Vort2D model

A step-by-step tutorial demonstrating data assimilation with NEDAS using the 2D vorticity (vort2d) model. The notebook walks through a complete OSSE (Observing System Simulation Experiment): generating a model truth, creating a perturbed ensemble, assimilating synthetic observations, and visualising error distributions and increments interactively.

Topics covered:

• Configuring an experiment with a YAML file and the Config object

• Generating a verifying truth and synthetic observations

• Building a perturbed ensemble (vortex position and background flow)

• Running one assimilation cycle: observation priors, increments, state update

• Assimilating a single wind observation to inspect ensemble error geometry

• Interactive widgets and animations for exploring observation–state correlations

The notebook can be run in several environments:

• Google Colab

• Docker (see below)

• Native Python — refer to the environment setup guide

docker pull myying/nedas-tutorials
docker run -it --rm -p 8888:8888 myying/nedas-tutorials

Then open the URL printed in the terminal and navigate to 1.step_by_step_with_vort2d_case.ipynb.

The full notebook is available on GitHub.