Ensemble Kalman Filter (EnKF) has that disadvantage that the spin-up time needed to reach its asymptotic level of accuracy is longer than the corresponding spin-up time in variational methods (3D-Var or 4D-Var). This is because the ensemble has to fulfill two independent requirements, namely that the mean needs to be close to the true state, and the ensemble perturbations need to represent the “errors of the day”. As a result, there are cases such as radar observations of a severe storm, where EnKF may spin-up too slowly to be useful. A scheme is proposed to accelerate the spin-up of EnKF applying a no-cost Ensemble Kalman Smoother, and using the observations more than once in each assimilation window in order to maximize the initial extraction of information. The performance of this scheme is tested with the Local Ensemble Transform Kalman Filter (LETKF) implemented in a Quasi-geostrophic model, whose original framework requires a very long spin-up time when initialized from a cold start. Results show that with this “running in place” scheme the LETKF spins-up and converges to the optimal level of error at least as fast as 3D-Var or 4D-Var. Additional computations (2-4 iterations for each window) are only required during the initial spin-up, since the scheme returns to the original LETKF after spin-up is achieved.

In addition, we developed and tested a scheme equivalent to the outer loop of 4D-Var.