Ensemble Kalman filter (EnKF) is a variant of Kalman filter, where pdf of a model state is represented by an ensemble of the model state (Evensen, G. [1994]). The mean of the model state is simply approximated by the average of the ensemble state. At analysis time, the forecast error covariance is computed by using the ensemble of model forecasts. In this way, its implementation becomes relatively easy. However, the accuracy of this technique subjects to the ensemble size: the larger the ensemble, the more accurate the estimate of error covariance.