ACCESS Visualisation Recipes

This repository hosts visualisation recipes developed for the ACCESS (Australian Community Climate and Earth-System Simulator) project. These recipes enable users to easily visualise climate model data and perform analysis using Python-based tools. The recipes are part of the Model Evaluation and Diagnostics (MED) team's efforts at ACCESS-NRI and were initially developed by Owen Kaluza at ACCESS-NRI.

The recipes make use of the accessvis package to create interactive visualisations of climate data, including outputs from ACCESS-ESM models and other CMIP6 datasets.

Running the Examples

Running on GADI (Australian Research Environment)

To run the examples from this repository on the Australian Research Environment (ARE), which is hosted on the GADI system at the National Computational Infrastructure (NCI), follow the steps below to set up a JupyterLab session:

1. Pre-requisites:

   • You will need an NCI account. If you do not have one, sign up on the MyNCI website.
   • To run the examples on Gadi, join project xp65. Log in to MyNCI website and request membership. Approval may take 1-2 days.

2. Open ARE on Gadi:

   • Go to the Australian Research Environment website and log in with your NCI username and password.

3. Start JupyterLab App:

   • Select JupyterLab under Featured Apps.

4. Configure JupyterLab session:

   • Complete the following fields:
     • Walltime: Set to 4 hours for the hackathon or your session's duration.
     • Queue: Select gpuvolta.
     • Compute Size: Select 1xGPU (1 gpu, 12 cpus, 95G mem).
     • Project: Use your research project, e.g., xp65.
     • Storage: Add the storage paths:

       scratch/xp65 + gdata/xp65...
       

     • Module directories: Add:

       /g/data/xp65/public/modules
       

     • Modules: Add the environment:

       conda/access-med
       

5. Launch your JupyterLab session:

   • After configuring the session, click Launch and wait for the JupyterLab instance to be ready.
   • Once started, click Open JupyterLab to begin working with the recipes.

---

Alternative (Not on Gadi)

If you're not running on Gadi, you can still use the recipes by installing the accessvis package locally. To do this, run the following command to install the package via pip:

pip install accessvis

Once the package is installed, you can proceed to use the visualisation recipes and interact with climate model data on your local machine or other computational environments.

Features and Examples

• Visualising global temperature data: Plot historical temperature data on the Earth's surface, apply colour schemes, and create interactive visualisations.
• Region-based plots: Crop data for specific regions, such as Australia, and visualise them on the Earth.
• Image overlays: Overlay additional data, such as satellite imagery of cloud or ice cover, on the Earth's surface.
• Interactive maps and animations: Generate animations showing climate model outputs over time.

Acknowledgements

The visualisation recipes were initially developed by Owen Kaluza at ACCESS-NRI, with contributions from the Model Evaluation and Diagnostics (MED) team at ACCESS-NRI. These tools are designed to make it easier for researchers to visualise and analyse climate data outputs from the ACCESS models and CMIP6 datasets.

For more information or to contribute, please check out the documentation or open an issue in this repository.