One of the goals of the COSIMA community is to create a common framework for evaluating and diagnosing model output. Our initial attempts to do this are encapsulated in the COSIMA Cookbook. The Cookbook aims to:
- Create standard Jupyter Notebooks for diagnosing model simulations, or groups of simulations;
- Documenting methods to efficiently interrogate large model datasets;
- Document model sensitivity studies; and
- Provide a gallery highlighting commonly used diagnostics.