Outline

Now that we’ve built the state vector ensemble and generated proxy estimates, we’re ready to run an assimilation. DASH supports three DA algorithms, which are implemented by the kalmanFilter, particleFilter, and optimalSensor classes. Feel free to skip the tutorials for any algorithms that are not of interest to you. This section includes the pages:

DA Algorithms in DASH

We’ll start with some brief comments about the data assimilation algorithms supported in DASH. We’ll also comment on the relationship of these algorithms with other classes in the toolbox.

kalmanFilter

Then, we’ll examine the kalmanFilter class in detail. We’ll examine the essential inputs to a Kalman filter, see supported variations of the algorithm, and explore the types of outputs produced by an assimilation.

Coding 8

We’ll next use the class to run a Kalman filter assimilation.

particleFilter

Next, we’ll examine the particleFilter class. We’ll examine its essential inputs, and different particle weighting schemes implemented by the class.

Coding 9

We’ll then do a coding session to run a particle filter.

optimalSensor

Finally, we’ll explore the optimalSensor class. We’ll explore its inputs, and the different types of analyses implemented by the class.

Coding 10

In the final coding session, we’ll run an optimal sensor analysis.