DART Tutorial

The slides for the Data Assimilation Research Testbed tutorial: DART_LAB part 6: Using the Real DART System, guide you through running ensemble data-assimilation experiments using the real DART software environment.

Building on assimilation concepts covered in the DART_LAB Tutorial part 1-5 you will:

• Install and configure DART, gaining an understanding of its directory structure and build system.

• Run the Lorenz-96 model to reproduce the DART_LAB assimilation experiments within the full DART framework.

• Explore configurable parameters such as ensemble size, localization, inflation method, and observation networks — and learn how they influence assimilation performance.

• Generate synthetic observations via OSSEs (Observing System Simulation Experiments), assimilate them, and diagnose results using MATLAB tools (time-series error/spread plots, rank histograms, bias/variance diagnostics, etc.).

• Examine observation-space diagnostics, including rank histograms, RMSE/bias evolution, and outlier-observation rejection (via a configurable threshold).

• Extend your familiarity to other low-order models (such as the Lorenz 63 system) and prepare for full-scale geoscience applications using DART’s model interfaces.

The diagnostics in the DART tutorial use MATLAB®. To learn how to configure your environment to use MATLAB and the DART diagnostics, see the documentation for MATLAB observation space diagnostics.