This open source project originated as a 'hands-on' tutorial for the RIKEN International School on Data Assimilation (RISDA2018).
The tutorial focuses on the Lorenz-63 (Lorenz 1963) model. The purpose is to assimilate observational data drawn from a 'true' driver system to synchronize a response system (the data assimilation system). Some of the most popular data assimilation methods are made available for experimentation, including: Optimal Interpolation 3D-Var Ensemble Kalman Filter 4D-Var Particle Filter Hybrid Filter
This is based on code developed for: Penny, S.G., 2017: Mathematical foundations of hybrid data assimilation from a synchronization perspective Chaos 27, 126801 (2017); https://doi.org/10.1063/1.5001819. https://aip.scitation.org/doi/full/10.1063/1.5001819
The second application currently available is the Modular Arbitrary Order Ocean Atmospehre Model (MAOOAM; https://github.com/Climdyn/MAOOAM). This is an example of a coupled data assimilation system (CDA; the model is a 2-layer atmosphere and 1-layer ocean), in which multiple scales must be addressed simultaneously. The code has been developed for and by the class of AOSC658E at the University of Maryland to study CDA.
~Stephen G. Penny
Currently:
Principal Research Scientist
Sofar Ocean
San Francisco, CA
Research Affiliate
Cooperative Institute for Research in Environmental Sciences at the University of Colorado Boulder
Visiting Professor
University of Maryland College Park