PowerPoint Presentation

• In 3D-Var B is assumed to be constant: it does not include “errors of the day” • 4D-Var is very expensive and does not provide the analysis error covariance. • In Kalman Filtering B is forecasted. It is like running the model N times, where N~106-8, so that it is impractical without simplifications

and the analysis error covariance is given by

As in the scalar case, the 3D-Var analysis is given by

where the weight matrix is