a small patch.

with Docker 🐳

1. docker run --detach --interactive --volume <base dir>:/altimetry_project cryoswath/cryoswath:nightly
2. connect with your favorite IDE or docker exec --interactive <container hash> sh

with conda 🐍

1. conda create --name env_name --file <base dir>/docker/conda_requirements.txt
2. conda activate env_name
3. conda install patch
4. find -name variables.py -path */env_name/*/xarray/coding/* -exec patch {} <base dir>/docker/custom_xarray.patch \; (the patch works for xarray=2024.9.0)

(The last two steps are necessary to change *= to x=x* in the xarray code.)

📖 documentation

j-haacker.github.io/cryoswath

dependencies

• requirements.txt
• reference elevation model
• glacier outlines

cryoswath will point you to the required resources.

🐛 known issues

• ! compatibility issues with xarray >= v2024.3.0
  -> apply patch as described in getting started -> conda -> steps 4+5
  -> \or downgrade to 2024.1.1
  If not fixed, xarray throws an error when you load the raw data. The error message points you to a line in the function _scale_offset_decoding where x*=y needs to be spelled out as x=x*y because the operator is not defined for the data type.

• projected RGI basins sometimes "invalid" -> add .make_valid() if it is missing somewhere

• it has mostly been tested for the Arctic

  Further: see open issues.

citation and attribution

You can cite this package using bibtex:

@misc{cryoswath,
  author = {J. Haacker},
  title = {cryoswath: CryoSat-2 swath processing package},
  year = {2024},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/j-haacker/cryoswath}}
}

Please mind that you likely used other resources on the way.

• ESA provides the L1b data under these Terms and Conditions
• RGI data is distributed under CC-BY-4.0 license
• if you (likely) used DEMs of the PGC, see their Acknowledgement Policy
• the many python packages and libraries this package depends on; some of which are indispensable.

📜 license

MIT. See LICENSE.txt.