nci / icenet

IceNet: Seasonal Arctic sea ice forecasting with probabilistic deep learing

This repository is forked from the official Icenet. For references and citations, please refer to the official repository.

This codebase accompanies the Nature Communications paper Seasonal Arctic sea ice forecasting with probabilistic deep learning. It includes code to fully reproduce all the results of the study from scratch. It also includes code to download the data generated by the study, published on the Polar Data Centre, and reproduce all the paper's figures. The demo notebook is available in the notebook folder. Additional information for setting up and running the notebook is available in the notebook and on the IceNet documentation page.

The data, model and result files are also available on Gadi storage at /g/data/wb00/icenet/. The notebook by default uses the data files from the gadi storage. However, we also provide an option for the user to switch between downloading the data locally. You need to be a memer of the wb00 project to access the data. You may ask to join NCI project wb00 on mancini.

Project structure: simplified output from tree

.
├── data
│   ├── ground_truth
│   ├── era5
│   ├── masks
├── model
│   ├── configs
│   └── pretrained
├── outputs
│   ├── forecasts
│   ├── results

Any additional instructions for running the notebook are provided in the notebook itself.

If you run into issues or have suggestions for improvement, please raise an issue or email us.