dash.ensembleFilter

Implement common utilities for ensemble-based data assimilation filters

---

Description

The ensembleFilter class implements utilities that are used by both the Kalman Filter, and the particle filter. The primary purpose of this class is to parse and error check the essential data inputs to these filters. Specifically, the observations, uncertainties, estimates, and prior. The class also includes a utilities for loading specific priors or R covariances, and also for checking that a filter object has the inputs required to run an algorithm

---

Methods

General

label - Set or return a label for the filter

name - Return a name for use in error messages

finalize - Ensure that a filter object has the required inputs to run an algorithm

Inputs

observations - Parse and process observations for a filter

prior - Parse and process priors for a filter

estimates - Parse and process estimates for a filter

uncertainties - Parse and process uncertainties for a filter

Error Checking

processWhich - Processes whichR and whichPrior inputs

assertValidR - Ensures that observations do not have NaN uncertainties

Query

Rcovariance - Returns an R covariance matrix for queried time steps and sites

loadPrior - Loads a prior from an evolving set

Console Display

dispFilter - Display details about a filter object in the console

Tests

tests - Unit tests for the ensembleFilter class