Author: Yue Ying (NERSC)
Data assimilation (DA) combines information from model forecasts and observations to obtain the best estimate of model state and parameters. NEDAS provides a light-weight solution for developing new DA algorithms for Earth-system models.
In this series of jupyter notebook tutorials, I demonstrate how to use NEDAS to perform DA research.
To run the notebooks, you can use one of these options: EDITO Datalab, Google Colab, Run in Docker, or Run in native environment.
A summary of purpose of each notebook:
- 1.step_by_step_with_vort2d_case.ipynb provides a step-by-step guide to running DA experiments using a simple 2D vorticity model as example.
- 2.validation_with_lorenz96.ipynb validates the NEDAS for correctness of implementation, using the classic Lorenz 1996 model cases.
Learn more from the NEDAS Documentation